This essay is available here in Adobe Acrobat PDF format.

This month’s Economic Outlook is dedicated to the memory of David F. Bradford, who wrote the blueprint for the consumption-based tax.

Alarmists who call for American households to save more point to a steady drop in the conventionally measured U.S. saving rate to about 1 percent at the end of last year and to a rise in household debt to a level well over 100 percent of personal disposable income. The current account deficit, our external deficit, measures national dis-saving at close to 6 percent of GDP. The federal government’s budget deficit contributes about 4 percentage points to national dis-saving and it, too, is the subject of considerable hand-wringing by those who point to a need for higher U.S. saving at both the household and national levels.
 
Government dis-saving, better known as the budget deficit, has received adequate attention elsewhere. Suffice it to say that the federal budget deficit has probably peaked at around 4 percent of GDP with no apparent damage having resulted in the form of higher interest rates, higher inflation, or slower growth--notwithstanding claims to the contrary emanating from the many critics of deficit spending. This essay will focus on household saving, the form it is taking, and whether households actually should increase their saving.
 
Saving is a good thing, but it is possible to overdo it. The uncritical acceptance of the notion that more saving is always better than less saving is a bad guide to individual behavior and a bad guide to public policy. Anyone who thinks that a nation whose people consistently work hard, save, and invest will be consistently better off than a nation whose people may work hard but save less need only compare the economies of Japan and the United States since 1990. Over the past fifteen years, America’s real net worth has risen by nearly 80 percent (about 4 percent per year), while Japan’s wealth has actually dropped despite its much higher saving rate.
 
The form that saving takes is also important. Saving involves forgoing current consumption in return for the security of having accumulated assets or for the earnings on investment that is financed by saving. The very low level of American saving, measured as the difference between income and consumption, suggests that a rise in the value of housing is being viewed as saving by many U.S. households. That may not be the best way to save, however. There are some ways to raise saving while correcting the U.S. bias toward housing-as-saving that the President’s Tax Reform Commission may want to consider. A consumption-based tax that taxes all saving only once, rather than twice as the current income tax system does, makes a lot of sense.

What Is Saving?

Saving is the difference between consumption per unit time and the sum of income plus the change in accumulated wealth over the same unit of time. When a household or individual begins a working life with no accumulated wealth, then saving is simply non-consumption, the portion of income that is set aside to accumulate wealth, perhaps for a rainy day (when no income is available) such as sickness, or for retirement, or to provide for one’s heirs. The bequest motive is strong, especially for those who began their working lives with no wealth, because appreciating or income-producing assets make it possible for one’s offspring to consume current income while still accumulating wealth. The frequent gifts by parents to children of a home or a down payment on a home come to mind.

Alfred Marshall reminds us in his Principles of Economics that the habit of saving for the future is not a constant. Where no secure means to store wealth exists, or where income is at a subsistence level, there is likely to be virtually no saving. Marshall notes that the earliest forms of accumulated saving or wealth were probably hunting or fishing implements along with clothing or huts in the colder climates. These items all represented stored-up sources of future use or enjoyment much as durable goods, including auto-mobiles and appliances, do today. More elaborate implements, huts, and clothing, together with perhaps growing herds of domestic animals, provided storage for future enjoyment and needs at the expense of maximum consumption as these items were being accumulated.
 
A necessary condition for saving is the realization by the prospective saver that provision of future wants can be stored. The more wealth one accumulates, the less rigid is the link between a variable and often uncertain stream of future income and future consumption. As wealth accumulates, provided the means exist to store it, the more it is possible to accumulate future wealth or to enjoy a steady stream of future consumption independently of a future stream of income.

Saving Begets Saving

It is difficult to escape the notion that wealth accumulation is self-reinforcing, at least up to a point. As more wealth accumulates, the means to store it safely, which in the fullest sense entails a modern nation-state whose primary responsibility is self-defense, is more likely to appear and grow in strength. More wealth, in turn, enhances the accumulation of productive capital that enhances labor productivity and provides its owners with an attractive rate of return. The payment of interest on accumulated capital further smoothes and enlarges the prospective stream of consumption that is possible for those who store wealth.
 
On all these points, nineteenth century Britain provides a helpful example. The combination of a powerful empire and the Industrial Revolution created a stock of wealth owned by the propertied classes that transformed Britain. The consol, a long-term liability of the British government paying an average of about 3 percent, displaced land as Britain’s primary asset. The owners of consols, Britain’s prosperous merchant-class, watched carefully the returns on their favorite asset, which rose upon the prospect of war, thereby depressing the value of existing consols, then fell when peace returned, thereby enhancing the value of consols acquired during the conflict. Of course it was essential that the war not be lost, at least not disastrously so, and the British were consistently successful on that score for over a century, until after the First World War, a disastrous conflict that eradicated many institutions of the nineteenth century, including much of the accumulated saving of the middle class.
 
