As unemployment persists at just above 6 percent and new job creation is sluggish or non-existent, polls show that more Americans are fearful of losing their jobs in the next year. Chiefly because of this, pollster John Zogby says, President George W. Bush‘s job-approval rating dropped in early September to 45 percent, the lowest since he took office.

Predictably, Democratic presidential candidates are using the occasion to step up their criticism of the president’s economic program. But does unemployment necessarily mean political problems for Bush’s re-election campaign?

Common wisdom would say “yes,” but a historical perspective is useful here. The 6.1-percent unemployment rate is above normal, but hardly unusual in modern times. Over the 103 years since reliable data have been computed, the median or typical unemployment rate is 5.5 percent. In the 33 years since 1970, the rate is 5.9 percent. Thus we are a bit higher than the normal rate, but not dramatically so.

On the other hand, unemployment is more than 2 percentage points higher than in the year in which Bush was elected and is the same as it was the year in which his father lost the presidency to Bill Clinton. Isn’t this bleak news for the president?

That is a possibility, but unlikely. For one thing, the fundamentals favor an upsurge in job creation. In the past few quarters, productivity growth—the amount of goods and services produced for each hour worked—has been strong, and real wage growth has been moderate, meaning the labor cost of making a dollar’s worth of widgets (or whatever) has fallen. This should increase the attractiveness of hiring more workers and unemployment rates should start heading south, certainly by the second quarter of next year, well before the November election.

Political history suggests that if this happens, Bush should not lose many votes over unemployment. In August 1983, at the same place in the electoral cycle, unemployment was 9.5 percent. Yet President Ronald Reagan won re-election handily the following year. Unemployment in the month before the election was 7.3 percent, still quite high but sharply down from its recession peak—and similar to many years in the previous decade.

The unemployment rate in October 2004 is likely to be well below 6 percent, down noticeably from its peak of 6.4 percent and close to a normal historic rate for the economy. It could even be below normal, judging from the recent fall in labor costs per dollar of output and the observed negative historic relationship between labor costs and unemployment.

Democratic strategists might point to three historic counterexamples: Gerald Ford, Jimmy Carter and George H.W. Bush—all losing incumbents who were hurt by rising unemployment, among other factors. In Ford’s case, unemployment had worsened sharply over his short presidency, and although it improved modestly before the election, it was well over 7 percent. Rising unemployment in an election year, to a high of 7.6 percent in the month before the election, certainly helped do in Jimmy Carter—along with double-digit inflation. The first President Bush had very slow growth and moderately high unemployment contribute to his unseating.

In none of these cases, however, was the incumbent a tax-cutter who was defeated during a period of normal and declining unemployment, the most probable scenario unfolding for President Bush next year. The popularity of the Bush tax cuts is likely to counterbalance any lingering dissatisfaction with unemployment.

Having said this, there is one major cloud on the horizon both for Bush and the U.S. economy. There is ample evidence that government resources are less efficiently allocated than those in the private sector. The private sector faces market discipline, competition and the bottom line of profits, which governments do not.

The boom in output and stock prices in the 1990s was partly fueled by rising productivity, stimulated by a reduction in federal government spending as a percentage of output. This reduced spending reflected falling defense spending with the end of the Cold War, the end of the savings and loan bailout and some congressionally mandated bipartisan fiscal discipline in Washington in the face of rising budget deficits.

This percentage fell continuously for eight years from 1992 to 2000, the longest such drop in modern American history. The conversion of resources from public to private use stimulated productivity, keeping labor costs low and promoting job creation and higher output.

By contrast, federal government spending has risen sharply over the past three years, crowding out more productive private activity. This, is turn, has kept productivity growth below what it would have been—and slowed down a needed decline in real unit labor costs (wage costs as a percent of sales) that is the basis for employment recovery. Estimated fiscal 2003 federal outlays are $422.9 billion higher than just three years earlier—an increase of 23.6 percent. The rise in public spending has contributed to the sluggishness in investment, as businesses are fearful to make long-term commitments in a period when activist government is clouding future private sector growth.

It is noteworthy that a majority of the federal spending increase is not related to national security, and that the spending binge had begun before the Sept. 11 tragedy. Since Sept. 11, in the name of national security, Congress has engaged in highly inefficient spending schemes, such as huge increases in farm subsidies. A vast expenditure to maintain military occupation in Iraq will not help either.

If this crowding out of private activity continues, the sluggishness in output growth and equity prices could well also persist, making recovery more tepid and potentially putting Bush’s presidency at great risk. The spending explosion, not tax cuts, are the largest cause of the big increase in deficits. From fiscal year 2000 to 2003 (finishing in less than three weeks), federal spending will have risen by an estimated $422.9 billion while revenue will have fallen only $268.9 billion—and much of that revenue decline was because of the recent recession, not the Bush tax cut. The budget deficit increases the demand for borrowed funds by the federal government; this in turn has contributed to the recent rise in interest rates, leading to some reduction in consumer spending and investment.

What should be done? Historically, in periods of above-normal unemployment, Congress rushes to pass a “stimulus package” to promote demand for goods and services. In a sense, we have done this already with the large increase in federal spending and shift from budget surpluses to massive budget deficits. On one thing modern economic history is very clear: Such Keynesian-style macroeconomic stimulus simply does more harm than good in the long run. We ran budget deficits throughout the 1970s and early 1980s—and had high rates of both unemployment and inflation. Japan has moved to massive budget deficits and government spending schemes for over a decade—and had stagnation and rising unemployment.

If I were giving political advice to President Bush, it would be not to panic, as market forces are working in his favor. However, I would urge him to erase one cloud on the horizon of recovery by moderating and hopefully stopping the spending surge. A resumption of the 1990s decline in federal spending as a percent of output would turn recovery from a tepid to a robust one. It would likely lead to rising equity prices and capital gains. To do this, we very well might have to reduce our commitment in Iraq, reverse unproductive increases in federal subsidies and engage in other forms of belt-tightening.

A massive increase in spending for Iraq and related anti-terrorist activities—and Bush has already asked Congress for $87 billion more for operations in Iraq—could leave the president as a political war casualty, not because of opposition to the military intervention itself, but because of its negative economic effects.