the contribution of ofw remittances to income …...ii. review of related literature much...

The Contribution of OFW Remittances to Income Inequality: A Decomposition Analysis

In fulfillment of the requirement of the course Economics 199

Submitted by

Nico M. Ravanilla Eduard Joseph P. Robleza

Submitted to the School of Economics

March 2003

ii

WAIVER

We, Nico M. Ravanilla and Eduard Joseph P. Robleza, allow any part of this thesis to be used or

reproduced in any form and for any purpose whatsoever, provided that the proper citation is given to the

authors.

Nico M. Ravanilla Eduard Joseph P. Robleza

iii

To the overseas Filipino worker.

iv

Acknowledgment For the greater glory of God. Ad majorem dei gloriam.

The authors would also like to thank the following:

- our families, for obvious reasons;

- Dr. Gwendolyn R. Tecson, for her guidance;

- Ms. Sharon Faye A. Piza, for her tolerance (we still owe you that free lunch);

- the kind people at the UPSE Library and Computer Laboratory;

- the people at the ETC tambayan for their love (Robert says hi!);

- Ms. Anne Imelda Pascua, for desperately wanting to be acknowledged;

- and to the people whom we may have forgotten, insert your name here:

“We also thank ____________________.”

Thanks to the people who are reading this. Thanks to those who aren’t.

v

Table of Contents

ABSTRACT ..................................................................................................................................... ix

I. INTRODUCTION...........................................................................................................................1

WORKING HYPOTHESIS 2 SIGNIFICANCE AND OBJECTIVES OF THE STUDY 3 OUTLINE OF THE STUDY 4

II. REVIEW OF RELATED LITERATURE........................................................................................5

FOREIGN WORKS 5 PHILIPPINE LITERATURE 8

III. THEORETICAL FRAMEWORK................................................................................................11

MIGRATION AND MOTIVATIONS 11 MIGRATION AND INCOME DISTRIBUTION 12

IV. METHODOLOGY .....................................................................................................................15

DECOMPOSITION ANALYSIS 15 THE DATA 17

V. PRESENTATION OF RESULTS AND ANALYSIS .....................................................................19

THE MIGRATION CHAIN 19 THE COMPOSITION OF INCOME INEQUALITY 23 THE CONTRIBUTION OF REMITTANCES TO TOTAL INCOME INEQUALITY 31

VI. CONCLUSIONS AND RECOMMENDATIONS.........................................................................40

POLICY RECOMMENDATIONS 40 RECOMMENDATIONS FOR FURTHER STUDIES 42

vi

SELECTED REFERENCES.............................................................................................................44

APPENDICES .................................................................................................................................47

APPENDIX A: CHOOSING A MEASURE OF INEQUALITY: THE FIVE PROPERTIES OF DECOMPOSABILITY 47

APPENDIX B: COMPARING THE DECOMPOSITION RESULTS OF USING

THE GINI COEFFICIENT AND THE GENERALIZED ENTROPY (GE) CLASS OF INEQUALITY MEASURES 49

APPENDIX C: BACKGROUND INFORMATION ON PHILIPPINE MIGRATION 51 APPENDIX D: TABLES A1 – A6: COMPOSITION OF HOUSEHOLD INCOME

INEQUALITY, BY YEAR 55

vii

List of Tables

Table Number

1 2

3A

3B 4 5 6 7 8 9

10

Title

Deployed Overseas Filipino Workers and Total Remittances from Overseas Filipinos, 1984 – 2000 Number of Households Receiving Income from Abroad Average Family Income (in Pesos), Urban – Rural Average Family Expenditure (in Pesos), Urban – Rural Shares of Income Sources in Total Income Gini Coefficients Percent of Households Receiving Income from Abroad Gini Correlation Ratios Contributions of Income Sources to the Gini (in Percent) Relative Concentration Coefficients Distribution of Overseas Filipino Workers, By Major Occupation, 1995 - 2000 (in Percent)

Page

20

20

21

21

24

27

28

29

33

36

38

viii

List of Figures

Figure Number

1 2 3 4 5

Title

Share of Remittances in Total Income Gini Coefficient of Remittances Gini Correlation Ratio of Remittances Contribution of Remittances to Total Income Inequality (in Percent) Relative Concentration Coefficients of Remittances

Page

24

27

31

33

37

ix

Abstract The paper aimed to determine whether remittances from overseas Filipino workers

improved or worsened income inequality. Using decomposition equations, the paper

divided total inequality into its four components, namely wages, entrepreneurial incomes,

other income, and remittances from migrants. The decomposition exercise revealed that

the contribution of remittances to overall income inequality depended on their share in

total income, their distribution among the population, and their correlation with total

income. Remittances were found to accrue mostly to higher- income classes, but they

were found to be gradually becoming less inequality- increasing over time. Therefore,

policies that would aim to reduce income inequality should consider making migration-

facilitating factors more accessible to those in the lower ends of the distribution, because

remittances would only tend to contribute less to income inequality if the lower- income

brackets were also able to migrate. Finally, further studies should look into the

implications of the changes in inequality, especially with regards to welfare.

1

I. Introduction

The Philippines has historically been one of Asia’s largest labor exporters. In the 1960s,

Filipino doctors and nurses took the opportunity to work abroad when the United States

liberalized its immigration law. This was followed by the migration of seamen, engineers and

construction workers to the Middle East during the oil shock in the 1970s and 1980s. The trend

has continued in the nineties and up to the present, as domestic workers, nurses and teachers

respond to increasing job openings all over the world.

One strong impetus that compels Filipino workers to migrate is the Philippines’ persistent

problem with unemployment and low economic growth rates. In fact, the Philippines’ GNP per

capita has grown at an average annual rate of just a little over 1% from 1965 to 1990, a rate that

is below those enjoyed by the country’s Southeast Asian neighbors (World Bank 1995). Faced

with little or no opportunities at home, Filipinos migrate to where the jobs are; migration, after

all, is a decision in response to opportunities available elsewhere.

The decision to migrate is also based on the expected benefits and costs associated with

taking a job abroad. One of such benefits that have strongly attracted Filipinos to overseas

employment is the higher wage being offered in foreign countries. With higher wages, overseas

Filipino workers (OFWs) obviously earn higher incomes that can be saved for future

consumption. Income from overseas employment also affords households the opportunities to

invest in other profitable pursuits such as in small family businesses or in their children’s

education—the benefits of which can be reaped later on. Indeed, this fact leads many to claim

that overseas employment lessens income inequality, because households with income from

abroad are believed to be able to afford a higher standard of living than otherwise possible.

2

The Philippine government has always acknowledged this positive effect of overseas

employment on the welfare of its citizens—and with good reason. OFW remittances have

contributed much to the country’s GNP and foreign exchange earnings. In fact, remittances

comprised, on the average, about 2% of GNP in the eighties, and that contribution has more than

doubled to 4.8% in the nineties (Tan 2000). These remittances are used to service the country’s

foreign debt, and the export of labor is thus seen as a means to achieve or sustain economic

development. In addition, the Philippine government has established secure safety nets—the

Philippine Overseas Employment Administration (POEA) and the Overseas Workers’ Welfare

Administration (OWWA)—to protect the interests of the millions of migrant Filipinos. These

agencies have also lowered the transaction costs associated with overseas employment. In effect,

not only has the government given everyone almost equal opportunities to migrate, but it has

made migration an even more attractive and worthwhile option for its workforce as well.

Working Hypothesis

Despite these obvious and substantial economic benefits, however, there is still reason to

believe that the effects of overseas employment and remittances may not always be positive.

Because migration entails both monetary and psychological costs, some workers in certain

classes are already precluded from migrating. Even if costs are made lower through government

assistance, these workers may still not be able to find jobs abroad, especially if they do not meet

the required qualifications of age, education and experience, because overseas employment is

basically demand-driven.

In other words, the chances for migration may not always be equally distributed among

the population. Studies indicate that Filipino migrants, whether temporary or permanent, are

3

most likely to come from only certain groups of people. More specifically, households with

remitting migrants predominantly come from urban areas and wealthier families, implying that

only these households can more readily assume the costs and risks of migration (Rodriguez and

Horton 1995; Rodriguez 1998). In addition, migrants are also seen to be more educated than non-

migrants. In 1990, OFWs had an average of 10.6 schooling years, while the entire labor force

had only 7.9 years (Tan 2000). This implies that only those with the required skills, training or

experience—which can be too costly for others to acquire—can more easily take jobs abroad and

hence benefit from higher incomes.

There is therefore reason to suspect that the benefits associated with migration,

particularly earnings and remittances, may not be distributed evenly as well. Remittances may

not always improve the distribution of income, and contrary to what many believe, they may

even contribute to income inequality within the country.

Significance and Objectives of the Study

The question of whether remittances from abroad serve to alleviate or worsen income

inequality is of great interest, particularly because of the high level of inequality in the country.

In addition, the country has faced massive outflows of workers in recent years. Considering that

these workers send in an amount almost equal to 5% of GNP in the 1990s, the overall effect of

these remittances on household income distribution appears to be substantial and thus worth

looking into.

There is also a dearth of empirical Philippine studies about the topic. Moreover, the few

studies that exist focused only on the short-term effects of international remittances on income

distribution, covering only one specific year, or up to three years at most. And because their

4

conclusions naturally varied, the relationship between remittances and income distribution

seemed unclear. At best, there is still no general consensus about the effect of migrants’

remittances on income inequality.

In this regard, this paper recognizes the importance of analyzing the said relationship

over a much longer period of time. Migration is a dynamic phenomenon; its flows and structure

change, as other variables, such as the demand for and the skill levels of migrant workers,

fluctuate over time. Hence, by looking at a longer time span, the paper aims to arrive at an

observable trend that allows for a firmer, more realistic statement about the relationship between

remittances and income distribution. This, in turn, will be of help in making policy

recommendations. Furthermore, the paper hopes to contribute to the existing literature on

remittances and income inequality in the Philippines.

