For over one hundred years, economists have understood the economic consequences of income inequality. They have worked out how, under differing specific conditions, income inequality can foster either unemployment or growth. They also understand how international trade distributes unemployment, even at times forcing countries into beggar-thy-neighbor policies. By the 1890s, for example, John Hobson in England and Charles Arthur Conant in the United States, the former what Americans today would call a liberal and the latter a conservative, had largely worked out the consequences of income inequality in ways that accorded fully with contemporary experience and historical precedents.

Income inequality is such a contentious topic, however, that it is still hard to separate myth from reality, even though the consequences that economists have fleshed out over the last century follow from basic economic identities. Ordinary people consume a larger share of their incomes than do the wealthy, so rising income inequality reduces the consumption share of GDP, which automatically forces up the national savings rate. The strongest argument in favor of allowing and even encouraging income inequality has always stressed inequality’s positive impact on savings.

Michael Pettis
Pettis, an expert on China’s economy, is professor of finance at Peking University’s Guanghua School of Management, where he specializes in Chinese financial markets.
More >

A similar form of income inequality occurs when businesses or states retain a growing share—and households a declining share—of GDP. This phenomenon has been especially noticeable in Germany and China over the past fifteen years as both countries have forced down the household share of GDP by implementing policies aimed at making domestic businesses more competitive in international markets. These policies, intentionally or unintentionally, have also forced down the consumption share of GDP, raising savings rates.

Rising income inequality and a declining household share of GDP, together referred to as “repressing household income,” both increase national savings rates. But an increase in the national savings rate is not the only consequence of these forces.

Savings must equal investment. If repressing household income causes the savings rate in one part of the economy to rise excessively (a “savings glut”), logically it must also cause the investment rate to rise or the savings rate elsewhere to decline (or some combination of both).

How can repressing household income cause investment to rise? If productive investment had been previously constrained by low savings, the answer is obvious. By increasing savings, repressing household income will foster more investment in productive projects. More productive investment benefits everyone, even if it benefits the wealthy more. This model is generally known as trickle-down economics.

In many undeveloped economies, productive investment may indeed be constrained by low savings. This is why developing countries often either turn to foreign savings to fund growth, as Argentina did in the 1990s, or repress consumption by reducing the household income share of GDP, as China did more recently.

Developed countries, however, generally do not face a savings constraint. If they fail to fund all productive investments, it is for other reasons—political gridlock, for example, or the difficulty of capturing externalities. For these countries, the higher savings associated with repressing household income will not result in more productive investment.

Former chairman of the Federal Reserve Board Marriner Eccles even suggested that during the 1930s, repressing household income reduced productive investment. “By taking purchasing power out of the hands of mass consumers,” he wrote in Beckoning Frontiers, “the savers denied to themselves the kind of effective demand for their products that would justify a reinvestment of their capital accumulations in new plants.”

Income Inequality and Bubbles

Mechanisms that force up the savings rate by repressing household income can still cause investment to rise, however, even if that investment is not productive. Hobson and Conant showed that if savings cannot find productive outlets, it will flow into speculative assets at home or abroad and can even create bubbles. These flows usually generate debt, so unproductive speculative investment can only continue until debt levels become excessive, after which it must stop.

But savings can balance in another way. If investment is not enough to absorb the higher savings caused by efforts to repress household income in one part of the economy, then logically savings elsewhere must decline. There are two ways this can happen.

Soaring asset prices driven by the investment of excess savings into speculative assets can set off a consumption boom either at home or, if excess savings are exported, abroad. If higher asset prices make ordinary households feel wealthier, their consumption might rise faster than their income. This increased consumption would offset the higher savings levels that resulted from repressing household income.

This happened most obviously in Europe in the past two decades. Germany swung from having savings below domestic investment in the 1990s to having an excess of savings in the subsequent period. Excess German capital poured into countries like Spain in the form of bank loans, and this influx of capital set off local stock market and real estate prices in Spain, making ordinary Spaniards feel wealthier and unleashing both a consumption boom and a real estate bubble.

A savings glut, therefore, does not require that overall savings rise, only that savings rise excessively in one part of the economy—a point that is often missed. If excess savings set off excess consumption elsewhere, usually unleashed by stock market and real estate booms, overall savings do not rise because of the wealth effect. Germany’s excess savings, in other words, spurred excess consumption in smaller countries like Spain, which meant that at the European level there was no significant rise in the savings rate.

But this resolution of the savings-investment imbalance is temporary. Excess consumption can continue only as long as markets surge. The game eventually must stop, in part because the consumption boom is supported by rising debt.

