|
Abstract
Why do some people receive higher incomes than others with similar talents and abilities? And why do certain sources of income, such as income from farm labor and income from growing sugarcane, go to different people? What steps can be taken to reduce the wide differences in income earned, so that the number of people living below the poverty line can be reduced? What roles do education and the migration of household members play in lifting people out of poverty and in improving income distribution?
To answer these questions, researchers in recent years have used various techniques to analyze the sources of income inequality and poverty in developing countries. These studies help policymakers understand the root causes of income inequality and poverty within their societies. Equipped with such information, policymakers can take specific measures to improve income distribution and the income-earning potentials of different groups of people.
In Sources of Income Inequality and Poverty in Rural Pakistan, Research Report 102, Richard H. Adams, Jr. and Jane J. He address these issues by analyzing longitudinal data from 727 households in four districts in rural Pakistan (Faisalabad and Attock in Punjab Province, Badin in Sind Province, and Dir in North-West Frontier Province). Whereas previous studies of income inequality and poverty have often been based on one round or one year of household interviewing, this data set encompasses 12 rounds of household interviews over a three-year time period, 1986-89. Detailed data were collected on income, expenditures, education, employment, migration, and landowning. Total income is then broken down into 5 sources of income in order to examine the contribution of each of these sources to income inequality and poverty.
Whether a household owns land has a major effect on the types of income it receives. On the whole, income inequality in the survey areas is moderate. The Gini coefficient, which ranks inequality on a scale of 0 to 1 with 1 highest, indicates that the distribution of total per capita household income among the sample population is 0.381, but the distribution of landownership is much higher, 0.769. In rural Pakistan, the lack of land--37 percent of the survey households own no land--forces the poor to earn the bulk of their livelihood from those sources of income that are not connected with land, such as nonfarm employment and livestock.
SOURCES OF INCOME
Although all of the survey households are rural, the report finds that agricultural income is not the most important source of income. Of the five sources of income--agricultural, nonfarm, transfer, livestock, and rental--nonfarm is the most important. Depending on the year, nonfarm income--from unskilled labor, self-employment, and government employmen--accounts for between 30 and 34 percent of total per capita household income. Agricultural income is the second most important source, accounting for between 23 and 27 percent of total per capita household income.
Agricultural, transfer, and rental incomes are of major importance only to the highest quintile, thereas nonfarm and livestock income are most important to the lower quintiles.
Moreover, nonfarm income is the leading source of income for poor households. When the households are ranked by income, poor households receive almost 50 percent of their total per capita income from nonfarm sources--more than twice that received by poor households from any other income source, and more than seven times that received from agriculture (Figure 1). Livestock income is the second most important source of income for the poor. According to the report, poor households receive almost 25 percent of their total per capita income from livestock.
INCOME INEQUALITY
Of the five sources of income, nonfarm and livestock income tend to decrease overall income inequality. However, additional increments of the other three income sources--agricultural, transfer, and rental--increase overall income inequality. Agricultural income makes the largest contribution to overall income inequality and livestock income makes the smallest. Depending on the year, agricultural income accounts for between 35 and 45 percent of overall income inequality, while livestock income accounts for only 1 to 11 percent. Land is the main reason for this difference. Because land is distributed so unevenly in the study areas and in rural Pakistan as a whole, the main benefits of landownership--such as crop income and land rent--tend to go to the rich. Livestock and nonfarm income, however, are negatively correlated with landownership.
These findings suggest that the direct, immediate effects of agricultural growth in Pakistan have gone mainly to those households that own land: the rich. To be sure, these findings do not negate results from other studies indicating that additional rounds of agricultural growth can and often do benefit the poor. If future agricultural growth in Pakistan increases the demand for labor more than the demand for land, then agricultural growth could have a positive impact on income distribution. Nevertheless, the results of this report show that agricultural wages represent on average less than 6 percent of total per capita agricultural income.
NONFARM INCOME
To present a more detailed view of the effects of various types of income on inequality, the report breaks down and analyzes each of the five sources of income into 35 separate sources of income.
