Earlier this month I gave a talk at an international conference in Vienna, Austria, The Haves and the Have Nots: Exploring the Global History of Wealth and Income Inequality. The keynote lecture was given by Branko Milanovic, the author of a book, not coincidentally titled The Haves and the Have Nots (in fact, the organizers borrowed the title for the conference).

Branko is one of the best economists working on inequality in the long run. At the very least, he understands the long-term dynamics of inequality better than Piketty (despite the latter having become a household name). Piketty’s theory is essentially that inequality always grows, except when it is knocked down by an exogenous event, such as a world war or a great depression. It’s a bit of a Deus ex Machina explanation, and it doesn’t explain why inequality never becomes too high, unless because something always happens to bring it down, accidental-like.

Milanovic started his talk by describing several alternative views on how economic inequality evolves in the long term. Vilfredo Pareto’s theory was the simplest one: inequality stays constant at a high level. But we now have abundant data to show that this is not correct. Historically, inequality has changed quite dramatically, and thanks to Piketty and coworkers we now have good quantitative data to prove this point.

Simon Kuznets

Simon Kuznets thought that the evolution of inequality follows an inverted U-curve. Or what might be called a Λ-curve. Although Kuznets did not have access to high quality data on incomes and wealth, we now know that he was correct in discerning that inequality increased during the second half of the nineteenth century, and then decreased during most of the twentieth century, until roughly 1980, which was of course many years after Kuznets wrote his 1955 article. But only 5 years before his death (I wonder, whether he had been able to comment on this second turn-around).

The rise in inequality in the last 30-40 years demonstrates that the inverted U-curve is not a correct description. This is where Branko steps in with his proposal that the dynamics of inequality is best described as repeated “Kuznets Waves.” The previous one, seen by Kuznets, ended in 1980, and now we are living through the ascending arm of the next one.

Here’s a graph from my forthcoming book on structural-demographic analysis of American history, illustrating this wave-like dynamics:

I think Branko is right, because my own historical work suggests that inequality goes up an down in a cyclic fashion. Our data, summarized in Secular Cycles, is of course nowhere near as detailed and quantitative as the data for the post-1800 period, which have been the subject of analysis by Piketty and other economists. However, it indicates that inequality moves in a predictable, cyclic pattern during each secular cycle. And it makes sense, because if inequality always grew, after 5,000 years of state-level complex societies it would have long ago reached the extreme, in which one individual owned all the wealth. This hasn’t happened, which suggests that there could be some kind of a dynamic feedback that would kick in when inequality got too high.