Risk.
Getty
Last week the Bureau of Labor Statistics (BLS) substantially altered the economic picture when it reported revised numbers of total jobs created over the last year. The revision was substantial—private-sector job creation had been overestimated by more than 500,000 jobs in 2018.
The revision helps explain some odd phenomena—like wages not increasing rapidly when the unemployment rate was reportedly so low. This historically high downward revision means economic performance is weaker than we thought and is consistent with growing pessimism and puzzlement among economists that the economy looked weaker than the jobs numbers last year. The major revisions included 321,000 fewer jobs in the low-wage sectors of retail and leisure and hospitality and about 163,000 fewer in the professions that serve businesses, like lawyers and accountants.
As it happens, the downward revisions were heavily weighted toward one of the sectors that was seen to have experienced rapid growth more recently. In the July jobs report, released August 2, the BLS reported ”notable job gains” in professional services, among other sectors. Now this recent job growth looks less impressive against the backdrop of diminished growth in 2018.
The sharp downward revision is neither part of a conspiracy nor does it indicate poor performance by government employees. Revisions are part of a normal annual process. Two monthly surveys assess the labor market: a survey of people and a survey of businesses. The payroll jobs numbers come from the business survey. Every March, the statisticians and economists at BLS benchmark those job survey estimates to state unemployment insurance tax records filed by employers, which provide a more comprehensive universe of payroll employment. Rarely, however, is this revision significant enough make the news. But this year is different. As Oxford Economics’ Gregory Daco tweeted, these downward revisions are the largest since 2009 and could signal lower total employment in the coming year.
The natural question to ask, given the downbeat revision, is what it says about recession forecasts. About three-quarters of economists surveyed in August by the National Association of Business Economists said they expected a recession no later than the end of 2021, with a slight belief in favor of it hitting in 2020. Although these new, lower employment estimates had not been factored into their economic projections, this new sign of weakness is consistent with economists’ pessimism. (Economic models are imperfect at forecasting recessions or other economic turning points, and other indicators aren’t much better.)
Other indicators, at least for GDP growth, are not signaling a negative short-term outlook. Two regional Federal Reserve banks compute a running prediction of quarterly growth based on how the components of overall GDP growth are performing. The New York Federal Reserve Bank’s “Nowcast” for the third quarter of 2019 is currently at 1.76%, and the Atlanta Fed’s “GDP Now” is at 2.3%. These are positive numbers; a recession prediction would be predicting a fall in GDP growth.
But what if a recession is around the corner, even if not looming in the current quarter? The Federal Reserve may try to bail us out with further interest rate cuts, although some economists agree with former Treasury Secretary Larry Summers that conventional central bank policy “is not an adequate response to the challenges now confronting the major economies.” And the recent federal budget agreement between Democrats and Republicans will add billions in deficit spending to the economy, providing a positive stimulus between now and the 2020 elections that might stave off a recession.
In conclusion, the downward payroll revision, while part of ordinary technical work of the BLS, is unusually large and negative. The revision is nothing but bad news about the economy. It is one more reason to worry about a coming recession.