An interesting new survey paper that brings together theoretical and empirical evidence on the effects of inequality on growth. Key conclusions :
The following conclusions can be drawn from this survey of the theoretical and empirical literature regarding the effects of initial inequality on future growth.
On the theoretical side, the main point that arises from the new literature on the relationship between inequality and growth is that a less unequal initial distribution of income and wealth can be associated with faster long-term growth. However, this does not mean that a progressive redistribution of income and wealth therefore increases necessarily the future growth rate of the economy. Indeed, on the one hand, a progressive redistribution can support long-run growth for the three following reasons: first, by relaxing the liquidity constraints, it permits the poor agents to accumulate more and more productive assets (for instance, it enables low-income agents to invest in human capital or for the landless peasants to buy plots of land on credit) and thus to reduce their fertility rate. Second, by raising their purchasing power, it also supports the progressive access of low-income individuals to a larger variety of new goods produced by modern industries. Last, it reduces finally the social dissatisfaction and the participation of poor agents in illegal and criminal activities. On the other hand, this redistributive measure can also slow down the long-term growth since it reduces the incentives of richer classes to accumulate productive assets and innovate in the production of new goods.
Nevertheless, as Bourguignon (2000) rightly suggests, a wealth redistribution which would be financed through a progressive taxation of incomes of some part of the population (in particular the upper class) and proceeds of which would be intended to support the accumulation of productive assets among the poorest members of the society, should contribute to accelerate both growth and poverty reduction. Under these conditions, the gains (more assets being accumulated by the poorest) should more than compensate the losses (i.e. less assets being accumulated by favoured classes) caused by this redistributive measure. In addition, the feasibility of this progressive redistribution policy supposes that the total wealth of the economy is large enough to prevent that this type of redistribution of the meagre national resources results in a null rate of accumulation of productive assets and thus a non-existent growth in the long run.
On the empirical side, the econometric estimates of the direct and indirect links from initial inequality to future growth led to overall rather mixed results: first of all, the cross-section reduced form regressions show that inequality of wealth (human capital and land) affects significantly and negatively the future growth rate. Asset inequality turns out to be a more robust determinant of growth than income inequality. Then, the findings from cross-section structural form estimates reveal that only the endogenous fertility approach and the explanation based on political instability receive convincing support from the data. Last, initial inequality of assets has a significant and negative effect on the future growth rate in panel data regressions.
According to Bourguignon (1998b), the general inconclusiveness of cross-sectional [and panel data19] studies of the relationship between income inequality and growth means that "on the whole, it is probably an exaggeration to consider that redistribution of income and/or productive assets is a panacea for economic growth […] [and] there is no reason to consider that redistribution is systematically inefficient. In other words, redistribution may not be an engine of growth, but, in some circumstances, it may be an adjuvant of it" (Bourguignon, 1998b: 20). More basically, as Bourguignon (2000, 2004) also points out, the appearance of a complementary relationship between wealth equality and growth suggests that if the ultimate goal of development is poverty reduction, then a "redistribution for growth" strategy including asset redistribution policies may be the most efficient for poverty reduction.