journal of sustainability science and management

JOURNAL OF SUSTAINABILITY SCIENCE AND MANAGEMENT

ADVISORY BOARD

CHAIRPERSONNor Aieni Haji Mokhtar, UMT, Malaysia

MEMBERS

MEMBERS

SECRETARYMohd. Effendy Abd. Wahid, UMT, Malaysia

EDITORIAL BOARD

CHIEF EDITORMohd Tajuddin Abdullah, UMT, Malaysia

DEPUTY CHIEF EDITORMhd Ikhwanuddin Abdullah, UMT, Malaysia

MANAGING EDITORAziz Ahmad, UMT, Malaysia

JSSM is indexed in the following databases:SCOPUS, ASEAN Citation Index, Zoological Record, Google Scholar, Chemical Abstract, Scimago, Directory of Open Access Journals (DOAJ) and MyCite.

COPYRIGHT AND PHOTOCOPYING. © 2018. Penerbit UMT. Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Penerbit UMT for libraries and other users registered with the Copyright Clearance Center. This consent does not extend to other kinds of copying such as copying general distribution for advertisement or promotional purposes, for creating new collective works or for resale. Special requests should be addressed to the Editor.

Abu Bakar Salleh, UPM, MalaysiaAileen Tan Shau Hwai, USM, Malaysia Faizah Mohd Sharoum, UMT, MalaysiaJane M. Hughes, Griffith University, AustraliaKadambot H.M. Siddique, University of

Western Australia, AustraliaMazlin Mokhtar, UKM, Malaysia

Abdul Manaf Ali, UniSZA, MalaysiaAziz Ahmad, UMT, MalaysiaFatimah Md. Yusoff, UPM, MalaysiaGhizan Saleh, UPM, MalaysiaHii Yii Siang, UMT, Malaysia

Mohd Azmi Ambak, UMT, MalaysiaMohd Effendy Abd. Wahid, UMT, MalaysiaNathalie Bourgougnon, Université de Bretagne- Sud, FranceNorhayati Mohd Tahir, UMT, MalaysiaPeter B. Mather, Queensland University of Technology, AustraliaRéjean Tremblay, Quebec University, Canada

Idris Abd. Ghani, UKM, MalaysiaMohd Lokhman Husain, UMT, MalaysiaMuhammad Suzuri Hitam, UMT, MalaysiaNik Fuad Nik Mohd Kamil, UMT, MalaysiaTg. Sifzizul Tg. Muhammad, UMT, Malaysia

Contents

Journal of Sustainability Science and Management Special Issue No. 4 2018The Role of Transparency and Management Integrity in Organisational Sustainability----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

i-vi Editorial: The Role of Transparency and Management Integrity in Organisational Sustainability

AKMALIA MOHAMAD ARIFF AND AZWADI ALI

1-14 Combating Corruption for Sustainable Public Services in Malaysia: Smart Governance Matrix and Corruption Risk Assessment

WAN MOHAMMAD TAUFIK WAN ABDULLAH, NURUL NADIAH AHMAD AND AKMALIA MOHAMAD ARIFF

15-33 Decision Making Approach of Direct Feeder Service Via Port Klang in Malaysia MUHAMAD NASIR RAHMATDIN, NOORUL SHAIFUL FITRI ABDUL

RAHMAN AND MOHAMMAD KHAIRUDDIN OTHMAN

35-52 Moderating Effect of Self-Efficacy on the Relationship Between Consumer Attitude and Willingness to Consume Healthy Food

HAYATUL SAFRAH SALLEH AND NOR AZILA MOHD NOOR

53-65 Enhancing Environmental Sustainability Over Fisheries Industry Through Proactive Risk Evaluation: A Case of Tok Bali Fishing Port

NURUL HAQIMIN MOHD SALLEH AND MOHD AZLAN ABDUL HALIM

67-79 An Assessment of the Perception and Satisfaction With Flood Evacuation Centre Service Quality in East Coast States of Peninsula Malaysia

SITI FALINDAH PADLEE, NIK NADRUL HISHAM NIK RAZALI, SITI NUR ATIKAH ZULKIFFLI AND NOOR ZATUL IFFAH HUSSIN

81-100 Foreign Direct Investment and Trade Openness: Do They Lead to Sustainable Development in Malaysia?

ABDUL RAHIM RIDZUAN, NOR ASMAT ISMAIL AND ABDUL FATAH CHE HAMAT

101-117 The Influence of New Economic Model on Intention to Work Abroad: An Empirical Study to Address Brain Drain Issue in Malaysia

NORHAYATI SHARIFF, JAYARAMAN KRISHNASWAMY, DAHLAN ABDULLAH AND ANG WOON CHAU

119-133 The Influence of Energy Consumption, Renewable Energy and Economic Growth on CO2 Emission in Malaysia and Indonesia

IZYAN NAZIRAH WAHID, AZLINA ABD. AZIZ, MAHIRAH KAMALUDIN AND ZURAINI ANANG

135-146 A Survey on the Willingness to Pay for Domestic Water Service Attributes in Terengganu, Malaysia

MAHIRAH KAMALUDIN, AZLINA ABD. AZIZ, NUR SYUHADA CHE IBRAHIM AND ALIAS RADAM

147-168 A Review of Corporate Governance and Corporate Social Responsibility Disclosure of Islamic Banks: A Call for Additional Transparency

SAED AHMED SULUB, ZALAILAH SALLEH AND HAFIZA AISHAH HASHIM

i

EDITORIAL: THE ROLE OF TRANSPARENCY AND MANAGEMENTINTEGRITY IN ORGANISATIONAL SUSTAINABILITY

AKMALIA MOHAMAD ARIFF* AND AZWADI ALI

School of Maritime Business and Management, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia.

*Corresponding author: [email protected]

Abstract: The International Conference on Accounting, Business and Economics (ICABEC), which is a conference held by the School of Maritime Business and Management, Universiti Malaysia Terengganu, is aimed at bringing together researchers to discuss emerging issues concerning organisational practices and behaviours. This Special Issue compiles ten selected articles that were presented during the 4th ICABEC (ICABEC 2016). With the theme “Thriving in the Emerging Economies: Transparency and Management Integrity”, ICABEC 2016 highlights good ethical conducts, exemplary best practices and top-notch alternatives that should be the focus of organisations that wish to remain viable in the ever challenging environment. In this Special Issue, three important aspects in organisational sustainability are identified. Firstly, ensuring high ethical conduct is a prerequisite towards efficient and effective utilisation of organisational resources. Secondly, development of sustainable organisational agenda is significant in sustaining the livelihood of the community and future generations. Thirdly, strengthening policies, rules and regulations might be required to align organisational goals towards achieving sustainable agenda. It is hoped that this Special Issue of ICABEC will be able to relay important messages to the organisations and related stakeholders on the importance of transparency and management integrity in ensuring the sustainability of organisations. Relevant policy implications are also expected to assist organisations in gearing their efforts towards practising high ethical conduct that can benefit them in its entirety.

Keywords: Transparency, management integrity, organisational sustainability.

Introduction

Globalisation, technological advances and internet revolution have created intense competition that require organisations to struggle to attain sustainable development. The endeavour towards getting and maintaining organisational success is made even challenging with the instability in the economic environment that is caused by economy, political, and social pressures. An aspect that is equally important towards organisational sustainability is the existence of good ethical conduct, through transparency and management integrity within the entities and their surroundings. Transparency and management integrity form a strong foundation for trust among organisations’ stakeholders. When trust is strongly established in an organisation, investors and other important

stakeholders tend to continue their support of the organisation. Conversely, the future of an organisation with poor trust among its stakeholders is questionable and this relates very much to the sustainability of the organisation. Moreover, transparency and integrity in the fields that affect society’s sustainability are continuously called for to reduce the expectation gap among the public.

In capturing this increasing importance, an international conference with the theme “Thriving in the Emerging Economies: Transparency and Management Integrity” was held in August 2016 which gathered interesting and relevant articles pertaining to the issue. This special issue of JSSM compiles ten best manuscripts to commemorate the success of the 4th International Conference on Accounting,

ii

Business and Economics (ICABEC 2016) which was organised by the School of Maritime Business and Management, Universiti Malaysia Terengganu.

Highlights of Research

This special issue begins with the article by Abdullah, Ahmad and Ariff (2018) titled “Combating Corruption for Sustainable Development of Public Services in Malaysia: Governance and Corruption Risk Assessment”. As asserted in many research findings on corruption, proactive measures should be taken to strengthen prevention and detection (Poerting & Vahlenkamp, 1998). Here, the authors propose the Smart Governance Matrix based on five governance measures to assess the corruption risk of organisations. Based on their sample, they find that enterprise governance, data governance and human governance can predict the likelihood of corruption in organisations, and that the matrix can be reliably used in typical for-profit and non-profit organisations. The study concludes that strengthening enterprise, data, and human governance is essential to eliminate corruption risk that hampers the sustainable development of the public sector in Malaysia.

The second article titled “Decision Making Approach of Direct Feeder Service via Port Klang in Malaysia” by Rahmatdin, Rahman and Othman (2018) sets a benchmark for current assessments of overall direct feeder service of route parameters which connect Port Klang and five potential ports or trade partners as alternatives. The research responds to the call by Hsu and Hsieh (2007) who had suggested that the optimal decision of direct service is more preferred than the hub and spoke service. Based on AHP-Fuzzy Link Based and Evidential Reasoning method, they find that the Port Klang – Laem Chabang route is the most preferred sailing route that can optimize feeder vessel utilisation, reduce environmental pollution, and potentially gain high returns for both feeder operators and ports.

Next, Salleh and Noor (2018)

investigate attitudinal factors that can predict consumer willingness to consume functional food. Their study is motivated by the aim to investigate the current eating habit that is linked with sustainable livelihood and good general well-being. This type of research is important because the World Health Organisation (WHO) has reported that unhealthy food, unhealthy eating habits and less physical activity may lead to the global burden of non-communicable diseases, which although non-infectious, are often referred to as chronic diseases (WHO, 2015). By using multiple regression analysis, they find that reward and necessity are predictors of willingness to consume functional food, whereas confidence and safety failed to explain the dependent variable. In addition, they have proved that self-efficacy is a significant moderator of the relationship between attitude dimensions and willingness to consume functional food. Therefore, to further encourage the public to consume healthy food, relevant authorities can focus on the reward and necessity of consuming healthy food consistently and that the access to obtain the food should be made easy.

The following study chooses a targeted fishing port in assessing its sustainability by evaluating proactively the associated risks. In this study, Salleh and Halim (2018) propose a model for analysing environmental risks arising from fisheries activities in Tok Bali Fishing Port, Malaysia. They find that water pollution is a focal environmental risk arising from fisheries activities in Tok Bali Fishing Port. They arrive at this finding by utilising a two-step approach that is prioritising environmental risk via Analytic Hierarchy Process method and suggesting mitigation strategies for minimising the risk losses from fisheries activities by using a decision-making rating tool. Long-term strategies including law enforcement, strategic fisheries management practices and environmental education to the fishermen are proposed in order to minimise environmental risks from fishing activities especially water pollution risk.

iii

Understanding the level of satisfaction among flood victims when sheltered in evacuation service centres during flood disaster is the motivation behind the study carried out by Padlee et al. (2018). They target the flood victims for their research because flood often occurs in Malaysia and the victims can be adversely affected; including potential loss of income and lives of family members (Akasah & Doraisamy, 2015). The researchers collect responses of questionnaires distributed to evacuees in three selected districts in East Coast states that were badly affected by flood events in 2014. The findings show that most evacuees perceived that the quality of the services provided to them was good and they were satisfied with the services provided in the evacuation centres. Hence, the facility provider should always aim to achieve high satisfaction among flood evacuees in future flood occurrence.

Do foreign direct investments (FDI) and trade openness lead to sustainable development in Malaysia? This is a very intriguing question posed by Ridzuan, Ismail and Hamat (2018). It is commonly known that FDI inflow is associated with the abundance of natural resources in a country (Poelhekke & van der Ploeg, 2013). Therefore, to maintain the balance between economic growth and environmental consideration in all parts of planning and management is imperative for a country’s sustainability. Their study uses annual data from 1970 to 2013 to provide evidence that foreign direct investments do lead to sustainable development via economic growth, income distribution and environmental quality. However, they find no evidence for similar influence by trade openness. They further suggest that Malaysian policymakers revise the existing policies to attract more sustainable foreign direct investment besides making trade more open by involving more free trade agreements in order to achieve the country’s sustainable development goals.

The employment and utilisation of resources must be effective and efficient to achieve business stability and sustainability.

Shariff et al. (2018) study an emerging sustainability issue in Malaysia, i.e., the migration of Malaysian intellectuals to work abroad; that potentially results in the nation’s brain drain. In addition, existing policies and strategies introduced by Malaysian government were inadequate to address the issue (National Economic Advisory Council, 2010). Based on survey, they find a significant positive relationship between higher expectation, perceived opportunities and lack of security in the intention of Malaysian intellectuals to settle abroad for work. Hence, economic policies such as the New Economic Model are essential in securing high quality human resource by offering sustainable business environment and income.

The eighth paper of this issue also deals with sustainability, where the authors, Wahid et al. (2018) examine the potential effects of energy consumption and economic growth in driving climate change pressure. This is alarming because the energy used in economic development has led to an upsurge in CO2 emissions in Malaysia (Begum et al., 2015). The findings reveal the presence of a long-run relationship among carbon dioxide emissions, consumption of energy resources and economic growth in Malaysia and Indonesia between 1980 and 2011. Their study implies that since emission through economic growth cannot be avoided, more efforts should be done to find, use and promote renewable energy. Minimising the negative impact of economy growth on emission level through the use of renewable energy is an agenda important to the sustainable development of both business and society.

Next, Kamaludin et al. (2018) present the results of their survey on consumers’ willingness to pay for an improved domestic water services in the state of Terengganu, Malaysia. Given that access to water is a basic human need, setting up a high water tariff would not be well accepted by society as the underprivileged would be denied of their basic need if they are not able to pay for it. However, by providing low water tariff, service providers

iv

are challenged with the inability to provide and recoup the cost of doing business involving the technological advances required to provide clean and continuous water supply. The survey discover that reducing the frequency of water interruption is the most important water service attribute to the population. They assert that the water services sustainability is essential in providing the population with safe and clean water. The desired level of sustainability in water services can be achieved when frequency of water interruption is minimized.

Lastly, the paper by Sulub, Salleh and Hashim (2018) looks at the validity of the hypothetical expectation of higher Corporate Governance Disclosure and Corporate Social Responsibility Disclosure in Islamic banks. Such disclosures are deemed important as they play a significant role in improving levels of corporate transparency, corporate image and firm sustainability (O’Dwyer & Gray, 1998). The authors find that while comparatively higher transparency levels of Corporate Governance Disclosure and Corporate Social Responsibility Disclosure are ideally predicted for Islamic banks, the disclosure levels were surprisingly lower in the annual reports of Islamic banks. In responding to this awkward finding, the authors recommend some possible remedies in order to reform Corporate Governance Disclosure and Corporate Social Responsibility Disclosure in Islamic banks. Due to the seemingly low disclosure level among the Islamic banks in Malaysia, this may hinder Islamic banks to be competitive and sustainable in the economic development.

Conclusion

After the three earlier ICABEC conferences, the organising committee had successfully selected relevant articles on organisational sustainability that are worth shared in printed publication. We believe that the readers of the Journal of Sustainability Science and Management would find the findings and suggestions presented by these ten articles useful.

In the spirit of these studies, achieving sustainable business agenda is a way to contribute towards sustainable service to society. The findings and suggestions presented in these articles can add new varying perspectives of sustainability as compared to the commonly published articles in the journal.

Four articles on livelihood sustainability reveal interesting findings on healthy food consumption, flood evacuation centre service quality, CO2 emission in the region and willingness to pay for water service. Two articles related to maritime environment in Malaysia describe the current situation at the Tok Bali fishing port and the direct feeder service in Port Klang. Next, two articles on the economic sustainability show the importance of foreign direct investment and not suffering from brain drain in the country. Finally, this special issue also presents two articles on the corruption and low transparency and why it is imperative to combat these issues to achieve efficient and effective utilisation of organisational resources.

We find that all of the included articles are important in the sustainability management of the country, especially when transparency and management integrity in organisations are upheld high. Through practising transparency and integrity, organisations can further prosper, while the nation can enjoy lasting stability and success. Nevertheless, this certainly requires actions, not merely reported for forming positive image among the public. Otherwise, such noble concept remains as a management fad rather than solution.

We hope that the information presented here will be useful to researchers, academicians and related stakeholders interested in the important role of transparency and management integrity in establishing long-term organisational sustainability. This should lead to the development of more effective policies, rules and regulations for improving the welfare of the existing public and future generations.

v

Acknowledgements

We thank the School of Maritime Business and Management and Centre of Transformation, Strategic Planning and Risk of Universiti Malaysia Terengganu for financial support.

References

Abdullah, W. M. T. W., Ahmad, N. N., & Ariff, A. M. (2018). Combating corruption for sustainable development of public services in Malaysia: Governance and corruption risk assessment. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Akasah, Z. A., & Doraisamy, S. V. (2015). 2014 Malaysia flood: Impacts and factors contributing towards the restoration of damages. Journal of Scientific Research and Development, 2(14): 53-59.

Begum, R. A., Sohag, K., Abdullah S. M. S., & Jaafar, M. (2015). CO2 emissions, energy consumption, economic and population growth in Malaysia. Renewable and Sustainable Energy Reviews, 41: 594-601.

Hsu, C. I., & Hsieh, Y. P. (2007). Routing, ship size, and sailing frequency decision-making for a maritime hub-and-spoke container network. Journal of Mathematical and Computer Modelling, 45(7): 899-916.

Kamaludin, M., Aziz, A., Ibrahim, N. S. C., & Radam, A. (2018). A survey on the willingness to pay for water service attributes in Terengganu, Malaysia. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

National Economic Advisory Council (2010). New Economic Model for Malaysia. Part 1: Strategic Policy Directions. Retrieved from http://www.jcci.or.jp/NEM%20for%20Malaysia%20-%20Part%20I_0.pdf, 5 January 2018.

O’Dwyer, B., & Gray, R. (1998). Corporate social reporting in the Republic of Ireland: A longitudinal study. Irish Accounting Review, 5(2): 1-34.

Padlee, S. F., Razali, N. N. H. N., Zulkiffli, S. N. A., & Hussin, Z. F. (2018). An assessment of the perception and satisfaction with flood evacuation centre service quality in East Coast states of peninsula Malaysia. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Poelhekke, S., & van der Ploeg, F. (2013). Do natural resources attract non-resource FDI? Review of Economics and Statistics, 95(3): 1047-1065.

Poerting, P., & Vahlenkamp, W. (1998). Internal strategies against corruption: Guidelines for preventing and combating corruption in police authorities. Crime, Law and Social Change, 29(2): 225-249.

Rahmatdin, M. N., Rahman, N. S. F. A., & Othman, M. K. (2018). Decision making approach of direct feeder service via Port Klang in Malaysia. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Ridzuan, A. R., Ismail, N. A., & Hamat, A. F. C. (2018). Foreign direct investment and trade openness: Do they lead to sustainable development in Malaysia?. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Salleh, H. S., & Noor, N. A. M. (2018). Moderating effect of self-efficacy on the relationship between consumer attitude and willingness to consume healthy food. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Salleh, N. H. M., & Halim, M. A. A. (2018). Enhancing environmental sustainability over fisheries industry

vi

through proactive risk evaluation: A case of Tok Bali fishing port. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Shariff, N., Krishnaswamy, J., Abdullah, D., & Chau, A. W. (2018). The influence of new economic model on intention to work abroad: An empirical study to address brain drain issue in Malaysia. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Sulub, S. A., Salleh, Z., & Hashim, H. A. (2018). A review of corporate governance and corporate social responsibility disclosure of Islamic banks: A call of

additional transparency. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

Wahid, I. N., Aziz, A. A., Kamaludin, M., & Anang, Z. (2018). The influence of energy consumption, renewable energy and economic growth on CO2 emission in Malaysia and Indonesia. Journal of Sustainable Science and Management, Special Issue 4: xx-xx.

World Health Organisation (2015). Non-communicable diseases. Fact Sheet 355. Retrieved from http://www.who.int/mediacentre/factsheets/fs355/en/, 5 January 2018.

Journal of Sustainability Science and Management Special Issue No. 4 2018The Role of Transparency and Management Integrity in Organisational Sustainability

ISSN: 1823-8556© Penerbit UMT

COMBATING CORRUPTION FOR SUSTAINABLE PUBLIC SERVICES IN MALAYSIA: SMART GOVERNANCE MATRIX AND CORRUPTION RISK ASSESSMENT

WAN MOHAMMAD TAUFIK WAN ABDULLAH1, NURUL NADIAH AHMAD1 AND AKMALIA MOHAMAD ARIFF2*

1College of Business Management and Accounting, Universiti Tenaga Nasional, Sultan Haji Ahmad Shah Campus, 26700 Bandar Muadzam Shah, Pahang, Malaysia

2School of Maritime Business and Management, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia


Abstract: In a developing country such as Malaysia, combating corruption is needed to ensure its sustainable social and economic development. While numerous initiatives are undertaken to increase integrity in Malaysia, corruption issues in public services are still alarming. Corruption involving public institutions disrupts service efficiency, hampers confidence in public institutions, and increases the cost of public transactions. This study examines the link between the quality of governance environment and the ability to assess corruption risk. The quality of governance environment is proxied by five governance initiatives (Smart Governance Matrix) while the ability to predict corruption is represented by corruption risk assessment that consists of five indicators. Thirty-nine responded questionnaires, gathered from Integrity Officers in public institutions in Malaysia, are used in the analysis. We find that enterprise governance model, data governance model and human governance model are positively associated with the corruption risk assessment. Good governance environment, which is represented by the Smart Governance Matrix, increases the ability to assess corruption risks. The findings provide insight into the need to improve the governance environment in public services in Malaysia as it can work towards controlling corruption activities through its ability to comprehensively assess corruption risk.

Keywords: Corruption risk assessment, integrity, public services, sustainable governance.

Introduction

This paper aims at providing input into the role of governance in combating corruption by focusing on public services in Malaysia. For a developing country such as Malaysia, tackling corruption is an essential role of the government who has major responsibilities in ensuring public sector efficiency through development of policies such as tax policy and development of economic regulation. In public services, corruption can be seen in cases involving the quality of service delivery system, public fund embezzlement, bribery in issuance of tender, favoritism and fraud in election process. ‘Grand corruption’ by high-level officials, which includes robbery of public budgets and over-investment in capital-intensive projects,

generates poverty and retards economic growth (Rose-Ackerman, 1999). Low-level corruption by civil servants, which include bribery to reduce bureaucratic delay or lower the cost of government taxes and regulations, undermines the legitimacy of the state and spreads the culture of corruption (Rose-Ackerman, 1999). Accordingly, corruption affects sustainable economic development, such as causing threat on macroeconomic stability and foreign direct investment, and social development, such as affecting the environment and causing the poor to suffer (Mauro, 1997).

In the context of Malaysia, there have been many initiatives to fighting corruption being implemented, especially under the National Key Results Areas (NKRA) initiatives. These

: 1-14

2COMBATING CORRUPTION FOR SUSTAINABLE PUBLIC SERVICES IN MALAYSIA:SMART GOVERNANCE MATRIX AND CORRUPTION RISK ASSESSMENT

Journal of Sustainability Science and Management Special Issue No. 4 2018The Role of Transparency and Management Integrity in Organisational Sustainability : 1-14

include the establishment of Integrity Units in all federal and state government agencies with a Certified Integrity Officer (CeIO) as the unit head and the establishment of Committee on Integrity Governance (CIG) in all government ministries and departments. Nevertheless, the corruption issues in Malaysia are still alarming perhaps due to the failure in implementing and enforcing the anti-corruption legislations. The 2014 Malaysian Corruption Barometer Results highlights that the perception on the effectiveness of government’s anti-corruption efforts has declined and the incidences of bribery in public services, especially in the education and medical/health services, have escalated with the main reason given for bribery is to ‘speed things up’ (Transparency International – Malaysia, 2014). The annually published Auditor-General Report reports many instances where projects, such as the Sultanah Nora Ismail Hospital upgrading project, are delayed and faced with many non-compliance issues. These evidences, while not conclusive, signify that corruption in Malaysian public services is a matter of concern.

An aspect that is important in tackling corruption is the quality of governance environment. The International Framework: Good Governance in the Public Sector by IFAC (2014) elaborates the fact that good governance in the public sector is achieved when the governing bodies and the players work towards achieving the organisation’s goals while adhering to the public interest; in line with the laws and policies of the government. Serving as a benchmark for good governance practices for public services, the framework aims to improve the delivery of services and accountability in public service organizations. However, there is no ‘one-size fits all’ governance framework, as there are different objectives and governance structures surrounding public service organizations. Further, prior studies in Malaysian corporate governance (Ariff et al., 2007; Hashim & Devi, 2008) have mainly focused on the private sectors and undermined the importance of good governance in the public sector. Thus, it is crucial to examine the governance framework that can work towards the development of

greater ethical conduct in the context of public service organizations in an emerging country such as Malaysia. Also, to add to the existing evidence that has mainly focused on individual governance measure (Shim & Eom, 2008), this study uses a holistic governance measure i.e. Smart Governance Matrix (SGM). Since corporate governance is a multi-dimensional phenomenon, a composite measure that captures multi-dimensions of corporate governance would be a better research approach and can add to the existing evidence involving the effectiveness of corporate governance. To our knowledge, this study is the first to employ the SGM in measuring the governance environment of public institutions.

In the context of corruption, where prevention is better than cure, it is important to tackle the issue at the earlier stage possible. Corruption risk assessment, therefore, is a prominent agenda for Malaysia. Corruption risk assessment involving public institutions measures the ability of public institutions to identify risks in their system, to act towards preventing corruption within their capacity, and the role of ministers and the cooperations within sectors (OECD Anti-Corruption Network for Eastern Europe and Central Asia, 2015). However, to date, there is no specific tool to assess corruption risk in public service organizations Malaysia. Our discussion with a government agency in Malaysia signifies the fact that the agency currently measures corruption risk in Malaysia based on the number of investigation papers issued, which is only when the corrupt practices are highly probable with very significant consequences and subsequent impact. As the input from the risk assessment involving corruptions is beneficial in order to propose and develop follow-up actions to manage the risks, this paper fills the gap in the literature by being the first to employ Corruption Risk Assessment (CRA) in measuring the ability to assess corruption risks in an organization. As compared to corruption risk that is measured based on perception of respondents, such as those of the Transparency International, the CRA in this study measures

3Wan Mohammad Taufik Wan Abdullah1 et al.

the corruption risk assessment from the actual incidences of corruption from the perspective of the regulators, mainly the related government agencies. The CRA has the potential to be used by public and private organizations - if the tool is adopted by Malaysian regulators in assessing corruption risks.

In this study, we investigate the relation involving the quality of governance environment and the corruption risk assessment. The aim is to test the role of governance environment in controlling corruption activities through its ability to predict corruption risk in public services. In the context of the literature on corruption risk assessment, prior studies are mainly concentrated on the methodological and systematic approach to measure corruption risk and the use of corruption risk assessment measure in analysis is uncommon. Accordingly, the posibility that the quality of governance environment can enhance the ability to predict corruption risk is an area that remains unexplored. In order to tackle the issue, we adopt Holmes’s (1997) three antecedents of corruption: institutional, economics, and cultural and historical factors, in setting the hypotheses below.

Institutional Elements

One of the measure of institutional elements is the monopoly power of the Board and the discretion level of their practice (Ali & Isse, 2003) that escalate the risk of corruption in the organization. Unethical attitude by the members of the board may trigger the same attitude from the subordinates. Therefore, it is vital for the quality of the Enterprise Governance Model (EGM) to be enhanced so that the Board members are committed towards achieving the objectives of the organization (Lees, 2007). In this context, high quality EGM can work to assist the organization towards focusing on drivers of value creation while at the same time ensure that effective control and monitoring are intact. High quality EGM is expected to enhance the ability to assess corruption risk. Thus, the first hypothesis is developed as follows:

Hypothesis 1 : There is a positive relationship between the quality of EGM and corruption risk assessment.

Further, the rules of the game in a society shape the institutional elements that empower corruption. This context is explained as the humanly-devised constraints that determine the interaction between humans (Knack & Keefer, 1995). To cater to this issue, improving the quality of data governance is essential in keeping track of the organizations’ formal rules and information as it is valuable in detecting potential risks related to corruption mismanagement (Power, 2013). The necessity for high quality data calls for a need to have a high quality Data Governance Model (DGM). The use of a system that can track and improve enterprise data will enhance the ability to assess corruption risk. Hence, the second hypothesis is hypothesized as follows:

Hypothesis 2 : There is a positive relationship between the quality of DGM and corruption risk assessment.

Another aspect of institutional elements that gives rise to corruption comes from ‘corruption of greed’ surrounding government servants with senior positions. While they have sufficient income, they desire for more because accumulating wealth is perceived as a normal activity of those in their position. Along with their power in influencing their working environment, these staff have the ability to overcome and bypass the computerized information systems that manage the significant data and information unless there are involvements from strong external agency (Wu, 2005). For this particular objective, management of IT requires a comprehensive control of corruption to monitor these issues. Hence, high quality IT Governance Model (ITGM) is required to support critical business functions and processes by providing prompt and safe feedback for effective management of complex technology. The existence of high quality ITGM is expected to enhance the ability to assess corruption risk. The third hypothesis is therefore developed as follows:



Hypothesis 3 : There is a positive relationship between the quality of ITGM and corruption risk assessment.

Economic Elements

Size of the organization is found to be as one of the economic elements that trigger corruption (Dreher et al., 2007). In an organization with significant size, the officials are able to apply a higher degree of discretion in their review and supervisory role, and hence increase the risk of corruption (Tanzi, 1998). Therefore, it is crucial for the board of directors to strictly perform their governance role in relation to overseeing the transparency and stability needed to increase the organization’s value creation and growth. High quality Corporate Governance Model (CGM), which revolves around the responsibilities of the board of directors in strategizing policies and guidelines towards achieving a company’s goals, is expected to increase the ability to assess corporate risk. Accordingly, the fourth hypothesis is developed as follows:

Hypothesis 4 : There is a positive relationship between the quality of CGM and the corruption risk assessment.

Cultural Elements

The cultural elements that consider psychological influences of individuals are also considered to be as one of the determinants of corruption. In the context of internal factors, one has to consider the fact that there are individuals who are “naturally evil” and willing to commit bribery (Voskanyan, 2000). In the context of external elements, factors such as the individual’s relationship to a group are vital (Voskanyan, 2000). In short, human governance is an important thing to focus in the establishment of moral culture that incorporates integrity within the human capital (Dreher et al., 2007). Accordingly, Human Governance Model (HGM), that brings back the initial intention that should be upheld during decision-making, is required to combat corruption. High quality HGM is expected to benefit in terms of

enhancing the ability to predict corruption risk. Hence, the fifth hypothesis is as follows:

Hypothesis 5: There is a positive relationship between the quality of HGM and the corruption risk assessment.

Methodology

A threshold model of collective behaviour, which employs the perspective of treshold effect, is used as the theoretical underpinning in the development of the hypotheses. Granovetter and Soong (1988) conclude that the threshold distribution predicts the results of the aggregate behaviour. Aidt et al. (2008) affirm that the causes of corruption can be implied as the antecedents of corruption risk. Accordingly, Holmes (1997) believes that there are three antecedents of corruption: economics, institutional and cultural and historical factors. Figure 1 depicts the theoretical framework of the study. For this study, the threshold model prescribes the determinants of corruption as the threshold distribution and the corruption risk assessment as the outcome of the aggregate behaviour. In the framework, quality of governance is represented by the present and future requirements of the organization in prioritizing the comfort of stakeholders. Numerous good governance indicators must be considered to ensure a holistic governance model is used. Thus, SGM outlines five governance models. The framework also depicts that the SGM predicts the ability to asssess corruption risks, which is presented by the CRA.



To develop the independent variables, SGM, this study puts together five governance models which all refer to the environment that is required for good governance in an organization. SGM is used in this study as it is a holistic governance initiative comprising all stages of management: from planning, implementing and evaluating. SGM represents the governance environment of an organization that allows one to measure the quality of governance from the perspective of enterprise, corporate, IT, data, and human governance. These five governance models are the outcome of the Confirmatory Factor Analysis (CFA) on all governance variables that are derived from focus group discussions with a government agency in Malaysia. The first model is Enterprise Governance Model (EGM) that refers to an integrated model to assist entities in focusing on drivers of value creation and ensuring effective control and monitoring. EGM provides enterprise governance capabilities to create value for shareholders through the effectiveness of the enterprises’ operations (Lees, 2007). The second model is Corporate Governance Model (CGM). CGM focuses on the transparency principles and stability required for the increment of growth and value creation. Board of Directors is responsible in strategizing for policies and guidelines towards achieving a company’s goals. The third model is IT Governance Model (ITGM) that provides

DEPENDENT VARIABLE

Corruption Risk Assessment

INDEPENDENT VARIABLES (SGM)

Institutional Elements -Enterprise Governance Model (EGM)

-Data Governance Model (DGM) -IT Governance Model (ITGM)

Economics Elements

-Corporate Governance Model (CGM)

Cultural Elements -Human Governance Model

(HGM)INDEPENDENT VARIABLES (SGM)

Figure 1: The link between quality of governance and corruption risk assessment

support for important business functions and processes through the creation of competitive advantage for effective management by using technology. It allows for a prompt and safe responses to business requirements. Data Governance Model (DGM), which is the fourth model, functions to improve data quality through the use of system for tracking and improving enterprise data (Loshin, 2013). The last model is Human Governance Model (HGM). According to Salleh and Ahmad (2010), HGM incorporates human well-being and interest in the process of making decision, by considering the initial intention that should be upheld during the decision-making.

The dependent variable, CRA, is developed from five indicators; policy, integrity, accountability, transparency, and monitoring. These five indicators are the outcome of the Confirmatory Factor Analysis (CFA) on all corruption risk assessment variables that are derived from focus group discussion with a government agency in Malaysia. Firstly, the formulation of policy is expected to decrease the corruption risks in organization. Secondly, the assessment of integrity assists management with feedback on the whether their efforts in combating corruption are effective or otherwise, and accordingly provides justification for adjustment to be done systematically (Public



Governance Committee, 2004). Thirdly, accountability is important as it refers to the responsibilities of each public officer such as in the formation, implementation, and enforcement of the law (Lanyi & Azfar, 2005). Fourthly, transparency requires information dissemination to be accurate, timely, and useful (GOPAC, 2005). The fifth indicator is monitoring which is a tool to keep an eye on corruption. Olken (2007) emphasizes that involvement of every member of the organization supplements important monitoring information that allows the check and balance in efforts to control corruption. This eventually provides better ability to monitor risk towards the organization.

The data on quality of governance and corruption risk assessment in Malaysian public services is derived from survey questionnaire that has 4 sections. A 5-point Likert scale is used to measure the response. The questionnaires were distributed to 43 respondents via email while 21 respondents were given the questionnaires by hand. There were only 64 respondents to which this study has access to in accordance with the information provided by the MACC as at 31 December 2014. The data was collected from the respondents on a continuous basis from January to March 2015. The determination of the means of distribution was based on the availability of email address and also upon the request from the respondents who preferred to receive the survey by hand. The respondents were the Chief Integrity Officer (CeIO)/Integrity Officer who are appointed by the MACC and attached to government agencies, federal departments, statutory bodies, and government-linked companies in Malaysia as well as Integrity Officer who are appointed by the respective organizations themselves. The Integrity Officer refers to accredited officers that are appointed to inculcate integrity and ethical values within the organization. They hold an important role in reducing the risk of corruption in organization. For example, they have the capability to develop integrity action plan as they are well-trained in the aspects of integrity management, legal framework, compliance and monitoring system.

To test the hypotheses of the study, statistical analyses were performed based on structural equation modelling (SEM) by utilizing the Partial Least Squares (PLS) method. PLS is suitable to be employed as it can be used to perform data analysis for both univariate and multivariate outlier especially when the sample size is small (Hair et al., 2010). PLS discloses the variables relationship in both measurement and structural model. This study follows the approach derived from Anderson and Gerbing (1988) to estimate the reliability and validity of the measurement model prior to testing the structural model. First, it assesses the measurement model for convergent validity. The assessment was made using factor loadings, composite reliability (CR), and average variance extracted (AVE) in accordance with Hair et al. (2006). Second, this study examines the discriminant validity to identify ‘the extent to which the measures are not a reflection of some other variables’. Discriminant validity is shown by the low correlations between the measure of interest and the measures of other constructs (Ali et al., 2015; Ramayah et al., 2013). In relation to the structural model, this study applies bootstrapping technique using 5,000 iterations to assess the statistical significance of the weights of sub-constructs and the path coefficients (Chin, 1998).

Results

Demographic Profile

From 64 questionnaires distributed, 60% of the questionnaires were suscessfully collected. The small number of sample is used due to the limited access to the information involving Integrity Officers. The demographic profile of the 39 respondents can be examined in Table 1 with the majority of the respondents, 46% are from federal departments. 64% of the respondents are the Chief Integrity Officer (CeIO) in the organization with majority of them, 36% have been in the position of the CeIO for less than 1 year. There are 43.6% respondents who have served the organization for less than 10 years.



Measurement Model

Table 2 shows that, with the exception of FIVE items that were finally deleted, all the other item loadings exceeded the recommended value of 0.6 (Chin, 1998). The deletion of the 5 exceptional items does not affect the reliablity of the factor. Composite reliability values, which represent the

Table 1 : Demographic profile

Items Frequency Percent Valid Percent

Cumulative Percent

Organization structure Ministries 15 38.5 38.5 38.5

Federal Departments 18 46.2 46.2 84.6

Government Agencies 2 5.1 5.1 89.7

Government-Linked Companies

(GLC)4 10.3 10.3 100.0

Are you the Chief Integrity Officer (CeIO)

/ Integrity Officer appointed by SPRM?

Yes 25 64.1 64.1 64.1

No 14 35.9 35.9 100.0

The number of years appointed as Chief

Integrity Officer (CeIO) / Integrity Officer in

this organization

1 year and below 14 35.9 35.9 35.9

1 – 2 years 7 17.9 17.9 53.8

2 - 3 years 2 5.1 5.1 59.0

More than 3 years 2 5.1 5.1 64.1

Not CeIO 14 35.9 35.9 100.0

degree to which the construct indicators indicate the latent construct, exceeded the recommended value of 0.7 (Hair et al., 2006). AVE, which identifies the overall amount of variance in the indicators accounted for by the latent construct, exceeded the recommended value of 0.5 (Hair et al., 2006).

The number of years served in this

organization

10 years and below 17 43.6 43.6 43.6

11- 20 years 15 38.5 38.5 82.1

21 - 30 years 6 15.4 15.4 97.4

31 years and above 1 2.6 2.6 100.0

Table 2 : Validity and reliability for construct

Variables Outer Loadings Cronbach Alpha Composite Reliability AVE

CG1 0.75 0.88 0.91 0.62

CG2 0.84



CG3 0.88

CG4 0.74

CG5 0.73

CG6 0.78

CR1 0.49 0.94 0.94 0.51

CR10 0.82

CR11 0.64

CR12 0.51

CR13 0.74

CR14 0.76

CR15 0.68

CR16 0.71

CR17 0.49

CR18 0.64

CR19 0.60

CR2 0.51

CR20 0.63

CR3 0.51

CR4 0.79

CR5 0.82

CR6 0.79

CR7 0.71

CR8 0.79

CR9 0.81

DG1 0.84 0.93 0.95 0.79

DG2 0.92

DG3 0.86

DG4 0.95

DG5 0.85

EG1 0.89 0.93 0.94 0.77

EG2 0.93

EG3 0.93

EG4 0.85

EG5 0.79

HG1 0.60 0.83 0.88 0.55

HG2 0.73



HG3 0.55

HG4 0.85

HG5 0.80

HG6 0.85

IT1 0.77 0.91 0.93 0.73

IT2 0.79

IT3 0.92

IT4 0.92

IT5 0.86

Table 3: Discriminant validity

Constructs CGM CRA DGM EGM HGM ITGM

CGM 0.787

CRA 0.589 0.714

DGM 0.526 0.667 0.888

EGM 0.772 0.692 0.620 0.877

HGM 0.418 0.644 0.610 0.479 0.741

ITGM 0.586 0.645 0.688 0.681 0.578 0.854

Note : CG : Corporate Governance; CR : Corruption Risk; DG : Data Governance; EG : Enterprise Governance; HG : Human Governance; and IT : IT Governance.

As shown in Table 3, the square root of the AVE (diagonal values) of each construct is larger than its corresponding correlation coefficients. As depicted by Fornell and Larcker (1981), these results indicate adequate discriminant validity.

Note : Corporate Governance Model (CGM), Corruption Risk Assessment (CRA), Data Governance Model (DGM), Enterprise Governance Model (EGM), Human Governance Model (HGM), and IT Governance Model (ITGM).

Overall, the measurement model shows an adequate convergent validity and discriminant validity.



* p < 0.05 (T-value > 1.645)

Figure 2: Results of the structural model

Table 4 : Structural estimates (Hypotheses testing)

Hypotheses Beta Standard Error t-statistics Decision

H1: EGM > CRA 0.375 0.171 2.197 Supported

H2: DGM > CRA 0.215 0.153 1.704 Supported

H3: ITGM > CRA 0.099 0.201 0.490 Not Supported

H4: CGM > CRA 0.062 0.146 0.423 Not Supported

H5: HGM > CRA 0.234 0.097 2.418 Supported

Note : Corruption Risk Assessment (CRA), Enterprise Governance Model (EGM), Data Governance Model (DGM), IT Governance Model (ITGM), Corporate Governance Model (CGM), and Human Governance Model (HGM).

Table 4 displays the full results related to the structural model and hypotheses testing. The results indicate a strong support for three out of the five hypotheses. More specifically, DGM, EGM and HGM are shown as having positive

Discussions

In this study, the issue of corruption that affects the sustainable social and economic development of a country is addressed by examining the relation between the quality of

relationship with corruption risk assessment (CRA). On the other hand, there is no significant associations shown for results involving CGM and ITGM and CRA.

governance environment (measured by five governance initiatives ie. SGM) and the ability to assess corruption risk (measured by CRA). The hypothesized relationship between enterprise governance and corruption risk assessment was significant and positive. This result explained



that enterprise governance is able to asssess corruption risk. According to Ali and Isse (2002), the behaviors of board members are able to influence the behavior of subordinates. If management puts more commitment towards company’s mission and vision, the ability to assess corruption risk will be higher. Therefore, hypothesis 1 is supported. There exists a significant positive relationship between data governance and corruption risk assessment. Good data governance is able to provide value on the formal rules and important information to detect the likelihood of any corruption and manipulation being done (Power, 2013). Shim and Eom (2008) support that data governance can reduce corrupt behaviors externally, through the ability to enhance relationships with the public, as well as internally, through more effective controlling and monitoring towards employees’ behaviors. Therefore, hypothesis 2 is supported.

The hypothesised relationship between human governance and corruption risk assessment is significant and positive. The results indicate that good human governance is able to enhance the corruption risk assessment. This finding is in line with the argument from Dreher et al. (2007) that human governance is able to develop moral culture in achieving all the goal, mission and mission. Salleh and Ahmad (2010) support that human governance could bring positive effect to the profession since human governance will be a part of the core guiding principles for professional conduct. Therefore, hypothesis 5 is supported. The other hypotheses were rejected may be due to the fact that corporate governance and IT governance structure only focuses on the processes, which may be important as the back stop, but may not be able to directly incorporate integrity in the organization. Therefore, hypothesis 3 and 4 are not supported.

Overall, our findings are in support of the view that good governance environment increases the likelihood of being able to assess corruption risk. Three SGM elements are found

to be related to corruption risk assessment: enterprise, data, and human governance. To enhance the role of governance in combating corruption, focus should be placed on improving governance in the context of enterprise, data and human in Malaysian public sector organizations. This is because good governance increases effectiveness of corruption risk assessment, thus allowing preventive and proactive mechanisms to be adopted to combat corruption accordingly.

Conclusions and Recommendations

This study finds that enterprise governance model (EGM), data governance model (DGM) and human governance model (HGM) are associated with the corruption risk assessment. The Smart Governance Matrix, which represents governance environment, predicts the ability of assessing corruption risks. The findings contribute towards the improvement of governance environment in public services in Malaysia as a tool in detecting red flags of corruption in the organizations through its ability to comprehensively assess corruption risk. In terms of EGM, our results indicate that the ‘tone from the top’ matters in combating corruption. A superior who is well-versed with the activities within the organizations would be able to delegate and monitor the assigned tasks efficiently, and therefore would easily identify irregularities occuring within the organizations that are related to corruption accordingly. In terms of DGM, our results emphasize on the importance of having high quality data as it ensures efficient and effective resource allocation. For example, the availability of high quality data on companies applying for logging license would allow the government to accurately evaluate them before the licenses can be awarded, so as to ensure that logging activities are performed in a manner that do not negatively affect the environment and livelihood of local communities. In terms of HGM, our results indicate that the core factor to combat corruption relies on the human capital i.e. the public servants themselves. A person with high integrity and ethical values would be able to



ensure policies and regulations are adhered to, and perform their tasks effectively, such as in monitoring the progress of government projects undertaken by suppliers and vendors. Strict monitoring would enable government projects to be completed in time, or for any discrepancies with the suppliers and vendors to be rectified, so that the projects serve their purpose in providing better quality of life to the public.

Nevertheless, the findings of our study are limited to the fact that the sample used is only Certified Integrity Officers (CeIOs) that may result in bias assessment of the findings. Future researchers can therefore extend the study by collecting feedback from other officials who are directly involved with the suppliers, vendors and customers i.e. the public. In the context of suppliers and vendors, for example, while the public procurement system is available in the Malaysian Government system of administration, there are weaknesses in the public procurement processes that may open avenues for corruption to take place (Othman et al., 2010). Besides, this study covers five models of governance in developing the Smart Governance Matrix. Future researchers can also explore other governance environment including enterprise architecture governance model and multi-stakeholder governance as the corruption control efforts. Also, the corruption risk assessment has a great potential to be adopted as a tool to assess corruption risk in both public and private organizations in Malaysia. Further input from Malaysian regulators shall be sought out to incorporate further refinement in the corruption risk assessment tool.

In terms of policy implications, findings of this study provide input to various efforts in the development of policy and enforcement mechanisms towards combating corruption in Malaysia. This is because, aside from having the right rules, controling corruption at the earlier stage possible is necesssary for the purpose of identifying the sources of informality that have caused divergence between the policies and the regulations as well as their effectiveness. The findings of this study calls for refinement in the

governance environment in public services in Malaysia as high quality governance enhances the ability to assess corruption risk so as to prevent the ‘cancer of corruption’. Special focus could be targeted to improve data governance within public services because accurate and reliable data increases the likelihood to predict corruption risk. In this aspect, data sharing between public sector organizations, that does not compromise confidentiality and control, should be implemented to enable an effective corruption risk assessment. On top of that, special emphasize is needed to improve the enterprise and human governance within the public services. These two governance initiatives are rather related as both involve human beings and their interactions. Albeit challenging, it is not impossible that combatting corruption can be enhanced by improving integrity and ethical values of the core players in the organizations i.e. the top management and the public servants.

Acknowledgements

We thank the Ministry of Education (MOE) for the Fundamental Research Grant Scheme (FRGS/1/2016/SS01/UNITEN/02/1) fund awarded. The assistance given by the Malaysian Anti-Corruption Commission (MACC) during focus group discussion session, to scrutinize issues with regard to the study, is highly appreciated.

References

Aidt, T., Dutta, J., & Sena, V. (2008). Governance regimes, corruption and growth: Theor and evidence. Journal of Comparative Economics, 36(2): 195-220.

Ali, A., Rahman, M. S. A., & Bakar, A. (2015). Financial satisfaction and the influence of financial literacy in Malaysia. Social Indicators Research, 120(1): 137-156.

Ali, M. A., & Isse, H. S. (2003). Determinants of economic corruption: A cross-country comparison. Cato Journal, 22(3): 449-466.



Anderson, J., & Gerbing, D. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3): 411-423.

Ariff, A. M., Ibrahim, M. K., & Othman, R. (2007). Determinants of firm level governance: Malaysian evidence. Corporate Governance: The International Journal of Business in Society, 7(5): 562-573.

Chin, W.W. (1998). Issues and opinions on structural equation modelling. MIS Quarterly, 22(1): 7-16.

Dreher, A., Kotsogiannis, C., & McCorriston, S. (2007). Corruption around the world: Evidence from a structural model. Journal of Comparative Economics, 35(3): 443-466.

