impact of wealth distribution on energy consumption in nigeria: a
TRANSCRIPT
1
IMPACT OF WEALTH DISTRIBUTION ON ENERGY CONSUMPTION IN NIGERIA: A
CASE STUDY OF SELECTED HOUSEHOLDS’ IN GOMBE STATE. By
Bello, Maryam
Department of Economics, Faculty of Arts and Social Sciences
Gombe State University, Gombe, Nigeria – West Africa
+234 802 374 2740
Abstract
The world is facing strong threat of climate change caused by carbon dioxide emissions from the use of unconventional
energy sources which have negative impacts on the environment and human health. Biomass is the primary source of energy
for majority of the population in Africa; in Sub-Saharan African Countries fuelwood and charcoal are the staple energy
sources for most rural and urban communities. Using Cross-Sectional survey data from a sample of 500 households’ in
Gombe State of Nigeria the study investigates the impact of wealth distribution on energy consumption, and analysis the
determinants of households’ energy choice for cooking. The simple descriptive statistics and multinomial logit model is used
in analysing the data obtained. An empirical result of the logit model reveals that the choice of cooking energy is mainly
determined by the economic wealth of households’. Besides the economic wealth, the analysis also shows that size of
households’ and level of education are found to be key factors in energy consumption behaviour, especially when dealing
with energy source switching. Economic wealth of households’ was found to be a major determinant of the type of cooking
energy used by households’ in Nigeria. Wood was used by the low-income households’ as the main source of cooking energy
while the modern fuels are used by the upper class in the society.
1.0 INTRODUCTION In developing countries, most of the rural as well as urban communities have less access to modern and clean energy sources
and mostly depend on traditional fuel /biomass (woods, twigs, leaves, charcoal, animal dung and crop residue) for virtually
all their energy requirements. It is estimated that approximately 2.5 billion people in developing countries rely on biomass
fuels to meet their cooking needs. For many of these countries, more than 90 percent of total household fuel is biomass.
Without new policies, the number of people that rely on biomass fuels is expected to increase to 2.6 billion by 2015, and 2.7
billion by 2030 (about one-third of the world‘s population) due to population growth (IEA 2006). While rural households‘
rely more on biomass fuels than those in urban areas, well over half of all urban households‘ in sub-Saharan Africa rely on
fuelwood, charcoal, or wood waste to meet their cooking needs (IEA 2006).
With increasing population and urbanization over time, urban household energy is an important issue for developing
countries in general. Heavy reliance of urban households‘ in Sub-Saharan Africa on biomass fuels (such as woody biomass
and dung) contribute to deforestation, forest degradation, and land degradation. This is partly because use of these fuels in
urban areas is an important source of cash income for people in both urban and rural areas. While use of woody biomass as
fuel and as construction material contributes to deforestation and forest degradation, use of dung as fuel implies that it might
not be available for use as fertilizer—thus contributing to land degradation and consequent reduction in agricultural
productivity.
Over 60% of Nigeria's population depends on fuelwood for cooking and other domestic uses (ECN, 2003). The rural areas
have little access to conventional energy such as electricity and petroleum products due to absence of good road networks.
Petroleum products such as kerosene and gasoline are purchased in the rural areas at prices very high in excess of their
official pump prices. The rural populace, whose needs are often basic, therefore depend to a large extent on fuelwood as a
major traditional source of fuel. It has been estimated that about 86% of rural households‘ in Nigeria depend on fuelwood as
their source of energy (Williams, 1998). Fuelwood supply/demand imbalance in some parts of the country is now a real
threat to the energy security of the rural communities (ECN, 2003).
2
Nigeria consumes over 50 million metric tonnes of fuelwood annually, a rate, which exceeds the replenishment rate through
various afforestation programme (ICCDD, 2000). Sourcing fuelwood for domestic and commercial uses is a major cause of
desertification in the arid-zone states and erosion in the southern part of the country (Sambo, 2009). The rate of deforestation
is about 350,000 hectares per year, which is equivalent to 3.6% of the present area of forests and woodlands, whereas
reforestation is only at about 10% of the deforestation rate (ICCDD, 2000). From available statistics, the nation‘s 15 million
hectares of forest and woodland reserves could be depleted within the next fifty years (ECN, 2003). These would result in
negative impacts on the environment, such as soil erosion, desertification, loss of biodiversity, micro-climatic change and
flooding. Most of these impacts are already evident in different ecological zones in the country, amounting to huge economic
losses (Sambo, 2009). The consumption of fuelwood is worsened by the widespread use of inefficient cooking methods that
are hazardous to human health, especially to women and children who mostly do the cooking in homes.
