Martin Feldstein. Foreign Affairs. Volume 85, Issue 3. May/June 2006.
The U.S. Savings Rate and the Global Economy
The savings rate of American households has been declining for more than a decade and recently turned negative. This decrease has dramatically reduced total national savings despite a rise in corporate saving. In 2003 and 2004, the combined net savings of households, businesses, and government were only about one percent of gross national income—the lowest level in at least 50 years.
This sharp decline in saving has had important implications for the United States and for the global economy. It has reduced productivity-enhancing net business investment in the United States to less than four percent of GDP and made the United States increasingly dependent on capital from the rest of the world to finance that investment. At the same time, the decreased national savings rate—and the increase in consumer spending that it implies—has induced a rise in U.S. imports. Those imports have contributed to the growth of output and employment in many countries around the world.
The downward trend in U.S. household saving will likely soon be reversed. In the long term, a substantial rise in household saving will have a positive effect on the U.S. economy. But the initial effects will pose problems for the United States and its trading partners. If these effects are not managed well, the result could be declines in output and employment and a corresponding rise in U.S. protectionism.
To appreciate the significance of these developments, it is helpful to have a more precise definition of net household savings. Household saving is the difference between what households receive in after-tax income (including wages, salaries, fringe benefits, interest, and dividends) and what they spend on goods and services. Those savings can take the form of bank deposits, purchases of financial assets such as stocks and bonds, or investments in real assets such as homes and unincorporated businesses. Contributions to individual retirement accounts (IRAs) and 401(k) plans as well as employer contributions to defined-benefit pension plans are also counted as household savings. So too is money used to pay down a mortgage or other loan. Borrowing is counted against saving, unless the borrowed funds are used to purchase financial assets or are converted into other types of savings.
Individuals in their 40s and 50s tend to save money, whereas many in their 60s and beyond are “dissavers,” people who are spending the assets that they accumulated in their working years. A negative U.S. net household savings rate implies that the saving of the savers is less than the dissaving of the dissavers. Net household savings in the United States fell from an already low 1.7 percent in 2004 to -1 percent in the second half of 2005. This low and declining savings rate stands in sharp contrast to the rate until the mid-1990s: 7 percent or higher, which was enough to finance most investment in business infrastructure and equipment and in housing.
That household saving has declined should come as no surprise. The rising prices of stocks and homes have made many Americans much wealthier. Despite the fall of the stock market in 2000, average share prices have more than doubled in the past decade. Housing prices have risen by more than 12 percent in the past year alone, and the total value of the real estate owned by households has risen by more than 50 percent in less than five years, an increase of more than $6 trillion. Between 2000 and the start of 2005, the overall net worth of the household sector increased by nearly $7 trillion.
As a result, many individuals in their 40s and 50s looked at the value of their IRAs, 401(k) accounts, and homes and rightly concluded that they did not have to save as much for their retirement as they would have had to in the past. And so they started saving less and spending more. Similarly, retirees looked at their substantial stock-market and housing wealth and concluded that they could dissave much more than previous retirees. This combination of saving less and dissaving more produced the substantial decline in the net savings rate.
The increase in consumer spending as a result of increased wealth has been reinforced by the process of mortgage refinancing. As mortgage interest rates have come down from nearly eight percent in the mid-1990s to less than six percent, individuals have been able to increase the size of their mortgages without having to raise their monthly mortgage payments. Although some of the extra mortgage borrowing has been converted into other savings, much of it has been used to finance additional consumption. The amounts involved have been enormous. In the past five years, the value of U.S. home mortgage debt has increased by nearly $3 trillion. In 2004 alone, it increased by almost $1 trillion. Net mortgage borrowing not used for the purchase of new homes that year amounted to nearly $600 billion, or almost seven percent of disposable personal income.
Why is the decline of saving significant for the economy? Household savings matter because those funds have traditionally been used to finance business investment in equipment, software, and buildings, and this investment increases productivity, the rate of economic growth, and the future standard of living. Although the rise in household wealth that comes from higher stock and real estate prices feels like “saving” to individual households, it does not free up resources to increase investment in business capital. The low savings rate has thus forced the United States to become increasingly dependent on funds from the rest of the world to finance domestic business investment. Such capital inflows now finance more than three-quarters of U.S. net investment.
U.S. dependence on foreign capital has increased very rapidly in recent years. The rate of net capital inflows in 2005 was more than twice that of 2001 and five times as much as that of 1997. Although American investors send capital abroad to buy foreign stocks and bonds and invest in overseas businesses, the flow of foreign capital to the United States has increased much more rapidly than U.S. investment abroad. Net inflows have grown from 1.7 percent of GDP in 1997 to an unprecedented 5.7 percent of U.S. GDP in 2004 and more than 6.4 percent in 2005.
Foreign investors are able to get the dollars to finance these inflows of capital to the United States because Americans import more goods and services from the rest of the world than they export. When Americans buy something that is made abroad, they pay in dollars; when they sell something to foreign buyers, they earn some of those dollars back. In 2004, U.S. purchases from the rest of the world exceeded U.S. export sales by $618 billion. Americans also made various other net transfers to the rest of the world (including personal remittances, net interest and dividend payments, and government grants) of $51 billion. Combining the two yields the total current account deficit of $669 billion. That amount—5.7 percent of GDP—is the total value of net capital inflows that came to the United States from the rest of the world in 2004.
This basic accounting identity—that the current account deficit is the amount of capital inflows to the United States—links the U.S. trade imbalance to U.S. dependence on foreign capital. Accordingly, the United States can reduce its dependence on foreign capital inflows only if it reduces its current account deficit. Doing so will require a slowdown in the growth of consumer spending.