Outline of the Study

The paper is divided into six sections. Section II reviews the related literature on the

topic. Sections III and IV outline the theoretical framework and the methodology used,

respectively. Section V presents and analyzes the results. Finally, Section VI summarizes the

main conclusions and offers policy recommendations.

5

II. Review of Related Literature

Much speculation has been made about the sources of income inequality in the

developing world ever since Kuznets (1955) wrote about economic development and income

inequality. Several empirical studies have decomposed the sources of income inequality by

economic sector (e.g., urban vs. rural), family characteristics (e.g., education and occupation of

workers), as well as by income source (e.g., income from labor vs. capital vs. land) (Adams and

Alderman 1992).

Foreign Works

It was not until the late eighties and early nineties when empirical studies attempted to

isolate the effects of labor migration on income inequality. Using labor migration survey data

from the 1980s, the relationship between international remittances and income inequality was

examined for rural Mexico, Egypt, Pakistan, and for some other Asian nations.

Stark, Taylor and Yitzhaki (1986) studied the effect of interna tional remittances on the

income distribution of two rural Mexican villages with different migration histories. The results

were contrasting. Remittances were found to reduce inequality in the village that had a longer

history of migration to the United States and hence a more ready access to United States labor

markets. On the other hand, remittances increased inequality in the village that had only a few

households that had experienced migration to the United States. Meanwhile, the more recent

study by Stark and Taylor (1991) further argued that the propensity of households to participate

in international migration from Mexico to the United States was directly related to the

households’ initial relative deprivation. This implied that relatively deprived households were

6

more likely to engage in international migration than were households more favorably situated in

their village’s income distribution.

A series of investigative studies organized by the International Labour Organization in

the mid-eighties examined the economic impact of overseas migration on labor-sending Asian

countries (Amjad 1989). As far as testing the relationship between income inequality and

international remittances was concerned, however, these studies were mainly descriptive, since

they did not use regression equations or decomposition analyses to test the said relationship. The

study on South Korea, for instance, merely reported of a highly positive correlation between

migration and the income disparity of urban households: The Gini ratio in the 1970s when

migration prevailed was higher than in the 1960s, implying a worsening of income distribution

(Hyun 1989). However, mere correlation does not necessarily mean that migration cannot

improve income distribution at all or will automatically aggravate it, since overseas employment

does allow some of those in the lower- income brackets to earn more than otherwise possible.

While also mainly descriptive, the studies on Bangladesh (Mahmud 1989), Thailand

(Tingsabadh 1989), and Sri Lanka (Rodrigo and Jayatissa 1989) suggested that the income-

distributional consequences of migration and remittances depended on the relative income

classes from which migrants’ families originally came. In Bangladesh and in Thailand, migrants

were found to have come from a more affluent background, since those from the poorer classes

could not afford the initial costs of migration. This implies that migration benefits the higher-

income classes and only serves to widen the income gap between them and the lower- income

classes. In contrast, most of Sri Lankan migrants came from the lower-income and low-skilled

groups, and this was seen to have moderated income disparities in Sri Lanka.

7

Meanwhile, Adams (1991) conducted a different study, this time focusing on the effects

of international remittances on poverty, inequality and development in rural Egypt. His study

used predicted income equations to estimate the changes that would occur between two

contrasting situations: excluding remittances, where the remittances of the correspondents with a

still-abroad migrant were excluded, and including remittances, where the remittances of these

still-abroad migrants were included. The study found that while international remittances helped

alleviate poverty, they nevertheless had a negative, worsening effect on income distribution.

Furthermore, the studies by Adams and Alderman (1992) and by Adams and He (1995)

on rural Pakistan provided a good foundation in the decomposition analysis of the sources of

income inequality. The decomposition analyses of these studies determined exactly what sources

of income contributed to the total income inequality, as well as how much these income sources

actually contributed to the observed inequality. It should be stressed that the main objective of

these two studies was only to analyze the determinants of rural poverty in Pakistan and not to

focus on and assess the impact of international remittances on income inequality, as evident in

both the data set and the methodology used. The study by Adams and Alderman, for instance,

used a 1986–1989 survey which, as the authors noted, “was not designed to be representative of

the rural population as a whole in Pakistan” (1992). Furthermore, in the decomposition of the

Gini coefficient, all transfers—international remittances, domestic or internal remittances,

pensions, and payments of the government to the poor—were lumped together as one source of

income. The effect of international remittances on income inequality was thus not isolated.

8

Philippine Literature

Just as empirical studies abroad remain in their early stages, Philippine studies about

international migration and its effects on income distribution have been few. In fact, some of the

more substantial studies did not come out until the early nineties. For instance, the work by

Untalan in 1980 looked at the composition and supply of workers who go to the Middle East and

analyzed the costs and benefits of labor exports. However, it barely dealt with the direct effects

of labor migration on domestic income inequality. Also, Macaraig and Sarino (1980) wrote an

undergraduate thesis on the remittances of contract workers, although the study focused more on

the direct effects of labor exports on welfare than it did on income distribution. By examining the

decile movements in income shares of overseas contract workers, the study showed that the

remittances of contract workers generally tended to decrease the welfare of families when the

30% mandatory remittance (under the 30% Remittance Law) served as a floor.

More than a decade later, Tan (1991) studied overseas employment, saving rate and

income distribution. Using the 1988 Family Income and Expenditures Survey, or FIES, she

observed that overseas employment had an equalizing effect on family income. The

improvement in income distribution was shown by the apparently much better distribution of

income by size for families with income from abroad than for families without such income.

More precisely, the statistical analysis showed that much fewer families with income from

abroad belonged to the category ‘poor’ (taking the poverty income to be approximately

PhP3000): 15.1% vs. 39.4% in urban areas, and 41.1% vs. 73.4% in rural areas.

However, the results of both Macaraig and Sarino’s and Tan’s studies could no t be

considered conclusive, because these studies did not specify whether the families sampled were

already poor even before the income from abroad was added in. In addition, these two studies

9

relied mainly on decile movements and size distributions rather than on some direct measure of

inequality, such as the Gini coefficient, which could be source decomposable; i.e., the total

inequality could be broken down into a weighted sum of inequality by various income sources.

In effect, these studies might not have considered that households also received incomes from

other sources, such as domestic wages or entrepreneurial activities. Thus, these two papers could

not conclusively determine or claim exactly how much of the overall inequality was actually due

to remittances, if at all.

Rodriguez and Horton’s 1995 working paper arrived at a conclusion very much at odds

with those of Macaraig and Sarino’s and Tan’s. Using the 1991 FIES, Rodriguez and Horton

compared households with migrants who remitted and households that did not have migrants or

remittances. The paper found out that the former group was predominantly urban and had a

higher total income and expenditure than the latter group. The paper also noted that the heads of

the former group were slightly older and more likely to be female, and that although these

household heads were less likely to be employed, they nonetheless had more education. All these

suggested that migration might exacerbate income inequality, since migrants came from urban,

wealthier, better-educated households.

Furthermore, by decomposing the Gini coefficient, Rodriguez, in his 1998 article, only

furthered the conclusion he made with Horton earlier that remittances actually worsened income

inequality in the Philippines. In the same paper, Rodriguez also pointed out some methodological

concerns with earlier Philippine studies. When income shares of households were ranked by their

pre-remittance income (defined as total income minus remittances from migrants), the procedure

implicitly assumed that if the migrants had stayed at home, their contribution to household

income would have been zero. This method of calculation would thus underestimate inequality

10

and make it appear that when the distribution was kept constant and remittances were added,

remittances would dramatically lift households from poverty.

The most recent study that was available was a paper by Balisacan and Piza (2003).

Unlike Rodriguez’s article, Balisacan and Piza’s paper decomposed inequality using the

Generalized Entropy class of inequality measures. The paper provided an excellent general

discussion about the nature and causes of inequality in the Philippines, although it did not focus

specifically on the impact of remittances upon income inequality. The results of the

decomposition, however, suggested that there was “weak evidence to the common claim that

overseas remittances in recent years (have) accentuated income inequality.” More importantly,

the paper could serve as a basis for comparison in checking whether the results would be

consistent even if different measures of inequality were used in the decomposition.

This paper takes all these into consideration. Before proceeding with a detailed analysis,

though, this paper first sets the context within which such analysis is to be made. The next

section therefore discusses the theoretical framework used.

11

III. Theoretical Framework

Migration is essentially a factor movement. Labor responds to employment opportunities

and wage differentials, which are determined by developments both in the sending economy and

in the receiving markets.

Migration and Motivations

Migrant labor responds to job opportunities abroad, but the opportunities tend to be

limited because of various economic, political and even cultural barriers. For example, workers

and labor unions in the host or receiving economies often resist the inflow of migrant labor due

to fears of unemployment among their ranks. Labor exporters like the Philippines, however,

respond to these openings by having a flexible or mobile labor force, since these openings tend

to be limited to a select set of occupations with strict qualification requirements. In this sense,

migration is seen as a human capital problem of simply maximizing the returns to the education

or training that the potential migrant has undertaken (Tan and Canlas 1989).

Relative wage rates also determine the direction of labor flows, and indeed, this monetary

benefit is often assumed to be the main motivation that encourages migration. The Philippine

case illustrates this concept. The recession after the balance-of-payments crisis in the 1980s

resulted in massive unemployment. Real wages then sank to levels that were lower than real

wages elsewhere, and as expected, many took jobs abroad (Tan 1991).

