The French economist Thomas Piketty scored an unlikely best-seller in 2013 with his book Capital in the Twenty-First Century, which examined what he saw as the almost inevitable growth of inequality under capitalism.
Senior Fellow Daron Acemoglu says that institutions matter more for inequality than general rules of economics.
The book inspired a tremendous amount of discussion, including a recent paper from Daron Acemoglu, a senior fellow in CIFAR’s program in Institutions, Organizations & Growth, and his Harvard colleague James Robinson. In their paper “The Rise and Decline of General Laws of Capitalism,” they argue that the increase in inequality has little to do with Piketty’s emphasis on the difference between the rate of return to capital and the growth rate of labor income. More important, they think Piketty doesn’t pay enough attention to the difference that institutions such as government can make to how much inequality changes and how much that matters. News & Ideas spoke with Acemoglu about his article.
Q. What’s interesting to you about Piketty’s work?
Piketty has done a tremendous amount of work on questions of inequality. It’s been a major topic of interest for economists to understand differences across countries and trends over time in inequality measured in a variety of different ways. I think Thomas, together with his long-time collaborator, Emmanuel Saez, has really shed new light on a particular snapshot of inequality, and it’s been very useful and it’s very been influential.
But the new book embeds this in a narrative about long run determinants of inequality, and implicitly also about costs of inequality for society. At the centre of the book is this sort of tension between capital and labour income. A few people earn a huge amount of capital income, and labour income is somewhat more equally distributed. And so trends in capital versus labour income are going to have major implications for how income inequality evolves and how we forecast it to evolve, the book argues.
Much of that is summarized in the inequality r>g, meaning the interest rate is greater than the rate of growth of the economy as a whole. The idea is that because capital is so unequally held, the income and wealth of the very rich grows at the rate r, whereas the rest of us earn labour income, and our income grows at the growth rate of the gross domestic products, g.
Q. Is another way of saying it that the rich get richer?
Yes, if r is greater than g, the capital owner rich get richer.
Q. Your paper talks about how it may be too simplistic to cast this in terms of a general law. Why do you think so?
The general laws refer to the ways that classical economists like Ricardo, Marx, and Malthus thought about the world. Marx explicitly used the words “general laws of capitalism”. And that’s also the language that Piketty uses in Capital in the Twenty-First Century.
And I think the way that that is presented in Piketty’s book is quite similar to the way that it is presented in Marx’s book, which is that those are physics-like laws that are very powerful, and all other economic, social and political relations are shaped by them.
But we fundamentally think that if we have learned anything from economic and social research over the last century, it is that nothing is as simple as these general or fundamental laws. All economic relations are embedded in certain institutional settings, and it is those institutional settings that determine how things evolve. They do so of course interplaying with economic relations. But ignoring these institutional settings and imagining that such general laws can shed light on the medium or long-run future of the economy is unsatisfactory.
Q. Are you saying that institutions can ameliorate this general law that he’s discovered?
I would say it’s that, first of all, it is not a general law of capitalism in the following sense: that r greater than g in no way implies increasing inequality. It may under some circumstances create an additional force towards inequality, or it may not. And second, even if it did create a force towards increasing inequality, this has to compete with a series of other economic relations, some of them more complex, and it has to compete with a series of institutional responses that could be much more powerful.
And for that reason, we also show in the paper that if you look at historical data, the very data that Piketty himself uses, there is no relationship between r greater than g and inequality.
But intuitively it makes sense. If I have income from labour and my brother has income from capital, and his income every year is going up faster than mine, why wouldn’t the inequality between us grow over time?
First of all, and most importantly, if you look at the data, much of the inequality that we see around us is not due to capital income. So if your brother is richer than you, that’s most likely because he’s earning more labour income than you. The capital is a sideshow.
Second, imagine that your brother is richer than you and is a capital holder, and you’re not. Whether his income over time grows faster than yours will depend on how much he consumes. So if he has consumption growth that is commensurate with the rate of return that he’s enjoying on his assets, then that will also eat into his assets, and you can have a stable distribution of income. And that’s actually in fact what happens in the majority of economic models that we have.
The third factor is that this example is deceptively simple, because it assumes that your brother and you are in fixed situations. In reality, there is social mobility. What social mobility implies is that your brother might be getting richer relative to you, but then when he has a child, his child may end up to being a labourer, and your offspring may be the capital earners. So that sort of social mobility is going to mix things up and act as a powerful force towards a more stable distribution of income.
Q. You used South Africa and Sweden as two case studies. Can you tell me what you found there and what points you were making?
I wanted to make a couple of points. First, this excessive focus that comes from Piketty’s work, and others who build on his work, on top income inequality can be quite misleading. South Africa is a beautiful illustration of that. It was recognized as one of the most unequal countries, and unfairly unequal countries, until apartheid collapsed. But it actually had fairly low and in fact declining top income inequality (top 1 per cent share), while the apartheid state was being built, and while the inequality between whites and blacks based on the vicious repression of blacks was exploding.
So it shows how sort of wrong-headed it can be to just focus on this number at the expense of everything else.
Second, we try to emphasize that major turning points in inequality are generally associated with major institutional changes, such as the emergence of social democratic state in Sweden, or the building of the apartheid state in South Africa.
If you want to understand inequality, not only do you have to look at the body of the income distribution, not just the share of the top 1 per cent, but you also have to look at the institutional settings, policies and their interactions with technologies in order to understand all of these changes.
Q. So I would take it if you’re interested in inequality, this is maybe a more hopeful analysis. You’re not fighting against some unstoppable force of capitalism.
Absolutely, absolutely, absolutely. I think even putting it as a general law of capitalism creates this gloomy view. And I think that’s not hopeful. That’s not helpful. There are lots of problems in the U.S. economy. There are lots of problems with the Canadian economy. But many of those are things we can identify, and deal with through policies and institutional safeguards.