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1): 39-50.

GOPAC (2005). Controlling corruption: A parliamentarian’s handbook. Ottawa, Canada: GOPAC.

Granovetter, M., & Soong, R. (1988). Threshold models of diversity: Chinese restaurants, residential segregation, and the spiral of silence. Sociological Methodology, 18: 69-104.

Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis, 6th edition. Prentice Hall, Upper Saddle River, New Jersey. 899 pp.

Hashim, H. A., & Devi, S. S. (2008), Board independence, CEO duality and accrual management: Malaysian evidence. Asian Journal of Business and Accounting, 1(1): 27-46.

Holmes, L. (1997). Corruption and the crisis of the post-communist state. Crime, Law and Social Change, 27: 275.

IFAC (2014). International Framework: Good Governance in the Public Sector. New York, USA : IFAC.

Knack, S., & Keefer, P. (1995). Institutions and economic performance: Cross-country tests using alternative institutional measures. Economics & Politics, 7(3): 207-227.

Lanyi, A. & Azfar, O. (2005). Tools for assessing corruption & integrity in institutions: A Handbook. The IRIS Center, University Research Corporation International, College Park, Maryland. 155 pp.

Lees, G. (2007). Enterprise governance: Restoring boardroom leadership (Discussion Paper). London, United Kingdom: Chartered Institute of Management Accountants.

Loshin, D. (2013). Big data analytics: From strategic planning to enterprise integration with tools, techniques, NoSQL, and graph. Morgan Kauffman, Waltham, Massachusetts. 120 pp.

Mauro, P. (1997). Why worry about corruption? (Economic Issues No. 6). International Monetary Fund, Washington. 18 pp.

OECD Anti-Corruption Network for Eastern Europe and Central Asia (2015). Prevention of corruption in the public sector in Eastern Europe and Central Asia. Paris, France: The Organisation for Economic Co-operation and Development.

Olken, B. A. (2007). Monitoring corruption: Evidence from a field experiment in Indonesia. Journal of Political Economy, 115(2): 200-249.

Othman, R., Zakaria, H., Nordin, N., Shahidan, Z., & Jusoff, K. (2010). The Malaysian



public procurements prevalent system and its weaknesses. American Journal of Economics and Business Administration, 2(1): 6-11.

Power, D. (2013). All the ingredients for success: Data governance, data quality and master data management. Hingham, Massachusetts, USA : Hub Designs.

Public Governance Committee (2004). Measures for promoting integrity and preventing corruption: How to assess?. Paris, France: The Organisation for Economic Co-operation and Development.

Ramayah, T., Yeap, J. A. L., & Igatius, J. (2013). An empirical inquiry on knowledge sharing among academicians in higher learning institutions. Minerva: A Review of Science, Learning and Policy, 51(2): 131-154.

Rose-Ackerman, S. (1999). Corruption and Government: Causes, Consequences, and Reform. Cambridge University Press, New York. 618 pp.

Salleh, A., & Ahmad, A. (2010). Human governance: Bringing the meaning of integrity in the life of professional accountants. In Articles of Merit E-Book (pp. 30–39). New

York: International Federation of Accountants.

Shim, D. C., & Eom, T. H. (2008). E-government and anti-corruption: Empirical analysis of international data. International Journal of Public Administration, 31(3): 298-316.

Tanzi, V., (1998). Corruption, governmental activities, and markets (Working Paper No. 94/99). International Monetary Fund, Washington. 23 pp.

Transparency International – Malaysia. (2014). Malaysian Corruption Barometer Survey 2014. Retrieved from http://transparency.org.my/wp-content/uploads/2014/05/MCB-2014-Press-Statement.pdf, 2 July 2016.

Voskanyan, F. (2000). A study of the effects of corruption on economic and political development of Armenia. Thesis of Masters Degree. American University of Armenia.

Wu, X. (2005). Corporate governance and corruption: A Cross-Country Analysis.Governance: An International Journal of Policy, Administration and Institutions, 18(2): 151–170.




DECISION MAKING APPROACH OF DIRECT FEEDER SERVICE VIA PORT KLANG IN MALAYSIA

MUHAMAD NASIR RAHMATDIN1, NOORUL SHAIFUL FITRI ABDUL RAHMAN2* AND MOHAMMAD KHAIRUDDIN OTHMAN2

1School of Ocean Engineering, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia,



Abstract: The feeder service plays an important role in the Malaysian shipping service industry to link between the main line service and ports. It appears to be an important mode of transport within intra-Asian countries, regional, and local trades, which allows access to certain ports with ship draught restrictions. The feeder service is efficient in terms of flexibility, travelling time consumed between ports and fast cargo delivery to consignors. The issue now is how can Malaysian feeder operators truly sustain the future via trade and route assessment perspectives. Efficient route selection and strategic trading partners is the key to boosting the Malaysian feeder economy by considering the five main factors of markets, time, miscellaneous, service and cost. The purpose of this research is to discover a benchmark of current assessments of overall direct feeder service of route parameters which connect Port Klang and five potential ports or trade partners as alternatives. The ports selected are Ho Chi Minh City, Laem Chabang, Jakarta, Yangon, and Chittagong due to their consistent previous years of container import and export activities. A combination method, known as the AHP-Fuzzy Link Based and Evidential Reasoning method, is applied in this research. The results show that the Port Klang – Laem Chabang route is the most preferred sailing route that can optimize feeder vessel utilization, reduce environmental pollution, and potentially gain high returns to both feeder operators and ports. This research contributes in terms of providing indicator of routes and a potential port appraisal case study of the feeder service in Malaysia by using a decision making approach.

Keywords: Analytical hierarchy process, evidential reasoning method, containers, short sea shipping, feeder hub, sustainability of feeder service.

Introduction

According to Abdul Rahman et al. (2014), the opening of the Northern Sea Route (NSR) had a potential prospect to decrease the number of vessels passing through Malaysian waterways. In 2006, a fluctuation of container demand and supply, by means of 83% import and 80% of export in North America, Europe and Asia, was reported to be influenced by the NSR (Verny & Grigentin, 2009). Accordingly, it is important to find a solution to maintaining the Malaysian maritime economy in the future.

The feeder service sector has played an important role of excellent cargo transportation in Malaysia and has been identified as a future solution that needs to be strengthened. This is largely due to outstanding annual feeder calling and container throughput at Port Klang, which is also known as a Malaysian feeder hub port. Port Klang is ranked first in the list of Malaysian container ports and was positioned 12th in the world between 2012 and 2014 (UNCTAD, 2015). More than 70% of feeder calling in Malaysia originated from Port Klang in 2010 (10,751 ships calling), 2011 (11,273 ships

: 15-35

16DECISION MAKING APPROACH OF DIRECT FEEDER SERVICE VIA PORT KLANG IN MALAYSIA

calling), 2012 (10,300 ships calling), 2013 (9,950 ships calling), and 2014 (9,601 ships calling) (ASEAN Ports Association Malaysia,

Table 1 shows that the container volumes of import, export and transshipment in Port Klang grew from 2011 to 2015. Over five consecutive years, from 2011 to 2015, Port Klang operated regularly as a transshipment port, and 64.9% average of its total container throughputs were derived from transshipment activities. According to UNCTAD (2015), the total container throughputs for Port Klang increased by 3.48% from 2012 to 2013 and 5.76% from 2013 to 2014. Port Klang appears to be outstanding as a Malaysian feeder hub, with consistent annual container throughputs. Further, Port Klang was deemed excellent because it is a feeder hub for transshipment point containers. An estimated 60 to 75% of all container through Port Klang in 2015 and previous years were transshipment containers (Westport, 2016). It attracted a lot of feeder operators to allocate their services and company bases in Selangor; especially in Klang and Shah Alam areas. Moreover, Port Klang is efficient in terms of proficient cargo handling facilities interrelated with rapid container inbound and outbound services provided. The excellent transportation services provided by the feeder operators have helped Port Klang to be one of the most superior and competitive ports worldwide. Therefore, the question is how to maintain this situation in the future.

Feeder service route plays an important role influencing the feeder service sector. There are several types of route design network for feeder shipping service and commonly, there is

2015). The container throughput of Port Klang from 2011 to 2014, using the feeder service, is shown in Table 1.

a mixture of service patterns based on specific route connectivity. However, according to Hsu and Hsieh (2007), the optimal decision of direct service is more preferable than the hub and spoke service. A direct service was indicated to have less transit time, less additional costs, more attractive service, more reliable, increased shelf life, and decreased transportation damage. Less transit time includes the ports of departure and destination with no waiting time at the next port (Polat, 2013). Polat (2013) justified that the direct service is better although somehow it depends on the specific area, whether it has a low or high current demand.

Therefore, the purpose of this paper is to analyse the routes of potential trading partners in order to maintain the Malaysian feeder and container port’s economy in the future by considering market, time, cost, service and miscellaneous factors using the Multiple Criteria Decision Making Approach (MCDM).

Methodology

Application of the Multiple Criteria Decision Making Approach

In applying the MCDM approach in this study, the factors considered are market, time, cost, service and miscellaneous factors as they were commonly highlighted in the literatures (Hsu & Hsieh, 2007; Lun, 2013; Polat et al., 2014; Tran et al., 2016; Polat & Gunther, 2016). Market factor includes the daily container supply

Table 1: Container throughputs of Port Klang using the feeder service (2011 – 2014)

Year Import (Teus) Export (Teus) Transshipment (Teus)

2011 1,330,621 1,673,333 6,022,454

2012 1,873,257 1,821,995 6,305,963

2013 1,915,603 1,860,613 6,574,193

2014 1,962,431 1,942,773 7,040,600

Source: ASEAN Ports Association Malaysia, 2015


17Muhamad Nasir Rahmatdin1 et al.


Table 2: Container throughput in deadweight tonnage of trading ports via Port Klang

No. Port/Country Year Container throughput in deadweight tonnage

Import Export

1. Ho Chi Minh City,Vietnam

2010 666,035 404,7802011 596,302 518,4492012 233,355 265,963

2. Laem Chabang,Thailand

2010 1,674,451 1,001,0962011 2,535,734 1,200,2392012 2,334,485 1,323,414

3. Jakarta,Indonesia

2010 1,829,204 2,179,3932011 1,751,283 2,057,3212012 1,801,213 2,063,369

4. Chittagong,Bangladesh

2010 626,760 1,819,2442011 916,184 2,064,7492012 745,897 2,311,520

5. Yangon,Myanmar

2010 868,253 883,1982011 1,317,825 1,030,8492012 1,273,156 1,231,866

Source: Port Klang Authority (2015)

(pick-up) of port and daily container demand (delivery) of port. Time factor covers on the duration of sailing season/days, maximum allowable voyage duration/hours, vessel set-up duration of ship type in port (pilotage, berthing, cleaning etc.), lay-up duration of ship, and berthing duration of ship at port. Cost factor covers on the chartering cost of ship, operating cost of ship (administration, maintenance, lubricant, insurance etc.), vessel set-up cost of ship at port, auxiliary fuel consumption of ship type at berth, and main fuel consumption of ship. Service factor covers on the available number of container feeder ship, port operation efficiency, and frequency of service in a sailing

Table 2 shows the container throughput in deadweight tonnage derived from feeder service connected via West Port and North Port to the five selected ports. For the import of containers from 2010 to 2012, Laem Chabang generated the highest container throughput in deadweight tonnage, followed by Jakarta, Yangon, Chittagong and Ho Chi Minh City.

season. Miscellaneous factor covers on the ship loading capacity, average voyage speed of ship, distance between ports, main fuel oil price, and auxiliary fuel oil price.

Selection of Alternative Ports

Five ports are selected as alternative ports. They are Ho Chi Minh City, Laem Chabang, Jakarta, Chittagong and Yangon ports. The selection is based on the statistical data provided by the Port Klang Authority (2015) concerning the import and export container data via Port Klang from 2010 to 2012. The summarized data is presented in Table 2.

For export containers during the same period, Jakarta ranked first, followed by Laem Chabang, Chittagong, Yangon and Ho Chi Minh City. During this three-year period, supply and demand generated by the customers fluctuated. In fact, supply and demand are intangible and can change depending on certain trade area market conditions.


Data Sources

A qualitative approach was used in this study to get the opinions from the industrial experts regarding the assessment of feeder routes

Table 3 shows that the respondents were from feeder service providers, port authority and port operator. Basically, they were selected because of their broad knowledge and experiences in feeder shipping services. Besides, they are among the decision makers or policy makers in the management and operational level of their organization which involves and have influences in the companies’ decisions process to operate their businesses.

service in Malaysia. Ten experts participated in this study and delivered their opinions on the subject matter. The respondents are described in Table 3.

Data Analysis

There are three types of data analysis used in this study. The first is Analytical Hierarchy Process (AHP), second is Fuzzy-Link Based Technique and third is Evidential Reasoning (ER). The outcome of the first method is linked to the second method. Then, the result of the second method is used in the analysis using the third method.

Table 3: List of respondents participated in the study

No. Feeder ServiceProviders Level of Position Year of Experience Routes

1.Malaysia Trade & Transport (MTT)

ShippingOperation Manager 17 year in shipping

operationMalaysian Peninsular to

Sabah and Sarawak

2. Perkapalan Dai Zhun Lines (PDZ) Operation Manager

12 years sailing and 10 years in shipping

operation

Malaysian Peninsular to Sabah and Sarawak

3. Regional Container Lines (RCL)

Manager of Operational

Support Division

20 years in shipping operation and logistics

East Asia, South East Asia, and Middle East

4.Evergreen Marine

Corp (MALAYSIA) Sdn Bhd

Deputy Manager Operation

16 years in shipping operation

East Asia, South East Asia, Southern Asia

5.X-Press feeder (Sea

Consortium Sdn. Bhd)

Senior Executive Operations

11 years in shipping operation

East Asia, South East Asia, East Asia, Southern

Asia and Oceania

6. Bengal Tiger Line (M) Sdn Bhd General Manager 16 years in logistics and

shipping operationSouthern Asia and South

East Asia

7. Q-Express Line (QEL) Operation Manager 19 years in shipping

operationSouth East Asia and

Oceania

8. Harbour-Link Group Berhad Operation Manager 22 years in shipping

operationSouth East Asia and East

Asia

9. Port Klang AuthorityPlanning and Development

Officer12 years Malaysian waterways

and ports

10. Westport (M) Bhd. Marketing Manager

15 years in shipping marketing

Domestic and International waters



assessment on pairs of criterion Ai and Aj are represented by a n x n matrix A as follows:

Application of Fuzzy Link Based Technique

Yang et al. (2009), Abdul Rahman (2012) and Abdul Rahman and Ahmad Najib (2017) stated the different criterion, use different grades of assessment, which need to be standardized by

Analytical Hierarchy Process (AHP)

According to Abdul Rahman (2012) and Abdul Rahman and Ahmad Najib (2017), the

Legends of items as i,j = 1, 2, 3,..., n and eachaij in respect of attribute Ai to attribute. Then, the weight value of pairwise comparison between

attributes can be calculated as below (Abdul Rahman, 2012):

Where, aij stands for the entry of row i and column j in a comparison matrix of order n. The validity of analysis is checked through

Consistency Ratio (CR) obtained from IDS software which must be equal or less than 0.10 (Yang & Xu, 2002).

using the transformation of belief degree from fuzzy input to fuzzy output. The transformation process is a process to convert parent of Lower Level Criteria (LLC) to the Upper Level Criteria (ULC). Equation 3, 4, and 5 are represented by referring to Figure 1:

and 5) and sum of = 1; = 1.Where, uj is fuzzy output; li = fuzzy input; = belief degree assigned by experts, ( j,i=1, 2, 3, 4



Fuzzy Output

u1 u3 u2

u5

u4

U1 U5 U4 U3 U2

L1 L5 L4 L3 L2

l1

l3

l2

l5

l4

Fuzzy Input

Fuzzy Output

Fuzzy Input

Figure 1: Transformation of fuzzy input to fuzzy output values (Abdul Rahman, 2012; Abdul Rahman & Ahmad Najib, 2017)

Evidential Reasoning (ER)

Referring to Zhou et al. (2013), ER is used to solve problem dealing with aggregate of Multiple Attribute Decision Analysis (MADA). There are top level and lower level attributes which are known as main criteria and sub criteria with L basic criterion ei(i = 1, 2,...,L)linked with a general criterion (Yang & Xu,

2002; Abdul Rahman, 2012). The weight values of criterion or attributes denoted as. wi(i = 1,2,...,L) Where, wi is the relative weight of the th basic attribute ei with 0 ≤ wi ≤ 1. The AHP’s algorithms is used to obtain the weight value of criterion. The assessment of attribute ei(i = 1,2,...,L) is denoted in Equation 6 (Yang & Xu, 2002; Abdul Rahman, 2012; Zhou et al., 2013).



Where, is the th is assumed to be collectively comprehensive of evaluation grade. is denoted by a degree of belief satisfying and (Abdul Rahman, 2012). Then, an assessment of is completed if (respectively, ) or otherwise it is considered incomplete (Zhou et

al., 2013). According to Abdul Rahman (2012), be a basic probability mass which indicates the degree to which the th basic criterion support the hypothesis that the criterion y is assessed to the th grade. is calculated following Yang and Xu (2002) and Abdul Rahman (2012):

Then, wi is required to be normalized. mH,i is denoted as (Abdul Rahman, 2012):

Then, the residual of probability mass is divided into two parts, and and the calculation process is conducted by referring to

Equation 9 and Equation 10 (Yang & Xu, 2002; Abdul Rahman, 2012; Zhou et al., 2013).

. is a fundamental probability mass representing the belief degree of the basic attributes , while is the residual of the belief degree assessment. The

recursive evidential reasoning algorithm can be summarised as Equation 11, 12, and 13 (Yang & Xu, 2002; Abdul Rahman, 2012; Zhou et al., 2013; Zhou et al., 2016):

Where, K is a normalising factor. The normalisation of the probability can be

computed by using Equation 14:

The degree of belief for specific of grade is stated in Equation 15 (Yang & Xu, 2002; Abdul

Rahman, 2012):



Steps in the Assessment of Direct Feeder Service via Port Klang in Malaysia

Step 1: Set up a goal

The main purpose of this study is to find the benchmark of current assessments of direct feeder service by using route parameters (refer Step 2). The assessment parameters are summarized in a route assessment model of feeder service, which are separated to goal, main criteria, sub criteria, and alternatives (Figure 2).

Step 2: Identification of main criteria and sub criteria

According to Hsu and Hsieh (2007), Maras et al. (2012), Lun (2013), Polat et al. (2014), Tran et al. (2016), Polat and Gunther (2016) and validated by industrial experts, the route assessment criterion include five factors which have been discussed above (in the section ‘Application of the Multiple Criteria Decision Making Approach’). They are summarized as follows: (Market’s sub criteria: Daily container supply (pick-up) of port (DCS); Daily container demand (delivery) of port (DCD)); (Miscellaneous sub criteria: Ship loading capacity (SLC); Average voyage speed of ship (AVS); Distance between ports (DBP); Main fuel oil price (MFO); Auxiliary fuel oil price (AFO)); (Cost’s sub criteria: Chartering cost of ship (CCS); Operating cost of ship (administration, maintenance, lubricant, insurance etc.) (OPC); Vessel set-up cost of ship at port (VSC); Auxiliary fuel

The selection of route assessment factors is conducted through a comprehensive discussion and brainstorming by the experts to ensure the factors obtained from the literature were relevant and acceptable. While some of these factors can be measured numerically, others are less tangible and must be assessed on a qualitative basis. Then, a model has been developed to evaluate the alternative ports via direct service by using route assessment. The main criteria act as parents to sub criteria as shown in Figure 2.

consumption of ship type at berth (AFC); Main fuel consumption of ship (MFC)); (Time’s sub criteria: Duration of sailing season, days (DSS); Maximum allowable voyage duration, hours (MAV); Vessel set-up duration of ship type in port (pilotage, berthing, cleaning etc.) (VSU); Lay-up duration of ship (LAU); Berthing duration of ship at port (BDS)); (Service’s sub criteria: Available number of container feeder ship (ANC); Port operation efficiency (POE); Frequency of service in a sailing season (FSS)).

Step 3: Identification of potential trading partners via route assessments

There are five trading partners which have been discussed above (in the section ‘Selection of Alternative Ports’). The first route is between Port Klang - Ho Chi Minh City, second route is Port Klang – Laem Chabang, third route is

Route Assessment of Direct Feeder Service

Market (Mt) Miscellaneous (Mc) Service (Svc)

AN

C

POE

FSS

LAU

DSS

VSU

MA

V

BD

S

Time (Tm)Cost (Cst)

MFC

AFC

CC

S

VSC

OPC

MFO

AFO

AV

S

DB

P

SLC

DC

D

DC

S

Ho Chi MinhLaem ChabangJakartaChittagongYangon

Goal

Main Criteria

Sub Criteria

Alternatives

Figure 2: The proposed model for route assessment of direct feeder service



Port Klang – Jakarta, fourth route is Port Klang – Chittagong, and fifth route is Port Klang – Yangon. All five routes are assessed from the perspective of route assessment parameters as shown in Figure 2.

Step 4: Data collection process

A comprehensive data set is obtained from eight feeder operators and three port authorities in

Step 5: Establishment of weight value for each criterion using AHP approach

The weight value or normalized principal eigenvector of the main criteria in Step 4 is

Port Klang, who have been detailed above (in the section ‘Data Sources’). The qualitative data set is extracted from the questionnaire’s answers conducted along with interview session. They also need to justify their answers related to the parameters used. Table 4 shows the route assessment using pairwise comparison which was extracted from primary data.

computed by using Equation 2. The weight values of A(SvcTmCstMtMc) are calculated in Table 5 and summarised in Table 6.

Table 4: Route Assessment using Pairwise Comparison

Main Criteria SVC TM CST MT MC

A(SvcTmCstMtMc)=

SVC 1.0000 4.4000 5.1250 0.8181 4.4000

TM 0.2273 1.0000 2.8333 0.2759 2.0000

CST 0.1951 0.3529 1.0000 0.2273 2.0000

MT 1.2222 3.6250 4.4000 1.0000 2.2500

MC 0.2273 0.5000 0.5000 0.4444 1.0000

SUM 2.8719 9.8779 13.8583 2.7657 11.6500

Note: SVC=Service; TM=Time; CST=Cost; MT=Market; MC=Miscellaneous

Table 5: Values of Normalized Principal Eigenvector of each main criterion

Main Criteria SVC TM CST MT MC

SVC 1.0000 ÷ 2.8719 = 0.3482

4.4000 ÷ 9.8779 = 0.4454

5.1250 ÷ 13.8583 = 0.3698

0.818 ÷ 2.765 = 0.2958

4.4000 ÷ 11.6500 = 0.3777

TM 0.2273 1.0000 2.8333 0.2759 2.0000

CST 0.1951 0.3529 1.0000 0.2273 2.0000

MT 1.2222 3.6250 4.4000 1.0000 2.2500

MC 0.2273 0.5000 0.5000 0.4444 1.0000

SUM 2.8719 9.8779 13.8583 2.7657 11.6500

Note: SVC=Service; TM=Time; CST=Cost; MT=Market; MC=Miscellaneous

An example, the weight value of main criteria “SVC” is shown as follows:

WSVC = (0.3482+0.4454+0.3698+0.2958+0.3777) / 5 = 0.3674



The weight value of the main criteria “SVC” is 0.3674. By using the similar technique in Step 4 and Step 5, the calculation algorithms are applied to all main criteria.

Then, the weight values of main criterion are summarized in Table 6. In addition, the weight values of sub criteria are calculated by using the Intelligent Decision Software (IDS). The weight

Step 6: Using Fuzzy-link based theory to convert lower level criteria (LLC) to upper level criteria (ULC)

Then, a set of assessment grades of ULC and

Table 6: Weight values of evaluation criteria

Main Criteria SVC TM CST MT MC Weight

Values

SVC 0.3482 0.4454 0.3698 0.2958 0.3777 0.3674

TM 0.0791 0.1012 0.2044 0.0998 0.1717 0.1310

CST 0.0680 0.0358 0.0722 0.0821 0.1717 0.0861

MT 0.4256 0.3670 0.3175 0.3616 0.1931 0.3330

MC 0.0791 0.0506 0.0361 0.1607 0.0858 0.0825

SUM 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000

Table 7: Assessment Grades of Main Criteria (ULC)

Main Criteria /Upper Level Criteria Assessment Grades

SVC Excellent Good Average Poor Worst

TM Short Reasonably Short Average Reasonably

Long Long

CST Low Reasonably Low Average Reasonably

High High

MT High Reasonably High Average Reasonably

Low Low

MC Excellent Good Average Poor Worst

values of sub criteria are obtained as follow: 0.5455 (DCD), 0.4545 (DCS), 0.5189 (DSS), 0.2892 (MAV), 0.0867 (VSU), 0.0536 (LAU), 0.0517 (BDS), 0.3641 (SLC), 0.2240 (AVS), 0.0909 (DBP), 0.2636 (MFO), 0.0574 (AFO), 0.1356 (ANC), 0.4731 (POE), 0.3913 (FSS), 0.2707 (CCS), 0.3434 (OPC), 0.1098 (VSC), 0.0688 (AFC), and 0.2073 (MFC).

LLC is established. It is determined by a brainstorming process conducted via industrial experts and academicians. The example of assessment grades used in this study can be referred to in Table 7.

An example of fuzzy input data for “Market” towards “Ho Chi Minh City”, referring to the assessment grade in Table 7 is shown as follows: Fuzzy input for sub criteria “DCD” under “MT”: (High = 0.4; Reasonably High = 0.4; Average = 0.2; Reasonably Low = 0.0;Low = 0.0). =1.0; = 0.2; = 0.8;

= 0.6; = 0.2; = 0.8; = 0.2; = 1.0

Then, the fuzzy input is converted to fuzzy output using Equation 3, 4, 5 and Figure 1. The example of calculation as follows:



Step 7: Application of Evidential Reasoning (ER) Algorithm and Intelligent Decision System (IDS) software

The aggregated assessment values of criterion can be computed using Intelligent Decision Making (IDS) software, which is similar to manual calculations using the algorithms shown in Step 5. In Step 7, the weight values and belief

High = (0.4 x 1.0) + (0.4 x 0.2) = 0.4800 Reasonably High = (0.4 x 0.8) + (0.2 x 0.2) = 0.3600Average = (0.2 x0.6) = 0.1200Reasonably Low = (0.2 x 0.2) + (0.0 x 0.0) = 0.0400Low = (0.0 x 0.0) + (0.0 x 0.0) = 0.0000

Table 8: The Belief Degree Values of the Main Criteria “Market” towards “Ho Chi Minh City”

Belief Degree (β) Assessment Grades

High Reasonably High Average Reasonably

Low Low

Market DCD 0.4800 0.3600 0.1200 0.0400 0.0000

DCS 0.0600 0.3600 0.3600 0.2000 0.0200

degree assessments of all main criteria and sub-criteria were completed using a fuzzy link based on manual calculation and computed using the IDS software. The IDS software compares between AHP and ER to analyze the constructed model. Example of the calculation is shown in Table 8.

The belief degree values of DCD and DCS are arranged using Equation 6:

S (DCD) = {(High, 0.4800), (Reasonably High, 0.3600), (Average, 0.1200), (Reasonably Low, 0.0400), (Low, 0.0000)}.

S (DCS) = {(High, 0.0600), (Reasonably High, 0.3600), (Average, 0.3600), (Reasonably Low, 0.2000), (Low, 0.0200)}.

The weight value of “DCD” is 0.5455 and “DCS’” is 0.4545. Both of weight values also known as normalized principal eigenvector have been obtained by applying same step in Step 6. By using the information given in Table 6 and the weight values, the basic probability masses are calculated using Equation 7:

mn,i of DCD = [m1,1 = 0.5455 × 0.4800 = 0.26184, m1,2 = 0.5455 × 0.3600 = 0.19638, m1,3 = 0.06546, m1,4 = 0.02182, m1,5 = 0.0000]

mn,i of DCS = [m2,1 = 0.4545 × 0.0600 = 0.02727, m2,2 = 0.4545 × 0.3600 = 0.16362, m2,3 = 0.4545 × 0.3600 =

0.16362, m2,4 = 0.09090, m2,5 = 0.00909] Then, the values for mH,i refer to mH,1 and mH,2 are calculated using Equation 8. can be computed using Equation 9, while can be calculated using Equation 10.

The calculation process to obtain the values for , , , , , and are shown as follow:

= 1 – 0.5455 = 0.4545= 0.5454 (1– (0.4 + 0.4 + 0.2)) = 0.0000= 1 – (0.2182 + 0.2182 + 0.1091) = 0.4545

=1– 0.4545 = 0.5455= 0.4545 (1– (1– (0.3 + 0.6 + 0.1)) = 0.0000=1 – (0.1364 + 0.2727 + 0.0454) = 0.5455

By applying Equation 11, K value is expressed in mathematical form:



Then, the normalised factor (K) is used to obtain probability mass (m) values by applying

Afterward, Equation 15 is applied to obtain the belief degree values of “Market” with respect to

Therefore, the aggregated assessment of “Market” with respect to the alternative “Ho Chi

m1 = K (m1,1 m2,1 + m1,1 mH,2 + mH,1 m2,1) = 1.24375 (0.00714 + 0.14284 + 0.01239) = 0.20195m2 = K (m1,2 m2,2 + m1,2 mH,2 + mH,1 m2,2) = 1.24375 (0.03213 + 0.10713 + 0.07437) = 0.26570m3 = K (m1,3 m2,3 + m1,3 mH,2 + mH,1 m2,3) = 1.24375 (0.01071 + 0.03571 + 0.07437) = 0.15023m3 = K (m1,4 m2,4 + m1,4 mH,2 + mH,1 m2,4) = 1.24375 (0.00198 + 0.01190 + 0.04131) = 0.06864

Equation 12:

The normalization of the probability is computed using Equation 13:

Next, using of Equation 14 to obtain the normalization of the probability :

the alternative “Ho Chi Minh City:

Minh City” is summarized as:



{(High, 29.20%), (Reasonably High, 38.41%), (Average, 21.72%), (Reasonably Low, 9.93%), (Low, 0.74%)}

Results and Discussions

According to discussions with industrial experts, the five alternative ports have the potential to be strong and sustainable future trading partners for the import and export containers of Malaysia and Port Klang. The different commodities are derived from demand and supply from all five ports (personal communication with Dev Prasad, Marketing Manager of Westport, Malaysia). However, not all container commodities exported from Port Klang originate, in terms of being manufactured or assembled, in Malaysia.

Further, based on interview session, out of ten respondents, nine of them (90%) agreed that the direct service is better than the multi-port calling service. Hsu and Hsieh (2007) stated that the optimal decision of direct service is more preferable than the hub and spoke service. In fact, there are several types of route design network; and commonly, there is a mixture of service patterns based on specific route connectivity. The 90% of respondents that agreed that multi-port calling was disadvantageous in terms of vessel delays could affect the fix window at ports and links to other ports to serve. A delayed sailing schedule totally influences the schedules of loading and discharging containers at all ports as a whole.

It is complicated because it involves multiple ports and the feeder operators need to verify estimated times of arrival (ETA) at the port, consider waiting times and queues at anchorage areas, waiting times of vessel’s turn to berth, and discharging or loading containers. Readjustment of schedules adversely affects the efficiency and frequency of service, customer reliability, operational cost, fuel cost, chartering cost, etc. However, a direct service has advantages of less transit time, less additional costs, more attractive service, more reliable, increased shelf life, and decreased transportation damage (Polat, 2013). Less transit time is involved because it specifically involves the ports of departure and destination with no waiting time at the next port (Polat, 2013). Three out of ten (30%) respondents justified that the direct service is better; somehow it depends on the specific area, whether it has a low or high current demand. Multiple port of calling is suitable for certain circumstances, such as the diversion of trade areas or depending on the demand created by the shipper, cargo owner, consignor and consignee. Ships can be fully utilized if the feeder operators are able to cater for the demand using several ports. In fact, the flexibility of route choices depends on the feeder operators themselves empowering their strategy and maintaining the sustainability of a company in the market.

The results of the data analyses are presented in the following bar charts. The assessment grade of belief degree in percentage is shown in Figure 3.



Figure 3: The evaluation of belief degree assessment grade of “market” towards alternative trade between Port Klang and Ho Chi Minh City

Figure 4: Route assessment of “Port Klang - Ho Chi Minh City”

The highest evaluation grade, referring to Ho Chi Minh City on “Market” is “Reasonably High” at 38.41%. Second was “High” at 29.20%, third

“Average” at 21.72%, fourth “Reasonably Low” at 9.93% and finally, “Low” at 0.74%.



Figure 5: Average score for Port Klang via five alternative ports towards route assessment of the feeder service

Table 9: The summary of results of direct feeder service via Port Klang in Malaysia

Main Criteria Summary of Results

Ranking of trading partners via Port Klang, Malaysia

Ports of service

Ho Chi Minh

Laem Chabang Jakarta Chittagong Yangon

Score 0.7202 0.7653 0.6919 0.5401 0.5020

Ranking 2 1 3 4 5

Figure 4 shows that the highest evaluation grade referring to Ho Chi Minh City on “Route Assessment” was “Reasonably Beneficial” at 56.06%. Second was “Most Beneficial” at 20.08%, third “Moderate” at 16.24%, fourth “Reasonably Less Beneficial” at 7.08%

The average score for Ho Chi Minh City is linked between Figure 3 and 4. The calculation

Most Beneficial : 20.08% ÷100 x 1.00 = 0.2008 Reasonably Beneficial : 56.06% ÷ 100 x 0.75 = 0.4205 Moderate : 16.24% ÷ 100 x 0.50 = 0.0812Reasonably Less Beneficial : 7.08% ÷ 100 x 0.25 = 0.0177Less Beneficial : 0.54% ÷ 100x 0.00 = 0.0000

and finally, “Less Beneficial” at 0.54%. All percentage values of evaluation grades on belief degree were used to calculate the ranking of alternatives as overall assessments. The calculation for alternative “Ho Chi Minh City” is shown as follows:

can be shown in mathematical form as: 0.2008 + 0.4205 + 0.0812 + 0.0177 = 0.7202



Ranking of criterion for feeder route assessment

Main criterion

Service (SVC)

Time (TM) Cost (CST) Market

(MT)Miscellaneous

(MC)

Score 0.3674 0.1310 0.0861 0.3330 0.0825

Ranking 1 3 4 2 5

The results shown in Figure 5 and Table 9 show the average score for Port Klang via five alternative ports towards route assessment of the feeder service. It can be seen that the average score for Laem Chabang is 0.7653, followed by Ho Chi Minh City in second place with 0.7202. Next, Jakarta is in third place at 0.6919, fourth is Chittagong (at 0.5401), and Yangon (at 0.5020) in fifth place. Based on Table 9, the ranking of main criterion in assessment of feeder routes and potential trading partners are “Service” in the first place (0.3674), followed by “Market” (0.3330), “Time” (0.1310), “Cost” (0.0861) and “Miscellaneous” (0.0825). These results are consistent with a Critical Ratio, CR: 0.0766 less than 0.10. SVC, or “Service”, is the most important factor of all main criteria. An efficient service, provided by port and shipping companies, results in efficient vessel use via increased service frequency, time and cost saving at the port, retained customer reliability, and catering to rapid market demand. In reality, shipping and port competition is very high (Westport, 2016). Shippers choose the best shipping services to deliver their cargos in time. Competition between commodity manufacturers also needs to fulfil current “Market” demand. “Service” is a key strength of feeder operators to remaining competitive in the “Market”. The high-quality “Service” of a port and feeder player also increases “Time” efficiency in both; thus reducing the “Costs” incurred by the shipping company. “Miscellaneous” factors, such as ship’s carrying capacity, main and auxiliary fuel prices, average voyage speed, distance between ports of service, are also important to be considered.

Laem Chabang port was identified as the most potential trading partner between the five selected ports. The current view of feeder operators and Port Klang authorities show Laem Chabang as a great prospect to be the

port of choice in maintaining the economy of feeder operators and Port Klang in the future. Therefore, it is a great opportunity for feeder operators to create future trading plans in effectively making Laem Chabang a pick-up and delivery point for cargo. Laem Chabang port is more efficient, in terms of cargo handling, than Ho Chi Minh City, Jakarta, Chittagong, and Yangon thus, reduce waiting time by a ship at port. Ship operators always expect the least waiting time for their container ships at port (Kiani et al., 2006). A long waiting time in berth increases the port’s profitability. This accounts for port charges, but incurs higher costs for the shipping operators (Malchow & Kanafani, 2001). Shipping liners also bear higher operating costs due to increasing fuel consumption. This results from increased ship speeds, in terms of following fixed sailing schedules (Notteboom, 2006). High fuel consumption produces high emissions of greenhouse gases (mainly Nitrogen Oxide and Sulphur Oxide) from shipping activities (Martinez & Sin, 2011). In Malaysian waterways, merchant ships are required to comply with International Conventions for the Prevention of Pollution from Ships (MARPOL - Annex VI Regulations for the Prevention of Air Pollution from Ships beginning from September 27, 2010 (Marine Department Malaysia, 2011). Therefore, efficient cargo shipments using the route Port Klang – Laem Chabang could potentially reduce air pollution and influence the sustainability of the environment between the ports of service.

Conclusion

In summary, the combined methods of AHP-Fuzzy Link Based Technique and ER are able to assess feeder route parameters. This results in a hierarchal route arrangement of potential trade partners. The managerial contributions of this study can be summarised as the feeder



operators in Port Klang needing to strengthen their services, optimise their sailing schedules and trade relationship with Laem Chabang as the first port of call choice, followed by Ho Chi Minh City, Jakarta, Chittagong and finally Yangon. This is considered a strategy that will maintain feeder operators, Port Klang, and Malaysia’s maritime economy in the long-term. However, the competition between feeder operators usually high to fulfil market demand due to the limited service windows available at ports. Therefore, to be a competitive shipping operator, the feeder operator must take the initiative and deploy an efficient shipping service with suitable carrying capacities. In addition, they should discover ports that offer a low bunker price. According to Stopford (1998), the bunker is the largest variable cost in shipping operations. An optimised consumption of ship fuel or bunker has also resulted in lower emissions of air pollutants (Martinez & Sin, 2011). Furthermore, a viable freight rate could be offered; and thus, generate the chance to penetrate the potential market.

The theoretical contribution of this paper is that the researchers have proposed a feeder route assessment model using a Multi Criteria Decision Making (MCDM) approach. Experts have been able to assess the tangible factors that result in the arrangement of feeder routes and potential trading partners. The methods and model applied in this research are flexible enough to deal with the uncertain conditions of future or potential prospects that cannot be determined using numerical or quantitative measurements. This assessment can also help feeder companies, terminal operators, and port authorities to enhance their capabilities and service marketability which influences maritime stakeholder’s profitability as a whole. Moreover, the constructed model can be used by maritime stakeholders and academicians based on their own test cases.

Acknowledgements

The authors would like to thank especially to the Ministry of Higher Education, Malaysia

and Universiti Malaysia Terengganu (UMT) by giving the financial support under the Fundamental Research Grant Scheme (FRGS) with vote number: 59415.

References

Abdul Rahman, N. S. F. (2012). Selection of the most beneficial shipping business strategy for containerships. European Journal of Business and Management, 4(17): 153-167.

Abdul Rahman, N. S. F., Shaharuddin, A. H., & Rasdi, R. (2014). Effect of the Northern Sea Route Opening to the shipping activities at Malacca Straits. International Journal of e-Navigation and Maritime Economy, 1: 85-98.

Abdul Rahman, N. S. F., & Ahmad Najib, A. F. (2017), Selection of the most practical Malaysian port for enhancing the Malaysia-China Kuantan Industrial Park business trade. International Journal of Shipping and Transport Logistics, 9(4): 500-525.

ASEAN Ports Association Malaysia (2015). Total container throughput by ports, Malaysia. Ports Statistic. Retrieved from http://www.apamalaysia.com/downloads /mar i t ime- t ranspor t -statistic/, 5 February 2016.

Hsu, C. I., & Hsieh, Y. P. (2007). Routing, ship size, and sailing frequency decision-making for a maritime hub-and-spoke container network. Journal of Mathematical and Computer Modelling, 45(7): 899-916.

Kiani, M., Bonsall, S., Wang, J., & Wall, A. (2006). A break-even model for evaluating the cost of container ships waiting times and berth unproductive times in automated quayside operations. WMU Journal of Maritime Affairs, 5(2): 153-179.



Lun, Y. H. V. (2013). Development of green shipping network to enhance environmental and economic performance. Polish Maritime Research, 79(1): 13-19.

Malchow, M., & Kanafani, A. (2001). A disaggregate analysis of factors influencing port selection. Maritime Policy and Management, 28(3): 265-77.

Maras, V., Konings, R., Radmilovic, Z., & Davidovic, T. (2012). Towards the optimal solution of feeder container ships routing with empty container repositioning. Journal of Maritime Research, IX (3): 11-20.

Marine Department Malaysia (2011). Malaysia Shipping Notice. Retrieved from http://www.marine.gov.my/jlmeng/pic/article/service/notice/NPM/2011/MARPOL_MSN2 42011.pdf, 8 February 2016.

Martinez, F., & Sin, M. (2011). Improvement of the energy efficiency of vessels as a measure for the reduction of greenhouses gases emission from sea shipping. Journal of Marine Technology and Environment, 1(4): 103-112.

Notteboom, T. (2006). The time factor in liner shipping services. Maritime Economics & Logistics, 8(1): 19-39.

Polat, O. (2013). Designing liner shipping feeder service networks in the new era of mega container ships. Thesis of Doctoral Degree. Technischen Universität Berlin.

Polat, O., & Gunther, H. O. (2016). The impact of seasonal demand fluctuations on service network design of container feeder lines. Journal of Transportation and Logistics, 1(1): 40-58.

Polat, O., Gunther, H. O., & Kulak, O. (2014). The feeder network design problem: Application to container services in the Black Sea region. Maritime Economics & Logistics, 16(3): 343-369.

Port Klang Authority (2015). Annual Report 2015. Retrieved from http://www.pka.gov.my/index.php/download/category/66-anual-report.html, 7 February 2016.

Stopford, M. (1997). Maritime Economics, 2nd Edition. Routledge, London. 562 pp.

Tran, N. K., Haasis, H. D., & Buer, T. (2016). Container shipping route design incorporating the costs of shipping, inland/feeder transport, inventory and CO2 emission. Maritime Economics & Logistics, 19(4): 667–694.

UNCTAD (2015). Review of Maritime Transport 2015. Geneva, Switzerland: United Nations.

Verny, J., & Grigentin, C. (2009). Container shipping on the Northern Sea Route. International Journal of Production Economics, 121(1): 107-117.

Westport (2016). Annual Report 2016. Retrieved from http:/ / ir.chartnexus.com/westportsholdings/docs/ar2016_lres.pdf, 2 February 2016.

Yang, J. B., & Xu, D. L. (2002). On the evidential reasoning algorithm for multiple attribute decision analysis under uncertainty. IEEE Transaction on System, Man and Cybernetics, 32(3): 289-304.

Yang, Z. L., Wang, J., Bonsall, S., & Fang, Q. G. (2009). Use of fuzzy evidential reasoning in maritime security assessment. Risk Analysis, 29(1): 95-120.

Zhou, M., Liu, X. B., Yang, J. B., & Fang, C. (2013). Group evidential reasoning




approach for MADA under fuzziness and uncertainties. International Journal of Computational ntelligence Systems, 6(3): 423-441.

Zhou, H., Wang, J. Q., Zhang, H. Y., & Chen, X. H. (2016). Linguistic hesitant fuzzy

multi- criteria decision making method based on evidential reasoning. International Journal of Systems Science, 47(2): 314-327.




MODERATING EFFECT OF SELF-EFFICACY ON THE RELATIONSHIP BETWEEN CONSUMER ATTITUDE AND WILLINGNESS TO CONSUME HEALTHY FOOD

HAYATUL SAFRAH SALLEH1* AND NOR AZILA MOHD NOOR2


2Othman Yeop Abdullah Graduate School of Business, Universiti Utara Malaysia, 06010 Sintok, Kedah, Malaysia


Abstract: The demand for healthy food consumption, such as functional food, has increased due to the prevalence of certain diseases that relate to unhealthy eating. Functional food is a new category of product that is fortified with nutrition and has been proven scientifically to prevent the risks of certain diseases and improve overall health. Healthy life will affect positively on household income, productivity and national economic prosperity because it may enable an individual to increase quality of life. Therefore, functional food consumption reflected the concept of sustainable diets that will contribute to the enhancement of sustainable livelihood, good health and well-being, as well as the economic and environment sustainability as a whole. Despite gaining acceptance among consumers, the information regarding the process of consumer behaviour, the influencing factors, and consequences are presently insufficient. Thus, there is a need to investigate the antecedent factors that influence the consumer willingness to consume healthy food in order to prevent diseases, increase good health and well-being as well as reducing medical costs. Attitude plays a role as guidance on how information regarding functional food is noticed, adopted, accepted or rejected. With this in view, attitude was found to be one of the key factors that influence the behavioural of individuals towards healthy food choice. This study aimed to investigate the attitude dimensions that can predict consumer willingness to consume functional food and to examine the effect of self-efficacy as the moderator on the relationship between attitude dimensions and consumer willingness to consume such food. Four dimensions of functional food-related attitudes were examined in this study, which are; a reward from using functional food, a necessity for functional food, a safety of functional food, and a confidence in functional food. A mall-intercept survey was conducted with the sample of 452 respondents by using the self-administered questionnaire. The results revealed that only reward and necessity significantly influence consumer willingness to consume functional food. However, confidence and safety have no significant influence on consumer willingness to consume functional food. The moderator of self-efficacy was statistically significant and moderates the relationship between attitude dimensions and consumer willingness to consume functional food. The results from this study would be valuable for the marketers in the healthy food industry to formulate effective marketing communication strategies in order to facilitate favourable attitudinal change towards good health and sustainable livelihood.

Keywords: Attitude, consumer willingness, self-efficacy, functional food, sustainable livelihood.

Introduction

Hippocrates, the father of medicine, once said: “Let food be thy medicine and medicine be thy food”. These words acknowledge the benefit of consuming healthy food and the capability of

certain foods for good health and well-being. Generally, product used to cure diseases known as medicinal products and not food. However, a healthy diet comprises of food with functional properties that can improve physiological

: 35-52

36MODERATING EFFECT OF SELF-EFFICACY ON THE RELATIONSHIP BETWEEN CONSUMER ATTITUDE AND WILLINGNESS TO CONSUME HEALTHY FOOD

functions and mental well-being, can also diminish the risk of developing certain diseases. As reported by World Health Organisation (World Health Organization, 2015), unhealthy food, unhealthy eating habits and less physical activity may lead to the global burden of non-communicable diseases (NCDs), which although non-infectious, are often referred to as chronic diseases. NCDs, such as hypertension, kidney disease, diabetes, chronic lung diseases, heart disease, high cholesterol and certain types of cancer, are increasing, affecting people of all age groups in all around the world.

Following the trends, consumers have changed the priority of consuming food from needs to satisfy hunger to the needs to become healthy and promote sustainable livelihood. Several studies demonstrate that consumers are progressively reflective concerning health issues and willing to adopt health-oriented changes in their eating routine (Niva, 2007; Kraus, 2015). Hence, to prevent NCDs, individuals must consume a healthy diet, such as consuming more functional foods and carry out physical activity. Apart from that, an increase in life expectancy, the desire for the better quality of lives, and the greater cost of medical and long-term rehabilitative care are the major motivations for an increasing demand for functional foods in the market designed to provide health benefits (Roberfroid, 2000). In this study, the term functional food was illustrated as food and drinks that provide health benefits that reduce the risk of some diseases in addition to its nutritional value (Urala & Lahteenmaki, 2007). Howlett (2008) and Kraus (2015) reported that functional food provides various health benefits and plays a role in several aspects of human health and well-being from the perspective of the availability of target functions that might be susceptible to dietary influence. Some of the functional food benefits are; to improve memory and eyesight, to promote optimal heart health and to promote optimal physical performance and health recovery (Kraus, 2015).

Previous studies discovered that individuals who consume more of functional

foods have been scientifically proven to reduce the rate of many diseases, thus, will improve the overall health of individuals (Pereira et al., 2004). Therefore, this study intended to develop more understanding towards factors that affect the issue of healthy consumption behaviour among Malaysian, which indeed leads to the healthy life and good well-being. The finding would shed further light on the knowledge in the area of functional food consumption behaviour, especially when it comes to discussing the relationship that attitude dimensions (i.e. reward, necessity, confidence, safety) have on willingness to consume functional food and the effect of self-efficacy on the relationship between consumer attitude and willingness to consume functional food.

