Public Investment and Poverty Reduction
Lessons from China and India
Growth in agricultural productivity, the rural non-farm sector and rural wages, which are the main sources of poverty reduction in both China and India, have been made possible by public investments in R&D, infrastructure (such as roads, power, irrigation, communication and education) and anti-poverty programmes. However, returns on these public investments, reckoned in terms of poverty reduction, vary drastically across different types of investment. The trade-off between agricultural growth and poverty reduction is generally small among different types of investments and between regions. Agricultural research, education, and infrastructure development have a significant growth impact as well as a large poverty reduction impact.
SUKHADEO THORAT, SHENGGEN FAN
B
The objective of this paper is to review and compare the evidence on the effects of public investment on agricultural growth and poverty reduction, and to draw lessons on how the Chinese and Indian governments can use public investment as an instrument to achieve their stated social objectives.
These two nations share some common features but also differ in other respects. Both are large countries and, coincidentally, underwent substantial political and economic changes at about the same time – in the late 1940s and early 1950s. India gained its independence in 1947 and initiated the experiment of planned economic development with the first Five-Year Plan in 1951. Following the 1949 communist revolution, China embarked on a new economic development path at the beginning of the 1950s. Thus, both countries experienced profound political and economic reforms in the past 50 years.
There are, however, at least three crucial differences that need to be recognised. First, following independence, India adopted a mixed economic system characterised by the presence of both the public and private sectors. Agriculture, which employs a major portion of the workforce and the population, is based on the principle of land belonging to the private domain. Similarly, a large portion of the industrial sector except for basic and key industrial activities has been in private hands. Thus, only a few critical industries along with a large part of primary services like education, health, and infrastructure such as irrigation, road and power are under the responsibility of the state. China, on the other hand, embarked on an economic development path with a socialist economic framework characterised at least until recently by a much larger ownership and control by the state over the agricultural and industrial resources.
Second, while the Indian state has pursued economic development and public investment policies within the framework and limitations of political democracy, China operated in a political framework, which gives the state relatively more freedom in deciding its investment priorities. Thus, compared to India, China has been functioning with a reasonable degree of control in terms of raising and allocating financial resources for public investment.
Third, China initiated the process of economic development with greater equality than India in terms of people’s access to income sources (land and non-land capital assets) and social needs (education and health). In these critical areas, the Indian population could derive relatively limited benefits from these measures, resulting in greater inequality relative to their Chinese compatriots with regard to access to land and non-land capital assets, as well as to basic services like education and health. All these similarities and disparities have had different impacts on the patterns of government spending, and therefore have led to different growth paths and poverty reduction outcomes.
Growth, Poverty and Government Spending
Growth and Poverty
Both China and India have experienced a remarkable drop in poverty levels. As a matter of fact, most of the reduction in the world poverty incidence over the last several decades has come from these two countries. It is also the case that together, China and India still account for more than 40 per cent of the total poor in the world today. Therefore, to reach the millennium goal of halving the global number of poor by 2015 largely depends on the performance of these two countries in poverty alleviation.
Reducing rural poverty has been the prime focus of the Indian government and has received a great deal of attention in the country’s economic agenda, beginning with the first Five-Year Plan adopted in 1951. However, progress has been highly uneven over time and space. The rural poverty rate, defined as the proportion of poor in the total rural population, fluctuated between 50 and 65 per cent in the 1950s to the late 1960s without any clear trend. It began to show a steady decline from the late 1960s until the 1980s, increased again to about 40 per cent between the late 1980s and the early 1990s, but experienced a steady decline thereafter, although the rate of this decline has slowed down during the last decade. Thus, over the last 50 years or so, rural poverty declined from about two-thirds of the rural population in the early 1950s to about one-third by the end of the 1990s [Planning Commission 2005].
This steady drop in rural poverty was strongly associated with agricultural growth, particularly in the “green revolution regions”, as well as with increased rural non-farm activities and wages, resulting from the massive public investments in agriculture and rural infrastructure like R&D, irrigation, roads, education and power implemented in the 1970s and 1980s. Government spending contributed to agricultural growth and thereby to poverty alleviation not only through increased income and lower food prices, but also by creating non-farm employment, which also increased wages. The lower decline in poverty during the past decade (as compared with the 1970s and 1980s), coincided with India’s liberalising economic policies, which neglected rural investment resulting in slower agricultural growth [Planning Commission 2005].
