Corporate payrolls increased by a robust 234,000 in January, according to this morning’s ADP Employment Report. The gain is slightly below December’s 242,000, although both numbers point to a solid rate of expansion in the labor market. Taking today’s update at face value points to a better-than-expected advance in Friday’s official employment report that’s due from the Labor Department. The mystery is why the government’s data to date has been trailing ADP’s estimates by a comparatively wide margin. Will the upcoming report from Washington close the gap? Or has the ADP data been overestimating the strength of the economy’s record on minting new jobs?
Consider the diverging paths for one-year changes in the ADP and government figures. For the past 12 months, ADP has been reporting a re-acceleration in employment growth in the private sector. At the end of 2016, ADP’s estimate of one-year growth slumped to 1.8% — well below the post-recession peak of 2.5% from two years earlier. Over the past year, however, the one-year trend via ADP has been rebounding and is now close to a two-year high.
The Labor Department’s numbers, however, tell a different story for private-sector employment: the trend has continued to decelerate. The 1.6% annual increase for December via the government’s reckoning is near a seven-year low.
Note, too, that the spread between the two data sets in monthly terms is usually relatively narrow. Although the pair rarely match exactly, any differences are usually short-lived and the spread tends to bounce within a limited range. But that history has given way in recent months and the difference between the two has exploded to unprecedented levels relative to the ADP data’s history that begins in 2002. (The Labor Department’s time series, by contrast, dates to 1939.)
Using the 2002-2017 track record for the two series, however, paints an upbeat forecast for this Friday’s report. Plugging the numbers into a simple regression model points to an expected gain of 228,000 for the government’s January update on private payrolls – substantially above Econoday.com’s consensus forecast of 172,000.
If the 288,000 rise materializes, the gain will lift the one-year trend for the government’s data to a 1.7% increase through January – a seven-month high. A weaker increase, on the other hand, would continue to raise questions about why the Labor Department is still reporting a conspicuously softer trend. The obvious answer: one data set is wrong. For now, it’s debatable which one deserves that label.
Reliable intelligence for monitoring the business cycle:
The US Business Cycle Risk Report