Promises, Promises: The elusive search for faster economic growth

September 1, 1996

The U.S. economy has performed superbly in recent years by way of reaching and maintaining both low unemployment and low inflation. But since the early 1970s, long-term economic growth in America has been disappointingly slow. And projections of future growth, including those of the Congressional Budget Office and the administration, foresee no pickup in the years ahead. While most Americans who want to work have jobs, their incomes and living standards havenþt been growing much, especially when compared with the halcyon days of the 1950s and 1960s when GDP was expanding 4 percent a year and incomes per family almost 3 ½ percent.

Not surprisingly, policymakers are looking about for ways to spur faster growth. From all along the political spectrumþfrom the National Association of Manufacturers to the AFL-CIO, from former Republican presidential hopeful Steve Forbes and GOP vice presidential nominee Jack Kemp to MIT economist Lester Thurow and investment banker Felix Rohatynþ come proposals to alter national economic policies to produce another 1 or 2 percent extra annual growth. (Needless to say, there is no agreement as to exactly what policy alterations are needed.)

If these policies could indeed produce those results, it would be a tremendous boon to the nation. Increasing annual GDP growth from its current 2 percent rate to 3 percent would, if continued for 10 years, add more than $4,000 annually to the $40,000 income of the average American family. A return to 4 percent growth would add $8,000. Notice, however, that a 1 percentage point speedup in growth requires a 50 percent increase and a 2 percent speedup a doubling in the growth that is now forecast!

Faster growth could do more than boost incomes. One percentage point of extra growth would raise federal revenues and lower interest payments on the debt enough to balance the budget at the end of the 10-year periodþso long as the growth-promoting policies didnþt themselves worsen the budget situation.

What changes in current national policies promise to produce higher growth? How much more growth are they likely to produce? And how much might each help in balancing the budget?

One often-mentioned change involves monetary policy. Current tight Federal Reserve policy, it is said, is a prime obstacle to growth. Whenever economic growth threatens to exceed 2 percent or so on a sustained basis, the Fed has shown itself ready to boost interest rates to prevent the higher growth from being realized. Perhaps a more venturesome policy of credit easing and lower interest rates would produce higher output without setting off inflationary pressures, giving U.S. citizens not only higher incomes, but lower budget deficits as well.

And even if the CBO, the administration, and the Fed are right about current limits to growth, perhaps policies could be devised to improve the nationþs future growth potential and allow the Fed to relax its caution. One pro-growth policy is a tax cut. A hotly debated issue here is the extent to which the initial revenue cost of a tax cut is offset by the revenue gains from the faster growth it generates. Two other pro-growth policies are relaxing and improving the economic efficiency of environmental regulation and expanding public investment in education and training.

The Federal Reserve

In times of economic þslack,” when workers and industrial capacity stand idle, a Federal Reserve policy of easy credit and low interest rates can stimulate spendingþespecially on big-ticket items like housing, cars, and business investmentþthat puts idle workers and capital resources back to work and causes the economy to expand. But once the nation has reached full employment, more big increases in spending spurred by continued easy money policies will overheat the economy and drive up inflation. Beyond that point the nationþs spending and production can grow on a sustained basis only as fast as its productive capacity, its supply of goods and services, expands. Evaluating Federal Reserve policy, therefore, requires knowing first, how much slack there is in the economy and second, how fast the economy’s productive capacity is capable of growing. Full employment—that is, zero slack—does not mean zero unemployment. Even in the best of times some unemployment exists. Millions of young people are looking for their first job, some firms and industries are having to lay off workers who then spend some time searching for a new job, and so on. Stimulating demand for goods, and the related demand for labor to produce those goods, to the point where unemployment is pushed below the full employment level will begin raising the rate of inflation in wages and prices.

Although the rate of unemployment consistent with full employment and stable inflation has varied a bit over time, most economists believe that today full employment is between 5 and 6 percent, and is probably not far from the 5.5 percent unemployment rate of midsummer. But it is at least conceivable that full employment today could be a little below its current level. That being the case, perhaps the Fed should carefully lower interest rates and probe the possibility that unemployment could be lowered a bit further without reigniting inflation. But even if unemployment could safely be reduced by, say, one-half percent, that would produce only a one-time rise in GDP of something like 1 percent. Thereafter GDP would resume its projected 2 percent growth rate. While any noninflationary gain would be welcome, this one-shot increase is a far cry from adding a sustained 1 percent to the annual growth rate. On the budgetary front it would lower the deficit in 2002 from $285 billion to something like $250 billion.

How Fast is U.S. Economic Potential Growing?

