Gordon's Guesstimate: May Jobs

Authored by Gordon Gray, AAF’s Director of Fiscal Policy

Congratulations are due to the hundreds of thousands of Americans who started new jobs in May – on that much presumably everyone can agree. Beyond that, observations will likely diverge. Last month’s jobs report came in well below prevailing expectations, which were admittedly, this guesstimator’s included, pretty optimistic. But has anything changed such that last month’s conventional wisdom should be abandoned? Or is the noise of a single month’s data worth overlooking?

Since last month’s disappointment, initial unemployment claims have fallen steadily to a new pandemic-era low of 385,000. This number is certainly high, but it is no longer in uncharted territory. The 4-week moving average for initial claims is down about 134,000 for the month of May, though that is less of a decline than was observed from March to April, when payrolls were well under expectations.

For a good part of this year, ADP often understated employment growth, and while certainly not regularly or reliably, was something of a good, if conservative, indicator for the direction of the labor market. That relationship doesn’t hold up well enough to give analysts particular confidence in the measure for any given month, however. Looking at other indicators, there is plenty of demand according to this month’s JOLTS data. Job openings, for example, stand at 8.1 million, the highest on record. That makes a 1 million jobs gain plausible. But demand is only one side of the labor market.

There is an ongoing debate as to what is constraining supply. Understandably, for parents childcare may figure into that decision, although recent research suggests that this issue may not be a major constraint in the current environment. The virus, while receding, remains a constraint on behavior, and certainly that extends to job seekers. Policy, too, can compete with market forces. According to research by Isabel Soto, federal unemployment policies are such that “50 percent of workers in the United States could make more on unemployment than in their jobs.”

This guesstimator certainly didn’t adequately discount payroll expectations within the complex of supply constraints that policy and circumstance have imposed on the labor market. It would be an error to abandon expectations at the first contrary evidence, but it’s clear that there are limitations that were not adequately considered in last month’s optimism. They may be transient, but they may not be. For this month, this guesstimator is assuming a payroll gain of 750,000 and a 0.2 percentage point decline in the U-3.