The report finds that while nonfarm income as a whole reduces income inequality, not all sources of nonfarm income have a favorable effect on income distribution. Of the three main sources of nonfarm income--unskilled labor, self-employment, and government employment--only unskilled labor tends to decrease income inequality because the poor depend heavily on it. Government employment, an inequality-increasing source of nonfarm income, accounts for a large proportion--21 to 31 percent--of nonfarm income inequality. This is because government jobs often have education requirements, and only members of richer households can usually afford education.
AGRICULTURAL INCOME
The report finds that food crops--like wheat and rice--and cash crops--like sugarcane--have different effects on inequality. Wheat and rice, the main food crops in Pakistan, account for a large share of the agricultural income of the poor, and therefore they are inequality-decreasing sources of agricultural income. At the other extreme, one cash crop--sugarcane--accounts for between 33 and 37 percent of agricultural income inequality (Figure 2). Because sugarcane stays in the ground all year, poor farmers with small plots, who need to grow food crops to feed their families, seldom grow sugarcane. This is unfortunate from an equity standpoint because government pricing policies have made sugarcane a very profitable crop. In rural Pakistan, sugarcane production is dominated by rich households who receive over one-third of their net agricultural income from this single cash crop.
TRANSFER INCOME
Most transfer income in this study comes from remittances, either from internal migrants working in urban areas in Pakistan or from external migrants working abroad. These two types of remittances have different effects on income distribution, however. Internal remittances are crucial for the poor, who receive over 75 percent of their total per capita transfer income from family members who work in Pakistan's cities. The better-off households are more likely to be able to afford the cost of sending workers abroad. As a result, external remittances represent an inequality-increasing source of income and account for between 58 and 91 percent of transfer income inequality.
LIVESTOCK INCOME
The report finds that different types of animals have different effects on inequality (Figure 3). Income from buffalo, which are more expensive, is concentrated in upper-income households. The poor are most likely to raise and receive income from a single female animal--a "local" cow (an indigenous type of cow that has not been improved by breeding with imported stock). The poor receive almost 60 percent of their total per capita livestock income from local cows. The poor prefer to own cows rather than male animals because they can use cows to plow and to produce milk and offspring.
RENTAL INCOME
The report finds that about 80 percent of rental income comes from land rent, either in-kind or cash rent. Since land is distributed so unevenly in Pakistan most rental income goes to the rich. Rich households receive almost 20 percent of their total per capita income from rental income; poor households receive less than 5 percent of their total income from this source.
POLICY CONCLUSIONS
The report shows that the incomes of poor households fluctuate considerably. When ranked by income data, only one-third of the 145 households in the lowest quintile group in the first year of the study were in that quintile in both of the next two years. This means that most of the poverty found in this study is temporary rather than chronic in nature. Households tend to move in and out of poverty for a variety of reasons. According to this report, changes in physical assets (such as landownership) and the household labor force (through education and migration) account for about one-quarter of the changes in income of the poor.
Until ways can be found to distribute land more evenly in Pakistan, the report suggests that future policies need to be designed that recognize the dependence of the poor on sources of income outside the agriculture sector, such as nonfarm and livestock. For the poor to benefit more from nonfarm government employment, the government needs to increase educational opportunities for the poor. In the livestock sector, since local cows are such an important source of income for the poor, Pakistani policymakers should initiate programs to upgrade the productivity of local cows through cross-breeding schemes and veterinary programs.
The breakdown of agricultural income shows that, because of government pricing policies one cash crop--sugarcane--accounts for most of agricultural income inequality. Since Pakistan has no comparative advantage in the production of this crop, sugarcane pricing policies need to be revised. At the same time, to improve equity, policymakers need to upgrade technologies for producing the main food crops, wheat and rice.
To improve the distributional effects of external remittances, policymakers should take steps to help poorer households meet the financial costs of sending a migrant abroad. Among the measures that might be considered is the establishment of rural migration centers that could process visas, work contracts, and loan arrangements for prospective external migrants.
|
