OVERALL EFFECTIVENESS OF POVERTY TARGETING

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 

To date there is limited evidence on the overall effectiveness of poverty

reduction programs in PRC and here we survey the main contributions. The

challenge of such assessments is to isolate the effect of poverty programs,

since progress or lack of progress in reducing poverty may refl ect factors other

than the targeting programs themselves. Some have argued, for instance, that

poor areas stood to gain more from market and commercialization reforms

since the planning system forced them into production patterns that went

against their comparative advantage to a greater extent than in richer areas

(Lardy, 1983). To back this up there is some evidence that income growth

in poor counties was greater than in non-poor counties in some regions,

allowing for other factors (Park et al., 1996; Tong et al., 1995). A number

of studies have looked specifi cally at the impact of poor county targeting

and we survey the more important of these.

Jalan and Ravallion (1998b) assess whether being located in an offi cially

designated poor county affects growth in household expenditures, controlling

for geographic externalities and other community variables that are likely

to determine income growth. Utilizing National Bureau of Statistics panel

data (1985 to 1990) on households in four southwest provinces (Guizhou,

Yunnan, Guangxi and Guangdong), they fi nd that living in a national poor

county increases consumption by 1.1 per cent per year above what would be

expected, although this gain is offset by a growing divergence in consumption

relative to non-poor counties, due to the unfavorable characteristics of poor

counties. Using this consumption growth as a measure of benefi ts the rate of

return on poverty investments is estimated to be 12 per cent.11 The authors

themselves point out that this may overestimate the poor county program’s

effect, since some public expenditure may not be included, funds may be

used for consumption rather than investment purposes, and the variables

used to control for area characteristics may omit factors that give poor

counties an advantage in growth.

Rozelle et al. (1998) examine the effectiveness of targeted poverty

interventions across economic sectors in 43 poor program counties of

Shaanxi province during 1986–91. The authors adopt three separate sectoral

growth models in which the rate of growth of output per capita is a function

of current-year poverty expenditures, poverty expenditures lagged one year,

government expenditures per capita (to control for other investments),

rural income per capita (to control for private investment), human capital

(represented by the share of the labor force that had graduated from middle

school in 1985), lagged output (to control for the initial size of the sectors),

and county and time dummy variables (to control for county characteristics

and time-related effects), as well as population density (as a proxy for the

relative abundance of labor). In the agricultural growth equation they also

include as a regressor changes in the availability of agricultural land, while

in the growth equation for state-owned industry, they include as regressors

fi xed investment in enterprise assets in both current and lagged form.

The results reveal that for the sample of national and provincial poor

area counties, targeted poverty funds allocated directly to households for

agricultural activities have a signifi cant and positive effect on growth. In

contrast, investments in township and village enterprises or county stateowned

enterprises do not have a discernible effect on growth. In a more

disaggregated part of the study, investments in agricultural infrastructure

(such as terracing or soil leveling) do not by themselves positively affect

growth in agricultural output. These results suggest that the poverty funds

targeted directly at households have a positive growth effect. The study is

based on data from only one province, and similar work using data from

other provinces is necessary to have confi dence that the results for Shaanxi

can be generalized to other parts of the country. Also an important source

of poverty funds, namely the food-for-work funds, are excluded from

the estimation, which is most likely to affect the impact of infrastructure

investment.

For another province, Sichuan, Zhang et al. (2002) analyze the impact of

participation in national and provincial poor county programs on income

growth. They classify all counties in Sichuan into poor program counties,

non-program poor counties and non-poor counties. Growth of income per

capita in poor program counties was positive and exceeded the very small

rise for non-program poor counties; however, it did not keep pace with

the increase in the non-poor counties. In terms of statistical signifi cance,

growth rates of poor program counties were statistically indistinguishable

from those of non-poor counties, whilst non-program poor counties had

signifi cantly slower growth than either of the other groups.

The authors use a regression model to identify the determinants of growth

and examine the impact of the poverty programs using data from 1990 to

1996 for 177 counties. The growth of income is regressed on independent

variables representing resource endowment, the economic structure of the

county, investment (by type) made through the fi scal system (which includes

some but not all of poor area expenditures) and program participation.

They fi nd investment in agriculture, health, education and electrifi cation

positively affects growth, though the effect on growth of other investments

(in ‘other infrastructure’) is not apparent. The presence of a poverty

program positively impacts on growth, keeping the relative growth of poor

program counties from falling as much as that of poor, non-designated

counties. After accounting for endowments, structure, and initial level of

income, poor program counties grew more slowly than non-poor counties

(by about 3 per cent per year). However, this slower income growth was

still faster than growth for non-program poor counties, which, allowing

for the county characteristics captured here, experienced growth nearly 5

per cent slower than non-poor counties. An important difference between

Zhang et al. (2002) and Rozelle et al. (1998) is that the former uses data

from all counties in Sichuan, while the latter only works with designated

poor counties. This allows the use of non-designated poor counties as a

comparator group, although incomplete coverage of poverty expenditures

is a limitation common to both studies.

