This paper examines the impact of fiscal and other economic policies on income distribution in the Philippines. Section II of the paper provides some details on the evolution of the distribution of income over time, and reviews some demographic information about families and individuals in the lower third of the distribution. Section III discusses some of the ways that past fiscal and macroeconomic policies have affected the distribution of income over time, as well as other characteristics of the economy that have tended to perpetuate the relatively uneven distribution of income in the Philippines. It, then, discusses some recent policy initiatives intended in part to improve the distribution of income. Section IV looks at how the distribution of income may have affected the formation of economic policy, paying particular attention to the role of oligarchies. Section V concludes.
Poverty is generally more widespread in the Philippines than in its ASEAN neighbors. The distribution of income (as measured by the Gini ratio) has been remarkably stable over time, and the country has experienced a much slower reduction in poverty than have some other Asian countries. Some of the blame for the slow progress in enhancing equity can be apportioned to economic policies that have either discriminated against agriculture—where most of the poor earn their incomes—retarded growth, or led to underinvestment in the most abundant resource of the poor, their potential for human capital development. In turn, these policies have sustained powerful interest groups which in the past have blocked or delayed economic reform. Recent political and economic policy reforms, including widespread deregulation of the economy, should improve the prospects for progress in poverty alleviation and increase equity. However, it will likely take some time before the poorest elements of Philippine society experience the benefits of recent rapid economic growth.
II. Income distribution in the Philippines
Among its ASEAN neighbors, the Philippines is notable both for its very high poverty incidence, especially when measured relative to total basic expenditure rather than just expenditure on food, and for its very slow progress in reducing the rate of poverty. As indicated in Table 1 below, although the poverty incidence in the Philippines was in the middle of a representative sample of Asian countries in the early 1970s, very slow progress in reducing the rate of poverty over the following two decades meant that by the early 1990s, the poverty rate in the Philippines was dramatically higher than in the other members of the group. In 1994, some 4.5 million families lived below the poverty line, about 300,000 fewer than in 1991, but about 300,000 more than in 1988.2
|Years||Reduction||First Year||Last Year|
Income distribution in the Philippines is significantly unequal. In 1994, the richest 20 percent of the population received 52 percent of the country’s total income, nearly 11 times the share received by the poorest 20 percent. The figures for the Philippines were little changed over the decade to 1994, and even worsened marginally: in 1985, the richest 20 percent of the population received about 52 percent of national income and had an average income that was approximately ten times that of the poorest 20 percent.4 Moreover, since 1957 the Gini ratio has barely changed, varying only between 0.45 and 0.51 over a period of nearly 40years.5 Between 1985 and 1994, the Gini ratio was virtually constant, despite the economic recovery that began during those years (see Table 2) and a small improvement observed between 1991 and 1994.6
|Percent of income, top 20 percent||48.6||56.5||56.0||54.0||52.1||51.8||53.9||51.9|
|Percent of income, bottom 20 percent||6.5||4.2||3.5||3.6||5.2||5.2||4.7||4.9|
It is interesting to note, however, that the disappointing experience of the Philippines in improving the distribution of income is not unique in Asia. In fact, none of the countries in the above sample have experienced large declines in Gini ratios in recent decades, and in the case of Thailand income distribution appears to have worsened (Table 3).7 Instead, it would appear that the other countries in the sample have reduced their rates of poverty by increasing income levels across the entire distribution of income. This suggests that decades of very slow growth, rather than inequality per se, may have been the most important factor in the persistence of poverty in the Philippines. Indeed, between 1970 and 1995 real GDP in the Philippines grew at about half the rate of these other countries and barely exceeded population growth (Table 4). To put the growth figures in some perspective, if two countries had identical GDP in 1970, but one consistently grew at a real rate 2.8 percent higher than the other, by 1995 real GDP in the faster-growing country would be double that in the slower-growing one.
|Years||First Year||Last Year|
In line with observations in other countries, the distribution of expenditure in the Philippines is somewhat less inequitable than that of income.8 Nevertheless, in 1994 the top 20 percent of the distribution accounted for nearly 48 percent of annual expenditure, almost eight times the share of the poorest 20 percent. In 1985, the richest 20 percent of the population accounted for 47 percent of expenditure, a little more than seven times the share of the poorest 20 percent.9 By comparison, in Indonesia, which had a roughly comparable per capita income, the richest 20 percent of the population accounted for 40.7 percent of expenditure in 1993, only 4.7 times the expenditure of the poorest 20 percent of the population.10
Looking beyond peso income and expenditure, the performance of the Philippines in improving human development indicators has also been slow relative to other Asian countries (Table 5). While other countries have made dramatic strides in limiting infant mortality rates, improving adult literacy, and increasing life expectancy, progress in the Philippines, while not negligible, has been slower. As a result, while social indicators in the Philippines were once quite good relative to other East Asian and lower-middle-income countries, by the early 1990s they were at best average and in some cases—such as childhood malnutrition—much worse.11
|Population Growth Rate (in percent annual average)|
|Child Malnutrition (percent under five years)|
|Infant Mortality (per 1,000 live births)|
|Life Expectancy at Birth (years)|
In the Philippines, as in many other countries, poverty levels are calculated both including and excluding nonfood necessities. Of course, the poverty threshold including only foodstuffs—often referred to as subsistence poverty—will always be lower than one based on total basic expenditures. Thus, poverty incidence measured on a subsistence basis is only about half that when nonfood expenditures are included, with a poverty incidence of about 18 percent of families in 1994 (Table 6). However, improvement in this indicator has also been slow, declining only by about 6 percent of the population between 1985 and 1994. In 1994 some 2.3 million families were below the subsistence level of income, about 140,000 fewer than in 1991, but about 165,000 more than in 1988.12
|Total basic expenditure||44.2||40.2||39.9||35.5|
At the same time, it is important to bear in mind that the relatively high level of poverty in the Philippines, compared to other ASEAN countries, is due in part to differences in the way the subsistence threshold is calculated among countries. Even if two countries define subsistence identically in terms of caloric consumption, there may still be differences in the composition of the baskets of goods used to translate calories into incomes. In the Philippines, nongrain items—which in the Philippines are seven times as expensive per calorie as grains—constitute an average of 34 percent of the calories, compared to 10 percent in China, 19 percent in Indonesia, and 27 percent in Thailand. As a result, subsistence income in the Philippines is higher than it would be using the consumption baskets of other countries. Reducing the share of nongrain items to 10 percent of the consumption basket, as in China, would reduce subsistence poverty by almost two-thirds.13 This exercise illustrates the difficulty of making cross-country comparisons of poverty rates.