It has been argued that households‘ with low income levels rely on biomass fuels, such as wood and dung, while those with
higher incomes consume energy that is cleaner and more expensive, such as Liquid petroleum gas (LPG). Those households‘
in transition consume what are called transition fuels, such as kerosene and charcoal. This fuel choice and consumption
behaviour of households‘ is known as the ―energy ladder hypothesis‖.
Apart from high income, one set of factors necessary for switching to other fuels households‘ is cheap and better availability
of alternative fuels other than traditional biomass fuels. Empirical evidence has shown that for many households‘, the
decision over which fuel to use or how much of the fuel to use, requires the consideration of several important factors. For
instance Narain et al (2008) found that fuelwood use and dependence (defined as its contribution to the total ‗permanent
income‘ of households‘) increases with forest biomass availability irrespective of income levels. Also, access to electricity
has been found to be another important determinant of the energy transition (Campbell et al. 2003; Davis 1998; Ouedraogo
2006). Others are house standard, level of education of husband and wife, occupation of wife, frequency of cooking certain
meals and household size (Alam et al. 1998; Ouedraogo 2005; Madubansi and Shackleton 2007; Pundo and Fraser, 2006).
The aim of this study is to investigate the impact of wealth distribution on energy consumption, and to analyse the
determinants of households‘ energy choice for cooking in Gombe State of Nigeria.
The paper is divided into five sections. The introduction in section one is followed by section two which presents the
conceptual framework, and a brief review of some theoretical and related literature. The data collected and methods of
analysis are discussed in section three while section four presents and discusses the results. Finally, the paper is concluded in
section five and some policy recommendations are also given.
2.0 REVIEW OF RELATED LITERATURE AND SOME THEORETICAL ISSUES
2.1 Defining the Concept of Wealth
In common use ‗wealth‘ means money, property, gold and so on. But in economics it is used to describe all things that have
value. Wealth is the total value of a person‘s net worth expressed as:
Wealth = assets − liabilities
Wealth may be held in various forms: these include money, shares in companies, debt instruments, land, buildings,
intellectual property, and valuables etc (Black, 2002).The wealth of individuals is believed to affect their choices of
consumption, thus the part of wealth that have direct influence on consumption is income. Understanding the effect of
income on consumption has been a central point of a great deal of theorizing work for example the Keynes ―Absolute
Income Hypothesis‖, ―Duesenberry‘s ―Relative Income Hypothesis‖, Modigliani and Brumberg ―Life Cycle Hypothesis‖ and
the ―Permanent Income Hypothesis‖ of Friedman.
3
The preferred theory here is the Keynesian – Absolute income hypothesis which is based on the assumption that,
consumption depends on income. Or simply, the major influence on personal consumption is an individual income. And this
is what portrays the real picture of the consumption pattern of the people living in the study area.
2.2 The Energy Ladder Model
Household fuel choice has often been conceptualized using the ―energy ladder‖ model (Figure1); the model emphasized the
role of income in determining fuel choices and fuel switching.
Figure 1: Energy Ladder Model
From W.H.O (2006) (Figure 1: The energy ladder: household and development inextricably linked)
Note: Ethanol and methanol are rarely, if ever, used.
Dash: estimate
WHO (2006) (Figure 1: The energy ladder: household energy and development inextricably linked)
Note: Ethanol and Methanol are rarely, if ever, used.
