Poonam Gupta

One of the Millennium Development Goals set by the United Nations is to reduce extreme poverty by half by 2015. How can such a goal be reached? Can growth by itself be pro-poor, or does poverty reduction depend on macroeconomic policies? How effective are policies directly targeted at the poorer segments of the population? Is the quality of institutions important in determining poverty reduction strategies? This article summarizes recent IMF research on these critical issues.

Kraay (2004), using household survey data for a sample of 80 developing countries, looks at the relationship between economic growth and poverty reduction. He finds that most of the variation in poverty can be explained by average income growth, and argues that growth is indeed pro-poor. Instead of looking at income-based measures of poverty reduction, Moser and Ichida (2001) rely on human development indicators of poverty, such as life expectancy, infant mortality, and primary school enrollment. Using a panel data set for 46 countries in sub-Saharan Africa, they also find a strong and robust relationship between growth and poverty reduction.

Linking poverty reduction to macroeconomic performance, Epaulard (2003) looks at how poverty has been affected in a number of boom and bust episodes in developing and transition economies. She shows that the elasticity of poverty reduction to growth (i.e., by how much a percentage increase in growth will translate into a percentage increase in poverty reduction) depends on the initial levels of both per capita income and income inequality. The higher the average income level, the higher the elasticity; and the higher the inequality, the lower the elasticity. The paper also finds the poverty response to growth to be symmetric across positive and negative macroeconomic shocks. Ghura, Leite, and Tsangarides (2002) also show that though growth is important for poverty reduction, the income of the poor does not rise one to one with per capita income. In addition to growth, macroeconomic policies, especially those aiming at lowering inflation, deepening the financial sector, and raising educational achievements, are important in reducing poverty.

The link between macroeconomic policies and poverty is further explored by Cashin and others (2001). Looking at the changes in the United Nations Development Program’s Human Development Index (HDI) between 1975 and 1998 in a large sample of developing countries, the paper shows that a macroeconomic environment leading to low and stable inflation, low budget deficits, low levels of external debt, openness to trade, rule of law, and higher education levels seems to be associated with greater improvements in HDI. However, the paper could not conclusively relate improvements in HDI with changes in specific macroeconomic policies. Berg and Krueger (2003) survey the literature that links trade policies to growth and poverty. They show that an increase in openness is an important contributor to growth, and that trade-led growth does not systematically worsen the income distribution. Combining these two propositions, they argue that trade openness would likely reduce poverty through its impact on growth.

Looking at the link between financial development and poverty alleviation, Holden and Prokopenko (2001) conclude that while financial development can contribute to poverty alleviation through the promotion of growth, two necessary conditions for the development of the financial sector are macroeconomic stability and strong institutions, particularly those concerned with the guarantee of property rights. Looking at the role of institutions, and indexing the quality of institutions through an indicator of corruption, Gupta, Davoodi, and Alonso-Terme (1998), using cross-country regressions, show that corruption increases income inequality and poverty through lower growth, lower social spending, poor targeting of social programs, biased tax systems, and increased inequality. One implication of the analysis is that policies that reduce corruption will also lower income inequality and poverty.

How is poverty affected during financial crises? Using cross-country and household survey data for Mexico during the 1994 crisis, Baldacci, De Mello, and Inchauste (2002) show that financial crises in developing and transition economies are associated with an increase in poverty and income inequality. They also show that, as expected, the provision of targeted safety nets and the protection of specific social programs from fiscal retrenchment can be pro-poor during financial crises. Loko, Nallori, and Kalonji (2003) analyze the link between external indebtedness and poverty in low-income countries. They show that the external debt has a small but significant effect on human development indicators, such as life expectancy, infant mortality, and primary school enrollment. To explain this finding, they suggest that high debt servicing crowds out government’s social spending, and is therefore associated with a worse outcome in term of HDIs.

Shedding light on the debate on poverty numbers in India, Aziz (2002) looks at state-level data for India in 1978–97 and shows that, while overall poverty declined, it increased somewhat during the early years of the 1990s reform period. He also shows that the states with higher growth and lower inflation have experienced faster reduction in poverty. In another case study, Thomas and Canagarajah (2002) look at the dynamics of poverty in Nigeria between 1985 and 1992 and examine the impact of macroeconomic policies on economic growth and welfare. Their main finding is that the decline in poverty during this period can be attributed mostly to the growth of the economy, rather than to changes in the income distribution.