An analysis of the regional pattern of poverty and income distribution reveals significant and persistent differences among regions. The National Capital Region had the lowest incidence of poverty (measured relative to total basic needs) in 1994 (8 percent of families), while Bicol had the highest rate (55 percent in 1994). While the poverty incidence has declined sharply in some regions in recent years, falling from 23 percent of families to 8 percent in Metro Manila between 1985 and 1994, and from about 60 percent to between 43 and 33 percent in the three Visayas regions over the same period, in others it has remained stable or even increased. In addition, the gap between the poorest and richest regions is largely unchanged: in 1985 the difference between the poverty incidence in Metro Manila and the (unweighted) average for the three poorest regions (Bicol, Western Visayas, and Eastern Visayas) was 36.8 percent, while in 1994 the gap between Metro Manila and these same three regions was 37.4 percent. The gap between Metro Manila and the three poorest regions in 1994 (Bicol, Central Mindanao, and the Autonomous Region of Muslim Mindanao) had grown to 48.6 percent.14 At least as far as the poor are concerned, the benefits of the economic recovery since the late 1980s appear to have been concentrated in specific regions. Moreover, the persistent and large disparity in poverty incidence among regions suggests the presence of structural impediments in the labor market that may exacerbate income inequalities. Of course, the high poverty rates in Mindanao also reflect the dire consequences of years of civil war. Of the 4.5 million families living below the poverty line in 1994, about 1.4 million (30 percent) lived on Mindanao, with an additional 1.1 million (23 percent) living in the Visayas. Only about 142,000 (3 percent) of poor families were located in the National Capital Region.15
Gini ratios for most regions improved between 1985 and 1994, although most gains were small and in some regions equality worsened. Even in Metro Manila, where the poverty incidence declined by nearly two-thirds, the Gini ratio improved only from 0.415 to 0.397. Although regions with the greatest reductions in poverty tended to have the sharpest improvements in their Gini ratios, this was not universally the case: for example, the Gini ratio in Eastern Visayas actually worsened, from 0.390 to 0.420 between 1985 and 1994, even as the poverty ratio declined from 59.0 percent to 37.9 percent over the same period. The improvements in income distribution have been concentrated in the last few years, as Gini ratios in almost all regions deteriorated between 1985 and 1991.16
A. Characteristics of the Poor
Drawing on the results of the 1991 Family Income and Expenditure Survey,Intal (1994) identified a number of important characteristics of the poor:
More than one-half of all households whose head has an elementary school education or less are poor. In fact, these households account for about three-quarters of all subsistence-poor households.
Larger households have a higher incidence of poverty than do smaller ones. Thus, the incidence of poverty among the population is greater than that among families. This finding needs to be interpreted carefully, however, since it is not clear whether the causality runs from large households to poverty or vice versa.17
Female-headed households have a much lower incidence of poverty than do male- headed households. Higher educational attainment and smaller family size probably account for this finding (see below).
The poor are disproportionately employed in agriculture, fishing, and forestry. Altogether, farmers, farm workers, fishermen, and forester households account for 62 percent of all poor households and 71 percent of the subsistence poor. Among the poorest three income deciles, about 74 percent of men and 56 percent of women over the age of 15 are employed in agriculture, forestry, logging, fishing, or hunting.18 By contrast, in 1996 the agriculture, fishery, and forestry sector accounted for only about 43 percent of the employed labor force.19
In Metro Manila, the poorest income deciles tend to receive a large percentage of income from wages and salaries (excluding agriculture) than do the wealthier deciles. In other words, the relative importance of wages and salaries tends to decline with income. However, in the rest of the country the opposite is true, with nonagricultural wages and salaries constituting an increasing share of total income as incomes rise. Agricultural wages and salaries are not a significant source of income for any decile in Metro Manila, but outside of the capital they account for a much larger share of income for the lowest income deciles than for the highest ones.
Remittances from abroad tend to constitute a larger share of income in more developed regions of the country than in less developed ones, and tend to account for a larger share of income in wealthier deciles than in poorer ones. Presumably, this reflects the fact that opportunities for overseas employment are more readily available for those with higher levels of human capital. In a partial equilibrium sense, this may suggest that widespread emigration—a defining characteristic of the Philippine labor market—has tended to exacerbate rather than reduce income inequalities. However, these transfers may have led to increased national savings and higher domestic investment, creating employment opportunities, and raising the marginal product of labor for nonemigrants. Of course, emigration also reduces oversupply in domestic labor markets directly. Accordingly, emigration has almost certainly had a beneficial impact on poverty and may even have a net positive impact on equality.