Dash: estimate
The energy ladder model envisions a three-stage fuel switching process. The first stage is marked by universal reliance on
biomass. In the second stage households‘ move to ―transition‖ fuels such as kerosene, coal and charcoal in response to higher
incomes and factors such as deforestation and urbanization. In the third phase households‘ switch to LPG, natural gas, or
electricity. The main driver affecting the movement up the energy ladder is hypothesized to be income and relative fuel
prices (Leach, 1992; Barnes, Krutilla, and Hyde, 2002; Barnes and Floor, 1999).
The major achievement of the energy ladder model in its simplest form is the ability to capture the strong income
dependence of fuel choices. Several studies have been conducted to test the energy ladder hypothesis. Hosier and Dowd
(1987) conducted a study in urban Zimbabwe using a multinomial logit model, the result revealed that although economic
factors do affect fuel choices, a large number of other factors such as culture, social desirability and security of supply are
also important in determining household fuel choice (Hosier and Dowd, 1987).
An investigation of household energy choices for a sample of households‘ residing in the city of Bangalore uses a binomial
logit model (Reddy, 1995). A binomial logit is defined as a model, which determines the choice between each pair of energy
sources. This model according to Reddy (1995) helps to explain the shift in the energy pattern of consumption of different
Very low
income
Low income Middle income High income
Electricity
Natural gas
Kerosene
Gas, liquefied petroleum gas
Ethanol, methanol
Coal
Charcoal
Wood
Crop waste, dung
Solid fuels
Non-solid
fuels
Increasing prosperity and development
Incr
easi
ng u
se o
f cl
ean
er,
more
eff
icie
nt
an
d m
ore
con
ven
ien
t fu
els
for
cook
ing
4
fuels used for cooking and water heating. The findings confirm that urban households‘ ascend an energy ladder and the
choice is determined by income. However, other factors worth noting that play a significant role in fuel switching amongst
households‘ is family size and occupation of head of the household (Reddy, 1995).
A similar study in India also employed a multinomial logit framework to represent household fuel choice (World Bank,
2003). However in the World Bank model households‘‘ decisions concerning the choice of both cooking and lighting fuels
are dealt with together. The World Bank took a closer look at a choice set that consists of all the key alternatives to different
energy sources combinations used by a household. The objective of this model was to study the effectiveness of the existing
price subsidies in facilitating a shift to cleaner and more efficient fuels like kerosene and LPG. The results showed that
subsidies are unsustainable in meeting social policy objectives and disproportionately favours the rich (World Bank, 2003).
The study done in India shows that degree of urbanisation has been shown to influence energy consumption such that
households‘ living in larger cities consume more electricity than the inhabitants of cities with less than one million
inhabitants (Horsier and Kipondya, 1993). According to (Madubansi and Shackleton, 2005) this is also true in South African
context where larger cities and townships have well developed infrastructure and sufficient supplies of electricity. As a
consequence, better employment opportunities that exist in large cities also enable households‘ to allocate more of their
income to modern fuels. However, the limited employment opportunities limit electricity consumption in poor urban
townships in South Africa (Madubansi and Shackleton, 2005). The changes in the consumption patterns in low-income urban
households‘ disprove the energy ladder model as they continue to consume energy sources at lower end of the ladder (Davis,
1998).
The energy ladder model has been criticized, since there is widespread use of multiple fuels for a particular purpose (such as
cooking) which results to fuel stacking for a given purpose (Davis 1998; Heltberg 2005).
3.0 DATA AND METHODOLOGY
This section presents the data source, followed by the specification of the model and description of the exogenous variables.
3.1 Data Source:
This study was carried out within Gombe State, located on Latitude 9030' and 12
030N and Longitude 8
045' and 11
045E in the
North East Region of Nigeria-West Africa.
The data for this study was obtained from a survey of 500 households‘ in 3 Local Government Areas of Gombe State; they
are Akko, Yamaltu – Deba and Gombe representing each senatorial district. The study was conducted between February
2009 and January 2010. A multi-stage cluster sampling technique was employed to select the households‘ of the study. In the
first stage, the households‘ were clustered into three: metropolitan, urban and rural with a distribution of 1 metropolitan, 2
urban and 4 rural areas. These were selected at random at this stage, and then the households‘ in the selected local
government areas were sub-clustered on the basis of high, medium and low income to see the impact of income in
determining the likelihood of households‘ to demand fuelwood. The procedure yielded a sample of 523 households‘.