In previous literature, the elements of attitude towards food choice, including functional foods were measured using general attitude scales such as “bad to good” and “pleasant to unpleasant” (Poulsen, 1999). Since functional food are known as a novelty product with credence attributes and categories as healthy food that come with special health benefits, thus, attitude measurements that directly emphasize the health-focus studies are needed (Urala & Lahteenmaki, 2004; 2007). In order to have a better prediction of consumer willingness to consume functional food, this study adopted four dimensions of attitude developed by Urala and Lahteenmaki (2007) that are more relevant to functional food-related attitudes, which are; reward from using functional foods, necessity for functional foods, confidence in functional foods, and safety of functional food.

These four dimensions of attitude were chosen because the functional food is different from conventional food in terms of extra benefits gained from consumption of such food. Individual consumed functional food not just to satisfy their hunger but also to obtain the reward from using functional food such as being healthy and lower the risk of getting certain diseases. Therefore, attitude towards necessity to consume functional food is important for individuals in order to obtain good health and


37Hayatul Safrah Salleh1 et al.

well-being (Urala & Lahteenmaki, 2007). In terms of attitude dimensions of confidence and safety, in order for consumption behaviour to be performed, individuals need to have a strong confidence and trust in food safety and benefit claimed by the manufacturer from using the functional food products (Landstrom et al., 2007; Urala & Lahteenmaki, 2007).

Urala and Lahteenmaki (2007) describe reward as a feeling of satisfaction and enjoyment in terms of health, mood, and well-being resulted from functional food consumption and they stated that perceived reward dimension is the most crucial factor in consumer acceptance towards functional food. Prior studies also found that consumer attitude that linked towards the perceived rewards, which mainly related to the healthiness of the food and well-being that they will get from consuming functional food, is the best predictor towards willingness to consume functional food (Bech-larsen & Grunert, 2003; Verbeke, 2005; Landstrom et al., 2007; Urala & Lahteenmaki, 2007; Barcellos & Lionello, 2011; Carrillo et al., 2013; Kraus, 2015). Hence, the hypothesis was proposed as:

Hypothesis 1: Perceived rewards of using functional food influence consumer willingness to consume such food.

The necessity for functional food measures how essential the consumers think that functional food is for themselves or for people in general (Urala & Lahteenmaki, 2004; 2007). In other words, the necessity for functional food is an attitude towards the perceived need for such foods due to the possible benefits for health. Urala and Lahteenmaki (2004; 2007) reported that there was a positive correlation between the necessity for functional food and the willingness to consume such food, however, the influence of necessity for functional food among Finnish consumers became weaker. Chen (2011) stated the necessity for functional food consumption significantly influences the willingness to use functional food among Taiwanese. However, a study by Landstrom et al. (2009) on Swedish consumers, aimed at examining consumer

attitudes, perceptions and the perceived need for functional food, reported consumers felt that functional foods were unnecessary if the person did not suffer from any health problems and that a lifestyle change would not result in optimal health. With this in view, the following hypothesis was proposed:

Hypothesis 2: Perceived necessity for functional food influence consumer willingness to consume such foods.

Confidence in functional food was measured as the confidence that the consumers have in functional food as food that promote health and the reliability of scientific basis and research of the promised health effect (Urala & Lahteenmaki, 2004; 2007). Thus, the term confidence is defined as consumer attitude towards the information and claims about functional food, particularly on functional food benefits related to health effects. Urala and Lahteenmaki (2004; 2007) reported apart from reward, confidence in functional food was also the important factor that influences the behavioural intention towards functional food consumption among consumers. Chen (2011) also found that confidence in functional food demonstrated a significant relationship towards consumer willingness to use functional food products in Taiwan. In addition, confidence and trust in food companies and the health benefits from using functional food are the most superior factor influencing the acceptance of functional food (Verbeke, 2005; Markovina et al., 2011; Siegrist et al., 2015) Moreover, Annunziata and Vecchio (2011) stated that the source of information plays an important role towards consumer confidence of the information about functional food. Previous results described that consumers had a lower level of confidence in information from newspaper, functional food’s manufacturer and retailer, while a high level of confidence is afforded to dieticians, nutritionists, doctors and government (Barcellos & Lionello, 2011). Thus, the hypothesis was proposed as:

Hypothesis 3: Perceived confidence in functional food influence consumer willingness to consume such food.



Finally, the safety dimension is operationalized as how respondents perceive the related risk or possible harmful effects when functional food are consumed (Urala & Lahteenmaki, 2004; 2007). Factors related to functional food safety, health benefits and functional food as medicine are the important determinants towards willingness to consume functional food among consumers in Thailand (Methakornkulnan et al., 2013). Some of the functional food was modified using sophisticated scientific methods by adding and/or removing the component of a food. Therefore, some consumers perceived functional food to be risky and were not safe to be consumed due to unnaturalness in food (Frewer et al., 2003). A study on functional food in Canada stated that consumers would disregard the health benefits of the functional food if the production technology seems frightening (Labrecque & Charlebois, 2011). Therefore, the perceived risks from functional food consumption would prevent the consumers to consume the functional food. However, studies on consumer attitudes by Urala and Lahteenmaki (2004), and Chen (2011) argued that the perceived safety towards the possible harmful effects of functional foods does not influence the consumer’s acceptance to use functional food in general. Hence, the following hypothesis was proposed:

Hypothesis 4: Perceived safety of functional food influence consumer willingness to consume such food.

The above explanation demonstrated that the results of the relationship between attitude dimensions and the consumer willingness to consume functional food from previous studies were inconsistent. Baron and Kenny (1986) stated that there may be some contingent (moderator) variable that has not investigated which could be the cause of this inconsistency. Therefore, the construct of self-efficacy was introduced as moderator in this study. Despite the health benefits obtained from consuming functional food, it is difficult to change

consumer behaviour, especially towards healthy consumption. Therefore, individuals need to have a strong and high level of internal motivation of self-efficacy in order to participate in healthy consumption behaviour, because high self-efficacy leads to a strong confident that they can perform the behaviour (Conner & Norman, 2005). Self-efficacy in this study deals with the strong internal believe that individuals have on their own capabilities and potential to perform the chosen behaviour as well as confidence in defeating the obstacles to achieving that behaviour (Bandura, 1986).

Prior studies also found that individuals who had strong confidence in their self-efficacy would change their behaviour towards a healthy diet and were less likely to relapse to their previous unhealthy diet (Mai & Hoffman, 2012; Fitzgerald et al., 2013). Supported by Conner and Norman (2005) and Hall et al. (2015) they stated that perceived self-efficacy is directly related to health-related behaviour. Moreover, Robles et al. (2014) and Nastaskin and Fiocco (2015) reported that the stronger the individual’s self-efficacy, the higher his/her obligation to consume healthier food. Therefore, the following hypothesis is proposed:

Hypothesis 5: Self-efficacy moderates the relationship between attitude dimensions and consumer willingness to consume functional food.

The main objectives of this study are; to examine the factors related to attitude dimensions that may influence the behaviour of the consumer towards willingness to consume functional food, and to examine the moderating effect of self-efficacy on the relationships between attitude dimensions and consumer willingness to consume such food.

Methodology

The research framework for this study is as shown in Figure 1.



This study uses the cross-sectional method to answer the study’s research questions in which data were gathered once. This study employs quantitative methodology involving individual consumers aged 18 and above who went shopping at 12 hypermarkets in Klang Valley, Malaysia. The convenience sampling technique was used in this study because a complete list of exact members of the population for this study cannot be determined, and the numbers of the population are more than 1,000,000 and the subjects are also widely scattered. Based on Krejcie dan Morgan (1970), it should be noted that as the population increases, the sample size increases at a diminishing rate and remains constant at 384 cases when the population reach 250,000 and above.

A self-administered questionnaire was distributed to the respondents personally

by using mall intercept survey method. Eight hundreds questionnaires were distributed and a total of 452 questionnaires were usable with a successful response rate of 56.5%. In the case where the respondent could not complete the questionnaire on site, they were given a pre-stamped envelope with the researcher’s address to return it by mail. Some of the questionnaires were not returned, incomplete and unusable.

The willingness to consume functional food, attitude dimensions and self-efficacy construct were measured on a five-point Likert scale, ranging from “strongly disagree” (1) to “strongly agree” (5). According to Table 1, willingness to consume functional food was measured using four items based on Verbeke and Vackier (2005) and five items were developed to examine self-efficacy based on Armitage and Conner (1999).

Figure 1: Research framework

Attitude Dimensions:1. Rewards2. Necessity3. Confidence4. Safety

Consumer Willingness to

Consume Healthy Food

Self-efficacy

Table 1: Measurement of the constructs

Construct Scale Items Statement Items Sources

Willingness to Consume Functional

Food

WC1 I will eat/drink functional food to make myself healthier

Verbeke and Vackier (2005)

WC2 I intent to eat/drink functional food to prevent me from certain chronic diseases

WC3 I want to eat/drink functional food product if I can trust it contains healthy component

WC4 I expect to consume more functional food in the future



Reward

R1 My mood improves when I consume functional food

Urala and Lahteenmaki

(2007)

R2 Functional foods helps to improve my performance

R3 Healthy lifestyle is easier to follow by eating functional foods

R4 Eating functional foods regularly can prevent certain disease

R5 The idea that I can take care of my health by consuming functional foods gives me happiness

R6 Functional foods can fix the damage caused by an unhealthy diet

R7 I am ready to compromise on the taste if the food is functional

R8 I am actively looking for information regarding functional foods

Necessity

N1 Eating functional foods are completely unnecessary (R)


(2007)

N2 Functional foods are fraud (R)

N3 The increasing number of functional foods in the market is a bad trend for the future (R)

N4 For a healthy individual, it is useless to consume functional foods (R)

N5** It is good that new technology enable the development of functional foods

N6** I only want to eat foods that do not have any medicine-like effects (R)

N7 Health effects are not suitable in delicacies (R)

N8 Functional foods are consumed mostly by individuals who have no need for them (R)

Confidence

C1** Functional foods upgrade my well-being


(2007)

C2 The safety of functional foods has been very thoroughly studied

C3 I believed that functional foods fulfil their promises

C4 Functional foods are science-based top products



Safety

S1 Consuming to much of function food can be harmful to health (R)


(2007)

S2 In some cases, functional foods may be harmful to healthy people (R)

S3* Consuming functional foods is completely safe

S4 The new components f functional foods carry unforeseen risks (R)

S5 Exaggerated information is given about health effects of functional foods (R)

Self-efficacy

SE1 Whether or not I consume functional foods is entirely up to me

Armitage and Conner (1999)

SE2 I am confident that I can consume functional food regularly

SE3 I am very sure that I would be able to consume functional foods next week

SE4 I am certain that I will be able to refrain myself from consuming foods that are not healthy

SE5 If I wanted to, it would be very easy for me to consume functional food every day

Note: (R) denotes items requiring reverse scoring. Item S3* was moved from the original safety dimension to confidence dimension and items N5**, N6** and C1** were deleted in the subsequent analysis.

Further, as shown in Table 1, attitude towards willingness to consume functional food in this study was operationalized by using four distinct attitude dimensions – reward, necessity, confidence, and safety – developed by Urala and Lahteenmaki (2007) with 25 functional food-related items. Accordingly, reward and necessity consist of eight items each, confidence with four items and safety with five items were produced.

Multiple regressions analysis was used to examine the amount of variance of consumers’ willingness to consume functional food explained by the attitude dimension, and hierarchical regression analysis was utilized to examine the effect of self-efficacy as a moderator on the relationship between attitude dimensions and consumer willingness to consume functional food. For the purpose of this

study, the three-step hierarchical regression was performed (Hair, 1998).

For ease of interpretation, for the results of descriptive statistics of the study, the range of the five-point Likert scale was grouped into similar sized categories of low, moderate and high. In which, scores of below than 2.33 [4/3 + lowest value (1)] are reflected as low, scores of 3.67 and above [highest value (5) – 4/3] are reflected as high, and those in between are reflected as moderate (Nor Azila et al., 2012). Based on recommendation from Hair et al. (1998), the measure of Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy will be explained using the following guideline: below 0.50, unacceptable; 0.50 to 0.59, poor; 0.60 to 0.69; moderate; 0.70 to 0.79, good; and 0.80 and above, excellent. Factor analysis is considered



appropriate if the value of KMO measure of sampling adequacy index with the minimum of 0.60 and Bartlett’s test of sphericity should be significant with p<0.05. Reliability analysis was performed on factors extracted to test the internal consistency of the measurement and the acceptability of reliability scale was determined by using the value of Cronbach’s Alpha coefficients. Sekaran and Bougie (2009) suggested that reliability coefficient less than 0.60 is poor, those in 0.70 ranges is acceptable, and over 0.80 are good.

Results

Characteristics of the Respondents

General demographics characteristics of the respondents were also collected as in Table 2.

There are large numbers of female respondents that demonstrated 70.4% of the samples and only 29.6% are male. The majority (74.6%) of the respondents are 18 to 40 years old. In terms of respondents’ religion, the majority are Muslim, with 76.8% and it is consistent with the ethnicity in which the percentage of Malays is 75.7%, 18.1% are Chinese, 5.1% are Indians and 1.1% are from other ethnic groups. The majority of the respondents are married (68.4 percent). In relation to education background, most of the respondents completed tertiary education (80.1%). Finally, the majority of respondents’ household income is more than RM3,000 (76.5%) per month.

Table 1: Measurement of the constructs

Characteristics Categories N %

GenderMale 134 29.6

Female 318 70.4

Age

18-30 years old 169 37.4

31-40 years old 168 37.2

41-50 years old 101 22.3

More than 50 years old 14 3.1

Religion

Muslim/Islam 347 76.8

Buddhism/Taoism 73 16.2

Hinduism 21 4.6

Christianity 10 2.2

Others 1 0.2

Ethnicity

Malay 342 75.7

Chinese 82 18.1

Indian 23 5.1

Others 5 1.1

Marital Status

Single 138 30.5

Married 309 68.4

Divorce/Widowed 5 1.1



Descriptive Analyses

Table 3 showed the results of descriptive statistics for the variables of the study. The mean values for consumer willingness to consume functional food and self-efficacy are 4.35 and 3.87 respectively. Clearly, according to Nor Azila et al. (2012), it shows that the respondents exhibit a high level of willingness to consume functional food, and their internal forces of

Factor Analysis

Table 4 represents the results of Factor analysis. The values of Cronbach’s Alpha were all higher than the lower limit of acceptability (α > 0.60), showing a highly reliable measurement for all of the variables. For the construct of consumer willingness to consume functional food, factor analysis was conducted based on four items. The result reveals a KMO value of 0.80, which shows that the data were free from multicollinearity and exceeds the recommended value of 0.50 (Hair et al., 1998). In addition, Bartlett’s test

self-efficacy are responsive to change towards healthy eating in order to be healthy. The mean values for attitude dimensions (reward, necessity and confidence) are generally high, which are in the range of 3.76 to 3.98, except for safety, falls in the moderate range (2.87). It explains that the respondents have doubts in trusting the safety of functional food.

of sphericity was highly significant (p<0.05), supporting the factorability of the correlation matrix. The principal components analysis indicated that only one component is present with an eigenvalue exceeding 1, which showed that the factor represents 68.28% of the total variance in the item. The factor loadings are between 0.80 and 0.87. The reliability for this factor is high, with the Cronbach’s Alpha value of 0.81. All of the items are included because the item-to-total correlation showed that removal of any item would not increase the alpha above 0.81.

Table 3: Descriptive statistics for all of the variables

Variables Mean Std. Deviation

Willingness to Consume FF 4.35 0.52

Reward 3.87 0.61

Necessity 3.98 0.87

Confidence 3.76 0.64

Safety 2.87 0.73

Self-efficacy 3.87 0.57

Education level

Primary/Secondary School 90 19.9

Certificate/Diploma 85 18.8

Degree 190 42.0

Masters/Doctorate 87 19.2

Monthly Income

RM3,000 or less 106 23.5

RM3,001-RM5,000 175 38.7

RM5,001-RM7,000 75 16.6

RM7,001-RM9,000 50 11.1

More than RM9,000 46 10.2



The attitude construct, which consists of four dimensions – reward, necessity, confidence, and safety - were factor analyzed based on 25 items altogether. The assumption of factor analysis was met, with KMO value was discovered to be 0.88 and the Bartlett test was highly significant (p<0.05). The attitude constructs produced 5 factors with eigenvalues more than 1, which explained 61.45% of the total variance of the items. The results of Varimax rotation indicated some cross-loadings among the items. Item N5 loaded on 3 factors, and item C1, which loaded on two factors, has a value of factor loading quite similar. Factor 5 was found to have only one item in it (N6). As a normal practice, item N5, C1 and N6 were deleted to improve the scale reliability and decrease the inconsistent correlation among the factors (Hair et al., 1998). With four factors remaining, the final factor loadings of the items were between 0.58 and 0.90, which are larger than the minimum value (> 0.30) needed for a sample size of 350 and above (Hair et al., 1998). The Cronbach’s Alpha value of the reliability analysis for reward (α = 0.86), necessity (α =

0.91), confidence (α = 0.82), and safety (α = 0.78), indicate high reliability. The results of item-to-total correlations demonstrated the deletion of any item would not increase the value of Cronbach’s Alpha; therefore all of the scale items were included. Item S3 from the original safety dimension was included in the confidence dimension.

The results for self-efficacy reveals the KMO value of 0.75 and Bartlett’s test of sphericity was highly significant (p<0.05), validate the factorability of the correlation matrix as recommended by Hair et al. (1998). Principal components analysis indicates that only one factor was extracted with an eigenvalue more than 1. This factor explained 49.51% of the total variance in the items, with the factor loading of 0.51 to 0.80. The value of Cronbach’s Alpha of 0.73 shows an acceptable reliability. Item-to-total correlations demonstrated that elimination of any items would not strengthen the value of Cronbach’s Alpha, thus supporting the inclusion of all scale items.

Table 4: Confirmatory factor analysis results

Construct Scale Items Factor Loadings

Cronbach’s Alpha

Kaiser-Meyer-Olkin

Bartlett’s Test of

Sphericity

% of variance

Willingness to Consume Functional

Food

WC1 0.84 0.81 0.80 0.00 68.28WC2 0.87WC3 0.80WC4 0.80

Reward

R1 0.78 0.86 0.88 0.00 61.45R2 0.77R3 0.69R4 0.58R5 0.71R6 0.69R7 0.59R8 0.61



Correlation Analysis

Correlation analysis was performed to enquire an understanding of the relationship between variables and the correlation coefficient values demonstrated the strength of the relationship. The result in Table 5 indicated that the overall values of the positive correlation among variables were between 0.05 and 0.56, thereby indicating weak and moderate associations between these variables. The correlations amongst the measures of reward, necessity, confidence and safety indicate weak associations between those variables with correlation values ranging from 0.10 to a high of 0.56. The results show that safety is significantly correlated with necessity and confidence with correlation values of 0.34 and 0.16, respectively. The reward is also

significantly correlated with confidence (r= 0.56) and necessity. However, the association with necessity is very weak (r= 0.10). On the other hand, it was found that both reward and safety, and, necessity and confidence were significantly not related. Generally, this suggested that those variables does not lead to each other and should be addressed independently. On the other hand, all of the variables except safety are statistically significantly correlated with consumer willingness to consume functional food with correlation values ranging from 0.23 to a high of 0.54. The results of these correlations indicated that higher consumer willingness to consume functional food is mostly associated with higher scores in reward, confidence, necessity, and self-efficacy.

Necessity

N1 0.88 0.91N2 0.90N3 0.80N4 0.85N7 0.71N8 0.76

ConfidenceC2 0.78 0.82C3 0.80C4 0.77

Safety

S3 0.61S1 0.80 0.78S2 0.80S4 0.74S5 0.68

Self-efficacy

SE1 0.51 0.73 0.75 0.00 49.51SE2 0.80SE3 0.79SE4 0.60SE5 0.77



Table 5: Pearson correlations

Factors WC R N C S SEWillingness to Consume 1.0

Functional Food (WC)

Reward (R) 0.54** 1.0Necessity (N) 0.23** 0.10* 1.0

Confidence (C) 0.37** 0.56** 0.08 1.0Safety (S) 0.05 0.05 0.34** 0.16** 1.0

Self-efficacy (SE) 0.49** 0.52** 0.20** 0.50** 0.11* 1.0

Note: *p< .05; **p< .01

Table 6: Factors influencing consumer willingness to consume functional food

Independent Variables B SE B ßReward 0.43 0.06 0.46**

Necessity 0.09 0.02 0.16**Confidence -0.03 0.04 -0.05

Safety -0.04 0.03 -0.07

Note: R²= 0.53; F= 56.68; Sig. F= 0.00; **p<0.01; B= Unstandardised coefficient beta; SEB= Standard error of regression coefficient; ß= Beta coefficient

Multiple Regression Analysis

Table 6 showed that the relationship between consumer willingness to consume functional food and attitude dimensions was significant (F= 56.68; p< .01). The model showed a moderate relationship with antecedent variables explained 53% of the variation in consumer willingness to consume functional food. The regression equation comprises four variables, and two of them appeared as significant predictors of consumer willingness to consume functional

Hierarchical Regression Analysis

Table 7 reported the results. In model 1, reward and necessity emerged as a predictor of consumer willingness to consume functional food, explaining 32.2% of the variance. Self-efficacy was then, included in step 2 as shown in Model 2. It was reported as an improvement with a significant increase in R-square of 37.1%. In model 3, a significant value of R-square is increased (R²= 0.412) with a significant F

food, which is reward and necessity. Reward and necessity positively influenced consumer willingness to consume functional food. Therefore, hypotheses 1 and 2 were accepted. Whereas, confidence and safety have no significant influence on consumer willingness to consume functional food and, leads to the conclusion that hypotheses 3 and 4 were not supported. The results also explained that reward (ß= 0.46) was a stronger predictor than necessity (ß= 0.16).

change value (R² change= 0.041, p<0.01). This indicated that the self-efficacy explained an increase of 4.1% of the variance in consumer willingness to consume after controlling for reward and necessity. The self-efficacy was statistically significant (ß= 0.783, p<0.01) and moderates the relationship between attitude dimensions (reward and necessity), and consumer willingness to consume functional food. Therefore, hypothesis 5 is accepted.



Table 7: Self-efficacy as a moderator in the relationship between attitude dimensions and consumer willingness to consume functional food

Independent Variable Model 1 Model 2 Model 3Reward 0.520** 0.397** -0.089Necessity 0.184** 0.142** 0.116Moderating Variable : Self-efficacy 0.264** -0.122Interaction Terms : Self-efficacy x Reward 0.783**R² 0.322 0.371 0.412R² change 0.322 0.049 0.041Sig. F change 0.000 0.000 0.000

**p< 0.01

Discussion

It was demonstrated that two out of four attitude dimensions influence consumer willingness to consume functional food; reward and necessity. This shows that Malaysian consumers perceive the consumption of functional food as personally rewarding in terms of health benefits and it is necessary for them to consume functional foods in order to maintain and promote good health and sustainable well-being. In other words, Malaysian consumers believe that functional foods are essential in promoting a healthy lifestyle and believe that people need to consume more functional food to stay healthy and ward off diseases. The result is congruent with Urala and Lahteenmaki (2007), Landstrom et al. (2007), Chen (2011), Carrillo et al. (2013) and Kraus (2015) who reported that consumers showed a positive attitude towards functional foods and regarded the foods as rewarding particularly in health benefits, health promoting and necessary to be consumed. Thus, this shows that Malaysian consumers’ willingness to consume functional food is linked to the strong beliefs and positive perceptions relating to the positive effects and benefits from functional food consumption.

Consequently, functional food consumption reflected the concept of sustainable diets that promoting food sustainability is important in encouraging a healthy diet among

individuals, which leads to a healthy lifestyle, sustainable livelihood, as well as the social and environment sustainability (Johnston et al., 2014). Supported by Bech-Larsen and Scholderer (2007) and Niva (2007), they mention that functional food represents a sustainable category in the food market. Some of the favourable benefits of sustainable diet through functional food consumption can be seen by means of reduced in related chronic diseases, economic sustainability in terms of reducing healthcare cost of individual household and government fund for medical cost, and social sustainability such as healthy lifestyle and increase in physical well-being.

The other two dimensions of attitude; confidence and safety were reported to be of no significance. This may be due to the lack of confidence and trust among Malaysian consumers in the safety of the food and in the claims of its benefits by the manufacturers. For instance, if the manufacturer claims that the functional food can reduce the risk of diseases or promote improved well-being, unfulfilled promises and undetectable effects could create dissatisfaction and lack of confidence among consumers towards such food. This might hinder them from buying and consuming functional food products. Therefore, Malaysian consumers’ impression in functional food safety and benefits would not determine the consumption



of such food. This is in line with Landstrom et al. (2007) who noted that safety of functional food and consumer confidence seems to fail to have an impact on consumer willingness to consume such food. In addition, Chen (2011) reported that perceived safety towards possible harmful effects of functional food does not influence the consumption of functional food in general. However, the result for the construct of confidence is contradicted with the findings of previous studies, which revealed that confidence in the functional food that was said to have positive health effects demonstrated a positive relationship with consumer willingness to consume functional food (Verbeke, 2005; Markovina et al., 2011; Siegrist et al., 2015). Likewise, Urala and Lahteenmaki (2004: 2007) and Chen (2011), reported that confidence in functional foods significantly influences the consumer willingness to consume functional foods after perceived reward.

Furthermore, this study reveals that the reward from using functional food is the strongest predictor of consumer willingness to consume functional food, followed by necessity. This means that the stronger the consumers’ belief in the rewards they will receive from using functional food (for example, promote their well-being, performance, health and mood), the greater their willingness to consume functional food would be. In addition, the strong belief among Malaysian consumers in the necessity to consume functional food in order to stay healthy as well as getting beneficial health rewards has increased their willingness to consume such food. This is consistent with Urala and Lahteenmaki (2007) and Chen (2011) who found that perceived reward and necessity are the most important dimensions of attitude in influencing consumer willingness to consume functional food.

Individuals may originally want to participate in healthy eating; however, many not succeed to achieve this target. Self-efficacy has been known to be an effective factor for behaviour to occur (Conner & Norman, 2005; Mai & Hoffman, 2012; Fitzgerald et al.,

2013; Robles et al., 2014; Hall et al., 2015; Nastaskin & Fiocco, 2015). The moderator of self-efficacy shows a significant positive effect on the relationship between attitude dimensions (reward and necessity) and willingness to consume functional food. This result demonstrates that the internal forces of self-efficacy occur when the individual feels the rewards and the necessity to consume functional food, thus, the chances of them consuming functional food would increase. Consequently, the predictive power of attitude dimensions towards willingness to consume functional food will also increase. In other words, the findings explain that the positive relationship between attitude dimension (reward and necessity) and consumer willingness to consume functional food will be stronger when their internal force of self-efficacy is high. Therefore, a strong self-efficacy among individual will determine the increase of healthy foods consumption among consumers.

The research findings show that feeling rewarded from the consumption of functional food, belief in the necessity for functional food to be consumed in order to be healthy, and having strong self-efficacy are important in influencing the behavioural intention towards functional food consumption among Malaysian consumers. Therefore, to formulate effective advertising and promotional strategies, marketers should emphasise the rewards that consumers will obtain from the consumption of functional foods, for example, lowering the cholesterol and blood pressure levels, and many other health advantages. In addition, marketers should provide an encouraging testimony and scientifically proven messages about health claims of functional food products to reduce the lack of trust and confidence among Malaysian consumers towards the health related product, especially concerning the health claims and side effect. Indirect monitoring, and verbal encouragement and reinforcement can increase an individual’s self-efficacy (Bandura, 1986). Thus, when applied to healthy consumption behaviour, an individual could be guided on the



steps required in avoiding illness through the concept of sustainable diets, such as consuming more functional food, fruits and vegetables and avoiding eating oily and fatty food. In addition, campaigns and advertisements relating to the benefits of consuming a healthy diet or the negative effects of not consuming a healthy diet should be continuously exposed to the target population.

In the planning of tailored health communication interventions towards social sustainability such as healthy lifestyle through improving sustainable dietary behaviour among Malaysian, marketers and the government, particularly the Ministry of Health, can develop the promising messages in the advertisement, media and health educational campaign to be delivered by the appropriate sources such as nutritionists, dieticians, physicians and trusted celebrities. The interest in losing weight, to maintain the ideal body figure and a captivating self-image among individuals becomes more important than being healthy when making a decision about food choice. Therefore, the government should also emphasise programmes that promote healthy body images and ideal weight rather than focusing on preventing diseases, especially for women and schoolgirls. In addition, Malaysian government should play its part in initiating policies to ensure that healthy food including functional food is available, accessible, sustainable and affordable for Malaysians to buy because this will facilitate and encourage consumers towards practising healthy consumption.

The limitations of this study may include the selection of samples. The collection of data was confined to only one area, the Klang Valley. These features may not be reflective of the overall population in Malaysia. Thus, the results cannot be used to generalise to the whole population of Malaysia. Future studies should, therefore, be extended with data collection in another part of states to portray the real picture of functional food consumption in Malaysia. This study explored the Malaysian consumers’ behaviour towards functional food in general.

Therefore, questions referring to all the constructs in this study referred to the general concept of functional food without focusing on different functional food categories. Thus, for future research, the study should focus on specific functional food categories with separate unique components that could cure different health problems or reduce the risk of different disease, because when consumers are making a choice towards functional foods, their reasons maybe dissimilar within the separate food categories.

In addition, the data in this study were obtained randomly from adult consumers aged 18 and above. Other segments, such as teenagers and university students, who may have different sustainable food consumption behaviour, should be investigated. Additional studies comparing consumption behaviour towards healthy food from the different region of Malaysia might produce different findings. It is also recommended that to investigate the impact of functional food processes and consumption on sustainable livelihood, socio-economic and environmental sustainability. Despite the limitations, the results and recommendations are important to consumer behavioural domain of sustainable food category, functional food industry and academic perspective.

Conclusion

The study was conducted to examine the influence of attitude dimensions (i.e. reward, necessity, confidence, and safety) on consumer willingness to consume functional food among Malaysian consumers, and at the same time to examine the moderating effect of self-efficacy on the relationship between attitude dimensions and consumer willingness to consume functional food.

With regards to the factors influencing consumer willingness to consume functional food among Malaysian consumers, it was found that the reward and the necessity are the attitude dimensions that influence the consumer willingness to consume functional food in their



daily diet. It can be concluded that consumer willingness to consume functional food appears to be facilitated by the perceived reward and necessity. In other words, Malaysian consumers believe it is necessary and important for them to consume functional foods in order to obtain the positive reward in terms of good health, well-being and sustainable livelihood. The result also shows that the reward from using functional food was the strongest predictor of the consumer willingness to consume functional food, followed by the necessity. Another two dimensions (i.e. confidence and safety) appear to have no significant relationship.

It appears that the relationship between attitude dimensions (reward and necessity) and the consumer willingness to consume functional food is improved with the presence of the self-efficacy as a moderator. This shows that a strong self-efficacy that internally motivates consumers to be healthy drives them to seek for healthy food, which they believe are associated with health benefits, wellness and longevity. This reveals that the strong self-efficacy that Malaysian consumers have does translate into healthy consumption.

The finding would shed further light on the knowledge in the area of functional food consumption behaviour, especially when it comes to discussing the relationship between attitude dimensions, self-efficacy and consumer willingness to consume functional food. Thus, marketers and the government can use this information in promoting functional food buying and consumption, as well as developing strategies and promotion campaigns towards practicing a healthy diet and a healthy lifestyle among Malaysians.

Acknowledgements

The author acknowledges a financial support for her Doctor of Business Administration study under Skim Latihan Akademik IPTA Fellowship from Ministry of Higher Education of Malaysia.

References

Annunziata, A., & Vecchio, R. (2011). Factors affecting Italian consumer attitudes toward functional foods. AgBioForum, 14(1): 20-32.

Armitage, C. J., & Conner, M. (1999). Distinguishing perceptions of control from self-efficacy: Predicting consumption of a low-fat diet using the theory of planned behaviour. Journal of Applied Social Psychology, 29(1): 72-90.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall, Englewood Cliffs, New Jersey. 617 pp.

Barcellos, M. D., & Lionello, R. L. (2011). Consumer market for functional foods in South Brazil. International Journal Food System Dynamics, 2(2): 126-144.

Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personal Social Psychology, 51(6): 1173-1182.

Bech-Larsen, T., & Grunert, K. G. (2003). The perceived healthiness of functional foods: A conjoint study of Danish, Finnish and American consumers’ perception of functional foods. Appetite, 40: 9-14.

Bech-Larsen, T., & Scholderer, J. (2007). Functional foods in Europe: Consumer research, market experiences and regulatory aspects. Trends in Food Science & Technology, 18: 231-234.

Carrillo, E., Fiszman, S., Prado-Gasco, V., Varela, P. (2013). Why buying functional foods? Understanding spending behaviour through structural equation modelling. Food Research



International, 60(1): 361-368.

Chen, M. F. (2011). The joint moderating effect of health consciousness and healthy lifestyle on consumers’ willingness to use functional foods in Taiwan. Appetite, 57(1): 253–262.

Conner, M., & Norman, P. (2005). Predicting health behaviour: Research and practice with social cognition models, 2nd edition. Open University Press, Maidenhead, Berkshire. 402 pp.

Fitzgerald, A., Heary, C., Kelly, C., Nixon, E., & Shevlin, M. (2013). Self-efficacy for healthy eating and peer support for unhealthy eating are associated with adolescent’s food intake patterns. Appetite, 63: 48-58.

Frewer, L., Scholderer, J., & Lambert, N. (2003). Consumer acceptance of functional food: Issues for the future. British Food Journal, 105(10): 714-731.

Hair, J. F. Jr., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate data analysis, 5th edition. Prentice Hall, Upper Saddle River, New Jersey. 767 pp.

Hall, E., Chai, W., Koszewski, W., & Albrecht, J. (2015). Self-efficacy is associated with healthy eating behaviours for fifth grade students. FASEB Journal, 29(1): 911-915.

Howlett, J. (2008). Functional foods: From science to health and claims. ILSI Europe, Brussels. 37 pp.

Johnston, J. L., Fanzo, J. C., & Cogill, B (2014). Understanding sustainable diets: A descriptive analysis of the determinants and processes that influence diets and their impact on health, food security, and environmental sustainability. American Society for Nutrition. Adv. Nutr, 5(4): 418-429.

Kraus, A. (2015). Development of functional food with the participation of the consumer: Motivators for consumption of functional products. International Journal of Consumer Studies, 39(1): 2-11.

Krejcie, R. V., & Morgan, D. W. (1970). Determining sample size for research activities. Educational and Psychological Measurement, 30: 607-610.

Labrecque, J., & Charlebois, S. (2011). Functional foods: An empirical study on perceived health benefits in relation to pre-purchase intentions. Nutrition & Food Science, 41(5): 308-318.

Landstrom, E., Hursti, U. K., Becker, W., & Magnusson, M. (2007). Use of functional foods among Swedish consumers is related to health-consciousness and perceived effect.The British Journal of Nutrition, 98(5): 1058-1069.

Landstrom, E., Hursti, U. K., & Magnusson, M. (2009). Functional foods compensate for an unhealthy lifestyle. Some Swedish consumers’ impressions and perceived need of functional foods. Appetite, 53(1): 34-43.

Mai, R., & Hoffman, S. (2012). Taste lovers versus nutrition fact seekers: How health consciousness and self-efficacy determine the way consumers choose food products. Journal of Consumer Behavior, 11: 316-328.

Markovina, J., Cacic, J., Kljusuric, J. G., & Kovacic, D. (2011). Young consumers’ perception of functional foods in Croatia. British Food Journal, 113(1): 7-16.

Methakornkulnan, P., Hsieh, Y. J., & Huang, L. Y. (2013). Consumer market for functional foods in Thailand. Paper



presented at The Asian Conference on Psychology & the Behavioural Science 2013, Osaka, Japan. Retrieved from http://iafor.org/archives/offprints/acp2013-offprints/ACP2013_0146.pdf, 9 January 2017.

Nastaskin, R. S., & Fiocco, A. J. (2015). A survey of diet self-efficacy and food intake in students with high and low perceived stress. Nutrition Journal, 14(42): 1-8.

Niva, M. (2007). All foods affect health: Understandings of functional foods and healthy eating among health-oriented Finns. Appetite, 48: 384-393.

Nor Azila, M. N., Zuriana, M. J., Norsiah, M., Norazwa, M., Azilah K., Azli M., & Hayatul Safrah, S. (2012). The relationships between environmental knowledge, environmental attitude and subjective norms on Malaysian consumers green purchase behaviour. Malaysian Journal of Consumer and Family Economics, 15: 1-20.

Pereira, M. A., O’reilly, E., Augustsson, K., Fraser, G. E., Goldbourt, U., Heitmann, B. L., Hallmans, G., Knekt, P., Liu, S., Pietinen, P., Spiegelman, D., Stevens, J., Virtamo, J., Willett, W. C., & Ascherio, A. (2004). Dietary fibre and risk of coronary disease: A pooled analysis of cohort studies. Archives of International Medicine, 164(4): 370-376.

Poulsen, J. (1999). Danish consumers’ attitudes towards functional foods. Retrieved from https://ideas.repec.org/p/hhb/aarmap/0062.html, 3 March 2017.

Roberfroid, M. B. (2000). A European consensus of scientific concepts of functional foods. Nutrition, 16: 689-691.

Robles, B., Smith, L. V., Ponce, M., Piron, J., & Kuo, T. (2014). The influence of gender and self-efficacy on healthy eating in a low-income urban population affected by structural changes to the food environment. Journal of Obesity, 2014(908391): 1-12.

Sekaran, U., & Bougie, R. (2009). Research methods for business: A skill building approach, 5th edition. Wiley, West Sussex. 468 pp.

Siegrist, M., Shi, J., Giusto, A., & Hartmann, C. (2015). Worlds apart. Consumer acceptance of functional foods and beverages in Germany and China. Appetite, 92: 87-93.

Urala, N., & Lahteenmaki, L. (2004). Attitudes behind consumers’ willingness to use functional foods. Food Quality & Preference, 15: 793-803.

Urala, N., & Lahteenmaki, L. (2007). Consumers’ changing attitudes towards functional foods. Food Quality and Preference, 18: 1-12.

Verbeke, W. (2005). Consumer acceptance of functional foods: Socio-demographic, cognitive and attitudinal determinants. Food Quality & Preference, 16(1): 45-57.

Verbeke, W., & Vackier, I. (2005). Individual determinants of fish consumption: Application of the theory of planned behaviour. Appetite, 44(1): 67-82.

World Health Organization (2015). Non-communicable diseases. Fact Sheet 355. Retrieved from http://www.who.int/mediacentre/factsheets/fs355/en/, 20 August 2016.




ENHANCING ENVIRONMENTAL SUSTAINABILITY OVER FISHERIES INDUSTRY THROUGH PROACTIVE RISK EVALUATION: A CASE OF TOK BALI FISHING PORT

NURUL HAQIMIN MOHD SALLEH* AND MOHD AZLAN ABDUL HALIM

School of Maritime Business and Management, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia


Abstract: The fisheries industry is a vital industry for Malaysia’s economy. There are growth potentials in deep-sea fishing, coastal fishing and aquaculture activities carried out along the Malaysian coast. However, the development of the fisheries industry threatens the sustainability of an environment if it is not managed systematically. The environmental risk of the fisheries industry can result in water and air pollutions, and leading to catastrophic ecosystem damages. As such, a new model is proposed for analysing environmental risks arising from fisheries activities. With this in view, this paper proposes a new analysis model using two methods which are Analytic Hierarchy Process and decision-making rating tool. Environmental risk factors are identified and prioritised by using the Analytic Hierarchy Process method. Mitigation strategies for minimising the risk losses from fisheries activities are suggested and highlighted by using a decision-making rating tool. To demonstrate the applicability of this proactive risk evaluation model, Tok Bali fishing Port is selected as a case study. The result shows that water pollution is a focal environmental risk arising from fisheries activities in Tok Bali Fishing Port. The proposed model is capable of helping related government agencies such as Lembaga Kemajuan Ikan Malaysia and Department of Fisheries Malaysia in conducting a proactive evaluation to enhance the environmental sustainability of the fisheries industry.

Keywords: Environmental sustainability, fisheries industry, risk evaluation, analytic hierarchy process.

Introduction

The fisheries industry is an important sector in Malaysia for the supply of animal protein (Food and Agriculture Organization, 2017). There are more than 10 types of fishing gears used in catching over 100 commercial marine fish species in the Malaysian waters (Samsudin et al., 2015). From these types of fishing gears, the two most efficient are trawlers and purse seines. In 2016, the production of marine fish from Malaysian waters was 1,574,447 metric tons with a value of RM10.176 billion (Department of Fisheries Malaysia, 2017). Since fish is considered as an importance source of protein for majority of the Malaysian population, the necessity to manage the fisheries industry sustainably is becoming crucial. Moreover, the development of the fisheries industry threatens

the sustainability of the environment, if, it is not managed systematically. Consequently, the environmental risk from fisheries industry can results in water pollution, air pollution, and ecosystem damages (GESAMP, 1991; Malaysia Fisheries Act, 1985; International Maritime Organization, 2009; 2014). The impact of these pollution and damages from fisheries industry can be worst if no action is taken to manage it systematically. As a result, this paper proposes a new model for analysing the environmental risk arising from fisheries activity and consists of three main areas which are water pollution, air pollution and ecosystem damages.

In order to demonstrate the applicability of this proactive risk evaluation model, Tok Bali Fishing Port is selected as a case study. This model is able to help related government

: 53-65

54ENHANCING ENVIRONMENTAL SUSTAINABILITY OVER FISHERIES INDUSTRY THROUGH PROACTIVE RISK EVALUATION: A CASE OF TOK BALI FISHING PORT

agencies such as Lembaga Kemajuan Ikan Malaysia and the Department of Fisheries Malaysia to conduct a proactive evaluation for enhancing the environmental sustainability of the fisheries industry. Practitioners and academicians will benefit the proposed methodology for analysing the environmental impact of the fisheries industry in a particular area. This paper presents a basic concept for making decision under uncertainty and explains the risk identification, the prioritisation and the mitigation for the environmental sustainability at Tok Bali Fishing Port. The main objective of this paper is to identify environmental risk factors in the fisheries industry and to prioritize them by using the Analytic Hierarchy Process (AHP) method. Finally, mitigation strategies for minimising the environmental risk in the fisheries industry are suggested and highlighted by using a decision making rating tool.

The Sustainability of Fisheries Industry

Fisheries sector in Malaysia is significant contributor to the Gross Domestic Product (GDP), and fish products are a main source of food and protein for the Malaysia population. This sector contributed around 1% to the Malaysian GDP, underlining its importance in this country (Department of Fisheries Malaysia, 2017). The fish demand is forecasted to increase from 1.57 million metric tons in 2016 to 1.9 million metric tons by the year 2020 (Department of Fisheries Malaysia, 2017). It is noteworthy to mention that the domestic and international demand for fish and fish products are growing year after year, but the captures are stagnant. This negative correlation is caused by several factors such as overfishing, ecosystem damages and pollutions.

The sustainability theory in the fisheries industry refers to the theoretical disciplines involving population dynamics of fisheries and practical fishing strategies (National Geographic, 2016). The purpose of this theory is to avoid overfishing and pollution arising from fishing activities. Several techniques

are recommended such as individual fishing quotas, curtailing destructive, illegal fishing enforcement and environmental risk protection (National Geographic, 2016). Crawford Global Technical Services (2016) defined environmental risk as the actual or potential threat of adverse effect on living organisms and the environment by effluents, emissions, wastes and resources depletion. Environmental risk exposures, whether physical, chemical, or biological, can induce a harmful impact and may affect soil, water, air, natural resources or entire ecosystem, as well as on plants and animals (US Environmental Protection Agency, 2017).

Water Pollution, Air Pollution and Ecosystem Damages

The environmental risk is evaluated by considering three main criteria which are water pollution, air pollution and ecosystem damage. Although, these risk criteria have originated from different sources, it is essential to consider them in a single model before proposing mitigation strategies, Moreover, the decision-makers are assisted with a single model that is capable of dealing with multi-elements. The classification of these three criteria follows the recommendations and adaptation of various studies.

Water pollution can be defined in many ways. Usually, it means one or more substances that have built up in water that can cause problems to people, animals and plants. Naturally, oceans, lakes, rivers and other inland waters can clean up a certain amount of pollution by dispersing it harmlessly. However, water pollution caused by oil spill from vessels (e.g. fishing and merchant), waste garbage and from waste from agriculture is difficult to be cleaned and almost impossible to be dispersed naturally. In fishing sector, cleaning boats and oil disposal are some of activities that caused water pollution. From these activities, sea water surface is covered and polluted with discharged oil. In addition, poor disposal facilities at fishing port also contribute to low awareness among


55Nurul Haqimin Mohd Salleh et al.

fishermen. Sea pollution is the problem faced by coastal communities, marine animals and plants. This pollution was introduced by human, whether directly or indirectly into the marine environment resulting in deleterious effects, such as harms to living resources, hazard to human health, hindrance to marine activities including fishing, impairment of quality for use of sea water and reduction of amenities (GESAMP, 1991).

Air pollution can result from human and natural actions. However, pollution from natural occurrences is not frequent compared to human actions. For example, an emission from vessels are a significant source of air pollution and it is estimated that carbon dioxide emission from various types of ships (i.e. currently were equal to 2.2% of the global human-made emission in 2012) will arise by as much as two to three times by 2050 if no action is taken (International Maritime Organization, 2009; 2014). Although, there is a perception that fishing vessel contributes a lower amount of air pollution, it still needs to be evaluated. Based on expert consultation, the use of chlorofluorocarbon (CFC) and jetty extension area by burning mangrove and ‘nipah’ trees are considered as the causes of air pollution instigated by fishing activities.

Fisheries industry can cause ecosystem damages if this sector is poorly controlled. Excessive fishing capacity, overfishing and oil discharge are some instigators of ecosystem damages. Oil spill is harmful especially for animals like sea birds when it swims on the surface of the water. Malaysia Fisheries Act 1985 argued that ecosystem damages are intended when harvesting extreme amount of fish, using prohibited or banned fishing gears, smuggling and intrusion of foreign fishermen, and damaging coral reef ecosystems or interfere with the natural life. As a result, it is worth mentioning that ecosystem damages in the fishery activities can be instigated by oil spill from fishing vessels, overfishing and fishery sewages.

Analytic Hierarchy Process (AHP)

AHP is a theory of measurement through pair-wise comparisons and relies on the judgments of experts to derive priority scales (Mohd Salleh et al., 2015). The comparisons are made using a scale of absolute judgments that represents the relative importance of one element to another element in a given attribute. The fundamental scale has been shown to be the one that captures individual preferences with respect to quantitative and qualitative attributes as well as, or better than, other scales (Saaty, 1980). In addition, the AHP approach is a well-structured approach for organizing and analysing complex decisions such as multilevel criteria or attributes. It has been developed based on precise mathematical structures of consistent matrices and their associated right-eigenvector’s ability to generate weights (Saaty, 1980). The AHP approach also has the capability of measuring inconsistencies and improving judgments (Saaty, 2008).

Several researchers have conducted risk assessment in the fisheries industry (Fletcher, 2005; Kolar et al., 2007; Ellis et al., 2009; Copping et al., 2015; Huchim-Lara et al., 2016). None of them however, deal with the assessment of environmental risks arising from fisheries industry, which highlights a significant research gap to be fulfilled. In addition, the application of the AHP approach in this risk assessment model is a new attempt since no studies have been found that employ this method.

Methodology

For the evaluation of the environmental risk in the fisheries industry, two methods are employed. Firstly, environmental risk factors are identified and prioritised by using the AHP method. Secondly, the mitigation strategies for minimising the risk losses from the fisheries industry are suggested and highlighted by using a decision-making rating tool. To develop the calculation process for prioritising the environmental risk factors in the fisheries industry, a flow chart of proposed methodology in sequential order is illustrated in Figure 1.



Figure 1: Methodological framework for analysing the environmental risk in the fisheries industry

Identification of the Environmental Risk Factors in the Fisheries Industry (Step 1)

As shown in Table 1, the risk factors are identified

through literature and further consulted with an expert.