Data suggest that the poor are disproportionately located in rural areas, with 60 percent of the poor (and more than 70 percent of the population in the three lowest income deciles) living in rural areas in 1991 compared to 51 percent of the total population. Since 1960, the percentage of the population that is classified as urban has increased from about 30 percent to nearly half, while the share of the poor population that is urban has grown from 30 percent only to about 40 percent. These figures could be interpreted as suggesting that rural to urban migration has led to a reduction in poverty among migrants. However, Balisacan (1995) indicates that these figures are in part a statistical artifact. Over the last 30 years, a large number of areas have been reclassified from rural to urban. Almost by definition, reclassified areas have experienced faster economic growth than have average rural areas, and it is, therefore, not surprising that the poverty incidence in these areas has declined. In 1990, about 36 percent of the population lived in areas that had been considered urban in 1970. In other words, the share of the population living in historically urban areas increased only slightly (about 4 percent) in the period 1970-90. Thus, it is not migration but rather reclassification of rural areas to urban status that has led to the large increase in the share of the population that is urban. Moreover, using the urban and rural classifications that applied in 1961, he finds that while poverty incidence declined only slightly in urban areas between 1961 and 1991 (from about 57 percent to about 51 percent), poverty declined markedly in rural areas over the same period (from 60 percent to 41 percent). In other words, while poverty has actually declined faster in areas that were rural in the early 1960s than in historically urban areas, the fact that fast-growing areas will tend to be reclassified as urban creates the opposite impression.
Balisacan (1995) conducted an econometric exercise to account for earnings differences among the urban population, drawing on the urban panel of the 1988 Family Income and Expenditure Survey. Dividing his sample between the self-employed and wage and salary workers, he found that for both modes of employment, earnings of workers who attended or completed high school were not significantly different from those with only a primary education, after controlling for age, sex of household head, access to electricity and roads, family size, and household assets.20 Among the self-employed, earnings of those who had some college education were significantly higher than those who did not, while earnings of those who graduated from college were significantly higher than those who had enrolled, but not graduated. Among wage and salary workers, only those who had graduated from college earned significantly more than those with lower levels of education. The rates of return on education were generally higher at all levels of education for wage and salary workers than for the self-employed, however.
Despite the finding of a higher poverty incidence for male-headed households, Balisacan found that for both self-employed and wage and salary workers, earnings are higher for urban males than for urban females. Presumably, the higher poverty incidence for male-headed households reflects differences in education and family size. In fact, the discrepancy between earnings for males and females in his sample is dramatic, with self-employed men earning 65 percent more than their female counterparts and wage and salary employed men earning 189 percent more than their female counterparts, after controlling for education, age, and other variables. Access to infrastructure—captured in two variables, the road density in a participant’s region of residence and the percentage of municipalities with electricity in his region of residence—has a positive and significant impact on the incomes of wage and salary workers, but not on the self-employed. Overall, Balisacan’s results suggest that improved access to physical infrastructure and increased educational achievement rates can substantially reduce urban poverty and improve the distribution of income.
Otsuka and others (1992) examine various factors affecting the income of land owning and landless households in rice growing villages in Central Luzon and Panay. As for the urban poor, education has a significant, positive impact on family income.
III. Economic policy and income distribution
As noted in the previous section, income distribution in the Philippines is highly uneven, and poverty rates are higher than in other Asian countries. In addition, while the poverty rate has declined over time, the rate of decline has been lower than in other Asian countries and income inequality, as measured by the Gini ratio, has been persistent. To some extent, the lack of progress in improving poverty indicators in the Philippines relative to its ASEAN neighbors can be attributed to the country’s relatively poor growth performance: during the 1980s, when most ASEAN countries enjoyed dramatic growth, GDP growth in the Philippines averaged only about 1 percent per year. The persistence of inequality and poverty in the Philippines can, therefore, be traced to economic policies that have hampered growth, many of which have been abandoned recently, as well as to policies that have more directly perpetuated income inequality. This section looks at how trade and exchange rate policies, the tax system, the pattern of government expenditure, and monetary policy have influenced poverty and inequality in recent years.
Blejer and Guerrero (1990) undertake a macroeconomic analysis of the impact of various macroeconomic variables in the Philippines. Focusing on the period 1980-86, they examine the impact of labor productivity, underemployment, the real exchange rate, real government consumption expenditure, the real interest rate, and inflation on income equality, with the last variable measured as income accruing to the poorest 30 percent of the population divided by income accruing to the wealthiest 10 percent of the population. Thus, they study the way various macroeconomic variables affect the relative, not absolute, incomes of the poor. For example, even if a cut in government expenditure lowers the absolute income of the poor, it could still have a positive coefficient if it reduces the income of the wealthy relatively more than that of the poor.
Blejer and Guerrero find that all of the variables included in their regression have a significant impact on income distribution, with high levels of labor productivity, depreciated real exchange rates and high real interest rates tending to improve income equality, and high levels of underemployment, inflation, and government expenditure tending to worsen income equality. The results for productivity and underemployment are not surprising, and the negative impact of inflation on income equality almost certainly reflects the fact that the poor have fewer real assets with which to insulate themselves from the effect of price increases.