Questionnaires were administered to the 523 households‘, during the process of cleaning the data, 9 questionnaires were
found to be missing and 14 questionnaires were rendered invalid due to one or more key variables missing or inconsistent
information and were discarded. The remaining 500 were valid. The data collected relate to households‘‘ sociological and
economical characteristics and their expenditures.
5
3.2 Model Specifications
The data collected was analysed first using the simple descriptive statistical methods of percentages and graphical
representations, followed by estimation of the binary choice models.
The study uses multinomial logit model to estimate the significance of the factors believed to influence a households‘‘ choice
of cooking fuel in Gombe State of Nigeria. Multinomial logit model describes the behaviour of consumers when they are
faced with a variety of goods with a common consumption objective. The choice of the model was based on its ability to
perform better with discrete choice studies (McFadden, 1974 and Judge, et al, 1985). However, the goods must be highly
differentiated by their individual attributes. For example, the model examines choice between a set of mutually exclusive and
highly differentiated cooking fuels such as fuelwood, kerosene, cooking gas, and electricity. If only two discrete choices
have to be analysed, the multinomial logit model reduces to a binomial logit model.
The probability that a household chooses one type of cooking fuel is restricted to lie between zero and one. The model
assumes no reallocation in the alternative set and no changes in fuel prices or fuel attributes. The model also assumes that
households‘ make fuel choices that maximize their utility (McFadden, 1974). The model can be expressed as follows:
Pr [Yi = j] =
j
j
ij
ij
X
X
0
)exp(
)exp(
………………………………………………………..(1)
Where:
Pr[Yi = j] is the probability of choosing either kerosene, cooking gas or electricity with fuelwood as the reference
cooking fuel category,
J is the number of fuels in the choice set,
j = 0 is fuelwood,
Xi is a vector of the predictor (exogenous) socio-economic factors
(variables)
βj is a vector of the estimated parameters.
When the logit equation in equation 1 above is rearranged using algebra, the regression
equation is as follows:
Pi = )....(
)....(
11
11
1 vvo
vvo
xbxbb
xbxbb
e
e
………………………………………………….(2)
The equation used to estimate the coefficients is
In [i
i
P
P
1] =b0 + b1x1 +…bv xv + µi................................................................................................................... (3)
From equation 3, the quantity Pi/ (1 – Pi) is the odds ratio. In fact, equation 3 has expressed the logit (log odds) as a linear
function of the independent factors (Xs). Equation 3 allows for the interpretation of the logit weights for variables in the
same way as in linear regressions. For example, the variable weights refer to the degree to which the probability of choosing
one fuelwood alternative would change with a unit change in the variables. For example, e bv(in equation 2) is the
multiplicative factor by which the odds ratio would change if X changes by one unit.
The model follows from the assumption that the random disturbance terms are independently and identically distributed
(McFadden, 1974). In addition, Judge et al (1985) shows that even if the number of alternatives is increased (from 2 to 3 to 4
6
etc) the odds of choosing an alternative fuel remain unaffected. That is, the probability of choosing the fuel remains the same
if it is compared to one alternative or if it is compared to two alternative fuels. The dependent variable is the cooking fuel
choice (fuelwood, kerosene or cooking gas) with fuelwood as the reference choice. Estimated coefficients measure the
estimated change in the logit for a one-unit change in the predictor variable while the other predictor variables are held
constant. A positive estimated coefficient implies an increase in the likelihood that a household will choose the alternative
fuel. A negative estimated coefficient indicates that there is less likelihood that a household will change to alternative fuel.
P-value indicates whether or not a change in the predictor significantly changes the logit at the acceptance level. That is, does
a change in the predictor variable significantly affect the choice of response category compared to the reference category? If
p-value is greater than the accepted confidence level, then there is insufficient evidence that a change in the predictor affects
the choice of response category from reference category. The exogenous variables are defined in Table 1.