The relatively rapid increase in consumer spending in the past decade and a half, and the resulting decline in the savings rate, is the primary reason for the large trade deficit and the associated capital inflows. But capital inflows have been possible only because foreign countries are willing suppliers of those funds. Low U.S. savings relative to U.S. business investment is necessarily mirrored by high savings relative to business investment in other countries. Their savings end up as investment in the United States rather than in other countries because of the United States’ large consumer-driven trade deficit.
Other People’s Money
The flow of funds to the United States should be a matter of concern not only because it has increased so rapidly but also because those funds now come in what may prove to be a much less sustainable form. In the late 1990s, most of the capital that flowed into the United States came in the form of equity investment by private investors buying stock in U.S. companies or buying U.S. companies outright. These private investors diversified their investments by moving into the U.S. market because they thought the potential equity returns in the United States were favorable in comparison to the risks. Now, in contrast, the flow of equity investment to the United States is very small—often less than the equity investment that Americans make in the rest of the world. Today, the capital inflows to the United States primarily go toward purchasing bonds and making bank deposits, including very sizable deposits by foreign governments at the Federal Reserve.
It is not clear how much of the overall capital inflow is coming from foreign governments and how much from private investors. The available data do not distinguish between purchases by banks on behalf of private investors and purchases by banks on behalf of foreign governments. But extensive conversations with officials and private bankers suggest that an overwhelming share of the foreign capital inflows in recent years has come from foreign governments or from institutions acting on their behalf. Such a change, from private investment in U.S. equities to government purchases of U.S. debt, could make the continued flow of funds less reliable.
Initially, foreign governments invested in dollar assets in order to increase their foreign exchange reserves, especially after the Asian financial crises in the late 1990s. South Korea now has more than $200 billion of official foreign exchange reserves (invested primarily in dollar bonds), Taiwan has more than $250 billion, and China has more than $800 billion. As a result, these countries no longer have to fear a speculative attack on their currencies or the possibility of being forced to accept an unwanted International Monetary Fund program. These large dollar reserves are also the result of the polices of these and other emerging-market countries to keep their currencies undervalued as a way of keeping their products relatively inexpensive and thus promoting exports.
Since governments do not make investment decisions by simply balancing risk and return the way private investors do, it is particularly difficult to anticipate their future behavior. How long will foreign governments want to keep running large current account surpluses and investing those surpluses in relatively low-yield foreign bonds? And how will they respond to developments in the United States or in the global economy in the future? These unanswerable questions add to the overall uncertainty of the global economic outlook.
What can be safely anticipated is that the savings rate of American households will start to increase. It is not clear when that increase will begin, but household saving cannot continue indefinitely at a zero or negative level. Saving will begin to rise because the forces that have promoted rapid consumption growth and depressed saving over the past decade will not continue. Share prices are not likely to double again in the next decade, nor will housing prices continue to rise at double-digit rates. Accordingly, households will be able to increase their wealth for retirement and other purposes only by reducing the growth of their spending so that it is less than the growth of their after-tax incomes—that is, by saving more.
Most important, mortgage interest rates have stopped falling, bringing an end to the dissaving that occurred when homeowners simultaneously reduced their monthly mortgage payments and extracted massive amounts of cash simply by refinancing their mortgages. This suggests that there will be a relatively rapid rise in the savings rate.
This coming rise in household saving will eventually be good for the U.S. economy. It will make the United States less dependent on capital from the rest of the world and permit American businesses to raise the rate of investment in the equipment, software, and structures that increase productivity and the future standard of living. But a higher U.S. savings rate will also pose a challenge for the rest of the world, because it will mean a reduction in the exports to and an increase in the imports from the United States.
This shift will be true even for countries, such as those in western Europe, that do not currently have a trade surplus with the United States. It will reduce the overall aggregate demand and employment in countries that trade with the United States. To prepare for the inevitable rise in the U.S. savings rate, those countries should start planning ways to stimulate their output and employment. The fact that the United States is not the only country in which rising asset prices have caused a sharp fall in saving over the past decade only adds to the urgency of such planning—since the savings rates of those countries are also likely to rise, with similar consequences.
The coming rise in the U.S. savings rate could temporarily cause a serious problem for the U.S. economy as well. As consumer spending falls, GDP and employment will also decline. This decline will last until net exports increase by an equal amount. If markets function well, the higher savings rate will quickly put downward pressure on real interest rates, causing the dollar to become more competitive—until the rise in exports and the decline in imports return total GDP to its full employment level. However, there is the danger of a time lag between when the savings rate rises and when the U.S. trade deficit declines. Such a lag would cause a slowdown in the growth of output and employment, potentially leading to a recession.
This downturn would be exacerbated if foreign governments restricted imports from the United States or prevented the natural downward adjustment of the dollar exchange rate. The political response in the United States could easily be the imposition of protectionist policies, including tariffs and quotas on imports.
Foreign governments would thus do well to anticipate the coming rise in the U.S. savings rate. By permitting more exchange rate adjustment, foreign governments could mitigate the adverse cyclical effects in the United States of a higher U.S. savings rate. Because it takes time for trade flows to respond to exchange rate adjustments, it would be best to allow exchange rates to adjust before the U.S. savings rate rises. Equally important, the United States’ trading partners must find ways to increase their domestic demand in order to maintain their output and employment even as exports to the United States decline. A failure to achieve both types of adjustments could lead to a wave of protectionist policies that would be in nobody’s interest.