China achieved immense success in reducing rural poverty during the past two decades, despite the global slowdown in poverty reduction. The incidence of rural poverty decreased sharply from 32.9 per cent in 1978 to 15.1 per cent in 1984, and dropped significantly to 3 per cent in 2000 [Ministry of Agriculture of China, various issues]. Even if the international benchmark of one dollar a day is used, China’s success in cutting poverty is still remarkable with poverty declining from 64 per cent in 1981 to 17 per cent in 2001 [Chen S and Ravillion 2004].
Contributing to this substantial achievement were a series of policy and institutional reforms, along with the promotion of equal access to social services and production assets, and public investments in rural areas. Yet, as China’s economy continues to grow, it is becoming harder to achieve further reduction in poverty while containing inequality. How the government can better design its policies, particularly public investment policy, to reduce the number of poor and narrow regional inequality without sacrificing growth is a hotly debated issue in both academic and policy circles.
Government Spending
Both countries have experienced erratic patterns in their total government spending as well as their spending earmarked for agriculture and rural areas. China and India dramatically increased their investments in agriculture in the 1950s, 1960s and 1970s, with the aim of solving food security problems. But both countries reduced their spending in agriculture either in relative (in the case of China) or in absolute (in the case of India) terms in the 1990s in response to macroeconomic reforms. Only until very recently, did both countries commit to allocate more resources to the agricultural sector and rural areas.
In the early years of economic planning in the 1950s and 1960s, public investment in agriculture accounted for almost half of the total investment in rural areas in India. Government spending in agriculture based on the CSO series (with base year 1993-94) increased at a per annum rate of 2.61 per cent from 1960-61 to 1969, compare with 9.50 per cent per annum from 1974-75 to 1980-81. Thus, public spending in agriculture grew three times faster during the 1970s than it did in the 1960s. During the 1980s however, public spending in agriculture decelerated at a rate of
3.89 per cent per annum, which continued in the 1990s although at a lower negative rate of 0.74 per cent per annum. Thus, the 1980s and the 1990s were characterised by declining public investment in agriculture with some recovery in the 1990s. The decline was much sharper when agriculture and allied activities are considered jointly, followed by irrigation and power for agriculture uses. It is this continuous declining trend, which has caused great concerns among policy-makers and planners with regard to the future growth of agriculture and its impact on poverty.
Further the pattern of government spending is such that with the exception of health the relative importance of overall spending across sectors has not changed much over the last several decades. For example, in 1966, the shares of education, health, irrigation, power, and roads were 15 per cent, 8 per cent, 5 per cent, 2 per cent and 3 per cent, respectively. In 1999, the corresponding shares were respectively 17 per cent, 4 per cent, 4 per cent, 2 per cent and 4 per cent. The only significant change is the rapid decline in the share of health expenditures.
Similar to India, over the past several decades, Chinese government spending in rural areas has experienced dramatic changes. Right after the foundation of the country in 1949, China’s investments in agricultural research were minimal, but grew rapidly in the 1950s. Growth in the 1960s was relatively small due to the three-year natural disaster (1959-61) and the Cultural Revolution (1966-76). Agricultural research investment increased steadily during the 1970s, but slowed down during the 1980s, increasing by only 2 to 3 percentage points during the entire 10-year period. In the 1990s, agricultural research expenditures began to rise again, largely due to government efforts at boosting grain production through science and technology.
The Chinese government gave top priority to irrigation immediately after 1949. In 1953, the government spending on irrigation was 10 times larger than the corresponding amount spent in agricultural research. Investment in irrigation continued to increase until 1966. Under the commune system, it was relatively easy for the government to mobilise a large amount of rural labourers in large irrigation projects. As a result, more than 10 million hectares of land were brought under irrigation during that period. From 1976 to 1995, investment in irrigation increased very little and even declined over the 1976-1989 period. During this period, there was no increase in irrigated areas in Chinese agriculture production. By 1989, irrigation investment was only 44 per cent of the 1976 level. In response to the grain shortfall and large imports in 1995, the government increased investment in irrigation sharply from 1997 to 1999. But further expansion would be difficult because of the competing industrial and residential uses of water. As a result, the returns to investment in irrigation may decline in the future.