In addition, when people compare incomes within a reference group—such as their local

community, for example—the comparisons create psychic costs or benefits, or feelings that Stark

and Bloom (1985) call “relative deprivation or relative satisfaction.” A person may then decide

to migrate from one location to another either to change his relative position within his reference

12

group or to move to another reference group altogether. Generally, a person who feels more

relatively deprived is expected to have a stronger incentive to migrate than a person who feels

less so. As particular individuals reap the benefits of migration, the relative deprivation

perceived by non-migrants may change, thereby inducing them to migrate as well. Migration can

thus be viewed alternatively as “a diffusion process [wherein] the level of migration at any point

in time is likely to be positively related to past migration” (Stark, Taylor and Yitzhaki 1986).

Migration and Income Distribution

At the beginning of a country’s migration history, when few households have just started

establishing contacts at the receiving economy, the distribution of remittances—and possibly of

income as well—is “necessarily unequal” (Stark, Taylor and Yitzhaki 1986), which can be

attributed to a number of observable facts. One is that, during this time, information about the

working conditions abroad is scarce and costly; because of the uncertainty that this creates,

migration, in a sense, becomes a high-return yet high-risk investment. Moreover, the initial costs

of migration—the fees of recruiting agents, the costs of processing passports, visas and airline

fares—preclude the poorer sectors from migrating. Thus, the first households that will invest in a

migration decision are likely to come from the upper rungs of the income distribution, since they

are more financially capable to assume its costs and risks.

In addition, given that migration is demand-driven, any inequality in the distribution of

income puts those in the lower- income classes at an even greater disadvantage. Working abroad

may require specific skills, a certain level of education or experience, and even English

proficiency, depending on which foreign labor market or occupation opens its doors to migrant

labor. Thus, unless the low-income classes can afford to acquire the skills and levels of education

13

demanded by overseas work, then migration may remain to be a valid option only for those in the

high- income classes.

The impact of remittances in the overall household income distribution at this initial stage

therefore depends firstly upon the magnitude of remittances in relation to income from other

sources. If remittance-receiving households rely significantly on remittances from abroad to

augment their income, then the distribution of remittances is expected to greatly contribute to the

distribution of income. If, on the other hand, remittances represent only a small part of household

incomes, then remittances will have a minimal effect on income distribution.

In addition, the impact of remittances on income inequality also depends on which

income classes migrants originally come from. If migrants primarily come from the relatively

well-off, then their remittances may worsen income inequality. However, if migrants also come

from the lower classes, then remittances may contribute to greater income equality.

Over time, successful migrants provide valuable information to those left behind, and this

may raise the latter’s propensity to migrate. These would-be migrants, now feeling more

relatively deprived, will also want to migrate in order to shift the distribution of returns to

migration in their favor. Moreover, the early migrants make migration a much more worthwhile

undertaking because the information they bring reduces the risks and uncertainty associated with

it. In some instances, too, early migrants provide direct assistance to new migrants by financing

the costs of migration, especially for family and friends. In other words, migration flows tend to

create a “migration chain” that generates opportunities for later migration (Tan 2000).

Therefore, the effect of remittances on income inequality over time depends on how

information and factors that facilitate migration become diffused through the population. If

access to these factors easily spread across households, then migration—and hence the receipt of

14

remittances—by those at the lower ends of the income distribution is likely to occur, possibly

reducing the initially unfavorable effect of remittances on income inequality (Stark, Taylor and

Yitzhaki 1986).

Furthermore, the types of labor or skills demanded abroad may change over time,

depending on the developments in the labor markets of the receiving economies. These shifts in

demand for migrant labor may in turn influence the income distribution in the sending countries.

For instance, if the demand for labor abroad shifts towards those occupations that require lower

skill and education levels, then migration opportunities are opened up especially for those in the

lower income brackets, who may be constrained from acquiring higher skill levels. In this case,

migrants’ remittances may contribute to greater income equality.

More importantly, if migration is indeed a diffusion process as suggested above, then its

impact must now be seen as stage-specific (Stark and Bloom 1985). The effect of remittances on

income distribution may vary depending on which point in a country’s migration history is being

considered, because the level of diffusion of migration-facilitating factors as well as the skills

demanded by overseas employment vary at each point. This should partly explain the divergence

of views about the effect of remittances on income inequality, because the empirical studies did

not temper their conclusions with the fact that the observations were made at distinct and specific

stages of the migration process. Moreover, because the impact of international remittances and

income inequality is stage-specific, looking at longer time spans logically suggests that this

impact may also change over time. More explicitly, while remittances do seem to worsen income

inequality at the start of the migration chain, they may be expected to gradually reduce

inequality, as more people from the lower ranks of the distribution are enticed and are able to

migrate.

15

IV. Methodology

Taking this stage-specificity into consideration therefore requires, not only looking at the

context of a specified period of time, but also dissecting the income stream of households over

this period. Households receive income from various sources, such as wages and entrepreneurial

activities. It is in this regard that migration is a source of remittance income, for the decision to

migrate only “reflects a desire to diversify income sources” (World Bank 1995). For example, a

rural household may choose to send one of its members to work in the city or abroad while the

rest stay and work in the family farm. In any case, international remittances are just one source

of household incomes for any period in time, and an income decomposition ana lysis will isolate

their impact on total income inequality. To account for migration as a stage-specific diffusion

process, the decomposition will also be done repeatedly for several periods.

Decomposition Analysis

While any measure can easily allocate income inequality among the components of total

income, Adams and Alderman (1992) have shown that it is appropriate to use the Gini

coefficient, since this measure satisfies the five basic properties or axioms of decomposability:

(1) Pigou-Dalton Transfer Principle, (2) Income Scale Independence, (3) the Principle of

Population, (4) Anonymity, and (5) Decomposability. (A discussion of these five properties is

found in Appendix A, while the implications of using different inequality measures are

summarized in Appendix B.)

Following Stark, Taylor and Yitzhaki (1991) and, more recently, Adams and He (1995),

the source decomposition based on the Gini coefficient of total income, G, can be written as

G = 2[cov(y,r)]/? , (1)

16

where y is the series of total incomes, and r is the cumulative distribution of total incomes. The

Gini coefficient of the ith source of income, Gi, can therefore be expressed as

Gi = 2[cov(yi,ri)]/? i, (2)

where yi and ri refer to the series of incomes from the ith source and their corresponding

cumulative distributions, respectively. Since total income is the sum of all source incomes, the

covariance between total income and its cumulative distribution can be expressed as the sum of

covariances between each source income and cumulative distribution of total income. Equations

(1) and (2) can then be used to express the total income Gini as a function of the source Ginis:

G = ? SiRiGi, (3)

where Si is the share of each source of income to total income, or

Si = ? i/? ? ? ? ? ? ? ???

and R is the Gini correlation ratio, expressed as

Ri = cov(yi,r)/cov(yi,ri) (5) = covariance between source income amount and cumulative distribution of total income covariance between source income amount and cumulative distribution of source income

The Gini correlation ratio, R, ranges between +1 and –1. An income source is an

increasing (decreasing) function of total income as R approaches +1 (–1). When the income

source is a constant, R equals zero, implying that that income source’s share of the Gini is zero.

The decomposition corresponding to the Gini coefficient can therefore be expressed by

the following terms:

? wigi = 1; wi = ? i/? ; gi = Ri(Gi/G) (6)

where wigi is the factor inequality weight of the ith source in overall inequality, and gi is the

relative concentration coefficient of the ith source in overall inequality.

17

Assuming that additional increments of an income source are distributed in the same

manner as the original units, an income source can be defined as inequality- increasing or

inequality-decreasing on the basis of whether additional shares of income from that source lead

to an increase or decrease in overall income inequality. Based on Equation (6), it follows that the

ith income source is inequality- increasing or inequality-decreasing according to whether gi or the

relative concentration coefficient is greater than or less than unity.

The Data

Data for constructing the decomposed Gini indices are taken from the various Family

Income and Expenditure Survey (FIES) rounds from 1985 to 2000, with each round conducted

every three years. The FIES is an integrated survey of households that seeks to gather data on

family income and expenditures, determine the degree of inequality among families, and provide

benchmark information for the estimation of the consumer price index. While the FIES for 1961,

1965, and 1971 are also available, they are excluded from the analysis since their unit-record

data are not available for the decomposition.

Definition of Terms

The empirical derivation of the decomposed Gini indices is accomplished using data from

the Philippines, as categorized into ‘all Philippines’, urban, and rural. The FIES of 1997 defines

urban areas as follows:

1. In their entirety, all cities and municipalities having a population density of at

least 1,000 persons per square kilometer.

18

2. Poblaciones or central districts of municipalities and cities which have a

population density of at least 500 persons per square kilometer.

3. Barangays having at least 1,000 inhabitants where the occupation of inhabitants

is predominantly non-farming or non-fishing.

All areas not falling under any of the above classifications are considered rural.

Four components of household income, and therefore of income inequality, are

considered: wages, entrepreneurial income, other income, and remittances from migrants or

income from abroad. Wages include all forms of compensation received by family members who

are either regular or occasional workers in agricultural and nonagricultural industries.

Entrepreneurial income includes earnings from farming, trading, manufacturing and other

entrepreneurial activities engaged in by any member of the family as operator or as self-

employed. Other income includes the family’s net share of crops produced, rental value of

owner-occupied dwelling units and other family sustenance activities. Finally, Income from

abroad includes cash receipts from family members working abroad as contract workers, as well

as gifts and other forms of assistance from other migrants, particularly family members with

permanent residence abroad (or emigrants).

Data Limitations

Because of the tendency of some households to underreport their incomes, especially

richer families, it is possible that income data from the FIES are understated (Rodriguez 1998).

As a consequence, conclusions about the contribution of international remittances to total income

inequality may reflect this downward bias; the magnitude of income inequality may be

underestimated.