Table 1: The summary of identified environmental risk criteria

Main Criteria Sub-criteria

Water PollutionTrash DumpingFishery Sewage DumpingOil Spill/Disposal

Air PollutionEmissions from the Use of Fossil FuelBurning Activities for Extension Area (Jetty Opening)The Use of Chlorofluorocarbon

Ecosystem DamageOil Spill/DisposalOverfishingFishery Sewages Discharging



Development of a Generic Model for the Proactive Risk Factors in the Fisheries Industry on Environmental (Step 2)

The kernel of developing a generic model is that it can be modified or adjusted to be used for a particular firm or industry (Mohd Salleh et al., 2015). The risk factors, as shown in Table 1 are used for developing a generic model in a hierarchical structure form. In step 2, the decision hierarchy from the top with the goal of the decision, followed by the intermediate level to the lowest level, is structured. As shown in the Figure 2, the environmental risk instigated

by fishery industry are classified into three main categories (i.e. as discussed in step 1), which are water pollution, air pollution, and ecosystem damages. The water pollution is caused by oil spill/ disposal, trash dumping and fishery sewage dumping. The second category is air pollution, instigated from burning activities for extension area, emissions from the use of fossil fuel, and the use of chlorofluorocarbon (CFC). The third category is ecosystem damages, triggered by fishery sewages discharging, oil spill/ discharge, and overfishing.

Figure 2: The generic environmental risk model in the fisheries industry


Establishment of Criteria Weight (Step 3)

A weight can be assigned to each criterion using established methods such as simple rating method or more elaborate methods based on pair-wise comparisons (Mohd Salleh et al., 2014). To compare the criteria or alternatives in a nature pair-wise comparisons mode, a fundamental scale of absolute number is used. Table 2 shows a fundamental scale of the ratio scale of pair-wise comparisons which consist of linguistic meaning and numerical assessment.

From this Table 2, expert should understand the ratio scale of the pair-wise comparisons before the assessment has been taken to avoid misjudgement. Qualitative data collections are used in this study and all necessary data are obtained from expert judgment and mathematical algorithms. All experts contributed their ideas and opinions in developing a scientific model, answering a set of pair-wise comparison as well as determining the parameters.


To determine the consideration pair of attributes Ai and Aj is presented by an n x n matrix d. The entries aij that are determined by rules of entry as follows (Saaty, 1980):

I. Rule 1: If aij = α, then aji = 1/α, α ≠ 0.

The calculated judgment of comparison on pair (Ai, Aj) is renowned as aij in the matrix D; next step is to allocate the weight vector for each criterion or alternative, as it indicates the

The decision may be inconsistent due to different views and beliefs with regard to criteria. By using a Consistency Ratio (CR), inconsistency of the pair-wise comparisons can be calculated. If CR value is 0.10 or less, the consistency of the pair-wise comparison is considered acceptable, and the AHP can continue with the computations

II. Rule 2: If Ai is judged to be of equal relative importance as Aj, then aij = aji =1.

According to above rules the matrix D is shown in Equation 1 as follows:

prioritisation of the criteria (Mohd Salleh et al., 2015). A weight value wk can be calculated by using Equation 2 as follows:

of weight vectors (Andersen et al. 2011; Riahi et al. 2012; Mohd Salleh et al., 2015). In contrast, a CR with a greater value than 0.10 indicates an inconsistency in the pair-wise judgements. To check the consistency of the judgements, a CR is computed by using Equations 3-5 (Andersen et al. 2011):

Table 2: The ratio scale of pair-wise comparisons (Mohd Salleh et al., 2015)

Numerical Assessment (Scale) Linguistic Meaning1 Equally Important (EQ)3 Weakly Important (WE)5 Strongly Important (ST)7 Very Strongly Important (VS)9 Extremely Important (EX)

2, 4, 6, 8 Intermediate Values Between The Two Adjacent Judgements



where CI is the consistency index, RI is the average random index (Table 3), n is the number

Case Study

To test the applicability of this evaluation model, Tok Bali Fishing Port has been selected as a case study. Tok Bali, which is located at Pasir Puteh, Kelantan, is intended to provide convenience to deep-sea fishermen in Kelantan and the surrounding area such as Terengganu and Pahang. Tok Bali is the most important fishing port to the State of Kelantan. The fisheries industry at Tok Bali has a potential prospect and become a vital industry for Malaysia’s economy in the future. This growth refers to the development of deep-sea fishing, coastal fishing and aquaculture activities carried out along the Malaysian coast.

To deal with qualitative data, 10 domain experts were approached to perform the pair-wise comparisons and to assign the appropriate grade for every environmental risk instigated by fishery activities. However, only five experts were able to commit during the interview session. Under the AHP approach, five experts

of items being compared, and is the maximum weight value of the n x n comparison matrix D.

are acceptable for further evaluation as there are studies that only used three experts (Mokhtari et al., 2012; Mohd Salleh et al., 2015). Selection of domain experts for subjective judgments was based on their qualifications and experiences of more than 15 years in the fisheries industry. The five domain experts are managers from Lembaga Kemajuan Ikan Malaysia (LKIM) at Tok Bali, fishing companies located at Tok Bali and Fisherman’s Association of Semerak Area. The methods of evaluation were conducted through interviews between the researcher and the domain experts, who are guided with an evaluation form which was given to them during the interviews session. The average time taken during the interview session is 70 minutes.

Due to difficulties in assigning weight for experts and to avoid prejudgement, they are assigned with equal weight. To obtain the aggregated comparisons matrices, geometric mean is used in this study to aggregate the judgement of experts (Equation 6).


N 1 2 3 4 5 6 7 8 9 10RI 0 0 0.58 0.9 1.12 1.24 1.32 1.141 1.45 1.49



Water Pollution Ecosystem Damage Air PollutionWater Pollution 1 1.241 6.119

Ecosystem Damage 0.806 1 2.804Air Pollution 0.162 0.357 1

where k is the number of experts and stands for the kth expert opinion for relative importance of the ith criterion to the jth criterion.

To demonstrate the aggregation process by using the GM, based on data obtained, a judgment (i.e. from five experts) of the importance of the criterion “Water Pollution” to the criterion “Ecosystem Damage” is calculated as (1/3 × 3

× 1/3 × 3 × 3)1/5 = 1.241. “Water Pollution” to the “Air Pollution” is aggregated as (5 × 7 × 7 × 7 × 5)1/5 = 6.119. Lastly, the importance of the criterion “Ecosystem Damage” to the “Air Pollution” is aggregated as (7 × 1/3 × 5 × 5 × 3)1/5 = 2.804. As a result, the matrix D for main criteria (i.e. water pollution, ecosystem damage and air pollution) is shown in Table 4.

kije


Based on Equation 2, weight calculations for main criteria are demonstrated as follows:

where ww is the weight for water pollution, wE and wA are the weights for ecosystem damage and air pollution. As a result, water pollution, ecosystem damage and air pollution are calculated as 0.534, 0.359, and 0.107. A further

The CR value for main criteria is 0.029. Saaty (1980) stated that a CR ≥ 0.1 indicates that the judgment is acceptable. As a result, the consistency of the pair-wise comparisons for main criteria is acceptable. The same calculation procedure is applied to rank the sub-criteria in the environmental risk evaluation model. The weight values and consistency ratio values for sub-criteria are shown in Table 5.

Results and Discussion

Based on Table 5, the result has shown that water pollution is the main environmental risk

Based on Table 3, RI for three criteria is 0.580. As a result, the CR is computed as follows:

W=(1×0.534)+(1.241×0.359)+(6.119×0.107)=1.634E=(0.806×0.534)+(1×0.359)+(2.804×0.107)=1.089A=(0.163×0.534)+(0.357×0.359)+(1×0.107)=0.322

step is to calculate and check the CR of the pair-wise comparisons. Firstly, λmax is calculated as to lead the RI and CR. Based on Equations 3-5, the CR is calculated as follows:

from fisheries activities in Tok Bali, followed by ecosystem damage and air pollution. Based on the AHP calculation, the weight for water pollution is calculated as 0.534, followed by ecosystem damage (0.359), and air pollution (0.107). It is noteworthy to mention that the weight for water pollution is closely associated with the perspective of experts’ judgment (questionnaires). They believed that fisheries activities and water pollution are directly proportional and are mostly directly linked.



Based on global weights as shown in Table 6, the most significant lowest level criteria are found to the water pollution instigated by oil spill or discharged (0.292), followed by the ecosystem instigated by oil spill/discharged (0.253) and

Table 5: Result of weight values and consistency ratios for all main and sub-criteria

Main Criteria Weight Sub-criteria Local Weights Global Weights

Water Pollution 0.534Oil Spill 0.546 0.292

Trash 0.137 0.073Fishery Sewage 0.121 0.065

CR = 0.058

Ecosystem Damage 0.359

Oil Spill 0.705 0.253Fishery Sewage 0.151 0.054

Over Fishing 0.145 0.052

CR = 0.029

Air Pollution 0.107

Using Fossil Fuel 0.705 0.075Extension Area 0.169 0.018

Using Chlorofluorocarbon 0.126 0.013

CR = 0.0009

Table 6: Ranking orders of the lowest-level criteria

Lowest-level Criteria Global Weight RankWater Pollution Instigated by Oil Spill 0.292 1

Ecosystem Instigated by Oil Spill 0.253 2Air Pollution Instigated by Using Fossil Fuel 0.075 3

Water Pollution Instigated by Trash 0.073 4Lowest-level Criteria Global Weight Rank

Water Pollution Instigated by Sewage 0.065 5Ecosystem Damage Instigated by Fishery Sewage 0.054 6

Ecosystem Damage Instigated by Over Fishing 0.052 7Air Pollution Instigated by Extension Area 0.018 8

Air Pollution Instigated by Using Chlorofluorocarbon 0.013 9

air pollution due to the emissions from the use of fossil fuel (0.075). The ranking orders of all other lowest-level criteria are also shown in Table 6.



The AHP results have shown that oil spill or discharge from fishing vessel are the main cause of environmental risk in Tok Bali Fishing Port. It is noteworthy to mention that the impact of oil spills from the fishery sector ranging from sea water pollution to the contamination of marine catches. These negative impacts lead to substantial losses if no proper practices are taken in fishing routines. Several useful practices that can be applied to minimise the risk of polluting the water are to avoid overfill the tank when refuelling, ensure the portable tanks are properly capped and locate them away from water. Fishermen should also use proper oil disposal facilities at fishing port, avoid pumping the polluted water into the sea, cover the engine and gearbox with a dip tray to ensure oil entering into the bilge and empty it regularly. Finally, avoid

The AHP results have shown that oil spill or discharge from fishing vessel are the main cause of environmental risk in Tok Bali Fishing Port. It is noteworthy to mention that the impact of oil spills from the fishery sector ranging from sea water pollution to the contamination of marine catches. These negative impacts lead to substantial losses if no proper practices are taken in fishing routines. Several useful practices that can be applied to minimise the risk of polluting the water are to avoid overfill the tank when refuelling, ensure the portable tanks are properly capped and locate them away from water.

Figure 3: The long-term strategies for enhancing environmental sustainability over fisheries industry

overuse detergent as it can cause water pollution too. In addition, effective contingency plans can be arranged to address oil spill response measures, prompt action can be taken to reduce the impact of oil spills, if water pollution occurs. Fishermen should be informed and trained on how to react when water pollution occurs.

To minimise the losses from overall risk causes, long term strategies are suggested. Based on interviews with the domain experts, three general strategies are recommended. As shown in Figure 3, the strategies are law enforcement, strategic fisheries management practices and environmental education to the fishermen. These strategies are discussed as follows:

Fishermen should also use proper oil disposal facilities at fishing port, avoid pumping the polluted water into the sea, cover the engine and gearbox with a dip tray to ensure oil entering into the bilge and empty it regularly. Finally, avoid overuse detergent as it can cause water pollution too. In addition, effective contingency plans can be arranged to address oil spill response measures, prompt action can be taken to reduce the impact of oil spills, if water pollution occurs. Fishermen should be informed and trained on how to react when water pollution occurs.



Conclusions

It is noteworthy to mention that fisheries industry and marine environment provide the important source (i.e. protein) for the majority of Malaysians. As a result, Malaysia fisheries industry aims to increase its production by one third from the current performance. However, since the fishing industry is one of the pollution contributors to the surrounding areas, the environmental risk and impact from this industry need to be proactively analysed and assessed by using an appropriate approach, thus enabling informed decisions to be made regarding mitigation strategies. In this paper, the proactive environmental risk model and decision support framework for prioritising

the environmental risk criteria are proposed. The proposed proactive risk evaluation model is tested at Tok Bali Fishing Port as a case study. The result has shown that water pollution is the key environmental risk from fisheries activities at Tok Bali. In addition the most effective strategies to enhance the sustainability of fishery industry at Tok Bali is found to be environmental law and regulation enforcement, followed by strategic fisheries management practices and environmental awareness to the fishermen. The proposed model can help related government agencies such as LKIM and Department of Fisheries Malaysia to conduct a proactive evaluation in order to enhance the environmental sustainability of the fisheries industry. For future research, a quantitative risk

To minimise the losses from overall risk causes, long term strategies are suggested. Based on interviews with the domain experts, three general strategies are recommended. As shown in Figure 3, the strategies are law

Based on the assessment by using decision making-rating tool, Table 7 shows environmental law and regulation enforcement is the best strategy that possibly to minimise the environmental pollution in the fishing activities at Tok Bali Fishing Port. Based on decision making rating tool, the implementation of “Environmental Law and Regulation Enforcement” is assessed as 5.069, indicates

Figure 4: The effective scales for suggested strategies

enforcement, strategic fisheries management practices and environmental education to the fishermen. These strategies are discussed as follows:

that this strategy is the best option compared to “Strategic Fisheries Management Practices” (4.041) and “Environmental Awareness to the Fishermen” (3.543). However, since the result values of these three strategies are not different from each other, it is suggested that these strategies can be implemented together for better impact.

PollutionStrategy

Water pollution

Ecosystem damage Air pollution Effectiveness

ValueEnvironmental Law and Regulation

Enforcement 3.097 1.651 0.321 5.069

Strategic Fisheries Management Practices 2.456 1.221 0.364 4.041Environmental Awareness to the Fishermen 1.816 1.149 0.578 3.543



evaluation for assessing the targeted or whole criteria in the environmental risk model can be implemented. Several mathematical methods such as Fuzzy Rule-Based and Evidential Reasoning for the assessment can be employed. Acknowledgements

The authors acknowledge the helpful comments of anonymous reviewers and participants of the 4th International Conference on Accounting, Business and Economics (ICABEC 2016) and the valuable suggestions of anonymous referees. The authors would also like to thank the editors of the Journal of Sustainability Science and Management for their helpful assistance throughout the preparation of this manuscript.

References

Andersen, D. R., Sweeney, D. J., Williams, T. A., & Martin, K. (2011). An introduction to management science: Quantitative approaches to decision making, 13th

edition. South-Western, Mason, Ohio. 896 pp.

Copping, A., Hanna, L., Van Cleve, B., Blake, K., & Anderson, R. M. (2015). Environmental risk evaluation system—an approach to ranking risk of ocean energy development on coastal and estuarine environments. Estuaries and Coasts, 38(1): 287-302.

Crawford Global Technical Services (2016). Environmental risk: Defined. Retrieved from https://crawfordgts.com/services/environmental-risk/environmental-risk-defined.aspx, 15 April 2017.

Department of Fisheries Malaysia (2017). Perangkaan Perikanan Tahunan 2016. Retrieved from http://www.dof.gov.my, 15 April 2017.

Ellis, B. J., Figueredo, A. J., Brumbach, B. H., & Schlomer, G. L. (2009). Fundamental dimensions of environmental risk: The impact of harsh versus unpredictable

environments on the evolution and development of life history strategies. Human Nature, 20(2): 204-268.

Fletcher, W. J. (2005). The application of qualitative risk assessment methodology to prioritize issues for fisheries management. ICES Journal of Marine Science: Journal du Conseil, 62: 1576-1587.

Food and Agriculture Organization (2017). Fishery and Aquaculture Country Profiles:Malaysia. Retrieved from http://www.fao.org, 9 July 2017.

GESAMP (1991). Global strategies for marines environmental protection. London, United Kingdom: GESAMP.

Huchim-Lara, R. O., Salas, S., Fraga, J., Mendez, N., & Chin, W. (2016). Fishermen’s perceptions and attitudes toward risk diving and management issues in small-scale fisheries. American Journal of Human Ecology, 5(1): 1-10.

International Maritime Organization (2009). Second IMO GHG Study 2009. London, United Kingdom: International Maritime Organization.

International Maritime Organization (2014). Third IMO Greenhouse Gas Study 2014. London, United Kingdom: International Maritime Organization.

Kolar, C. S., Chapman, D. C., Courtenay, Jr., W. R., Housel, C. M., Williams, J. D., & Jennings, D. P. (2007). Bigheaded carps: A biological synopsis and environmental risk assessment. American Fisheries Society, Bethesda, Maryland. 204 pp.

Malaysia Fisheries Act (1985). Retrieved from http://www.agc.gov.my, 15 January 2016.

Mohd Salleh, N. H., Riahi, R., Yang, Z., & Wang, J. (2014). Risk assessment of liner shipping from a business



environment perspective. Proc. of the 2nd International Conference on Vulnerability and Risk Analysis and Management and the 6th International Symposium on Uncertainty, Modelling, and Analysis. 2320-2329.

Mohd Salleh, N. H., Riahi, R., Yang, Z., & Wang, J. (2015). Business environment-based risk model for the container liner shipping industry. European Journal of Business and Management, 7(27): 15-27.

Mokhtari, K., Ren, J., Roberts, C. & Wang, J. (2012). Decision support framework for risk management on sea ports and terminals using Fuzzy Set Theory and Evidential Reasoning approach. Expert Systems with Applications, 39(5): 5087-5103.

National Geographic (2016). Sustainable fishing encyclopedia. Retrieved from http://www.nationalgeographic.org/encyclopedia/sustainable-fishing, 15 February 2017.

Riahi, R., Bonsall, S., Jenkingson, I., & Wang, J. (2012). A seafarer’s reliability assessment incorporating subjective judgement. Proceedings of the Institution of Mechanical Engineers, Part M: Journal of Engineering for the Maritime Environment, 226(4): 1-22.

Saaty, T. L. (1980). The Analytic Hierarchy Process. McGraw-Hill, New York. 287 pp.

Saaty, T. L. (2008). Decision making with the analytic hierarchy process. International Journal of Services Sciences, 1(1): 83-98.

Samsudin, B., Sallehudin, J., & Azlin M. N. (2015). Malaysia national report to the Scientific Committee of the Indian Ocean Tuna Commission for 2014. Kuala Lumpur, Malaysia: Department of Fisheries Malaysia.

US Environmental Protection Agency (2017). Risk Assessment. Retrieved from http://www.epa.gov., 16 October 2017.




AN ASSESSMENT OF THE PERCEPTION AND SATISFACTION WITH FLOOD EVACUATION CENTRE SERVICE QUALITY IN EAST COAST STATES OF

PENINSULA MALAYSIA

SITI FALINDAH PADLEE*, NIK NADRUL HISHAM NIK RAZALI, SITI NUR ATIKAH ZULKIFFLI AND NOOR ZATUL IFFAH HUSSIN

School of Maritime Business and Management, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu., Malaysia


Abstract: Floods occur every year in Malaysia, particularly in the east coast states of Kelantan, Terengganu and Pahang. During these floods, thousands of flood victims are evacuated to temporary evacuation centres such as community halls or schools. In each evacuation centre, services are provided to evacuees in order to ensure that they feel comfortable during their stay. Nevertheless, there are reported cases of some evacuees being dissatisfied with the services offered. This study identifies seven types of services that are provided at evacuation centres. To measure the perceived quality of these services and the evacuees’ overall satisfaction with them, a total of 600 questionnaires were distributed in one district of each of the three east coast states that were badly affected by flood events in 2014. A total of 477 questionnaires were returned and a descriptive analysis was performed on the collected data. The findings show that most evacuees perceived that the quality of the services provided to them was good and they were satisfied with the services provided in the evacuation centres. This study is beneficial in terms of assisting managers in the management of such services as well as for the Government and non-governmental organisations in terms of evacuees’ perceived quality of evacuation centres and their level of satisfaction. The managers can also monitor their level of satisfaction based on current findings for future planning in regards to evacuation centres in Malaysia.

Keywords: Descriptive analysis, evacuation centre, perceived quality, satisfaction, service management.

Introduction

In Malaysia, the months of November to March are known as the monsoon season with rainfall that is heavier than usual is recorded. The states of Kelantan, Terengganu and Pahang are usually affected by the monsoon season and often experience severe flooding. Some causes of severe floods include improper drainage systems and a high amount of rainfall in a short period of time (Chan, 1997). The negative impact of severe floods can be seen from many perspectives. Floods can adversely affect the livelihood (i.e. loss of income) and health of those affected; and most victims of floods are left traumatised for some time, especially those who lose family members or loved (Akasah &

Doraisamy, 2015). Severe floods also damage residential and commercial property, disrupt the supply of electricity and clean water, and lead to loss of livestock and environmental degradation. Also, the damage caused to public property and infrastructure, such as roads and rail networks, could pose a significant effect on the economies of the east coast states in the long term. It is estimated that the total cost of damage to both personal and public property caused by the severe floods of 2014 is close to RM400 million (Akasah & Doraisamy, 2015). Although floods can be of benefit to agriculture land as they sometimes can rejuvenate the fertility of the soil, they can also submerge and destroy agricultural land such as paddy fields and palm oil and rubber plantations. Agriculture is one of the principal

: 67-79

68AN ASSESSMENT OF THE PERCEPTION AND SATISFACTION WITH FLOOD EVACUATION CENTRE SERVICE QUALITY IN EAST COAST STATES OF PENINSULA MALAYSIA

economic contributors to the economies of the east coast states; however, the floods of 2014 damaged most oil palm and rubber plantations and led to estimated losses of RM21 billion (Felda Golden Ventures, 2014). Severe floods have affected the lives of the people residing in the east coast states of Peninsular Malaysia for many years (Chan, 1995) and if there is no continuous formal system of supporting the community that lives in these flood-prone areas, this may lead to general insecurity and the people living in those states may become trapped in poverty (Tran et al., 2008).

In any cases of crisis, such as severe floods, the emergency response (Gilbert, 2001) and the provision of safe evacuation and/or temporary shelter is the priority and the main concern of any government (Bologna, 2010). The evacuation centre is one of the important elements to have in place whenever a disaster happens. It is a safe place for the community to assemble (Somasundaram & Davies, 2014). At this place, all essential necessities such as food, accommodation and information are provided (Somasundaram & Davies, 2014). Services such as social assistance as well as emotional, technical and financial support are also provided to disaster victims (Smith & Parsons, 2015). “Evacuation and the use of evacuation centres, involves the planned and coordinated movement of persons from an unsafe or potentially location to a safer location…” (Queensland Government, 2013). Countries such as Japan and Australia have comprehensive evacuation guidelines for managing evacuation centres. They have an effective and efficient standard of operations so that victims feel safe and comfortable while in evacuation centres. The need to maintain a high standard of operations during a crisis has been underlined, so that the evacuation process works effectively and efficiently. In the context of Malaysia, evacuation centres were commonly located in community halls and schools and were chosen based on their location and relative safety. In Malaysia, the responsibility for providing safe places, temporary shelters and aid to flood disaster victims lies with The Malaysian

Disaster Preparedness Centre (MDPC) whose stakeholders include the National Security Council and other government agencies.

The evacuation centre presents some challenges, and these have been studied to some extent by some researchers (Hosseini et al., 2014). The issues are mainly related to the services provided in evacuation centres such as the quality of services, the evacuees’ satisfaction with those services and the management of those services. In the case of the severe floods in Malaysia in December 2014, all the parties in Malaysia worked hard to contribute in any way possible to ensure that the flood victims received appropriate assistance. The most severe flooding at that time occurred in the east coast states, which witnessed the evacuation of thousands of people. As been reported by The Straits Times (2014, December 28); in Kelantan, 124,966 were evacuated, while in Terengganu, it involved the evacuation of 36,410 and in Pahang there were 33,601 evacuees. As a result of the severe flood and the huge number of evacuees during the 2014 disaster, the management of evacuation centres became more complex and weaknesses in the management of evacuation centres were reported. These weaknesses included lack of suitable facilities, shortage of food, overcrowding and no understanding among the victims (Said et al., 2013). Issues involving the lack of emotional support and insufficient number of counsellors were reported by Sinar Harian (2015, Februari 13), while poor electricity and water supply and no proper sleeping areas (Hamidi et al., 2015) were also among the problems highlighted in the media. When such problems exist in evacuation centres, they should have been taken seriously by the authorities. These problems need to be resolved to ensure that the management of evacuation centres is maintained at the highest level and the centres would then be perceived as having the highest quality by disaster victims.

Issues related to evacuation centre might affect the evacuees’ experience and therefore, it is necessary that managers of evacuation centre manage the services wisely


69Siti Falindah Padlee* et al.

based on the evacuees’ expectations. If managers of evacuation centres fail to handle and resolve these problems, it might affect the overall management of the centres and lead to a perception of poor management among disaster victims. To assist managers in this endeavour, this paper measures the perceived quality of services provided at flood evacuation centres and the evacuees’ overall satisfaction with those services.

Perceived quality involves the customers’ overall evaluation of the quality of services (Zeithaml, 1988) and customers’ judgement about the service offering and its delivery (Korda & Snoj, 2010) that differ based on individual influences and experiences of particular services offered (Kayeser & Abdur, 2014). In this study, the specific focus is the quality of services in evacuation centres. Evacuation centres act as the temporary shelters for evacuees for a short period of time, so, the quality of services must meet a recognised standard and the expectations of evacuees for this short period (Hossain & Islam, 2012). It is therefore important to ensure that all the services in evacuation centres are available, easy to use and in excellent condition. Evacuees’ negative or positive perceptions about the services provided should be the benchmark or the yardstick by which to assess the performance of services at evacuation centres (Forsythe, 2016).

Customer satisfaction is the customer’s post-purchase comparison between their pre-purchase expectation and the performance/product they received (Oliver, 1980; Zeithaml et al., 1993) as well as customer’s intentions and behaviour result from customer satisfaction (Olsen & Johnson, 2003). Recent researchers have proposed that customer satisfaction be viewed from two different perspectives: transaction-specific satisfaction and overall satisfaction (Jones & Suh, 2000; Negi & Ketema, 2013). Overall satisfaction or cumulative satisfaction is defined as the customer’s overall satisfaction with the organisation based on all their encounters and experiences with that organisation. Transaction-specific satisfaction,

on the other hand, relates to the customer’s satisfaction based on their evaluation of a particular product or specific service.

The main objective of this study is to examine whether the quality of services received by evacuees in evacuation centres led to their overall satisfaction with those services (Parasuraman et al., 1985). Customer satisfaction is often associated with perceived quality. In this study, perceived quality and satisfaction are seen as separate, where customer satisfaction results from the customer’s evaluation of service quality. Previous studies agree that perceived quality significantly influences overall satisfaction (Chang et al., 2013; Kayeser & Razzaque, 2014).

Methodology

A quantitative research design was used in this study. The researchers undertook a comprehensive investigation to get relevant information regarding the sampling and data collection, data analysis and item measures.

Sampling and Data Collection

Three east coast states were involved because these states experience severe flooding and many people are forced to leave their homes and evacuate to proper evacuation centres. This study focused on three districts, one each from the three states of east coast Malaysia mostly affected by the severe flood event of 2014. The three districts were Temerloh (Pahang), Kemaman (Terengganu) and Kuala Krai (Kelantan). The National Security Council of Malaysia gave permission to distribute a questionnaire to people who had been the evacuees in these three districts. 600 questionnaires were distributed by six enumerators. Before the respondents answered the questionnaire, they were given proper instructions so that they would be able to understand the questions.

Measurement of Items

To measure service quality, previous studies have proposed different dimensions to address the



measurement and identify the issues pertaining to service quality. There is a wide variance in the definition (Giese & Cote, 2000; Fečiková, 2004) and no consensus in the literature on the actual definition of satisfaction (Høst & Knie-Anderson, 2004). Various models have also been developed to measure service quality, such as SERVQUAL and SERVPERF (Kayeser & Abdur, 2014). Both of these models have been discussed in depth in relation to their application in service quality studies. This study referred to several sources to identify the services provided at evacuation centres in Malaysia and these sources were derived from websites, reports in local and international and research journals. It should be acknowledged that studies that measure the perceived quality of and satisfaction with evacuation centres in Malaysia are very limited in number, so this study had to use those sources to develop the items for the questionnaire. Previous studies such as Kayeser and Razzaque (2013), Bigne et al. (2003) and Huddleston et al. (2009) were used to guide the development of the items. In this study, the researchers identified seven dimensions or services provided at flood evacuation centres in order to measure the perceived quality and satisfaction with those services (DeShields et al., 2005). The seven services offered at evacuation centres are thus identified as (1) food, (2) health and safety, (3) transportation, (4) volunteers, (5) site services, (6) telecommunications and (7) special facilities for people with special needs.

In total, 34 items were used to describe the seven dimensions in the questionnaire. After the questionnaire items were developed, the questionnaire was sent to three academicians for validation. When the three academicians had reached a consensus regarding the items and when some improvements had been made, the questionnaire was distributed to 50 participants for a pre-test to confirm its reliability. The Cronbach’s alpha for the seven key dimensions was above 0.969. Therefore, the results of the reliability analysis showed that the items were reliable and ready to be distributed to respondents (Hair et al., 2010).

There were five sections in the final version of the questionnaire: section one was designed to measure the importance of services (‘1’ = not very important to ‘7’ = very important), section two, the respondents’ perceptions about the services provided (‘1’ = strongly disagree to ‘7’ strongly agrees), and section three, the respondents’ overall satisfaction with the services provided (‘1’ = very dissatisfied to ‘7’ = very satisfied), while section four contained open-ended questions to gather data on the evacuees’ overall perceptions of flood evacuation centres and the fifth and final section consisted of questions to obtain a demographic profile of the respondents. For section three, where respondents were asked about their level of overall satisfaction or dissatisfaction based on the scale in the questionnaire, respondents who chose scale 1 to 3 were considered “dissatisfied”, whereas respondents who chose scale 4 to 7 were considered “satisfied”. The analysis of frequencies was run on this part to see the propensity of percentage of evacuees’ satisfaction or dissatisfaction.

Data Analysis

In this study, SPSS (version 21.0) was used to analyse the data. Descriptive analysis, factor loading and Pearson’s correlation were used in the analyses. The descriptive analysis included an analysis of the seven dimensions of perceived quality and overall satisfaction mean score. A descriptive analysis was also performed on the demographic profile of the respondents.

Principal component analysis was performed on 54 items for data reduction. Only 34 items remained with a factor loading above 0.5; items with a factor loading below 0.5 were deleted (Hair et al., 2010). The Kaiser-Meyer-Olkin value was 0.965, which indicated that the sample size used for the analysis was sufficient (Kaiser, 1974), and the Bartlett’s test was 561 with a significance level of 1%. Varimax rotation was applied in the analysis and the results showed that all 34 items were loaded into seven factors with the extracted factors accounting for 81.57% of the variance. Cronbach’s alpha



was also calculated to examine the internal data consistency and reliability. If the value of the Cronbach’s alpha results were above 0.9 (Nunnally & Bernstein, 1994), this was

considered to show that the item was reliable and had high consistency. Table 1 shows the factor loading and Cronbach’s alpha for each dimension and item.

Table 1: Factor analysis and Cronbach’s alpha

Key dimensions and items Loading αF1: FCQ = Food Services Quality 0.9231. Enough food 0.7202. Food cleanliness 0.7653. Caterer behaviour 0.7634. Nutritious food 0.6985. Food for special needs 0.733F2: HSQ = Health and Safety Services Quality 0.9241. Friendly doctors and nurses 0.7212. Availability of medicines 0.7153. Medical care 0.7024. Adequacy of medical equipment 0.7275. Sufficient number of doctors 0.771F3: TQ = Transportation Services Quality 0.9321. Safety 0.6852. Fast services 0.6883. Efficient 0.6864. Friendly staff 0.624F4: VQ = Volunteer Services Quality 0.9401. Volunteer commitment 0.7672. Volunteer performance 0.8273. Support and information provided by volunteers 0.7924. Volunteer diligence 0.7515. Volunteer readiness 0.7406. Volunteer experience 0.782F5: SSQ = Site Services Quality 0.9231. Management efficiency 0.5522. Wide area 0.5773. Adequacy of facilities 0.5424. Cleanliness of environment 0.608F6: TLCQ = Telecommunication Services Quality 0.9741. Fixed-line phone availability 0.7932. Variety of phone providers 0.8413. Telephone facilities 0.7954. Internet access 0.8425. Reload card availability 0.8306. Power banks and chargers 0.822



Key dimensions and items Loading αF7: SFSNQ = Special Facilities for Special Needs Services Quality 0.9361. Strict monitoring 0.6092. Safety for special people 0.6713. Special care for special people 0.5874. Basic needs for special people 0.525

Table 2: Profile of respondents in the East Coast states of Malaysia

Frequency Percent (%) Frequency Percent (%)Sex Age

Male 238 49.90 30 and below 183 38.36Female 234 49.06 31–35 52 10.90Missing 5 1.04 36–40 48 10.06

Race 41–45 55 11.53Malay 462 96.86 45–50 47 9.85

Chinese 6 1.26 51 and above 86 18.03Indian 2 0.42 Missing 6 1.25Others 1 0.21 Status

Missing 6 1.25 Married 309 64.77Level of education Single 129 27.04

None 36 7.55 Widow/Widower 19 3.98Primary school 59 12.37 Missing 20 4.21

Secondary school 250 52.41 ProfessionDiploma 60 12.58 Government 102 21.38

Bachelor’s degree 58 12.16 Private Sector 32 6.71Master’s degree 3 0.63 Self-employed 136 28.51

Missing 11 2.3 Not Working 185 38.78Missing 22 4.62

Results

A total of 481 questionnaires were returned. However, only 477 questionnaires (a response

There were more male respondents (238, 49.90%) than female respondents (234, 49.06%). Most of the respondents were aged 30 years old and below (183, 38.36%), while those who were 45–50 years old (47, 9.85%) made up the smallest group of respondents by age. Most of the respondents were Malay (462, 96.86%), followed by Chinese (6, 1.26%) and Indian (2, 0.42%). Meanwhile, the highest level of education of the majority of the respondents was secondary school (250, 52.41%). Only a few hold a Bachelor’s degree (58, 12.16%). About 64.77% of the respondents were married,

rate of 80.2%) could be used in the analysis. Table 2 shows the demographic profile of these respondents.

27.04% of the respondents were single, and 3.98% were widows/widowers. Most of the respondents were not working (38.78%) because of the age factor. About 28.51% of the respondents were self-employed and 21.38% worked as government officers. Another 32% of the respondents were working in the private sector.

The dimensions of seven services of evacuation centre were analysed to get the mean score of perceived quality. The results are presented in Table 3.



Table 3: Mean score of perceived quality

Dimensions and itemPerceived

quality (PQ; mean)

Standarddeviation (SD)

Food ServicesEnough food 6.10 1.281

Food cleanliness 6.04 1.170Caterer behaviour 6.03 1.055

Nutritious food 5.98 1.213Food for special needs 5.95 1.161

Health and Safety ServicesFriendly doctors and nurses 6.15 1.106Availability of medicines 6.11 1.048

Medical care 6.09 1.055Adequacy of medical equipment 6.08 1.069

Sufficient number of doctors 6.06 1.088Transportation Services

Safety 5.96 1.309Fast services 5.92 1.379

Efficient 5.89 1.255Friendly staff 5.86 1.140

Volunteer ServicesVolunteer commitment 6.14 1.052Volunteer performance 6.13 1.064

Support and information provided by volunteers 6.12 1.068Volunteer diligence 6.11 1.042Volunteer readiness 6.06 1.107

Volunteer experience 6.05 1.064Site Services

Management efficiency 5.99 1.187Wide area 5.94 1.257

Adequacy of facilities 5.92 1.262Cleanliness of environment 5.88 1.286

Telecommunication ServicesFixed-line phone availability 5.43 1.678Variety of phone providers 5.43 1.673

Telephone facilities 5.43 1.761Internet access 5.29 1.742

Reload card availability 5.29 1.738Power banks and chargers 5.26 1.764



Dimensions and itemPerceived

quality (PQ; mean)

Standarddeviation (SD)

Special Facilities for Special Needs ServicesStrict monitoring 0.609 1.142

Safety for special people 0.671 1.155Special care for special people 0.587 1.199Basic needs for special people 0.525 1.226

Note : PQ = perceived quality; mean score: Likert-scale; ‘1’ = strongly disagree to ‘7’= strongly disagree.

The items with the highest mean score of perceived quality were ‘friendly doctors and nurses’ (mean = 6.15), ‘volunteer commitment’ (mean = 6.14), ‘volunteer performance’ (mean = 6.13) and ‘support and information provided by volunteers’ (mean = 6.12). The item of ‘friendly doctors and nurses’ was under the ‘health and safety services’ dimension while the items ‘volunteer commitment’, ‘volunteer performance’ and ‘support and information provided by volunteers’ were items in the ‘volunteer services’ dimension. The items with the lowest mean score of perceived quality were all under the ‘telecommunication services’ dimension, and included ‘internet access’ (mean = 5.29), ‘reload card availability’ (mean = 5.29) and ‘power banks and chargers’ (mean = 5.26).

Based on the findings reported here, the services offered by evacuation centres that were highly rated by evacuees in terms of perceived quality were (1) friendly doctors and nurses, (2) volunteer commitment, (3) volunteer performance and (4) support and information provided by volunteers. Meanwhile, the services that were perceived to be of low quality were all under the telecommunication services dimension. With reference to volunteer services, the victims said that the voluntary service provided at the evacuation centres was as expected. During the severe flooding of 2014, a number of volunteers from public and private agencies came to help flood victims. These public and private agencies consist of volunteers that help voluntarily by providing material assistance such as food, drink, clothing,

medicines as well as moral support (Ministry of Home Affairs, 2015).

One of the issues that have been given priority in evacuation centres is the health and safety of evacuees. To ensure that the evacuees are safe, comfortable and healthy during their stay in an evacuation centre, this service must be of excellent quality. Since each evacuation centre can house thousands of evacuees, it is obligatory to prevent infectious diseases from breaking out and protect evacuees’ security from being compromised. In addition, services such as food services, transportation services and special facilities for special needs services must also be taken seriously. Food services, for instance, must comply with recognised food standards and an adequate level of nutrition should be present in the food and drink provided, as well as a clean dining environment; and the Government is very concerned about the quality of this service. This is because it is important to ensure that the victims get enough food, i.e. eat healthy meals, during their stay in an evacuation centre. Special facilities have also been provided to certain groups of evacuees such as infants, disabled people and the elderly. The services for these groups have been highlighted as a serious concern as they are considered as special evacuees.

The service dimension found to be with lower item ratings was the telecommunication services. Managers should recognise the importance of this service to evacuees. It has been suggested that telecommunication



services be updated and improved. The internet service provider, for instance, should offer broader coverage and unlimited usage. Also, phone reloads by local telecommunication providers should always be available so that evacuees can constantly keep in touch with

In Figure 1, the measurement of overall satisfaction is represented by the items of overall service performance, the fulfilment of evacuees’ needs and wants and overall service quality. The results show that most evacuees (86.9%) were very satisfied with the service performance at the evacuation centres in the east coast states of Malaysia and about 13.1% of evacuees were not satisfied. Regarding the needs and wants of evacuees, 416 (87.3%) respondents were satisfied and only 61 respondents were not satisfied (12.7%). As for the overall quality of services, 421 (89.3%) respondents were satisfied and only 56 (10.7%) respondents were dissatisfied.

their family, relatives and friends. Additional telecommunication services such as phone power banks and chargers should also be provided and given to evacuees as and when needed.

Most evacuees were satisfied with the overall performance and quality of services they received at evacuation centres. Evacuees were also satisfied with the management of the evacuation centres; agreeing that the centres had catered to their needs and wants accordingly. Although this finding indicated a positive result, the managers of evacuation centres should not become complacent; rather, they must continue to endeavour to provide the best services possible to evacuees. On the other hand, managers should also be aware of the needs of those evacuees who were not satisfied with the services available in evacuation centres. These evacuees need to be focused on in order to determine the services with which they were dissatisfied.

Figure 1: Overall level of satisfaction with services at evacuation centres



Discussion

The results of the findings reported herein add to the understanding of evacuees’ perception and satisfaction with the evacuation centers. The findings indicate that evacuees’ perceptions on each of the dimensions of quality services are relatively similar. They perceived that all services provided in evacuation centers are of excellent quality. They also perceived that the quality of services should be continued and upgraded from time to time. The findings also reaffirmed a larger gap between satisfied and dissatisfied percentage of evacuees’ perception on three elements of overall satisfaction with regards to overall performance, needs and wants and overall perceived quality. Overall, the management of evacuation centres must identify all obstacles that inhibit evacuees to obtain better services. Perhaps, it may assist managers and policy makers to make an improvement on the particular aspects (i.e. new policy, strategies and measures), in order to enhance overall evacuees’ satisfaction.

Conclusion and Recommendation

As an overall conclusion, although incidents were reported of weak management at evacuation centres during the flood event of 2014, a proper investigation such as this study has provided the answer to particular problems. Therefore, the findings of this study can be used to inform managers about the evacuees’ perceptions of and level of satisfaction with evacuation centres. It is evident from the literature that perceived quality is a factor that strongly influences satisfaction (Baker & Crompton, 2000; Caruana et al., 2000; Heung & Cheng, 2000; Kozak & Rimmington, 2000; Marinkovic et al., 2014). Thus, it is important to upgrade and continuously improve the services provided to evacuees at evacuation centres so that the flood disaster management in the country can be managed effectively and efficiently. Appropriate policies and guidelines should be developed by the Malaysian Government and non-governmental agencies to ensure that the services at evacuation centres

in Malaysia meet evacuees’ expectations and achieve the desired level of satisfaction.

As a recommendation for future research, this research could be extended by adopting a qualitative research approach so that additional factors influencing the perceived quality and overall satisfaction can be determined. By identifying additional factors, this could lead to further improvements in the services provided by evacuation centres, in the east coast states specifically, and in Malaysia generally. This study has some limitations due to the difficulty of obtaining cooperation from the evacuees, as well as the time constraint. Due to time constraint, the researchers were not able to reach a larger number of respondents from different parts of the three states. If more time were available, the researchers would have been able to gather data and consider the responses of a greater number of diverse respondents.

Acknowledgements

This work received a grant from the Ministry of Higher Education Malaysia under Flood Research Grant 2015, Project Number 59376.

References

Akasah, Z. A., & Doraisamy, S. V. (2015). 2014 Malaysia flood: Impacts and factors contributing towards the restoration of damages. Journal of Scientific Research and Development, 2(14): 53-59.

Baker, D. A., & Crompton, J. L. (2000). Quality, satisfaction and behavioural intentions. Annals of Tourism Research, 27(3): 785–804.

Bigne, E., Moliner, M. A., & Sanchez, J. (2003). Perceived quality and satisfaction in multiservice organisations: The case of Spanish public services. Journal of Services Marketing, 17(4): 420-442.

Bologna, R. (2010). Planning and designing post disaster transitional housing: The case



of the Abruzzo (Italy) earthquake. Paper presented at the 18th CIB World Building Congress 2010, Salford, United Kingdom. Retrieved from https://www.irbnet.de/daten/iconda/CIB19005.pdf, 26 January 2016.

Caruana, A., Money, A. H., & Berthon, P. R. (2000). Service quality and satisfaction–the moderating role of value. European Journal of Marketing, 34(11/12): 1338-1353.

Chan, N. W. (1995). Flood disaster management in Malaysia: An evaluation of the effectiveness of government resettlement schemes. Disaster Prevention and Management: An International Journal, 4(4): 22-29

Chan, N. W. (1997). Increasing flood risk in Malaysia: Causes and solutions. Disaster Prevention and Management: An International Journal, 6(2): 72-86.

Chang, C. S., Chen, S. Y., & Lan, Y. T. (2013). Service quality, trust, and patient satisfaction in interpersonal-based medical service encounters. BMC Health Services Research, 13(1): 22.

DeShields Jr, O. W., Kara, A., & Kaynak, E. (2005). Determinants of business student satisfaction and retention in higher education: Applying Herzberg’s two-factor theory. International Journal of Educational Management, 19(2): 128-139.

Fečiková, I. (2004). An index method for measurement of customer satisfaction. The TQM Magazine, 16(1): 57-66.

Felda Global Ventures (2014). Malaysia floods 2014: Preliminary damage and impact assessment to FGV. Retrieved from http://ir.chartnexus.com/fgv/docs/presentation/2014%20Flood%20Assessment%20Report%20to%20FGV.pdf, 9 July 2016.

Forsythe, P. J. (2016). Construction service quality and satisfaction for a targeted housing customer. Engineering, Construction and Architectural Management, 23(3): 323-348.

Giese, J. L., & Cote, J. A. (2000). Defining consumer satisfaction. Academy of Marketing Science Review, 2000(1): 1-12

Gilbert, R. (2001). Doing more for those made homeless by natural disasters. Disaster Risk Management Series, No. 1. Washington, USA: World Bank.

Hair, F. J., Babin, B., Anderson, R. E., & Black, W. C. (2010). Multivariate Data Analysis: A Global Perspective, 7th edition. Pearson Education, Inc., Upper Saddle River, New Jersey. 800 pp.

Hamidi, M. H., Shaharir, S., Shazalli, S. A., Sharom, I. N. S., & Aziz, N. (2015). Strategic initiatives in flood disaster preparedness and mitigation for Malaysia. Retrieved from http://www.uia-architectes.org/sites/default/files/STRAT-INI_Feb2015.pdf, 15 January 2016.

Heung, V. C., & Cheng, E. (2000). Assessing tourists’ satisfaction with shopping in the Hong Kong special administrative region of China. Journal of Travel Research, 38(4): 396-404.

Hossain, M. J., & Islam, M. A. (2012). Understanding perceived service quality and satisfaction: A study of Dhaka University Library, Bangladesh. Performance Measurement and Metrics, 13(3): 169-182.

Hosseini, A. K., Hosseini, M., Hosseinioon, S., O. Izadkhah, Y., Shaw, T., & Takahashi, R. (2014). A survey on evacuation planning and its challenges for potential earthquake in Tehran. International Journal of Disaster Resilience in the Built Environment, 5(1): 38-52.



Høst, V., & Knie-Andersen, M. (2004). Modeling customer satisfaction in mortgage credit companies. International Journal of Bank Marketing, 22(1): 26-42.

Huddleston, P., Whipple, J., Nye Mattick, R., & Jung Lee, S. (2009). Customer satisfaction in food retailing: Comparing specialty and conventional grocery stores. International Journal of Retail & Distribution Management, 37(1): 63-80.

Jones, M. A., & Suh, J. (2000). Transaction-specific satisfaction and overall satisfaction: An empirical analysis. Journal of Services Marketing, 14(2): 147-159.

Kaiser, H. F. (1974). An index of factorial simplicity. Psychometrika, 39(1): 31-36.

Kayeser, F. J., & Abdur, R. M. (2013). Roles of customer involvement in rapport and satisfaction. Asia Pacific Journal of Marketing and Logistics, 25(3): 452-471.

Kayeser, F. J., & Abdur, R. M. (2014). Service quality and satisfaction in the banking sector. International Journal of Quality and Reliability Management, 31(4): 367-379.

Korda, A. P., & Snoj, B. (2010). Development, validity and reliability of perceived service quality in retail banking and its relationship with perceived value and customer satisfaction. Managing Global Transitions, 8(2): 187-205.

Kozak, M., & Rimmington, M. (2000). Tourist satisfaction with Mallorca, Spain, as an off-season holiday destination. Journal of Travel Research, 38(3): 260-269.

Marinkovic, V., Senic, V., Ivkov, D., Dimitrovski, D., & Bjelic, M. (2014). The antecedents of satisfaction and

revisit intentions for full-service restaurants. Marketing Intelligence & Planning, 32(3): 311-327.

Ministry of Home Affairs (2015). Misi bantuan kemanusiaan dan kemasyarakatan. Retrieved from http://www.moha.gov.my/index.php/ms/kdn-sana-sini/2036-mis i -ban tuan-kemanus ian-dan-kemasyarakatan, 23 January 2016.

Negi, R., & Ketema, E. (2013). Customer-perceived relationship quality and satisfaction: A case of Ethiopian telecommunication corporation. African Journal of Economic and Management Studies, 4(1): 109-121.

Nunnally, J. C., & Bernstein, I. H. (1994). The assessment of reliability. Psychometric Theory, 3(1): 248-292.

Oliver, R. L. (1980). A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of Marketing Research, 17(4): 460-469.

Olsen, L. L., & Johnson, M. D. (2003). Service equity, satisfaction, and loyalty: From transaction-specific to cumulative evaluations. Journal of Service Research, 5(3): 184-195.