The education level of the Chinese population was one of the lowest in the world four decades ago. In 1956, less than one-half of the children in the age group eligible for primary and secondary school education were enrolled. Since 1978, China has promoted the education policy of the “nine-year compulsory schooling
Economic and Political Weekly February 24, 2007 system”, which requires all children to attend school for at least nine years to finish both the primary and junior middle-school programmes. But the policy was never seriously implemented, particularly in rural areas. In 1986, an education law of nine years compulsory education was formally issued. By 2000, the enrolment ratio of school-aged children had risen to over 98 per cent, and the percentage of primary school entering junior high school to 85 per cent in rural China.
Consequently, labour quality has improved substantially over the past two decades, with a decline of the illiteracy rate of agricultural labour from 28 per cent in 1985 to 10 per cent in 1997. This improved human capital in rural areas provided a great opportunity for farmers to use modern farming technology, and to engage in non-farm activities in both rural township enterprises and urban industrial centres.
Despite these successes, the government investment in education is still not sufficient. In terms of expenditures the government has spent roughly 2.6 per cent of the total national GDP on education, which is lower than in most developing countries with exceptions such as Bangladesh, Indonesia and Myanmar. In particular, many poor were not reached by the government efforts [Fan S et al 2002a].
For the past several decades, China has given higher priority to electricity than to road development in its investment portfolio. Investment in power has increased 90-fold. The percentage of villages that have access to electricity was 98 per cent in 1998, and more than 97 per cent of the households have electricity connections. During most of the pre-1980 period, growth in government investment in telecommunication was very slow, increasing from 166 million yuan (measured in 1990 prices) in 1953 to only 738 million yuan in 1980. However, large-scale development happened in the last several years when the number of rural telephone sets increased from 3.4 million in 1992 to 51.7 million in 2000. This is largely a result of both public and private investments in the sector. From 1989 to 2000, public investment in telecommunication alone increased more than 20-fold.
In 2000, rural education spending accounted for 33 per cent of total expenditure in rural areas. Irrigation was next, accounting for 30 per cent. Investment in rural infrastructure took about 33 per cent of total government spending in rural areas, with 15 per cent for rural power, 5 per cent for rural roads, and 14 per cent for rural telecommunications. Agricultural research accounted for only a small fraction (2.2 per cent) of total government investment in rural areas.
Public Investment and Poverty Linkages – Conceptual Framework
There have been many attempts to measure the impact of public investment on growth and on rural poverty reduction. A significant feature in the literature on public investment and rural poverty is that most of the previous studies have considered only one type of government spending (or investment). This approach makes comparison across the relative returns to both growth and poverty reduction of different type of investments difficult. Moreover, most of these empirical studies used a single equation approach. There are at least four disadvantages associated with this method. First, many poverty determinants such as income, production or productivity growth, prices, wages and non-farm employment are generated from the same economic process as rural poverty. In other words, these variables are also endogenous variables, and ignoring this characteristic leads to biased estimates of poverty effects. Second, certain economic variables affect poverty through multiple channels. For example, improved rural infrastructure will not only reduce rural poverty through improved agricultural productivity, but also affect rural poverty through improved wages and non-farm employment. It is very difficult to capture these different effects in a single equation approach. Third, including only one type of public investment in estimating poverty reduction will lead to upward bias on the impact of that particular investment. Finally with the single equation approach, it is difficult to rank the effects of different types of investment on both growth and poverty reduction.
In order to systematically assess the impact of different types of public investment on both agricultural growth and poverty reduction, several studies conducted at the International Food Policy Research Institute (IFPRI) have tried to estimate simultaneous econometric systems that consist of many equations, using pooled time-series and cross-regional data. These studies differ from previous ones in several aspects. First, most types of public investment and expenditures are included in the assessment to avoid biased estimates of returns when only one single type of investment (e g, agricultural research) is considered, and to compare and rank returns of various types of investment. Second, the model has the ability to identify different channels through which government investments impact growth, inequality, and poverty. Understanding these different effects provides useful policy insights to improve the effectiveness of government’s poverty alleviation strategies. Lastly, the model permits calculation of economic returns measured by the number of poor people raised above the poverty line for additional units of expenditure on different items.
