When nations compete by suppressing wages, globalization can leave us all poorer

RECOMMENDED READING
CHIPS Won’t Help China
Cutting China Tariffs Will Offer No Respite From Rising Prices
Trading It All Away

“China will compete for some low-wage jobs with Americans,” lectured Nobel laureate Robert Solow from the White House podium, amidst the U.S. debate over China’s ascension to the World Trade Organization. “And their market will provide jobs for higher wage, more skilled people. And that’s a bargain for us.” More than 20 years later, economists and policymakers are still searching for that bargain. Belatedly, they are discovering that whether or not trade benefits the global economy or any particular nation depends, like most things in economics, on the specific underlying economic conditions.

When it comes to trade, the key conditions are the approaches that countries take in their quests for international competitiveness. Trade can directly boost production and indirectly boost demand, so that the global economy is generally better off. But trade can also make the global economy worse off by directly constraining demand and so indirectly constraining production. The outcome depends on whether a country’s higher export revenues are recycled into higher consumption and imports or into higher savings.

In the traditional view, international trade allows a country or region to specialize in producing things that it can produce relatively more efficiently than its trading partners, and so trade shifts production to the locale in which a given amount of labor and capital yields the greatest output. In a world of scarce inputs, this allows the global economy to maximize production. According to that model, by definition, anything that impedes or distorts free trade, whether a regulation or tariff or quota, reduces global production. Mainstream economists accept that the benefits of free trade may be badly distributed, even to the point at which free trade can leave some sectors worse off, but this, they insist, is a distribution problem that should be solved politically. The world, they argue, is always collectively worse off when any country intervenes in the free flow of goods.

This goes for each individual country’s policy choices too. “The economist’s case for free trade is essentially a unilateral case,” Nobel laureate Paul Krugman wrote in 1997. “A country serves its own interests by pursuing free trade regardless of what other countries may do.”

"What happens if, instead, the country achieves competitiveness by suppressing wage growth relative to increases in productivity, which means paying workers a lower share of what they produce than its trade partners pay their workers? This creates a form of “bad competitiveness,” which results in weakening demand for goods and services."

Lurking behind these convictions is a series of assumptions that are simply not true, at least not in the regime that currently governs global trade and capital flows. In a well-functioning trading system, a country should become internationally more competitive by exploiting natural and geographical advantages and investing in research, manufacturing facilities and infrastructure to increase the productivity of workers. This “good competitiveness” results in rising production, and as the benefits of rising production are distributed to workers in the form of rising wages and benefits, workers are then able to increase their demand for other goods and services.

But that is not the only option. What happens if, instead, the country achieves competitiveness by suppressing wage growth relative to increases in productivity, which means paying workers a lower share of what they produce than its trade partners pay their workers? This creates a form of “bad competitiveness,” which results in weakening demand for goods and services, which in turn either reduces global production, or requires surging debt to maintain demand and production at its existing level. Perhaps that rings a bell, because it is the world we live in.

How to Become “Competitive”

In our hyperglobalized world, in which transportation, communication and financial transaction costs have dropped to almost zero, production shifts easily to countries in which wages are relatively low. What matters—and this is a point often missed in the debate about trade—is not the absolute level of wages. What drives competitiveness is the total compensation households receive in the form of wages, salaries, investment income and transfers relative to their productivity levels.

A simple example illustrates the point: Suppose workers in the United States earn $20 per hour in a t-shirt factory while in China they earn $1 per hour. Where should the production locate? The casual observer might say it will move to China, which has “cheap labor.” But if American t-shirt workers are earning 20 times more than their Chinese counterparts because (as economists would assume) they are 20 times as productive, then the Chinese labor is not actually cheaper at all. There is no competitive advantage in paying $1 per hour instead of $20 if it takes 20 hours to get the same output.

