Growth From the Middle Out, and How It Works

Jared Bernstein is a senior fellow at the Center on Budget and Policy Priorities in Washington and a former chief economist to Vice President Joseph R. Biden Jr.

At the core of President Obama’s economics speech on Wednesday was the notion that as rising inequality diverts income growth from the many Americans in the broad middle class, their diminished buying power leads to slower growth.

It’s an intuitive idea with solid theoretical backing, but is it true?

The theory is simple enough and has its roots in basic microeconomics, specifically the decline in the marginal propensity to consume as incomes rise. That may sound tricky but it’s just common sense: an extra dollar that finds its way to an “income-constrained” (or “poor,” or these days even “middle-class”) person is more likely to be spent than saved.

Bureau of Labor Statistics statistics show, for example, that in 2012, households with income of $40,000 to $50,000 spent 92 percent of their after-tax income. Those with incomes above $150,000 spent 53 percent of their income.

Couple that fact with the fact that consumer spending as a share of gross domestic product in the United States is about 70 percent and you see both the problem and the motivation for the president’s framing. It’s interesting to note that Europe, where the consumption share of G.D.P. is 55 percent, is somewhat less vulnerable to this dynamic.

Another germane point is that back in the late 1970s, before inequality starting climbing, the United States spending share of G.D.P. was 62 percent. So we’ve become more unequal, and the part of G.D.P. that depends on spending has grown. That certainly seems like a recipe for slower growth.

And in the current moment, I agree with the president that it is. The fact that economic growth over this recovery has pretty solidly eluded the poor and middle class is one reason the United States is slogging along at subpar growth rates.

But what about the last recovery? Inequality grew sharply in the 2000s and did so at the expense of the middle class. Using the highly regarded household income series from the Congressional Budget Office, which factors in capital gains (very important when you’re measuring inequality), taxes and the value of government transfers, real middle-class incomes were up 13 percent, 2000-7, while top 1 percent incomes rose 31 percent.

Yet consumer spending wasn’t notably slow in those years. And if you include home-buying, spending was of course through the (overleveraged) roof.

And here’s where things get a lot more interesting. In this interpretation of recent events, higher inequality actually lifts middle-class spending, but it does so based on loose credit leading to the inflation of debt bubbles that ultimately damage growth even more than sluggish consumer spending.

With inequality squeezing paychecks, the only way for middle-class families to get ahead was to borrow. At the same time - and I’d assert this is also a dangerous symptom of wealth concentration and its impact on American politics - bank regulators and credit raters were asleep at the switch, paving the way for financial engineering that artificially depressed the price of risk. Put those toxic factors together and you get the economic shampoo cycle - bubble, bust, repeat - that’s by now all too familiar.

Is there any evidence for this chain of events? There is, from an interesting recent paper by the economists Barry Cynamon and Steven Fazzari. It’s by no means bulletproof; there are lots of moving parts to these arguments, and the fact that we’ve had debt bubbles in periods of low inequality is a substantive challenge to their hypothesis. But I find their circumstantial evidence convincing.

Mr. Cynamon and Professor Fazzari find that as the income shares of the top 5 percent soared relative to the bottom 95 percent, both the debt burden and the spending of the lower-income group grew quickly relative to that of the wealthiest households. The savings rates of the very wealthy increased over the 2000s expansion, while that of the bottom 95 percent fell sharply.

These dynamics drove much stronger consumer spending than one would have expected looking only at median income growth, inflated a huge housing bubble and ensured that once the bubble burst and the deleveraging cycle began, the United States was in for a deep and protracted recession.

My point is that the inequality affects growth through many channels. In demand-constrained periods like the present, it weakens consumer spending and growth because the small share of beneficiaries of what growth there is have lower consumption propensities than everybody else. But in debt-driven expansions, inequality appears to play an integral role in terribly damaging bubbles inflated by stagnant incomes, underpriced risk and leveraged consumption.

So I’m with the president. Let’s get back to broadly shared growth. It’s better for the economy and for most of the people in it. How to do that, especially with this Congress, is a very big problem for another day.