The results for the real exchange rate, government expenditure, and the real interest rate are perhaps more interesting. Most likely, the beneficial (from a distributional perspective) impact of real devaluations reflects the fact that the poor are disproportionately involved in agricultural export activities, which would benefit from a devaluation. The beneficial impact of cuts in government expenditure could have occurred because government expenditure primarily benefited the wealthy or because higher levels of expenditure tended to be linked to higher levels of regressive taxation. For example, if the wealthy have relatively better access to public education, cuts in spending on schools could conceivably improve the distribution of income. Finally, Blejer and Guerrero speculate that the positive impact of higher real interest rates could be due to the fact that the poor have less access to credit, and therefore are less affected by high interest rates, or to the fact that the informal sector is less likely than the formal sector to experience a slowdown in activity when interest rates rise. However, the impact of higher interest rates on the distribution of income is complicated, and it will be argued below that high domestic interest rates may have a regressive impact. The following sections explore the possible impact of exchange rate, fiscal, and monetary policies in more detail.
A. Exchange and Trade Policies
For many years, the Philippines pursued an industrial policy based on the idea of fostering import substitution, with little effort to encourage exports. Until the tariff reforms of 1991, trade policies heavily penalized the primary and agricultural sectors—where the poor are predominantly employed—and benefited the manufacturing sector. Throughout the 1960s, 1970s, and 1980s, the manufacturing sector enjoyed an effective rate of protection (ERP) much greater than that of the agricultural sector: in 1965 the average ERP for manufacturing was 29 percent higher than that of agriculture, and by 1985 the gap had widened to 43 percent. However, following the tariff reform of 1991 the difference in relative protection fell to just 17 percent.21 In addition, the overvalued exchange rate of the 1950s through 1980s depressed the prices of export products in peso terms and shifted resources toward import-substituting manufacturing and away from agriculture.22
Moreover, the overvalued exchange rate coupled with fiscal and other incentives such as tax exemptions for imported capital equipment reduced the cost of capital and encouraged the substitution of capital for labor. By one estimate, these incentives reduced the cost of capital by as much as 70 percent in the 1970s and 80 percent in the early 1980s.23 As a result, the share of the industrial sector in total employment has grown only from about 13 percent in 1960 to 16 percent in 1996. Between 1990 and 1996 nearly 80 percent of the five million person increase in total employment can be accounted for by agriculture and services. Manufacturing employment increased by only about 500,000 over the period, or only about 10 percent of the total increase in employment. In addition, total employment growth was slower than labor force growth, so that the unemployment rate in 1996 (at 8.6 percent) was slightly higher than in 1990 (8.3 percent).24
These developments are reflected in the declining real wages in the manufacturing sector. In the first three quarters of 1996, average real compensation in the manufacturing sector stood at only 75 percent of its 1978 level. After peaking in 1988 real compensation in the manufacturing sector has declined in each subsequent year, to the point where it now stands at less than 60 percent of its 1988 level.25
Agricultural incomes—and therefore the incomes of the poor—were also depressed by the heavy regulation of the agricultural sector. Beginning in the 1970s, price controls on rice and other products were imposed, and the importation of wheat and soybeans was monopolized. Controls on the production, marketing and processing of coconuts were put in place, and a price stabilization fund was created. In addition, fertilizer and pesticide imports were controlled through licensing requirements. Under the Aquino administration many of these restrictions were relaxed or eliminated, which Balisacan (1991) finds has led to a reduction in poverty. However, reforms in this area have not yet been completed, and important regulations remain in effect that restrict the ability of farmers to increase their earnings or acquire inputs at the lowest possible prices.26
The bias against agricultural investment introduced by the overvalued exchange rate and the heavy regulation of agriculture have had disastrous effects on productivity in the sector. For example, in the period 1982-85, productivity in the coconut sector—long the county’s most important agricultural industry in terms of export earnings and employment—averaged 1.0 metric ton per hectare, exactly the same as in the period 1962-66.27 The stagnation of agricultural productivity is borne witness by the startling fact that between 1860 and 1996, productivity in the sugar industry (which employs 600,000 people) has increased by less than 20 percent (from 50 tons per hectare to 59 tons per hectare).28
On several occasions, beginning in very limited ways in the 1950s and 1960s, the government has undertaken land reform programs intended to break up large landholdings and improve the distribution of rural incomes by encouraging a more equal distribution of farmland. In 1972 a land reform program was instituted that allowed tenants to purchase land via 15 annual payments equal to 25 percent of the yield for three normal crop years.29 However, coverage was restricted to rice and corn fields, which allowed landholders to exempt themselves simply by shifting production out of these crops. Perversely, the reform covered only tenant farms, giving landowners an incentive to evict tenants and hire landless workers, thereby actually reducing access to land. In addition, substantial bottlenecks in the process meant that very little of the land that was covered by the program was actually redistributed.
In 1987 the Aquino administration introduced a Comprehensive Agrarian Reform Program (CARP) that was intended to redistribute about three-quarters of all agricultural land to landless farmers and farm workers. However, CARP has also been plagued by bottlenecks, owing to a lack of financing for the enormous costs of the program as well as cumbersome administrative requirements, and at end-1994 only about 30 percent of the 10.3 million hectares scheduled for redistribution over the period 1987-98 had actually been redistributed. Strikingly, the Gini coefficient for landholding actually worsened between 1960 and 1990, going from 0.53 to 0.57, despite these land reform efforts.30
Despite its ambitious goals, land reform has thus far done little to alter the distribution of land in the Philippines, or to improve rural incomes. Achievement of the goals of land reform would require a tremendous commitment of resources (on the order of about ₱120 billion) that would likely stifle infrastructure investment and social spending in rural areas. Given that land distribution in the Philippines is not notably more unequal than in some other ASEAN countries with more equal income distributions of rural income (such as Malaysia and Indonesia) it is at best an open question whether successful implementation of the CARP would ultimately improve or worsen the distribution of rural incomes. On the other hand, there can be little doubt that increased expenditure on education, health, and infrastructure in rural areas would reduce poverty and improve the distribution of income.