When we say colloquially that production shifts to China in that scenario because of cheap labor, what we mean is that the labor costs one twentieth as much and is perhaps a tenth as productive. In America, you have to pay $20 to get $20 of output. In China, you can pay $1 and get $2 of output. The “competitiveness” of China’s “cheap labor” is in fact just the opportunity for the producer to capture more of the value for itself.

That is why “competitive” wages can be an advantage in both high-wage countries, like Germany, Japan, the Netherlands, and South Korea, and in lower-wage countries, like China and Vietnam. Their workers and middle classes receive less relative to their levels of productivity than do the workers of their trading partners. At the macroeconomic level, this means that the ratio of household income to GDP is lower than among their trading partners.

There are many ways to reduce household income relative to GDP, all of which depend on policies that explicitly or implicitly transfer income from households to businesses. Most obviously, a nation can keep wages “competitive” by suppressing wage growth directly, as happened in Germany with the “Hartz” labor reforms of 2003–05. These reforms cut worker protections and reduced unemployment benefits, setting off a several-year period during which wage growth slowed significantly relative to the growth in worker productivity, effectively transferring income from workers to businesses.

"It is strange, then, that economists complain about the distorting effect of tariffs and other direct trade interventions while ignoring all the other indirect mechanisms that do exactly the same thing."

There are other ways. Policymakers can achieve the same effect by reducing pensions, suppressing the returns households receive on their savings, or undermining the social safety net. In the 2000s, for example, China repressed interest rates on household savings, imposed residency restrictions that severely weakened the bargaining power of migrant workers, and gradually dismantled the so-called “iron rice bowl” of lifetime job security, forcing down the share of GDP retained by Chinese households from roughly 65% in the mid-1980s to just over 50% by 2010. As that happened, Chinese workers consumed a declining share of what they produced and China’s export “competitiveness” automatically increased.

Policies that weaken labor movements and other sources of worker power are also economic tools for shifting national income from labor to capital. Less intuitively, tolerance for environmental degradation reduces operating costs for businesses and increases health costs for households, creating an implicit transfer from households to businesses that makes the latter more “competitive” in the global marketplace.

Depreciating the currency is yet another way of increasing international competitiveness by forcing households to subsidize businesses, although the mechanism is often misunderstood. A cheaper currency is basically a transfer from net importers, who must pay more for the products they consume, to net exporters, who benefit from higher foreign prices. Because all households are effectively net importers, and because most net exporters are manufacturers, farmers, and commodity producers, depreciating the currency is just another way of transferring income from households to subsidize producers. The same can work with interest rates. In economies like Japan’s in the 1980s and China’s in the 2000s, artificially repressing interest rates caused a direct transfer of income from households, who were net savers, to manufacturers, state-owned enterprises, and local governments, who were net borrowers. Needless to say, this borrowing subsidy, paid for by Japanese or Chinese household savers, raised the international “competitiveness” of their manufacturers.

Even tariffs should be viewed through this lens. When a country imposes tariffs on a foreign import, it boosts the profits of domestic manufacturers at the expense of domestic consumers of the product. In aggregate, tariffs work like all the other mechanisms, by transferring income from household consumers to the manufacturing sector. It is strange, then, that economists complain about the distorting effect of tariffs and other direct trade interventions while ignoring all the other indirect mechanisms that do exactly the same thing. Policies that put downward pressure on the currency, reduce interest rates on household savings, undermine the social safety net, encourage environmental degradation, weaken labor unions, and so on, leverage the same dynamic as tariffs: they increase a country’s international competitiveness and suppress domestic demand by forcing households to subsidize production.

Subsidies, Savings, and Surpluses

One of the greatest sources of confusion in the discussion of trade is the role of cultural and household preferences in a country’s savings rate. Many analysts point approvingly to what they believe is a culture of thrift in high-savings countries like Germany, China, and Japan, while making invidious comparisons with the spendthrift ways of low-savings countries like the United States, Spain, and England.

