The U.S. unemployment rate fell to 4.5 percent in March, the lowest level in nearly a decade, from 4.7 percent, and an encouraging sign for the job market. Yet employers added just a meager 98,000 jobs — barely half the pace of January and February.
How did the unemployment rate manage to drop so much when the job gain was so weak?
The reason relates to a little-known fact about the government's monthly jobs report: The government conducts one survey to learn how many jobs were created and another to determine the unemployment rate. And the two surveys sometimes produce conflicting results.
One is called the payroll survey. It asks mostly large companies, nonprofits and government agencies how many people they employed during the month. This survey produces the number of jobs gained or lost. For March, the payroll survey showed that employers added 98,000 jobs.
The other is the household survey. Government workers ask whether the adults in a household have done any work for pay that week. Those who have not are asked whether they're looking for a job. If they are, they're considered unemployed. If they have not looked in the previous four weeks, they're not considered part of the workforce and aren't counted as unemployed. The household survey produces each month's unemployment rate.
In March, the household survey showed that 472,000 more people found jobs. That lowered the number of unemployed and cut the unemployment rate.
The household survey, unlike the payroll survey, captures farm workers, the self-employed and people who work for new companies. It also does a better job than the payroll survey of capturing hiring by small businesses.
But the household survey is more volatile from month to month. The Labor Department surveys just 60,000 households, out of more than 100 million U.S. households.
By contrast, the payroll survey seeks information from 147,000 companies, nonprofits and government agencies. They employ roughly one-third of non-farm employees. The employers send forms to the Labor Department or fill out online surveys, noting how many people they employ. They also provide pay, hours worked and other details.
Most Americans focus more on the unemployment rate, which comes from the household survey. But economists generally prefer the jobs figure from the payroll survey.
While they can diverge each month, the two surveys tend to even out over time.