A distinguishing feature of the case studies conducted at IFPRI is the use of a simultaneous multiple equations system. In the system, rural poverty is modelled as a function of growth in agricultural production, changes in rural wages, growth in rural non-farm employment, and changes in agricultural prices. Increases in agricultural production or wages should reduce poverty. Improvements in relative prices for agriculture (a domestic terms of trade variable) are expected to hurt the poor in the short run because they increase food prices. In the long run, however, higher agricultural prices may stimulate farmers to invest more in technology and infrastructure, thereby increasing agricultural production and reducing rural poverty.
Agricultural production is modelled as a function of conventional inputs such as labour, land, fertiliser, machinery, and public investment variables such as the use of high-yielding varieties, public irrigation, roads, electrification, and education.
Rural wages and no farm employment are modelled as functions of growth in agricultural production as well as public investment variables. Numerous studies have shown the important linkage between agricultural and non-agricultural growth. Ignoring the effect of public investment on rural poverty through this linkage could lead to underestimation of the poverty reduction effects of government investment in agriculture. A similar downward estimation bias could arise from overlooking the impact of improved infrastructure and education on wages and employment.
Impact of Public Investment on Growth and Poverty Reduction
Fan, Hazell and Thorat (1999) used the system of econometric equations described in the previous section to identify the relative role of different type of government spending in agricultural growth and rural poverty reduction in India. The authors relied on state level data for 1970 to 1993 to conduct their analysis. The results from the model show that additional government expenditures on roads have the largest impact on poverty reduction as well as a significant impact on productivity growth (Table 1). Additional government spending on agricultural research and extension has the second largest effect on rural poverty, and the largest impact on agricultural productivity growth. The third largest impact on rural poverty reduction arises from additional government spending on education, largely as a result of the increases in non-farm employment and rural wages that it induces.
Additional irrigation investment has the third largest impact on growth in agricultural productivity but has only a small impact on rural poverty reduction, even after allowing for trickle-down benefits. Additional government spending on rural and community development, including on the Integrated Rural Development Programme contributes to reduction in rural poverty, but its impact is smaller than expenditures on roads, agricultural R&D, and education. Additional government expenditures on soil and water conservation and health have no impact on productivity growth, and their poverty effects through employment generation and wage increase are also small.
The results of an extended data period (1971-2000) throw additional light on the impact of public spending on rural poverty [Fan and Thorat 2003]. Fan and Thorat analysed the sources of poverty reduction over four different sub-periods: the 1960s, 1970s, 1980s and 1990s. Most categories of government spending and subsidies contributed to reducing rural poverty in the 1960s, 1970s, and 1980s. Investment in roads, fertiliser subsidies, HYV, power, and credit subsidies had the largest poverty reducing impact in the 1960s and 1970s. In the 1980s, roads, education, and irrigation investment had a relatively larger poverty reducing impact than any other type of government investment. While the impact of road and education spending on poverty reduction remained very strong in the 1990s, the effect of irrigation investment declined significantly. Agriculture research also showed a declining impact, but had a greater impact if it were defined more broadly.
Although informative, these two studies present some limitations as they analyse the impact of government spending on rural poverty at the aggregate level without bringing the regional variations into the picture. The latter is an important feature of India’s agricultural and economic development. Aggregate level analysis in some respect hides more than it reveals. In a third study, Fan and Hazell (1999) attempted to estimate the returns of various categories of public investment in different regions of India, including irrigated and two types of rainfed regions (Table 2). For every type of investment (particularly roads, canal irrigation, electricity and education), the highest marginal impact on agricultural production and poverty alleviation occurs in the high-potential or low-potential rainfed areas, while irrigated land ranks second or last. Moreover, many types of investments in the low-potential rainfed lands give some of the highest production returns, and all spending except education have the largest impacts on poverty. These results strongly support the hypothesis that more investment should now be channelled to the lessfavoured areas of India. This finding also implies that the lower marginal impact of irrigation, power and other expenditures in the aggregate level analysis (observed in the two studies discussed previously) is presumably due to the lower return in the high irrigated regions where the optimal effect has already been reached and therefore any additional investment would not yield higher returns. Conversely in the rainfed regions where the level of irrigation and power is low, additional spending on these types of infrastructure and on other supplementary inputs would generate much higher returns.