B. Federal Government Expenditure
Because the poor are relatively better endowed with labor than with physical capital, public expenditure on education and on health can exert an important influence on poverty and income distribution.31 While virtually all primary education in the Philippines is financed by the public sector, about one-third of secondary school students are enrolled in private schools. At the university level, nearly 90 percent of students are enrolled in private schools. Although enrollment rates are very high for all income groups at early ages, rates begin to drop for lower income groups by age 13 in urban areas and by age 11 in rural areas. By age 15, the enrollment rate for the highest income decile in urban areas remains at 0.97, while for the poorest income decile it has fallen to 0.63. In addition, for every income decile, enrollment rates are lower in rural than in urban areas. Scores on elementary school achievement tests, and figures on the average number of textbooks per elementary school pupil, both reveal a significant bias not only toward the upper income deciles, but also toward urban students versus rural ones.32
Overall, public education in the Philippines is underfunded relative to other ASEAN countries (Table 7), with the exception of Indonesia, relative both to GDP and to total government expenditure.33 Moreover, the distribution of education spending among levels is suboptimal. In 1987, the government nationalized village high schools, and in 1989 began providing vouchers for students to attend private high schools. These measures were introduced in order to improve access of the poor to secondary education. However, as a result of them, the share of secondary education in total education spending rose from 10 percent in 1987 to 17 percent in 1994, while the share of primary education dropped from 60 percent to 56 percent.34 More strikingly, the government now subsidizes a greater share of educational expenses at the tertiary level than at the primary level (Table 8). Altogether, Yoingco and others (1994) estimate that about 52 percent of the budget on public education went to programs participated in predominantly by middle class families, with less than 6 percent of the budget flowing to households in the lowest two income deciles in 1990.
|In Elementary Education|
|In Secondary Education|
|In Tertiary Education|
Unfortunately, the Philippines is not alone in this regard. Psacharopoulos (1994) reviews empirical studies from many countries that suggest that secondary and especially tertiary education are oversubsidized, and primary education undersubsidized. In addition, participation rates are probably lower for low-income students than for wealthy ones in almost all countries, because of the high opportunity cost of education. Moreover, it is conceivable that the shift in the allocation of the educational budget that occurred in the late 1980s and early 1990s could have been justified by the very high coverage already enjoyed by primary education. However, a reallocation of spending toward primary education could contribute to a significant improvement in the quality of primary schooling and in retention rates for lower-income students, especially in rural areas, and in the process help reduce poverty levels.
As with education, spending on health and nutrition is relatively low by ASEAN standards (Table 7), and the composition of expenditure is biased away from lower income consumers. Only 25 percent of barangays (the smallest administrative unit in the country) had health stations in 1990, and each health station served 14,200 people in the National Capital Region, but between 21,000 and 44,000 people in other regions.35 This bias is reflected in the fact that per capita government health expenditures in 1986 were three times as high in Metro Manila as the national average, and more than five times as high as in Central Mindanao.36 In addition, expenditure has increasingly concentrated on expensive, tertiary level care that lower income families cannot afford, with the share of expenditure on preventative care falling from 35 percent in 1981 to 14 percent in 1990, while the share of curative care rose from 55 percent to 65 percent during the same period. These developments are reflected in the fact that the infant mortality rate in Metro Manila in 1989 was lower than in any other region and only about one-third the rates in Central and Western Mindanao. Similarly, life expectancy at birth in Metro Manila and Central Visayas was about a decade greater than in Mindanao.37 Housing programs and infrastructure investment also reflect a pro-urban (especially Manila) bias that aggravates existing income discrepancies between urban and rural areas and worsens problems of income inequality.38
C. Tax Policy
Beginning in 1994, the government embarked on a Comprehensive Tax Reform Program (CTRP), and undertook a major expansion of the coverage of the VAT, motivated in large part by a desire to increase the contribution of the tax system to equity. The CTRP has effected major changes to the tax code since 1996, and is now nearly complete. The first component of the tax reform, modifying the taxation of oil products, was introduced in 1996. The second, which increased excise taxes on cigarettes and alcohol, went into effect on January 1, 1997. The remaining components, which will affect personal and corporate income taxes as well as tax administration more generally, were signed into law in December 1997. Changes to the VAT through the enactment of the VAT Reconstruction Act in 1994 greatly expanded the previously limited base of the VAT, extending it to many services consumed largely by upper income groups in the population.
An analysis of the incidence of the Philippine tax system (Yoingco and others, 1994) found that the overall tax system, as structured prior to that time, was regressive at the very bottom of the income scale, but progressive thereafter. Although the results (like those of all incidence studies) depend on assumptions about how various taxes—such as the corporate income tax—are distributed among consumers, shareholders, and employees, in general that study indicated that the overall tax burden on the poorest income group was greater than that on those in the second to fifth income groups. Most direct taxes, including the individual income tax, were progressive. Indirect taxes were found to be regressive at the bottom of the income distribution. Because the poor consume a relatively larger percentage of their incomes than do the wealthy, consumption-based taxes will nearly always be found in the first instance to be somewhat regressive in their impact, even when exemptions are granted to food.39 For example, the study found that alcohol taxation was regressive, reflecting the fact that the poorest income groups spent a relatively higher fraction of their incomes on alcohol products than did the wealthy.