If the economy does not have enough slack to accommodate a large immediate increase in GDP, is it possible that the economyþs capacityþits potential GDPþcan grow faster than the 2 percent per year rate envisioned in the standard projections? Unfortunately, the evidence seems overwhelming that under current conditions and policies, the CBO projection of capacity growth is unlikely to be far off the mark. The 2 percent projected growth rate consists of a roughly 1 percent annual expansion in labor input and a little over 1 percent a year growth in productivity, or output per worker hour. The strong and steady expansion in labor force participation by women after World War II slowed after 1989, and there is no warrant to assume the earlier growth rate will resume. The labor input projection is likely to be reliable over the next 10-15 years.

Productivity is a little more open to question, but not much. From warþs end until about 1973, productivity grew almost 3 percent a year. Thereafter, despite the information revolution, productivity growth slowed sharply to little more than 1 percent. Economists can explain only part of the declineþthe reasons for much of it remain a mystery. But whatever the reasons, the productivity trend has remained remarkably steady for more than 20 years. Not knowing what caused the decline, we can’t be sure productivity growth wonþt one day pick up again. But it would be foolhardy to plan national policy on that hope.

Policies to Increase Capacity Growth

If the prospective growth of the nationþs economic capacity puts a lid on what the Federal Reserve can do to raise GDP, could changes in national policy boost capacity growth? Such changes are possible, and I will examine several in a moment. But the cold facts of economic life tell us that no politically feasible policy is going to raise the current rate of economic growth by more than several tenths of a percentage point a year. We should not dismiss the benefits of what look like small increases in economic growth, especially if they persist for several decades. For example, a 20 percent increase in labor productivity growth would lift the overall economic growth rate from 2.1 percent a year to 2.3 percent. Continued for a generation, that increase would raise average family income more than $2,000 a year in todayþs dollars. But the increase in GDP and national income will not generate enough revenue gains to offset more than a modest fraction of the budgetary cost of these policies. To avoid widening the budget deficit, and thereby slowing growth, tax cuts or new public investments will have to be matched by cuts in federal spending over and above the massive cuts already needed to wipe out the budget deficits now in prospect. Whether the nation is willing to pay the costs and whether the gains are worth paying them are what we should be debating.

Tax Cuts

Federal income tax cuts are an ever-popular way to increase economic growth. Few if any serious supporters of tax cuts today would claim that the economic gain will be so large that the tax cuts will largely pay for themselves. But exactly how much additional GDP would be induced by a supply-side tax cut, and how much of the initial revenue loss would be offset by the revenue yield from that added GDP?

Broad tax cuts can increase growth in two ways: by expanding the labor supply and by stimulating private saving and investment. The critical issue in each case is how much? Economists measure the responsiveness of labor supply to a tax cutþthat is, to higher after-tax wagesþby laborþs supply elasticityþthe percentage change in labor supply induced by a 1 percent change in after-tax wages. Needless to say, not all studies produce the same answers. It is generally agreed that the elasticity is quite low for men, a good bit higher for married women, and somewhere in between for single women. The elasticity for the labor force as a whole probably lies within a range of 0.2 to 0.33. A 15 percent across-the-board cut in federal income tax rates would raise after-tax wages by some 5 percent. An elasticity of 0.2 would mean a 1 percent increase in labor supply; an elasticity of 0.33, a 1.7 percent increase. Because labor costs constitute a little under 70 percent of GDP, the resultant gain in GDP would lie between 0.7 percent and 1.1 percent. But that would be a one-time gain; thereafter, GDP will grow at the same rate as before but at a higher level.

What about the effects of a tax cut on private saving? Many economists find little effectþþsmall and hard to find” is how one textbook puts it. One of the higher estimates of the responsiveness of saving to changes in the after-tax return was made many years ago by Michael Boskin (later, chairman of the Council of Economic Advisers during the Bush administration). He estimated that historically a 1 percentage point change in the return to saving in the United States tended to lower consumption spending out of a given income by a little over 2 percent. Using a high response based on the Boskin estimate and a lower one half that size to accommodate the range of views, I estimate that a 15 percent cut in federal income tax rates would raise the private saving rate from its current 6.0 percent to somewhere between 6.15 and 6.3 percent.

The standard economic model of the growth process suggests that a 1 percentage point increase in the share of GDP flowing to saving and then into domestic investment would add about 0.1 percent to the annual growth of productivity and GDP. (If some of the extra saving went into investment overseas, the resulting inflow of earnings from abroad would raise the growth of the nationþs income about the same.) In recent years a þnew growth theory,” still much in dispute, has suggested that investment yields a higher return than implied by the standard model. That possibility could yield an increased payoff, with each 1 percentage point increase in the new investment share adding 0.2 percent to the GDP growth rate. The combination of labor supply and saving increases from a 15 percent tax cut enacted this year would, on the conservative assumptions, raise GDP by 2002 some 0.8 percent above the level it would otherwise reach; on the optimistic assumptions the gain would be 1.4 percent. That gain would expand to a range of 0.8 to 1.6 percent 10 years from now. By 2002 the revenue loss would be $150 billion a year. The resulting gain in GDP would bring in added taxes of $20-30 billion. Thus the net cost to the budget of the tax cut would be $120-130 billion in 2002. If these costs were not matched by spending cuts (over and above those already needed to balance the budget), the deficit would rise, an increased share of private saving would be diverted to financing that deficit, and private investment would fall. In that case, the tax cut would slow economic growth.

