In response to:

The Clinton Budget: Will It Do? from the July 15, 1993 issue

To the Editors:

Benjamin Friedman [NYR, July 15, 1993] apparently endorses a “widespread concern…that if the President’s program ends up delivering little if any deficit reduction, but instead only a big increase in domestic spending financed by a combination of new taxes and defense cuts…the consequences will be damaging both economically and politically.”

The political damage is unlikely to be very great if the effect of increased spending were to bring about genuine recovery. But Friedman apparently believes that a failure to reduce the deficit will ensure economic disaster.

“When the economy is at or near full employment, as it was during much of the 1980s, the principal factor limiting its investment is the amount of savings available to finance it.” This happens because the borrowing the deficit forces “…the Treasury to do absorbs private saving, saving that the financial markets would otherwise make available to businesses undertaking investment in new factories…”

Professor Friedman does not tell us exactly how the Treasury takes funds away from private investment. If an outstanding investment project emerged today, calling for “new factories” that everyone agreed would be exceptionally profitable, would it be impossible to fund such construction domestically unless the deficit were reduced? Even though there is unemployed labor and underutilized capacity? That amounts to saying that Wall Street is not smart enough to figure out how to get in on a good thing.

It is often argued that a deficit-induced crunch in financial market bids up interest rates and so “crowds out” the private sector. However, there is no evidence that large or rising deficits are associated either with high or rising interest rates, real or nominal. (The interested reader should check the appropriate tables in the Statistical Abstract.) Nominal interest rates are largely determined by the Fed’s policies; real rates depend in addition on inflation, which, pace Monetarism, depends on cost changes, but not on the deficit.

Moreover it is a misuse of words to contend that the US economy was operating at or near full employment “during much” of the 1980s. The official unemployment rate never fell below 5.2 percent, and averaged 6.8 percent in the BOOM years 1983–1989, (compared to an average of 4.1 percent in the Sixties.) To obtain a more nearly correct measure of the actual slack in labor markets, it is necessary to add a correction for “discouraged workers” and another for part-time work. In the 1980s these corrections added several percentage points to the official rate. There was plenty of labor available for expansion.

Perhaps even more important, capacity utilization was at or below 80 percent until 1987, when it rose to 81 percent, after which it crept up to only 84 percent for two years before falling again. There was plenty of capacity available.

The problems that truly concern Friedman, and which appeared to be Clinton’s chief concerns during and after the campaign, are slow productivity growth, stagnating real wages, rising unemployment, and a slowdown in investment. These are serious problems, but they are not caused by the deficit: it is rather that the changes in the economy which have brought us these problems have also left us with a deficit. To attack the deficit is therefore to focus on a symptom rather than the disease.

Worse, this misguided focus on the deficit directs our political discourse away from a central question. The US has the lowest ratio of total government spending (including transfers) to GNP of any major industrial nation. A recent study in the Journal of Economic Literature (Barr, 1992) showed that between 1960 and 1980 total government spending as a percentage of GDP rose in all countries surveyed. But it rose from 31.4 to 46.6 in Germany, from 31.3 to 61.1 in Sweden, from 30.0 to 56.9 in the Netherlands, but from 28.1 to only 35.4 in the US.

Moreover, included in the US measure is the highest ratio of defense spending to GNP of any advanced nation. This means that excluding defense, we have by far the smallest public sector plus transfer system, in relation to GNP, in the developed world. Moreover our public sector has grown more slowly than that of any other developed nation. Could it be that many of our problems arise because our government is spending too little, in relation to GNP, rather than too much? Can we really manage the problems of a modern economy with less government participation than other nations require? Our social and economic statistics do not paint an encouraging picture.

These issues call for analysis and interpretation. Yet in public political discussion it is all too commonly assumed that government spending must be reduced further. However, recent studies have provided evidence that public investment in education, R&D and infrastructure will add more to future GNP than private spending, and further is likely to stimulate private productive investment. Far from trying to curb or cut government spending, the Clinton Administration needs to do just what Professor Friedman apparently fears—“…deliver…little if any deficit reduction, but instead [provide us with] a big increase in domestic spending, financed by a combination of new taxes and defense cuts…”


Edward Nell

Malcolm B. Smith Professor of Economics

The New School for Social Research

New York City

Benjamin Friedman replies:

Professor Nell’s letter raises an important issue that is central to the ongoing discussion of US tax and spending policies, and I welcome the opportunity to address it more fully.

The consequences of government deficits and borrowing—consequences for output, for investment, for interest rates, and so on—differ importantly according to whether or not an economy’s resources are fully employed. Professor Nell quotes the beginning of my brief discussion of how these effects operate “when the economy is at or near full employment” (my words), but he then goes on to question whether these same mechanisms would still be at work “even though there is unemployed labor and underutilized capacity” (his words). In doing so he misses the central matter at issue. If someone indicates how traffic flows when the light is green, it makes no sense to challenge his statement by asking if cars move in the same way when the light is red.