The impact of any particular tax, and indeed of the tax system as a whole, on the economic equity of a society must, of course, be viewed in the context both of the alternatives for raising revenues, and of the progressivity of the expenditure side of the overall government budget—that is, what persons benefit from the taxes raised. While the structure of the Philippine individual income tax is quite progressive, the opportunities for tax avoidance and reduction by the relatively wealthy, as in all countries, are much greater than they are for the less well-off. For example, wage earners’ incomes are effectively taxed through withholding, but the earnings of businessmen are far harder for the tax administration to effectively measure. Similarly, the wealthy have more opportunity to engage in perfectly legal, but effective tax planning, for example, by the use of tax-free investments and deposits. In such a case, in countries in which the administration of taxes on the wealthy is difficult and where there are many tax incentives and preferences that benefit the wealthy, shifting the tax burden toward consumption and away from income may actually result in the wealthy paying a higher proportion of tax revenues, and of their incomes in tax, than where reliance is placed upon income taxes which they can evade or avoid.
It is very difficult to effect a major redistribution of income through the tax system in any country, in large part because many of the very poorest citizens are outside the tax net. For this reason, redistribution to the poor is more effectively undertaken through targeted expenditure programs, using revenues raised in as economically efficient manner as possible. In keeping with this problem, it appears that the tax system in the Philippines, at least prior to 1994, did not directly contribute to an improvement of the income distribution for the poorest groups. In fact, the tax system tended to redistribute income away from the lowest income groups and toward the middle class. Partially to improve overall progressivity, the government undertook the reforms described.
Determining the effects of the tax system on income distribution is always difficult. However, it would seem likely that the CTRP’s increase in excise taxes on cigarettes and alcohol will have a regressive component, since the poor tend to spend a larger portion of their income on these products than do the wealthy. However, because the tax collected on premium brands is greater than that on inferior brands, the regressive nature of the system may be ameliorated somewhat. In addition, the aim of the reform of the direct taxes is in part to increase the overall progressivity of the income tax system, through changes both to the tax structure for individuals and to the corporate tax system.
The existing personal income tax is already somewhat progressive. The reform of the personal income tax is intended to result in a decrease in income taxes for all current taxpayers.40 Because the very lowest income earners are already exempt from the income tax, reforms cannot have a direct impact on their welfare. However, there are households below the poverty line that do have income tax liabilities under the existing system. The reform is intended to ensure that no family living below the poverty line would have any income tax liability. More generally, the income tax reform seeks to bring into the tax net many individuals who were previously outside the net. To the extent that these individuals tend to have relatively higher incomes, the reform would also likely improve the distribution of income.
The impact of the corporate tax reform on income distribution ultimately depends on assumptions about who bears the burden of the tax. However, certain features of the reform, such as the introduction of a tax on employee fringe benefits, would likely improve the progressivity of the overall tax system. In addition, the government’s proposals to eliminate many corporate tax breaks and investment incentives, which were originally included in the CTRP but excluded from the final version, would help eliminate distortions in the economy and ensure that resources flow into their most productive uses, which would ultimately increase growth and benefit the poor.
Finally, it is important to remember that to the extent that the tax reform increases revenues, this will create additional room for government expenditures that could raise incomes of the poor. Ultimately, it is very much a question not just of where the revenue comes from, but also where it goes that determines the progressivity of fiscal policy.
D. Monetary Policy
In recent years, the Philippine authorities have pursued a monetary policy that has resulted in relatively high real domestic interest rates. As noted above, Blejer and Guerrero (1990) found that higher real interest rates led to an improvement in income distribution in the period 1980-86, in part because the poor had limited access to credit. However, there is some reason to expect that this may no longer be the case. In recent years, the country’s open capital account and the relatively few restrictions placed on banking transactions conducted in foreign currency have led to a rapid rise in the volume of dollar lending in the economy. Dollar loans generally have much lower interest rates than peso loans.41 If dollar loans are only available to members of the upper-income classes, then an unintended consequence of the tight monetary policy and open capital account would be to raise borrowing costs for the middle classes while keeping relatively cheap credit available for the wealthy. Although poor farmers have difficulty gaining access to formal sources of credit in many countries, this appears not to be the case in the Philippines, where in recent years many farmers have increasingly been able to receive credit from cooperatives, rural banks, and suppliers and marketers. The relatively high real costs of credit are thus felt throughout the lower portions of the income distribution.
IV. Income distribution and government policies
While the previous section looked at how economic policy has affected income distribution, it is clear that the present unequal distribution of income in the Philippines has also constrained the ability of the authorities to enact reforms that would increase equity. As Boyce (1992) puts it, “unchecked by a more equitable distribution of wealth and power, the ruling elite was free to pursue its self-interest even at the expense of the public interest (p. 19).”
There is a widespread public perception in the Philippines that to a greater extent than in many other countries, the political process has been captured by elites. To some extent, this is an inevitable outgrowth of an unequal distribution of income: the financing of political activity is expensive, and the greater is the concentration of income, the smaller will be the group of individuals exerting political influence. Hand in hand with this, the highly regulated, import-substituting based industrial strategy pursued by the country created many opportunities for rent-seeking behavior, which also contributed to a concentration of power in the hands of an elite.