Parasuraman, A., Zeithaml, V. A., & Berry, L. L. (1985). A conceptual model of service quality and its implications for future research. The Journal of Marketing, 49(4): 41-50.

Queensland Government (2013). Queensland Evacuation Centre Field Guide. Retrieved from http://www.disaster.q ld .gov. au /D i sa s t e rResou rce s /Documents/RED647%20Field%20Guide_FA_Web.pdf, 24 January 2016.

Said, M. Z., Gapor, S. A., Samian, M. N., & Aziz, A. M. A. (2013). Konflik di pusat pemindahan banjir: Kajian kes di Daerah Padang Terap, Kedah.



Geografia: Malaysian Journal of Society and Space, 9(1): 69-78.

Sinar Harian (2015). Kaunselor di setiap pusat pemindahan banjir. Retrieved from http:/ /www.sinarharian.com.my/nasional/kaunselor-di-setiap-pusat-pemindahan-banjir-1.360193, 15 January 2016.

Smith, C., & Parsons, C. (2015). Preferred sheltering practices for emergency sheltering in Australia. Retrieved from https://www.redcross.org.au/getmedia/cdbdc563-84fc-4af8-aa65-61414f84b01c/2015-Preferrred-Sheltering-Practices-for-Emergency-Sheltering-in-Australia.pdf.aspx, 13 December 2016.

Somasundaram, T., & Davies, B. J. (2014). Collaboration to improve evacuation centre operations in Queensland. International Journal of Disaster Resilience in the Built Environment, 5(3): 305-317.

The Straits Times (2014). Number of evacuees in Malaysia’s worst flooding in a decade jumps to 203000. Retrieved from http://www.straitstimes.com/asia/se-asia/number-of-evacuees-in-malaysias-worst-flooding-in-a-decade-jumps-to-203000, 15 February 2016.

Tran, P., Marincioni, F., Shaw, R., & Sarti, M. (2008). Flood risk management in Central Vietnam: Challenges and potentials. Natural Hazards, 46(1): 119-138.

Zeithaml, V. A. (1988). Consumer perceptions of price, quality, and value: A means-end model and synthesis of evidence. The Journal of Marketing, 52(3): 2-22.

Zeithaml, V. A., Berry, L. L., & Parasuraman, A. (1993). The nature and determinants of customer expectations of service. Journal of the Academy of Marketing Science, 21(1): 1-12.




FOREIGN DIRECT INVESTMENT AND TRADE OPENNESS: DO THEY LEAD TO SUSTAINABLE DEVELOPMENT IN MALAYSIA?

ABDUL RAHIM RIDZUAN1*, NOR ASMAT ISMAIL2 AND ABDUL FATAH CHE HAMAT2

1Faculty of Business Management,Universiti Teknologi MARA, Melaka City Campus, 75300 Melaka, Malaysia

2School of Social Science,Universiti Sains Malaysia, 11800 Pulau Pinang, Malaysia


Abstract: In this research paper, the role of foreign direct investment and trade openness as drivers of sustainable development, are tested against three main pillars of sustainable development, namely; economic growth, income distribution and environmental quality. Foreign direct investment was cited as driver of sustainable development in Investment Policy for Sustainable Development by United Nations Conference on Trade and Development (2012), while trade openness was highlighted by World Trade Organization as another driver for sustainable development especially for developing countries. Three econometric models, which represent the three pillars of sustainable development, are formulated to examine the impact of foreign direct investment and trade openness. The study used annual data from 1970 to 2013 and the analysis was conducted by using Autoregressive Distributed Lag estimation technique. The findings, based on calculated long run elasticity, indicated that foreign direct investment inflows have successfully led to higher growth, better income distribution and lower pollution level in Malaysia. However, for trade openness, the result shows that it is only able to lead to higher growth and lower income gap but insignificant effect on environmental quality. The study found mixed evidence for the effects of other macroeconomic variables in the three estimated models. With the rise in competition from around the world, the study suggests that there is an urgent need for the Malaysian policymakers to revise the existing policies to attract more sustainable foreign direct investments besides making trade more open by involving in more free trade agreements in order to achieve the country’s sustainable development goals.

Keywords: Sustainable development, foreign direct investment, trade openness, Bound test, growth.

Introduction

As one of the progressive developing countries in the Asia region, Malaysia is considered as the fourth most open economy in the world with its export and imports accounting for most of its gross domestic product. According to Kim and Lin (2009), trade openness (TO) has played a crucial role in shaping the stages of economic development for Malaysia over many years by acting as a driver of growth. Beside TO, foreign direct investment (FDI) is another vital driver of growth, and is considered as the backbone of growth for this country. The active participation of Malaysia in ASEAN (Association of South East Asian Nation), has transformed this country as one of the favorite

destinations for FDIs in the region through the introduction of ASEAN Investment Area (AIA), and has facilitated higher level of trade activities due to the implementation of Common Effective Preferential Trade Agreement (CEPT) scheme. Recently, according to World Investment Report by the United Nations Conference on Trade and Development (2012) and World Trade Organization, the roles of FDI and TO have expanded to include not just the effect on growth but also on income distribution and environmental quality. These three elements, namely economic growth, income distribution and environmental quality, form the three main pillars of sustainable development as addressed by the Commission on Sustainable Development of the United Nations. The concept

: 81-100

82FOREIGN DIRECT INVESTMENT AND TRADE OPENNESS: DO THEY LEAD TO SUSTAINABLE DEVELOPMENT IN MALAYSIA?

for sustainable development as defined here has been adopted mainly by all organizations within the United Nations system, as well as most other international and national agencies involved in the cooperation with the third world countries (Martinussen, 2004). According to Martinussen (2004), sustainability in development means that we are able to meet the needs of the current generation without compromising the needs of future generations.

Achieving sustainable development has become the goal for every country in today’s world including Malaysia. Malaysia has integrated the concept of sustainable development in various policies. For example, Malaysia introduced National Development Policy (NDP) which focused on eradicating poverty. The programs under this policy are merged into the national planning process when the Sixth Malaysia Plan (1991 to 1995) was revised in 1995 (Saadatian et al., 2009). Besides NDP, Malaysia also introduced National Conservation Policy (NCP) which was said to function as a framework for the development of natural resources. Moreover, according to Hassan (2004), Malaysian National Vision Policy (NVP) which was realized through the Third Outline Perspective Plan (2001 to 2010), has implanted the concept of sustainable development in its second, fourth and last articles as follows ; 1. Encouraging more equitable society, 2. Sustaining economic development, and 3. Pursuing environmental protection. Towards this end, despite the many policies that have been introduced by the Malaysian government to reach sustainable development goals, there is still a lack of empirical assessments on detecting the presence of sustainable development through its three main pillars. Such assessments are needed as the steady inflow of FDI in the recent decade and the deepening of trade liberalization in Malaysia could be the potential drivers for this country to achieve sustainable development as set under ASEAN Vision 2025.

With more than 40 years’ history of FDI inflows and increasing openness to trade

that Malaysia has experienced, this paper aims to determine the impact of FDI and TO along with selected macroeconomic variables on the above-mentioned three pillars of sustainable development. To the best of the writers’ knowledge, this study is perhaps the only study available that investigated the evidence of sustainable development for Malaysia by using these three pillars of sustainable development employing a comprehensive econometric method of assessment.

Selected past studies on the three pillars of sustainable development are reviewed with the condition that the model incorporated both FDI inflows and TO, besides other various determinants in a single framework. Even though FDI inflows and TO are being used as one of the determinants of the three pillars of sustainable development, the previous studies do not relate the outcome of this variable as a potential driver for sustainable development. Besides, the studies only focus on one area or topics (economic growth, income distribution or environmental quality) which, accordingly, can’t hardly be made related with the presence of sustainable development. However, the previous studies are able to give greater insight to the researcher on the scope of research on each pillar of sustainable development. A number of past studies are briefly discussed below.

Solarin and Shahbaz (2015) have added new variables, natural gas consumption to capital, FDI, and TO in a growth model to study Malaysian growth for the period of 1971 until 2012. The inclusion of natural gas in the model has become essential in today’s world as it emits 50% less environmental pollution compared to other fossil fuels. The outcome of the estimation showed that natural gas consumption, FDI, capital, and TO have a direct relationship to Malaysia’s economic growth during the period of the study. The positive association between FDI and Malaysia’s growth is similar to the finding of an earlier study by Anwar and Sun (2011). Further, Anwar and Sun (2011) have carried out the Granger causality tests and found that the outcome supported


83Abdul Rahim Ridzuan1* et al.

the presence of feedback hypothesis between natural gas consumption and economic growth, FDI and economic growth, and lastly between natural gas consumption and FDI. Using the World Inequality Database and a static and dynamic panel data analysis, Faustino and Vali (2011) have analyzed the relationship between income inequality and economic globalization, measured by trade openness and FDI, in 24 OECD countries for the period spanning from 1995 to 2007. Faustino and Vali (2011) have used both fixed effect estimators and GMM estimators in the analysis, with the latter being able to avoid endogeneity problems that may exist in the model. The result of the statistical analysis which was conducted using fixed effect estimators shows that higher TO lead to lower inequality whereas higher FDI inflows have increased the inequality. Some control variables introduced in the model, such as inflation and unemployment, also show a positive impact on inequality. However, the results are slightly different when GMM estimation was used. In this case, trade was found to decrease income inequality while the effect of FDI was no longer significant. It was shown for both static and dynamic panel data analyzed that economic growth has caused inequality to increase.

Meanwhile, Shahbaz et al. (2013) have used the model of environmental quality to test the impact of FDI, TO, and another macroeconomic variable on pollution for Malaysia based on the period of 1971 to 2011. Their model was tested using Bound test. Other earlier studies on Malaysia, such as Ang (2008) and Lean and Smyth (2010), have included energy consumption as one of the important determinants beside TO and FDI. This is in line with some recent studies that found that development of financial market may also influence energy emissions, while at the same time, motivating technological progress in the energy sector that works to cut emissions. The results of long run elasticity in this study indicated that financial development has led to lower air pollution. This means that the positive and significant influence of financial development is able to counter the opposing

environmental deprivation in the country as more financings at lower costs are channeled to investments in environmentally friendly projects. Meanwhile, energy consumption, FDI, and economic growth have been found to worsen the air quality. A more recent study by Ridzuan et al. (2017) tested the role of FDI towards economic growth, income distribution and environmental quality on Singapore. The study used ARDL estimation and the covered period were tested from 1970 until 2013. The empirical findings showed that higher FDI inflows improved the economics growth and environmental quality, however, it may also lead towards greater income distribution in the country.

Methodology

Model of Economic Growth

The model of economic growth was derived from the endogenous growth model by extending the simple Cobb-Douglas production function and transforming all variables into logarithmic linear form (LN).

LNGDP = β0 + β1LNLABt + β2LNDIt + β3LNFDIt + β4LNHCt +β5LNTOt + β6LNFDt + εt --- (1)

LAB (total labor force) is added in the model as a control for other determinants of growth, represented by GDP (real gross domestic product per capita, constant at 2005), while, also helping to reduce the problem of omitted variable bias. Domestic investment, DI (proxied by gross fixed capital formation, constant at 2005) is included, following the extended model proposed by Barro (1999). With DI in the model, we are able to test whether DI is better or not in generating growth as compared to foreign investment under the dependency hypothesis introduced by Amin (1974) and Frank (1978). To strengthen the empirical results, HC (human capital, proxied by secondary school enrollment rate), following the work of Blomstrom et al., (1994) was added as it can play a critical role in absorbing foreign knowledge or skills brought about by foreign investors via the FDI inflows. Besides LAB, we



also added financial development, FD (proxied by the ratio of M2 over GDP) as another control variable in Equation 1. This is in line with Hermes and Lensik (2003), which said that the positive growth impact of FDI is dependent on the extent of financial sector development in host countries. Lastly, the main variables of our study, FDI (foreign direct investment inflows, given as a percentage of GDP) and TO (trade openness, proxied by trade share of GDP, as highlighted by Balasubramanyam et al., 1996), are added in order to determine whether they can potentially bring positive impact on Malaysia’s economic growth. According to Bhagwati (1994), the impact of FDI on economic growth tends to be greater in developing countries such as Malaysia, where the export promotion trade strategies are preferred over import-substitution strategies. Generally, based on Malaysia’s development performance so far, all coefficients (β1, β2, β3, β4, β5, β6) are expected to have a positive relationship with the country’s GDP. Most of the variables selected above can be regarded as having elements of absorptive capacities that enable the host country to absorb and implement new incoming technologies from foreign countries. Besides, it can help to improve the investment climates, promote the efficiency of FDIs, and thus generate growth for the host country.

Model of Income Distribution

Following the model proposed by Tsai (1995) and Mah (2003), we derived Equation 2 as below:

LNGINI = β0 + β1LNGDPt + β2LNDIt + β3LNFDIt + β4LNTOt + β5LNFDt + εt ---(2)

Higher growth in a country will usually lead to higher income inequality within the society unless the government carefully monitors the situation and implements appropriate and effective policies such as a fiscal policy to curb the problem. However, GDP is expected to have a positive influence on income distribution in Malaysia. The inclusion of DI is crucial as it plays the role of a variable with absorptive capacity that can influence the

efficiency of FDI in the model. Meanwhile, the impact of FDI on income distribution can be either way, positive or negative. For example, based on Mundell hypothesis, an increase in FDI inflows will reduce income inequality in middle-income countries. This may be due to the fact that higher movement of capital from other countries will lead to higher wages for the local workers. Given that Malaysia can be considered as labor abundant country, TO is expected to have a negative correlation with GINI (GINI Coefficient) as a result of trade liberalization. Another variable included in the model is FD, which is another important factor that affects income inequality. Greenwood and Jovanovich (1990) has postulated that FD will increase income inequality initially but later lead to its decline once the financial sector matures. To conclude, β1 is expected to have a positive sign while for the others (β2, β3, β4, and β5), the signs are expected to be mixed, either positive or negative. As was similarly argued in the case of the model for economic growth, the selection of variables in this model is also based on their absorptive capacity properties since such variables can ensure the optimal impact of FDI on income distribution in Malaysia.

Model of Environmental Quality

Equation 3 is a modified version of the model of environmental quality proposed by Lee (2013).

LNEVQ = β0 + β1LNGDPt + β2LNFDIt + β3LNFDt + β4LNENt + β5LNTOt + εt ---(3)

In this model, the real GDP per capita is expected to have a positive influence on the level of environmental quality (EVQ) which captured by carbon emissions (β1 > 0) while effect of FDI on EVQ is expected to be mixed (positive or negative). For example, FDI may have a negative effect on air pollution when it (FDI) facilitates the adoption of modern technologies in the production processes that ensures the best possible environmental practices to energy efficiency, better knowledge transfers, labor training, skill acquisition, introduction of alternative management practices, and etc. (Prakash & Potoski, 2007). The effect of TO



on EVQ is also mixed, as its increase may deteriorate or improve environmental quality (Antwieler et al., 2001). For example, an increase in trade openness that result from higher import of environmentally friendly and more efficient technologies will reduce carbon emissions and enhance environmental quality. Another potential determinant added into the model is EN (per capita energy consumption, measured in kg of oil equivalent) (Ang, 2008). Since higher level of energy consumption leads to greater economic activity and stimulate more carbon emissions, its coefficient (β4) is expected to have a positive sign. Besides FDI and TO, FD is also expected to have a mixed sign. Frankel and Rose (2002) for example, have found that FD could improve EVQ in countries where more financial resources are allocated to firms that utilize environmentally friendly technologies. In summary, β1 and β4 are expected to have positive signs while for other variables (β2, β3, and β5), the signs are expected to be mixed, either positive or negative.

In this time series analysis, the variables are tested for stationarity, by using

the conventional Augmented Dickey-Fuller (ADF) and the Phillips–Perron (PP) test. The Autoregressive Distributed Lag (ARDL) bounds test is only allowed to proceed if the tested variables are stationary at I (0) or I (1) but not allowed if the variables are stationary at I (2) to avoid spurious results. Next, ARDL co-integration, also known as Bounds test, developed by Pesaran et al. (2001) is applied to test for the existence of long run relationships in the model. The Bounds test is mainly based on the joint F-statistic whose asymptotic distribution is non-standard under the null hypothesis of no co-integration. The null hypothesis of no co-integration is rejected when the value of the F-statistic exceeds the upper critical bounds value and not rejected if the F-statistic is less than the lower bounds value. Otherwise, the co-integration test is inconclusive.

The unrestricted error correction model (UECM) form of ARDL model that contains both short run and long run dynamics are listed as follows:






where ∆ is the first-difference operator and vt is the white-noise disturbance term. Residuals for all the UECM models should be serially uncorrelated and the models should be stable. The above final model for economic growth can be viewed as an ARDL or Bound of order (p q r s t u v) while for the model of income distribution and model of environmental quality, it is of order (p q r s t u).

Beside estimating short run and long run elasticity, this research has also performed the Toda and Yamamoto’s (1995) Granger non-causality test through vector autoregressive (VAR) model to verify the direction of causality between the variables for the model of economic growth, model of income distribution and model of environmental quality. The information on the direction of causality between the variables in the models will provide additional inputs for policy recommendations. According to the authors, this method is valid regardless of whether the series is I (0), I (1) or I (2), non-co-integrated or co-integrated of any arbitrary order. Before running the test, the optimum lag for each model is determined using VAR lag order selection based on AIC. Ganger causality test is very sensitive to the selection of lag length. If the chosen lag length is less than the true lag length, the omission of relevant lags can cause biased results, while in the case of higher chosen lag length, the presence of irrelevant lags in the equation will cause the estimates to be inefficient.

This study used annual data from 1970 to 2013. The data for DI, EN, FD, FDI, GDP and TO were extracted from World Development Indicator (WDI) (2016). The sources for other data are as follows: data for HC was taken from Barrolee database (Barro & Lee, 2011), data for LAB from both WDI (2016) and International Labor Organization (ILO), data for EVQ is from both WDI (2016) and Emissions Database

for Global Atmospheric Research (2015), and finally, data for GINI coefficients from University Texas Income Project and Global Consumption Income Project.


Table 1 below shows the results of the unit root tests for all variables in the three models. The results confirm a mix of stationarities at I (0) and I (1) for all variables, and thus fulfill the criteria to proceed with further analysis using ARDL technique.

Table 2 displays the result of ARDL bounds F-test for co-integration. The maximum lag of 4 was set in each model as shown by the Akaike Information criterion (AIC). The critical values for model of income distribution and model of environmental quality are given under the number of variables, k=5, while for the model of economic growth, the critical value is given under k=6. The F-statistic for each model (5.33, 3.90 and 3.87) is higher than the corresponding upper I (1) critical values in the table for k=6 and k=5, and this renders the model significant at 1 and 5 percent levels, respectively, thus, validates the existence of long run relationship.

The results from Table 3 confirm that all models have passed all diagnostic tests, and this implies that the long-term estimates of these models are reliable. The models show no evidence of serial correlation and have no heteroscedasticity effects in the disturbance terms. Besides, the Jarque-Bera normality test suggests that the errors of each model are normally distributed and all the models are well specified. In addition, the results of the cumulative sum of recursive residuals (CUSUM) and the cumulative sum of square of recursive residuals (CUSUMSQ) based on Figure 1 reveal that all models are stable.



Table 1: Results of ADF and PP Unit Root test

Model VariableADF test statistic PP test statistic

Intercept Trend and intercept Intercept Trend and

intercept


Level LNGDP -1.52 (0) -1.99 (0) -1.48 (1) -2.08 (2)LNLAB -1.29 (0) -1.60 (0) -1.29 (1) -1.64 (1)LNDI -2.48 (1) -2.50 (1) -2.40 (2) -2.37 (1)

LNFDI -2.84 (8)* -5.54 (0)*** -5.58 (1)*** -5.54 (1)***LNHC -1.21 (0) -2.56 (0) -1.29 (7) -2.48 (5)LNTO -0.89 (0) -0.869 (0) -0.89 (2) -0.86 (0)LNFD -2.54 (0) -2.90 (0) -2.74 (5)* -2.81 (3)

First difference LNGDP -5.61 (0)*** -5.77 (0)*** -5.621 (1)*** -5.77 (0)***

LNLAB -6.50 (0)*** -6.48 (0)*** -6.50 (1)*** -6.49 (2)***LNDI -4.71 (0)*** -4.64 (0)*** -4.65 (3)*** -4.58 (3)***

LNFDI -2.84 (9)* -2.64 (9) -25.10 (24)*** -24.83 (25)***LNHC -5.85 (1)*** -5.80 (1)*** -5.89 (2)*** -6.01 (3)***LNTO -5.70 (0)*** -5.75 (0)*** -5.67 (2)*** -5.72 (3)***LNFD -5.92 (1)*** -6.11 (1)*** -6.44 (4)*** -7.00 (6)***

Model of Income

Distribution

Level LNGINI -5.99 (0)*** -5.91 (0)*** -6.04 (2)*** -5.98 (2)***LNGDP -1.52 (0) -1.99 (0) -1.48 (1) -2.08 (2)LNDI -2.48 (1) -2.50 (1) -2.40 (2) -2.37 (1)

LNFDI -2.84 (8)* -5.54 (0)*** -5.58 (1)*** -5.54 (1)***LNTO -0.89 (0) -0.869 (0) -0.89 (2) -0.86 (0)LNFD -2.54 (0) -2.90 (0) -2.74 (5)* -2.81 (3)

First difference LNGINI -7.11 (2)*** -7.55 (2)*** -22.97 (41)*** -36.14 (20)***

LNGDP -5.61 (0)*** -5.77 (0)*** -5.621 (1)*** -5.775 (0)***LNDI -4.71 (0)*** -4.644 (0)*** -4.65 (3)*** -4.58 (3)***

LNFDI -2.84 (9)* -2.64 (9) -25.10 (24)*** -24.83 (25)***LNTO -5.70 (0)*** -5.752 (0)*** -5.67 (2)*** -5.72 (3)***LNFD -5.92 (1)*** -6.11 (1)*** -6.44 (4)*** -7.00 (6)***



Model of Environmental

Quality

Level LNEVQ -0.81 (0) -2.17 (0) -0.81 (0) -2.20 (2)LNGDP -1.52 (0) -1.99 (0) -1.48 (1) -2.08 (2)LNFDI -2.84 (8)* -5.54 (0)*** -5.58 (1)*** -5.54 (1)***LNFD -2.54 (0) -2.90 (0) -2.74 (5)* -2.81 (3)LNEN -1.032 (0) -2.122 (0) -1.545 (11) -2.191 (1)LNTO -0.89 (0) -0.869 (0) -0.89 (2) -0.86 (0)

First difference LNEVQ -7.53 (0)*** -7.45 (0)*** -7.52 (1)*** -7.44 (1)***

LNGDP -5.61 (0)*** -5.77 (0)*** -5.621 (1)*** -5.77 (0)***LNFDI -2.84 (9)* -2.64 (9) -25.10 (24)*** -24.83 (25)***LNFD -5.92 (1)*** -6.11 (1)*** -6.44 (4)*** -7.00 (6)***LNEN -6.63 (0)*** -6.71 (0)*** -6.97 (7)*** -9.63 (14)***LNTO -5.70 (0)*** -5.752 (0)*** -5.67 (2)*** -5.72 (3)***

Note: 1. ***, ** and * are 1%, 5% and 10% of significant levels, respectively. 2. The optimal lag length is determined by using the Akaike Information Criteria (AIC) for ADF test and the bandwidth had been selected by using the Newey–West method for PP unit root test.

Table 2: Results of ARDL bounds F-test for cointegration

Model Max. lag Lag order F-StatisticModel of Economic

Growth 2 (2,2,2,0,2,2,1) 5.332***

Model of Income Distribution 4 (4,0,4,1,0,0) 3.906**

Model of Environmental Quality 4 (4,3,3,4,3,4) 3.869**

k = 6 k = 5

Critical Values for F-statistics# Lower I (0) Upper I (1) Lower I (0) Upper I (1)

1% 3.15 4.43 3.41 4.685% 2.45 3.61 2.62 3.7910% 2.12 3.23 2.26 3.35

Note: 1. # The critical values are based on Narayan (2004), case III: unrestricted intercept and no trend. 2. k is a number of variables. 3. *, **, and *** represent 10%, 5% and 1% levels of significance, respectively. 4. k =5 for the model of income distribution and model of environmental quality, while k =6 for the model of economic growth. Max refers to maximum.

Model Serial correlation Functional form Normality HeteroscedasticityModel of Economic

Growth1.33

[0.28]1.72

[0.20]1.29

[0.52]1.53

[0.16]Model of Income

Distribution2.09

[0.14]0.07

[0.79]0.96

[0.61]1.22

[0.31]Model of Environmental

Quality1.98

[0.19]0.01

[0.93]1.98

[0.19]0.38

[0.97]

Note: The numbers in brackets [ ] are p-value.




-15

-10

-5

0

5

10

15

90 92 94 96 98 00 02 04 06 08 10 12

CUSUM 5% Significance




-0.4

0.0

0.4

0.8

1.2

1.6

03 04 05 06 07 08 09 10 11 12 13

CUSUM of Squares 5% Significance




-10.0

-7.5

-5.0

-2.5

0.0

2.5

5.0

7.5

10.0

2001 2002 2003 2004 2005 2006 2007 2008 2009 2010

CUSUM 5% Significance

Figure 1: Cumulative Sum of Recursive Residuals (CUSUM) and Cumulative Sum of Square Of Recursive Residuals (CUSUMSQ)



Table 4 below presents the results of the long run elasticity for each model in this study. Based on the model of growth, DI, FDI, HC, TO and FD have a positive and significant influence (at 1%, 5% and 10 % levels of significance) on the growth of Malaysian economy. Since the coefficient value of DI is greater than that of FDI, it means that domestic investment has contributed more to economic growth than FDI based on the differential productivity argument as proposed by dependency theorists. Rising trade openness (TO) as well as higher inflows of investment was due the active participation of Malaysia as one of the original member of ASEAN group towards the implementation of CEPT scheme to reduce the trade barriers in order to increase the international trade activities and the promotion of AIA that attracted higher FDI inflows into the country. In this case, a 1 percent increase in FDI and TO will increase Malaysia’s economic growth by around 0.02 percent and 0.51 percent, respectively. The significance of DI, HC, TO and FD highlights the importance of absorptive capacities of these variables in enhancing the impact of FDI on growth in Malaysia. For example, the important role of human capital development as a crucial facilitator in technology transfers associated with FDI was recorded by Zhang and Markusen (2009).

Next, the results from the model of income distribution show that higher values of DI, FDI and TO will improve income distribution (lower GINI) while higher GDP can worsen it (higher GINI). As for FD, the result shows that it does not have any significant relationship with income distribution in Malaysia. The finding of a negative relationship between FDI and TO to GINI in Malaysia is in line with the studies by Choi (2004) and Reuveny and Li (2003). This can be explained by referring to Mundell hypothesis (Mundell, 1957) in the following

way: higher FDI inflows in developing countries like Malaysia will result in a reduction in income inequality due to the higher marginal physical product of labor since there is more capital for each worker to work with. As for TO, the Heckscher-Ohlin and Stolper-Samuelson models explain that, a rise in TO can increase the demand for less skilled workers in the production of labor-intensive products, and this will raise the wages of these workers and decrease income inequality.

In the case of the model of environmental quality, the results show that all variables are found to have a significant influence (at 1% and 5 % levels of significance) on the EVQ in the country except FD. Higher GDP and TO will lead to higher release of carbon emissions in the air (higher air pollution) and thus decrease the environmental quality. An increase in TO for example, will lead to greater economic activities, and as these activities inherently entail environmental costs, they worsen the environmental quality. This result is similar to the recent findings on environmental quality model for Malaysia by Ridzuan et al. (2017). The findings on FDI inflows and EN, on the other hand, indicate a negative relationship between these variables and EVQ, thus leading to an improvement in environmental quality. However, it is possible for the combined effects of both FDI and EN on EVQ to outweigh the impact of TO, meaning that there is a chance for Malaysia to enjoy better environmental quality if the government can utilize the FDI mechanism properly and appropriately. Overall, FDI seems to be a better lead driver of sustainable development for Malaysia than TO as it (FDI) is able to successfully promote more growth, better income distribution, and better environmental quality, while TO can bring about favorable impact on growth and income distribution only.



In this section, we discuss the results of the short-run elasticity estimations as can be viewed in Table 5. Here, we will focus only on the short run impacts of FDI and TO. The result shows a mixture of signs for the variables at different lags for all three models with the variables of the first two models (model of economic growth and model of income distribution) are mostly significant. Based on the model of economic growth, both FDI and TO were found to have a positive relationship with

growth. As for the model of income distribution, the result shows that both FDI and TO have significantly improved income inequality. For the model of environmental quality, higher FDI inflows, based on the first lag, appears to improve environmental quality, but later on, seems to lean towards more adverse effect on the environment, towards the last lag. As for TO, based on first lag and last lag, shows that higher openness to trade could also worsen the environmental quality.

Table 4: Estimation of Long-Run Elasticity

Model of Economic GrowthLNGDP

Model of Income DistributionLNGINI

Model of Environmental QualityLNEVQ

Variables Coefficient Variables Coefficient Variables CoefficientLNLAB 0.0007 LNGDP 0.461** LNGDP 1.575***LNDI 0.051* LNDI -0.282** LNFDI -0.082**

LNFDI 0.016* LNFDI -0.037* LNFD -0.044LNHC 0.605*** LNTO -0.415** LNEN -0.637**LNTO 0.510*** LNFD 0.027 LNTO 0.611***LNFD 0.047** C 7.091*** C -9.604***

C 0.082

Note: 1. ***, **, * indicate significance at 1%, 5% and 10% levels of significance, respectively.

Table 5: Estimation of Short Run Restricted Error Correction Model (ECM)

Model of Economic Growth Model of Income Distribution Model of Environmental QualityVariables Coefficient Variables Coefficient Variables CoefficientΔLNGDPt 0.247* ΔLNGINIt-1 -0.649** ΔLNEVQt-1 1.266*ΔLNLABt -0.013*** ΔLNGINIt-2 -0.781*** ΔLNEVQt-2 0.823*ΔLNLABt-1 -0.007* ΔLNGINIt-3 -0.599*** ΔLNEVQt-3 0.571**

ΔLNDIt 0.110*** ΔLNGDPt 0.310** ΔLNGDPt 0.542ΔLNDIt-1 -0.097*** ΔLNDIt -0.134 ΔLNGDPt-1 -1.063ΔLNFDIt 0.008** ΔLNDIt-1 0.071 ΔLNGDPt-2 -1.266*ΔLNHCt 0.330*** ΔLNDIt-2 -0.330*** ΔLNFDIt -0.047**ΔLNHCt-1 -0.136** ΔLNDIt-3 0.298*** ΔLNFDIt-1 0.017ΔLNTOt 0.118** ΔLNFDIt -0.039*** ΔLNFDIt-2 0.096**ΔLNTOt-1 -0.250*** ΔLNTOt -0.279** ΔLNFDt 0.207ΔLNFDt -0.046** ΔLNFDt 0.018 ΔLNFDt-1 0.152ECTt-1 -0.509*** ECTt-1 -0.672** ΔLNFDt-2 0.161

ΔLNFDt-3 0.186



ΔLNENt 0.605ΔLNENt-1 -0.165ΔLNENt-2 0.734*ΔLNTOt 0.989*ΔLNTOt-1 0.430ΔLNTOt-2 -0.691ΔLNTOt-3 0.724*

ECTt-1 -0.289***R square 0.99 R square 0.73 R square 0.96

Ad.R square 0.99 Ad.R square 0.58 Ad.R square 0.96

Note: ***, **, * indicate significant at 1%, 5% and 10% significant level. Ad refer to adjusted.

The estimates of long run elasticity in each model were supported by the negative and significant value of error correction term (ECT). ECT represents the speed of adjustment for each model and the negative value means that the variables in each model will converge in the long run. The model of income distribution has the highest speed of adjustment (-0.67), followed by the model of economic growth (-0.51) and the model of environmental quality (-0.29). This means that approximately, 67 percent, 51 percent and 29 percent disequilibria, respectively from the previous year’s shock of those models, converge back to the long run equilibrium in the current year. Overall, the R-squared values suggest that almost 99 percent, 73 percent and 96 percent of the variables in the three respective models are able to explain the corresponding dependent variables (GDP, GINI, and EVQ).

As for the Toda-Yamamoto Granger non-causality test, which is to test the causality between variables on each model, the optimum lag detected for a model of economic growth is 4, while for the model of income distribution and the model of environmental quality, it is 5. One extra lag (dmax=1) was added to the optimal lag of VAR model for implementing the Granger non-causality test using Toda and Yamamoto approach. Next, to ensure that each model is dynamically stable, inverse roots of AR polynomial was performed. The result is reported in Figure 2, where all the inverted roots (dots) for each model are shown to lie strictly inside the unit circle, which confirmed that all models are dynamically stable.

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5Model of Economic

GrowthModel of Income

DistributionModel of Environmental

Quality

Figure 2: AR Roots Graph



Table 6: Toda Yamamoto Granger Non-Causality Test

Dependent variable

Direction of causalityModel of Economic Growth

Short RunMalaysia LNGDPt LNLABt LNDIt LNFDIt LNHCt LNTOt LNFDt ALLLNGDPt - 2.144 1.653 1.587 1.889 1.005 1.207 10.203LNLABt 1.022 - 2.782 1.737 0.742 2.048 0.826 5.495LNDIt 15.100*** 22.761*** - 6.814 16.550*** 17.458*** 9.707** 54.252***

LNFDIt 10.217** 6.853 10.494** - 12.851** 6.528 6.101 40.390**LNHCt 1.995 1.715 1.187 1.921 - 1.955 0.820 9.043LNTOt 8.620* 7.011 11.336** 12.441** 11.669** - 8.872* 60.991***LNFDt 4.449 5.548 4.424 1.941 7.625 3.264 - 104.501***


Malaysia LNGINIt LNGDPt LNDIt LNFDIt LNTOt LNFDt ALLLNGINIt - 17.459*** 6.283 20.916*** 14.432** 18.822*** 207.096***LNGDPt 1292.147*** - 809.982*** 726.646*** 2352.265*** 481.094*** 4000.027***LNDIt 7.832 2.535 - 6.487 3.953 2.767 18.229***

LNFDIt 17.654*** 10.350* 19.295*** - 13.968** 7.303 140.178LNTOt 32.283*** 50.520*** 7.319 8.574 - 11.025** 130.502***LNFDt 47.451*** 33.396*** 17.543*** 36.167*** 31.263*** - 136.077***

The outcomes of Toda-Yamamoto Granger non-causality test are displayed in Table 6 while the figure illustration can be viewed in Figure 3. Based on the model of economic growth, the result shows unidirectional causalities between the following pairs of variables: (a) LNGDP and LNDI, (b) LNGDP and LNFDI, and (c) LNGDP and LNTO. In addition, it is found that LNFD causes LNDI and LNTO, LNHC causes LNFDI and LNTO, while LNLAB and LNFDI, on the other hand, cause LNDI. The feedback hypothesis is only confirmed between LNDI and LNTO. Although no causality running from LNFDI to LNGDP is detected, there is causality from LNGDP to LNFDI. This suggests that per capita income growth can be taken as a requirement in

attracting FDI into the country. In addition, FDI is also important for the country to boost up its DI, since producing better products using more advanced foreign techniques or technologies brought about by FDIs, would encourage more local firms to compete by increasing their domestic investment. The causality running from LNHC to LNFDI means that more FDIs would pour into the economy where there is a higher level of knowledge and awareness as stated by Miyamoto (2003). Realizing this, Malaysian government has allocated a huge amount of the budget into programs such as MyBrain 15 to increase the total enrolment in postgraduate studies in various areas in an effort to boost human capital development in the future.



For the model of income distribution, there are at least 5 bi-directional causalities detected between the variables, which are between: (a) LNGINI and LNGDP, (b) LNGINI and LNFD, (c) LNGINI and LNFDI, (d) LNGDP and LNFD, and lastly (e) LNGDP and LNFDI. In addition, the following causalities are also identified: LNTO causes all other variables except for LNDI; LNDI causes LNGDP, LNFD and LNFDI; and LNFDI causes LNFD. For the model of environmental quality, a strong bidirectional causality is detected between: (a) LNGDP and LNTO, (b) LNGDP and LNFD, (c) LNGDP and LNTO, and lastly (d) LNTO and LNEN. The finding of bidirectional causality between financial development and economic growth, which confirms the existence of both supply-side and demand-side hypotheses, is

consistent with Ndoko (2010), who observed the same for South Africa. Other causalities uncovered include: LNEVQ causes all other variables except LNFDI; LNFDI causes LNGDP, LNTO and LNFD; and LNEN causes LNGDP and LNFD.

Unlike other studies on Malaysia, such as the study by Tang and Tan (2012), the authors have found a bi-directional causality between LNEN and LNGDP. The inconsistent Granger causalities detected between the variables in these models are not only due to differences in time spans and econometric techniques employed, but also due to the omission of some variables and the use of proxy variable for energy, where the latter are seen to play even more important role.


Malaysia LNEVQt LNGDPt LNFDIt LNFDt LNENt LNTOt ALLLNEVQt - 1.238 0.435 2.356 0.680 0.677 9.231LNGDPt 1110.017*** - 520.819*** 536.032*** 900.136*** 1117.185*** 3532.421***LNFDIt 4.343 4.560 - 2.951 3.290 5.670 13.977LNFDt 121.809*** 86.297*** 27.560*** - 56.206*** 31.867*** 296.170***LNENt 19.273*** 6.123 3.447 7.885 - 11.227** 86.854***LNTOt 928.519*** 591.780*** 342.228*** 584.203*** 930.637*** - 5000.243***

Note: 1. ***, **, * show significance at 10%, 5% and 1% levels of significance. 2. The optimum lag detected is 4 based on AIC, and K+dmax = 5 for the model of economic growth. 3. For the model of income distribution and the model of environmental quality, the number of optimum lag detected is 5 based on AIC, and K+dmax = 6.



Note: 1. 1 refers to the model of economic growth, 2 represents the model of income distribution, and 3 is the model of environmental quality. 2. The line refers to unidirectional causality while the dotted line (feedback hypothesis) refers to bi-directional causality.

Figure 3: Toda-Yamamoto Granger Non-Causality Test

Conclusions

Based on the empirical findings of this research paper, the role of FDI inflows as a driver for sustainable development in Malaysia has been confirmed. The FDI inflows have successfully generated higher growth, improved income distribution and enhanced environmental quality in Malaysia based on the period of study. This outcome indirectly indicates the successfulness of Malaysian government agencies, such as Malaysian Investment Development

Authorities, in selecting and monitoring all the foreign investment activities by providing various technical assistance and supports to the foreign investors so that the spillovers of the FDI inflows were not only seen in the economic development aspect, but also reaching the social and environment aspects. On the other hand, TO, which is mostly monitored by Ministry of International Trade and Industry is only able to generate favorable growth and improve income distribution but has no significant effect on



environmental quality in the country. Thus, in order to achieve the goal of sustainable development for the country, Malaysia, through its government agencies as mentioned above, has to continue to improve its policies to attract more FDIs, while at the same time, monitor closely the trade activities that have led to the degradation of the environmental by adhering closely to the multilaterally agreed set of guidelines among countries participating in the emission targets of Kyoto Protocol. Another alternative is to impose import penalties on some industries that make heavy use of fossil energy sources such as cement, steel, and chemicals. These measures can help to reduce the air pollution that result from trade activities, and thus improve the environmental quality in the country.

Even though FDI has a favorable impact on the three pillars of sustainable development in Malaysia, there is still an issue facing the government, which is, how to ensure a continuous flow of FDI into the country. In this regard, the government needs to continue to come up with more innovative and effective policies from time to time such as revision of Economic Transformation Plan, which was formulated as part of Malaysia’s National Transformation Program that can sustain the higher flow of FDI into the country in the face of competition from other emerging countries which are offering the same or better investment packages to attract FDIs to their countries. To conclude, it is observed that Malaysia has undertaken serious efforts to achieve the goal of sustainable development by introducing various relevant policies and developing comprehensive plans, geared towards meeting the three pillars of sustainable development. The outcome from this research paper has given some insights to the policymakers on the role that FDI and TO can play in helping to achieve sustainable development for the country, thus hopefully, motivating them to come up with more innovative and more relevant policies that would direct towards that end. Furthermore, a better understanding about the potential roles of FDI and TO are important for planning suitable

policies that may be able to promote both short and long run sustainable development plans in this country.

Acknowledgements

The authors acknowledge the helpful comments of anonymous reviewers and participants of the 4th International Conference on Accounting, Business and Economics (ICABEC, 2016) and the valuable suggestions of anonymous referees. The authors would also like to thank the editors of the Journal of Sustainability Science and Management for their helpful assistance throughout the preparation of this manuscript.

References

Amin, S. (1974). Accumulation on a world scale: A critique of the theory of under development. Monthly Review Press, New York. 666 pp.

Ang, J. (2008). Economic development, pollutant emissions, and energy consumption in Malaysia. Journal of Policy Model, 30(2): 271-278.

Antweiler, W., Copeland, B. R., & Taylor, M. S. (2001). Is free trade good for the environment? American Economic Review, 91(4): 877-908.

Anwar, S., & Sun, S. (2011). Financial development, foreign investment and economic growth in Malaysia. Journal of Asian Economics, 22(4): 335-342.

Balasubramanyam, V. N., Salisu, M., & Sapsford, D. (1996). Foreign direct investment and growth in EP and IS countries. The Economic Journal, 106(434): 92-105.

Barro, R. J. (1999). Notes on growth accounting. Journal of Economic Growth, 4(2): 119-137.

Barro, R., & Lee, J. W. (2011). Educational attainment dataset. Retrieved from http://www.barrolee.com, 10 December 2014.



Bhagwati, J. (1994). Free trade: Old and new challenges. Economic Journal, 104(423): 231-246.

Blomström, M., Lipsey, R. E., & Zejan, M. (1994). What explains developing country growth? In W. Baumol, R. Nelson, & E. Wolff (Eds.), Convergence and productivity gross-national studies and historical evidence (pp. 246-262). Oxford: Oxford University Press.

Choi, C. (2004). Foreign direct investment and income convergence. Applied Economics, 36(10): 1045-1049.

Emissions Database for Global Atmospheric Research (2015). CO2 time series 1990-2014 per region/country dataset. Retrieved from http://edgar.jrc.ec.europa.eu/overview.php?v=CO2ts1990-2014, 18 November 2015.

Faustino, H., & Vali, C. (2011). The effects of globalization on OECD income inequality: A static and dynamic analysis. DE working papers; No12/2011/DE. Retrieved from h t t p s : / / w w w. r e p o s i t o r y. u t l . p t /handle/10400.5/3512, 2 April 2016.

Frank, A. G. (1978). Dependent accumulation and underdevelopment. Palgrave Macmillan, London. 226 pp.

Frankel, J., & Rose, A. (2002). An estimate of the effect of common currencies on trade and income. Quarterly Journal of Economics, 117(2): 437-466.

Global Consumption and Income Project (2015). Income inequality dataset. Retrieved from http://gcip.info/graphs/download, 10 February 2016.

Greenwood. J., & Jovanvoic, B. (1990). Financial development, growth, and the distribution of income. Journal of Political Economy, 98(5): 1076-1107.

Hassan, A. A. G. (2004). Growth, structural change, and regional inequality in Malaysia. Ashgate Publishing, Burlington, North Carolina. 245 pp.

Hermes, N., & Lensink, R. (2003). Foreign direct investment, financial development, and economic growth. Journal of Development Studies, 40(1): 142-163.

International Labor Organization (2015). International labor dataset. Retrieved from http://www.ilo.org/global/statistics-and-databases/lang--en/index.html, 8 December 2015.

Kim, D. H., & Lin, S. C. (2009). Trade and growth at different stages of economic development. Journal of Development Economics, 45(8): 1211-1224.

Lean, H. H., & Smyth, R. (2010). CO2 emissions, electricity consumption and output in ASEAN. Applied Energy, 87(6): 1858-1864.

Lee, W. J. (2013). The contribution of foreign direct investment to clean energy use, carbon emissions and economic growth. Energy Policy, 55: 483-489.

Mah, J. S. (2003). A note on globalization and income distribution - The case of Korea. Journal of Asian Economics, 14(1): 161-168.

Martinussen, J. (2004). Society, state, and market: A guide to competing theories of development, 4th edition. Zed Books, London. 386 pp.

Miyamoto, K. (2003). Human capital formation and foreign direct investment in developing countries, OECD Development Centre Working Papers 211. Paris, France: The Organisation for Economic Co-operation and Development.

Mundell, R. A. (1957). International trade and factor mobility. American Economic



Review, 47(3): 321-335.

Narayan, P. K. (2004). Reformulating critical values for the bound F-Statistic Approach to Cointegration: An application to the tourism demand model for Fiji. Department of Economics Discussion Papers. Victoria, Australia: Monash University.

Ndako, U. (2010). Stock markets, banks, and economic growth: Time series evidence from South Africa. The African Finance Journal, 12(2): 72-92.

Pesaran, M. H., Shin, Y., & Smith, R. J. (2001). Bound testing approaches to the analysis of level relationship. Journal of Applied Econometrics, 16(3): 289-326.

Prakash, A., & Potoski, M. (2007). Investing up: FDI and the cross-country diffusion of ISO 14001 Management Systems. International Studies Quarterly, 51(3): 723-744.

Reuveny, R., & Li, Q. (2003). Economic openness, democracy, and income inequality: An empirical analysis. Comparative Political Studies, 36(5): 575-601.

Ridzuan, A. R., Ismail, N. A., Hamat, A. F. C., Md Nor, A. H. S., & Ahmed, E. M. (2017). Does equitable income distribution influence environmental quality? Evidence from developing countries of ASEAN-4. Pertanika Journal of Social Sciences and Humanities, 25(1): 385-400.

Ridzuan, A. R., Ismail, N. A., and Che Hamat, A. F. (2017). Does foreign direct investment successfully lead to sustainable development in Singapore?, Economies, 5(3): 1-20.

Saadatian, O., Salleh, O. M. T. & Dola, K. (2009). Observations of sustainability practices in Malaysian research universities: Highlighting particular

strengths. Pertanika Journal of Social Sciences and Humanities, 17(2): 225-244.

Shahbaz, M., Solarin, S. A., Mahmood, H., & Arouri, M. (2013). Does financial development reduce CO2 emissions in Malaysian economy? A time series analysis. Economic Modelling, 35: 145-152.

Solarin, S. A., & Shahbaz, M. (2015). Natural gas consumption and economic growth: The role of foreign direct investment, capital formation and trade openness in Malaysia. Renewable Sustainable Energy Review, 42: 835-845.

Tang, C. F., & Tan, B. W. (2012). The linkages among energy consumption, economic growth, relative price, foreign direct investment, and financial development in Malaysia. Quantity Quality, 48(2): 781-797.

Toda, H., & Yamamoto. T. (1995). Statistical inference in vector autoregressions with possibly integrated processes. Journal of Econometrics, 66: 225-250.

Tsai, P. L. (1995). Foreign direct investment and income inequality. World Development, 23(3): 469-483.

World Bank (2016). The World Development Indicators 2016 Database. Retrieved from http://data.worldbank.org/data-catalog/world-development-indicators, 5 December 2016.

United Nations Conference on Trade and Development (2012). World Investment Report 2012: Towards a new generation of investment policies. Geneva, Switzerland: United Nations Publication.

Zhang, K. H., & Markusen, J. R. (1999). Vertical multinationals and host-country characteristics. Journal of Development Economics, 59(2): 233-25.




THE INFLUENCE OF NEW ECONOMIC MODEL ON INTENTION TO WORK ABROAD: AN EMPIRICAL STUDY TO ADDRESS BRAIN DRAIN ISSUE IN MALAYSIA

NORHAYATI SHARIFF1, JAYARAMAN KRISHNASWAMY2*, DAHLAN ABDULLAH3 AND ANG WOON CHAU4


2Taylor’s Business School, Taylor’s University, 47500 Subang Jaya, Selangor, Malaysia

3Faculty of Hotel and Tourism Management, Universiti Teknologi MARA, 13500 Permatang Pauh, Penang, Malaysia

4Graduate School of Business, Universiti Sains Malaysia, 11800 USM, Penang, Malaysia


Abstract: Although Malaysia is developing at a faster rate, it faces to some extent the problem of brain drain. Since independence in 1957, more than two million people have emigrated. In 2016, Malaysia’s brain drain issue has reached critical stage. In the present study, the Malaysia’s New Economic Model is used to address the brain drain issue. The purpose of the present study is to explore the factors influencing Malaysian intellectuals’ intention to work abroad that contributes to brain drain issue in Malaysia. A sample of 105 usable survey questionnaires were analysed using partial least square approach via Smart PLS 3.0. The study reveals a significant positive relationship between higher expectations, perceived opportunities and lack of security on the intention of Malaysian intellectuals to settle in abroad for work. Interestingly, the New Economic Model makes Malaysian intellectuals to remain in the country for work due to better prospects for business environment, scope for higher income and economic sustainability.

