Monthly job numbers from the Bureau of Labor Statistics (BLS) continue to surprise economists, this time by moving in opposite directions than usual. The rate of unemployment rose by 0.2 percentage points to 5.4 in data released today, reversing its January decline, while payroll job growth finally beat expectations with a solid increase of 262,000 jobs in February. The dissonance between the two top-line indicators over the last year - tightening unemployment rates alongside flat payroll growth - reveals deeper trends at work in the economy.

The workforce is aging, while society is getting wealthier and healthier. How these forces shape labor supply cannot be known with certainty, but the lesson for data watchers is clear: look at demographically neutral indicators if your goal is to assess the macro economy. It does no good to obsess over the total number of jobs if labor supply is contracting because people simply feel wealthier. Likewise, a booming number of jobs would be misleading if the cause is higher birth rates from two decades ago.

Instead, the best way to assess the health of the labor market itself is to measure the number of jobs relative to the available labor supply. Meet the unemployment rate. It is defined as the number of people who are unemployed divided by the labor force of people who have or want a job.

In February, the unemployment rate nationally was 5.4 percent, in line with what economists call the "full-employment" or "natural" rate. During President George W. Bush's first term, which included a mild economic downturn in 2001, the unemployment rate actually averaged 5.53 percent, which compares very favorably with the average since Harry Truman's second term in 1948 of 5.67 percent.

A comparison of unemployment rates during presidential terms in Table 1 shows that the first Bush term was sixth best. Few of the five "better" terms, however, included a recession, and all of them preceded recessions because they reflected overheated economies - most notably under President Lyndon B. Johnson in 1968 and President Bill Clinton in 2000. Few economists forecast a looming downturn as a result of today's labor market.

The other demographically neutral indicator to watch is the labor force participation rate (LFPR). For men, the LFPR has been on a gradual and steady decline for decades. For women, the rate has been rising. Both trends reflect that today's labor market is different than ever before. The big surprise in the data since 2000 is a collapsing LFPR for teenagers, which hit an all-time low in 2004.

There are many implications of this data, but the most important lesson is for policymakers who are considering increasing the minimum wage. Such a move may attract more teens into the labor force. But economists also expect a mandatory wage increase to restrain the supply of jobs. The inevitable result will be a higher unemployment rate, which may not be the legacy President Bush wants to build for his second term.

Tim Kane, Ph.D., is Bradley Research Fellow in Labor Policy in the Center for Data Analysis at The Heritage Foundation.