19

V. Presentation of Results and Analysis

As evident in Equation (3), the decomposition exercise outlined above can illustrate how

(a) the magnitude of remittances to total income, (b) the unequal distribution of remittances, and

(c) the correlation of remittances with total income all contribute to the observed inequality. The

first part of this section sets the context of the analysis against the theoretical framework by

describing the changes and trends in the decomposition variables. Having set the context, the

paper then proceeds to discussing the contribution of remittances to overall income inequality.

The Migration Chain

The Philippines has had a fairly long history of migration, which dates back to American

colonial times (Sto. Tomas 1984). With increasing demand for migrant labor and with some

encouragement from the Philippine government, migration has gradually become a worthwhile

option for Filipino workers over the years. (Appendix C discusses the country’s history of

migration and policies on remittances in more detail.) This supports the theory that migration is a

diffusion process wherein the benefits of migration induce non-migrants to migrate in the future.

Because of this diffusion process, one may expect to see a gradual rise in the number of

deployed workers over time—or a migration chain. This phenomenon can be observed from

Table 1, which shows that the number of deployed OFWs grows as time passes. Assuming that at

least some of these new migrants remit some fraction of their earnings back home like permanent

migrants and contract workers do, one may infer that this migration chain logically—although

not always necessarily—results in more households receiving income from abroad, as Table 2

indicates.

20

Table 1. Deployed Overseas Filipino Workers and Total Remittances from Overseas Filipinos, 1984 – 2000.

Year Deployed OFWs* Total Remittances (US$ Million)

1984 350982 602.9 1985 372784 687.2 1986 378214 680.4 1987 449271 791.9 1988 471030 856.8 1989 458626 973.0 1990 446095 1181.1 1991 615019 1500.3 1992 686457 1769.5 1993 696630 2229.6 1994 719602 3008.1 1995 654022 3868.3 1996 660122 4306.6 1997 747696 5741.8 1998 831643 4926.0 1999 837020 6794.6 2000 841628 6050.4 Source: Department of Labor and Employment. Yearbook of Labor Statistics, 2000.

* Land-based and Sea-based OFWs.

Table 2. Number of Households Receiving Income from Abroad.

1985 1988 1991 1994 1997 2000 Urban Philippines 343909 332890 493600 557121 523704 681266 Rural Philippines 224615 273805 281400 304404 357558 425239

All Philippines 568524 606695 775000 861525 881262 1106505

Urban/Rural 1.5 1.2 1.8 1.8 1.5 1.6 Source: National Statistics Office. FIES, various years.

Thus far, the statistics in Tables 1 and 2 have shown that past migration flows have

created opportunities for later migration. The growing numbers of deployed OFWs may imply

that the risks associated with migration have diminished, possibly because the information about

the working conditions and successes experienced by earlier migrants has accumulated

sufficiently to become public information (Tan 2000). However, the growing numbers of

deployed OFWs cannot yet be automatically attributed to an improvement in the labor mobility

of those in the lower ends of the income distribution in the sense that they have gained more

21

access to education or skill training. What can be implied from the statistics, especially from

Table 2, is that the opportunities for migration have not been equally distributed among urban

and rural households. In fact, the urban-rural ratios in Table 2 show that remittance recipients in

urban areas have continued to outnumber those in rural areas, and that they may even continue to

do so since the ratios show hardly any decline. All in all, urban households are more able to

pursue jobs abroad, as various other studies have already claimed and proven (see, for example,

Rodriguez and Horton 1995, and Tan 1991).

Among the reasons found regarding why more migrants typically come from urban

households is that urban households are wealthier (Rodriguez and Horton 1995), which Tables

3A and 3B confirm. With higher incomes, urban households can more easily assume the initial

costs of migration, such as those of acquiring the education required by overseas jobs.

Table 3A. Average Family Income (in Pesos), Urban – Rural.

1985 1988 1991 1994 1997 2000 All Philippines 31,052 40,408 65,186 83,161 123,168 144,039

Urban Philippines 46,127 60,330 89,571 113,121 178,121 204,977 Rural Philippines 21,875 28,284 41,199 53,483 73,319 85,373

Urban/Rural 2.11 2.13 2.17 2.12 2.43 2.40

Source: National Statistics Office. FIES, various years.

Table 3B. Average Family Expenditure (in Pesos), Urban – Rural.

1985 1988 1991 1994 1997 2000 All Philippines 26,865 32,521 51,991 67,661 99,537 118,002

Urban Philippines 39,134 47,299 70,551 91,115 140,955 164,794 Rural Philippines 19,397 23,529 33,733 44,427 61,966 72,953

Urban/ Rural 2.02 2.01 2.09 2.05 2.27 2.26

Source: National Statistics Office. FIES, various years.

22

Because of differences in wealth, urban and rural households are also presumed to differ

in education and skill levels (although the causation between wealth and education may have

worked the other way as well). In addition, urban households also have an advantage over rural

households because the former group faces lower transaction and placement costs when it comes

to migration. Placement offices are usually located in urban centers. Information about overseas

employment openings and conditions is also more difficult to access from rural areas because of

the limited availability of media, especially access to the Internet. Because of these, Tan (1991)

surmises that workers in rural areas probably resettle into urban centers first before migrating to

foreign destinations. This hypothesis is not entirely unexpected, given that wages are also higher

in urban areas, since the theory of human capital does predict that workers tend to move to where

wages are higher and where employment is available (Ehrenberg 2000).

Given also that urban households generally have better opportunities to migrate than rural

households, an analysis of the distributional impact of remittances as migration opportunities

spread across the population can therefore be carried out by considering separately the effect of

overseas remittances on urban and on rural income inequality. Such an analysis is actually a

variation of the one done by Stark, Taylor and Yitzhaki (1986) on the income distribution of two

Mexican villages, since the villages also differ in terms of their access to foreign labor markets.

In this sense, comparing the composition of urban and rural income inequalities will be

analogous to considering two contrasting conditions where migration opportunities in one

scenario are relatively more diffused than in the other.

23

The Composition of Income Inequality

The theoretical framework argues that the impact of migrants’ remittances upon income

inequality tends to become more favorable (or less unfavorable) as migration opportunities are

diffused through the population over time. When viewed in this context, the results of the

decomposition exercise seem to support the argument.

Shares of Income Sources in Total Income (Si)

Table 4 shows the shares of the four income sources in total income for 1985 – 2000. The

proportions are similar to those found by Rodriguez (1998) for 1991 Philippines. Paid wages

have been the most important income source in urban areas, representing, on the average, about

47% of total household incomes. In the rural areas, however, entrepreneurial income is a more

significant source of income, averaging about 38% as against only 35% for paid wages. This is

so because of a higher incidence of self-employment in the rural areas (Rodriguez 1998).

Nevertheless, paid wages exhibit an increasing trend for the entire country, while entrepreneurial

income shows the opposite. Other sources of income consistently represent just about a fifth of

total incomes for the entire Philippines.

Remittances, on the other hand, represent less than 10% of total incomes, but it appears to

be a more important income source in the urban areas than in rural areas. Moreover, the share of

remittance income in total income does not exhibit a clear trend for the entire Philippines, which

Figure 1 illustrates. It was as high as 8.3% in 1985 then it dropped to 6.8% in 1997. The share

then spiked back up to 8.6% in 2000. This erratic trend may most probably be due to movements

in the shares of the other domestic income sources. The share of remittances in total income has

dropped, for instance, because the shares of domestic incomes like wages have increased.

24

Table 4. Shares of Income Sources in Total Income.

Income Source 1985 1988 1991 1994 1997 2000

All Philippines Wages 0.3825 0.4336 0.4300 0.4518 0.4647 0.4586 Entrepreneurial income 0.2836 0.2822 0.2844 0.2530 0.2618 0.2444 Other income 0.2513 0.2139 0.2018 0.2197 0.2056 0.2109 Remittances from migrants 0.0826 0.0703 0.0838 0.0755 0.0680 0.0861

Total 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000

Urban Philippines Wages 0.4213 0.4671 0.4585 0.4756 0.4939 0.4843 Entrepreneurial income 0.2247 0.2328 0.2476 0.2173 0.2287 0.2121 Other income 0.2625 0.2228 0.2036 0.2266 0.2058 0.2119 Remittances from migrants 0.0915 0.0773 0.0903 0.0804 0.0716 0.0916

Total 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000

Rural Philippines Wages 0.3044 0.3627 0.3356 0.3707 0.3732 0.3776 Entrepreneurial income 0.4023 0.3867 0.4063 0.3746 0.3654 0.3459 Other income 0.2286 0.1953 0.1960 0.1963 0.2047 0.2077 Remittances from migrants 0.0646 0.0553 0.0621 0.0585 0.0567 0.0688

Total 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000

*NOTE: Figures may not add up to totals correctly due to rounding.

Figure 1. Share of Remittances in Total Income.

0.00

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.10

1985 1988 1991 1994 1997 2000

Year

Sha

re

All Philippines

Urban

Rural

25

Furthermore, that the share of remittances is higher for urban households is expected.

Stark, Taylor and Yitzhaki (1986) saw that the shares of international remittances were higher in

the Mexican village that had had more migrants to the United States. Hence, it seems reasonable

to expect the share of remittances in urban households’ incomes to be higher than in rural

households’, since migration is a more practicable or possible employment option in urban areas.

This, in turn, may ultimately be due to the fact that migration-facilitating factors, such as access

to information, are more available and relatively more diffused in urban areas.

