What's new since 1982?

When I moved to Boston in 1982, no one talked about the declining share of national income going to labor. Or increasing income inequality. Now these are the hottest topics in economics. Much of the discussion seems to implicitly suggest that there is some sort of “mystery” to be explained. Perhaps corporations are getting better at lobbying in Washington. Or maybe there is cultural change that makes CEOs bolder in demanding high pay.

I find those sorts of explanations to be unsatisfactory. Too vague. Let’s go back to my move to Boston. The city I moved to in 1982 was radically different from the Boston of today, even though superficially it doesn’t look all that much different. It has not gone through the sort of radical physical transformation that you see in places like Shanghai, or even Austin, Texas. But nonetheless Boston is radically different, and has changed in ways that seem to correspond quite closely to the growing inequality in America.

I’m going to suggest that maybe there is no big mystery to explain. I won’t present any new ideas, but rather bundle together some innovative work done by others. It seems to me that the economy has changed in such a way that it would be surprising if the labor share of income had not fallen. Indeed perhaps the real surprise is that the ratio has not fallen by even more.

I believe that major societal changes don’t just happen randomly, they have causes. Here are three reasons that others have pointed to:

1. The growing importance of rents in residential real estate.
2. The vast upsurge in the share of corporate assets that are “intangible.”
3. The huge growth in the complexity of regulation, which favors large firms.

Kevin Erdmann did some very important posts on the share of income going to capital, which haven’t gotten anywhere near the attention they deserve. Here are a few excerpts, but read his whole series of posts:

We start with Profit (the blue line at the bottom). I have extended the time frame further back. The green line represents all returns to corporate capital, both to debt and equity. The debt portion peaked in the early 1980’s when corporate leverage was at its highest. When we make this correction, we find that corporate returns to capital have been flat for 40 or 50 years. If we add in proprietors’ income, we find that returns to capital have been flat or declining for a century. From 1929 to about 1985, there was a trend of profit claims moving from proprietors to creditors. From 1985 to the present, there was a trend of profit claims moving from creditors to equity owners. But, there is no trend of increasing total returns to capital over the past 30 years.

So in recent decades the rising equity income is offset by falling interest income, leaving total income to the corporate sector fairly stable, as a share of GDP. But then why is labor losing out? It turns out that more income is going to the residential real estate industry, but it’s implicit income from rents:

First, this is a little tricky, because 60% of American households own their homes. So, in effect, this is a measure of rent we are paying ourselves. Or, put differently, this is a measure of the income share we capture because home ownership tends to provide excess returns.

The trend in Compensation has dropped from about 57% in 1970 to about 53% – a 4% drop. But, the trend in Rent + Compensation has dropped from about 59% to 57%. Rental income explains about half the drop in Compensation Share, and in fact, accounts for more than all of the drop in Compensation Share since the previous low point in 2006.

To the extent that Rental Income supplements Compensation, this income is probably distributed mostly to middle and upper-middle class households. So, both the level and the distribution of household Compensation Share are probably helped by reducing excess returns to Rental Income.

Boston has extremely tight regulations on building, and a strong high tech economy. Put the two together and you have extremely high rents. And it wasn’t like that when I moved to Boston in 1982. They had lost a lot of their older industries, and people were moving away to the Sunbelt. Rents were not all that high.

What about corporations? We all know that the capital-intensive businesses of yesteryear like GM and US steel are an increasingly small share of the US economy. But until I saw this post by Justin Fox I had no idea how dramatic the transformation had been since 1975:

The rise of companies like Apple, Facebook and Uber affect the economy in two ways. Intellectual property rights create more monopoly power than manufacturers of TVs and refrigerators had back in the 1960s, and this boosts corporate profits. In addition, the individuals with the creative ideas and/or the financiers who picked the winners in the high tech race can make much larger personal incomes than a CEO at an appliance maker in the 1960s. How hard is it to figure out how to make washing machines? How hard is it to figure out the next WhatsApp? These are totally different skills.

However, I’d guess that it’s not just about high tech. We’ve also seen companies like Starbucks do increasingly well against the local corner coffee shop. There could be lots of reasons for this, but one might be the rapid growth in regulations. When regulations are highly complex, there are enormous economies of scale in dealing with the complexities. This favors larger firms. And as this article at Free Exchange points out, anything that favors the growth of larger firms tends to increase inequality:

The standard explanation says that technology plays a big role: modern economies require more skilled workers, raising the pay premium they can demand. A new paper* by Holger Mueller, Elena Simintzi and Paige Ouimet adds a new and intriguing wrinkle to this: the rising size of the average firm. Economists have long recognised that economies of scale allow workers at bigger firms to be more productive than those at smaller ones. That, in turn, allows the bigger firms to pay higher wages. This should not, in theory, cause a rise in inequality. If the chief executive and cleaner at a larger firm are both paid 10% more than their counterparts at a small firm, the ratio between their wages–and thus the overall level of inequality–should remain the same.

But the paper shows that the benefits of scale are not shared equally among all workers. Using data on wages at British firms, they divide workers into nine groups according to how skilled they are. Over time, they find that the proportional difference in wages between the groups grows as firms get bigger. This trend is driven entirely by a rising gap between wages at the top compared with the middle and bottom of the distribution. As the authors note, this is very similar to the trend in income inequality in America and Britain as a whole since the 1990s, when pay for low and median earners began to stagnate (see chart).

What do all three of these explanations have in common? Regulations. Building restrictions are increasing rental income as a share of national income. Intellectual property rights are barriers to entry that tend to create a winner-take-all situation (although other factors like network effects also play a role). And other types of regulations (financial, human resources, etc.) are especially burdensome for small firms, and this favors the growth of inequality-intensive large firms.

All of these changes have hit Boston in a big way since 1982. Boston has itself become more unequal, and it’s also moved further ahead of the American average income. Building restrictions here are almost comically excessive (which means that living standards aren’t that high, despite the high incomes). Industry is dominated by knowledge-intensive sectors.

I would not argue that these forces explain everything about inequality. My first full-time job at St. Bonaventure (assistant professor) paid $19,300 in 1981. I’d guess that was less that autoworkers made back then. Today econ professors start at almost $100,000 at some schools, whereas newly hired autoworkers are forced to accept much lower pay than the more senior members of the UAW. Technology and trade have widened the gap between blue-collar workers and professionals. But two of the types of inequality that are most frequently discussed are very much impacted by these trends; the falling share of national income going to labor, and the rising share of labor income going to the top 1%.

Given the three factors cited above, the only mystery I can see is why the inequality hasn’t gotten even greater.

Update: I forget to mention Matt Rognlie’s excellent work in this area. He makes some of the same points about rental income that Erdmann made.

Update#2: I also forget to mention Scott Winship.