Jobless people are classified into one of two categories by the Bureau of Labor Statistics (BLS)—either unemployed or not in the labor force. To be classified as unemployed in the month they are surveyed, people must be actively looking for work. If they are not actively looking, they are classified as not in the labor force. This is a reasonable distinction to make. In 2017, about 40 percent of American adults were not working in a given month. If all of these adults were actively looking for work and not finding it, this would be an economic crisis. If none of these adults wanted work and were instead voluntarily retired, taking classes, or looking after dependents, it would not be a problem at all.

The unemployment rate is defined as the number of unemployed workers divided by the sum of employed and unemployed workers (or, the labor force). This rate tries exactly to capture what share of the adult population wants work, but hasn’t found it. This is why it is the most commonly referenced measure of “slack” in the labor market, or how many adults need new jobs created for them to fill. Today’s low unemployment rate has led many to conclude that the American labor market has very little slack left in it, and that new jobs going forward will be hard to fill without raising wages (perhaps even to inflationary degrees, the worries go).

The figure below provides a reason to be wary about accepting the unemployment rate as the only measure of labor market slack. It shows the share of newly employed workers who were not looking for work in the previous month. Two things stand out from the graph. First, the share of newly employed workers who were previously not searching for work is always high—well over half. Second, recent years have seen this share hit historic highs.

There are a couple of policy takeaways here. First, because more and more jobs are being filled by people claiming to not have been looking for work it seems like the unemployment rate is becoming less useful as a clear-cut measure of labor market slack—this means we shouldn’t rely on it alone to decide whether or not the economy is at full employment. Second, even with unemployment low, it seems like Americans have plenty of appetite for new jobs (and particularly for good jobs). This means we should still be thinking hard about strategies for job creation—like I did in a recent paper. Finally, we don’t need policymakers who are committed to slowing the economy because they think unemployment has fallen low enough. Instead, we need policymakers willing to aggressively test just how fast the economy can grow before sparking inflation. The most important policymakers in this regard are the Federal Reserve, and this makes the next pick to head the Federal Reserve Bank of New York crucially important.