In addition, Stark and Taylor (1991) hypothesized that the more relatively deprived—in

this case, the rural households relative to the urban households—were more inclined to migrate

to improve their living conditions. We can also expect the same to happen in the Philippines,

since rural households do appear to be more deprived with their lower average incomes and

limited access to education and media. The decomposition results as shown in Table 4, however,

do not necessarily suggest that rural workers have not migrated as predicted. It is highly possible

that rural households migrate to the cities first, where information about and placement for

overseas jobs are more accessible. With rural-to-urban migration, the otherwise rural incomes

will be considered urban incomes, and this is necessarily reflected in higher incomes—and

higher income shares—for urban areas. It should be stressed, though, that this is just a

hypothesis, and that it may need to be tested further.

On the other hand, the shares of remittances in total household incomes do not exhibit a

clear-cut trend, both for urban and rural Philippines. This, however, does not mean that the

irregular trend in the share of remittances has little or no impact on the contribution of

remittances to total income inequality, as will be seen later.

26

Gini Coefficients (G i)

The contribution of an income source to overall income inequality does not rely solely on

its share in total incomes. It also depends on the distribution of income from that particular

source.

Table 5 summarizes the inequality in total incomes, as well as the inequality within the

four income sources. It shows that total urban income is more unevenly distributed than total

rural incomes. In addition, the total- income Ginis indicate that, over time, income distribution

has improved in urban areas but it has worsened in rural areas. For the entire Philippines,

however, it seems that income distribution has generally improved.

Table 5 also shows that remittance income is the least evenly distributed income source;

that is, of the four income sources, it has the highest average Gini coefficient of about 0.94 for

the entire Philippines. The inequality even appears to be more acute in the countryside: The rural

Gini coefficient s have an average of about 0.96, while the urban Gini coefficients have an

average of about 0.92.

Table 5 further reveals that the changes in the coefficients are very minimal—in fact, the

changes do not exceed 0.01 between the three-year periods. Thus, it should be prudent to remark

that the inequality in remittance incomes has remained relatively unchanged. The remittance

Gini coefficient for urban areas, for example, remains almost constant at 0.92, although the rural

coefficients show slightly more evident deviations. Indeed, looking at the trend lines plotted in

Figure 2 below will support these observations.

27

Table 5. Gini Coefficients.

Income Source 1985 1988 1991 1994 1997 2000


Total Income 0.4970 0.5080 0.4998 0.5012 0.4841 0.4769


Total Income 0.5052 0.5153 0.4927 0.4963 0.4621 0.4536


Total Income 0.3878 0.3910 0.3999 0.3921 0.4187 0.4115

Figure 2. Gini Coefficient of Remittances.

0.00

0.50

1.00

1985 1988 1991 1994 1997 2000

Year

Gin

i Co

effic

ien

t

All Philippines

Urban

Rural

28

Table 5 and Figure 2 show that the distribution of remittance income is very skewed,

which means that very few Philippine households actually receive remittance income. In fact,

less than 10% of Philippine households receive some income from abroad. Moreover, the Gini

coefficients for rural Philippines are higher than for urban Philippines because no more than

about 5.5% of rural households receive remittances, whereas as much as 9.2% of urban

households receive income from abroad. One may also imply from the figures in Table 6 below

that the change in the Gini coefficient of remittance income is negligible since the changes in the

proportion of remittance-receiving households within populations are small.

Table 6. Percent of Households Receiving Income from Abroad.

1985 1988 1991 1994 1997 2000 All Philippines 5.8 5.8 6.5 6.8 6.2 7.2

Urban Philippines 9.2 8.4 8.3 8.8 7.8 9.1 Rural Philippines 3.7 4.2 4.7 4.8 4.8 5.5

Urban/Rural 2.5 2.0 1.8 1.8 1.6 1.7

Source: Authors’ estimates based on NSO-FIES, various years.

One may recall that the urban-rural ratios in Table 2 indicate that remittance recipients in

urban areas continue to outnumber their rural counterparts. The urban-rural ratios in Table 6

show a more complete picture, though. The ratios in Table 6 show an evident downward trend,

which implies that of the total number of workers deployed for overseas employment, the

proportion that comes from rural areas seems to have increased over the years. Migration thus

seems to have gradually become a valid labor option for the relatively poorer rural areas as well.

Before discussing its impact on income distribution, though, the paper considers the last

determinant of an income source’s contribution to income inequality, or the Gini correlation

ratios.

29

Gini Correlation Ratios (Ri)

According to Equation (3), an income source’s share in total incomes and its distribution

among the population reflect only a part of its contribution to total income inequality. The

contribution also depends on where the recipients of that income source are situated in the

income spectrum.

Table 7 reports the Gini correlation ratios between income sources and total income. The

significance of these Gini correlation ratios depends on the distribution of the income source,

which its Gini coefficient indicates. Given a very unequal distribution of an income source,

higher Gini correlation ratios imply that more households at the upper ends of the distribution

receive income from that income source. Lower correlation ratios, on the other hand, imply that

even households at the lower ranks of the income distribution receive income from that source.

Table 7. Gini Correlation Ratios.

Income Source 1985 1988 1991 1994 1997 2000


Urban Philippines

Wages 0.7206 0.7474 0.7427 0.7705 0.7529 0.7539 Entrepreneurial income 0.6022 0.6297 0.6033 0.5563 0.5682 0.5059 Other income 0.8159 0.7961 0.7605 0.7925 0.7363 0.7108 Remittances from migrants 0.7112 0.6587 0.6954 0.6469 0.6146 0.6675

Rural Philippines


30

According to Table 7, all income categories are positively correlated with total income.

In the case of remittances, the correlation ratios indicate that inequality in remittance income is

positively and highly correlated with overall income inequality, and even more so for rural areas,

where remittances are more unevenly distributed, than in urban areas. This implies that in rural

areas, primarily households from the higher classes receive remittances from migrants. For urban

areas, though, remittances also accrue mostly to the upper classes (since the ratios all exceed

0.5), but to a lesser degree. It appears that more households at the lower ends of the distribution

receive remittance income in urban areas than in rural areas.

Although remittances accrue mostly to the upper classes for the Philippines as a whole,

the ratios shown in Table 7 illustrate a downward trend for the entire country. Figure 3 illustrates

this trend, which suggests that the proportion of lower-class households receiving income from

abroad has increased over time. This lends further support to the observation that an increasing

proportion of migrants has come from rural households, especially since rural households have

been shown to be less wealthy than urban households and at a disadvantage when it comes to

migration opportunities.

Migration theory actually predicts these results. The reason why remittances primarily

accrue to those in the upper ranks of the income spectrum is because migration entails huge

costs, both monetary and psychic. Hence, although the less-privileged are more inclined to

migrate, they are more constrained from doing so.

In addition, the lower urban ratios, when compared with the higher rural values, further

imply that as migration opportunities become more diffused, more people from the lower ranks

are able to migrate and benefit from remittance income. This should explain the observed

31

downward trend as well, because as more people from the lower ranks seize job opportunities

abroad, the Gini correlation ratio drops.

Figure 3. Gini Correlation Ratio of Remittances.

The Contribution of Remittances to Total Income Inequality

The context is now set for describing the contribution of remittances to total income

inequality. Two other variables can be derived from the three decomposition variables to

describe the contribution of each income source to overall income inequality. One variable is an

exact measure of the contribution of an income source to overall income inequality, expressed as

a percentage of the total- income Gini. The other is the relative concentration coefficient, which

tells exactly if an income source increases or decreases total income inequality.

0.00

0.10

0.200.30

0.400.50

0.60

0.700.80

0.901.00

1985 1988 1991 1994 1997 2000

Year

Gin

i Cor

rela

tion

Rat

io

All Philippines

Urban

Rural

32

Percentage Contributions to the Gini

Table 8 reports the percentage contributions of each income source to the total Gini,

which depend on the magnitudes of all the three decomposition variables. The table reports that

paid wages make the biggest contribution to total income inequality for the entire country, and

that their contribution even increases over time. This result is to be expected because of the trend

in the share of paid wages in total income: Wages are the most significant income source,

especially in urban areas, and their share in total income for the entire country also demonstrates

an upward trend.

On the other hand, although entrepreneurial incomes make up the bulk of rural incomes,

their contribution to rural income inequality still falls below that of wages, and it exhibits a

downward trend. This is so because even in rural areas, the share of entrepreneurial income

declines over time, while that of wages rise. Moreover, for rural areas, entrepreneurial income is

less correlated with total income than wages are, as shown by the higher Gini correlation

coefficients for wages. Rural entrepreneurial incomes are also more evenly distributed than rural

wages, as they have lower Gini coefficients.

In the case of remittances, these have made the smallest contribution to total income

inequality, even if they are the most unevenly distributed income source. Figure 4 helps describe

this contribution of international remittances to total income inequality. The shapes of the lines

closely resemble those in Figure 1, which show that the share of remittances in total incomes has

fluctuated over the time frame considered. This emphasizes the fact that an income source’s

share in total income determines its percentage contribution to total income inequality, and that

the two generally move together. Remittances have thus made the smallest contribution to

overall inequality because they have always had the smallest share in total income.

33

Table 8. Contributions of Income Sources to the Gini (in Percent)*.

Income Source 1985 1988 1991 1994 1997 2000


Total 100.00 100.00 100.00 100.00 100.00 100.00


Total 100.00 100.00 100.00 100.00 100.00 100.00


Total 100.00 100.00 100.00 100.00 100.00 100.00

* Contributions are computed as [(Si × Ri × Gi)/G] × 100, where G is the total-income Gini. NOTE: Figures may not add up to totals correctly due to rounding.

Figure 4. Contribution of Remittances to Total Income Inequality (in Percent).

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

1985 1988 1991 1994 1997 2000

Year

Con

trib

utio

n (%

)

All PhilippinesUrbanRural

34

There is one important difference between Figures 1 and 4, though. While the plot line

for urban Philippines appears above that for rural Philippines in Figure 1, one can observe the

opposite in Figure 4. This implies that the contribution of an income source to total income

inequality does not rely solely on that income source’s share in total income, as the

decomposition equations above actually assert.