Of course mature wealthy nations are often tempted by the prospect of even higher prospective returns from investment in newer, more vigorous economies. During the nineteenth century, British investors were tempted to invest heavily in the United States in ventures including canals and railroads. While some British investments in the United States did well, the results were not uniformly positive. A wealthy nation, or at least a nation’s wealthiest households, can save too much, at least judged after the fact, if they are tempted by high prospective rates of return that divert savings from lower, less risky ventures.
 
Too Much Saving in One Nation

The problems faced by modern China’s growing class of the newly wealthy provide a reminder that limits do exist to the self-reinforcing aspect of wealth accumulation. Very rapid income growth in China has, given the desire of many Chinese to build wealth rapidly, coupled with the rapid inflow of foreign capital, boosted investment to a level reportedly above 40 percent of GDP. Such investment is probably above the level that can be absorbed profitably inside China. The result has been a helter-skelter rush to store wealth through questionable investments in unoccupied apartment complexes and often-empty high-rise office buildings, just to mention a few, instead of in the state-owned banking system, the insolvency of which is well known. The flow of too much savings into China can produce overinvestment that drives the return on capital to zero or, after the fact, below zero. Savers beware.
 
Japan’s negative experience after its high-growth period in the 1970s and 1980s reminds us that a high level of national saving can turn out to be oversaving. Japan’s remarkable post–World War II recovery, a testimony to the national ethic of working, saving, and investing, was highly successful in the 1960s and 1970s as capital accumulated and earned an attractive rate of return. As the process continued, and Japan’s prodigious national savings drove the accumulation of more and more capital inside Japan, the return on that capital began to fall. By the end of the 1980s, Japan’s high level of national saving spilled over into purchases of property until the Japan bulls proudly declared that the plot on which the emperor’s palace sat in the middle of Tokyo was “worth more than California.” Japanese savers, when on rare occasions they were able to extract money to invest outside of Japan, were so eager to diversify their holdings that they became legendary as overpayers for prime sites such as the famous Pebble Beach Golf Links in California. But broadly, Japan’s trapped savers drove up the price of income streams from wealth, specifically the price of stocks and land, until the stock and property bubbles burst after the 1980s. Japan’s property market has still not recovered, nor has its stock market. During the 1990s, Japan tried to engineer a recovery by having the public sector finance overinvestment in railways and public works projects, thereby driving the return to savers and investors even lower.

American Saving Experience

Turning to the experience of the United States as a saver and investor, some disquieting signs have arisen that there may be too much saving (not all of it American saving) chasing investment in U.S. assets. The rapid run-up in high technology stocks from 1996 to 2000 followed by the March 2000 bursting of the tech-stock bubble suggests, in retrospect, overinvestment in that sector. Overinvestment of course can only be fueled by what retrospectively comes to be known as oversaving. Those who lost money in the tech-stock bubble, after the fact, wished that they had either invested elsewhere or simply consumed the funds that flowed into the tech sector.

Of course for American savers there is an asset that provides a tax-sheltered way to enhance consumption while, simultaneously, storing wealth. U.S. residential real estate has surged in value by 44 percent, over 10 percent per year, since 2000. Even adjusting for inflation, the real gains in the value of U.S. residential real estate between 2000 and 2004 are estimated at 38 percent. For purposes of comparison, the real gains during the late 1970s real estate boom between 1976 and 1980 totaled 26 percent.
 
Thinking about residential real estate as a store of value for U.S. households suggests a number of insights about U.S. saving behavior. The tax preferences for U.S. residential real estate are well known. Interest on mortgages up to $1 million is fully deductible from income tax. Capital gains on sales of residences are exempt from tax in amounts up to $500,000. So, too, are state and local real estate taxes, although the alternative minimum tax may be starting to atrophy this benefit. Still, the largest tax benefit lies with the fact that the consumption services from owning real estate and living in it, either as a primary residence or a vacation home, constitute a non-taxed form of consumption. These extraordinary tax preferences for residential real estate amount to $1 trillion in tax revenue losses over the next five years. Those funds could be used to finance a move toward a far more efficient consumption-based tax system wherein all forms of saving are treated the same--and more favorably--than under the current system.

