POVERTY IN THE PHILIPPINES – AN OVERVIEW

К оглавлению
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 
85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 
136 137 138 139 140 141 142 

The past quarter century has seen the Philippines lagging behind most of the

major East Asian countries in practically all aspects of economic and social

development. The chapters in Balisacan and Hill (2003) discuss in detail

broad development issues and concerns in the Philippines. The country’s

average economic growth was only slightly higher than its population

growth, which was, and continues to be, comparatively high by most Asian

standards. Further blunting the impact on poverty of whatever growth that

occurred has been its persistently high level of economic inequality. Indeed,

not a few observers characterize the country’s social structure, especially with

reference to land distribution, as a largely Latin American rather than an

East Asian feature (see, for example, Hayami, 2001 and Ranis and Stewart,

1993). The co-existence of huge plantations and industrial enclaves owned

by a few families and of several million semi-subsistence small farmers and

vast colonies of urban poor depicts a highly inequitable Latin American-like

society that is quite uncommon elsewhere in East Asia.

An almost regular pattern of boom and bust has characterized the

Philippine economy during the period from 1960 to 2002 (see Figure 6.1).

Bust and stagnation soon followed each episode of boom, fueled largely

by massive foreign borrowing and capital-intensive import-substituting

industrialization. The period also saw heavy government regulation of

the market economy, as well as political instability, natural disasters, and

major shocks in global trade and fi nance. However, notwithstanding the

interruption in the late-1990s owing to the combined impact of the Asian

economic crisis and the El Niño phenomenon, the growth episodes since

the second half of the 1980s appear to have a fundamentally different

character from previous ones. Economic growth, albeit meager compared

with that in any of the country’s South East Asian neighbors, has taken

place in an environment of political stability, economic deregulation

and institutional reform. While domestic political squabbling and policy

coordination problems persisted, it could not be denied that the Philippines

at the beginning of the new millennium was closer to a market economy than

it had ever been in the past. One could ask: how have the poor benefi ted

from the growth process?

Table 6.1 provides estimates of three dimensions of poverty – incidence,

depth and severity – from 1985 to 2000.2 All the poverty indices show

signifi cant reductions during periods of relatively rapid growth of mean

expenditure (1985–88 and 1994–97). The highest three-year poverty

reduction was achieved during the ‘economic boom’ of 1994–97, when

real per capita expenditure rose by 21 per cent. But poverty also fell when

the growth of mean expenditure was negative (1991–94). Surprisingly

too poverty depth and severity increased even when the growth of mean

expenditure was positive (1988–91), though at a comparatively low rate. It

thus appears that the observed poverty changes are related to the growth

(and stagnation) of real mean consumption, while obviously also infl uenced

by other factors. Indeed, as Table 6.1 indicates, another ‘proximate’ cause

for poverty changes may well be the evolution in expenditure distribution.

After falling slightly to 0.40 in 1988 from 0.41 in 1985, the expenditure Gini

coeffi cient rose to 0.43 in 1991. It fell back to its 1988 level in 1994, only

to rise to 0.43 in 1997 and 0.45 in 2000. The same pattern emerged for two

other simple inequality indicators – the share of the richest 10 per cent and

poorest 20 per cent of the population in total expenditures. While these

changes are not spectacular, they had a considerable impact on aggregate

poverty, as shown in Balisacan (2003).

The evolution of poverty, inequality and average welfare (given by average

per capita expenditure) in the 1980s and 1990s may well be related also to

movements in price levels. Infl ation averaged 25 per cent in 1983–85. The rate

dropped from 18 per cent in 1985 to 9 per cent in 1988, possibly benefi ting

the majority of the poor, who tended to be fi xed-income earners or selfemployed

workers in rural areas. Infl ation surged once more to an average

of 15 per cent per year at the end of the decade. This was accompanied

by an increase in inequality. Infl ation decelerated to only 7.9 per cent per

–10

–8

–6

–4

–2

0

2

4

6

8

10

1962

64

66

68

70

72

74

76

78

80

82

84

86

88

90

92

94

96

98

00

2002

%

6000

7000

8000

9000

10 000

11 000

12 000

13 000

14 000

Pesos

GDP growth (%) Per capita GDP growth (%)

Per capita GDP (1985 Pesos)

Figure 6.1 Growth of Gross Domestic Product, 1960–2002

year during 1992–94, 7.7 per cent during 1995–97, and 6.9 per cent during

1998–2000. As shown elsewhere (Balisacan, 1995), high infl ation during a

period of low growth increases aggregate poverty. Particularly vulnerable

to commodity (particularly food) price increases are the numerically large

small-scale agricultural producers and landless workers who are net buyers

of food.

In an earlier study (Balisacan, 2003), it was shown that the national poverty

profi le was quite robust for periods of relatively high mean consumption

growth, such as for 1985–88, 1994–97, and 1997–2000, regardless of the

choice of poverty indicator or of aggregation procedure. The same thing,

however, cannot be said for periods of low or negative mean consumption

growth. Moreover, the same study showed that for the entire 1985–2000

period, the increase in inequality reduced the impact of growth on poverty,

but this effect was proportionately small relative to the reduction of poverty

attributable to consumption growth. It is thus the changes in real mean

consumption, rather than changes in its distribution, that have mainly

contributed to the observed changes in poverty in recent years. This result

runs counter to the common claim in policy dialogue in the Philippines that

recent episodes of growth have not benefi ted the poor.

