The Japanese economy has hardly grown for the past quarter century, following forty years of spectacular growth. Japan's gross domestic product (GDP), a measure of the size of its economy, was only 12 percent higher (after correcting for price changes) in 2015 than it was in 1997, a growth of only 0.6 percent a year over two decades. Such a period is called "secular stagnation," a term first brought into popular use by Harvard professor Alvin Hansen in the late 1930s, but recently revived by another Harvard professor, Lawrence Summers (2016), as well as others. The question posed by Summers and others is whether secular stagnation has extended beyond Japan to Europe and even to the world as a whole.
There are two quite different (but not incompatible) reasons for secular stagnation. The first concerns the "supply side" of an economy: its capacity to grow has become impaired, for example through a decline in the growth of the labor force not over-compensated by increases in labor productivity. The latter factor in turn may arise because productivity-increasing investment has reached limits and/or measurable technical improvements in products, and production processes are also limited. This seems to have been the case in Japan, where the labor force has declined for some years.
Robert Gordon (2016) has recently argued that the United States has entered a period of declining growth—if not literally secular stagnation—because of a prospective decline in the growth of its labor force and an exhaustion of sources of technical change that lead to significant productivity-increasing investments, such as occurred historically with the introduction of steam power, electricity, and the petroleum-based internal combustion engine, which revolutionized travel on land and by air. The third industrial revolution, in computation and communication based on the integrated circuit, may have been life transforming in many ways, but not by producing such dramatic increases in measured productivity and growth as in earlier periods.
Gordon's analysis operates very much within the context of existing measurement and national accounts, and in that framework he may well be correct. But what we should conclude from this is that the existing framework is inadequate to capture the many, often revolutionary and welfare-improving, technological changes that are taking place, especially but not only in the domain of medical care.
Take a stylized example to make the point: suppose an economy consists of only two sectors, manufacturing and education. We comfortably measure productivity increases in manufacturing, but we are so uncomfortable measuring them in education, despite tremendous increases in scientific knowledge in recent decades, that the U.S. official statisticians measure outputs by inputs, in effect assuming that productivity increases in education are zero. As this over-simplified economy grows through the productivity increases in manufacturing, labor will shift over time to the education sector, permitting its consumers to have more manufactured goods and more education. Its measured growth rate may steadily decline, even approaching zero if consumers choose to take a disproportionate share of their rising real income in the form of education. Yet their well-being is steadily increasing.
The general point is that we have much greater difficulty measuring adequately the real value of services than we did for agricultural and manufactured goods, and this difficulty seriously affects our measured growth rates as modern economies shift their expenditures from goods to services.
The second possible source of secular stagnation is inadequate growth in demand. In modern economies the decisions to save and to invest are typically made by quite different entities. In the end, saving must equal investment in a closed economy. But initially desired saving may not equal initially desired investment. What brings them together? Main-line economics texts focus on interest rates, which adjust to assure the ex post equality of saving and investment. That claim is questionable even in normal times, since much business investment seems not very sensitive to interest rates (as opposed to credit conditions); and it is highly doubtful at present when low-risk interest rates are near zero in all of the rich countries. (Negative interest rates might induce some investment, but negative interest rates have so far reached only governments, whose economic decisions are governed by other considerations, and not private investors.)
In the absence of reconciliation between desired saving and investment by interest rates, real economic activity adjusts to assure ex post equality between them. If desired saving persists in exceeding desired investment, adjustment over time will lead to secular stagnation, arising from persistently deficient aggregate demand.
Of course real modern economies are not closed. In an open economy a discrepancy between desired saving and desired investment can be resolved through an export surplus (in the case of excess saving) or deficit (in the case of excess investment). Of course not all countries can compensate for deficient domestic aggregate demand in this way: the world economy is closed, and surpluses by some countries must be matched (apart from measurement errors) by deficits in other countries.
At present most rich countries, and many emerging markets, experience current account surpluses in their international payments, often large relative to GDP. Germany's surplus, for example, is around eight percent of GDP, and that of the Netherlands, Switzerland, and Taiwan is even higher. Japan's surplus has dropped from its highs some years ago, but remains nearly three percent of GDP, similar to that of China.
As would be expected, many poor developing countries run current account deficits, financed in part by foreign assistance, as well as by foreign bank lending. But the largest current account deficit by far is that of the United States, over $400 billion in 2015, although that is under three percent of GDP. That number is roughly doubled by adding the deficits of Britain, Canada, Australia, Brazil, and Turkey, in that order.
How are the persistent U.S. current account deficits financed? Contrary to general impression, they are not financed mainly by foreign central bank acquisition of dollar assets—which in fact actually declined in 2015. The United States generates assets, both debt and equity, which private investors around the world want to acquire—often in preference to investing at home. This of course includes U.S. Treasury securities, which are highly liquid and low risk; but it also includes equity in new start-ups with a promising future. One way of putting it is that the United States has a comparative advantage in producing new financial assets, which foreigners are happy to buy in exchange for their exports of employment-creating goods and services.
So is the world entering a period of secular stagnation? There is lots of evidence for it. Unemployment rates exceed ten percent in continental Europe and growth is weak. Excess capacity exists in many industries. Inflation rates are low, well below two percent, in all of the advanced economies. Money wages are hardly increasing. Long-term interest rates are at historic lows, barely above one percent in the G-7 countries since 2008, and recently actually negative in Japan. There have also been positive output gaps in Europe and the United States for over eight years (IMF, 2015). World population growth has slowed, and become negative in several large countries, including Japan, Germany, and Russia. All of this suggests that stagnation may have arrived.
On the other hand, the 2008 financial crisis and subsequent recession represented huge negative shocks to the world economy. The American economy has largely recovered, but Europe had a second recession in 2012-13, from which it is recovering only slowly. We will have to wait several years to discover whether Europe recovers, or whether it enters a period of stagnation like Japan.
If we are entering a period of stagnation, what might be done about it? Three possibilities come to mind. First, income could be redistributed domestically from the wealthy to poorer people, who have lower propensities to save. This could be done directly, or indirectly through public expenditures on the poor (such as through pre-school education) financed by the rich.
Second, world infrastructure needs are high, mainly but not exclusively in developing countries. McKinsey Global Institute (2013) has estimated that the world need for infrastructure investment over 2014-2030 varies from $57 trillion-$67 trillion on various methods of calculation. On its criteria, only Japan and China have exceeded their normal needs for infrastructure investment; Brazil and India fall far short. But "needs" do not represent effective demand, which requires finance. This is where the World Bank, the regional development banks, and the new Asia Infrastructure Investment Bank can play an important role. Their bonds, ultimately backed by member governments, can help satisfy the demand by savers around the world for high quality assets, which the United States now disproportionately provides.
Third, the capital requirements for mitigating climate change would require further investment. With the right incentives—effectively a sufficient charge on emissions of carbon dioxide and other greenhouse gases—private capital will provide much of the required financing, supplemented with some official support for reducing emissions in developing countries.
Thus, appropriate globally-coordinated public policy can overcome a global imbalance between desired saving and desired investment. There is no better time to start now, when long-term interest rates are unusually low.