As Figure 4 shows, migrants’ remittances contribute more to rural inequality than they do

to urban inequality, even if remittances make up a smaller part of rural incomes, because

remittances are more unevenly distributed in rural areas than they are in urban areas. This is

reflected in a more evident, more-than-proportional increase in inequality in rural areas:

Remittances from abroad represent, on the average, only about 6% of rural incomes, but they

contribute an average of about 11.4% to rural income inequality. On the other hand, remittances

from abroad represent about 8% of urban incomes, but they contribute only about 10.5% to

urban income inequality.

Still, these percentage contributions do not seem to fully answer whether remittances—or

any income source for that matter—serve to increase or decrease income inequality. Since the

contributions of all income sources are positive, it may appear at first glance that all income

sources increase inequality. While these contributions may become negative because the Gini

correlation ratio can acquire a negative value, this possibility seems unlikely given that the Gini

correlation ratios of all four income sources have been consistently positive.

In addition, it is difficult to establish a discernible trend based on these percentage

contributions alone. Figures 1 and 4 seem to show tha t the contribution of remittances over time

depends solely on their share in total incomes; the contribution of remittances to the total- income

Gini fluctuates just because their share in total incomes also fluctuates. The plot lines in Figure 4

35

also cannot account for the declining trend in the Gini correlation ratios (and the theoretical

implications thereof), which, as Equation 3 states, should also explain the contribution of

remittances to the overall inequality.

Relative Concentration Coefficients (gi)

To describe more accurately the contribution of remittances to total income inequality

over time, one must therefore use another measure or variable. Derived from Equation (6) above,

the relative concentration coefficients of each income source are presented in Table 9. An

income source either increases or decreases income inequality, depending on whether the

coefficient is greater or less than unity.

Interestingly, paid wages appear to increase inequality, especially for rural areas. The

coefficients also show an increasing trend, with values again peaking in 2000 for the entire

country. This trend becomes even more apparent for urban Philippines, where paid wages

account for the bulk of household incomes: Wages become inequality- increasing only in 2000

after being inequality-decreasing since 1985.

In the case of entrepreneurial income, they have always been inequality-decreasing,

especially for rural Philippines where it accounts for the biggest share in total income. The

coefficients even exhibit a decreasing trend, which implies that entrepreneurial income actually

contributes to greater income equality over time.

36

Table 9. Relative Concentration Coefficients*.

Income Source 1985 1988 1991 1994 1997 2000


Urban Philippines


Rural Philippines


* Relative Concentration Coefficient = gi = Ri×(Gi/G), where G is the total-income Gini.

With values consistently greater than 1, remittances from migrants are undoubtedly

inequality- increasing. Furthermore, the fact that the rural coefficients are larger than the urban

values implies that remittances worsen rural more than they do urban income inequality, an

observation which is consistent with the calculated percentage contributions to the Gini.

Unlike the computed values for the percentage cont ribution of remittances, however, the

relative concentration coefficients of remittances reveal a more perceptible trend when graphed,

as seen in Figure 5. The coefficients for migrants’ remittances exhibit an obvious declining trend,

especially for rural areas and the Philippines as a whole; the urban trend is a bit more erratic,

although later years generally yield values lower than the 1985 level. This downward tendency

implies that, over time, remittances have gradually become less inequality- increasing, as

predicted by theory.

37

Figure 5. Relative Concentration Coefficient of Remittances.

Therefore, if the relative concentration coefficients reported in Table 9 are as predicted,

then it must be the case that the low-income groups have had greater access to migration-

facilitating factors such that an increasing proportion of migrants have now come from their

ranks. For instance, more placement centers may have been established in rural areas, although

statistics to support this claim are not available. In any case, the conclusion still seems to be

supported by the declining urban-rural ratio in Table 6 above, as well as by the downward trend

in the Gini correlation ratio of remittances in Figure 3—all of which show that low-income

groups have also benefited from remittances. This result would actually agree with those of other

studies on Asian countries, especially the one on Sri Lanka (Rodrigo and Jayatissa 1989), which

0.00

0.50

1.00

1.50

2.00

2.50

1985 1988 1991 1994 1997 2000

Year

Rel

ativ

e C

on

cen

trat

ion

Co

effi

cien

t

All PhilippinesUrbanRural

38

suggested that remittances would tend to improve the income distribution if migrants also came

from low-income households.

Moreover, the conclusion that remittances have gradually become less inequality-

increasing over time must now be seen in light of the fact that migration is a demand-driven

phenomenon and of the view that its impact on income distribution is stage-specific. The time

period covered in the paper was actually a stage in the migration history of the Philippines when

the prevailing demand for overseas workers had just shifted from the professional fields in the

sixties and seventies to the relatively low-skilled service sector in the eighties (see Appendix C).

In fact, the 1996 Survey of Overseas Filipino Workers by the NSO claimed that the service

sector “[had] been the sector with the most demand as early as 1990”—a trend that has continued

well into the nineties, as Table 10 indicates.

Table 10. Distribution of Overseas Filipino Workers, by Major Occupation, 1995 – 2000 (in Percent).

Major Occupation Group 1995 1996 1997 1998 1999 2000 Professional, Technical and Related Work 10.8 12.0 11.0 11.8 14.7 14.0 Administrative, Executive and Managerial 0.3 0.4 0.4 0.6 1.0 0.7 Clerical and Related Workers 2.9 3.4 3.0 4.3 3.3 4.3 Sales Workers 1.7 1.0 1.6 1.5 1.8 1.7 Service Workers 41.4 40.1 40.2 41.9 39.2 38.0 Agricultural, Animal Husbandry and Forestry Workers, Fishermen and Hunters 1.3 1.1 0.6 0.3 0.4 0.8 Production and Related Workers, Transport Equipment Operators and Laborers 41.0 41.1 43.2 39.3 39.6 40.1 Occupation Not Adequately Described 0.8 0.8 0.2 0.2 0.1 0.3

Total 100 100 100 100 100 100 Source: Authors’ estimates based on NSO-Survey of Overseas Filipino Workers, various years. NOTE: Figures may not add up to totals correctly due to rounding. Figures for years prior to1995 were unavailable.

Hence, it also seems plausible that remittances have become less inequality- increasing

over the period considered because the Philippines has faced a demand for migrant labor that

favors the relatively low-skilled occupations and the low-income classes. More importantly, if

39

the Philippines were to face the same demand in the future, then OFW remittances may be

expected to become even less inequality- increasing (or more inequality-decreasing).

The relative concentration coefficients seem to show further that of the three

decomposition variables, the positions of remittance-receiving households in the income

distribution—as manifested in the Gini correlation ratios—greatly determine whether

remittances will alleviate or worsen income inequality. This is so because over time, fluctuations

in the share of remittances in total income do not explicitly mean that remittances have actually

improved or exacerbated income inequality. The distribution of remittance incomes has also

hardly changed, even though migration-facilitating factors have presumably become diffused

through the population. Finally, the changes in the demand for certain types of overseas workers

are more likely to affect the income distribution by determining from which income classes

migrants will most likely come, which ultimately affects the Gini correlation ratios accordingly.

40

VI. Conclusions and Recommendations

In summary, decomposing total income inequality into its sources revealed that

remittances from international migrants contributed to income inequality. Remittances appeared

to increase inequality—and more so for rural than urban areas. However, remittances had

become less inequality- increasing over time, as predicted by the theory of stage-specific

migration and income distribution.

The paper likewise showed how the magnitude of remittances’ contribution to total

inequality in the Philippines was determined by the fluctuating share of remittances in total

income, the highly unequal distribution of remittance incomes, as well as the rank or position of

remittance-receiving households in the income distribution. Of these three determinants,

however, the income class from which migrants came appeared to be the most significant

element in explaining whether remittances increased or decreased total income inequality over

time.

Policy Recommendations

Therefore, if the goal of policy were to reduce income inequality, then policymakers

should best focus on how the income classes from which migrants came would contribute to

overall inequality. More explicitly, policies should aim to allow those in the low-income brackets

greater degrees of access to migration-facilitating factors, for remittances would only reduce

inequality if the less-privileged classes were also able to migrate. An example of such a policy

would include the establishment of more placement centers, especially in rural areas. This would

lower the transaction and placement costs faced by would-be rural migrants, since it would

reduce the need to travel to urban centers. However, doing so might seem inadequate in the long

41

run, especially since migration seemed to have become a viable option for the lower classes and

that the proportion of migrants coming from lower income brackets could even be expected to

continue increasing. In this regard, a complementary policy would be to upgrade existing

training centers and programs for overseas workers.

On the other hand, it would seem imprudent to continue encouraging overseas

employment without giving due consideration to the long-term effects of labor outflows. The

country might continue sending its workers abroad, since their remittances were found to be

becoming less inequality- increasing over time. However, if migrants were to continue coming

from urban areas and from higher- income households, migration might eventually exacerbate

income inequality. In addition, if only the more-skilled workers were able to migrate, an active

deployment of workers for overseas employment might result in brain drain, which had been

“often viewed as increasing inequality in the source [countries]” (Davies and Wooton 1992).

Most importantly, the migration of workers would seem very vulnerable to unanticipated

changes in the skills and occupations demanded abroad. In this regard, policymakers might want

to focus on fostering conditions that would either create domestic employment or make working

domestically more attractive relative to working abroad, with a long-term goal of making the

productivity of labor rise. This approach could take many forms, but two could be cited as

pertinent.

