First, we should not assume that the mere fact that inequality reduces economic growth will be enough to convince the rich to reduce it.
On Twitter, @richisglorious objects:
@SeanMcElwee “fact that inequality reduces economic growth” FACT? Strong language for a piece based on a paper that repeatedly…
— Colin (@richisglorious) October 4, 2014
Initially, I was tempted to respond simply that this is a slightly pedantic point – my argument is that even if the wealthy accept the proposition that inequality reduces growth overall, they may still oppose redistribution for other reasons.
As I thought about the question though, I realized the sentence is true: inequality reduces growth. To see why, simply imagine a perfectly unequal society where one person controls all income and everyone else has nothing. There will be no economic growth. Imagine a slightly less unequal society, where ten people control nearly all the income, and everyone else struggle to survive. Again, no growth. The question then, is not whether inequality reduces growth, it’s at what point inequality reduces growth.*
The evidence I summarize in my essay strongly suggests we might be at the point. Other research suggests this as well. But let us not re-litigate this and instead make an important distinction between good inequality and bad inequality.
In my interview with Branko Milanovic, we discussed this distinction. He tells me:
These two extremes [absolute inequality and absolute equality] are clearly not good. We know that the optimum must be at some point in-between. Now, we don’t know what it is. Is this the Gini of “X” or “Y”? We have no idea, but we have in mind a certain “good” range simply by observing things empirically or, sort of, heuristically looking at the world. We see that there are countries that do well economically and socially and what type of inequality they have. Moreover, we have now studies, and I’m quite encouraged by them, which for the first time try to empirically ascertain what percentage of overall inequality is caused by so called bad inequality. And that leads us to the issue of “deserved” and “undeserved” inequality and social mobility.
The bad or undeserved inequality would be the one that arises from the factors over which you have no control: what was the income level of your parents, whether you were born male or female, what is your race, and things like that. Actually, these studies do show that, in some countries, a large chunk of inequality is due to these factors. Thus we can rank countries by their bad and good inequalities. The good inequality, calculated as the residual between overall and bad inequality, is inequality we effort, work, luck and so on.**
The studies to which he refers can be found here and here. They find, after studying Europe and the United States, that inequality which stems from factors beyond an individual’s control (father’s education/race) are correlated with lower growth. Their study of 23 U.S. states over a two decade periods finds, “Inequality is good for growth when that comes from differences in the returns to effort, while it is harmful for growth when that comes from differences in opportunity.” They also find a strong correlation between changes in total inequality and inequality of opportunity.
I believe that there are persuasive reasons to believe that the much of the exploding income inequality in the U.S. is due not to the good inequality, but bad inequality. As inequality increases, the opportunities for rent-seeking through the political system, education system and labor market become greater.
It is also worth noting that good inequality quickly become bad inequality unless mitigated (see: Walton family). Call it the The First Law of Inequality: inequality tends to perpetuate itself unless acted upon by an outside force. Estate taxes and capital gains taxes can limit such rent-seeking.
To summarize: in a state of perfect inequality, there will be no growth. In the state of perfect equality there will be little growth. At some point, inequality will become high enough to choke off growth, mainly by reducing demand and opportunity. Most of the research here is being done by economists who aren’t as cocky as Reinhart/Rogoff and therefore uncomfortable declaring a cut-off point. Instead, we’ll have to use Potter Stewart’s maxim: “I know it when I see it.” I think we see it.
* We might have a similar thought experiment with perfect equality, in which no one is motivated to work since they earn the same amount as everyone else. Even if we assume that Amour-propre provides some incentive, we can assume growth would be slower.
** The full interview will be available next week on the Demos Policyshop blog.