But how much a country saves has little to do with culture and is mostly a function of the distribution of income between households and other sectors of economy. By definition, savings comprise everything that is produced but not consumed. Because different sectors within the economy have different consumption propensities, a country’s savings rate—or the obverse, which is its consumption rate—is determined largely by how these sectors share the national income. Workers and middle-class households, for example, typically consume most of what they earn, and save very little. Businesses consume none of what they earn or, put another way, save all of it. The rich consume very little of what they earn and save most of it. Most government revenues, rather than being spent directly on consumption, are either recycled to other sectors via transfer payments or else invested in development of public assets. This is especially the case in countries like China, where the government allocates a substantial share of national income to investment in infrastructure and state-owned or -subsidized enterprises.

What determines a country’s overall savings rate is the way in which income is distributed among these different groups. Policies that transfer income from workers or the middle class to the rich or to businesses, for example, automatically force up the national savings rate regardless of culture. That is why it is not a coincidence that high-savings countries like Germany, China, and Japan are also countries in which households retain a lower share of GDP than other countries at similar stages of development.

"Whether the global economy benefits from surplus countries exporting their excess savings depends crucially on where those excess savings go."

Nor is it a coincidence that high-savings countries run persistent trade surpluses. Total domestic demand consists of household and government consumption plus business and government investment. Because businesses invest mostly to serve domestic demand, if domestic consumption is low, this often means that private business investment is also low. The consequence of households receiving a low share of what they produce is that domestic demand tends to be weaker than otherwise.

Normally, weaker demand should mean slower growth, but this is why trade surpluses matter so much to these economies. Exporting the excess production is what allows them to continue manufacturing, even when they cannot absorb what is produced domestically. Exporting excess production, in other words, is the same thing as exporting excess savings, and countries with trade surpluses are by definition exporting their excess savings to the rest of the world. Or, looked at another way, countries with trade surpluses are effectively exporting their domestic demand deficiencies to the rest of the world.

Whether the global economy benefits from surplus countries exporting their excess savings depends crucially on where those excess savings go. A rapidly growing developing economy may have very high domestic investment needs that are constrained by the scarcity of domestic savings. Exporting savings to such countries can allow them to invest more, in which case the deficient demand of the surplus country is matched by higher demand abroad as investment in developing countries rises. This leaves the world better off because both total global production and total global demand are higher, even with a demand deficiency in the surplus country. This was the story in the 19th century, when the United States was rapidly developing and British and other European capital inflows allowed it to invest far more than it could have done just with domestic savings.

The Problem of Unbalanced Trade

But—and here is what many economists miss—when a country exports excess savings it doesn’t necessarily lead to higher investment elsewhere. When excess savings are exported to advanced economies, like the United States today, domestic investment doesn’t rise. For the past several decades, American businesses have been reluctant to invest mainly because of weak expected demand for their products, not because they lacked access to cheap capital. On the contrary, capital has been cheap and plentiful, and businesses sit on large hoards of cash which they use mainly for share repurchases or acquisitions of existing business.

Under these conditions, foreign inflows of excess savings won’t cause U.S. investment to rise. But as a net recipient of foreign capital inflows, the U.S. must nonetheless run a current account (trade) deficit, which is another way of saying that its investment must exceed its savings. The conclusion seems surprising but is inescapable: if net capital inflows do not cause investment to rise, they must cause savings to fall.

For many Americans, and even many economists, this is totally counterintuitive. We are used to thinking that the U.S. controls every aspect of its economic fate, and that the very low American savings rate is wholly a function of faulty American attitudes towards thrift.

But this simply isn’t true. A country, like the United States, with open capital markets, high-quality governance, and a flexible financial system, has little to no control over the extent of net capital inflows, which occurs as savers abroad purchase assets from Americans. Foreign central banks looking to manage their currencies, wealthy oligarchs protecting their wealth, Asian fund managers scrambling for investment opportunities, European speculators playing currency and interest games, along with an enormous variety of other investors who collectively export 30–50% of the world’s excess savings into the U.S. are determined to convert that savings into American assets. So long as they are willing to keep bidding higher, they will find willing sellers, and the capital will flow stateside.

