What’s the deal with those payroll revisions?
First off, it’s key to remember these revisions aren’t a pressing risk for investors. Revisions are backward-looking. That said, they can make markets more reactive to data releases, so it’s still something investors should keep on their radar.
In August, the Bureau of Labor Statistics (BLS) announced payroll gains between April 2023 and March 2024 were 818,000 fewer than originally reported. These revisions stemmed from the annual benchmarking process, where the BLS compares its survey data to more comprehensive administrative data sources, such as unemployment insurance records.
Why did this happen? In stable times, the BLS’s “birth-death” model has tended to be more accurate, but during turning points in the cycle, those assumptions can be off. For instance, as growth slows, the BLS might overestimate jobs from new startups – “births” – and underestimate losses from business closures – “deaths”. Another factor behind larger-than-usual revisions could be that undocumented immigrants are excluded from the data, as revisions are based on unemployment claims, which only account for those eligible for benefits.
One of the reasons why we don’t see revisions as a major market risk is because the data is available for everyone, and the market can sniff it out. One of my favorite conspiracy theories is that the BLS inflates jobs numbers to boost the sitting administration. While no one can guarantee it’s not happening, I believe the market is sharp enough to sniff out any manipulation. If there were a consistent pattern of inflating numbers and quietly revising them later, the market would have already modeled a shadow figure and traded accordingly.
On a monthly basis, revisions are driven by response rates getting weaker
Another challenge for data analysis is the steady decline in response rates to federal surveys (not only labor surveys), a trend that accelerated during the pandemic as household and business patterns were disrupted. If this trend were to worsen, lower response rates could undermine the credibility of data releases. Federal agencies are working to improve response rates, but until they do, (initial) data releases might be less reliable.
The chart below shows the drop in BLS survey response rates, with the second release – typically lagging a month behind – offering a more accurate reflection of employment stats. It’s no surprise then that monthly revisions have increased as well.
On the other hand, consumer data revisions show resilience
In September, the Bureau of Economic Analysis (BEA) revised its estimates for wages, profits, and economic growth upward. The update also showed that households saved more than previously reported, driven by stronger wage growth, government benefits, and investment income. Government transfer payments, including Social Security and unemployment benefits, further boosted overall income levels.
The upward revision to overall economic output and the downward adjustment in labor market data likely suggest evidence of higher productivity. Like payroll revisions, the growth revisions are backward-looking, but they suggest a stronger outlook for consumers. As our economic dominoes framework highlights, a stable consumer can support corporate profits and keep labor growth intact.
Portfolio strategy - What’s the point of all this?
So, should investors care? What should they do? Investors should be prepared for more market swings as data continues to drive sentiment. The best defense against volatility and potential drawdowns is staying diversified – and that means more than just stocks and bonds. Consider adding real estate, infrastructure, or alternative investments like private equity and hedge funds, which can behave differently from traditional assets.
Related Thought Leadership
By subscribing you are consenting to receive personalized online advertisements from New York Life Investments.