Keywords: New economic model, brain drain, economic sustainability, push factors, pull factors.

Introduction

Brain drain is a phenomenon that occurs when people who have high qualifications and skills leaving their home country to another country. This phenomenon is not new and has been discussed extensively since the past ten years. This topic has become popular among researchers due to various factors including globalization, politics, economics, sociology, and technology (Baruch et al., 2007). In general, brain drain is a global issue faced by many developing countries. Migration of educated intellectuals and highly skilled professionals has increased significantly since 1960 (Jauhar et al., 2015). Globalization has resulted in the migration of professionals and intellectuals from many developing countries around the world such as Ghana (Dadson & Kato, 2016), Kenya (Nelson et al., 2016), India (Walton-Roberts, 2015), and

China (Ha et al., 2016). Brain drain typically involves migration from a developing country to a more developed country to work (Beine et al., 2001). This phenomenon is also a direct impact of the increasing demand for professionals in the developed country. For instance, Malaysia’s brain drain issue is worsened by the weakening local economy but attractive work opportunities in developed countries like United Kingdom (UK) and Taiwan (Nadaraj, 2016). Hence, if the intellectuals migrate to work in other countries, it will bring negative impacts on the development of their countries of origin. The impact of their migration is the decrease in number of educated and highly skilled employees that is so critical to the productivity and economic growth in developing countries (Lowell & Findlay, 2002). However, when they return to the country of origin, they will bring back new knowledge, skills and experience from abroad that can be

: 101-117

102THE INFLUENCE OF NEW ECONOMIC MODEL ON INTENTION TO WORK ABROAD: AN EMPIRICAL STUDY TO ADDRESS BRAIN DRAIN ISSUE IN MALAYSIA

used to develop and improve their own country (Lowell & Findlay, 2002). Therefore, it is very important for developing countries to retain the talents of their intellectuals and to influence them to return to homeland to assist the country’s development.

Many research has been carried out among intellectuals in different countries about their intentions to work abroad (Tessema, 2010; Tahir et al, 2011; Khoo et al., 2011; Nadaraj, 2016). However, there are only few researches on brain drain being carried out in South East Asia especially in Malaysia, though the country is facing serious brain drain issue (Jauhar & Yusoff, 2011; Nadaraj, 2016; Azman et al., 2016). Thus, the present study develops and tests two separate models to investigate the significant factors influencing Malaysian intellectuals to work abroad, that contributes to brain drain issue. In this study, New Economic Model (NEM) was used as a possible measure to address the issue (National Economic Advisory Council, 2010a). It includes several policies to address the brain drain issue. Based on NEM, the present research develops the second model to investigate the effectiveness of NEM to transform the brain drain issue to brain gain to Malaysia.

Brain Drain

Jauhar et al. (2015) conceptualized ‘brain drain’ as the movement or migration of skilled and talented individual from the country of origin to other country for work. The term ‘brain’ refers to knowledge and skill of highly competent individual. Whereby, the word ‘drain’ implies to the shortage of talented workforce due to migration rate beyond normal level (Bushnell & Choy, 2001). Most of the time, brain drain is perceived to have negative impact especially to the homeland. Scholars argued that the impact is higher on the developing countries because in those countries, the number of skilled workforce is normally limited. Although migration of

workforce also provides positive impact to home country due to remittance sent home by emigrants (Beine et al., 2001; Stark, 2004), the findings are inconclusive. Researchers have already understood that brain drain problem also provides other benefits to the country of origin because it enables the transfer of knowledge from the host country (Dustmann & Kirchkamp, 2002) as well as greater opportunity to boost entrepreneurial initiatives and self-employment (Mesnard & Ravallion, 2001). The migrant normally creates networks abroad, thus, it helps to reduce informational barriers with regards to foreign direct investments (FDIs). As a result, foreign investors have become more interested to invest in their home country (Kugler & Rapoport, 2007). Furthermore, workforce migration has shown to influence better public health and reducing the rate of mortality in the home country. These are key determinants of a country’s productivity (Hildebrandt & McKenzie, 2005).

Brain drain issue in Malaysia has begun since 1960s when many local students went to overseas to pursue for higher education. Upon graduation, many of them denied coming back and work in the home country (Jauhar et al., 2015). According to economic report on brain drain in Malaysia (World Bank, 2011), the Malaysian diaspora has quadrupled over the last three decades. It was reported that almost one million people migrated in 2010. About fifty five percent of the entire diaspora resides in Singapore. The remaining include countries like Australia, Brunei, United Kingdom and United States. Unfortunately, almost a third of all migration is brain drain. At present, Malaysia’s rate of brain drain among skilled workers is three times larger than two decades ago and is growing at the rate of six percent per annum. Based on the literature review, various factors have been found to influence the intention to migrate abroad. In this study, factors associated to brain drain is summarized in Table 1.


103Norhayati Shariff1 et al.


New Economic Environment in Malaysia

According to World Bank (2011), more than one million Malaysians have migrated. While one third of them are highly educated intellectuals, half of them resides in Singapore. Though the overall migration rate has recently slowed down, the migration rate for skilled workforce remains high due to better pay in the neighbouring country. In 2010, the gross national income (GNI) per capita was USD7,900 in Malaysia. In contrast, the GNI per capita in Singapore was USD40,920, almost 5 times higher than Malaysia. Thus, it creates a

challenging moment to Malaysian government to retain the intellectuals and professionals in the home country. Most of the policies and strategies introduced by Malaysian government were inadequate to address the issue (National Economic Advisory Council, 2010a). Hence, in March 2010, the NEM was introduced with the objective of ensuring Malaysia to be recognized as a developed nation by the year 2020. Malaysian citizen is predicted to receive high income level, better quality of life, and sustainable growth by the year 2020.

Table 1: Literature on factors causing brain drain

Factors AuthorsExpectation (high income/

financial advantage/wages)

Kangasniemi et al. (2007); Hendel and Kagan (2010); Connell et al. (2007); Tansel and Gungor (2002); Barrett and Goggin (2010); Tahir et al. (2011); Dahl and Sorenson (2010); Oosthuizen and Ehlers (2007); Kazlauskiene and Rinkevieius (2006); Baruch (1995); National Economic Advisory Council (2010a, 2010b)

Career advancement Kangasniemi et al. (2007); Carr et al. (2005); Connell et al. (2007); Gaiduk et al., (2009); Clark et al. (2006); Lu et al. (2009); Oosthuizen and Ehlers (2007); Kazlauskiene and Rinkevieius (2006)

Access to good technology (training)

Connell et al. (2007); Gaiduk et al. (2009); Tansel and Gungor (2002); Mpinganjira (2009); Imran et al. (2011); Tahir et al. (2011); Oosthuizen and Ehlers (2007)

Working conditions/business environment

Connell et al. (2007); Gaiduk et al. (2009); Tansel and Gungor (2002); Imran et al. (2011); Clark et al. (2006); Oosthuizen and Ehlers (2007); Kazlauskiene and Rinkevieius (2006); Serour (2009); National Economic Advisory Council (2010a, 2010b)

Political/ethnic problems Carr et al. (2005); Connell et al. (2007); Tansel and Gungor (2002); Imran et al. (2011); Tessema (2009); Clark et al. (2006); Tahir et al. (2011); Kazlauskiene and Rinkevieius (2006)

Perceived opportunity Tansel and Gungor (2002); Mpinganjira (2009); Khoo et al. (2011)Sustainability Hendel and Kagan (2011); Carr et al. (2005); Connell et al. (2007); Tansel

and Gungor (2002); Tessema (2009); Khoo et al. (2011); National Economic Advisory Council (2010a, 2010b)

Security/instability Hendel and Kagan (2011); Connell et al. (2007); Tansel and Gungor (2002); Tessema (2009); Oosthuizen and Ehlers (2007)

Social support Baruch et al. (2007); Khoo et al. (2011); Sidebotham and Ahern (2011)

Better quality of life Hendel and Kagan (2011); Clark et al. (2006)


Figure 1: Goals of Malaysia’s New Economic Model(Source: National Economic Advisory Council, 2010)

Figure 1 illustrates the goal of the NEM that is to ensure a successful transformation of Malaysia into a sustainable, inclusive and high earning nation (National Economic Advisory Council, 2010a). The concluding part of the NEM report represents a significant milestone, as it completes the roadmap of the Economic Transformation Program (ETP). ETP is the two-component approach to achieve the economic growth target of NEM. The first component, the National Key Economic Areas, is an effort to identify, priorities and align support for the key sectors and major projects that will drive quantifiable economic growth that is already moving forward. The second component, a set of Strategic Reform Initiatives (SRIs) that is designed to remove blockages to the growth of the country (National Economic Advisory Council, 2010b). Despite NEM, it is challenging to retain Malaysian professionals who are looking forward in moving abroad. Hence, the present study is undertaken to investigate the attraction of new policies which may lead to reduction in the intention of intellectuals and professionals to migrate.

Conceptualization of a Model

Baruch (1995) developed ‘push-pull’ model, a two-factor model to understand the reasons behind brain drain issue. The model provides a clear explanation for cross-border movements of highly educated and skilled individuals. The individual migration decision is influenced by several ‘pull’ factors, including active recruitment by well-known employers, interesting compensation rate, world-class working facilities, and better quality of live. On the other hand, the ‘push’ factors originally come from the home country and consist of factors like political uncertainty, high living cost, and the difficulty to get jobs. One of the major destinations of Malaysian migrants is Singapore (World Bank, 2011). Singapore is believed to provide better education, more job opportunities and higher pay (Gooch, 2010). Besides, the crime rate in Singapore is also lower than Malaysia (Numbeo, 2013). These factors are considered as pull factors in the proposed model (Figure 2a). According to World Bank (2011), other than differences in



Figure 2: Conceptual Research Framework

Hypotheses Development

The ‘expectation’ in Figure 2a is defined as an individual’s belief on the future and prospect that he or she predicts to happen neither optimistic nor pessimistic ways. Jauhar and Yusoff (2011) investigated the factors contributing to brain drain issue among accountants in Malaysia. In line with Khoo et al. (2011), they found that the main reason for most of the skilled workers to leave the home country is to get better promotion and higher salary abroad. Many other studies have also reported that higher pay for working abroad is considered as the most influential factors for professionals to migrate (Tansel & Gungor, 2003; Connell et al., 2007; Carr et al.,

2005; Dahl & Sorenson, 2010; Hendel & Kagan, 2011). Besides low salary, lack of personal growth opportunity and the various barriers to career advancement also influence the highly intellectual and skilled workers to migrate (Oosthuizen & Ehlers, 2007). Accordingly, the following hypothesis is developed:

Hypothesis 1: Malaysian residents’ expectations has a positive relationship with their intention to leave Malaysia to work abroad.

‘Perceived opportunity’ may be defined in this study as an individual perception that he or she has a better chance to gain something. Although many researchers have investigated the link

earnings potential, career prospects, quality of education and quality of life, displeasure with Malaysia’s inclusiveness policies is also a key factor, particularly among the non-Bumiputeras that cause the diaspora. Hence, Malaysia needs to manage the push factors of migration by updating its inclusiveness policies. Today, over 90 percent of all inequality is a function of socio-economic differences within ethnic groups, rather than between them. Productivity and inclusiveness lie at the heart of Malaysia’s transformation programs. Implementing these forcefully, will go a long way towards turning the brain drain into a gain. Due to this, the

Malaysian government introduced NEM with three pillars, namely high income, inclusiveness, and sustainability. These three pillars can be considered as one of the Malaysian government initiative to manage push factors that causing Malaysian diaspora. Therefore, in the present study these push factors are conceptualized into three different variables, namely, sustainability and inclusiveness, business environment, and high income. The second model aims to investigate the influence of these initiatives on the intention of Malaysian intellectuals to work abroad. The model is illustrated in Figure 2b.



between perceived opportunity and intention to go abroad in developed countries, empirical research about this matter is limited in developing countries, especially in South East Asian region. Khoo et al. (2011) and Tansel and Gungor (2003) have reported that working experience abroad affects the individual’s intention to reside abroad. Furthermore, Connell et al. (2007) stated that individual intention to migrate abroad is caused by their intention to upgrade academic qualification, to gain experience, lack of career advancement opportunity, and heavy work load in the country of origin.Thus, the following hypothesis is developed:

Hypothesis 2: Malaysian residents’ perceived opportunity has a positive relationship with their intention to leave Malaysia to work abroad.

The ‘perceived sustainability’ refers to the long-term concerned about economic and social aspects. In this article, sustainability includes concerned about social support, economic and political stability abroad. The socio-economic factors have been recognized as the push factor toward migration (Kazlauskiene & Rinkevieius, 2006). Hence, Serour (2009) suggested that by improving the working conditions, the brain drain issue can be significantly addressed. Political and economic stability are also important in talent retention. It includes the government policy on taxation, minimum pay, and the freedom of movement (Carr et al., 2005). Therefore, the following hypothesis is proposed:

Hypothesis 3: Malaysian residents’ perceived sustainability has a positive relationship with their intention to leave Malaysia to work abroad.

In addition to political and economic stability, growth of a country is highly dependent on the level of security. Therefore, the government should ensure that everyone in a country is protected. Research has shown that, there is a link between national security and migration rate (Hendel & Kagan, 2011; Tansel & Gungor, 2003). In Malaysia, there are concerns about the levels of crime and violence that has led the intent to go abroad increased. According

to Oosthuizen and Ehlers (2007), a work environment that is not secure can also lead to migration. Therefore, it is believed that:

Hypothesis 4: Malaysian residents’ perceived security has a positive relationship with their intention to leave Malaysia to work abroad.

Since Malaysia is not among the world top performing economies, majority of its people are stucked in the middle-income group. The present economic growth has not benefited most of the population, even though it has come with considerable environmental cost. For that reason, the NEM was designed to unleash potential growth of the country and drive Malaysia forward from the present stagnant situation into high income nation with both inclusive and sustainable (National Economic Advisory Council, 2010a). The main aim of the Malaysia NEM is to create a sustainable and high-in come country by the year 2020. The NEM sets the goals such as reducing crime, fighting corruption, improving students’ outcome, strengthening the public sector, improving urban public sector, becoming higher income group, raising living standards of low-income households. These goals were classified into three main constructs namely; sustainability and inclusiveness policy, environment policy, and high-income policy.

‘Sustainability and inclusiveness’ construct in the present conceptual model refers to the aim to meet the present needs without compromising future generation and wealth of the country are shared by all communities. The NEM also consists of strategies to transform private sector to make Malaysia globally competitive for investors. “Business environment’ construct refers to the needs to reform the private sector, to boost private investment by removing distortions and the high costs of doing business, and to rebalance the roles of the public and private sectors. It aims to uphold a level playing field and fair business environment. Finally, in the ‘high income’ economy, the people of Malaysia can expect for more choices and higher purchasing power,



better quality of life, opportunities for upward mobility, reward for innovation and creativity, and greater confidence in the robustness of the economy. Based on these discussion, the following hypotheses are developed to investigate the influence of NEM policies conceptualized in the conceptual model on the intention of Malaysian intellectuals to work abroad or to remain in the country. Accordingly, this study hypothesizes that:

Hypothesis 5: Sustainability and inclusiveness policy of NEM has negative relationship with the intention to leave Malaysia to work abroad.

Hypothesis 6: Business environment has negative relationship with the intention to leave Malaysia to work abroad.

Hypothesis 7: High income has negative relationship with the intention to leave Malaysia to work abroad.

Methodology

The study employed a quantitative research with a primary survey. A structured questionnaire has been used with each question item is measured on a 5-point Likert scale. Scale 1 means strongly disagree and scale 5 means strongly agree. The content validity was considered in the study to measure the representativeness and comprehensiveness of the measurement items used to measure the models’ constructs. To ensure content validity is established, a literature review was carried out to identify the dimensions for each construct. The items for the constructs were developed based on Akl et al. (2007), Baruch et al. (2007), Mansour (2008), and National Economic Advisory Council (2010a), thus satisfying content validity. In addition, the subject experts in the areas of brain drain and NEM were contacted for face validity of the developed survey questionnaire. The comments and suggestions provided by the experts were used to enhance the survey questionnaire. Following Viechtbauer (2015), a pilot study and an exploratory factor analysis (EFA) were carried out to identify the possible factor structure. Fifty-six respondents participated

in the pilot study and the principal component method with varimax rotation was used in the analysis. Based on the EFA, two measurement items were deleted, and the remaining items were categorized into three constructs namely sustainability and inclusiveness policy, business environment policies, and high-income policy. The main study was then, conducted in Malaysia by administering the questionnaire in the hard copy and emails to the Malaysian citizens who are currently studying either undergraduates or have minimum qualification of degree and above in local universities. Out of 275 questionnaires distributed, 105 respondents from different parts of Malaysia have correctly filled in the questionnaire and were used for data analysis. The response rate reported was 38.2%.

The study applies Partial Least Squares (PLS) method of analysis using Smart PLS 3.0 software (Ringle et al., 2014). Nonparametric bootstrapping (Wetzels et al., 2009) was applied to 5,000 resamples as suggested by Hair et al. (2014). This technique has been applied to investigate the appropriateness to the exploratory nature of the study where some of the hypothesised relationships between the variables have not been previously tested. Furthermore, PLS is an appropriate method to overcome the limitations of covariance-based SEM with respect to sample size, measurement model and model complexity, (Wetzels et al., 2009). The sample size needed for the study depends on the number of studied variables and the number of questions in the questionnaire. Barclay et al. (1995), postulated by Chin (1998), suggested that the sample size depends on the number of predictors that are involved in the multiple regression models. Consequently, one must look for (a) the largest number of formative indicators (b) the largest number of independent variables, and (c) the maximum of both the numbers in (a) and (b) multiplied by 10 to get the minimum sample size. In the present study the intention to go abroad has the largest number of predictors in both models (four indicators in the first model and three indicators in the second) and consequently, 105 samples need to be more than sufficient.



Results

Table 2 presents demographic profile of the respondents. The final sample has 50 (47.62%) males and 55 (52.38%) females. The respondents’ age ranged from 20 to 50 years old. However, majority of the respondents were under 30 years old (71.42%). The ethnic group

Podsakoff and Organ (1986) have claimed that common method bias can be problematic towards respondent’s bias especially when a single latent variable represents most the explained variance. In the present study, results of the un-rotated factor analysis show that the first normalized linear combination accounts for 28.61% out of 62.00% total variance in the first model and 36.86% from 67.99% of total variance in the second model. The results indicate that there is no serious issue of common method bias exist in the present research. Hence, the respondent’s bias on the understanding of the questionnaire items was at a minimum level.

Measurement Model

The measurement model needs to be measured for reliability and validity such as main loading,

of the respondents consisted of 48 (45.71%) Malays, 41 (39.04%) Chinese, and 16 (15.23%) Indians. Regarding educational status, 74.28% of the respondents hold bachelor degree, 18.10% had master degree 14.29%, and 7.62% hold doctorate degree.

cross loadings, composite reliability (CR) and average variance extracted (AVE). Igbaria et al. (1995) recommend that a question item can be retained for further analysis if it has a loading of at least 0.5. As presented in Table 3, all loadings for the scales used in this study were larger than 0.5, suggesting that individual item validity was clearly demonstrated. With regards to the reliability test to determine a construct’s internal consistency, Cronbach’s Alpha values must be determined. Recently, however, scholars have recommended the use of Composite Reliability (CR) as a better estimate of Cronbach’s Alpha (Henseler et al., 2009). Thus, the CR values were provided in Table 2, which ranging from 0.772 to 0.882 for the economic model before NEM and from 0.814 to 0.896 for new economic model. Therefore, they are well above the threshold

Table 2: Demographic profile of the respondents

Demographic Frequency (n=105) PercentageGenderMale 50 47.62%

Female 55 52.38%Age

20 – 30 years 75 71.42%31 – 40 years 18 17.14%41 – 50 years 7 6.67%Ethnic group

Malay 48 45.71%Chinese 41 39.04%Indian 16 15.23%

Education levelDoctorate degree 8 7.62%

Master degree 19 18.10%Bachelor degree 78 74.28%



Table 3: Measurement model evaluation for the proposed research model

Construct Items Loadings CR AVEEconomic Model (before NEM)

I chose to work abroad to…….Expectation (E) earn higher income 0.811 0.847 0.583

get a job with a well-paid salary 0.642take opportunity for personal growth and development 0.792

search for better career advancement 0.797

Perceived Opportunity (PO) get better working conditions 0.770 0.857 0.600search for better job opportunity 0.755get access to better research facilities 0.782

experience better scientific training 0.791Sustainability (S) enjoy family benefit 0.922 0.803 0.582

experience higher living standard 0.637experience better social support 0.699

Security (SE) I am safer in abroad 0.595 0.772 0.537relative social security 0.675lower crime rate 0.896

Intention to go abroad (I) I am looking forward to go abroad to work 0.814 0.882 0.651

I have intentions to work abroad 0.839There is a likelihood that I will go abroad to work 0.812

I am confident that I will work abroad 0.759

value of 0.7 (Straub et al., 2004). To establish the convergent validity, the average variance extracted (AVE) for all constructs exceed the 0.5 threshold value (Fornell & Larcker, 1981). Both discriminant validity and convergent validity determine the construct validity.

In order to verify that all constructs in the research framework have differed from each other, discriminant validity was tested. As recommended by Gefen and Straub (2005), the measurement items for each construct must an order of magnitude larger than their loadings on other constructs which have to be satisfied to examine the discriminant validity. Fornell and Larcker (1981) claimed the correlations between

items in any two constructs must be lower than the square root of the average variance shared by items within a construct. As depicted in Table 4, the square root of the variance shared between a construct and its items appearing in bold along the diagonal is greater than the correlations between the construct and any other construct in the model, confirming to Fornell and Larcker’s (1981) criteria for discriminant validity. Thus, given the above analysis, the scales used to measure the constructs in this study demonstrated sufficient evidence of construct validity, internal consistency, and discriminant validity to ensure that the measurement model of the data has been well established.



New Economic Model (NEM)I prefer to stay in Malaysia to……..

Sustainability and Inclusiveness (SI)

achieve its goal of fighting for corruption 0.774 0.867 0.525

achieve its goal for reducing crime 0.746enhance the sources of growth 0.525achieve its goal of inclusiveness 0.721ensure sustainability of growth 0.801build the knowledge base and infrastructure 0.745

Business Environment (BE) build a viable, market-friendly and favorable position 0.786 0.896 0.682

improve the public sector 0.806create a competitive domestic economy 0.836

reform the private sector 0.874

High Income (HI) raise standard of living among low-income earners 0.853 0.814 0.687

achieves its goal of becoming higher income group 0.803

Intention to go abroad in the light of NEM (NI)

There is a likelihood that I will work abroad 0.831 0.855 0.597

I am confident that I will work abroad 0.728

I have intention to go abroad to work 0.817

I am sure that I will be going abroad to work 0.709

Note : CR= Composite Reliability; AVE= Average Variance Extracted

Table 4: Summary of the Discriminant Validity

Economic model (before NEM) New economic modelE I PO S SE SI BE HI NI

E 0.763 SI 0.724I 0.708 0.807 BE 0.381 0.826

PO 0.601 0.628 0.775 HI 0.378 0.466 0.829S -0.017 0.098 0.008 0.763 NI -0.427 -0.662 -0.560 0.773

SE 0.434 0.601 0.498 0.014 0.733

Note: Diagonals represent the square root of the AVE and the off-diagonals represent the correlations.



Structural Model

Since results of the measurement model have been satisfied, the next step is the evaluation of the structural model to determine whether the proposed hypotheses in the present study are supported. Therefore, by using Smart PLS method, the explanatory power of the research model was examined. It was found that total explainable variation (R2) was 64.1% for the first model, thus highly significant statistically. Besides, 53.5% of the variance in Malaysian intention to go abroad has been explained in the light of NEM and was found to be highly statistically significant. Other than estimating the R2, scholars have proposed the measurement of predictive relevance as an additional model fit assessment (Stone, 1974). This estimate represents the model adequacy, which is to predict the manifest indicators of each construct. Using the blindfolding procedure, the cross-validated redundancy (Stone-Geisser Q2) was determined to identify the predictive relevance. Based on Chin (2010) recommendation, a model has predictive relevance if the Q2 value

Figure 3a

is greater than zero. In the present study, a value of 0.390 (for the first model) and 0.257 (for the second model) were obtained which are greater than zero fulfilling the criteria. The non-parametric bootstrapping (Wetzels et al., 2009) using 5,000 resamples were used to evaluate the structural model. The significance and relative strength of the pull factors in host countries were evaluated (Figure 3a). The results reveal that expectation (β=0.457, p<0.001), perceived opportunities (β=0.203, p<0.01), and security (β=0.300, p<0.001) have significantly and positively influence on Malaysian intellectuals’ intention to go abroad for work. Hence, H1, H2, and H4 are supported whereas H3 is not supported. Furthermore, the analysis of the second proposed model (Figure 3b) reveals that sustainability and inclusiveness (β=-0.137, p<0.05), business environment (β=-0.477, p<0.001), and high income (β=-0.286, p<0.01) have significantly and negatively influence on intention to go abroad. Therefore, H5, H6, and H7 are all supported.



Figure 3b

Path Beta t-value DecisionH1 E → I 0.457 7.135*** SupportedH2 O → I 0.203 2.771** SupportedH3 S → I 0.100 1.243 Not SupportedH4 SE → I 0.300 4.783*** SupportedH5 SI → NI - 0.137 1.667* SupportedH6 BE → NI - 0.477 3.743*** SupportedH7 HI → NI - 0.286 2.860** Supported

Note ; *p<0.05, **p<0.01, ***p<0.001

Discussions and Implications

About one million Malaysian citizens are working abroad for better opportunities, lifestyle and higher standard of living. For instance, most of the educated intellectuals were in Singapore since Singapore has been doing well in currency rate as compared to Malaysia. Furthermore, the sustainable growth in terms of economy, social and environment are brighter at Singapore. The various measures taken by the government in Malaysia do not prevent people to go abroad mainly due to insufficient opportunities for highly trained professionals and skilled executives in Malaysia. In addition, those Malaysian citizens staying abroad may not wish to come back to their mother land as they enjoy multiple benefits outside. One

of the primary objectives of NEM is to bring Malaysian workforce to a higher income group and retain the highly talented professionals and intellectuals within Malaysia.

The main aim of the present research is to examine the factors influencing Malaysian intellectuals’ intention to leave Malaysia for work abroad. Results of the PLS analysis suggest a significant positive relationship between higher expectations such as higher pay, greater personal growth opportunity, or better career advancement and Malaysian intellectuals’ intention to go abroad for work. Malaysian intellectuals remarked that the lack of opportunity and low wages in Malaysia were the reasons to work abroad. Besides, lack of opportunity for career advancement in



Malaysia has influenced them to migrate abroad. As expected, findings of the present study are consistent with the literature (Kangasniemi et al., 2007). As one of the implications, like other developed nation, Malaysian government and private sector can consider seriously to invest more in high-tech industries and put greater effort on research and development to boost the Malaysia economy. Thus, income of the people can be increased and number of people looking forward to work abroad can be reduced.

Finding of the present study also shows a significant and positive relationship between perceived opportunities abroad such as better working conditions, better job opportunity, better training opportunity and intention to work abroad. The multinationals companies (MNCs) are dominating in Malaysia as compared to local Malaysian companies. Undoubtedly, it fetches more FDI to Malaysia, but the FDI has been considerably reducing in the recent years, which is a real concern to Malaysia. Furthermore, there is a wide gap in the tax structure for MNCs and local Malaysian companies, which makes local business more difficult. Hence, the time has come to concentrate and enhance the local Malaysian business to sustain self-contended economic growth. At the same time, more effort needs to be given to attract FDI and offer good packages to the incoming professionals in future. Besides, to retain real talents in the home country, government agencies and private sector need to offer the appropriate trainings to enhance the knowledge and skill of Malaysian intellectuals by offering them competitive remuneration and incentives. In addition, the government may encourage entrepreneurs and small and medium enterprise (SMEs) by giving some special subsidy in the tax structure. Although Malaysia is a peace-loving country with relatively stable political and economic situation, there is a significant positive relationship between security situations of other visiting countries and resulted in the intention of professionals to leave Malaysia. These findings are consistent with the previous study by Oosthuizen and Ehlers (2007).

Regardless of the influence on the pull factors that have been discussed earlier, the result also indicates that the NEM has negative influence on the intention to work abroad among Malaysian intellectuals. The result suggested that Malaysians have shown a significant level of confidence in the government. They believe that the NEM will be able to transform Malaysia into a higher income economy, which can finally be turned into higher income nation, improved business environment, better management and preservation of resources, and fair distribution of wealth among its people. Therefore, for the people to feel involved in a developed and high-income economy, they would expect to live and work in safe surrounding, have an equal and easy access to information, practice mutual respect and hold to individual dignity. On the other hand, the people should expect that the various benefits of the NEM policy will only be realized when the policy measures are consistently and fully implemented over a period of time. In the meantime, some segments of the population may perceive greater negative impact than benefit, or that they are receiving less benefits than others. The government must be able to convey the ultimate equity of the benefits over time, urge a commitment to the process, and create a vision of the long-term common good. Once the full benefits of the NEM policy have been realized, brain drain issue would be able to be managed.

Conclusions

The emergence of NEM in Malaysia during the year 2010 is not only a boom to most Malaysian nationals but also to the country in terms of prospects and progress. The factor analysis for the components of NEM splits the model into three factors namely; business environment, sustainability and inclusiveness, and high-income policy. All these factors were found to be statistically significant which reveal that the Malaysian intellectuals have confidence in NEM and have intention to remain in the home country to work for better prospects. The long-term effect of NEM can be seen in the light of brain gain to Malaysia since it supports the



government efforts to transform the Malaysians, especially the middle class socio-economic group into higher income group. However, other than retaining the existing intellectuals in the home country, future studies are recommended to investigate whether NEM can influence some of the Malaysian intellectuals and professionals staying abroad to go back and work in the home country.

Acknowledgement

The authors would like to express their sincere gratitude to all reviewers for lending their constructive feedback and unconditional support. We are also thankful to all our respondents who have participated in our survey.

References

Akl, E.A., Maroun, N., Major, S., Afif, C., Chahoud, B., Choucair, J., Sakr, M. & Schunemann, H.J. (2007). Why are you draining your brain? Factors underlying decisions of graduating Lebanese medical students to migrate. Social Science & Medicine, 64(6): 1278-1284.

Azman, N., Sirat, M., & Pang, V. (2016). Managing and mobilizing talent in Malaysia: Issues, challenges and policy and implications for Malaysian universities. Journal of Higher Education Policy and Management, 38(3): 316-332.

Barclay, D., Thompson, R., & Higgins, C. (1995). The Partial Least Squares (PLS) approach to causal modeling: Personal computer adoption and use an illustration. Technology Studies, 2(2): 285-309.

Barrett, A., & Goggin, J., (2010). Returning to the question of a wage premium for returning migrants. National Institute Economic Review, 213(1): R43-R51.

Baruch, Y. (1995). Business globalization: The human resource management aspect.

Human Resource Management, 14(4): 313-326.

Baruch, Y., Budhwar, P. S., & Khatri, N. (2007). Brain drain: Inclination to stay abroad after studies. Journal of World Business, 42(1): 99-112.

Beine, M., Docquier, F., & Rapoport, H. (2001). Brain drain and economic growth: Theory and evidence. Journal of Development Economics, 64(1): 275–289.

Bushnell, P. and Choy, W. (2001). ‘Go West, Young Man, Go West!’. People and Place, 9(3): 66-83.

Carr, S. C., Inkson, K., & Thorn, K. (2005). From global careers to talent flow: Reinterpreting “brain drain”. Journal of World Business, 40(4): 386-398.

Chin, W. W. (1998). The partial least squares approach to structural equation modelling. In A. G. Marcoulides (Eds.), Modern Methods for Business Research (pp. 295-336). New York : Psychology Press.

Chin, W. W. (2010). How to write up and report PLS analyses. In V. E. Vinzi, W. W. Chin, J. Henseler, & H. Wang (Eds.), Handbook of partial least squares: Concepts, methods and applications in marketing and related fields (pp. 655–690). Berlin : Springer.

Clark, P. F., Stewart, J. B., & Clark, D. A. (2006). The globalization of the labor market for healthcare professionals. International Labour Review, 15(1-2): 37-64.

Connell, J., Zurn, P., Stilwell, B., Awases, M., & Braichet, J. M. (2007). Sub-Saharan Africa: Beyond the health worker migration crisis? Social Sciences & Medicine, 64(9): 1876-1891.

Dadson, I., & Kato, R. R. (2016). International



remittances and brain drain in Ghana. Journal of Economics and Political Economy, 3(2): 211-241.

Dahl, M. S., & Sorenson, O. (2010). The migration of technical workers. Journal of Urban Economics, 67(1): 33-45.

Dustmann, C., & Kirchkamp, O. (2002). The optimal migration duration and activity choice after re-migration. Journal of Development Economics, 67(2): 351 – 372.

Fornell, C. & Larcker, D.F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1): 39-50.

Gaiduk, R., Gaiduk, J., & Fields, D. (2009). Limiting the brain drain: Determinants of employee organizational attachment in Lithuania. Baltic Journal of Management, 4(2): 149–168.

Gefen, D. & Straub, D. (2005). A practical guide to factorial validity using PLS-graph: Tutorial and annotated example. Communications of the AIS, 16: 91-109.

Ha, W., Yi, J., & Zhang, J. (2016). Brain drain, brain gain, and economic growth in China. China Economic Review, 38: 322-337.

Hair, J. F., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2014). A primer on partial least squares structural equation modelling (PLS-SEM). Sage, Thousand Oaks, California. 384 pp.

Hendel, T., & Kagan, I. (2011). Professional image and intention to emigrate among Israeli nurses and nursing students. Nurse Education Today, 31(3): 259-262.

Henseler, J., Ringle, C.M. & Sinkovics, R.R. (2009). The use of partial least

squares path modeling in international marketing. Advances in International Marketing, 20: 277-320.

Hildebrandt, N., & McKenzie, D. (2005). The effects of migration on child health in Mexico. World Bank Policy Working Paper No. 3573. Washington, USA: World Bank.

Imran, N., Azeem, Z., Haider, I., Amjad, N., & Bhatti, M. (2011). Brain drain: Post graduation migration intentions and the influencing factors among medical graduates from Lahore, Pakistan. BMC Res. Notes, 4: 417.

Jauhar, J., Ghani, A. B. A., Joarder, M. H. R., Subhan, M., & Islam, R. (2015). Brain drain to Singapore: A conceptual framework of Malaysian’s diaspora. The Social Sciences, 10(6): 702-711.

Jauhar, J., & Yusoff, Y. M. (2011). Brain drain: Propensity to leave by Malaysian professionals. International Journal of Innovation, Management and Technology, 2(2): 119-122.

Kangasniemi, M., Winters, L. A., & Commander, S. (2007). Is the medical brain drain beneficial? Evidence from overseas doctors in the UK. Social Science & Medicine, 65(5): 915-923.

Kazlauskiene, A., & Rinkevieius, L. (2006). Lithuanian “brain drain” causes: Push and pull factors. Engineering Economic, 1(46): 27-37.

Khoo, S. E., Hugo, G., & McDonald, P. (2011). Skilled migration from Europe to Australia. Population, Space and Place, 17(5): 550-566.

Kugler, M., & Rapoport, H. (2007). International labor and capital flows: Complements or substitutes? Economics Letters, 94(2): 155-162.

Lowell, B. L., & Findlay, A. M. (2002).



Migration of highly skilled persons from developing countries: Impact and policy responses (Synthesis report). International Labour Office, Department for International Development, United Kingdom. 41 pp.

Lu, Y., Zong, L., & Schissel, B. (2009). To stay or return: Migration intentions of students from People’s Republic of China in Saskatchewan, Canada. Int. Migration & Integration, 10(3): 283-310.

Mansour, A. Y. (2008). Factors affecting locational decisions of Saudi health care professionals. Thesis of Doctoral Degree, University of Illinois.

Maslow, A. H. (1943). A theory of human motivation. Psychological Review. 50(4): 370-396.

Mesnard, A., & Ravallion, M. (2001). Wealth distribution and self-employment in a developing country. CEPR Discussion Paper No. DP3026. London, United Kingdom: Centre for Economic Policy Research.

Mpinganjira, M. (2009). Comparative analysis of factors influencing the decision to study abroad. African Journal of Business Management, 3(8): 358-365.

Nadaraj, V. (2016). Malaysia’s brain drain reaching critical stage. Retrieved from http://www.aseantoday.com/2016/03/malaysias-brain-drain/, 9 January 2017.

National Economic Advisory Council (2010a). New Economic Model for Malaysia. Part 1: Strategic Policy Directions. Putrajaya, Malaysia: National Economic Advisory Council.

National Economic Advisory Council (2010b). New Economic Model for Malaysia: Concluding Part. Putrajaya, Malaysia: National Economic Advisory Council.

Nelson, M., Judith, O., Okurut, D., Ewk, N., Metto, E., Sammy, K., Gracie, C., & Makewa, L. N. (2016). Brain drain in private universities in Kenya: Seeking for an innovative solution. Journal of Advance in Social Science and Humanities, 2(12): 20171-20785.

Numbeo (2013). Crime comparison between Malaysia and Singapore. Retrieved from https://www.numbeo.com/cr ime/compare_count r ies_resul t . j sp?country1=Malays ia& country2=Singapore, 9 January 2016.

Oosthuizen, M., & Ehlers, V. (2007). Factors that may influence South African nurses’ decision to emigrate. Health SA Gesondheid, 12(2): 14-26.

Ringle, C. M., Wende, S., & Becker, J. (2015). SmartPLS 3. Bönningstedt: SmartPLS. Retrieved from http://www.smartpls.com, 23 August 2016.

Serour, G. I. (2009). Healthcare workers and the brain drain. International Journal of Gynecology and Obstetrics, 106(2): 175-178.

Sidebotham, M., & Ahern, K. (2011). Factors influencing midwery migration from the United Kingdom to Australia. International Nursing Review, 58(4): 498-50.

Stark, O. (2004). Rethinking the brain drain. World Development, 32(1): 15-22.

Stone, M. (1974). Cross-validatory choice and assessment of statistical predictions. Journal of the Royal Statistical Society. Series B (Methodological), 36(2): 111-147.

Straub, D., Boudreau, M.C. & Gefen, D. (2004). Validation guidelines for IS positivist research. Communications of the Association for Information Systems, 13(24): 380-426.



Tahir, M. W., Kauser, R., & Tahir, M. A. (2011). Brain drain of doctors: Causes and consequences in Pakistan. World Academic of Science, Engineering and Technology, 51: 406-412.

Tansel, A., & Gungor, N. D. (2003). “Brain drain” from Turkey: Survey evidence of student non-return. Career Development International, 8(2): 52-69.

Tessema, M. (2010). Causes, challenges and prospects of brain drain: The case of Eritrea. International Migration, 48(3): 131–157.

World Bank (2011). Malaysia economic monitor: Brain drain. Washington, USA: World Bank.

Viechtbauer, W., Smits, L., Kotz, D., Budé, L., Spigt, M., Serroyen, J., & Crutzen, R. (2015). A simple formula for the calculation of sample size in pilot studies. Journal of Clinical Epidemiology, 68(11): 1375-1379.

Walton-Roberts, M. (2014). International migration of health professionals and the marketization and privatization of health education in India: From push-pull to global political economy. Social Science & Medicine, 124: 374-382.

Wetzels, M., Odekerken-Schroder, G., & van Oppen, C. (2009). Using PLS path modeling for assessing hierarchical construct models: Guidelines and empirical illustration. MIS Quarterly, 33(1): 177-195.




THE INFLUENCE OF ENERGY CONSUMPTION, RENEWABLE ENERGY AND ECONOMIC GROWTH ON CO2 EMISSION IN MALAYSIA AND INDONESIA

IZYAN NAZIRAH WAHID1, AZLINA ABD. AZIZ2*, MAHIRAH KAMALUDIN2 AND ZURAINI ANANG2

1Academic Department, President College (Sultan Ismail), 50400 Kuala Lumpur, Malaysia2School of Social and Economic Development,

Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia


Abstract: The main objective of this study is to analyse linkages between carbon dioxide emissions, consumption of energy resources and economic growth in Malaysia and Indonesia between 1980 and 2011. With a motivation to recognize the potential effects of energy consumption and economic growth in driving the climate change pressure, it further aims to examine the Environmental Kuznet Curve hypothesis. Annual data over 1980-2011 period were used to examine the long-run relationship and the causality between the variables. The findings revealed the presence of long-run relationship among variables. In addition, analysis on causality tests suggests that there is a causality run from energy use to renewable energy in Malaysia. The results also provide evidence of causality running from renewable energy to carbon dioxide emissions and to economic growth in Indonesia. However, the results of the Environmental Kuznet Curve hypothesis show no evidence in favor to this hypothesis for both countries. The implication of this study is that in order to reduce emission without sacrificing economic growth, policies should be aimed at promoting renewable energy and energy efficiency.

Keywords: Carbon dioxide emissions, energy resources, economic growth, Environmental Kuznet Curve, sustainable consumption.

Introduction

It is widely recognized that energy resources are vital to the world’s economic development. They are essential ingredients in most of the sectors of existing economies. Energy is demanded as one of the products that a consumer (individuals, households, producers or firms) decides to buy for utility maximization. For instance, energy provides the consumer with heat and electricity for lightning and domestic appliances in the residential sector, and to power manufacturing products in the industrial sector. Furthermore, energy is considered as important factor for production on top of other inputs such as labor, materials and capital (Stern, 2010). However, a negative outcome of energy consumption is the emission of carbon dioxide (CO2) through mainly from industrial activities, economic growth and increase in the world population. To

date, issues on increased level for CO2 emissions are well discussed by most countries in the world because it has become a major contributor to the problem of climate change (Sebri & Ben-Salha, 2014, Al-Mulali et al., 2013).

Economic growth in Association of Southeast Asian Nations (ASEAN) countries have been well progressing in recent years. Up to now, Singapore ranked the highest per capita income among the ASEAN countries, followed by Malaysia and Indonesia (World Bank, 2015). As a developing country, Malaysia has experienced outstanding growth in economic progress for the past three decades. As a leading exporter of commodities such as natural gas and palm oil in the world, economic activities in Malaysia are globalized and there is a connection between energy consumption and economic development that contribute to environmental

: 119-133

120THE INFLUENCE OF ENERGY CONSUMPTION, RENEWABLE ENERGY AND ECONOMIC GROWTH ON CO2 EMISSION IN MALAYSIA AND INDONESIA

problems, which can be translated to the emission of carbon dioxide. Besides, there are structural changes in the economy which is undergoing an economic transition from different sectors. For instance, a change from agriculture to industry sector, and then to services sector which served a significant role for emerging growth in energy consumption. The industry sector is the leading energy user compared to other sector and its energy consumption is very high indeed (Energy Information Administration, 2017). Therefore, industry sector contributes considerably to the CO2 emissions. As reported by Carbon Dioxide Information Analysis Centre (2012), a statistics of CO2 production in 2009 shows that Malaysia, with an estimate of 198,348 (Kt), ranked 24th from 188 countries in the world.

On another side, Indonesia, which is the most populous country in Southeast Asia, is also experiencing strong economic growth that has headed to a surge in the total energy consumption throughout the country (World Bank, 2014). International Energy Agency (2015) reported in 2014, Indonesia ranked at the 7th place as the largest exporter for natural gas and was the main exporter of coal in the world. Oil, gas, and coal are the three major resource wealth that contribute to domestic growth and foreign exchanges earning. Economic activities in Indonesia are also heavily dependent on energy consumption and this inevitably contributes to the CO2 emissions. In 2008, CO2 emission in Indonesia was reported at 406,028.58 (Kt) (World Bank, 2014). As a result of being seen as a major contributor of CO2 emissions, Indonesia has committed to decrease the greenhouse gases (GHG) emissions ranging from 41% to 26% (Copenhagen Accord, 2009).

Energy use has been identified as one of the factors that contributes to the CO2 emission which subsequently gives a negative impact to the environment. The dependency on fossil fuels in energy consumption is driving the continued rise in CO2 emissions thus leading to climate change. As can be shown from Figure 1 and 2, there is an increasing trend of energy consumption for both countries. The figures

also reveal that over time, CO2 emissions increase continuously. In general, when energy consumption increases, CO2 emissions increase dramatically. A rapid increase in energy use will increase CO2 emissions which subsequently gives a negative impact to the environment. Prior studies have confirmed that energy use was associated to economic development and leads an upsurge in CO2 emissions as evidenced by Jalil and Mahmud (2009) and Chang (2010) in the case of China; Ghosh (2010) in the case of India; Azlina and Nik Mustapha (2012), Chandran and Tang (2013), Begum et al. (2015) in the case of Malaysia; Chandran and Tan (2013) in the case of both China and India and Tang and Tan (2015) in the case of Vietnam.

Since economic development may have a contrary effect on the climate change in the long term, the negative impact on the burning of fossil fuel on the environment is slowly shifting the world economies towards using renewable energy (Jamaluddin et al., 2013). In recent years, renewable energy, such as wind, hydropower, geothermal, biomass and solar have been used as alternatives for non-renewable energy resources. In fact, ASEAN countries as a whole are rich not only in fossil fuels which are not renewable such as coal, natural gas and petroleum but also the vast potential in renewable resources (Lidula et al., 2006). Climate change concerns coupled with energy scarcity and an increasing cost of energy are strong impetus behind increased demand of global renewable energy resources. Renewable energy offers the opportunity for climate mitigation, contributing to sustainable development.

Over the past centuries, studies on the Environmental Kuznets Curve (EKC) revealed that environmental degradation occurs in tandem with economic growth. The EKC hypothesis suggests that in the early stages of economic growth degradation and pollution increases, but beyond a certain level of income per capita, which will vary for different indicators, the trend tend to reverse, so that at high income levels economic growth leads to environmental improvement (Stern, 2004). Current literature


121Izyan Nazirah Wahid1 et al.

on EKC hypothesis shows that there are linkages between environmental degradation and economic growth. A previous study by Azam and Khan (2016) provided further support to the EKC hypothesis which proposed that in the event of increases in income, it will also results in increases in pollution up to a certain level and then decreases. In addition, the study by Jebli et al. (2016) provided more proof to the EKC hypothesis in which the findings supported the causal relationship between income and carbon dioxide emissions. In fact, the increase in energy consumption increases carbon dioxide

emissions both in the short run and the long run. The extension study of economic growth-environmental pollution causality nexus and energy consumption-economic growth causality nexus has convinced economists to understand the role of economic growth and energy consumption in climate change. Therefore in this study, it is necessary to analyze EKC hypothesis between economic growth, energy consumption, CO2 emissions and renewable energy to determine the effects of energy consumption, renewable energy, and economic growth on CO2 emissions.

Figure 1: Carbon dioxide emissions per capita and energy consumption per capita in Malaysia, 1980-2011 (Source: World Bank, 2014)

Figure 2: Carbon dioxide emissions per capita and energy consumption per capita in Indonesia, 1980-2011 (Source: World Development Indicator, 2014)



Thus, this study refers to two major research problems as a motivation for investigating the empirical links between CO2 emissions, energy resource consumption and economic growth. Firstly, as ASEAN has experienced spectacular economic development in recent years, then industrialization and services plays an essential role for the emerging development in energy consumption. Since Malaysia and Indonesia are developing countries in which most electricity sectors are particularly dependent on energy consumption such as electricity use; this sector contributes to the emission of greenhouse gas. As a consequence, due to the existence of causality between energy use and economic growth, the growing use of energy particularly affects economic growth while considerably contributes to environmental degradation. Hence, this study finds it is noteworthy to focus on causality between CO2 emissions along with energy resource which comprise of renewable and non-renewable energy consumption and economic growth, given the fact that energy consumption is increasing with both economic growth and CO2 emissions. So, as to assume the lower of CO2 emissions was a result of increasing use of renewable sources, the current study considers the renewable energy consumption as one of the efforts to avoid environmental pollution. An analysis of renewable energy use and carbon dioxide emissions nexus proposes that the relationship between the two can either be a positive or a negative relationship. A positive relationship suggests the higher renewable energy use will increase the carbon dioxide emissions while a negative relationship indicates that the higher renewable energy use will lower the carbon dioxide emissions. Besides that, it is also essential to examine the EKC hypothesis because an increase in income per capita may have a significant contribution to environmental degradation. For that reason, testing EKC hypothesis is needed given the remarkable growth rate for these particular ASEAN countries. The findings are anticipated to provide an imperative reference in formulating long term energy policies.