If the U.S. cannot control the extent of net capital inflows, then by definition it cannot control the gap between domestic investment and domestic savings. In the 19th century it was mostly American domestic investment that fluctuated in response to changes in net foreign capital inflows. Today, when investment is constrained by weak demand, it is mostly American savings that must fluctuate in response to changes in net foreign capital inflows.

"As long as countries can improve their international competitiveness by directly or indirectly suppressing wages and reducing domestic demand, and as long as they can externalize the resulting cost by exporting savings abroad, the incentive for countries to increase their international competitiveness ends up depressing global wages and global demand."

Net capital inflows, in other words, force down American savings, and this can occur through several channels. If foreign capital inflows cause the U.S. dollar to strengthen, for example, the stronger dollar effectively transfers income from manufacturers (net exporters) to households (net importers), and so raises the consumption share of GDP. If a stronger dollar causes foreign manufacturers to price American manufacturers out of business, and these American manufacturers respond by laying off workers, this also lowers the American savings rate (unemployed workers have a negative savings rate).

What if the Federal Reserve tries to counter this rise in unemployment by lowering interest rates? In that case, lower interest rates encourage households to borrow more and so maintain consumption levels. Similarly, Washington may try to counter the rise in unemployment by expanding the fiscal deficit. Because debt is negative savings, in either case the American savings rate must decline.

Still another channel is if foreign purchases of American stocks, bonds, and real estate cause their prices to rise, triggering a wealth effect in which American asset-holders feel more secure with their existing savings. This will encourage them to increase their consumption and reduce the amount of their income they save.

The point is that as long as countries can improve their international competitiveness by directly or indirectly suppressing wages and reducing domestic demand, and as long as they can externalize the resulting cost by exporting savings abroad, the incentive for countries to increase their international competitiveness ends up depressing global wages and global demand. This is especially a problem for countries like the U.S., whose deep, flexible, open and well-governed capital markets mean that they must automatically absorb these excess savings, in which case the American savings must decline to accommodate the flood of foreign savings. This decline must occur either in the form of higher unemployment, higher household debt, or higher fiscal deficits.

The problem is not free trade per se. The problem is a system in which trade depresses wages rather than raises productivity. The United States should take the lead in reforming this very unbalanced global trade regime, not by turning against trade but rather by eliminating the conditions that allow grossly unbalanced trade. Either Washington should pioneer new trade agreements that directly restrict the ability of countries to run large and persistent surpluses, much as John Maynard Keynes proposed during the Bretton Woods conference, or it should unilaterally refuse to continue playing its role of absorber of last resort of global excess savings and instead force its own trade and capital flows into balance.

When trade between countries directly boosts production and indirectly boosts demand, the global economy is better off. When it directly constrains demand and so indirectly constrains production, not only is the global economy worse off, but countries like the United States bear an especially heavy cost to keep the system afloat. The purpose of international trade should be to maximize overall productivity and, with it, to increase welfare. It should not allow individual countries to maximize domestic production at the expense of their trading partners.

Michael Pettis
Michael Pettis is a senior fellow at the Carnegie Foundation and a finance professor at Peking University. His most recent book is Trade Wars Are Class Wars (Yale University Press, 2021).
@michaelxpettis
Recommended Reading
CHIPS Won’t Help China

American Compass executive director Oren Cass argues that demanding perfect legislation is a convenient excuse for voting no, and a standard by which everyone would always vote no.

Cutting China Tariffs Will Offer No Respite From Rising Prices

American Compass executive director Oren Cass makes the case against rolling back tariffs on China in response to inflation.

Trading It All Away

Adapted from remarks delivered by Senator Marco Rubio on the 20th anniversary of China’s ascension to the WTO.