Table 1: Definition of Exogenous Variables
S/N Variable Meaning Value
1. EFW Monthly Expenditure on Fuelwood in N
2. EKR Monthly Expenditure on Kerosene in N
-
3. ECG Monthly Expenditure on Cooking Gas in N
-
4. PSC Price of Stove or Cooker in N -
5. INC Monthly Income of Household in N
-
6. SZH Size of Household (No. of people in a
residence)
-
7. HHE Head of Household Education 1 if head of household has at least post
secondary education, 0 otherwise.
8. HWE Housewife Education 1 if housewife has at least post secondary
education, 0 otherwise.
4.0 EMPIRICAL RESULTS AND DISCUSSIONS
In this section, the empirical analysis starts by the presentation of the demographic characteristics of the 500 households‘ in
other to gain insight into their socioeconomic features which may guide the analysis. The data collected was analysed first
using the simple descriptive statistical methods of percentages and diagrammatic representations, and then the consumption
function of households‘ was estimated using the multinomial logit choice model.
7
4.1 Socio-Economic Characteristics of Households’
S/N VARIABLE FREQUENCY
PERCENTAGES
1. Location (place of origin)
Rural 399 79.8
Urban 101 20.2
2. Gender (sex of the respondents)
Female 380 76
Male 120 24
3. Age (Age in years of the respondent)
Below 20 13 2.6
20-30 75 15
31-40 227 45.4
41-50 152 30.4
51-60 21 4.2
61 Above 12 2.4
4. Occupation (occupation of respondents)
Civil servant 120 24
Farmer 215 43
Trading 110 22
Others 55 11
5. Size of households‘
1-3 78 15.6
4-6 126 25.2
7-9 207 41.4
10-12 36 7.2
13-15 28 5.6
Above 15 25 5.0
6. Level of education of respondents
Informal only 187 37.4
Primary 56 11.2
Secondary 115 23
Diploma/NCE 85 17
Higher National Diploma (HND) 17 3.4
Graduate 31 6.2
Post graduate 9 1.8
8
4.2 Households’ Cooking Choice
Fig.2 indicates that the dominating source of household cooking energy in the study area is wood-energy, which is used by
70% of the households‘. Kerosene is mainly used by 21% of the households‘, 7% of the households‘‘ uses cooking gas and
only 2% utilize other forms of cooking energy. Although Electricity was included as energy options in the survey
questionnaires, it recorded zero response as none of the respondents utilized it as their main source of fuel but rather as a
backup.
Fig. 2 Distribution of households‘ by cooking energy choice
4.3 Household Wealth and Fuel Choice
Fig. 3 depicts the relationship between a fuel choice and a households‘‘ wealth (income). Fig 3 shows that the low-income
households‘ are the main users of fuelwood with an average income of about N 8000.00. The reversed pattern is observed for
kerosene and cooking gas. The use rate of kerosene and cooking gas is highest among the richest household with an average
total monthly income of about N15000.00 for kerosene and N35000.00 for cooking gas indicating a movement to cleaner
fuel as income increases.
Fig. 3 Energy Choice and Household Income
70%
21%
7%
2%
Fuelwood
Kerosene
Cooking gas
Others
0
5000
10000
15000
20000
25000
30000
35000
40000
Fuelwood Kerosene Cooking gas
Me
an t
ota
l ho
use
ho
ld m
on
tnly
in
com
e (
Nai
ra)
Energy types
Income
9
4.4 Multinomial Logit Analysis for Kerosene and Cooking Gas as Compared to Fuelwood
Variable Kerosene Cooking gas
Parameter
coefficient
p-value Odds
ratio
Parameter
coefficient
p-value Odds ratio
Expenditure on
energy for cooking
0.653
(0.152)
0.182 1.921 0.081
(0.002)
0.966 1.084
Monthly Income
of Households‘
(INC)
0.863
(0.419)
0.039 2.370 0.848
(0.172)
0.494 2.335
Size of household
(SZH)
-0.304
(0.031)
0.000 0.738 -0.212
(0.961)
0.028 0.808
Price of stove or
cooker (PSC)
-0.892
(2.742)
0.000 2.44 -0.003
(0.200)
0.000 0.997
Head of household
level of education
(HHE)
0.418
(0.207)
0.090 1.519 0.232
(0.018)
0.214 1.261
House wife level
of education
(HWE )
0.689
(0.295)
0.019 1.99 0.613
(0.230)
0.022 1.845
Households‘ increased consumption of each fuel type as their total expenditure increased. The price of stove or cooker has a
negative estimated coefficient for both kerosene and cooking gas, implying that an increase in the price of stove or cooker
will decrease the tendency of using fuelwood alternatives and increase the probability of using fuelwood.