One of the reasons that the conventionally measured U.S. saving rate is so low, having dropped virtually to 1 percent from a still-low 4 percent in the late 1990s, lies with the fact that the consumption of housing services is subsidized, while increases in the value of owner-occupied real estate are not counted as conventional saving. The measure of household net worth as a multiple of personal disposable income, which includes the value of real estate holdings, has risen to 5.4, not far below 6, the level seen at the peak of the stock market bubble. For purposes of comparison, that ratio averaged about 4.5 during most of the 1980s. Once again, after the bursting of the stock market bubble, Americans are saving by overinvesting in tax-preferred housing as a store of value. Unfortunately, housing does not add as much to labor productivity as more traditional capital does, and so the rush of American saving into housing raises questions about the durability of America’s decade-old rise in productivity growth from about 1 percent to about 2.5 percent annually.

U.S. households are also heavy purchasers of durable goods. Durables as a source of a stream of services over time have both a consumption component and an investment component, just as hunting and fishing implements did in Marshall’s primitive-saving societies. The fact that an automobile or a washing machine lasts considerably longer than a year is not captured in the official U.S. savings statistics. The entire purchase price of a durable good is counted as consumption and subtracted from income when estimating U.S. saving. At a time when U.S. income is growing rapidly, and interest rates are low, it is not surprising to see accelerated spending on housing and durables that, in turn, drives down the measured saving rate.
 
Too Much Saving?

The discussion of the experience with saving and investment in advanced industrial countries suggests a crude criterion to use when answering the question of how much saving is too much. The rough answer is that the appropriate level of savings is related to the return that can be earned on investments financed with those savings. For example, one of the reasons that the U.S. runs a large current account deficit financed by a heavy net flow of foreign investment into the United States lies with the perception that the real return on investment in U.S. assets is higher than the real return on investment elsewhere. The Chinese government, which does not allow large-scale capital outflows by its citizens, is investing about $200 billion a year abroad, much of it in U.S. Treasury securities on behalf of its population, which is generating more savings than can be profitably absorbed inside China.
 
The result is, of course, to accommodate U.S. consumption, especially of owner-occupied real estate given the ability of U.S. households to increase their leveraging of purchases of real estate through low-interest, tax-advantaged mortgages. In effect, the Chinese government is helping to subsidize more U.S. saving in the form of an increase in the value of the stock of real estate in the United States, which in turn builds an ever-larger stream of untaxed benefits to U.S. households while simultaneously enabling them to store and enhance wealth.
 
The problem lies with the fact that real estate bubbles are only supported by the prospects that the next buyer of a property will pay even more than the last since real estate does not actually produce anything other than residential services for its owner. The value of those services shows up in the rental value of a property and, ominously, rental returns on properties in the stronger real estate markets in the United States are extraordinarily low, as is often the case in the late stages of a real estate bubble.

The suggestion that too much saving worldwide is chasing investments with returns that are too low is permeating world markets. It is not surprising to find in a world of aging societies that are growing wealthier, that the search for ways to store wealth and thereby generate future streams of income for retirement is intensifying. There is a global market for income streams from wealth and the price of those streams, the inverse of the rate of return on investments, has been driven higher and higher, especially in areas such as real estate, where tax preferences reinforce the attractiveness as a store of value for U.S. households.

Rising House Prices Not the Best Way to Save

The definition of savings as the difference between consumption per unit time and the sum of income plus the change in accumulated wealth over the same unit of time reminds us that the simple, conventional notion of saving, not spending out of current income, is inadequate and misleading. If the substantial rise in house prices over the past four years is counted as saving, Americans are saving enough, although they are accumulating wealth in a highly illiquid and potentially risky way. Who wants to sell their home to cushion consumption if income falls? It is not like taking money out of the bank.

In a society that is aging and growing wealthier, the price of income streams from wealth can rise rapidly. In those circumstances it is not surprising to see asset managers ignoring risks in order to earn a higher rate of return. Eventually, this search for return irrespective of risk results in too much asset acquisition, and when reality dawns, a sharp drop follows in the value of some assets. Such could easily be the fate of America’s much-favored real estate. A gradual transition to a consumption-based tax system that favors all forms of saving, not just housing, would accomplish the dual goal of encouraging more U.S. saving in all forms while reducing the risk of disruptive real estate bubbles.

It is important to remember that, over the long run, the real return to saving has averaged around 3 percent on low-risk assets after inflation. Those planning on consistently earning a higher rate of return may be disappointed. When asset values fall, saving is more urgently desired at the very time achieving it requires a cut in consumption that exceeds realized saving defined to include changes in asset values. More specifically, a drop in U.S. house prices would sharply reduce U.S. consumption, the world’s major source of demand growth. Perhaps that is why many analysts are uncomfortable about seeing higher real estate values as the major saving vehicle of American households while increasing claims on U.S. wealth, in the form of the current account deficit, is a major saving vehicle of Asian and European households.

Note

1. Alfred Marshall, Principles of Economics, 9th ed., (London: Macmillan, 1961), 220–39.