When an economy is at or near full employment, the supply of available saving is the effective limit restraining how much investment can be done over sustained periods of time. Nobody can invest more unless somebody else invests less, or somebody saves more. Anything that absorbs saving without using it for investment—for example, government borrowing—reduces the economy’s investment in total.

By contrast, when an economy has ample unemployed resources, the problem in this regard is not too little saving but too much. In those circumstances anything that increases overall economic activity will also raise people’s incomes, and probably also increase the total amount of saving and investment. Hence additional government spending, or tax cuts, need not diminish the amount of investment in an underemployed economy, and they may even increase investment. For example, I suggested in Day of Reckoning that President Reagan’s tax cuts and greater military spending not only did not crowd out US private investment during the high-unemployment years of 1981–1984 but probably crowded in some investment by increasing overall economic activity and hence people’s incomes and their demand for many products.

The question at issue today—and it is an important one—is whether in designing appropriate tax and spending policies for the United States for the balance of the 1990s we should plan for a more or less fully employed economy, or one that has ample unemployed resources. Professor Nell claims that the US economy was significantly underemployed throughout the 1983–1989 business expansion, and remains so today. His argument against President Clinton’s deficit reduction program has force on the assumption that the economy will continue to be underemployed right through the period over which these deficit reduction measures will take effect.

I disagree with this assessment of the 1980s expansion and the present situation, as well as with the corresponding assumption about the future. The main reason US price and wage inflation began to pick up again in the late 1980s is that the rates of utilization of the economy’s labor and capital resources were then at, and probably above, what constitutes “full employment” nowadays. This does not mean, of course, that every American who wanted a job had one, or that every existing factory was running three shifts. At any time some people are newly entering the labor force, others are between jobs, and others simply lack the skills to find work at all in today’s increasingly high-technology economy. Some factories are outmoded, and others are inefficient. The fundamental economic changes—in demographics, in technology, and in labor market institutions—that explain why “full employment” meant a 4 percent unemployment rate thirty years ago but means a higher rate today are, by now, well understood. To ignore them is a sure recipe for poor policy prescriptions. In the same vein, the Federal Reserve Board recently revised its utilization of industrial capacity statistics to indicate that the US economy today has less unutilized plant capacity than earlier supposed.

Today unemployment the United States is still above 6 percent of the labor force, which is probably as good a “full employment” benchmark as any. But it is not much above 6 percent, and it is certainly well below what this economy has typically experienced during out-and-out business recessions. What matters for policy, however, is to look forward, not back: What is the best way to view the economic background against which the President’s program will limit government spending and raise taxes? Should we picture an economy that will have substantial amounts of usable but unused resources through 1998, and hence push for higher spending and reduced taxes to provide needed stimulus? Or should we suppose that the economy will be sufficiently close to full employment that the supply of saving will effectively constrain total investment, and therefore aim to limit spending and raise taxes so as to free up scarce saving and thereby boost investment?


I supported President Clinton’s deficit reduction package (indeed, I argued in these pages for a more aggressive program than the President’s) not merely because I believe increased investment is essential to achieving productivity growth and with it a rising standard of living but also, and importantly, because I share the President’s assessment that on average the US economy will be near to full employment during the period in question. The President’s budget projected that unemployment will decline from 7.1 percent of the labor force on average this year to 5.7 percent in 1998. The corresponding forecast from the Congressional Budget Office is identical. This seems to me a plausible working assumption.

It is especially so in light of how gradually Mr. Clinton’s program will limit spending and raise taxes over this period. The spending cuts and tax increases legislated for the 1994 fiscal year, compared to what would have happened under previous policies, add up to approximately one half of one percent of US national income. The next year the combined total is just three fourths of one percent. Not until 1996 will the President’s program reduce the deficit by more than one percent of national income compared to the previous trajectory. With a lead time that long, the Federal Reserve System can readily adjust monetary policy if doing so appears necessary to resist any general downturns in business activity due to fiscal retrenchment.

In the end, I was therefore pleased that Congress passed the President’s program. No one likes paying higher taxes, of course, or doing without the programs curtailed by cuts in government spending. But I believe that reducing the government’s borrowing, and so increasing our country’s investment, is the surest way we know to begin to correct the dual problem of declining average living standards and widening inequalities about which I have previously written here.

Finally, in closing this response to Professor Nell, I must firmly reject his attempt to buttress (somehow) his argument by claiming that I believe “that a failure to reduce the deficit will ensure economic disaster.” Such a statement is wholly unsupported by any of my writings on this subject.

I feel strongly about this matter because a major aim of my own work has been to undo the damage done to the credibility of valid arguments for a lower deficit by those who, a decade ago, chose exaggeration and hyperbole over measured debate. I have also suggested that a major reason for our inability to correct our now-chronic fiscal imbalance is precisely that the adverse consequences of sustained government deficits at full employment unfold gradually and subtly over time. Ironically, if chronic deficits did produce “economic disaster,” with highly visible, tangible consequences capable of arresting the public’s attention, perhaps we would have solved the problem long ago and by now be better off as a result.

This Issue

February 17, 1994