Don't Sweat The Sickly Employment Numbers

Job data can be unreliable. But runaway spending is real

By Robert J. Barro

The employment report for December was weaker than expected, with the payroll number nearly unchanged and the household survey number falling by 50,000. Although disappointing, these results should not carry too much weight. A forecasting equation I use shows that monthly changes cannot be predicted accurately. This equation considers the history of employment and gross domestic product and recent claims for unemployment insurance. The current prediction is for a January gain of 65,000 payroll jobs. But the range could be -90,000 to +225,000.

In the long run, the payroll and household surveys show similar growth. However, the two can diverge for long periods. In the late 1990s, payroll grew faster. For the past couple of years -- though not last month -- the pattern was reversed, and household climbed more. For example, in the past four months, payroll employment advanced by 240,000, while household employment rose by 790,000. Over the past year, payroll fell by 70,000, while household grew by 2 million.

Recent disparities between payroll and household numbers are not well understood. The payroll figure should be more accurate, because the survey is larger and based on harder data, and the household survey estimates the labor force in a rough manner. However, some conceptual differences favor the household survey. For example, the payroll data include multiple job holders and exclude the self-employed, farm, and domestic workers. Still, adjustments for measurable differences do not explain the discrepancies. Thus, it seems best to weigh both when evaluating the state of the labor market.

Some observers say recent drops in unemployment do not count because they represent decreases in the labor force, notably withdrawals of discouraged workers. The labor force is volatile from month to month, and one should not make much of December's decline of 310,000. In the past year, the increase in the labor force by 1.7 million represented 1.2% growth -- close to the annual growth rate of 1.4% seen since 1980. So, there is no validity to the argument that the labor force behaved in an unusual way during 2003.

More generally, the economy has been performing strongly, as seen in robust gdp growth. Other favorable factors are the strong pickup in investment, the sharp growth in labor productivity, and the recent increases in real personal income. Given the strength of the overall economy, it is puzzling that the labor market has not been stronger. One possible explanation is the rapid growth of productivity related partly to the outsourcing of jobs. Strong productivity explains why output can grow without commensurate increases in employment. However, in the long run, employment growth should correspond to the growth of the labor force -- 1% to 1.5% per year -- no matter that happens to productivity.

The strong economy reflects the Bush Administration's two best policies: tax cuts, especially the incentive-based ones in 2003, and the pursuit of international security through the operations in Afghanistan and Iraq. Another plus is the expansionary monetary policy. Low interest rates will probably continue until inflation or employment rises. These two variables -- not gdp growth, the dollar exchange rate, the current-account deficit, or the stock market -- are the ones to which the Fed has responded since the early 1980s.

Despite my approval of Bush's policies, I am concerned about the failure to restrain spending. For President Reagan in the 1980s, tax cuts and deficits were used to halt the growth of spending, particularly outside of defense and the interest on the national debt. But in the past three years, federal outlays aside from defense and interest on the federal debt have exploded at an annual rate of more than 8%.

The big question is would the Administration be more prudent in a second term. This seems relevant since the Democrats seem intent on committing electoral suicide by nominating a candidate who resembles George McGovern from 1972. (In my opinion, Howard Dean is clearly qualified to be President...of France... or, perhaps, of Boston University.)

The favorable view is that a more Republican government would avoid the political necessity of enacting programs such as expanded Medicare, enlarged outlays for education, steel tariffs, textile quotas, and farm and ethanol subsidies. The pessimistic view is that divided government worked well to restrain the growth of government in the 1980s and '90s and only a return to that division would work now.

Robert J. Barro is a professor of economics at Harvard University and a senior fellow of the Hoover Institution (

    Before it's here, it's on the Bloomberg Terminal.