Household size has a negative estimated coefficient for both kerosene and cooking gas. This supports the theoretical
expectation that larger households‘ will prefer to use fuelwood since it is comparatively cheaper to use fuelwood to cook for
many people as it has a lower consumption rate per unit of time compared to kerosene and cooking gas (Punder and Fraser,
2006). Moreover, it is believed that larger household sizes may mean larger labour input, which is needed in fuelwood
collection. Larger households‘ are more likely to have extra labour (for example children‘s labour) that can be used to freely
collect fuelwood from public fields and thus may lower the price of fuelwood relative to alternatives which cannot be
obtained freely.
10
The level of education concurs with the hypothesized theoretical expectation of a positive effect on the choice of kerosene
and cooking gas due to an increase in the level of education of respondents. This may be associated with the assumption that
higher levels of education are associated with better income.
A statistical significance test was conducted using the standard error test and also by considering the respective probabilities
of the variables. Standard error test entails a situation where if the standard error of a parameter is smaller than half the
numerical value of the parameter estimate, then the variable is statistically significant (Koutsoyiannis, 2004). From the
regression result the variables EKR, ECG, INC, HHE and HWE are statistically significant and therefore explain the
probability of choosing kerosene and cooking gas as compared to fuelwood. The variables SZH, and PSC, were found not
significant in explaining the probability of choosing kerosene and cooking gas as compared to fuelwood.
The count r squared is about 45%, which implies that about 45% of the determinants of energy choice are being accounted
for by the explanatory variables. A goodness of fit test was conducted using the Hosmer-Lemeshow and Andrew statistics.
The results are H-L (12.47) and Andrew (72.39) shows that the regression exhibit a very good fit (though coefficient of
determination is of secondary importance in probability modelling, since they are not meant for forecast).
5.0 CONCLUSION AND POLICY IMPLICATIONS
The aim of this study is to investigate the impact of wealth distribution on energy consumption, and to analyse the
determinants of households‘‘ energy choice for cooking in Gombe State, Nigeria. The Multinomial logit model was
employed to identify the determinants of energy for cooking as well as sociological and economical variables influencing
major energy sources in the study area.
Empirical investigation revealed that apart from household income, household cooking energy choices also depends on
sociological and other economical factors such as household size, levels of education and the prices of appliances for a
particular energy type. The study shows that fuelwood is by far the fuel of choice for a majority of households‘ in the study
area. The study further revealed that as household income increases, households‘ switch to cleaner fuels; from fuelwood to
kerosene as implied by the energy ladder hypothesis. The dependence on fuelwood in this region has far-reaching
implications on the environment: deforestation, soil erosion and declining agricultural productivity and lose in the natural
habitat.
In the light of the above, the study suggest that apart from improving household income, policy design also need to focus on
other factors in addressing the challenges of energy exploitation. One solution to the environmental consequences of
unsustainable wood exploitation requires that modern cooking fuels be made more accessible and affordable and fuelwood
and charcoal use be made sustainable.
Moreover, improvement in income and education enhance the likelihood of the household to increase the consumption of
other fuels. This will help reduce consumption of wood, implying a reduction in the pressure of wood resources and
contributing towards mitigating deforestation.
Furthermore, measures should be taken by stakeholders in the energy sector to develop and promote renewable, clean
technologies to lessen the burden of economic activities on the ecosystem, reduce pollution and meet the demand of
households‘. Such measures should promote the use of energy carriers other than biomass as well as the use of biomass in
modern ways.
