In the middle of a recession, one of the most closely watched barometers of economic health is the unemployment rate.

But the unemployment rate, particularly at the regional level, can be misleading. Surprisingly, a low rate of unemployment isn't always a sign of economic health, and the change in the unemployment rate isn't necessarily a good measure of how well a region's economy is doing.

The Pittsburgh Region's unemployment rate in December was 6 percent. That's significantly higher than the 4.4 percent rate a year earlier, and reflects the fact that we have more than 20,000 more people unemployed than a year ago.

Although it's little comfort to the large and growing number of unemployed workers in the region, our 6 percent unemployment rate was lower than the U.S. rate (7.2 percent) and lower than Pennsylvania's (6.4 percent); in fact, it was the 11th lowest rate among the largest 40 regions, and we had the eighth smallest increase over the past year.

But are all of the 10 regions with lower unemployment rates really doing better than we are? Milwaukee's unemployment rate in December was only 5.8 percent, and its unemployment rate went up by only 1.1 percent over the previous 12 months, much less than the 1.6 percent increase in Pittsburgh. Yet from December 2007 to December 2008, the number of people working in Milwaukee decreased by more than 2 percent. That's faster than the decline in U.S. employment, and it was the 16th biggest decline among the top 40 regions. In contrast, employment in Pittsburgh actually increased slightly (by 0.3 percent) over that same time period.

How can Milwaukee have lost employment faster than Pittsburgh and the nation as a whole, but still have a lower unemployment rate and a smaller increase in the unemployment rate than either Pittsburgh or the United States?

The reason is Milwaukee's labor force decreased, whereas the labor force grew in Pittsburgh. Milwaukee had more than 7,000 fewer total workers (both employed and unemployed) in December than a year earlier, whereas Pittsburgh had 24,000 more workers than it did a year earlier.

A decrease in a region's unemployment rate doesn't always mean that its residents found work -- unemployment in a region also can decrease (or increase more slowly) if people get discouraged and stop looking for work or if they leave the region entirely, and that's what appears to be happening not only in Milwaukee, but also in Detroit, Cleveland, Minneapolis, and a number of other regions.

Although there's no way to know for sure, it's entirely possible that some people who lost their jobs in such places as Milwaukee, Cleveland and Detroit moved to Pittsburgh to look for work, since we had job growth here through most of 2008, whereas those other regions lost jobs in every month of the year. Someone can be unemployed in Pittsburgh without necessarily having lost their job here.

Similarly, some regions with bigger increases in unemployment actually can be doing better than we are. For example, Virginia Beach, Va., had a bigger increase in the unemployment rate over the past year than did the Pittsburgh Region (1.8 percent vs. 1.6 percent), but Virginia Beach also had the highest growth in employment of any large region in the country, thanks to a lot of military and government jobs. The labor force there grew faster than employment -- probably due to job-seekers coming from other parts of the country -- and that's why unemployment went up.

So when Pittsburgh's new unemployment rate comes out in mid-March, don't assume an increase is all bad news -- some of it may be new workers coming here. The only way to know for sure is to look at changes in jobs and the labor force, too, which you can find at the PittsburghToday Web site (www.pittsburghtoday.org).