As every Friday, from Raj Nallari and Breda Griffith's lecture notes.
Variations in Poverty Responses
While economic growth is critical, especially in the long run, the response of poverty to economic growth differs across countries. Key in this respect is the initial level of income inequality and how this inequality changes over time. Lopez (2005) cites the cases of Senegal and Burkina Faso with similar levels of economic growth (2.2% per capita per annum) over a similar time frame poverty declined by 2.5% per annum in Senegal but by just 1.8% in Burkina Faso. Relatedly, the sectoral composition of economic growth can explain why given economic growth rates can lead to differences in poverty reduction rates (see below). Understanding these factors enables us to view growth in a pro-poor context.
The composition of economic growth also has implications for poverty reduction. Growth that derives from the intensive use of unskilled labor will have a greater impact on reducing the incidence of poverty given that the poor are more likely to be unskilled. A recent paper by Loayza and Raddatz (2005) shows that not only does the incidence of poverty fall in the face of aggregate economic growth but that poverty reduction is stronger when growth has a labor-intensive inclination (Loayza and Raddatz, 2005). Furthermore, the Operationalizing Pro-Poor Growth group found that most of the absolute reduction in poverty in countries experiencing economic growth occurred in the rural areas, where the majority of the poor live and draw their livelihoods from agriculture. Thus land and labor policies will affect participation by the rural poor with consequent implications for economic growth and poverty reduction.
From the previous section, we know that the incidence of poverty responds to changes in economic growth. We also know that how receptive poverty is to economic growth depends on the initial level of income inequality and how this changes with economic growth. In recent years the concept of pro-poor growth was introduced to provide greater insights to the relationship between economic growth, poverty and income inequality. The concept of pro-poor growth attempts to capture the way in which economic growth improves the welfare of society’s poor given the accompanying changes in income inequality. Thus assessing whether growth is pro-poor requires knowledge of the distributional changes in income and the extent to which this has impacted the welfare of the poor.
What do we mean by ‘pro-poor growth’? Put simply, ‘pro-poor’ growth is the type of growth process that is most effective in raising the incomes of the poor. More formally, the literature suggests two definitions. The first focuses on whether the distributional shifts accompanying growth favors the poor, i.e. whether poverty falls by more than it would have had all incomes grown at the same rate. The second focuses on whether economic growth benefits poor people in absolute terms as reflected by changes in an appropriate measure of poverty. The measures of poverty generally used are the headcount index, the Watts index and the poverty gap and the concepts of relative and absolute poverty are also used to detect underlying levels of inequality.
Measuring how pro-poor growth is means measuring both how poverty responds to economic growth and to changes in income distribution. To do this we break down poverty, e.g. the headcount index, into a component attributable to economic growth and a component attributable to the change in income distribution. This method has proved to be informative about the reasons for variations in the rate of poverty reduction across countries and has been developed by Datt and Ravallion (1992) and more recently by Kraay (2005). According to Kraay (2005) there are three potential sources of pro-poor growth – (i) a high rate of growth of average incomes, (ii) a high sensitivity of poverty to growth in average incomes, and (iii) a poverty-reducing pattern of growth in relative incomes. The outcome from the large sample of developing countries considered by Kraay during the 1980s and 1990s is that most of the variation (greater than 90 percent) in changes in poverty over the medium to long term is due to the growth in average incomes. The remainder of the variation comes from changes in relative income. A high sensitivity of poverty to growth in average incomes was not found to be relevant for the sample considered.
Two conclusions emerge – the first suggests quite strongly that the search for pro-poor growth should begin by focusing on the determinants of growth in average incomes. Second, although inequality and poverty elasticity were found to be weak sources of pro-poor growth, a deeper understanding of what drives growth will shed further light on its distributional impact and the response of poverty. Undeniably high income inequality hampers the ability of growth to reduce poverty and is in fact a barrier. Ravallion found that for very high inequality countries, a 1 percent increase in income levels had a much lower impact (0.6% reduction) in poverty levels than for low inequality countries (4.3%). The interpretation then must be that faster economic growth is necessary to reduce poverty in countries with high inequality. In other words, the concept of pro-poor growth becomes relevant.
Poverty will be more responsive to growth the greater the equality of income distribution. While the literature largely agrees that growth tends to be distribution neutral, i.e. changes in inequality are not related to changes in economic growth; distributional changes have been an important factor in explaining differing rates of poverty reduction at a country level. The 1990s have been characterized by rising within-country inequality in every region except the Middle East and North Africa. The Gini coefficient for Sub-Saharan Africa and Latin America is similar. Between 1993 and 2002, the Gini coefficient grew by 2.3 percent annually in Vietnam and by 2 percent per annum in China between 1990 and 2001. Ravallion (2001) suggests that ‘the median rate of decline in the “$1/day” headcount index is 10% per year among countries that combined growth with falling inequality, while it is only 1% per year for those countries for which growth came with rising inequality. Rising inequality reduces the impact of future economic growth on poverty reduction. Bourguignon (2004) finds that a fall in the Gini coefficient from 0.55 to 0.45 (still a relatively high inequality country) would cause poverty to drop by more than 15 percentage points in 10 years. It would take 30 years to achieve the same reduction in poverty if inequality remained unchanged.
To conclude, we note that:
(i) growth is the fundamental factor for poverty reduction,
(ii) growth accompanied by progressive distributional change is, however, better for poverty reduction than growth alone; and
(iii) high initial inequality reduces the impact of growth on poverty reduction.
The existing literature suggests that both growth and inequality are important for pro-poor growth and that their relative importance depends on country conditions, most notably the initial level of income and inequality. In order for growth to be pro-poor, it will have to disproportionately benefit the poorer levels of any given society. In low-income countries, the need to accelerate pro-poor growth for poverty reduction has emerged as critical, particularly in the context of poverty reduction strategies and the efforts to accelerate progress towards the Millennium Development Goals (MDGs).
Next week: Economic Growth and Inequality