Ron Wirtz of the Federal Reserve Bank of Minneapolis calls it “a hole in the employment doughnut” — the growing number of noncovered jobs in the US. EMSI tracks data on workers not covered by unemployment insurance, and thus not captured in traditional job statistics (examples include self-employed workers, independent contractors, and commission-only salespeople).
Trying to pin down a tricky segment of the workforce to measure, the Federal Reserve focused on our noncovered data in a fedgazette post.
Economic Modeling Specialists, Inc., for example, is an economics and labor market consulting firm in Moscow, Idaho. It shared its estimates on noncovered workers with the fedgazette, as well as its methodology, “quite a bit (of which) is based on the BEA’s much-appreciated work in this area,” said Jared Miller, an EMSI data analyst.
Unlike the population of covered workers, which has taken a big and well-publicized drop during the recession, EMSI’s data show steady, if modest, growth among noncovered workers in every district state, as well as the nation as a whole …
The Minneapolis Fed has six states in its district — North and South Dakota, Minnesota, Montana, Michigan, and Wisconsin — and Wirtz graphed out the growth of the noncovered population in each state since 2006 (see chart below). Montana, as we mentioned in an earlier post, has the highest percentage of noncovered workers in the US (nearly 30% of its overall workforce). But Wirtz’s index also shows North Dakota has seen a dramatic increase in the raw number of noncovered workers.
Wirtz also outlines two important caveats to consider with this data:
For starters, the noncovered population is a count of jobs, not employed individuals, which means there are multiple plausible interpretations for job trends. From an optimistic standpoint, a growing number of noncovered jobs might mean there are more employed individuals in this gray area of employment. However, a growing number of noncovered jobs might also be indicative of growing part-time jobs and outsourced labor in the form of independent labor contracts. Were this the case, total noncovered workers could well be stagnant or even falling as they take on more of these jobs to make ends meet.
Secondly, noncovered job counts cannot distinguish employment duration; in many cases, someone self-employed for a single month would count the same as a year-long job, whereas QCEW is more precise, reporting monthly job counts as well as quarterly and annual averages.