Turning to China, Fan et al (2002) adopted a simultaneous equations model to estimate the effects of different types of government expenditure using provincial-level data for 1970-97. The model not only ranks the marginal effects of public investments on growth, inequality, and poverty, but it also tracks various channels of investment along with their effectiveness. The latter is important because it enables policy-makers to focus on strengthening weak links in the poverty reduction chain. The results show that government’s production-enhancing investments, such as agricultural R&D, irrigation, rural education, and infrastructure (including roads, electricity, and telecommunications) contributed not only to agricultural production growth, but also to reduction of rural poverty and regional inequality (Table 3).
However, variations in the magnitude of the effects are large among different types of spending and across regions. Based on actual investments in 1997 and on the parameters estimated from the model, the authors calculated the marginal returns of various investments to (i) growth in agricultural and non-farm production; and (ii) reduction in rural poverty and regional inequality. These returns were calculated for the nation as a whole and for three different economic zones (coastal, central and western regions). Since the estimated returns are recent, they can serve as a direct input into the current policy debate.
Government expenditure on education had the largest impact in reducing rural poverty and regional inequality and had significant impact on production growth. Increased rural nonfarm employment was accountable for much of this poverty and inequality-reducing effect. Government spending on agricultural R&D substantially improved agricultural production. In fact, this category of expenditure had the largest impact on agricultural production growth, which is much needed to meet the increasing food demands of a richer and larger population. Benefits of agricultural production growth also trickled down to the rural poor. The poverty-reduction effect per unit of additional agricultural R&D investment ranked second after investment in rural education (Table 3).
Government spending on rural infrastructure (roads, electricity and telecommunications) had a substantial impact in reducing poverty and inequality, owing mainly to improved opportunities for non-farm employment and increased rural wages. Investments in irrigation had only a modest impact on agricultural production growth and an even smaller impact on rural poverty and inequality, even after trickle-down benefits were taken into account. A striking finding is the minimal impact of specifically targeted government anti-poverty loans. In fact, the poverty reduction impact of these loans was the smallest of all the types of government spending considered in the study.
Table 1: Returns to Rural Government Spending, India State-level Analysis
Returns in Rupees No of Poor Reduced Per Rupee Spending Per Million Rupee Spending
R&D 13.45 84.5 Irrigation 1.36 9.7 Roads 5.31 123.8 Education 1.39 41 Power 0.26 3.8 Soil and water conservation 0.96 22.6 Health 0.84 25.5 Anti-poverty programmes 1.09 17.8
Note: Marginal returns are calculated in 1993. Sources: Fan, Hazell and Thorat 1999.
Economic and Political Weekly February 24, 2007
| Disaggregating the analysis into different regions reveals that | access to agricultural land made the poverty reduction elasticity |
| for all types of government spending, returns to investments in | of agricultural growth much larger in China than in India. |
| poverty reduction were the highest in the west (the less-developed | (d) It is also the case that increased agricultural production may |
| region), while the returns in agricultural production growth were | result in downward pressure on agricultural (food) prices. This |
| the highest in the central region (the more developed region). | downward pressure is likely to offset part of the poverty reduction |
| Furthermore, investments in the western region led to the greatest | effect of agricultural growth in the case of China because farmers |
| reductions in regional inequality for all types of government | may be negatively affected by lower prices. In the case of India, |
| spending, while investments in either the coastal or central | landless poor may gain from lowered food prices because they |
| regions worsened existing large regional inequalities. | are usually net buyers of food. |
| Like India, China being a large country with vast and diverse | (e) Differences in the indirect impact of some variables in China |
| natural resource endowments, and different socio-economic | and India also need to be recognised in order to understand the |
| conditions across regions, a more disaggregated level analysis | different outcomes in poverty reduction in two countries. Similar |
| of public investment is necessary. Fan et al (2002b) in a county | factors seem to have promoted agricultural productivity in both |
| level analysis based on data from more than 2000 counties, | countries. In China, however, agricultural productivity does not |
