In recent decades, income inequality in the United States has soared. This isn’t the first time. Such high levels of wealth concentration have happened before in U.S. history, most notoriously in the 1920s but also several times in the nineteenth century. Like during these previous periods, the economy today seems to grow only when debt is also growing quickly.

This is no coincidence. In any economy, income inequality tends to reduce consumption and force up the savings rate by effectively transferring income from those who mostly consume to those who mostly save. There are times when this tendency can be positive for growth, but other times it can be negative, depending mainly on whether or not investment in the economy is constrained by low savings and expensive capital.

Specifically, higher savings can boost growth when an economy has less investment than businesses desire because of insufficient savings. In such cases, by saving income that would otherwise have been consumed, the rich create the higher savings that businesses use to boost investment, driving the economy to grow more quickly and wealth eventually to trickle down to even the poorest workers. This was the case for much of U.S. history, when savings were inadequate and investment was constrained mainly by the high cost of capital.

For economies in which investment is constrained not by the lack of savings but mainly by weak demand, however, rising income inequality can actually suppress investment and reduce growth by reducing consumption. This point is very poorly recognized by most economists, but to understand why this is true, it is enough to see that there are basically two kinds of investment:

  • Investment that is not sensitive to changes in demand. This kind of investment—typical of underinvested economies (usually developing countries) or underinvested sectors (such as a new technological sector in its early stages) in which there is not enough investment to satisfy existing demand—is limited by the high cost of capital, so when additional savings cause the cost of capital to decline, businesses will increase their investment.
  • Investment that is sensitive to changes in demand. In such cases, typical of advanced economies, businesses justify investing in additional productive capacity only when they expect demand for their products to grow.

Much investment in the United States is of the second type,1 meaning that if increases in income inequality in advanced economies reduce expected consumption growth, businesses typically respond by cutting back on investment. This is probably what happened in Germany after the Hartz reforms of 2003–2005 and what has been happening in the United States and other advanced economies over the past several decades. Because of weak growth in demand, business investment has grown more slowly than most economists would have liked, and even then much of it has occurred in the high tech sector (which in its early stages is more like the first kind of investment than the second). Demand growth would have been even weaker if rising household debt and fiscal deficits hadn’t boosted consumption.

The seeming paradox under these conditions is that rising income inequality cannot raise total savings even as it raises the savings of the rich. This is because in a closed economy, like the global economy, savings by definition must equal investment, and if investment does not rise, savings cannot rise.

This dynamic can also apply to open economies, like the United States, in which the capital account is driven mainly by foreign investment decisions. Given the United States’ status as the automatic recipient of nearly half of the world’s excess savings, this condition is even stricter: savings must grow more slowly than investment whenever there is an increase in the current account deficit, which is itself determined largely by the need for the rest of the world to find a safe haven for its excess savings.2 Because most rich countries and some developing countries are unable to find productive use for their savings, they export the excess to countries like the United States, whose deep, flexible, and well-governed financial markets make them the best places to park excess savings.

This is why U.S. investment exceeds U.S. savings, with the gap between the two determined largely by foreign investors.3 To put it another way, the United States must reconcile the tendency of income inequality to drive up the savings of the rich and the tendency of investment growth to decline due to the associated weakness of consumption growth. Because savings cannot rise more quickly than investment—and this is just a basic accounting condition that cannot be violated—more savings among the rich must suppress savings in another sector.

There are broadly two ways an increase in income inequality can suppress savings elsewhere in the economy:

  • It can cause unemployment to rise and GDP to contract. In this case, the downward pressure on consumption caused by income inequality can close factories and businesses, which must then lay off workers, who in turn dis-save.
  • Or it can facilitate more household or government debt to maintain GDP growth. In this case, to counter rising unemployment, either the Federal Reserve can loosen money conditions, and so encourage private debt, or Washington can increase public debt by expanding the fiscal deficit. Debt is simply negative savings.

Put differently, without rising debt, the United States would suffer rising unemployment and a contracting economy. Understanding this makes it easy to see why much conventional economic thinking fails to make the obvious connection between income inequality, trade imbalances, debt, and unemployment. Economists’ models still implicitly assume that investment is constrained by relatively scarce savings, but—unless perhaps the United States were to implement a major infrastructure spending program—clearly this assumption no longer applies.