Table 6.1 Poverty and inequality, 1985–2000

1985 1988 1991 1994 1997 2000

Average per capita

expenditure1 (at 1997 prices) 17 197 18 926 20 049 19 600 23 694 22 865

Dimension of poverty

Incidencea (%) 40.9 34.4 34.3 32.1 25.0 27.5

Depthb (%) 13.2 10.1 10.6 8.7 6.4 7.2

Severityc (%) 5.8 4.2 4.5 3.4 2.3 2.7

Inequality

Gini 0.412 0.400 0.428 0.397 0.427 0.450

Share of richest 10% 26.2 24.2 26.9 24.9 28.4 28.2

Share of poorest 20% 8.6 9.1 8.4 9.7 8.8 8.8

Notes:

1 Adjusted for provincial cost-of-living differences. Reference province is Metro Manila.

a Incidence is the headcount ratio.

b Depth is the poverty gap.

c Severity is the squared poverty gap.

Source: Balisacan (2003), based on Family Income and Expenditure Survey data.

Table 6.2 shows poverty and inequality estimates for both the urban and

the rural sector.3 A high mean consumption disparity between urban and

rural areas is apparent. Mean consumption in urban areas was nearly twice

that in rural areas, rising signifi cantly during the high-growth periods of

1985–88 and 1994–97 for both sectors. Correspondingly, all poverty indices

declined signifi cantly during both periods. The direction of inequality for

both sectors also generally followed the overall pattern reported in Table 6.1.

Clearly, poverty reduction during high-growth periods was quite broadly

based – that is, taking place in both urban and rural areas.

Table 6.2 Poverty and inequality in urban and rural areas, 1985–2000 (%)a

1985 1988 1991 1994 1997 2000

Urban

Incidence 21.7 16.0 20.1 18.6 11.9 13.2

Depth 5.9 3.8 5.7 4.4 2.6 3.0

Gini 0.41 0.39 0.42 0.39 0.43 0.41

Share of richest 10% 26.9 24.5 28.0 25.7 31.1 30.7

Share of poorest 20% 7.5 8.0 7.0 8.2 6.7 7.0

Rural

Incidence 53.1 45.7 48.6 45.4 36.9 41.3

Depth 17.8 14.0 15.6 13.0 9.8 11.3

Gini 0.35 0.35 0.36 0.34 0.35 0.41

Share of richest 10% 23.2 23.1 23.9 23.1 24.3 24.3

Share of poorest 20% 9.6 9.7 9.8 10.7 10.3 10.7

Note: a Poverty and inequality estimates are based on per capita consumption expenditure

adjusted for provincial cost-of-living differences. Poverty lines employed to calculate poverty

indices are fi xed in terms of living standards.

Source: Balisacan (2003).

Table 6.3 provides poverty estimates across regions of the country

from 1985 to 2000, as well as the importance of each region to national

poverty. While considerable variation exists, Metro Manila consistently

had the lowest poverty, and Bicol, Western Mindanao and the Visayas the

highest. In 2000, poverty incidence in Bicol was nine times higher than in

Metro Manila. Some signifi cant re-rankings also occurred, such as Central

Mindanao becoming the fourth poorest region in 2000 when it was only the

ninth poorest in 1985. Even more signifi cant is the differential evolution

of poverty over time. In two regions, Central and Western Mindanao,

poverty – in all three dimensions – was higher in 2000 than in 1985. Toward

the close of the 1990s, these two regions, particularly Western Mindanao,

were at the center of violent confrontations between the military and

armed dissidents.

Not a few observers contend that income disparity between urban and

rural areas, across regions and between economic sectors, is at the core of

the poverty problem. Mean income in urban areas was at least twice that

in rural areas during the 1980s and 1990s. Metro Manila, which accounted

for about 14 per cent of the population, had the highest mean income. In

2000, its mean income was more than twice the national average or about

three to four times the mean income for Bicol and Eastern Visayas. Except

for Bicol and Cagayan, mean income for the Luzon regions was higher than

for most of the regions in Visayas and Mindanao.

In the 1990s, average income in agriculture, where the large majority of the

poor are located and where about 40 per cent of the labor force are employed,

was much less than one-half of those in virtually all other sectors (except

construction). It is thus claimed that the key to winning the war against

poverty is to focus development priorities on agriculture, so as to raise

incomes in that sector vis-à-vis those in other sectors of the economy.

If indeed spatial and sectoral income disparities are at the core of the

poverty problem in the Philippines, then policy reforms aimed at reducing

these disparities have to be central elements of the country’s poverty

reduction program. This may also promote effi ciency goals: important

dynamic externalities can arise from targeting by area or according to

sector-specifi c characteristics (Bardhan, 1996; Ravallion and Jalan, 1996).