| revealed the regional variations in the impact of government on | indirectly lead to increased landlessness since the ownership of |
| poverty reduction. The study divides China into eight different | agricultural land is regulated by the state. In India, on the other |
| types of rural economies using the following combined criteria: | hand, improvement in total factor productivity does induce |
| natural resources, agricultural production, rural and overall | landlessness and adversely impacted income distribution at least |
| economic development. Results suggest that most public capital | in the beginning of the green revolution. A similar difference |
| inputs show high returns in both the agriculture and the non | is observed with respect to factors associated with growth of rural |
| farm sectors for all regions. However, improved education provides | non-farm employment. In China, roads, education, rural telecom |
| the largest return to growth in both the agricultural and non-farm | munication and electrification have all contributed to growth of |
| sectors in the less-developed areas, such as the north-west and | non-farm employment. The most interesting feature, however, |
| south-west regions. Agricultural research also has the highest | is the strong and positive impact of agricultural productivity on |
| returns in the north-west border and the south-west regions, which | the growth of the non-farm sector (off-farm employment |
| are the least developed regions of China. | elasticity with respect to agricultural productivity is 0.716). On |
| Lessons from Experiences | Table 2: Marginal Returns to Public Investments in Rural India:District-level Analysis |
| The reviewed case studies on China and India show that public | Investment Irrigated High-potential Low-potential |
| investments in rural areas have contributed significantly to | Areas Rainfed Areas Rainfed Areas |
| agricultural growth and rural poverty reduction. Without these investments, agricultural growth and national economic growth would have been much slower, and there would be greater rural poverty in both countries. There are some lessons to be drawn from each other’s experience with respect to the impacts of public investment on poverty alleviation. | Production return in rupees per unit of investment HYV (Ha) 6 3 243 688 Roads (Km) 100,598 6,451 136,173 Canal irrigation (Ha) 938 3,310 1,434 Private irrigation (Ha) 1,000 (2,213) 4,559 Electrification (Ha) (546) 9 6 1,274 Education (Worker) (360) 571 102 Number of poor lifted out of poverty per unit of investment |
| HYV (Ha) 0.00 0.02 0.05 | |
| Factors affecting Poverty Reduction | Roads (Km) 1.57 3.50 9.51 Canal irrigation (Ha) 0.01 0.23 0.09 |
| The differential rates of poverty reduction between China and India come from the following factors: | Private irrigation (Ha) 0.01 (0.15) 0.30 Electrification (Ha) 0.01 0.07 0.10 Education (Worker) 0.01 0.23 0.01 |
| (a) Different growth rates in agricultural productivity, agricul- | Source: Fan and Hazell 1999. |
| tural wages, and non-farm employment. For instance, agricultural GDP grew by more than 4 per cent per annum in China from 1978 to 2002, compared with 2.9 per cent per annum in India from 1981-82 to 2002-03. Moreover, non-farm employment in | Table 3: Returns to Agricultural Research in China,Provincial-level Analysis Coastal Central Western Average |
| China has grown by more than 10 per cent per annum over the past two decades, while in India it grew by only 2 per cent during the same period. On the other hand, growth rates in rural regions were similar in the two countries. | Returns to agricultural GDP Yuan per Yuan Expenditure R&D 8.6 10.02 12.69 9.59 Irrigation 2.39 1.75 1.56 1.88 R o a d s 1.67 3.84 1.92 2.12 Education 3.53 3.66 3.28 3.71 |
| (b) Different elasticities of poverty reduction with respect to the above factors. For given growth rates in agricultural production (or productivity), wages, and non-farm employment, the poverty reduction impact in China is much larger. These poverty reduction elasticities with respect to agricultural production, | Electricity 0.55 0.63 0.4 0.54 Telephone 1.58 2.64 1.99 1.91 Returns to poverty reduction No of Poor Reduced per 10,000 Yuan Expenditure R&D 1.99 4.4 33.12 6.79 Irrigation 0.55 0.77 4.06 1.33 |
| wages, and non-farm employment amounted to 1.13, 0.56, and 0.863 in China respectively, compared with 0.171, 0.185, and 0.594 in India. | R o a d s 0.83 3.61 10.73 3.22 Education 2.73 5.38 28.66 8.8 Electricity 0.76 1.65 6.17 2.27 Telephone 0.6 1.9 8.51 2.21 |
| (c) Different land distribution policies have affected the poor | Poverty loan 0.88 0.75 1.49 1.13 |
| directly and indirectly. In India, many landless farmers are poor while China does not have landless cultivators. Fairly equal | Note: Marginal returns are calculated in 1997. Source: Fan, Zhang and Zhang 2002b. |
the other hand, unlike China, agricultural growth was less significant in inducing growth in non-farm employment in India. Rural development programmes, soil and water conservation, roads, and education matter most in creating additional rural non-farm employment in India. With the exception of a few regions, agricultural growth was not an important variable in the growth of non-farm employment.