This explanation is not new. Amid the extreme wealth concentration of the late nineteenth century, the British economist John Hobson and an American contemporary named Charles Arthur Conant understood and explained the dynamic. The same principle was systematically elaborated nearly four decades later, during another period of extreme wealth concentration, by John Maynard Keynes. At the same time, Friedrich Hayek—whose contempt for rich people did not prevent him from providing the most rigorous argument as to why, given the unequal distribution of business ability, spontaneous accumulations of wealth are a condition of well-functioning markets—justified the accumulation of wealth, not rising wealth concentration, which tends to impede the disruptive nature of markets.4 In a well-functioning market economy, for different reasons, both Keynes and Hayek argued that the income of the poor should grow more quickly than that of the rich, not the other way around, as has been the case for decades.5

Yet the sharpest and most insightful exposition, also made in the 1930s, was by another conservative economist, Marriner Eccles, chair of the Federal Reserve Board. He explained the relationships this way in his memoirs:

By taking purchasing power out of the hands of mass consumers, the savers denied to themselves the kind of effective demand for their products that would justify a reinvestment of their capital . . . That is what happened to us in the twenties. We sustained high levels of employment in that period with the aid of an exceptional expansion of debt outside of the banking system. This debt was provided by the large growth of business savings as well as savings by individuals, particularly in the upper-income groups where taxes were relatively low.

In the United States of the 1920s, the invention of new forms of consumer lending, powered by soaring gold reserves, drove the economy in the way Eccles described, until debt could no longer grow enough to offset the growth in savings set off by extreme wealth concentration. When this happened, during the three years after the 1929 stock market crash, surging unemployment replaced surging debt as the only way to balance the high savings of the rich.

As Eccles went on to explain in his 1933 Congressional testimony:

It is utterly impossible, as this country has demonstrated again and again, for the rich to save as much as they have been trying to save, and save anything that is worth saving. They can save idle factories and useless railroad coaches; they can save empty office buildings and closed banks; they can save paper evidences of foreign loans; but as a class they cannot save anything that is worth saving, above and beyond the amount that is made profitable by the increase of consumer buying. It is for the interests of the well-to-do—to protect them from the results of their own folly—that we should take from them a sufficient amount of their surplus to enable consumers to consume and business to operate at a profit. This is not soaking the rich; it is saving the rich.

The United States is in the same position today: surging debt is the only way to balance the high savings of the rich without resorting to higher unemployment. While rising concentrations of wealth can create adverse social and political outcomes, they are still justified by some economists on the grounds that, by boosting savings, income inequality leads to higher investment, which in turn leads to the higher growth that eventually benefits even the poor. The point, however, is that this can only happen under economic conditions that no longer apply to the United States (or to other advanced economies).

Today, because it is weak demand, not high costs of capital, that restrains business investment, income inequality does not lead to higher investment. On the contrary, it leads to slower growth, more debt, and perhaps even less investment. If the United States were to reverse the conditions that since the late 1970s have indirectly encouraged rising income inequality, the U.S. economy would grow more soundly with less debt and, as Eccles promised, growth would eventually trickle up even to benefit the rich.

My Twitter account is @michaelxpettis. Aside from this blog, I write a monthly newsletter that covers some of the same topics covered on this blog. Those who are interested in receiving the newsletter should write to me at, stating affiliation.


1 That is not to say that there is no potential investment of the first type in the United States. U.S. infrastructure is clearly very poor in some respects, and it could use significant upgrading, at the very least. This would be needed investment no matter how quickly or slowly demand grew. The reason much of this investment is not being made is not because the United States is unable to fund the investment at an acceptable cost, but rather because political considerations prevent the country from doing so.

2 The U.S. current account deficit is equal by definition to its capital account surplus, which consists of the net amount of foreign capital inflows into the United States. Because the country acts as the stabilizer that automatically balances global demand for savings with global supply, the United States has little control over its capital account and consequently little control over its savings rate. To see how this stabilizing process works, note that until the 1970s, when the global economy was being rebuilt from the ravages of two world wars, the global demand for savings exceeded the global supply, and as a result the United States exported excess savings while running seemingly permanent current account surpluses. Once this rebuilding was more or less complete, and investment needs among the world’s rich-again developed economies dropped relative to their higher savings, the world suffered from excess savings. As a result, the United States became an importer of foreign excess savings, while the current account swung automatically into seemingly permanent deficits. This almost certainly is not a coincidence, and as long as foreigners place their excess savings into the U.S. economy, its savings rate must be below its investment rate.

3 This is also why an expanding U.S. current account deficit typically raises unemployment or debt, while a contracting U.S. current account deficit typically lowers unemployment or debt. Mainstream economists often have trouble understanding this mainly because, usually without realizing it, they implicitly assume that savings are generally scarce and constrain productive investment.

4 “We ought to be much more aware that if we regard a man as entitled to a high material reward that in itself does not necessarily entitle him to high esteem,” Hayek once explained.

5 At the risk of excessive simplifying, Keynes was interested in understanding the systemic tendencies toward wealth concentration and its impact on demand, while Hayek worried more about how persisting concentrations of wealth might create rent-seeking and lead to legal privileges and government-secured advantages.