Development Economics
Kuznets’ incessant empirical drive prompted him to dive into developing countries, seeking additional data to further test his long-term hypothesis on income inequality. As we saw in the previous post, the scant data available for a few developed countries only showed declining inequality matched by increased economic growth. Evidence regarding trends in inequality in the early phases of capitalist development was thus missing. Consequently, looking at developing countries to see whether inequality increased in parallel with economic growth during those inaugural stages made a lot of sense. That is partly why section IV of the paper focused on them.
If relevant inequality data were scarce for developed economies, the situation was comparatively much worse for poorer nations. Here, the paper offers six data points for three countries: India, Ceylon, and Puerto Rico for 1948-1950. In all three, the top 20 percent have income shares of 50 percent or more, which are higher than those of the same percentile group in the U.S. and the UK. At the same time, the lower 60 percent have income shares below 30 percent, about 5 percentage points lower than the average for those same countries. In that context, income inequality is larger in developing countries. Note that here we are getting a snapshot of these countries, not a historical series.
Aware of such a limitation, the author then highlights some characteristics of developing countries.
- Margins of error on inequality data collection might be higher
- Income levels are more prone to “transient disturbances” (business cycles, crises, etc.) as “primitive material and economic technology prevail.” (pg. 21)
- Low-income levels prevail, which raises the question of how local populations survive
- Consequently, income shares of the lowest percentiles cannot go below certain levels (six or seven percent)
- Both the lower and higher income percentiles receive larger income shares
- That is the result of the absence of a middle class
- Some of these countries were economic leaders in the past, but have been unable to grow as fast as advanced nations in the last two centuries
From here, Kuznets tentatively concludes that income inequality in developing countries is “somewhat” greater than in developed countries, while their income per capita is comparatively lower. In addition, the rate of growth of per capita income is also low. On this basis, he speculates that income inequality in developing countries has not decreased in the last few decades, unlike in industrialized nations. That is the result of the lack of local conditions to promote economic growth, combined with the absence of political will to implement policies supporting the poorest sectors of the population.
On this basis, he suggests that income inequality in these countries might have actually widened over the past decades. Clearly, he was not pushing for an income inequality “law” that would work around the globe, 24/7, forever.
As for studying developing countries, he recommended a more empirical approach grounded in history and current internal conditions. He also warned us that “Because they may have proved favorable in the past, it is dangerous to argue that completely free markets, lack of penalties implicit in progressive taxation, and the like are indispensable for the economic growth of the now underdeveloped countries. Under present conditions, the results may be quite the opposite…” (pg. 26).
Unsurprisingly, some place Kuznets among the early pioneers of modern development economics.
A Kuznets “Law”?
The parallels between Kuznets’ 1955 article and W.A. Lewis’s “Economic Development and Unlimited Supplies of Labor,” published 7 months earlier, are striking. While the two authors were unaware of each other’s work, they both deployed dual models to explore capitalist development. Lewis was entirely focused on developing economies and provided a mechanism to explain capital accumulation in the presence of stagnant wages. That can indeed lead to increased inequality during the early stages of development, exactly what Kuznets proposed. While the authors have different academic targets in mind, the interconnection between their work is undeniable.
On the other hand, mainstream economics took a very different path. The classic early example was Rostow and his 1960 anti-communist manifesto, which dominated development economics in the 1960s. He clearly did not get the full memo from Kuznets. Instead, he suggested five stages that, if followed sequentially, will inexorably lead to development, regardless of geography or history. Such instances were drawn from the experience of advanced countries that proudly showed the future to poorer nations. That implicitly assumed that long-term capitalist development entailed decreased income inequality, thus showing the superiority of capitalism over socialism or communism.
I have already mentioned the role the World Bank and other international organizations and foundations played in making Kuznets’ hypothesis a development “law.” Essentially, they cherry-picked Kuznets’ research to find hypotheses that fitted their economic and ideological purposes, thus ignoring all its critical elements. History and detailed analysis of internal conditions went out the window. Here, we should not underestimate the impact of the Cold War on the development agenda. In most cases, economic development was superseded by political interventions, covert and overt, to stop the spread of “communism,” especially after the Cuban revolution.
In hindsight, both Latin American structuralists, led by Prebisch, and dependency theorists, drawing on diverse theoretical frameworks, were very close to Kuznets’ development economics approach. That said, they usually took a more radical approach in numerous instances. However, there is no evidence they were aware of Kuznets work.
The origins of the inverted-U Kuznets curve can be traced to 1970, when independent researchers began using econometric models to test the income inequality hypothesis. They almost accidentally found such a shape, using the basic model that deployed a second-degree polynomial to capture the potential inflection point. Kuznets himself was not part of this work, as he had already moved on to other topics. And as far as I know, he never endorsed such a christening of his highly speculative hypothesis.
One has to wonder if his early followers read the full paper. Or maybe they also went cherry-picking.
Raúl