Public Investment Priorities
Growth in agricultural productivity, which is the main source of poverty reduction in both China and India, was in turn made possible by public investments in R&D and infrastructure like road, power, irrigation, communication, education and anti-poverty programmes. However, returns on public investments in terms of poverty reduction vary drastically across different types of investment.
The trade-off between agricultural growth and poverty reduction is generally small among different types of investments and between regions. For agricultural research, education, and infrastructure development, they have a significant growth impact as well as a large poverty reduction impact.
Among the categories of investments, those in agricultural research, education, and rural infrastructure are found to be the three most effective types of public spending in promoting agricultural growth and poverty reduction in both countries. This implies that there is a great deal of potential for more growth and poverty reduction by raising the level of public investments in these categories.
Regional analyses conducted for China and India suggest that more investments in less-developed/favoured areas not only offer the largest poverty reduction per unit of spending, but also lead to the highest economic returns. Thus for a same set of investments (such as irrigation, power, and even R&D) in India and China, returns in highly irrigated regions seem to have reached their upper limit, whereas in less-developed areas returns are found to be higher. In the developed irrigated regions, the lower productivity growth in the 1990s is closely associated with lower returns on irrigation and agriculture technology and therefore there is a need to improve both efficiency of existing irrigation projects and expansion of the irrigation facilities to new areas. There is need for more investment in existing technology upgradation in developed regions. Expansion of irrigation in new areas through public investment should come in less-favoured areas such as rainfed regions in India and in the less-endowed areas of China.
This calls for the development of a clearer regional investment policy by the Chinese and Indian governments in relation to the choice of the right investment mix to be implemented in different parts of their respective countries.
Government spending on anti-poverty programmes has a poverty reduction impact but its impact is targeted and narrowly focused. The impact of the anti-poverty programmes through agricultural growth is weak mainly due to the improper selection of the projects under public employment and self-employment. More efforts are needed to better target the funds at the poor, and reorient the investments toward the improvement of rural education and infrastructure. There is also need to use public employment programmes for improving the privately owned land and water resources of the small and marginal farmers. Thus use of anti-poverty programmes for expansion of public and infrastructure, and public and private land and water resources will help not only reduce poverty directly but also indirectly through promoting agricultural growth and thereby offer a longer run solution to poverty reduction.
What Can China and India Learn from Each Other?
The China-India comparison offer lessons to each other to achieve greater poverty reduction in the future.
Firstly, the nature of rural non-farm enterprises matters. The Chinese rural non-farm employment is characterised by township and village enterprises, and migration of farmers to urban areas to work in the service or industrial sectors. Given the nature of work and earning the farmers are most likely to be out of poverty. However, rural non-farm activities in India occur largely in the small and in the informal sectors with low earnings and therefore even if a rural resident is engaged in these activities, he is not guaranteed to be out of poverty. For quite some time therefore the rural and urban informal sectors are likely to be the residual sector for employment of small and marginal farmers and landless households. China’s experience indicates that higher productivity in rural non-farm sectors can be a main source of rural poverty reduction in India.
The related aspect of China’s experience is that Chinese rural enterprises began in areas where agriculture was also the most advanced. In contrast, rural non-farm growth was linked with government spending in rural infrastructure in large part of India and less driven by agriculture-led growth. Thus, it is important that India promotes and strengthens the linkages between the farming and rural non-farming sectors.
