Will the rich always get richer?

BY Laurence Kotlikoff  May 16, 2014 at 12:40 PM EDT
Larry Kotlikoff disputes Thomas Piketty's inequality thesis that the rich will keep getting richer and richer. Photo by Torresigner/360/iStock via Getty Images.

Larry Kotlikoff disputes Thomas Piketty’s inequality thesis that the rich will keep getting richer and richer. Photo by Torresigner/360/iStock via Getty Images.

A Note from Paul Solman: Larry Kotlikoff is a regular here on Making Sen$e, especially with his “Ask Larry” Social Security column on Mondays.

Friday, he takes on the subject we’ve been exploring all week, the inequality thesis of French economist Thomas Piketty. Larry is no fan, as he makes clear.

I have just one point to make before he takes the floor, a point Piketty made when I asked him if one has to buy the theoretical crux of this thesis and book – that the return to capital is and will continue to outpace economic growth – in order to worry about rising inequality.

Paul Solman: Do we need to believe your story about inequality being a function of the high return on capital compared to a low growth of the economy in order to think that we ought to be doing something about inequality? Regardless of why it’s happening?

Thomas Piketty: If the top is rising three times faster than the average, whatever the reason for this in the long run, this is a problem. Where will this stop? Nobody knows and there’s no natural force that guarantees that this will stop somewhere that is acceptable and compatible with our democratic institutions. So I think there is a lot of evidence that the key force is the rate of return to capital [rising faster] than the growth rate. But even if some other combination of forces was to explain [inequality], the point is that you don’t want the top and the middle to completely [diverge] and evolve at completely different rates in the long run.

With that out of the way, here’s Larry’s critique. Watch my interview with Piketty at the bottom of this post.


Extreme private wealth inequality is a fact of economic life in most countries. The U.S. is a prime example. The top 1 percent of wealth holders own almost 40 percent of all wealth. And the top 10 percent hold almost 90 percent.

What should we make of this?

Thomas Piketty’s new book, “Capital in the Twenty-First Century,” claims that ever-rising wealth inequality is an inevitable feature of capitalism. His argument? The growth of wealth is set by its rate of return, denoted by the symbol r, which is much higher than the growth rate of wages, denoted by the symbol g. Thus capitalists become “more and more dominant over those who own nothing but their labor,” inviting “terrifying consequences.” Or, as Marx put it, capitalism “carries the seeds of its own destruction.”

Marx’s solution was “Abolish all private property.” Piketty’s answer is twofold. First, tax large wealth holdings at up to 10 percent per year so that they disappear, albeit gradually.

And second, don’t trust economists’ mathematical models. The statement r > g (r exceeds g) is the only math needed — that plus the data.

We can’t rely on Piketty’s analysis and numbers as a cure for capitalism. His “theory” is deeply flawed, his private wealth data are highly misleading, and his policy prescription is over the top. His book may set back public understanding of resource inequality for years to come.

Piketty’s r > g economic law of ever-rising wealth inequality would make a good Disney cartoon. Picture Scrooge McDuck and his nephew Donald stuck on an island. Each day two bags of corn drop from the sky. Donald eats his corn as it lands. Scrooge eats nothing. Instead, he plants every kernel, gets a big yield of new corn, and stacks this output in an ever-taller pile.

But Scrooge hates working. So he pays Donald a kernel a day to plant and stack. Scrooge, now perched high atop his growing mountain, gets richer and richer (thanks to r, the return on his investment). Donald’s wealth gets smaller and smaller from Scrooge’s dominant, ever-rising perspective.

Now Donald realizes he should save, just like his uncle. But he can’t. His wages are fixed because overall economic growth -– g — is zero. And he needs all the calories he can get to fuel is ability to work. As a result, he never has a kernel to his name. Eventually, Donald realizes his oppression and walks off the job — leaving Scrooge stuck sky-high and squawking. The weather’s changed – from warm to frigid.

The real world is much different from this caricature, however. In the U.S., the two richest people are Bill Gates and Warren Buffett. Most of their fortunes are going to the Gates Foundation, whose main mission is eliminating disease in the developing world — perhaps the greatest cause of global poverty and inequality.