This eventuality leaves only one other way to reduce savings—unemployment must rise. As Eccles predicted, waning demand resulting from repressing household incomes causes inventory to pile up, and factories respond by firing workers. Because unemployed workers must still consume, savings automatically decline when workers are fired.

Unemployment, in other words, is the other way to offset the impact of higher savings among the rich, and it often follows a consumption boom. This progression occurred in Europe, where a savings glut led first to a surge in consumption and later to an increase in unemployment.

This process is fairly automatic and does not require households in one part of the economy, such as Germany, to become thriftier or households in another part, such as Spain, to become profligate, which is why much of the moralizing that surrounds trade imbalances is beside the point. As the German government, working with the labor unions and large businesses, forced down German wage growth after 2000 to one-third of its previous level, the income, and with it the consumption, of ordinary German households dropped sharply relative to GDP. The result of this drop in German consumption was a rise in the national savings rate, although German households were no thriftier.

Germany could not balance higher domestic savings with more domestic investment, so it exported the excess savings to the rest of Europe. It did so in such large quantities that even though speculative investment soared in many countries, much of it into empty buildings, these countries simply could not increase investment fast enough to absorb Germany’s excess savings. Because these countries were unable to depreciate their currencies (which would have protected their workers at the expense of German workers), their savings rates had to fall. No alternative was possible.

One way these savings rates might have fallen was through higher unemployment, as the tradable goods sectors in these countries faced withering German competition. But that did not happen initially. Instead, interest rates dropped sharply after these states joined the euro, and the wealth effect that accompanied the resultant soaring asset prices ignited a domestic consumption boom that went on until debt reached crisis levels in 2008. At that point, the only way for savings to adjust downward was through a surge in unemployment, which is exactly what happened in most peripheral European countries after 2008.

Overcoming the Pattern

It is hard to escape the logical consequences of repressing household income. By forcing up the savings rate in one part of the economy, it can permanently increase productive investment if such investment had been constrained by low savings. Otherwise, repressing household income must result either in a temporary increase in speculative investment, a temporary surge in consumption (justified by the increase in speculative investment), or a permanent increase in unemployment.

These are the only possible consequences of a rise in savings. It is not a coincidence that in developed countries, periods of rising income inequality have often started with a jump in productive investment, later to be accompanied and reinforced by speculative investment bubbles and soaring consumer debt and ultimately to end in a surge in unemployment. It is a very common pattern.

So what can countries do to reverse the employment consequences of a savings glut? Unfortunately, free trade creates significant constraints on their abilities to react. If countries raise domestic wages or increase tax transfers, global demand rises but the international competitiveness of domestic producers declines, causing these producers to lose sales and fire workers. Raising wages at home, in other words, means creating jobs abroad and worsening the trade deficit.

Individual countries would be better served if they actually lowered wages, but this approach would only make the global problem worse. For this reason the world is stuck in a position much like it was in the 1930s: the global economy needs all countries together to increase global demand.

One way to do so is to increase consumption by raising the median household income.1 As this happens, more consumption will justify more production, thus encouraging profitable investment.

But no country can go first. On the contrary, reducing labor costs, perhaps by lowering wages, is the easiest way to reduce domestic unemployment. Spain, for example, is trying this approach. While lowering wages reduces global demand, countries with declining wages can claim a bigger share of the smaller total. What is good for individual countries, in other words, is bad for the world. So, like the beggar-thy-neighbor policies of the 1930s, the world is locked in competitive austerity.

The dirty little secret of international economics is that while government intervention in trade is bad for the global economy, it can often be good for individual countries, especially for diversified economies with large trade deficits and high unemployment. In these cases trade intervention nearly always leads to faster GDP growth and lower unemployment because it allows countries with large deficits both to increase their share of global demand at the expense of countries with trade surpluses and to raise median household income without suffering the consequences of reduced international competitiveness.

The arithmetic of excess savings and open trade is pretty straightforward, but the politics is complicated. A powerful constituency benefits from the repression of household income. And while there are certainly legitimate economic arguments in favor of higher income inequality and a greater state role, at least in specific circumstances, what is often lost in the debate is an understanding of the conditions under which repressing household income leads either to higher productive investment or to higher unemployment, as it has done in the United States and Europe today.

When the world is suffering from insufficient demand and excess capacity, international trade can itself prevent adjustment by forcing each country into a brutal race to repress household income further. In that case, support for free trade is likely to wither away. This shift, after all, has happened many times before during global crises—and the logic is hard to circumvent, even if trade intervention is nearly always taken to harmful extremes.

So while global free trade is almost always good for the world, there are times when it can prevent the necessary adjustments in individual countries from taking place. Now is one such time.

1 The other sustainable way is to increase productive investment, perhaps by engaging in substantial infrastructure spending in countries that are underinvested in infrastructure.