One way to increase the productivity of labor would be through the inflow of capital or

investments, whether foreign or domestic. The inflow of capital would, at least in theory,

increase the productivity of labor because it would increase its demand, and hence domestic

wages. This should address the issues of how to make skilled labor stay and how to absorb the

excess supply of labor if ever the demand for migrant labor slackened. Considering that wages

42

appeared to have a growing importance to households and an increasing contribution to

inequality over the years, though, discussions should be careful with how policies that aim to

stimulate investments would affect the wage structure. In the final analysis, too, domestic

investments should be encouraged primarily, for foreign direct investments would most probably

account for only a small part of total investments. In any case, policies should be geared towards

establishing and sustaining favorable macroeconomic conditions.

The other way to increase the productivity of labor would be by having an efficient

educational system—with an aim to reduce the spread of educational differences and skill levels

between workers and between income classes—so that those in the lower classes would be better

equipped if foreign demand shifted away from low-skilled labor. In this regard, the government

might want to look into its public education system, where many of the low-income classes

acquire the minimum skill requirements needed in securing a job, whether here in the Philippines

or abroad. All in all, by reducing the spread of educational differences, an efficient educational

system might also reduce the spread of migration opportunities between the upper and lower

income classes.

Recommendations for Further Studies

The paper has opened up many areas for future research. The paper has limited itself to a

discussion of income inequality in the context of international migration, but the analysis can be

taken a few steps further. For one, regression analysis can determine the effects of individual

households’ characteristics on income and its distribution. It may even be used to test the

significance of the results of the decomposition.

43

There also has been some mention of the possibility of rural-to-urban migration before

actual international migration takes place. The paper has not dwelled on the issue, although it

remains an important avenue towards understanding the impact of migration and remittances on

Philippine income inequality. In this regard, future studies may opt to decompose income

inequality on a smaller level—perhaps regional—instead of on a national level.

Finally, the paper has not given attention to the further implications of the changes in

inequality. Changes in the overall inequality may have ambiguous implications especially with

regard to welfare. For instance, a small increase in the income of household A, leaving all other

households’ incomes unchanged (ceteris paribus), may result in a worsening of income

inequality depending on household A’s relative position in the distribution. This may be

interpreted as a welfare gain, at least for household A, but not necessarily so for the entire

society. It is thus reasonable to investigate the effect of a small change in income from

remittances—or from any other source, as the decomposition equations allow it—on social

welfare. Doing so, however, will require some knowledge of social welfare functions, which is

beyond the scope of this paper.

•

44

Selected References

Adams, Richard H. Jr. “The Effects of International Remittances on Poverty, Inequality, and Development in Rural Egypt.” Research Report 86. International Food Policy Research Institute, 1991.

Adams, Richard H. Jr., and Harold Alderman. “Sources of Income Inequality in Rural Pakistan:

A Decomposition Analysis.” Policy Research Working Paper 836. World Bank. Agriculture and Rural Development Department, 1992.

Adams, Richard H. Jr., and Jane J. He. “Sources of Income Inequality and Poverty in Rural

Pakistan.” Research Report 102. International Food Policy Research Institute, 1995. Amjad, Rashid. Ed. To the Gulf and Back: Studies on the Economic Impact of Asian Labour

Migration. Geneva: ILO-ARTEP, 1989. Balisacan, Arsenio M., and Sharon Faye A. Piza. “Nature and Causes of Income Inequality in the

Philippines.” Paper presented at the “Conference on Comparative Analyses of East Asian Income Inequalities,” Bangkok, 27 – 28 January, 2003.

Davies, James B., and Ian Wooton. “Income Inequality and International Migration.” Economic

Journal. 102 (1992): 789 – 802. Department of Labor and Employment. Yearbook of Labor Statistics. 2000. Ehrenberg, Ronald G., and Robert S. Smith. Modern Labor Economics: Theory and

Public Policy. 7th ed. Boston: Addison-Wesley, 2000. Hyun, Oh-Seok. “The Impact of Overseas Migration on National Development: The Case of the

Republic of Korea.” To the Gulf and Back: Studies on the Economic Impact of Asian Labour Migration. Ed. Rashid Amjad. Geneva: ILO-ARTEP, 1989. 143 – 166.

Kuznets, Simon. “Economic Growth and Income Inequality.” American Economic Review. 45

(1955): 1 – 28. Layard, Richard, D. Piachaud, and M. Stewart. “The Causes of Poverty.” Tackling Inequality.

Comp. Richard Layard. London: Macmillan, 1999. 91 – 104. Macaraig, Mel A., and Michael B. Sarino. “An Analysis of the Remittances of Contract Workers

and Its Effects on Income Distribution and Welfare.” Thesis. UP School of Economics, 1980.

Mahmud, Wahiduddin. “The Impact of Overseas Labour Migration on the Bangladesh Economy:

A Macro-Economic Perspective.” To the Gulf and Back: Studies on the Economic Impact of Asian Labour Migration. Ed. Rashid Amjad. Geneva: ILO-ARTEP, 1989. 55 – 94.

45

Mangahas, Mahar. “Income Inequality in the Philippines: A Decomposition Analysis.”

Discussion Paper 74-15. Institute of Economic Development and Research. UP School of Economics, 1974.

National Statistical Coordination Board. Philippine Statistical Yearbook. 2002. National Statistics Office. Family Income and Expenditures Survey. Various years. ---. Survey on Overseas Filipino Workers. Various years. Available online. National Statistics

Office Website. http://www.census.gov.ph. Nygard, Fredrik, and Arne Sandstrom. Measuring Income Inequality. Hango, Finland: Hango

Trycken Ab, 1981. Ray, Debraj. Development Economics. Princeton University Press: New Jersey, 1998. Rodrigo, Chandra, and R.A. Jayatissa. “Maximising Benefits from Labour Migration: Sri

Lanka.” To the Gulf and Back: Studies on the Economic Impact of Asian Labour Migration. Ed. Rashid Amjad. Geneva: ILO-ARTEP, 1989. 255 – 302.

Rodriguez, Edgard R. “International Migration and Income Distribution in the Philippines.”

Economic Development and Cultural Change. 46 (1998): 329 – 350. Rodriguez, Edgard R., and Susan Horton. “International Return Migration and Remittances in

the Philippines.” Working Paper. University of Toronto, 1995. Sicat, Gerardo P. “Philippine Income Distribution: Facts for Economic and Social Policy.”

Discussion Paper 69-13. Institute of Economic Development and Research. UP School of Economics, 1969.

---. “Myths and Fallacies in Economic Policy Debates.” Discussion Paper 0001. UP School of

Economics, 2000. Stark, Oded, and David E. Bloom. “The New Economics of Labor Migration.” American

Economic Review. 175 (1985): 173 – 178. Stark, Oded, and J. Edward Taylor. “Relative Deprivation and Migration: Theory, Evidence, and

Policy Implications.” Policy, Research and External Affairs Working Paper 656. World Bank. Population and Human Resources Department, and Agriculture and Rural Development Department, 1991.

Stark, Oded, J. Edward Taylor, and Shlomo Yitzhaki. “Remittances and Inequality.” Economic

Journal. 96 (1986): 722 – 740.

46

Sto. Tomas, Patricia A. “Overseas Employment in the Philippines: Policies and Program.” Migration from the Philippines. Comp. Anthony Paganoni. Quezon City: Scalabrinians, 1984. 101 – 121.

Tan, Edita A., and Dante B. Canlas. “Migrants’ Savings, Remittances and Labour Supply

Behaviour: The Philippines Case.” To the Gulf and Back: Studies on the Economic Impact of Asian Labour Migration. Ed. Rashid Amjad. Geneva: ILO-ARTEP, 1989. 223 – 254.

Tan, Edita A. “Overseas Employment, Savings Rate and Income Distribution: The Philippine

Case.” Proceedings of Second Japan ASIAN Forum on International Labor Migration in East Asia, The United Nations University, Tokyo, September 26 – 27, 1991.

---. “Filipino Overseas Employment—An Update.” Discussion Paper 0003. UP School of

Economics, 2000. Tingsabadh, Charit. “Maximising Development Benefits from Labour Migration: Thailand.” To

the Gulf and Back: Studies on the Economic Impact of Asian Labour Migration. Ed. Rashid Amjad. Geneva: ILO-ARTEP, 1989. 303 – 342.

Untalan, Teodoro S. “The Middle East Demand for Philippine Labor: A Cost-Benefit

Evaluation.” Thesis. UP School of Economics, 1980. World Bank. World Development Report 1995: Workers in an Integrating World. New York:

Oxford University Press, 1995. ---. “Inequality Measurement.” World Bank Official Website. Accessed on 12 March 2003.

http://www.worldbank.org/poverty/inequal/methods/measure.htm. © 2000.

•

47

Appendix A Choosing a Measure of Inequality: The Five Properties of Decomposability

The decompositions of the measures of inequality are generally divided into two types:

(1) a functional decomposition by income-determining characteristics (subgroups), and (2) a

functional decomposition by income sources (Nygard and Sandstrom 1981).

Decomposition by income-determining characteristics determines how much of total

inequality is attributable to variability in each income-determinant or subgroup, as well as to in-

between-groups inequality. On the other hand, decomposition by income sources determines

how much of total inequality is attributable to income from specific sources such as wages,

entrepreneurial income, and remittances from abroad.

While there are many ways of measuring inequality, some measures are sensitive to

changes in the distribution and in the values of the data being computed. For example, the

variance, which is a basic measure of inequality, is not independent of the income scale:

Doubling all incomes would result in a quadrupling of the estimate of income inequality. Thus, it

would be prudent to choose a measure of inequality that would conform to a set of axioms,

which would in turn characterize a fair measure of inequality.