Investment in physical infrastructure (such as roads, communications and

irrigation) in backward areas, or in the rural sector in general, may improve

the productivity of private investment, infl uence fertility through its effect

on labor allocation and educational investment decisions, promote the

development of intangible ‘social capital’ (in the form of social networks, peer

group effects, role models, and so on), and mitigate erosion in the quality of

life in urban areas through its effect on rural–urban migration decisions.

However, if disparity in incomes and human achievement within each of

the regions or areas of the country were itself the major problem, a different

approach to poverty reduction would have to be found. It is possible, for

example, that systematic differences in levels of human capital between lowand

high-income groups within a geographic area translate into considerable

differences in earning opportunities between these groups within each area.

In this case, the policy prescription to reduce overall income inequality

and poverty would have to involve expanding the access of low-income

groups to basic social services, technology and infrastructure, regardless

of their location.

Table 6.3 Poverty incidence by region, 1985–2000 (%)a

Contribution

to total

1985 1988 1991 1994 1997 2000 poverty, 2000

National Capital 11.6 9.5 5.9 5.6 3.5 5.5 2.9

Ilocos 33.1 27.6 27.3 26.5 21.0 19.4 4.6

Cagayan Valley 44.9 39.7 42.2 39.8 29.5 29.7 4.4

C. Luzon 19.1 15.3 15.4 24.3 13.2 16.1 5.8

S. Luzon 35.4 31.7 22.9 28.6 19.6 19.5 10.3

Bicol 67.0 60.9 62.2 50.2 45.6 53.3 14.4

W. Visayas 49.4 34.4 31.6 34.5 21.8 28.1 8.1

C. Visayas 66.5 55.2 53.2 42.8 35.2 39.4 10.2

E. Visayas 59.3 53.7 54.4 51.5 50.6 46.8 8.0

W. Mindanao 52.5 43.8 44.0 53.7 44.6 56.0 10.5

N. Mindanao 52.6 41.4 54.2 37.9 29.9 30.1 6.2

S. Mindanao 51.8 43.7 53.9 30.7 27.8 25.8 7.0

C. Mindanao 35.8 30.1 42.3 39.8 32.9 39.9 7.6

Note: a The regional classifi cation of provinces and cities is kept fi xed to that existing in

1985.

Source: Balisacan (2003).

A parametric procedure, such as that suggested in Fields (2002), might be

useful for exploring systematically the relative contribution of location- and

household-specifi c attributes to the observed variation in household income.

Specifi cally, one can estimate a standard set of income-generating functions

and use the parameter estimates to calculate the relative contribution of each

factor to differences in household income. Table 6.4 summarizes the results

of such an exercise for the six survey data covering the 1985–2000 period,

giving the proportions in the total variance of (log) income accounted for by

location- and household-specifi c attributes. Together all variables included

in the regression explain 55–58 per cent of the variance of log-incomes for

the six survey years (Balisacan and Piza, 2003).

Household composition and the household head’s attributes, most

especially educational attainment, accounted for one-third of the total

variance of (the log of) income. Educational attainment contributed about a

fi fth of the observed variation in income. After controlling for the effects of

other factors, location (regional and urban location) contributed only about

15 per cent of the observed variation in income. Economic attributes (sector

of employment and class of worker) represented only a small amount,

roughly 10 per cent, of the total variance. This suggests that, by and large,

it is the differences in income levels within a sector or location, rather than

differences in mean income levels between sectors or locations, that account

for a signifi cant proportion of the variation in household income nationally,

and this weakens the case for the type of location targeting discussed below.

This conclusion holds for each of the survey years.

Table 6.4 Relative contribution of factor in total variance of (log) income (%)

1985 1988 1991 1994 1997 2000

Household attributes 34.4 33.6 33.1 33.4 34.8 33.6

Family size 10.2 9.5 9.5 9.5 8.9 8.3

Household type 1.5 1.1 1.3 1.2 1.3 1.5

Child dependency ratio 1.3 1.5 1.5 1.8 1.6 1.4

Employment ratio –0.3 0.0 0.0 0.2 0.2 0.0

Spouse employed 0.2 0.4 0.5 0.5 0.8 0.6

Experience 0.5 0.7 0.6 0.3 0.5 0.4

Gender 0.0 0.1 0.1 0.1 0.1 0.1

Marital status 0.6 0.8 0.7 0.5 0.6 0.7

Education 20.4 19.5 18.9 19.3 20.8 20.6

Economic sector 5.4 7.6 8.7 9.6 9.9 10.1

Class of worker 4.1 4.5 5.4 5.5 6.2 5.4

Sector of employment 1.3 3.1 3.3 4.1 3.7 4.7

Location 15.8 14.1 15.8 15.5 13.1 14.3

Urban 3.1 3.6 2.7 2.8 3.9 4.4

Region 12.7 10.5 13.1 12.7 9.2 9.9

Residual 44.4 44.7 42.4 41.5 42.2 42.0

Total 100.0 100.0 100.0 100.0 100.0 100.0

Source: Estimates of income-generating functions based on unit record data from the Family

Income and Expenditures Survey (various years) of the National Statistics Offi ce; see Balisacan

and Piza (2003).