Finally, since fuelwood is the fuel of choice by a majority of the rural populace, a permanent programme of reforestation that
provides for the planting of wood species that are ecological suitable, socio-culturally compatible and economically feasible
11
and products harvested under controlled and best practices should be adopted by the government as an avenue to address
energy demand issues and other-interrelated concerns like food production, soil erosion and desertification. These can be
achieved through establishing micro-credit facilities for entrepreneurs, especially women groups, for the establishment and
operation of commercial fuelwood lots and the production of renewable energy devices and systems.
12
REFERENCES
Alam M., Sathaye J., and Barnes D. (1998): Urban Household Energy use in India: Efficiency and Policy Implications.
Energy Policy, 26:885–891
Barnes, D. and Willem F. (1999): ―Biomass Energy and the Poor in the Developing Countries‖. Journal of International
Affairs, fall, 53(1): 237-259.
Barnes, D. F., Kerry K, and William H. (2002): ―The Urban Energy Transition – Energy, Poverty, and the Environment: Two
Decades of Research‖. Forthcoming book volume.
Black, J (2002): Dictionary of Economics. Oxford University press New York.
Campbell, B.M., Vermeulen, S.J., Mangono, J.J., Mabugu, R.,(2003): The Energy Transition in action: Urban Domestic Fuel
Choice in a changing Zimbabwe. Energy Policy 31, 553–562.
Davis, M. (1998): ―Rural Household Energy Consumption: The Effects of Access to Electricity—Evidence from South
Africa,‖ Energy Policy 26(3): 207–217
Energy Commission of Nigeria (2003): National Energy Policy, Federal Republic of Nigeria, Abuja.
Heltberg, R. (2005): ―Factors Determining Household Fuel Choice in Guatemala,‖ Environment and Development
Economics 10: 337–61
Hosier, R.H., and J. Dowd. (1987): ―Household Fuel Choice in Zimbabwe: An Empirical Test of the Energy Ladder
Hypothesis,‖ Resources and Energy 9: 347–61
Horsier, R. and Kipondya, W., (1993): Urban Household Energy use in Tanzania: Prices,
Substitution and Poverty. Energy Policy, 453-473.
International Energy Agency (2006) World Energy outlook, Paris OECD
Inter-Ministerial Committee on Combating Deforestation and Desertification (ICCDD) Report of August 2000.
Judge G, Hill RC, Griffiths WE, Lǘtkepohl H and Lee TC (1985): Theory and Practice of Econometrics. John Wiley and
Sons, New York.
Koutsoyiannis, A (2004): Theory of Econometrics Second Edition, Palgrave Publishers New York.
Leach, G. 1992. ―The energy transition‖ Energy Policy, February: 116-123.
Madubansi, M.and C.M. Shackleton (2005): "Changes in Fuelwood use and Selection Following Electrification in the
Bushbuckridge lowveld, South Africa." Journal of Environmental Management, 83:416–426.
McFadden D (1974): Conditional Logit Analysis of Qualitative Choice Behaviour. In: Zarembka P (ed), Economic Theory
and Mathematical Economics. Academic Press, New York.
Narain, U., S. Gupta and K. van‘t Veld (2008):‗Poverty and Resource Dependence in Rural India‘, Ecological Economics 66
(1): 161-176.
Ouedraogo B. (2006): Household Energy Preferences for cooking in Urban Ouagadougou, Burkina Faso, Energy Policy.
Pundo M. O and Fraser GCG (2006): Multinomial logit Analysis of Household Cooking Fuel Choice in Rural Kenya: The
case of Kisumu district, Agrekon, Vol 45, No 1 (March 2006)
13
Reddy, B. S., (1995): A Multi-logit Model for Fuels Shifts in the Domestic Sector. Energy,
20(9), 929-936.
Sambo S. A. (2009): Strategic Developments in Renewable Energy in Nigeria International Association for Energy
Economics, 3rd quarter.
Williams, C. E. (1998): ―Reaching the African Female Farmers with Innovative Extension Approaches: Success and
Challenges for the Future ―Paper Presented To The International Workshop on Women Agricultural Intensification
and Household Food Security At University of Cape Coast, Ghana, 25th-28
th June.
World Bank (a) (2003): Household Energy Use in Developing Countries, a Multicountry Study, UNDP/World Bank Energy
Sector Management Assistance Programme (ESMAP)