Axiomatic Approach in Choosing a Measure of Inequality

The Axiomatic Approach outlines five key axioms in choosing the appropriate measure

of inequality for decomposition: (1) The Pigou-Dalton Transfer Principle, (2) Income Scale

Independence, (3) the Principle of Population, (4) Anonymity, and (5) Decomposability (World

Bank 2000).

48

The Pigou-Dalton Transfer Principle. The inequality measure should rise (or at least not

fall) in response to a mean-preserving spread. An income transfer from a poorer person to a

richer person should register as a rise (or at least not as a fall) in inequality and an income

transfer from a richer person to a poorer person should register as a fall (or at least not as an

increase) in inequality.

Income Scale Independence. The inequality measure is scale independent if it is invariant

to uniform proportional changes. If each individual’s income changes by the same proportion

then inequality should not change.

Principle of Population. The inequality measure should also be invariant to replications

of the population. Merging two identical distributions should not alter inequality.

Anonymity. This axiom requires that the inequality measure be independent of any

characteristic of individuals other than their income (or the welfare indicator whose distribution

is being measured).

Source Decomposability. This requires overall inequality to be related consistently to

constituent parts of the distribution, such as specific income sources. For example, if inequality

is seen to rise amongst each income source, then overall income inequality is expected to rise as

well.

Inequality measures that satisfy all of these axioms provide a good foundation for

decomposition analysis. The Generalized Entropy (GE) class of measures, the Atkinson class of

inequality measures, and the Gini coefficient are the more popular inequality measures that

satisfy these axioms.

•

49

Appendix B Comparing the Decomposition Results of Using the Gini Coefficient and the Generalized Entropy (GE) Class of Inequality Measures

Varying measures of inequality will almost always rank distributions in different ways,

especially when using measures that are sensitive to changes that affect distributions. The Gini

coefficient, for example, appears to be a perfectly satisfactory index because it satisfies the five

axioms (or their equivalent, the Lorenz consistency) discussed in Appendix A (Ray 1998).

However the problem arises when two Lorenz curves that are being compared cross each other

(i.e., when sub-vectors of income overlap), in which case the Gini may fail to meet the

decomposability axiom and may therefore yield results that are not intuitively or mathematically

appealing (World Bank 2000).

Even the parameter used in the Generalized Entropy (GE), which represents the weight

given to distances between incomes at different parts of the income distribution, may give biased

results (World Bank 2000). For lower values of the parameter, the GE is more sensitive to

changes in the lower tail of the distribution, and for higher values the GE is more sensitive to

changes that affect the upper tail.

Clearly, there is no intuitive reason to prefer one measure to another. The logical move

then is not to rely entirely on one particular measure of inequality, but to compare the results of

using different inequality measures.

Figure A below plots the percentage contribution of migrants’ remittances to total income

inequality, computed using the Gini coefficient and the Generalized Entropy class of inequality

measures over the period of interest. The Gini coefficient yields values that are generally higher

than those of the GE class values. More interestingly, while the two measures indicate

contrasting trends, specifically from 1988 to 1994, a comparison of the results of using two

50

different inequality measures captures the fact that remittances’ percentage contribution to total

inequality has generally been erratic. Furthermore, even without reference to the theoretical

explanation as to where the inconsistency may have come from, this finding shows that

comparisons and conclusions may be sensitive to the choice of the index.

Figure A. Contribution of Remittances from Migrants to Total Income Inequality (in Percent).

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

Year

Per

cen

t

GE* 8.28 7.82 5.15 6.57 5.06 8.39

Gini** 12.65 9.70 11.79 9.92 8.97 12.21

1985 1988 1991 1994 1997 2000

*Source: Balisacan and Piza. 2003. **Source: Authors’ computations.

In this sense, it is important to realize that coming up with stricter criteria for choosing an

index of inequality measurement will be inevitably subjective and controversial. Relative

inequality may be perfectly discernible some of the time and difficult to judge at other times

(Ray 1998).

•

51

Appendix C Background Information on Philippine Migration

Historical Background of International Migration

First Wave

The first wave of migration occurred as early as the American colonization period. In the

1900s, the Philippines became a source of cheap plantation manpower when Hawaii experienced

a severe shortage of labor. The Hawaiian Sugar Planters Association increased its incentives for

recruiters to cope with the labor shortage, and the recruitment for overseas employment thus

became a lucrative business. However, there were cases of illegal recruitment for non-existent

jobs, or of low wages and inadequate living conditions.

Responding to the need to protect the interests of the Filipino worker, the Philippine

legislature passed Act 2486 in February 1915, the first manifestation of government regulation of

overseas employment and which remained the reference legislation for almost sixty years. The

Act provided, among others, that (a) no contracting of minors under fifteen and minors of

eighteen years without the written consent of parents or guardians was allowed; (b) return

transportation for workers who had finished their contracts or had become unfit for work was

guaranteed; and (c) recruitment of members of non-Christian tribes for exhibition purposes was

strictly prohibited (Sto. Tomas 1984).

Then, after World War II, Filipino contractors, engineers and technicians were recruited

to rehabilitate and reconstruct American strongholds in the Pacific region, particularly in Guam

and Vietnam.

52

Second Wave

The second wave of migration occurred in the 1960s as a response to the liberalization of

US immigration policy, which replaced racial quotas with occupation-based quotas. The

liberalization also aimed to weaken the monopoly of the American Medical Association on the

supply of medical workers. Filipino doctors and nurses responded favorably to this opportunity

to work abroad; in fact, according to Tan (2000), nearly half of the country's graduating doctors

went to the US in the said decade. Nevertheless, up until 1970, the volume of overseas

employment was assumed to be unsubstantial. Thus, Act 2486 was seen to suffice as the

government's main policy regarding overseas contract employment (Sto. Tomas 1984).

Third Wave

Then the Middle East phenomenon began in the 1970s, which started a new wave of

migration. The oil shock provided oil-producing countries with resources for construction and

development projects. However, these countries required more skilled manpower than their

natives could supply; hence, they started hiring overseas workers, such as seamen, construction

workers and engineers, from neighboring Asian countries like the Philippines.

At this time, however, the Philippines experienced worsening deficits in its current

account. This made the foreign exchange remittances of overseas Filipino workers valuable.

Thus, Presidential Decree 442-also known as the Labor Code of the Philippines-was signed into

law on 01 May 1974, ushering in a new period in government supervision of overseas

employment. Under this law, the government set up the Overseas Employment Development

Board, which aimed to lower the costs of emigrating through a reduction in the travel tax and the

establishment of one-stop shops that simplified the processing of travel and employment papers.

53

All these were aimed to encourage migration and hence the inflow of much-needed foreign

exchange.

Presently, the Philippine Overseas Employment Administration (POEA) and the

Overseas Worker's Welfare Administration (OWWA) have been taking care of the interests of

the OFWs. The POEA, for example, has worked to reduce the number of illegal recruitment

cases (Tan 2000). The OWWA has also established a fund to lend money to overseas workers

and their families, to provide pre-departure training grants for the migrant, and to extend

scholarships for the migrants' children. However, no information about the actual impacts of

these OWWA programs is available, whether in terms of the amount of funds spent or the

number of beneficiaries served (Rodriguez and Horton 1995).

Government Policy on Overseas Contract Workers’ Remittances

Government policy has underscored the importance of international remittances to the

Philippine economy.

This section draws heavily from Rodriguez and Horton (1995), because no other source

that described Philippine policy on remittances was available. In 1982, an interagency body

formed by the Central Bank of the Philippines, the Ministry of Foreign Affairs and Ministry of

Labor and Employment, passed Executive Order (EO) 857, which required the mandatory

remittance of 50% to 70% of overseas contract workers’ salaries. Sanctions imposed on

delinquent contract workers included the non-renewal of the worker’s passport or contract, and

the disapproval of new contracts for overseas employment.

After the government received numerous complaints from the contract workers, EO 1021

repealed the provisions of EO 857, although the target remittance figure remained in effect. The

54

government’s policy on compulsory remittances actually remains valid up to now, but it isn’t

widely observed because it is difficult to implement.

Perhaps to increase the overseas workers’ willingness to remit, the government started

giving incentives to migrants instead of imposing sanctions in the early nineties. There were

numerous programs, such as the Overseas Workers Investment program in 1991, but no data

about their performance or success were made available. (No data about the goals and objectives

of the said program were available.)

There were other programs that provided alternative channels that facilitated remittances.

For example, the OWWA provides remittance assistance to overseas contract workers: It takes

deposits abroad, then it delivers the money within 24 hours in Metro Manila (it takes two to three

more days for provinces). Many other Philippine banks work similarly.

Rodriguez and Horton conclude that the change from an imposing policy to a more

supportive one reveals the difficulty of increasing remittances through legislation, especially if

the determinants of overseas workers’ willingness or propensity to remit is not yet well

understood.

•

55

Appendix D Tables A1 – A6: Composition of Household Income Inequality, by Year

Table A1. Composition of Household Income Inequality, 1985.

Income Source

Share in Total Income (Si)

Gini Correlation (Ri)

Gini Coefficient (Gi)

Contribution Gini ̂ Contribution to Gini (%)

Relative Concentration Coefficients*

All Philippines


Total 1.0000 1.0000 0.4970 0.4970 - - - - - - Urban Philippines


Total 1.0000 1.0000 0.5052 0.5052 - - - - - - Rural Philippines


Total 1.0000 1.0000 0.3878 0.3878 - - - - - - NOTE: Figures may not add up correctly due to rounding.

^ computed as (Si) * (Ri) * (Gi) * computed as Ri*(Gi/G), where G = total-income Gini

56


Income Source






All Philippines








57


Income Source






All Philippines








58


Income Source






All Philippines








59


Income Source






All Philippines








60


Income Source






All Philippines








the contribution of ofw remittances to income …...ii. review of related literature much...

Documents