Karnow (1989) documents the important role that clans and oligarchies played for decades in Philippine politics and economics. At the turn of the 20th century, Philippine towns were controlled by small groups of leading citizens with the ability to manage many aspects of public life. Although elections were introduced, and the electorate gradually broadened, during the early years of the century, Karnow argues that the political parties
…were identical ideologically. They fought partly for prestige—but, more often, for the power to sequester land, steal tax receipts, demand kickbacks for licenses or hand out profitable and frequently fake public works contracts for their relatives. Officials grew rich in subsequent years by pocketing funds allocated by the Manila government to develop municipalities (pp. 228-29).
Although traditional power structures and family ties have weakened in urban areas, in many rural parts of the country traditional elites continue to dominate.
While government officials doled out economic favors to supporters and family members for much of the century, the process reached its most notorious stage during the martial law regime. Karnow provides numerous examples of tax breaks, low-cost loans, debt bailouts, monopoly rights, and other benefits provided by presidential decree in order to enrich President Marcos’s supporters at the cost of taxpayers, workers, or small farmers.
Boyce (1992) discusses the effect of the concentration of wealth and power on the distribution of income in the coconut industry, traditionally the largest employer and largest producer of agricultural export earnings in the economy. Beginning in the early 1970s, a series of presidential decrees created a single institution with control of all stages of coconut production from the purchase of coconuts from farmers to the sale of processed oil on the domestic and international markets. Although these measures were purportedly motivated by a desire to improve conditions for coconut growers, in fact the main beneficiaries seem to have been a very small number of close associates of Marcos. Aided both by a substantial levy on coconut producers and by its monopsony power (which enabled it to increase profit margins on its production from a normal 50 centavos per kilo of copra to P-2-3 per kilo) the United Coconut Oil Mills/United Coconut Planters’ Bank complex amassed vast sums of money (much of it alleged to have been appropriated by Marcos cronies).42 Summarizing the changes in coconut marketing in the 1970s and early 1980s, Boyce states
The primary motive for this transformation, and its major effect, was redistribution of the income generated by the country’s leading agricultural export. The beneficiaries were a handful of politically powerful individuals. The losers included millers and traders driven out of the industry, and above all the coconut growers, who experienced an intensification of the monopsonistic environment in which they sold their product (pp. 17-18).
Although the coconut monopoly was dismantled by the Aquino administration, support for reform and progress in changing the status quo remains lacking in agriculture. As discussed above, land reform efforts have met with very little success in the Philippines, in large part because of the power of landed elites. Similarly, while tariffs have been lowered dramatically in almost all sectors of the Philippine economy, agricultural tariffs remain at the maximum level permitted by WTO agreements, with potentially disastrous effects on productivity.
Income distribution in the Philippines is highly unequal, and poverty rates are higher than in other ASEAN countries. In addition, while the poverty rate has declined over time, the rate of decline has been lower than in other ASEAN countries and income inequality, as measured by the Gini ratio, has been stagnant. To some extent, the lack of progress in improving poverty indicators in the Philippines relative to its ASEAN neighbors can be attributed to the country’s relatively poor growth performance: during the 1980s, when most ASEAN countries enjoyed dramatic growth, GDP growth in the Philippines averaged only about 1 percent per year. However, the persistence of inequality and poverty in the Philippines can also be traced to economic policies that have favored capital over labor, and import-substituting industries over agriculture, and that have led to underinvestment in the human capital of the poor. The effects of these policies on the agriculture sector in particular have been devastating, and the low productivity of the agricultural sector (where most of the poor earn their incomes) remains a drag on the country’s growth performance. The persistence of these policies owes much to the important role of elites in Philippine politics and society. Although a number of reforms have been undertaken in recent years, it will likely take some time before a full liberalization of the economy will erase the effects of past policies.
AlonzoRuperto P.1994 “A Review of Equity-Oriented Investment Criteria and the Medium-Term Public Investment Program” in Understanding Poverty and Inequity in the Philippinesed. by Intal and Bantilan (Manila: National Economic and Development Authority).
BalisacanArsenio M.1995 “Urban Poverty in the Philippines: Nature, Causes and Policy Measures” Asian Development ReviewVol. 13 pp. 117-52.
BalisacanArsenio M.1991Deregulation in Philippine Agriculture: Has it Benefited the Rural Poor? Working Paper91-06 (Manila: Research and Training Program for Agricultural Policy).
BautistaRomeo M.1987 “Production Incentives in Philippine Agriculture: Effects of Trade and Exchange Rate Policies” International Food Police Research Institute Research Report 59 (Washington: IFPRI).
BlejerMario I. and IsabelGuerrero1990 “The Impact of Macroeconomic Policies on Income Distribution: An Empirical Study of the Philippines” Review of Economics and StatisticsVol. 72 pp. 414-23.
BoyceJames K.1992 “Of Coconuts and Kings: the Political Economy of an Export Crop” Development and ChangeVol. 23 pp. 1-25.
DeiningerKlaus and LynSquire1996 “A New Data Set Measuring Income Inequality” World Bank Economic ReviewVol. 10 pp. 565-91.
Department of Labor and Employment1997Current Labor Statistics (Manila: DOLE).
Department of Labor and Employment1994Yearbook of Labor Statistics (Manila: DOLE).
Faculty of Economics at the University of the Philippines at Los Baños1994 “An Integrated Analysis of Past and Existing Policies and Programs Directed Toward Alleviating Poverty and Improving Income Distribution” in Understanding Poverty and Inequity in the Philippinesed. by Intal and Bantilan (Manila: National Economic and Development Authority).