Using a different methodology, Fan et al. (2002) develop a simultaneous

equation model to estimate the effects of government expenditure on

output and poverty through different channels. They conclude that poverty

funds matter for growth and poverty alleviation, but not nearly as much

as investments in other sectors. The study, using provincial data for the

past 26 years between 1970 and 1995, shows that government spending on

production-enhancing investments, such as agricultural R&D, irrigation,

rural education and infrastructure (including roads, electricity and

communications) have all contributed not only to growth, but also to poverty

reduction. One of the most striking results is that large parts of the poverty

effect are realized through improved access to rural non-farm employment.

Government anti-poverty loans specifi cally targeted for poverty alleviation

have the smallest impact on poverty reduction of any of the investment

programs considered. The strength of the study lies in its adoption of a

simultaneous equation model and the use of a provincial panel data set.

However, of the poverty-targeting funds only subsidized loans are taken

into consideration. Also methodologically there may be a bias against

the impact of poverty loans, as they do not enter into the simultaneous

‘production equations’ and therefore do not generate feedbacks in the way

that infrastructure and other non-poverty investments do.

One of the most detailed and disaggregate approaches is provided by

the author and others using Ministry of Agriculture county level data,

for all counties where relevant data exist (Park et al., 2002). To estimate

the impact of poverty reduction policy on average income growth in poor

counties, growth in a county’s rural income per capita from one period to

another is modeled as a function of the county’s status as a designated

poor county at the beginning of the period, initial income per capita, other

initial characteristics (principally grain production per capita), county timeinvariant

characteristics, and prefecture time-varying factors. A panel is

constructed from data for each county for four time periods: 1981–85,

1985–89, 1989–92, and 1992–95. The fi rst period pre-dates the poor county

programs, and the fi rst poor county designations occurred during the second

and third periods, with additional designations being made during the

fourth period. Information on growth rates before the poverty program

began makes it possible to identify the effects of poor county status while

also controlling, through a county dummy variable, for unobservable

characteristics that have persistent effects on growth. The coeffi cients on the

poverty status variable can be interpreted as the program effect allowing for

all other factors. The results suggest that poverty designation and the funds

that come with it have a positive growth effect, allowing for other factors.

Rural household net income per capita is found to increase 2.2 per cent

faster annually in poor counties than in non-poor counties over 1986–92,

and 0.9 per cent faster over 1992–95, other things held equal.12 This means

that although growth was nonetheless lower in poor areas, their relative

disadvantage due to their starting point and characteristics was reduced

by the poverty funds. The effects are larger than those found by Jalan and

Ravallion (1998b) for the period of 1985–90 in four southwest provinces,

discussed above. Furthermore, these fi gures are based on national data, not,

as with earlier studies, on data from one province alone.

Using this estimate of program impact on rural income growth, it is

possible to estimate the rate of return on poverty investment. In real terms,

the poverty spending included here fell during 1985–92 and then recovered

to about its initial level, averaging 9.5 billion yuan per year (in 1995 yuan),

equivalent to 89 yuan per person or 14 per cent of rural income. Based

on the 2.2 per cent impact on incomes, the poverty program on average

increased rural income by 13.8 yuan per person per year. This suggests a rate

of return of 15.5 per cent, somewhat higher than the 12 per cent estimated

by Jalan and Ravallion (1998b). For the 1992–95 period, the approximate

doubling of the program’s coverage reduced spending per capita to 55 yuan

and the implied rate of return is 11.6 per cent.

These results of Park et al. (2002) are open to different interpretations.

Total poverty expenditure may be considerably greater than the central

government funds included in this study, which do not cover items like the

administrative costs of programs, matching or supplementary funds provided

by local governments and international donor funds. Some argue that the

total of such spending is much greater than offi cial central government

poverty alleviation funds (Xie, 1994). Indirect evidence of low repayment

rates on subsidized loans and suspected substitution effects, discussed above,

may make the relatively high rate of return surprising. Most critically the

results provide no evidence on the distribution of benefi ts within counties,

so high impacts do not necessarily benefi t the poor within poor counties.

Other factors, however, may bias the estimates downward, particularly if

targeted programs also benefi t poor counties not offi cially designated as

poor, as leakage will dilute the measured impact on targeted counties. This

will be reinforced if provincial governments shift other budgetary allocations

away from counties supported by national poverty alleviation funds.

Evidence from all of the above studies surveyed here suggests that poverty

programs in PRC have had a positive impact on household income and

poverty reduction in poor areas, although there is also evidence that the

impact from other non-poverty focused investment has been even greater.

There is also some evidence that during the 1990s the impact of targeted

poverty funds on poverty reduction decreased with the decrease of the

rural poor population and the increase in such funds, possibly because

of the worsening targeting problems and the diversion of some poverty

funds. Nonetheless a major omission of all studies is that due to the lack

of reliable poverty data at the disaggregate level within counties, none of

the above has managed to disaggregate gains to poor counties into those

to the poor and to the non-poor within the counties. It is clear that this

requires further research.