Furthermore, with the growing concern about GHG emissions and climate change, a number of developing economies have started to explore the use of renewable energy as an alternative to the current energy sources and to reduce the environmental problem as well. In fact, countries in ASEAN region are richly endowed with renewable energy sources. Renewable energy sources are a good alternative for electricity generation in ASEAN region. ASEAN countries as a whole are rich in energy sources ranging from the oil reserved, natural gas, and coal to the large potential in renewable energy (Lidula et al., 2006). Among the renewable energy sources available in the region are wind, hydro, and geothermal. Therefore, this study also takes a look at renewable energy use in the attempt to analyze the impact of renewable energy on carbon dioxide emission. This approach is of interest because most of previous literature still paid less attention to the renewable energy use as a variable that might affect the findings of this study.

The linkage between energy use and economic growth has attracted a considerable body of literature over the past century. This linkage is closely associated not only to analyze the causality between the two variables, but also helps to examine the existence of the EKC. In order to examine economic growth-energy consumption nexus, previous studies agreed to propose that there are four main hypotheses related to the nexus. First, is the ‘Growth hypothesis’ - the validity of this hypothesis is confirmed when causality is found to run from energy use to economic growth, which implies energy use is vital in economic development. In this situation, limitations on energy use could affect economic growth negatively. In contrast, an increase in energy use could provide benefits to the development of economics. Under this hypothesis, energy consumption plays as a controlling factor to economic development. Any shocks to the supply of energy will negatively correlated with economic development. The second hypothesis refers to ‘Conservation hypothesis’. It refers



to a situation where economic development contributes to energy use. It is valid if there is causality runs from economic development to energy use. Following this, a policy related to energy conservation may be executed with minimum effect on economic development. Concisely, it proposes an energy independent economy. Next is the ‘Feedback hypothesis’. This hypothesis indicates a mutual relationship between economic development and energy use. The hypothesis is valid with the existence of bi-directional causality between energy use and economic development. In this regard, both variables are mutually determined and affected at the same period. For instance, reduction on economic growth may decrease energy consumption. Similarly, any changes in energy use will be transmitted to economic growth. Finally, ‘Neutrality Hypothesis’ is supported by the nonexistence of causality between energy use and economic development. It suggests that energy use does not affect economic development. Thus, neither conventional nor extensive policies which refer to energy use will have any influence on economic development.

Numerous empirical evidences support the growth hypothesis. This hypothesis exists in energy dependent countries as evidenced by Soytas and Sari (2003) in the case of Turkey, France, Japan and Germany, Lean and Smyth (2010) in the case of Indonesia, Malaysia, Philippines, Singapore and Thailand, Tsani (2010) in the case of Greece and Tiwari (2011) in the case of India. According to Lean and Smyth (2010), consumption of energy will results in direct and indirect effect. For instance, the direct effect of energy use from commercial and industrial sectors will generate economic growth at a higher rates, while higher consumption of electricity which a results from an increase in energy production, will generate indirect effect in energy services by creating employment and providing infrastructure.

Conversely, conservation hypothesis proposes that economic development causes energy use in one way causality. Previous studies have found the support for this hypothesis in

less energy dependent economies as shown by Yoo (2006) in Thailand, Azlina et al. (2014), Azlina (2011) and Ang (2008) in Malaysia, Zhang and Cheng (2009) in China and Ozturk et al. (2010) for low income countries. According to Yoo (2006), economic development causes enlargement in main sectors such as commercial and industrial sectors. In these sectors, electricity consumption has increased because electricity served as major input in factories and large scale plants in the production process.

With respect to the feedback hypothesis, some of the evidence can be found in Chen et al. (2007), Ozturk et al. (2010) and Tang and Tan (2013). They found consistent evidence showing that when energy consumption increases, economic growth will increase while a rise in economic growth directs to an increasing dependence on energy consumption. In contrast, studies by Jobert and Karanfil (2007), Halicioglu (2009) and Soytas and Sari (2009) in Turkey and Payne (2009) in USA reported that energy use and economic development have not given an effect on each other significantly. Their findings proved that no relationship exists among energy use and economic development either in short or long run.

Methodology

Data

This study uses secondary data of carbon dioxide emissions, energy consumption, real GDP per capita and renewable energy extracted from the World Development Indicators 2014 database from World Bank (2014). The annual data covers the period from 1980 to 2011, which correspond to 32 years of data that were used in this study.

Table 1 and 2 show the descriptive statistics of the variables used in the study. CO2 emission in metric tonnes per capita ranges from 6.32 to 42.43 in the case of Malaysia and 0.65 to 1.90 in the case of Indonesia. In Malaysia, real gross domestic product ranges from US$1,729.18 to US$6,512.13 while in Indonesia, it ranges from US$555.93 to US$1,650.63. With



regards to the energy consumption per capita, it ranges from 11,883.61 to 73,023.53 kilogram of oil equivalent in Malaysia, while in Indonesia, it is between 56,251 and 201,999. Finally, the renewable energy use ranges from 1,657.93 to 3,503.61 metric tonnes of oil equivalent in Malaysia and ranges from 0.651 to 1.902 in Indonesia.

Descriptive results showed a negatively skewed distribution between all the variables with skewness values at -0.2920, -0.2622, -0.1836, and -1.3508 respectively in the case of Malaysia. Skewness is a measure of asymmetry and describes the shape of the Probability

Distribution Function. If the skewness value is positive, the probability distribution function is at right or positively skewed. If it is negative, it is at left or negatively skewed. All of the variables showed suitable kurtosis since the value is not larger than 3. Meanwhile, Jarque-Bera normality test showed that CO2, energy consumption, GDP, and renewable energy are normally distributed. With respect to Indonesia, the descriptive results showed a positively and negatively skewed distribution between all the variables. The Jarque-Bera normality test also showed that CO2, energy consumption, GDP, and renewable energy are normally distributed.

Table 1: Summary of descriptive statistics for each series (Malaysia)

Variable

Carbon dioxide emission (CO2)

(Metric tonnes per capita)

Income per capita (GDP) (2005 constant US$)

Energy consumption (EC) (Kilogram of

oil equivalent)

Renewable energy consumption (RE) (Billion Kilowatt

hours) Mean 107934.7 3896.66 38446.48 2341.781

Median 111060.6 3813.18 37532.97 2375.000 Maximum 208267.3 6512.13 73023.53 2987.000 Minimum 27997.54 1729.18 11883.61 1606.000 Std. Dev. 62465.46 1474.77 20382.34 435.6944 Skewness 0.218384 0.2087 0.250282 -0.049170 Kurtosis 1.662677 1.6856 1.682823 1.828886

Jarque-Bera 2.638931 2.9320 2.647359 1.841572Probability 0.267278 0.2308 0.266154 0.398206

Table 2: Summary of descriptive statistics for each series (Indonesia)

Variable

Carbon dioxide emission (CO2)

(Metric tonnes per capita)

Income per capita (GDP) (2005 constant US$)

Energy consumption (EC) (Kilogram of

oil equivalent)

Renewable energy consumption (RE) (Billion Kilowatt

hours) Mean 1.172 1026.636 128919.3 1.171

Maximum 1.902 1650.629 201999.0 1.902 Minimum 0.65 555.927 56251.00 0.651 Std. Dev. 0.398 312.758 49787.27 0.398 Skewness 0.756100 0.514913 -0.430252 0.626283 Kurtosis 2.887311 2.154690 1.625602 1.878681

Jarque-Bera 3.065931 2.366785 3.505915 3.768371 Probability 0.215895 0.306238 0.173261 0.151953Probability 0.267278 0.2308 0.266154 0.398206



Model Specification

Previous studies show that the EKC hypothesis has generally been used as a baseline estimate model where it is used to find out the relationship

Where t, and ε denote time and error term, respectively; CO2 is per capita carbon dioxide emission, GDP and GDP2 indicate real income per capita and the squared value of real income per capita, respectively.

In order for the EKC to be valid, one would expect α1 is positive and α2 is

In this analysis, we closely follow the methodology of Azlina et al. (2014) which adopted three nexus into a single multivariate framework, (i) economic growth - energy use nexus (ii) economic growth - CO2 emissions

Where, EC is per capita energy consumption and RE stands for renewable energy consumption. The residuals are assumed to be normally distributed and white noise. The sign of α3 is expected to be positive and have been found to affect CO2 emissions, as shown by Azlina et al. (2014). A positive association reflects higher level of energy consumption can enhance the scale of economy and as a consequence, more CO2 emissions will be produced. The sign of α4 is expected to be negative because the use of renewable energy resource does not contribute to environmental degradation and are favorable to environmental quality while their consumptions are not subject to exhaustion. A negative association reveals that renewable energy consumption can reduce the emissions level.

Estimation Method

Following the established procedure, the test of the causal relationship between the series is

among numerous indicators of environmental problem and income per capita. This model can be formally represented as follows:

negative. The combination of these two effects (α1>0α2<0) will produce the inverted U- shaped relationships between CO2 emissions per capita and real income per capita. To assess the impact of energy consumption and renewable energy use on CO2 emissions, the new model for testing the EKC hypothesis can be shown as follow:

nexus, and (iii) renewable energy - CO2 emissions nexus. The following log quadratic EKC equation is used to study the impact of economic development and energy resources consumption on CO2 emissions.

conducted in three stages. First, a test is carried out to ascertain the order of integration in all variables. The presence of unit roots that is when the series is non-stationary in their level form can be examined by applying the Augmented Dickey-Fuller (ADF) and Phillip-Perron (PP) tests. As noted by Granger (1987), to make sure that the procedures of classical econometrics are applied appropriately, one must ensure that the variables of interest do not exhibit unit roots because using non-stationary data to conduct an analysis can lead to spurious results. Next, having established the order of integration in the series, the cointegration test developed by Johansen and Julius (1990) is carried out to investigate the existence of long run relationship between the variables. In the third step, the Vector Error Correction Model (VECM) is estimated to assess the direction of causality between the variables. The VECM equations are shown as follows:



(4)

(5)

(6)

(7)

(8)

Where symbol ∆ indicates first difference, α1 are intercepts and p is the lag lengths determined by the AIC statistics. The error term (ε1t,ε2t,ε3t,ε4t,ε5t) are assumed to be sequentially independent with zero mean and fixed covariance matrix. The terms ECT are the error correction terms. The coefficients of ECT measure speeds of adjustments. These terms were derived from the long run cointegrating relationships. The VECM model was estimated with the variables in first differences and including the long run relationships as ECT terms in the system.

To detect the short run Granger causal relation between carbon dioxide emissions and its determinants, the F test was applied on the first difference of lagged independent variables.

This test is based on the null hypothesis that there is no Granger causality. For instance, the test for Granger causality of economic growth (in equation (4)), is H01: y1i = 0, which implies that economic growth does not cause carbon dioxide emissions. Therefore, equation (4) is used to test the causal relation from economic growth, renewable and non-renewable energy consumption to carbon dioxide emissions. Similarly the hypotheses in equation (5) to (8) examined the sequence hypothesis.

Results and Discussion

Unit Root Test

The results of the unit root test are shown in Table 3. The results show that the null of a unit



root in both ADF and PP tests cannot be rejected in their level form, implying that all variables are non-stationary. Nevertheless, after the first

Johansen-Juselius Cointegration and Long-Run Equilibrium Relationships

Since all series are integrated of order of one, Johansen-Juselius cointegration analysis can be applied for further analysis. The results of the cointegration tests are shown in Table 4. The table reports country by country trace statistics and maximum eigenvalue statistics. The results show evidence of one cointegrating relationship for Malaysia and Indonesia at 5% significance level, implying that there is clear evidence of a long run relationship between these variables.

Given that there is an evidence of cointegration, the long run coefficients of energy resource use and economic development can be derived in the CO2 emission equation. Table 5 summarizes the long run coefficients

difference of each variable, the test suggests that all series are stationary.

and the t-statistics. The results show that the EKC hypothesis is not valid among these two countries. Since Malaysia and Indonesia are developing countries, the EKC hypothesis is not valid because they may not have achieved a level of income to apprehend the inverted U shaped. This finding is consistent with results from previous study conducted by Lean and Smyth (2010), Narayan and Narayan (2010), Chandran and Tang (2013) and Azlina et al. (2014). Conversely, EKC hypothesis is significant in case of developed countries because the developed countries have higher turning point of inverted U shaped than developing countries. Furthermore, higher economic growth may lead to different impacts on environmental quality (Al-Sayed & Sek 2013).

Table 3: Augmented Dickey Fuller (ADF) and Phillips Perron (PP) unit root test results

Country/ Variable Augmented Dickey Fuller (LL) Philips Perron (BW)

At level At first difference At level At first differenceIntercept & Trend Intercept & Trend

MalaysiaCO2 0.8289(1) -7.1806(0)*** -0.0072(0) -7.2570(2)***EC -0.9527(0) -7.0271(0)*** -0.6829(2) -7.0271(0)***

GDP -0.6472(0) -4.4293(0)*** -0.8758(2) -4.4200(2)***GDP2 -0.8998(0) -4.4973(0)*** -1.1291(2) -4.4923(2)***TREG -3.0656(1) -3.7197(7)** -2.9460(27) -6.0549(26)***

IndonesiaCO2 -2.9365(0) -5.8682(0)*** -2.9679(2) -7.5988(10)***EC -0.4959(0) -6.0890(0)*** -0.4959(0) -6.0890(0)***

GDP -1.9022(1) -4.0319(0)*** -1.7519(2) -4.0319(0)***GDP2 -1.9824(1) -4.0053(0)** -1.7913(2) -4.0303(1)***TREG -2.9661(0) -8.5241(0)*** -2.8802(3) -8.8506(3)***

Note: An asterisk ***,** and * represent 1% significance level, 5% significance level and 10% significance level, respectively. LL and BW denote optimum lag length selected based on Akaike Info Criterion and the Bandwidth, respectively.



Table 3: Augmented Dickey Fuller (ADF) and Phillips Perron (PP) unit root test results

Country / Hypothesized no. of CE(s) Trace statistics Maximum eigenvalue statistics

Malaysia

r = 0 79.98896* 34.85670*r ≤ 1 45.13226 21.41627r ≤ 2 23.71600 11.32994r ≤ 3 12.38606 7.692097r ≤ 4 4.693965 4.693965

Indonesiar = 0 86.05905* 41.23018*r ≤ 1 44.82887 22.09054

Note: An asterisk ***,** and * represent 1% significance level, 5% significance level and 10% significance level, respectively.

Dependent variable: CO2 Explanatory variables Coefficients Standard error t-statistics

Malaysia

Constant -9.6753** 4.9010 -1.9741GDP 0.1208 0.0885 1.3657

GDP2 -0.0004 0.0003 -1.2093EC 0.5403 0.3238 1.6690RE 0.0377 0.0484 0.7800

IndonesiaConstant 4.3890* 2.4680 1.7784

GDP -5.3690*** 1.7198 -3.1219GDP2 0.9786*** 0.2749 3.5604

EC 0.5940*** 0.1436 4.1376RE -0.0756 0.0576 -1.3126


Granger Causality Based on VECM

Since all variables are found to be integrated of order one and there is evidence of cointegration, this implies the existence of causal relations between variables. Nevertheless, the direction of causality is not detected at this stage. In order

to identify the direction and causal relationship among variables, the Granger causality test was performed in the VECM. The VECM allows a distinction to be made between short-run causality and long-run causality.



The results for joint Wald F-statistics of the lagged explanatory variables of the VECM which indicates the significance of short run causality is shown in Table 6. Meanwhile, the long run causality is shown by the t-statistics for coefficients of the ECT. The findings of short run causality in Malaysia indicates that F-statistics i) for energy consumption (equation 4) is significant at 5% level respectively, ii) for CO2 emission (equation 5) is also significant at 5% level, respectively and iii) for energy consumption and economic growth (equation 8) is significant at 1% level. The result of the short run causality suggests that causality running from energy consumption to CO2 and renewable energy, CO2 to GDP and GDP to renewable energy in Malaysia. Besides, the results of t-statistics found that the coefficient of ECT is significant in the renewable energy at 1% significance level. This result revealed that there are long run equilibrium relationship exists in renewable energy in Malaysia.

With respect to the emissions-growth

nexus, the findings of this study show evidence of causality from emissions to income. Our results are consistent with Ang (2008) and Azlina and Nik Mustapha (2012) for Malaysia and Yoo (2006) for Indonesia. As for energy-emission nexus, there is causality from energy to CO2 emissions. Furthermore, we also found the evidence of Granger causality from economic growth and energy use to renewable energy. Therefore, it can be concluded that Malaysia relies on energy consumption to boost its economic development. Furthermore, industrialization and services plays a vital role for the emerging growth in energy consumption that contributed to environmental problem indirectly. In fact, the increase of energy consumption that contributed to increase the emissions of CO2 gives an adverse impact to environmental. Therefore, most of ASEAN countries realized the importance of renewable energy use in order to minimize the level of CO2 emission. Hence, Malaysia is working towards the use of green technologies to reduce the environmental problems.

Table 6: Granger causality estimated under VECM approach in Malaysia and Indonesia

Dependent variable

Sources of causation (independent variables)Short-run Long-run

ΔCO2 ΔGDP Δ GDP2 Δ EC Δ RE Δ ECTMalaysia

ΔCO2 - 1.7097 1.5350 5.0030** 0.1134 0.0437ΔGDP 4.0061** - 4.4772 0.7305 0.1045 2.2237

Δ GDP2 3.9531 4.1280 - 0.7705 0.1045 535.8371Δ EC 0.3949 2.0282 2.0284 - 0.0280 0.1552Δ RE 1.6029 6.3394*** 7.2813 21.1103*** - -1.1436***

IndonesiaΔCO2 - 0.2308 0.2222 1.8370 3.4040** 0.2840ΔGDP 1.1042 - 1.1889 0.5151 3.4004** 0.0432

Δ GDP2 1.1089 1.1531 - 0.5400 3.5526 0.3740Δ EC 0.2272 1.3588 1.3217 - 1.3525 0.1397Δ RE 0.3430 0.2030 0.1886 0.5290 - -0.7066




On the other hand, in Indonesia, there is evidence of causality runs from renewable energy to carbon dioxide emissions. This can be referred from the short run causality which indicates that F-statistics for renewable energy in the carbon dioxide emissions and economic growth equation is significant at 5% level. Moreover, there is evidence of causality from renewable energy to GDP. The finding of this study suggests that renewable energy use is a crucial component to encourage the economic growth while implementation of renewable energy contributes to CO2 emissions. Following this, we can conclude that Indonesia is a developing country that is dependent on energy consumption in pursuit to expand the economic growth towards developed country. Thus, the increased of carbon dioxide emissions is the factor that contributes to environmental problem is noteworthy. Thus, renewable energy seems to be promising alternative for the current energy sources to reduce the carbon dioxide emissions in the future.

Conclusion

This study examines the link between CO2 emissions, consumption of energy resources and economic development and for Malaysia and Indonesia over the period 1980 to 2011. The unit root test shows all the variables were stationary at first difference. The results of Johansen’s test indicate that the existence of long run equilibrium relationships between variables. The causality test revealed the causality was running from energy use to CO2 and renewable energy, CO2 to GDP and GDP to renewable energy in Malaysia. Besides, the result provides an evidence of long run equilibrium relationship between variables. Conversely, we found short run causality from renewable energy to GDP and CO2 emission in Indonesia. In addition, the EKC hypothesis is not valid among these two countries.

There is an urgent need for strengthening a policy recommendation on both energy and economic sector in Malaysia and Indonesia. With respect to Malaysia, the government should develop local energy policies

to reduce environmental degradation. Any policy to combat CO2 emissions should ensure that it will not affect economic growth in the future. Since Malaysia is gifted with abundance supply of renewable resources, there is large potential for renewable energy utilization to become an alternative to current energy sources. Renewable energy should be used for major sectors not only to conserve non-renewable energy but also to combat emission. In the case of Indonesia, the government may consider to put a limitation for exporting gas and coal in order to conserve its fossil energy reserves. In other words, renewable energy should be used as alternatives to oil and must be utilized as well to stop the exports of gas and coal. In the light of these findings, developing countries seem to have an imperative need to promote and explore the use of renewable energy as an alternative to the current energy sources and for sustainable growth.

To conclude, this study on the effect of energy consumption, renewable energy consumption and economic growth on CO2 emission in Malaysia and Indonesia revealed that both countries should focus more on optimizing the energy use through renewable energy and energy efficiency that particularly contribute to achieve high income or GDP. These countries should realize the important of energy efficiency in order to combat pollution emissions and to achieve the sustainable development. Moreover, these countries should realize that renewable energy use is the one of alternative that should be promoted to major sectors. In fact, the use of renewable energy is not only to conserve energy but also to reduce emission. Meanwhile, the government incentives in investment of high technologies in major sectors also can help to improve the environmental problems in order to achieve sustainable development.

Acknowledgements

The authors acknowledge the helpful comments of anonymous reviewers and participants of the 4th International Conference on Accounting, Business and Economics (ICABEC, 2016) and



the valuable suggestions of anonymous referees. The authors would also like to thank the editors of the Journal of Sustainability Science and Management for their helpful assistance throughout the preparation of this manuscript.

References

Al-Mulali, U., Fereidouni, H. G., Lee, J. Y., & Sab, C. N. C., (2013). Examining the bi- directional long run relationship between renewable energy consumption and GDP growth. Renewable Sustainable Energy Review, 22: 209-222.

Al-Sayed, A. R. M., & Sek, S. K. (2013). Environmental Kuznets Curve: Evidences from developed and developing economies. Applied Mathematical Sciences, 7(22): 1081-1092.

Altinay, G., & Karagol, E. (2004). Structural break, unit root, and the causality between energy consumption and GDP in Turkey. Energy Economics, 26(6): 985-994.

Ang, J. B. (2008). Economic development, pollutant emissions and energy consumption in Malaysia. Journal of Policy Modelling, 30(2): 271-278.

Azam, M., & Khan, A. Q. (2016). Testing the Environmental Kuznets Curve hypothesis: A comparative empirical study for low, lower middle, upper middle and high income countries. Renewable and Sustainable Energy Reviews, 63, 556–567.

Azlina, A. A. (2011). On the causal links between energy consumption and economic growth in Malaysia. International Review of Business Research Papers, 7(6): 180-189.

Azlina, A. A., & Nik Mustapha, N. M. (2012). Energy consumption, economic growth and pollutant emissions nexus: The

case of Malaysia. Procedia – Social and Behavioral Sciences, 65: 1-7.

Azlina, A. A., Law, S. H., & Nik Mustapha, N. H. (2014). Dynamic linkages among transport energy consumption, income and CO2 emission in Malaysia. Energy Policy, 73: 598-606.

Begum, R. A., Sohag, K., Abdullah S. M. S., & Jaafar, M. (2015), CO2 emissions, energy consumption, economic and population growth in Malaysia. Renewable and Sustainable Energy Reviews, 41: 594-601.

Carbon Dioxide Information Analysis Centre (2012). CDIAC Data. Retrieved from http://cdiac.ess-dive.lbl.gov/2012, 2 March 2017.

Chandran, V. G. R., & Tang, C. F. (2013). The impacts of transport energy consumption, foreign direct investment and income on CO2 emissions in ASEAN-5 economies. Renewable Sustainable Energy Revision, 24: 445-453.

Chang, C. C. (2010). A multivariate causality test of carbon dioxide emissions, energy consumption and economic growth in China. Applied Energy, 87(11): 3533-3537.

Chen, S.T., Kuo, H.I., & Chen, C.C. (2007). The relationship between GDP and electricity consumption in 10 Asian countries. Energy Policy, 35(4): 2611-2621.

Copenhagen Accord (2009). Fifteenth session of the Conference of the Parties (COP 15). Retrieved from http://unfccc.int/meetings/copenhagen_dec_2009/meeting/6295.php, 2 March 2017.

Ghosh, S. (2010). Examining carbon emissions – Economic growth nexus for India: A Multivariate cointegration approach. Energy Policy, 38: 2613-3130.



Granger, C. W. J. (1987). Some recent development in the concept of causality. Journal of Economics, 39: 199-211.

Halicioglu, F. (2009). An econometric study of CO2 emissions, energy consumption, income and foreign trade in Turkey. Energy Policy, 37: 1156-1164.

International Energy Agency (2015). Data and statistics manual. Retrieved from https://www.iea.org/countries/non-membercountries/indonesia/, 13 January 2018.

Energy Information Administration (2017). International Energy Outlook 2017. Retrieved from https://www.eia.gov/outlooks/ieo/exec_summ.php, 13 January 2018.

Jalil, A., & Mahmud, S. F. (2009). Environment Kuznets curve for CO2 emissions: A cointegration analysis for China. Energy Policy. 37: 5167-5172.

Jamaluddin, S., Azlinda, A., & Behnaz, S. (2013). The potential of renewable energy: Using the Environment Kuznets curve model. American Journal of Environmental Science, 9(2): 103-112.

Jebli, M. B., Youssef, S. B., & Ozturk, I. (2016). Testing Environmental Kuznets curve hypothesis: The role of renewable and non-renewable energy consumption and trade in OECD countries. Ecological Indicators, 60: 824–831.

Jobert, T., & Karanfil, F. (2007). Sectoral energy consumption by source and economic growth in Turkey. Energy Policy, 35: 5447-5456.

Johansen, S., & Juselius, K. (1990). Maximum likelihood estimation and inference on cointegration-with applications to the demand for money. Oxford Bulletin of Economics and Statistics, 52(2): 169-210.

Lean, H. H., & Smyth, R. (2010). CO2 emissions, electricity consumption and output in ASEAN. Applied Energy, 87(6): 1858-1864.

Lidula, N. W. A., Mithulananthan, N., Ongsakul, W., Widjaya, C., & Henson, R. (2006). ASEAN towards clean and sustainable energy: Potentials, utilization and barriers. Renewable Energy, 32: 1441-1452.

Narayan, P. K., & Narayan, S. (2010). Carbon dioxide emissions and economic growth: Panel data evidence from developing countries. Energy Policy, 38: 661-666.

Ozturk, I., Aslan, A., & Kalyoncu, H. (2010). Energy consumption and economic growth relationship: Evidence from panel data from low and middle income countries. Energy Policy, 38(8): 4422-4428.

Payne, J. E. (2009). On the dynamics of energy consumption and output in the US. Applied Energy, 86(4): 575-577.

Sebri, M., & Ben-Salha, O. (2014). On the causal dynamics between economic growth, renewable energy consumption, CO2 emissions and trade openness: Fresh evidence from BRICS countries. Renewable and Sustainable Energy Reviews, 39: 14-23.

Soytas, U., & Sari, R. (2003). Energy consumption and GDP: Causality relationship in G-7 countries and emerging markets. Energy Economics, 25: 33-37.

Soytas, U., & Sari, R. (2009). Energy consumption, economic growth, and carbon emissions: Challenges faced by an EU candidate member. Ecological Economics, 68(6): 1667-1675.

Stern, D. I. (2004). The rise and fall of the Environmental Kuznets curve. World Development, 32: 1419-1438.



Stern, D. I. (2010). The role of energy in economic growth. Centre for Climate Economics & Policy working paper 3.10. Canberra, Australia: Crawford School of Economics and Government, The Australian National University.

Tang, C. F., & Tan, E. U. (2013). Exploring the nexus of electricity consumption, economic growth, energy prices and technology innovation in Malaysia. Applied Energy, 104: 297-305.

Tang, C. F., & Tan, B. W. (2015). The impact of energy consumption, income and foreign direct investment on carbon dioxide emissions in Vietnam. Energy, 79: 447-454.

Tiwari, A. K. (2011). A structural VAR analysis of renewable energy consumption, real GDP and CO2 emissions: Evidence from India. Economics Bulletin, 31(2): 1793-1806.

Tsani, S. (2010). Energy consumption and economic growth: A causality analysis for Greece. Energy Economics, 32: 582-590.

World Bank (2014). The World Development Indicators 2014 Database. Retrieved from https://data.worldbank.org/products/wdi, 18 January 2017.

World Bank (2015). The World Development Indicators 2015 Database. Retrieved from https://data.worldbank.org products/wdi, 18 January 2017.

Yoo, S. (2006). The causal relationship between electricity consumption and economic growth in ASEAN countries. Energy Policy, 34: 3573-3582.

Zhang, X. P., & Cheng, X. M. (2009). Energy consumption, carbon emissions, and economic growth in China. Ecological Economics, 68(10): 2706-2712.




A SURVEY ON THE WILLINGNESS TO PAY FOR DOMESTIC WATER SERVICE ATTRIBUTES IN TERENGGANU, MALAYSIA

MAHIRAH KAMALUDIN1,2, AZLINA ABD. AZIZ1*, NUR SYUHADA CHE IBRAHIM1 AND ALIAS RADAM3

1School of Social and Economic Development, Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia

2Institute of Oceanography (INOS), Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu, Malaysia

3Faculty of Economics and Management, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia


Abstract: This study evaluated consumers’ willingness to pay for an improved domestic water services in the state of Terengganu. The following issues are among the challenges faced by water supply provider; excessive water consumption, frequent interruptions, urbanization, climate change, as well as rapid social and economic growths. Choice modelling method was applied in analysing the preferences of domestic water services to improve the standard of current water services. Four attributes, namely water quality, water interruption, water pressure, and water prices were examined to improve water supply services. In order to enjoy the water service transformation, respondents were required to trade-off their income and select their preferences from the attributes based on their willingness to pay. Conditional Logit model was estimated to account for heterogeneity in the selected choices of the consumers from the options provided in the water service attributes. A survey was conducted from March until June 2015 on 1200 respondents from every district (cluster) in the state. The survey discovered that reducing the frequency of water interruption is the most important water service attribute to the population. Consumers are willing to pay about RM1.29 to lessen the state’s water disruptions. An appropriate water price facilitates the water supply provider to overcome the challenges towards providing a better service. The water services sustainability is essential in providing the population with safe and clean water. A sustainable water service is helping the economic, environmental, and social sustainability of the communities served by the water utilities. Thus, the water provider must focus on long term sustainability in bringing about meaningful transformation in their organizations and service for the public.

Keywords: Choice modelling, Conditional Logit Model, water service, willingness to pay, sustainable water service.

Introduction

In Malaysia, water supply service is typically provided by either local government agency or private organization. Urbanization, large population, industrialization, and the expansion of irrigated agriculture increase the demand and pressure on water resources. Service interruption and aging infrastructure affect the operation and capital investment, for higher costs are incurred in delivering an excellent water service. Good water management

improves human health, reduces poverty, and sustains human development. Changing urban metabolism through population growth, urbanization and climate change urge the water supply professionals and other key actors to revamp the water management particularly in urban areas. Given these issues, it is likely that the current problems of water services will increase in the future unless very efficient strategy for expanding water services is developed. Application of effective technologies and upgrading of services are needed as a key

: 135-146

136A SURVEY ON THE WILLINGNESS TO PAY FOR DOMESTIC WATER SERVICE ATTRIBUTES IN TERENGGANU, MALAYSIA

component in any strategy aimed at improving water services in developing countries. These require a vast amount of capital to be invested, which later promote a strong economic growth, as well as better health and social equity. The water industry is also facing increasing costs for it becomes very costly to bring the best water provision to its users. Sufficient investment is needed to execute numerous programs that should not only be borne by water supply providers because the water supply provider has already had to suffer low revenues/losses as a result of low price of water.

The price of water in Terengganu was last revised two decades ago in 1997, whereas population growth is known to be one of the many challenges in the state’s water supply management. At the moment, accelerating depletion of water resources occurs as a result of

increasing demand of water. The total estimated population of Terengganu is around 1.15 million people in 2015 and it is expected to rise to two million by 2035. Urbanization and economic developments in Terengganu have increased the demand of water in various sectors such as potable water, non-paddy agriculture crops, irrigation, fisheries, and animal husbandry. The highest demand comes from drinking water and it shows that the demand has increased annually in Terengganu. Table 1 tabulates the water demand for various sectors in Terengganu from 2010 until 2015. The motivation of this study is to work out an effective strategy for water supply development especially in Terengganu. Hence, an application of economic valuation in this study is able to recognize which specific attributes that are concerned by consumers in Terengganu to improve water supply services.

Table 1: Water demand for various sectors in Terengganu

SectorsWater demand million cubic meter

per yearWater demand millimetre rainfall

per year2010 2020 2030 2040 2050 2010 2020 2030 2040 2050

Drinking water demand 230.7 354.4 399.3 442 481.4 17.7 27.2 30.6 33.9 36.9

Irrigated paddy 464 428 373 351 323 35.6 32.8 28.6 26.9 24.8Non-paddy crops 184 184 184 185 185 14.1 14.1 14.1 14.2 14.2

LivestockTotal 5.1 8.1 12.9 21.4 36.5 0.4 0.6 1.0 1.6 2.8

consumptive demand 883.8 974.5 969.2 999.4 1025.9 67.8 74.7 74.3 76.6 78.7

Fisheries 35.2 42.9 52.3 127.6 155.5 2.7 3.3 4.0 9.8 11.9Total demand 919 1017.4 1021.5 1127 1181.4 70.5 78.0 78.3 86.4 90.6

Source: National Water Resources Study (2011).

Terengganu lies 400 kilometres from the east of Kuala Lumpur, the capital city of Malaysia. The state is located in the East Coast of Peninsular Malaysia, having many beautiful beaches and marine parks such as Pulau Redang, Pulau Perhentian and Pulau Kapas. Figure 1 presents the map of Terengganu with water treatment plants’ location in the districts of the state, namely Kemaman, Dungun, Marang, Kuala Terengganu, Hulu Terengganu, Besut, and Setiu.


137Mahirah Kamaludin1,2 et al.

Currently, the only water provider in the state is Syarikat Air Terengganu Sdn. Bhd. (SATU). The company’s mission is to supply sufficient high quality of clean water and to improve the quality of services in order to fulfil consumer satisfaction. Meanwhile, their corporate vision is towards becoming a leading water supply provider in the country. Two main sources of water supply in Terengganu are from the river extraction and dams. In the year 2012, about 440 million litres of water per day are direct extraction from river and 187 million litres of water per day from storage dams (Malaysian

Figure 1: Water treatment plants in Terengganu (Source: Syarikat Air Terengganu Sdn. Bhd., 2014)

Water Association, 2013). SATU serves 85.2% of water supply for domestic users and 14.8% of water supply for non-domestic users. The statistics from Malaysian Water Association (2015) showed that 99.1% in urban areas and 92.9% in rural areas get accessed to water supply in 2014.

Water price is inexpensive, therefore the water provider is usually unable to gain sufficient revenues to cover the full costs of capital investment, maintenance and operation. In 2015, the state was ranked second lowest



in terms of domestic water prices in Malaysia (MYR0.52/USD0.13 applies for the first 30m3). Such low price of water limits the implementation of infrastructure projects to improve the facilities; hence, it demonstrates low value of water supply business although the sources are insufficient and valuable. There are many difficulties in pricing the water as it is a public good and does not have a well-defined market. Water is still very costly and involves high amount of investment. Below cost tariff makes consumers take for granted upon how precious the water sources are. In Malaysia, the average water consumption is very high at 211 litres per capita per day. The World Health Organization (WHO) sets it should be at 160 litres per capita per day. In Terengganu, the consumption is increasing every year and in 2015, it recorded a usage of 216 litres per capita per day compared to 204 litres per capita per day in the year 2013 (Malaysian Water Association, 2015). Similarly, operation and maintenance costs are increasing as these costs are mostly spent on energy cost such as electricity to operate the water treatment plants.

The present study estimated the monetary trade-off between the level of attribute and price, in which it determined the appropriate level of service attributes relative to price. Thus, information was required based on the value that the respondents put on each attribute. Respondents were presented with a set of preferences, and then they chose the best preference for improvement. The marginal willingness to pay was calculated by the differences in the coefficients between the two attribute levels. Households were asked about how much they would pay for a given service level, describing at which level they were willing to contribute to experience a transformation of water provision.

Economic valuation presents the assignment of monetary values to non-marketed assets, where the money value has a particular and precise meaning. Non-marketed goods and services might not be directly sold and bought in the market place because they do not have

‘price tag’. Pearce and Ozdemiroglu (2002) stated that goods and services that are positively contributing to the well-being of humans have economic values. The contributions made are estimated by respondents’ preferences through their willingness to pay for the goods and services. They are willing to pay high prices if they attain high social values from the services. Previous studies by Hensher et al. (2005), MacDonald et al. (2005) and Kanyoka et al. (2008) focused on an improved water services in developing and developed countries. Other authors such as Mahirah et al. (2013) and Mohd Rusli et al. (2011) used willingness to pay (WTP) approach and conducted Choice Modelling (CM) survey to determine respondents’ willingness to pay towards reforming the water services in Malaysia. Nevertheless, the CM method is still being criticized although it offers important information about the value of specific attributes of a proposed program/policy. Blamey (1998) stated that the CM method contains bias in terms of quantity and quality of information offered to the respondents.

Methodology

In order to measure the value that households in Terengganu place on water service, this study examined the amount of money that the households are willing to pay according to three key results of service improvement: (i) increasing water quality; (ii) reducing water supply disruptions; and (iii) increasing water pressure. Better water quality, water supply, and water pressure are not the items that are bought and sold in the markets. Thus, specific non-market valuation techniques were used to estimate their value. The orthogonal design allowed generating 15 water service options. Hence, this study applied CM approach. Respondents were given with a number of choice card sets which described the three proposed improvements in water service policies. With reference to Table 2, this study exhibits a range of impacts that these policies might have on the water quality, water disruptions and water pressure. The choice cards also listed the amount of money that the policy options would cost a household. In this study,



the WTP could indirectly be estimated from the choices, inclusive of water price as one of the attributes. This method aimed to increase

the attribute level, for instance, water quality at QUAL1 (status quo) to improve the QUAL2 or QUAL3 level.

Table 2: Water service attributes and their levels

Attributes Levels Descriptions

Water Quality (QUAL)

QUAL1 Satisfactory Connection of water supply at homes is of high quality. It is safe for direct human consumption, odourless, colourless, and tasteless. It also complies with the standard of drinking water

quality.

QUAL2 Good

QUAL3 Very Good

Water Disruption (DIST)DIST1 Always

Reducing the frequency of water supply disruptionsat homes.DIST2 Sometimes

DIST3 NeverWater Pressure (PRES)

PRES1 LowCondition of water pressure when the water gushes out from

the tap water.PRES2 ModeratePRES3 High

Water Price (PRICE)PRICE1 Current Price

Water price describeshouseholds’ monthly water bills charged by water provider,

Syarikat Air Terengganu Sdn. Bhd.

PRICE2 Low (Increased 28% from current price)

PRICE3 Moderate (Increased 56% from the current

price)PRICE4 High (Increased 90%

from the current price)


Table 3: Attributes and expectation sign

Attributes Descriptions Expected Sign

Water QualityConnection of water supply at homes is high in quality, safe for direct

human consumption, colourless, tasteless, and odourless. It also complies with the standard of drinking water quality.

Positive

Water Disruption Reducing the frequency of water supply disruptions at homes. Positive

Water Pressure Condition of water pressure when the water gushes out from the tap water. Positive

Water PriceWater price refers to household monthly water bills charged by water

provider; Syarikat Air Terengganu Sdn. Bhd, presented as amount increase over the current bill.

Negative



The respondents were then asked to tick the box showing the options they would prefer. Figure

McFadden (1973) highlighted that the theoretical foundation of CM lies within the random utility theory (RUT). RUT proposes that individuals select their preferred characteristics of goods based on random components that are unique. The theory maintains that the utility Uij

The Conditional Logit (CL) model assumes that the error disturbance has a Type 1 extreme value

Uij = Vij + εij (1)

exp[-exp(-εij)] (2)

2 demonstrates a choice card which portrays options to choose by respondents.

of an individual i serves as a dependent random variable, attributes of the investigated program and their attribute levels is Vij and unobservable random component is εij. The model shows that the utility in which an individual i correlates with alternative j is as follows (1);

distribution as in (2);

Attribute Water ServiceOption 1

Water ServiceOption 2 Current Water Service

Frequency of water supply disruption

Good Good Satisfactory

Increasing water pressure

Sometimes Always Always

Increasing water pressure

Water Price

High Low Low

Choose your best preference

Increases 90% Increases 28% Current PriceChoose your best

preference √

Figure 2: Sample of show card demonstrating a choice set



By employing CL, the probability of selecting an alternative j among n choices for individual

Equation 3 demonstrates the probability that the individual i selects j is equal to the probability that the utility derived from j is larger than the utility derived from any other alternatives (Whittington et al., 1990). The next step is to

Consumer Surplus (CS) was used to estimate the WTP or the marginal values of attributes for a

Where bP is the marginal utility of price (monetary value), VC demonstrate the utility of the current situation, and VN is the utility of new option.

This survey was conducted from March until June 2015 at eight districts in the state of Terengganu, Malaysia. The households were randomly selected. In order to collect quantitative and qualitative data about the households, a structured questionnaire was distributed to conduct the CM method. Samples of 1,532 households were contacted and 1,200 households agreed to take part in the survey. In addition, information was gathered regarding on the households’ perceptions and their opinions about the current water service. They were informed that the water service improvement strategy would depend on the most popular option chosen by the citizens and the results would determine the amount of money that each household must pay for the water service. Besides, each attributes and levels demonstrate different marginal willingness to pay for the improvements of water services in accordance with their preferred selection levels to move from status quo (Level 1) to either Level 2 or Level 3. The in-person interview for this section took about 15 to 20 to allow respondents make their decisions.

Probi(j) = Prob [x’ij β +εij ≥ maxk ∞ci (x’ik β + εik)] = exp(x’ij β) / ∑ ∞ciexp(x’ik β) (3)

CS = - 1/bP(VC – VN) (5)

i is;

estimate the choice probability and welfare measure. The ratio of an attribute’s coefficient and the price coefficient signifies the marginal implicit price of the attributes as in the Equation 4.

change from current to alternative situation. It could be estimated as follows;


Socio Demographic Characteristics of the Respondents

Table 4 presents the results of the respondents’ demographic profiles. The respondents were heads of the households (female, 50.1% and male, 49.9%). The percentages for both genders were about the same. Most of them (34.5%) were between 41 and 50 years old, where the age group mean was 42 years old. The present study found that the participants mostly obtained university/college education (55%) as their highest education level, followed by secondary school (31%), primary school (12%) and no formal education (2%). The results reported that their average monthly income was RM4,182.50. Meanwhile, the household size ranged from one to five members, with a mean of six for the whole sample. A majority of 424 respondents (35.3%) were government servants, private sectors (20.1%), managing businesses (12.5%), and others (32.1%). 94% of them with jobs had one to three family members.

Probi(j) = (4)



Table 4: Descriptive statistics for respondents, n=1200

Demographic Characteristics Frequency Percentage (%) Mean Standard Deviation

GenderMale 599 49.9

Female 601 50.1Age Group (Years)

20 – 30 207 17.4

42.13 11.42

31 – 40 336 2841 – 50 413 34.551 – 60 185 15.461 – 70 42 3.7

> 70 17 1.6Education Level

University/College 662 55Secondary School 367 31Primary School 141 12

No Formal Education 30 2Household Income

Less than RM2000 312 26

RM4182.50 RM4242.80

RM2001 - RM4000 446 37RM4001 - RM6000 263 22RM6001 - RM8000 102 9

RM8001 - RM10,000 58 5More than RM10,000 19 2

Size of Household1 - 5 701 58.4

6.444 3.0696 - 10 471 39.3> 10 28 2.4

Occupation of the Head of Household Government sector 424 35.3

Private sector 241 20.1Businessman 150 12.5

Others 385 32.1Working Family Members

3 members and below 1129 94.04 – 6 members 65 5.5 1.958 1.047

More than 7 members 6 0.5



Determinants of Households’ Willingness to Pay for Water Services

The estimation procedures for CM were employed using econometric software, NLogit Version 5. A CM model was applied to interpret the data collected through the CM. The dependent variable for this model was the choice

Where β1 until β7 demonstrate the main attribute coefficients, and X1 until X7 demonstrate the main attributes. The Conditional Logit (CL) model includes seven variables, namely QUAL2, QUAL3, DIST2, DIST3, PRES2, PRES3 and PRICE. Table 5 illustrates the estimated willingness to pay from the current condition to an optimal outcome, signifying the estimated benefits in monetary terms regarding the state’s water service improvements.

The overall coefficients and signs were as expected and showed positive signs except for High (PRES3) in the water pressure attribute (Table 5). All coefficients for attribute levels were significant at 1% level except for Moderate (PRES2) water pressure which was significant at 5% level. Next, the highest level of the attribute (PRES3) portrayed a negative relationship towards the dependent variable, and the preferences over water quality attribute levels showed a similar pattern. Also, the conditions of water quality such as colourless, tasteless, and odourless were mostly favoured by the respondents during the interview. In terms of water disruptions, preferences in Sometimes and Never indicated that the respondents preferred lessening the frequency of water supply interruptions. For water pressure, both Moderate and High water pressures had effects on the willingness to pay for water services but in different signs, indicating that the respondents were more prone towards Moderate water pressure compared to High water pressure. This is probably due to High water pressure could cause water leakage in plumbing and piping system. Besides, the respondents did not expect that the water services could be improved to the highest level, because at that time, the current water pressure was worsening.

of a water alternative, whereas the independent variables were presented by the water services attributes. Then, the alternatives to be selected by respondents were in different combination of attribute levels. The model is stated in the Equation 6;

Table 5 demonstrates the estimated coefficients for all variables which were used to measure the outcomes of change in attributes regarding the price that households were willing to pay for service improvements. The result of marginal willingness to pay at the level “good” to “very good” for water quality attribute was at RM1.20. A service improvement in terms of water disruptions from the level of “sometimes” to “never’” was at RM0.11. These values illustrated how much respondents were willing to pay for water service improvements at each level. Meanwhile, the differences in marginal values in water pressure at the level “low” to “moderate” was at RM0.09 and the level “moderate” to “high” was at RM0.92 (negative value). ‘Water pressure’ attribute exhibited a relatively low values (negative value) compared to other attributes. Surprisingly, households were mostly concerned about water disruptions at their homes as the attribute generated the highest willingness to pay, which was at RM1.29 compared to other attributes. The marginal rates of substitution of water disruption attribute in Level 2 (Sometimes) was the highest which, demonstrating that the consumers were willing to pay for water disruption improvements. However, the main problem in the service was water pressure as the demand was very high in the state. At the moment, there are 12 water plants that cater the needs of 1.15 million population in Terengganu. Similarly, Mahirah et al. (2016) posited that respondents chose to improve water interruptions to reduce the frequency of disruptions in Kelantan from moderate level to high level (less water service disruptions); it generated the highest marginal values based on the CM model.

Choice of Water Services = β1X1+ β2X2+ β3X3+ β4X4+ β5X5+ β6X6+ β7X7+ ε (6)



Table 5: Estimated coefficients for households’ willingness to pay in Terengganu, Malaysia

Variables Coefficient t-valueWater Quality (QUAL)

Good (QUAL2) 3.7682 7.190*** Very Good (QUAL3) 3.8076 7.573***

Water Disruption (DIST) Sometimes (DIST2) 4.0899 9.251***

Never (DIST3) 0.3394 3.984***Water Pressure (PRES)

Moderate (PRES2) 0.2881 2.100** High (PRES3) -2.9269 -11.384***

Water Price (PRICE) -3.1706 -4.658***Marginal Values of the Attributes: (β’s coefficients/-β’s price)

Good (QUAL2) 1.18 12.636*** Very Good (QUAL3) 1.20 9.827*** Sometimes (DIST2) 1.29 9.133***

Never (DIST3) 0.11 6.556*** Moderate (PRES2) 0.09 1.932*

High (PRES3) -0.92 -7.263***

Note: Significance Level: (***) 1% Level, (**) 5% Level, (*) 10% Level.

A study on economic valuation that took place in Kazakhstan signified that water price improvements could be used to identify the proper selection of technologies and service delivery to make rural water projects both sustainable and applicable at a larger scale. Revising water price supported the improvements of their water supply services. However, a study on the willingness to pay in Karachi, Pakistan demonstrated that cost recovery was indeed possible via the increment of tariffs for higher income households. The study recommended imposing cross subsidization among users which will make some users pay more than cost and others poor households pay less. Overall, previous willingness to pay studies revealed the different stories regarding socio demographic of consumers. For instance, cross-subsidy tariffs are imprecise instruments as poor households are least likely to have connections at homes. In those cases, it demonstrated that water prices are

important to promote financial and environmental sustainability. The price must cover at least the costs of operation and maintenance in the water service provision. Increasing water price would most likely discourage an excessive water use and a replication of the marginal cost of expanding water supply capacity. An effective water rate structure for cost recovery and its sustainability are important because the water sources are vital in our lives. Human beings require a sustainable water supply to help them achieve a better quality of life in the future.

