Growth and Poverty

The Importance of Growth and Inequality to Changes in Poverty

A series of recent studies have explored the relative contribution of income growth and distributional changes to changes in poverty. All in all, the papers suggest that the extent to which governments should focus on growth or distributional change to achieve poverty reduction depends on country conditions, and in particular the levels of economic development and initial inequality, as well as the society’s level of tolerance for inequality.

There is plenty of evidence suggesting that growth is important for poverty reduction (Deininger-Squire, 1996; Foster and Szekely, 2001; Dollar and Kraay, 2002; Ravallion 2002; Bourguignon 2003). Building on his earlier work, Kraay (2004) provides the case for pro-growth focus. He disentangles the impact of growth on poverty reduction by identifying three potential sources of pro poor growth (understood as growth that leads to a fall in a given poverty measure). These are: (i) a high growth rate; (ii) a high sensitivity of poverty to growth; and (iii) a poverty reducing pattern of growth. His results suggest that roughly 70 percent of the variation in short-run changes in poverty can be explained by growth in average incomes. In the medium- to long-run, growth would account for an impressive 97 percent of the changes in (headcount) poverty. Virtually all of the remainder of the variance would be due to changes in relative incomes, with the cross country sensitivity of poverty to growth accounting for little of the variation. He also finds that the relevance of growth for poverty reduction declines as one move from headcount poverty to the squared poverty gap. He explains this finding by noting that more bottom sensitive poverty measures place more weight on changes in the distribution of income than on growth.

Other studies state the case for looking also at distribution. Focusing on the expected change in poverty (rather than on the share of variance explained) that would be associated with a one percent growth rate (i.e. the growth elasticity of poverty), and how this impact is affected by inequality, Ravallion (1997) presents a parsimonious empirical model of the relationship between poverty and growth where the rate of poverty reduction associated with a given growth rate depends on a distributional correction (one minus the initial gini index). In Ravallion (2004) the model is improved (in empirical terms) by using an adjustment for possible nonlinearities in the relationship between the growth elasticity of poverty and the initial inequality. His estimates suggest that depending on the initial level of inequality a one percent increase in income levels could result in a poverty reduction of as much as 4.3 percent (very low inequality countries) or as little as .6 percent (high inequality countries). Against this background, Ravallion (2004) concludes that "growth will be quite a blunt instrument against poverty unless that growth comes with falling inequality".

Bourguignon (2003) also emphasizes the importance of the growth elasticity of poverty and how it is affected by distributional changes, as well as by initial inequality and a country’s level of development. Specifically he explores alternative specifications for the relationship between poverty, inequality and growth and concludes that, at least for headcount poverty, assuming that income follows a log normal distribution may prove satisfactory. He finds that distributional changes is as much responsible for variation in poverty reduction across spells as the heterogeneity in growth rates itself. Bourguignon pinpoints two channels as to how redistribution affects growth: a permanent redistribution of income reduces poverty instantaneously through what was identified as the distribution effect; but also redistribution contributes to a permanent increase in the growth elasticity of poverty reduction - therefore accelerating the rate of poverty reduction for a given rate of growth.

Finally, the relative importance a country places on growth and inequality may depend on societal preferences. Certain societies have higher levels of aversion to inequality, which make it advisable to concentrate on re-distribution policies, whilst others have a much higher tolerance to inequality which allows governments to focus on maximizing growth in the first stages of development and correct later on for distribution imbalances.

Next: Growth and Inequality