The second lesson from China’s experiences is the more widespread poverty reducing impact of agricultural growth due to more equal access to agriculture land. More equitable access to land has enabled the farmer in China to take the benefit of income gains of technological change in agriculture. Therefore for India there is a need to improve access of agricultural land through land distribution and tenancy reforms. The Chinese experience shows that land reforms in the 1950s and the introduction of the household responsibility system (redistribution of land use rights to individual farm households) led together to a sharp improvement in agricultural production and productivity. Equal land distribution also avoids the negative effects of improved agricultural terms of trade on landless farmers and
Table 4: Public Investment and Poverty Reduction
China India Thailand Vietnam
Ranking of Returns in Agricultural Production
Agricultural R&D 1 1 1 1 Irrigation 5 4 5 4 Education 2 3 3 3 Roads 32 42 Telecommunications 4 Electricity 6 8 2 Health 7 Soil and water conservation 6 Anti-poverty programmes 5
Ranking of Returns in Poverty Reduction
Agricultural R&D 2 2 2 3 Irrigation 6 7 5 4 Education 1 3 4 1 Roads 31 32 Telecommunications 5 Electricity 4 8 1 Health 6 Soil and water conservation 5 Anti-poverty programmes 7 4
Sources: Fan, Zhang and Zhang (2002); Fan, Hazell and Thorat (1999); Fan, Jitsuchon and Methakunnavut (2002); and Hao and Fan (2001).
Economic and Political Weekly February 24, 2007 marginal and small farmers who are usually net buyers of grains.
The third lesson relates to education. Compared to India, higher educational achievement levels and increased agricultural labour productivity bring added advantages in China. By 1997, the literacy rate of the general population and of agricultural labour was about 90 per cent in China. Consequently, labour quality has improved substantially, which provided greater opportunity for farmers to use modern technology and to engage in non-farm activities in both rural township enterprises and industrial centres. This also provided more equity in rural areas. On the other hand in India, the rural literacy rate was about 59 per cent in 2001, which is much lower than in China. Greater literacy over time results in higher poverty reducing impact in China than in India.
There are lessons for China too. The experience of India with respect to institutions relating to marketing, finance and information which China lacks in the initial stages can be quite useful. India’s experience with respect to subsidies in agriculture provides an important lesson for China in deciding future government spending priorities. Over the last several decades, the Chinese government has implicitly heavily taxed agriculture through its central procurement and pricing system. As the economy gradually gets liberalised, these implicit taxes have been reduced and evidence has shown that China is in transition from taxing to subsidising agriculture. But whether and how agriculture and farmers should be supported is a hotly debated issue. India’s experience clearly points out that some, if not all, direct input and output subsidies do bring loss in economic efficiency while the poverty impact is small. India’s experience and efforts to rationalise the subsidies in term of selected spheres and targeting are probably useful lessons for China.
EPWEmail: S.Fan@cgiar.org skthorat@hotmail.com
[Where sources are not specified, the data used in this paper are from Fan Hazell and Thorat (1999) and Fan, Zhang and Zhang (2002).]
References
Chen and Ravillion (2004): ‘How Have the World’s Poorest Fared since theEarly 1980s’, World Bank Research Observer, 19(1): 141-169.
Fan, S and P Hazell (1999): ‘Are Returns to Public Investment Lower inLess-Favoured Rural Areas? An Empirical Analysis of India’, EPTD Discussion Paper No 43, International Food Policy Research Institute,Washington DC.
Fan, S, P Hazell and S Thorat (1999): Linkages between Government Spendingand Poverty in Rural India, Research Report No 110, International FoodPolicy Research Institute, Washington DC.
Fan, S Jitsuchon and N Methakunnavut (2002a): ‘Rural InfrastructureDevelopment and Poverty Reduction in Rural Thailand’, Project Reportsubmitted to ADB by IFPRI and TDRI.
Fan, S, L Zhang and X Zhang (2002b): Growth, Inequality and Poverty inRural China: The Role of Public Investments, Research Report 125,International Food Policy Research Institute, Washington DC.
Fan, S and S Thorat (2003): Public Investment, Subsidy, Agricultural Growthand Poverty Linkages, IFPRI/DFID Paper, Washington DC.
Hao, C T and S Fan (2001): ‘How Can Vietnam Better Target Its PublicSpending for Poverty Reduction’, paper presented at InternationalConference on Rural Investment, Growth and Poverty Reduction, November5-6, Beijing.
Ministry of Agriculture of China: China Agricultural Development Report, various issues, China Agricultural Press, China.
Planning Commission (2005): Government of India, 10th Mid-Term Appraisal,2002-07.