Neither Gates nor Buffett came from big money. And even were they, instead, to leave all their wealth to their children, it would, over time, become dispersed as their descendants multiply, marry “down,” get divorced, mismanage their money, pay taxes, donate, pursue lives of leisure, lose their jobs, live longer than expected, incur major medical expenses, lose big at Black Jack, and so on. As the saying went in 19th century England, “Shirtsleeves to shirtsleeves in three generations.”

But the main reason why wealth distributions stabilize in all realistic r > g economic models is that the rich, like the poor, spend money. Becoming rich doesn’t suddenly transform us, let alone all our heirs, into Scrooge McDucks with spending propensities that are below average, let alone zero.

One quick glance at the massive yachts docked or the private jets parked in Nantucket, Massachusetts, or other playgrounds of the rich, makes this plain. And without this assumption – that heirs are permanently transformed into non-spending Scrooges – all r > g economies end up with stable wealth distributions.

This doesn’t mean bequests vanish. It means that, as a group, those who inherit spend the excess of r over g earned on their inherited wealth. This keeps wealth holdings from becoming ever more skewed.

The math here isn’t that tough and takes us a long way. Computer modeling helps us see that Piketty’s “theory” is at odds with actual theory. (See, for example, this paper). So does a glance at Forbes’ list of the 400 wealthiest Americans. Sixty percent of the people listed in 2001 weren’t on the list in 1989. And nearly a quarter of those listed in 2001 weren’t listed in 1998. This is the future devouring the past, not “the past devouring the future,” which is Piketty’s version of “the rich getting richer.”

Moreover, Piketty’s supposed law of growing private wealth inequality is refuted by U.S. data, which shows, if anything, higher wealth inequality in the early part of the last century than prevails today. Piketty says shocks — the Great Depression, the New Deal and World War II – temporarily interrupted the force of r > g. But g > r events aren’t all that rare. Remember 2008? Proclaiming a “law” and then saying it hasn’t worked for over a century because it only works when it works isn’t serious scholarship.

Moreover, all reasonable economic models in which wealth keeps rising relative to the supply of labor will produce an ever-smaller value of r as capital becomes ever more abundant compared to labor and competes for ever-shrinking investment opportunities, relative to the value of labor. Thus, even if Piketty were right, he’d be wrong. Past wealth accumulation would devour future wealth accumulation.

Another inconvenient truth for Piketty’s thesis is that the ratio of total inheritances to total net worth has dropped by one third over the last two decades. This finding comes care of New York University’s Edward Wolff and the Bureau of Labor Statistics’ Maury Gittleman. Their careful study of Federal Reserve survey data also shows that “inheritances and other wealth transfers tend to be equalizing in terms of the distribution of household wealth.”

How can that be? One answer is that the parents of the poor die at earlier ages than parents of the rich and aren’t, therefore, able to spend down as much of their wealth before they die and bequeath it.

But the biggest empirical problem with Piketty’s work is relying on private wealth data to measure wealth inequality. In so doing, he leaves out much, if not most, of the total wealth of poor and middle-class households – their claims to future Social Security, Medicare and Medicaid benefits (or their equivalent in other countries). See Paul’s story with economist Bob Lerman on this very point. The fact that households acquired these assets by saving through the government doesn’t justify excluding them in analyzing total wealth inequality. Were Piketty to include them, he might find declining wealth inequality in recent decades.

The deep flaws in parts of Piketty’s book don’t mean that inequality is either small or benign. But the real source of inequality these days is not due to capitalists saving every penny and workers spending every cent or to r always exceeding g. It’s due to labor earnings becoming ever more skewed. This is happening for a variety of reasons, including the advent of smart machines. This rising wage inequality, which Berkeley’s Emmanual Saez and co-author Piketty have spent years carefully documenting, doesn’t pit capitalists against workers. It pits workers against workers.

Paul Solman interviewed Piketty about his thesis and the success of his book on the NewsHour.