Conclusion

This study applied CM method in Terengganu and used the utility of stated preference methods to determine households’ demand characteristics and their willingness to pay for various aspects of water services. The households had provided valuable information to decision makers



according to their willingness to pay in different scenarios of water service improvements. This study discovered that the water service disruption was mostly concerned by the respondents as the attribute had the highest marginal willingness to pay compared to other attributes. Water provider in the state should take into account an efficient and equitable allocation of water to urban and rural consumers. Furthermore, the survey found that the households were not satisfied with the water services at that time, particularly in terms of water pressure.

This valuation technique offered in-depth information about consumers’ preferences and willingness to pay for different attributes of water services. This information is significant for water providers and authorities outline the provision of this service and the corresponding price based on the households’ demands. The level and structure of new water price should be carefully designed in order to increase the success rate of this transformation program. This is important to promote and educate communities regarding the importance of water in our lives and encourage water saving practices. Cheaper water price makes households take for granted the valuable natural resources. An appropriate water pricing results in a much better water consumption by consumers. Additionally, reliable and efficient water services are also a vital ingredient for economic growth. Effective management of water services can prominently reduce water losses, improve water billing and increases revenue collection; thus, it will have a substantial effect on water demand.

Acknowledgments

We thank the anonymous referees for their useful suggestions. This study was supported by the Ministry of Higher Education (MOHE) Malaysia, FRGS2014/1 research grant Vote no. 59348.

References

Blemey, R. K. (1998). Contingent valuation and the activation of environmental norms. Ecological Economics, 24(1): 47-72.

Hensher, D. A., Shore, N., & Train, K. (2005). Households’ willingness to pay for water service attributes. Environmental & Resource Economics, 32(4): 509-531.

Mahirah, K., Alias, R., & Khalid, A. R. (2016). Household preferences for improved water services in Kelantan, Malaysia: A choice experiment approach. Journal of Business and Social Development, 4(1): 43-54.

Mahirah, K., Khalid, A. R., Alias, R., & Mohd, R. Y. (2013). Improvements in domestic water services in Kelantan: Are people willing to pay?. Journal of Sustainability Science and Management, 8(2): 61-70.

Kanyoka, P., Farolfi, S., & Morardet, S. (2008). Households’ preferences and willingness to pay for multiple use water services in rural areas of South Africa: An analysis based on choice modelling. Water SA, 34(6): 715-723.

MacDonald, D. H., Barnes, M., Bennett, J., Morrison, M., & Young, M. D. (2005). Using a Choice

Modelling approach for customer service standards in urban water. Journal of the American Water Resources Association, 41(3): 719–728.

Malaysian Water Association (2015). Malaysia water industry guide. Kuala Lumpur, Malaysia: Malaysian Water Association.

Malaysian Water Association (2013). Malaysia water industry guide. Kuala Lumpur, Malaysia: Malaysian Water Association.

McFadden, D. (1973). Conditional logit analysis of qualitative choice behavior. In P. Zarembka, (Ed.), Frontiers in Econometrics (pp. 105-142). New York : Academic Press.



Pearce, D., & Ozdemiroglu, E. (2002). Economic Valuation with stated preference techniques:

Summary guide. Department for Transport, Local Government and the Regions, London. 87 pp.

Syarikat Air Terengganu Sdn. Bhd. (2014). Maklumat Takat Lokasi Intake. Kuala Terengganu, Malaysia: Syarikat Air Terengganu Sdn. Bhd.

Whittington, D., Davis, J. & McCelland, E. (1990). Implementing a demand-

driven approach to community water supply planning: A case study of Lugazi, Uganda. Water International, 23(3): 134-145.

Yacob M. R., Radam, A., & Samdin, Z. (2011). Willingness to pay for domestic water service improvements in Selangor, Malaysia: A Choice modeling approach. International Business and Management, 2(2): 30-39.




A REVIEW OF CORPORATE GOVERNANCE AND CORPORATE SOCIAL RESPONSIBILITY DISCLOSURE OF ISLAMIC BANKS: A CALL FOR ADDITIONAL

TRANSPARENCY

SAED AHMED SULUB, ZALAILAH SALLEH* AND HAFIZA AISHAH HASHIM

School of Maritime Business and Management,Universiti Malaysia Terengganu, 21030 Kuala Nerus, Terengganu

* Corresponding author: [email protected]

Abstract: The purpose of this paper is to investigate the validity of the hypothetical expectation of higher corporate governance disclosure and corporate social responsibility disclosure in Islamic banks. Islamic banks are expected to have higher corporate governance disclosure and corporate social responsibility disclosure levels due to the unique characteristic of maintaining requirements for compliance with the Shariah in safeguarding stakeholder interests and economic sustainability. In order to achieve this purpose, the theoretical expectations of Islamic banks’ practices in relation to corporate governance disclosure and corporate social responsibility disclosure are reviewed and compared with the empirical findings from previous literature. Previous studies published primarily in the major general interest accounting journals are reviewed and the findings are summarised and compared. While comparatively higher transparency levels of corporate governance disclosure and corporate social responsibility disclosure are ideally predicted for Islamic banks, the disclosure levels were surprisingly lower in the annual reports of Islamic banks. In addition, some of the evidence obtained from corporate governance disclosure and corporate social responsibility disclosure of conventional banks has shown that these disclosures were made more extensively than Islamic banks, even though the degree of comparability between conventional and Islamic banks is relative. Furthermore, the evidence provides impressions that Islamic banks were less transparent than conventional banks in providing governance and social information to stakeholders. Accordingly, the implications of these lower disclosures are further discussed and, therefore, some suggestions are recommended to be possible remedies in order to reform corporate governance disclosure and corporate social responsibility disclosure in Islamic banks.

Keywords: Islamic banks, disclosure, sustainable corporate governance, corporate social responsibility.

Introduction

Disclosures of information over statutory and voluntary requirements have become increasingly important among company stakeholders. A transparency of governance and social information are of high concern among investors and corporate regulators because of their contribution in the fulfilment of effective corporate governance (CG) and companies’ sustainability objective. In particular, corporate governance disclosure (CGD) and corporate social responsibility disclosure (CSRD) play a significant role in improving levels of corporate transparency, corporate image and

firm sustainability (O’Dwyer & Gray, 1998; Alnajjar, 2000; Deegan & Blomquist, 2006; Eweje, 2014). These disclosures reduce the extent of information asymmetry discussed in the context of the agency relationships, which is an outcome of the separation between ownership and management. Accordingly, CGD and CSRD provide useful information for investment decision making (Friedman & Miles, 2001; Deegan & Blomquist, 2006; Aribi & Gao, 2010), which can satisfy the owners’ need. In Islamic banks, the risk caused by this information gap is exacerbated by the existence of unrestricted investment account holders.

: 147-168


148A REVIEW OF CORPORATE GOVERNANCE AND CORPORATE SOCIAL RESPONSIBILITY DISCLOSURE OF ISLAMIC BANKS: A CALL FOR ADDITIONAL TRANSPARENCY

Although the investment account holders partly share risks and rewards with equity owners, they are not often represented in the board. This circumstance restricts their access to Islamic banks’ information. Furthermore, previous research has observed that the rights of the unrestricted investment account holders are not properly addressed by the banks and the investment account holders have lack of power to control the management decision in relation to investment activities (Magalhães & Al-Saad, 2013). Therefore, in order to overcome this problem, financial statements of high quality with extensive disclosure are expected by the stakeholders.

Apart from this consideration, corporate transparency is of utmost importance for Islamic banks because of their nature to comply with the requirements of the Shari’a. In addition, ‘in the context of Islamic business world, businesses have to demonstrate their accountability to God and fulfilment of social objectives’ (Aribi & Gao, 2010, p.76). Therefore, Islamic banks are expected to publish information that reflects the demonstration of the accountabilities in their annual reports. Furthermore, previous CG failures in Islamic banks represent another factor for Islamic banks to strive for more transparency in order to regain stakeholders’ confidence (Grais & Pellegrini, 2006). As a result, CGD and CSRD are placed in a crucial role for corporate transparency of Islamic banks. Despite this, in regards to the crucial role of CGD and CSRD in Islamic banks, little is known of CGD and CSRD practices of Islamic banks.

The main objective of this study is to investigate the validity of the hypothetical expectation of higher CGD and CSRD in Islamic banks. In order to achieve this purpose, the theoretical expectations of Islamic banks’ practices, in relation to CGD and CSRD are reviewed and compared with the empirical findings of previous literature. This study reviews previous studies published primarily in the major general interest accounting journals. CGD, CSRD, conventional banks and Islamic banks are used as dominant keywords in

searching for relevant articles because the study focuses on corporate disclosures. The study considers all published papers and working papers available up to 2016. Subsequently, the finding of previous studies concerning disclosure levels and less disclosed areas are deliberated. In addition, the study provides comparisons of previous findings for corporate disclosures of Islamic banks and conventional banks. In this regard, Aribi and Gao (2010) provided evidence that Islamic banks have greater CSRD than conventional banks. However, several studies that were made in the context of governance and social reporting disclosure in Islamic banks showed surprisingly that this is not the case. The disclosure of these aspects in annual reports was considerably low (e.g., Maali et al., 2006; Hassan & Harahap, 2010; Farook et al., 2011; Sulaiman et al. 2015; Darmadi, 2013; Abdulrahman & Bukair, 2013; Wan Abdullah et al., 2013; Kamla & Rammal, 2013; Yussof et al., 2013; Wan Abdullah et al., 2015). Another observed problem is the discrepancy between the risks faced by Islamic banks and their actual practices regarding CGD and CSRD. These risks emanate from the complexity of Islamic banking financial innovations and the severity of reputation risks for their activities (El-Hawary et al., 2004).

While this study does not primarily investigate all the literature, to a certain extent, the results of the review highlights the low disclosures and, therefore, shows the need for a reform for more transparency in these respects. In addition, this paper adds to the literature by providing a collective review of the CGD and CSRD studies.

Islamic banks have become an important sector that contributes to economic prosperity and development in many countries. The recently issued competitiveness report of world Islamic banks, showed that the sector achieved a growth of 16% in terms of assets in the period between 2010 and 2014 (Ernst & Young, 2016). Interestingly, the growth was significantly higher for Islamic banks than for conventional banks, and is expected to reach


149Saed Ahmed Sulub et al.

an estimated value of USD 1.8 trillion in 2020 in the major important markets, which are the Gulf Cooperation Council (GCC) countries, Malaysia, Indonesia, Pakistan, and Turkey (Ernst & Young, 2016).

Furthermore, the higher growth is partially a consequence of the strong loyalty of Islamic banks’ customers to their services, because Muslims perceive these institutions from the perspective of their faith and therefore, place a higher degree of confidence in them (Maali et al. 2006). As a result, this consideration leads the study to presume that Islamic banks should, in exchange, be well governed and provide the social responsibility that they undertake by the rule of Shariah and expectations of the population. Alternatively stated, they should be accountable to Allah first, and then to the community (ummah) (Yusoff et al., 2013). The wider accountability concept is an indication of the concepts of CG and social responsibilities, which are different from conventional institutions, and therefore, have Shariah implications on Islamic banks.

In general, if any organisation has good governance and takes part in the provision of social services to the community in which it operates, it is expected to reflect its achievements in the annual report, which is provided to their stakeholders. Therefore, the transparency of information on the governance and social support is likely to be associated with the extent of its quality in the entity. Likewise, if an Islamic bank is well-governed and actually does participate in social activities, it would be expected to disclose that to the respective stakeholders.

Corporate Governance, Social Responsibility and Transparency

CG is simply defined to be, ‘the system by which companies are directed and controlled’ (Cadbury, 2000 p.8). Therefore, CG is a tool that enables investors, who are the providers of the organisation’s capital, to assure that their investments are well-managed and make a return

on their capital (Shleifer & Vishney, 1997). The appropriate form, structure, and relationships of corporations’ boards, shareholders, managers, and other stakeholders form the point of concentration for CG. For this reason, good CG practices is associated with higher information transparency, lower capital costs, higher operational efficiency and better allocation of resources (Maher & Anderson, 2000; Barako et al., 2006; Aksu & Kosedag, 2006).

On the other hand, CSR activities are concerned with the provision of social services to the community in which their company operates. Dahlsrud (2008) claims that the social service definition is not comprehensive enough and accordingly, analysed thirty-seven different definitions of CSR. The study has provided five-dimensional descriptions for CSR, which are environmental, social, economic, stakeholder, and voluntariness. Therefore, it is expected that when a company provides social activities to the community, the disclosures of such activities are normally made in annual reports, which show the extent of social responsibility performance in any organisation (Cochran & Wood, 1984).

Even though good CG and CSR practices are highly encouraged in both Islamic and non-Islamic models, Islam places greater emphasis on them, which would make Muslims liable by religious rules rather than human-set regulations. More importantly, Islam provides further requirements through the five-dimensional concept of Maqasid al-Shariah, which protects both self and public interests and does not allow any deviation from ethics (Hassan, 2009; Choudhury & Alam, 2013). In this section, the theoretical aspects of CG, CSR and transparency are briefly presented.

Firstly, governance in Islam emanates from the very concepts of Tawhid (oneness of Allah), the Shurah (consultation) principle, and Amana (integrity), which are articulated in the Quran and Sunnah (Hassan, 2009). Therefore, compliance with Shariah is a major element that makes corporate governance in Islamic banks very unique. As a result, the Shariah



Supervisory Board (SSB) is required to govern Shariah compliance in Islamic banks, in addition to the role of board of directors. Another component of ownership that exists in Islamic banks also adds to the uniqueness of CG in this sector, which is the ownership of unrestricted investment account holders. The critical issue for these types of owners is that their capital is mixed with the shareholder’s capital, because it is unlikely that there funds are invested in isolation. Additionally, they are not represented in the board, a matter that creates a greater need for good governance. Collectively, it can be said that the nature of being an Islamic bank is certainly enough to make its CG different.

Secondly, a focus on developmental and social goals is considered one of the fundamental features of Islamic institutions, based on the rule of Shariah. This can be observed for example, from the obligatory requirement of Zakah in Islam, the encouragement of voluntary charities (Sadaqa) and the provision of interest-free loans (Haniffa & Hudaib, 2007). Furthermore, self-interest and public interest must be in balance, and whenever contradictions exist, public interest has to be given a priority. This highlights a higher concern given to society in Islam.

Therefore, as a result of these unique attributes, higher corporate transparency becomes important for Islamic banks in order to enhance their corporate ethical nature and accountability (Haniffa & Hudaib, 2007; Smolo & Mirakhor, 2010). In general, transparency is an essential Shariah aspect in Islam, which is required by the Quran and Hadith as well as the industry-specific regulators, such as the statement of Islamic Financial Services Board (IFSB) stating that ‘any form of concealment, fraud or attempt at misrepresentation violates the principles of justice and fairness in Shariah (IFSB, 2007). Therefore, disclosure on governance and social responsibilities should be of utmost priority, because as mentioned above, Islam gives more consideration to these aspects (Alsaati, 2003).

Where true and fair disclosures are not made, the fulfilment of a Shariah-compliant

contract is violated. The prophet said: ‘Both parties in a business transaction have a right to annul it so long as they have not separated; and if they tell the truth and make everything clear to each other they will be blessed in their transaction, but if they conceal anything and lie, the blessing on their transaction will be eliminated’. In addition, Gharar, which means uncertainty and misrepresentation, is one of the constraints of contracts and dealings in Islam that may lead to a transaction being invalidated and cancelled.

The managers and owners of a business are often not the same, and an entity is managed by a delegated manager ‘agent’ who is supposed to be accountable for an owner’s investment. Therefore, it would be difficult to inform the owner ‘principal’ about every transaction and event that takes place. Consequently, owners are communicated through a periodic disclosure of summarised information.

Generally, the most used disclosure instrument in literature is an annual report, because it is a formal communication of information to stakeholders and can be easily accessed (e.g., Collet & Hrasky, 2005; Barako, 2007; Farook et al., 2011). However, the use of other instruments was also observed, such as internet reporting (e.g., Kelton & Yang, 2008), although these were not utilised in Islamic banks thus far.

In measuring the extent of disclosure, these instruments are either compared with a constructive index, which is built based on accounting standards, prior studies and observation, or with an index developed by financial companies, such as the Standard and Poor’s. For scoring purposes, an unweighted method, in which ‘1’ is given if the item is disclosed and ‘0’ if not, is the most prevalent method of scoring in the literature (e.g., Craig & Diga, 1998; Barako, 2007; Mainghot & Zeighal, 2008; Samaha & Dahawy, 2010; Farook et al., 2011; Wan Abdullah et al., 2015). Furthermore, a weighted method is used in some studies, in which the scores are weighted accordingly to the extent of its disclosure (Botosan 1997; Barako



et al., 2006). Nevertheless, the results of the two methods were found to be consistent (Barako et al., 2006).

In this section, the disclosure studies relating to CG and CSR are briefly explored. It is important to note that CGD and CSRD have Shariah components in Islamic banks. Therefore, there are Shariah requirements in both CGD and CSRD in Islamic banks.

Corporate Governance Disclosure Studies

Despite the rapid growth of Islamic banks in the past decades, the studies of CGD are still limited and very few. In addition, even though good corporate and Shariah governance practices are expected as a result of the uniqueness referred above, levels of CGD disclosure were significantly lower (Sulaiman et al. 2015; Wan Abdullah et al., 2015).

Hassan and Christopher (2005) investigated the 2003 annual reports of three banks in Malaysia, one full-fledged Islamic bank and two banks providing both conventional and Islamic banking in order to find out the relationship between the extent of CGD and the fact of being an Islamic bank. Their findings revealed that CGDs in the Islamic bank were not comparatively different from those of conventional banks. Additionally, they found that the existence of Muslim directors in the board has no impact on CGD practices.

Using a larger sample than this study, Sulaiman et al. (2015) studied the CGD practices of sixteen Islamic Financial Institutions (IFIs) operating in Malaysia. They found that their CGD level was only 51% on average in regards to the items contained in the index they used. Wan Abdullah et al. (2015) used cross-country data from 67 Islamic banks in Southeast-Asian

and GCC regions in 2009. She found that the disclosure of CG was less than 40% of the items contained in the checklist. Similarly, Wan Abdullah et al. (2013) examined Shariah governance disclosures in the annual reports of 23 Islamic banks in Malaysia and Indonesia for the year 2009, and they found that Shariah disclosure levels were less than half.

The most favourable empirical finding is the study of Darmadi (2013) on seven Indonesian banks in 2010, in which the sample banks disclosed about 60% of the study’s disclosure checklist. However, it was observed that the checklist used was less detailed than the ones used by Sulaiman et al. (2015) and Wan Abdullah et al. (2015). More recently, Srairi (2015) investigated the CGD practices of 27 Islamic banks located in five GCC countries in the period from 2011 to 2013. From the CGD checklist constructed for the study, the average disclosure level was just above half, at 54% with Shariah governance information being the most repeatedly disclosed. Again, the index used for the study was comparatively less detailed.

Table 1 summarises the disclosure levels of these studies and the less disclosed areas. It was observed that the disclosures related to investment account holders were identified to be a less disclosed category of CG by both Sulaiman et al. (2015) and Srairi (2015). Given the prescribed risks embodied in this ownership type, it would be very important to carefully consider this area. Moreover, the disclosures of internal controls and the methods used to avoid any conflicts of interest, which are necessary to protect owners’ investments and form significant instruments for effective governance, were the least disclosed categories identified by Darmadi (2013) and Wan Abdullah et al. (2015).



Table 1: Summary of corporate governance disclosure studies

Study Avg. DisclosureLevel Less Disclosed Area

Sulaiman et al. (2015) 51% Investment account holdersDarmadi (2013) 60% Internal control and external audit

Wan Abdullah et al. (2015) 43% Avoidance of conflicts of interestSrairi (2015) 54% Investment account holders

Corporate Social Responsibility Disclosure Studies

The study by Maali et al. (2006) was a pioneer in the literature of CSRD in Islamic banks. The study’s sample covered the 2000 annual reports in twenty-nine Islamic banks located in sixteen countries. The study concluded that no concern was given to social reporting, where the average disclosure was only around 13% with a maximum score of about 35%.

Hassan and Harahap (2010) surveyed the CSR reporting levels of seven Islamic banks, using reports from 2006 and found an average score of less than 50%. In a more comprehensive survey, Farook et al. (2011) investigated the CSRD of 47 Islamic banks in 19 different countries, and found a disclosure score of about 17% on average. Similarly, the results of a few other studies that included CSRD in Islamic banks were not significantly different from these studies, which were Ousama and

Fatima (2010), Zubairu et al. (2011), Kamla and Rammal (2013), Yusoff et al. (2013), Mallin et al. (2014), and Platonova et al. (2016).

Table 2 presents a summary of some of the studies discussed above. As shown in the table, information about insolvent clients, which is crucial to financial institutions, was the least disclosed area of CSRD in Maali et al. (2006). Additionally, environmental disclosures were the areas which contribute the smallest amount of information that was made available in the annual reports in Hassan and Harahap (2010), Yusoff et al. (2013), and Mallin et al. (2014). This may be because of the relative irrelevance of this category to financial services’ entities. However, information about the expenditures made on the social awareness of a community about environmental aspects and environment-harming industries funded by a bank would be of a major significance.

Table 2: Summary of corporate social responsibility disclosure studies

Study Avg. DisclosureLevel Less Disclosed Area

Maali et al. (2006) 13.0% Information on insolvent clientsHassan and Harahap (2010) 37.6% Environment

Farook et al. (2011) 17.0% Not detailedYusoff et al. (2013) 37.5% EnvironmentMallin et al. (2014) 44.4% Environment

Platonova et al. (2016) 49.9% Not detailed



Some Evidences on Corporate Governance Disclosure and Corporate Social Responsibility Disclosure in Conventional Banks

Although the disclosure levels in conventional banks were not absolutely comparable to those of Islamic banks, because of the material differences stated above, it would however, provide some indication of their status in non-Islamic banks. Therefore, some empirical findings were explored in order to provide some evidence about the CGD and CSRD in conventional banks.

Mainghot and Zeghal (2008) investigated CGD scores in the annual reports of 8 Canadian banks and found a disclosure level of about 89% on average. In a different setting, Sharma (2014) surveyed the CGD scores of 59 financial institutions listed on the Nepal stock exchange and found that the sample banks disclosed about 71% on average of the disclosure index, which contains both mandatory and voluntary requirements.

For the CSRD, Doughlas et al. (2004) surveyed six Irish financial institutions and found that that they disclosed most of the checklist requirements, which was a combination of four European codes of ‘good practices’. Moreover, the evidence of CSRD in a developing country has been shown by a study by Khan (2010) on 30 banks in Bangladesh. He found that the sample banks disclosed about 47% of the used index. However, it is important to note that the sample they used included a few Islamic banks.

In summary, the CGD and CSRD scores seem to be higher in conventional banks. Notwithstanding, these scores cannot be compared directly with the disclosure levels in Islamic banks, because of massive divergences emanating from the nature of their operations, however, it still can be indicated that these practices were relatively of higher quality in conventional banks than in Islamic banks. Recent research by Wan Abdullah et al. (2015) suggests that the level of voluntary corporate

disclosure of Islamic banks is positively associated with strong corporate governance. Thus, using the same premise we posit that lower disclosure among Islamic banks is possibly due to weak corporate governance. In addition, the Islamic banks are considered emerging and less regulated as compared to conventional banks. Therefore, lower corporate disclosure among Islamic banks is justified.

Implications of the Literature and Directions to Future Research

Regardless of the theoretical predictions of Islamic banks’ transparency, the disclosure levels of CG and CSR were low. Therefore, this would have theoretical and practical implications which need to be carefully studied in order to obtain further understanding as well as improve CGD and CSRD practices in Islamic banks.

Theoretically, the bases on which the predictions of higher disclosures are made are still limited to the general concepts of Islam rather than specific requirements, and, therefore, might have led to lower consideration given to CGD and CSRD. In addition, since the unique products of Islamic finance would be accompanied with higher risks, transparency of information is necessary in order to reduce these risks. Therefore, the lack of transparency, especially for CG, may increase the risks embodied in Islamic instruments.

Practically, the consistency of low scores of CGD and CSR among the literature may highlight that the standards of disclosure, issued by regulators of Islamic banking, are still limited and need to be more comprehensive than its current status.

There are many limitations that could be taken into consideration when interpreting the results of the empirical literature stated above. Firstly, the use of different indices or checklists might probably culminate in massive differences between these studies. As a consequence, the lack of a united disclosure checklist would limit the comparability of these



studies. Therefore, for future research purposes, it is suggested that consideration has to be taken in using more standardised checklist for a representative sample taken from Islamic banks.

Secondly, the implicit assumption of ‘one size fits all’, embodied in the checklists used for cross-country studies may be invalid, because of the differences in environment, culture, regulations and needs of stakeholders among different countries. Accordingly, there is a need for considering these differences when comparing between banks in different countries in future research.

Thirdly, relatively small samples, which were concentrated in Southeast Asian and GCC countries, were used by most of the literature; this would limit the generalisation of these studies to the whole Islamic banking sector. Similarly, there is a lack of a longer trending analysis of transparency levels. Therefore, it would be useful for future researchers to expand samples and the studied period in order to reduce the sampling error, which would reduce the generalisability of the study.

Conclusion

The purpose of the paper is to study the degree of agreement between the ideal transparency level of CGD and CSRD in Islamic banks, given by Islamic principles, and the findings of the previous literature. We conclude that the expectations of higher disclosure, which emanates from the nature of being an Islamic bank, did not match the available empirical results of CGD and CSRD studies. Even though there are ideological motives of higher CGD and CSRD transparency, the disclosures of CG and CSRD, which are critical for Islamic banks, were found to be low. This may hinder Islamic banks to be competitive and sustainable in the economic development. Therefore, reforms need to be made by the regulators of Islamic banking and directors.

Accordingly, we recommend a number of suggestions. Firstly, the harmonisation of requirements and the invention of a standard

code of good practices of CGD and CSRD in Islamic banking sector may improve the degree of transparency and adoption. In this regard, Kasim et al. (2013) conducted a comparative study on Shariah governance standards, issued by the Accounting and Auditing Organization for Islamic Financial Institutions (AAOIFI), IFSB and Bank Negara Malaysia and found that there were many discrepancies between these standards. Consequently, they emphasized the need for standard Shariah governance framework in Islamic banks.

Secondly, the currently issued codes need to be revised in order to give more consideration to these issues. As it is clearly stated in the study by Maali et al. (2006), where many banks followed the requirements of AAOIFI, whilst the conceptually expected CSRD levels were still lower. Therefore, it may be useful to increase the obligatory requirements of CGD and CSRD, because of their significance in Islamic banks.

Thirdly, the regulatory agencies at the national and global levels may establish a committee to provide an annual rating of CGD and CSRD in Islamic banks, and thereby grant awards to banks with the highest transparency.

In summary, CGD and CSRD practices seem to be understated in Islamic banks regardless of the theoretical considerations discussed, and therefore, an improvement should be made as referred to above through sustainability initiatives for a prosperous economy. However, the review we undertook may have some limitations. Firstly, there may be undetected CGD and CSRD studies in the literature that were not taken into consideration; this would limit the degree of generalising the findings of the paper. Secondly, most of the studies in the literature used the checklist and mark-off procedure, which would have resulted in comparable percentages of disclosure. However, some studies used the number of sentences disclosed on a specific item as a measure of disclosure level, such as Abdulrahman and Bukair (2013) and Kamla and



Rammal (2013). This would make it difficult to be considered in a comparative analysis where the level of disclosure was not determined by the study to be either low or high. Accordingly, the investigation of CGD and CSRD using a large representative sample and similar scoring method would be of the utmost importance in order to give a better understanding.

Acknowledgments

The authors acknowledge the helpful comments of anonymous reviewers, and the participants of the 4th International Conference on Accounting, Business and Economics (ICABEC, 2016). The authors also gratefully acknowledge Universiti Malaysia Terengganu (UMT) for the financial assistance.

References

Abdurahman, A. A., & Bukair, A. A. (2013). The influence of the Shariah supervision board on corporate social responsibility disclosure by Islamic banks of Gulf Co-operation Council countries. Asian Journal of Business and Accounting, 6(2): 65-105.

Aksu, M., & Kosedag, A. (2006). Transparency and disclosure scores and their determinants in the Istanbul Stock Exchange. Corporate Governance: An International Review, 14(4): 277-296.

Alnajjar, F. (2000). Determinants of corporate social responsibility disclosure of US Fortune 500 firms: An application of content analysis. Advances in Environmental Accounting and Management, 1: 163-200.

Alsaati, A. (2003). The permissible Gharar (risk) in classical Islamic jurisprudence. Islamic Economics, 16(2): 65-104.

Aribi, Z. A. & Gao, S. (2010). Corporate social responsibility disclosure: A comparison between Islamic and conventional financial institutions. Journal of Financial Reporting and Accounting, 8(2): 72-91.

Barako, D. G. (2007). Determinants of voluntary disclosures in Kenyan companies annual reports. African Journal of Business Management, 1(5): 113-128.

Barako, D. G., Hancock, P., & Izan, H. Y. (2006). Relationship between corporate governance attributes and voluntary disclosures in annual reports: The Kenyan experience. Financial Reporting, Regulation and Governance, 5(1): 1-25.

Botosan, C. A. (1997). Disclosure level and the cost of equity capital. Accounting Review, 72(3): 323-349.

Cadbury, S. A. (2000). The corporate governance agenda. Corporate Governance: An International Review, 8(1): 7-15.

Choudhury, M., & Alam, M. N. (2013). Corporate governance in Islamic perspective. International Journal of Islamic and Middle Eastern Finance and Management, 6(3): 180-199.

Cochran, P. L., & Wood, R. A. (1984). Corporate social responsibility and financial performance. Academy of Management Journal, 27(1): 42-56.

Collett, P., & Hrasky, S. (2005). Voluntary disclosure of corporate governance practices by listed Australian companies. Corporate Governance: An International Review, 13(2): 188-196.

Craig, R., & Diga, J. (1998). Corporate accounting disclosure in ASEAN. Journal of International Financial Management & Accounting, 9(3): 246-274.

Dahlsrud, A. (2008). How corporate social responsibility is defined: An analysis of 37 definitions. Corporate Social Responsibility and Environmental Management, 15(1): 1-13.

Darmadi, S. (2013). Corporate governance disclosure in the annual report: An



exploratory study on Indonesian Islamic banks. Humanomics, 29(1): 4-23.

Deegan, C., & Blomquist, C. (2006), Stakeholders influence on corporate reporting: An exploration of the interaction between WWF-Australia and Australian minerals industry. Accounting Organization and Society, 31(4): 343-72.

Douglas, A., Doris, J., & Johnson, B. (2004). Corporate social reporting in Irish financial institutions. The TQM Magazine, 16(6): 387-395.

El-Hawary, D., Grais, W., & Iqbal, Z. (2004). Regulating Islamic financial institutions: The nature of the regulated. World Bank Policy Research Working Paper No. 3227. Retrieved from https://ssrn.com/abstract=610268, 23 March 2017.

Ernst & Young (2016). World Islamic Banking Competitiveness Report. Retrieved from http://www.ey.com/Publication/vwLUAssets/ey-world-islamic-banking- compet i t iveness-report-2016/$FILE/ey-world-islamic-banking-competitiveness-report-2016.pdf, 23 February 2017.

Eweje, G. (2014). Introduction: Trends in corporate social responsibility and sustainability in emerging economies. In G. Eweje (Eds.), Corporate social responsibility and sustainability: Emerging trends in developing economies (Critical Studies on Corporate Responsibility, Governance and Sustainability) (pp.3 – 17). Bingley: Emerald Group Publishing Limited.

Farook, S., Kabir Hassan, M., & Lanis, R. (2011). Determinants of corporate social responsibility disclosure: The case of Islamic banks. Journal of Islamic Accounting and Business Research, 2(2): 114-141.

Friedman, A. L. & Miles, S. (2001). Socially responsible investment and corporate social and environmental reporting in the UK: An exploratory study. British Accounting Review, 33(4): 523-548.

Grais, W., & Pellegrini, M. (2006). Corporate governance in institutions offering Islamic financial services: Issues and options. World Bank Policy Research Working Paper No. 4052. World Bank, Washington. 45 pp.

Haniffa, R., & Hudaib, M. (2007). Exploring the ethical identity of Islamic banks via communication in annual reports. Journal of Business Ethics, 76(1): 97-116.

Hasan, Z. (2009). Corporate governance: Western and Islamic perspectives. International Review of Business Research Papers, 5(1): 277-293.

Hassan, A., & Harahap, S. (2010). Exploring corporate social responsibility disclosure: The case of Islamic banks. International Journal of Islamic and Middle Eastern Finance and Management, 3(3): 203-227.

Hassan, S., & Christopher, T. (2005). Corporate governance statement disclosure of Malaysian banks and the role of Islam. Asian Review of Accounting, 13(2): 36-50.

Islamic Financial Standards Board (2007). Disclosures to promote transparency and market discipline for institutions offering Islamic financial services. Retrieved from http://www.ifsb.org/standard/ifsb4.pdf, 23 April 2016.Kamla, R., & Rammal, H. G. (2013). Social reporting by Islamic banks: Does social justice matter? Accounting, Auditing & Accountability Journal, 26(6): 911-945.

Kasim, N., Nuhtay, S., & Salman S. (2013). Comparative Analysis on AAOIFI, IFSB and BNM Shari’ah Governance



Guidelines. International Journal of Business and Social Science, 4(15): 220-227.

Kelton, A. S., & Yang, Y. W. (2008). The impact of corporate governance on Internet financial reporting. Journal of Accounting and Public Policy, 27(1): 62-87.

Khan, H. U. Z. (2010). The effect of corporate governance elements on corporate social responsibility (CSR) reporting: Empirical evidence from private commercial banks of Bangladesh. International Journal of Law and Management, 52(2): 82-109.

Maali, B., Casson, P., & Napier, C. (2006). Social reporting by Islamic banks. Abacus, 42(2): 266-289.

Magalhães R., & Al-Saad S. (2013). Corporate governance in Islamic financial institutions: The issues surrounding unrestricted investment account holders. Corporate Governance: The International Journal of Business in Society, 13(1): 39-57.

Maher, M., & Andersson, T. (2000). Corporate governance: Effects on firm performance and economic growth. Retrieved from http://ssrn.com/abstract=218490, 20 June 2016.

Maingot, M., & Zeghal, D. (2008). An analysis of corporate governance information disclosure by Canadian banks. Corporate Ownership and Control, 5(2): 225-236.

Mallin, C., Farag, H., & Ow-Yong, K. (2014). Corporate social responsibility and financial performance in Islamic banks. Journal of Economic Behavior & Organization, 103: S21-S38.

O’Dwyer, B., & Gray, R. (1998). Corporate social reporting in the Republic of Ireland: A longitudinal study. Irish Accounting Review, 5(2): 1-34.

Ousama, A. A., & Fatima, A. H. (2010). Voluntary disclosure by Shariah approved companies: An exploratory study. Journal of Financial Reporting and Accounting, 8(1): 35-49.

Platonova, E., Asutay, M., Dixon, R., & Mohammad, S. (2016). The impact of corporate social responsibility disclosure on financial performance: Evidence from the GCC Islamic banking sector. Retrieved from https://link.springer.com/article/10.1007/s10551-016-3229-0, 1 December 2016.

Samaha, K., & Dahawy, K. (2010). An empirical analysis of corporate governance structures and voluntary corporate disclosure in volatile capital markets: The Egyptian experience. International Journal of Accounting, Auditing and Performance Evaluation, 7(1-2): 61-93.

Sharma, N. (2014). Extent of corporate governance disclosure by banks and finance companies listed on Nepal Stock Exchange. Advances in Accounting, 30(2): 425-439.

Shleifer, A., & Vishny, R. W. (1997). A survey of corporate governance. The Journal of Finance, 52(2): 737-783.

Smolo, E., & Mirakhor, A. (2010). The global financial crisis and its implications for the Islamic financial industry. International Journal of Islamic and Middle Eastern Finance and Management, 3(4): 372-385.

Srairi, S. (2015). Corporate governance disclosure practices and performance of Islamic banks in GCC countries. Journal of Islamic Finance, 4(2): 001-017

Sulaiman, M., Majid, N. A. & Ariffin, N. M. (2015). Corporate governance of Islamic financial institutions in Malaysia. Asian Journal of Business and Accounting, 8(1): 65-93.



Wan Abdullah, A. W., Percy, M. & Stewart, J. (2013). Shari’ah disclosures in Malaysian and Indonesian Islamic banks: The Shari’ah governance system. Journal of Islamic Accounting and Business Research, 4(2): 100-131.

Wan Abdullah, A. W., Percy, M. & Stewart, J. (2015). Determinants of voluntary corporate governance disclosure: Evidence from Islamic banks in the Southeast Asian and the Gulf Cooperation Council regions. Journal of Contemporary Accounting and Economics, 11(3): 262-279.

Yusoff, H., Darus, F., Fauzi, H., & Purwanto, Y. (2013). Exploring the environmental reporting practices of Islamic banks: A case of Malaysia and Indonesia. Journal of Energy Technologies and Policy, 3(11): 440-445.

Zubairu, U. M., Sakariyau, O. B., & Dauda, C. K. (2011). Social reporting practices of Islamic banks in Saudi Arabia. International Journal of Business and Social Science, 2(23): 193-205.


GUIDELINE FOR AUTHORS

Submission of articles on subjects of relevance to the journal is welcome at any time. Authors interested in submitting a paper, should send an abstract of the article using an online submission. Submission should adhere to our guidelines. UMT currently publishes two issues per year (June and Dicember- JSSM).

Manuscripts should be in English or Bahasa Melayu and submitted to: scholarjournal.umt.edu.my.

Papers are accepted on the understanding that they are original work that has not been and will not be published elsewhere. All submissions will be peer reviewed. The editors have the final decision on publication. Papers on acceptance become the copyright of the journal. The journal will be printed in black and white.

PREPARATION OF MANUSCRIPTS

JSSM accepts research/ original article, short communications & reviews article. All papers should adhere to the scope and policy of JSSM and should relate to social sustainability.

Manuscripts should be typed with a wide margin and double spaced, Times New Roman (Font 12), on one side of A4 (30 x 21 cm) paper and usually should not exceed 20 pages in length. In the text, scientific names should be italicized; do not underlines any words. Avoid excessive use of italics to emphasize text. In general, manuscripts should be organized in the following order:

Research/Original Article

Title (should be clear, descriptive and concise)Name of authorsComplete postal address of affiliationsE-mail address of the corresponding authorAbstractKeywords (indexing terms): 3 – 5 itemsIntroductionResearch MethodsResults and discussionConclusion

Acknowledgements (additional information concerning research grants)References

Review Article (It does not have a methods or a results section, no study to describe & no data to be analyzed)

Title (should be clear, descriptive and concise)Introduction (where the topic or problem are introduced & previous research is described)Discussion (main points are discussed)Conclusion (article’s strengths & weaknesses)Reference

TITLE AND SUBTITLES

Titles and subtitles should not be run within the text. They should be typed on a separate line without indentation. Use bold face, lower-case letter type for titles. Use italics letter type for subtitles.

LENGTH

Including illustrations, tables and references, original contributions should not exceed 25 manuscript pages or 18,000 characters. Short communications are limited to 5 manuscript pages or 7,500 characters.

UNITS AND ABBREVIATIONS

SI-units should be used wherever possible. If other units and non-standard abbreviations cannot be avoided, they should be defined at first mention.

ABSTRACT

The abstract should be clear, descriptive and not longer than 400 characters. It should provide a very brief introduction to the problem and a statement about the methods used in the study. This should generally be followed by a brief summary of results, including any important numerical data (means and standard errors, for example). The abstract should end with an indication of the significance of the results.

KEYWORDS

Please include relevant keywords in the manuscript. Keywords should not be the same as the word appears in the title.

TABLES

Authors should take notice of the limitations set by the size and layout of the journal. Large tables should be avoided. Reversing columns and rows will often reduce the dimension of a table. Tables should be placed and typewritten in the text. Each table should have a brief self-explanatory title & numbered accordingly. Column headings should be brief, but sufficiently explanatory. Standard abbreviations of units of measurement should be added between parentheses. Any explanation essential to the understanding of the table should be given as a footnote at the bottom of the table.

FIGURES

Figures should be saved in a neutral data format such as TIFF or ESP and a printout should always be included. All figures should be placed and typewritten in the text. Figures should be numbered consecutively according to their sequence in the text. Reference should be made in the text to each figure. Lettering should be clear and large enough to allow a reduction of 50% without becoming illegible. The lettering should be in English. Each figure should have a caption. The captions to all illustrations should be typed on a separate sheet of the manuscript. Explanation should be given in the typewritten legend. Drawn text in the illustrations should be kept to a minimum. Photographs are only acceptable if they have good contrast and intensity. Sharp and glossy copies are required. Reproductions of printed photographs cannot be accepted.

REFERENCES

JSSM uses APA system. All publications cited in the text should be presented in a list of references following the text of the manuscript. The manuscript should be carefully checked to

ensure that the spelling of author’s names and the dates of publication are exactly the same in the text as in the reference list. If reference is made in the text to a publication written by more than two authors the name of the first author should be used followed by “et al.,”. References cited together in the text should be arranged chronologically. The list of references should not exceed 25 for regular article and should be arranged alphabetically by authors’ names, and chronologically per author. Papers accepted for publication but not yet published should be cited as “in press”. References concerning unpublished data and ‘personal communications’ should not be cited in the reference list but may be mentioned in the text. Use the following system for arranging your references:

[Listed in numerical order in the References, Times New Roman, font size 12, justified with APA 6th edition format system]:

Articles in Journal

Author, A. (year). Title of paper. Journal title, volume(issue number), pages

Author, A., & Author, B. (year). Title of paper. Journal title, volume(issue number), pages

Author, A., Author, B., & Author, C. (year). Title of paper. Journal title, volume(issue number), pages

Dame, R., & Libes, S. (1993). Oyster reefs and nutrient retention tidal creeks. Journal of Experimental Marine Biology and Ecology, 171: 251-258.

Book

Authors, years, book title. publisher name, country. volume, issue number: and pages

Zar, J. H. (1999). Biostatistical analysis, 4th edition. Prentice-Hall, Englewood Cliffs, New Jersey. 718 pp.

Multi-author books

Authors, years, book title. publisher name, country. volume, issue number: and pages

Kernis, M. H., Cornell, D. P., Sun, C. R., Berry, A., Harlow, T., & Bach, J. S. (1993). There’s more to self-esteem than whether it is high or low: The importance of stability of self-esteem. Journal of Personality and Social Psychology, 65: 1190-1204

Thesis

Author, years, thesis title, degree, and university

Khalik, W. M. A. W. M. (2012). Status kualiti air Tasik Temenggor, Perak. Thesis of Master Degree, Universiti Kebangsaan Malaysia.

Webpage

Author or institution, year, title, link website, date access

Kenneth, I. A. (2001). How to be an entrepreneur. Retrieved from http://www.business.com/money/how.html, 20 August 2016.

SPECIAL ISSUES

Special issues are papers from symposiums/conferences.10-20 selected papers can be accommodated in one issue. Papers should not exceed 8-10 printed pages, or 5,000 words.

Procedure for Special Issues in JBSD/JSSM is as follows:

• The Guest Editor from the Seminar/Symposium is the Chair of the Scientific Committee of the Seminar/Symposium. The Guest Editor and the Guest Editorial Board are responsible for collecting the papers (should adhere to JSSM format). The Board shall also be responsible in identifying 2 reviewers for each paper, and in completing the blind peer-review process.

• When the reviews are received, the Editorial Members of JSSM are responsible for the final approval of the papers for publication in the special issue.

• Research Publication Section may not cover the cost of printing for the Special Issues. The Board/Committee are responsible for every cost needed.

ALTERATION IN PROOF

Authors will receive an email message from the Managing Editor of Journal of Sustainable Science and Management providing directions for checking of galley proofs. Authors should return any corrections to the galleys to the Managing Editor as soon as possible. A limited number of alterations in proof are unavoidable, but the cost of making extensive corrections and changes after an article has been composed will be charged to the author. Final responsibility for correcting errors lies with the author. Article is freely available in JSSM’s website.

LIST OF REVIEWERS

Ahmad Suhaimi Baharudin, USM, MalaysiaAminuddin Md Arof, UniKL, Malaysia

Asan Ali Golam Hassan, UTM, MalaysiaDr. Tajul Ariffin Masron, USM, Malaysia

Fatimah Kari, UM, MalaysiaHasnah Binti Haron, UMP, Malaysiati

Jamaliah Said, UTM, MalaysiaK. Kuperan Viswanathan, UUM, Malaysia

Mohamad Rosni Bin Othman, UMT, MalaysiaNazatul Faizah, UniSZA, Malaysia

Noorhayati Binti Mansor, UniSZA, MalaysiaNor Azila Mohd Noor, UUM, MalaysiaNormaz Wana Ismail, UPM, Malaysia

Safiek Mokhlis, UMT, MalaysiaSaharuddin Abdul Hamid, UMT, Malaysia

Siti Hasnah Hassan, USM, MalaysiaSuriyani Muhamad, UMT, Malaysia

Wan Amalina Wan Abdullah, UniSZA, MalaysiaZainudin Awang, UniSZA, Malaysia

ASSOCIATE EDITORSAkmalia Mohamad Ariff, UMT, Malaysia

Azwadi Ali, UMT, Malaysia

164A REVIEW OF CORPORATE GOVERNANCE AND CORPORATE SOCIAL RESPONSIBILITY DISCLOSURE OF ISLAMIC BANKS: A CALL FOR ADDITIONAL TRANSPARENCYJournal of Sustainability Science and Management Special Issue No. 4 2018The Role of Transparency and Management Integrity in Organisational Sustainability----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

i-vi Editorial: The Role of Transparency and Management Integrity in Organisational Sustainability

AKMALIA MOHAMAD ARIFF AND AZWADI ALI

1-14 Combating Corruption for Sustainable Public Services in Malaysia: Smart Governance Matrix and Corruption Risk Assessment

WAN MOHAMMAD TAUFIK WAN ABDULLAH, NURUL NADIAH AHMAD AND AKMALIA MOHAMAD ARIFF

15-33 Decision Making Approach of Direct Feeder Service Via Port Klang in Malaysia MUHAMAD NASIR RAHMATDIN, NOORUL SHAIFUL FITRI ABDUL

RAHMAN AND MOHAMMAD KHAIRUDDIN OTHMAN

35-52 Moderating Effect of Self-Efficacy on the Relationship Between Consumer Attitude and Willingness to Consume Healthy Food

HAYATUL SAFRAH SALLEH AND NOR AZILA MOHD NOOR

53-65 Enhancing Environmental Sustainability Over Fisheries Industry Through Proactive Risk Evaluation: A Case of Tok Bali Fishing Port

NURUL HAQIMIN MOHD SALLEH AND MOHD AZLAN ABDUL HALIM

67-79 An Assessment of the Perception and Satisfaction With Flood Evacuation Centre Service Quality in East Coast States of Peninsula Malaysia

SITI FALINDAH PADLEE, NIK NADRUL HISHAM NIK RAZALI, SITI NUR ATIKAH ZULKIFFLI AND NOOR ZATUL IFFAH HUSSIN

81-100 Foreign Direct Investment and Trade Openness: Do They Lead to Sustainable Development in Malaysia?

ABDUL RAHIM RIDZUAN, NOR ASMAT ISMAIL AND ABDUL FATAH CHE HAMAT

101-117 The Influence of New Economic Model on Intention to Work Abroad: An Empirical Study to Address Brain Drain Issue in Malaysia

NORHAYATI SHARIFF, JAYARAMAN KRISHNASWAMY, DAHLAN ABDULLAH AND ANG WOON CHAU

119-133 The Influence of Energy Consumption, Renewable Energy and Economic Growth on CO2 Emission in Malaysia and Indonesia

IZYAN NAZIRAH WAHID, AZLINA ABD. AZIZ, MAHIRAH KAMALUDIN AND ZURAINI ANANG

135-146 A Survey on the Willingness to Pay for Domestic Water Service Attributes in Terengganu, Malaysia

MAHIRAH KAMALUDIN, AZLINA ABD. AZIZ, NUR SYUHADA CHE IBRAHIM AND ALIAS RADAM

147-168 A Review of Corporate Governance and Corporate Social Responsibility Disclosure of Islamic Banks: A Call for Additional Transparency

SAED AHMED SULUB, ZALAILAH SALLEH AND HAFIZA AISHAH HASHIM