Income inequality is harming the economy. Most people spend whatever extra money they earn. The rich, however, are disproportionately likely to save any additional income, which means that income concentration saps consumer demand and threatens the viability of new investments—just as in the 1920s and 1930s.

We used to know this. Marriner Eccles, the former Federal Reserve chairman, explained in his 1951 memoir that the Great Depression had been preceded by an enormous concentration of American wealth, as if by a “great suction pump,” which took purchasing power away from those who spend most of what they earn on goods and services and transferred income to those who are most likely to save it.

That, Eccles argued, ultimately reduced the value of capital. After all, “as mass production has to be accompanied by mass consumption, mass consumption, in turn, implies a distribution of wealth to provide men with buying power equal to the amount of goods and services offered by the nation’s economic machinery.” Factories are worthless if no one can afford to buy what they produce.

The rising concentration of income meant that consumption could grow only if it were boosted by borrowing. “High levels of employment” were possible in the 1920s only “with the aid of an exceptional expansion of debt.” After the 1929 stock market crash, however, the private sector’s ability to borrow collapsed. So did consumer spending. Unemployment soared.

Income inequality in the U.S. has risen sharply since the 1980s, and especially since 2000. Until 2008, this had coincided with rapid increases in private indebtedness. In 1984, household debt was worth about 64% of disposable personal income. By the end of 2007, indebtedness had more than doubled to 134% of income. The bulk of this increase occurred during the housing bubble, as rising valuations and loosening credit standards enabled American homeowners to borrow against their homes to pay for everything from hot tubs to cars to family vacations. Alan Greenspan and James Kennedy estimated that this boosted consumer spending by 3% per year from 2001 to 2005.

The debt binge did not cause consumption to grow unusually quickly: average consumer spending grew at about the same steady pace, as it had from 1947 to 2000. The debt was sufficient only to prevent consumption from stagnating in line with consumer incomes. That lower spending path would have caused much of the productive capacity of the country to lose its value. From this perspective, the financial crisis was simply the recognition of past losses. Since then, both debt and spending have grown sluggishly.

The economists Michael Kumhof, Romain Rancière, and Pablo Winant drew this parallel in a recent paper. In both the 1920s and the years before 2008, “top earners allow bottom earners to limit the drop in their consumption” by lending to the masses. However, “the resulting large increase of bottom earners’ debt-to-income ratio generates financial fragility that eventually makes a financial crisis much more likely.” “The crisis,” in their view, is therefore “the ultimate result” of rising inequality.

This is not to say that inequality is always bad. When resources are scarce, it can be necessary to limit consumption to prioritize worthwhile investment projects. This was the case when the U.S. was a developing country in the 19th century, for example. Back then, income concentration helped make everyone richer, as new wealth “trickled down” to even the poorest members of society.

Read the rest of Michael Pettis’ article at Barron’s