IntalPonciano S.Jr.1994 “The State of Poverty in the Philippines: an Overview” in Understanding Poverty and Inequity in the Philippinesed. by Intal and Bantilan (Manila: National Economic and Development Authority).
International Monetary Fund1997International Financial Statistics Yearbook (Washington).
International Monetary Fund1996Government Finance Statistics Yearbook (Washington).
KarnowStanley1989In Our Image: America’s Empire in the Philippines (New York: Ballantine Books).
MackenzieG.A. and others1997The Composition of Fiscal Adjustment and Growth IMF Occasional Paper No. 149 (Washington: International Monetary Fund).
ManasanRosario G.1986 “Impact of BIO Incentives on Rate of Return, Factor Prices and Relative Factor Use” Staff Paper 86-01 (Manila: Phillippine Institute for Development studies).
National Statistical Coordination Board1996 “Philippine Poverty Statistics” (Manila: NSCB).
National Statistics Office Republic of the Philippines (various years) Family Income and Expenditure Survey (Manila: NSO).
OtsukaKeijiro and others 1992 “Green Revolution, Land Reform and Household Income Distribution in the Philippines” Economic Development and Cultural ChangeVol. 40 pp. 719-41.
PsacharopoulosGeorge1994 “Returns to Investment in Education: A Global Update” World DevelopmentVol. 22 pp. 1325-43.
VirtucioFelizardo K.Jr.1994 “A Household Survey of the Lowest Income Groups in the Philippines” in Understanding Poverty and Inequity in the Philippinesed. by Intal and Bantilan (Manila: National Economic and Development Authority).
World Bank1997World Development Indicators. Available on CD-ROM (Washington: The World Bank).
World Bank1996 “A Strategy to Fight Poverty: Philippines” Working Paper 16118 (Washington: The World Bank).
I thank John Hicklin, Subir Lall, Sandy Mackenzie, David Nellor, Michael Sarel, Amrit Singh, and Victoria Summers for comments on earlier drafts of this paper. I also thank Ofie Templo and other staff at the Philippine National Economic and Development Authority for providing data and comments. All errors are my own.
Defined as percent of families living below the poverty line.
Data for the Philippines come from the Family Income and Expenditure Surveys for 1985 and 1994. In 1985, the richest 10 percent of Filipinos earned 18 times as much as the poorest 10 percent.
However, because of changes in coverage and methodology of surveys, especially for the earlier years, caution should be exercised in comparing results from year to year.
National Statistics Office (1994).
Because of differences in methodology and coverage, it is difficult to make comparisons of Gini ratios across countries. For example, data for Indonesia are based on expenditure rather than income.
This is generally thought to be because expenditure does not fluctuate with changes in transitory income, and because the marginal propensity to consume tends to decline with income.
National Statistics Office (1994 and 1985).
The World Bank (1996).
Data in this section come from National Statistical Coordination Board (1996). Subsistence poverty is virtually nonexistent in Metro Manila (0.7 percent of families in 1994), but about one-third of all families are below the subsistence level in Bicol and Central Mindanao.
Data in this section come from the National Statistics Office (1994).
For example, it could be the case that when families are poor, children are less likely to move out or grandparents and other relatives are more likely to move in with their children and grandchildren.
National Statistics Office, Household Surveys Section, Current Labor Statistics, Department of Labor and Employment.
This finding may be due to the fact that household assets were included in the regression. If assets and education levels correlate positively, omitting assets from the regression might have led to a finding that education has a significant, positive impact on earnings.
Bautista (1987) provides evidence on the effect of exchange rate overvaluation on agricultural incentives.
Data come from the Philippine authorities, and are drawn from the Labor Force Survey of the National Statistical Office, the Quarterly Survey of Establishments (NSO), the Annual Survey of Establishments (NSO), and the National Statistics and Coordination Board.
According to the World Bank (1996), among the most important of these are limits on the area farmers can plant with bananas, import bans on seeds, restrictions on importing cattle feeder stock, export bans on buntal and ramie planting materials, a ban on the slaughter of carabaos, and export restrictions on unendangered animals and their products.
In Indonesia, the yield is 77 tons per hectare.
For a discussion of the role of fiscal policy in promoting economic growth, see Mackenzie and others (1997).
International Monetary Fund (1996). To some extent, the low share of education relative to total expenditure reflects the very high level of interest expenditure in the Philippines.
In terms of regional comparisons, it should be noted that as of 1993, the VAT in the Philippines actually contributed a substantially lower proportion of total tax revenues than that in Indonesia, and an equivalent proportion to that of the Thai VAT.
In fact, the tax reform could lead to higher total tax payments from some individuals for a variety of reasons. First, if the wage elasticity of labor supply is high, workers could increase their labor supply sufficiently to raise their total tax obligations even at the lower tax rate. Second, the lower marginal tax rate reduces the incentive to engage in tax avoidance strategies, which could offset a substantial portion of the revenue theoretically lost from the cut in marginal tax rates. Finally, at lower tax rates workers may choose to take more of their income in the form of taxable wages rather than tax-free benefits or deferred compensation. The proposed introduction of a fringe benefits tax would also have this effect on income tax revenues.
Although the recent depreciation of the peso certainly increased the real cost of some dollar loans in peso terms, in fact in some cases borrowers were reportedly required to enter into forward contracts for their dollars at the time the loans were granted, shifting the currency risk back to the lenders.