August payrolls were revised up to 186,000 from 133,000, and September payrolls were revised up to 60,000 from 29,000, a combined upward revision of 84,000 jobs that says the summer was stronger than the first prints suggested. The revisions, buried in the October jobs report released on Friday, are the most important numbers in the release after the headline itself, because they change the story of the past three months, not only of the past month. The labour market was not weakening through the summer. It was stronger than anyone knew, and the first estimates were wrong.
Why it matters
The reason revisions matter is that the first payroll estimate is an early reading of an evolving dataset. The Bureau of Labor Statistics collects payroll data from employers over several weeks, and the first estimate is based on an incomplete sample, which is then revised as more employers report. The revisions are usually small, a few thousand jobs in either direction, and they are usually ignored. A combined revision of 84,000 is not small, and it is not ignorable. It is the difference between a labour market that was stalling and a labour market that was growing, and it changes the judgement of every policymaker who acted on the first prints.
The October headline makes the revision story even stronger. Employers added 165,000 jobs in October, more than two and a half times the 65,000 economists had expected, and the unemployment rate dipped to 4.1 percent from 4.2 percent. The October number is preliminary, and it too will be revised, but the direction of the revisions so far suggests that the first estimate is more likely to be revised up than down. The average monthly gain over the past year had been only 48,000, and the October gain is more than three times that average, which is the kind of number that makes forecasters rewrite their models.
What happens next
For the Federal Reserve, the revisions change the argument without settling it. A labour market that was stronger through the summer than first reported gives policymakers more room to focus on inflation, which ran at 3.7 percent over the year to September, and traders raised the odds of a December rate increase to about 78 percent. The Fed held its benchmark rate at 3.75 to 4.00 percent at its October 28 meeting, and its next decision comes on December 8 and 9, after the November inflation report on November 10. The revisions say the economy does not need relief. The inflation data will say whether it needs restraint.
What happens next will be decided by the next revisions, and by the next report. The November jobs report, due on December 4, will show whether October strength holds, and whether the revisions continue to run upward. For workers, the message is simpler: the jobs were there all along, even when the first estimates said they were not, and the labour market is stronger than the headlines suggested. For forecasters, the message is humbler: the first estimate is a guess, and the revision is the truth, and the truth, this time, was better than the guess.