From the standpoint of economic growth, the $125 billion in spending cuts needed to finance the net cost of a 15 percent tax reduction would be better used to reduce the budget deficit (or if the budget were already balanced, to produce a budget surplus). The spending cuts would reduce the deficit directly and gradually produce further large and growing budgetary savings by lowering interest rates and slowing the rise in federal debt. Initially the tax cuts would produce greater GDP gains. But six years after the spending cuts were fully in place, the gain in GDP from deficit reduction would equal the gain from the tax cut, and thereafter exceed it by a growing amount. After 15 years deficit reduction would add over 3.5 percent to GDP compared with a gain of 1.4 percent with the tax cut. The package of tax reductions proposed by presidential candidate Robert Dole last July included a 15 percent cut in personal income taxes only, whereas the 15 percent across-the-board tax cut analyzed here applies to both personal and corporate taxes. But the Dole proposal included a 50 percent reduction in the tax on capital gains. There is no reason to believe that the effect on economic growth from the Dole plan would be significantly outside the range estimated above (given the important caveat that the net costs of the tax cuts were fully offset by spending reductions).

Investment In Education and Training

Improving the quality of the workforce through investment in education and training has long been recognized as a spur to economic growth.

The combination of slow productivity growth and declining real wages for lower-skilled men over the past 20 years has called forth many proposals to upgrade education, especially for those who begin work right after high school. Could big public investments in education and training produce what tax cuts cannot, a substantial gain in the level or growth rate of the economy? Again, unhappily, the answer is no.

Well-known labor economist James Heckman has concluded that a good starting point for estimating the payoff to additional public investment in education and training would be to assume that it yields a rate of return of 10 percentþabout the same as the return to investment in business capital. The federal government now spends almost $30 billion a year on education and training (excluding student loans). If it launched a massive increase in that spending, amounting to 1 percent of GDP, its education budget by 2002 would grow to $130 billion, and extra spending cuts of $100 billion would be needed to keep the new spending from worsening the deficit. Given the assumption of a 10 percent rate of return, the extra spending would raise the long-term growth of GDP by 0.1 percent a year, about the same as growth-oriented tax cuts of the same magnitude.

Regulatory Rollback and Reform

Arguing that excessive and inefficient regulation depresses economic growth, the Republican Congress set about in 1995 to redesign and scale back the regulatory apparatus of the federal government, particularly environmental protection. How much added growth can realistically be expected from such a rollback?

Dale Jorgenson and Peter Wilcoxen have estimated that removing all environmental controls would eventually raise measured GDP by 3.2 percent, as labor, capital, and raw materials devoted to cleanup were shifted to producing the goods and services counted in GDP. Adjusting that estimate to account for statistical peculiarities involving the net economic costs resulting from the mandated installation of emission control devices gives a figure closer to 2.9 percent.

Of course no one proposes scrapping all environmental controls. But costs could be reduced in two ways. First, some environmental laws and regulations have economic costs that well exceed the value of their environmental benefits. Those laws and regulations could be scaled back. Second, relying less on detailed rules and more on economic incentives to control pollutionþfor example, charging effluent fees or auctioning off effluent and discharge permitsþcould significantly lower the cost of achieving the environmental goals we do set.

Realistically we should not expect to achieve all the theoretically available saving. It would be impossible to fine-tune the environmental cleanup so as to take all actions whose benefits exceed costs and none whose costs exceed benefits. Besides, voters have widely differing views as to the value of the benefits. An ambitious target might be to pare the economic costs of environmental regulation by 25 percent. That, according to the adjusted Jorgenson and Wilcoxen estimates, would yield a gain of 0.7 percent in the level of GDP. In turn, realizing most of the gains by the end of 10 years would temporarily raise GDP growth a little less than 0.1 percent a year. After that GDP would grow at its earlier rate but along a higher path.

Promising the Moon

Certainly policies exist that will increase the level or the rate of growth of the nation’s economy—but only modestly. Even small increases in economic growth, if they persist for several decades, can boost noticeably the living standards of the next generation. But nothing that is realistically possible, not tax cuts, nor public investments, nor regulatory reform, will generate the large gains that many politicians are seeking—and promising. Moreover, neither tax cuts nor public investments will generate enough economic gain to pay for more than a small fraction of their budgetary costs. To avoid raising the deficit, and thereby lowering growth, they would have to be accompanied by spending cuts over and above the massive ones already required to erase the budget deficit. After witnessing the difficulties and shortfalls over the past two years in nailing down budget-balancing spending cuts, a reasonable person could well question whether now is the time to launch new policies that would, at best, provide small increments to growth, while running the risk of actually lowering growth by worsening the budget deficit.