The Government Accountability Office report also offers the Bureau of Labor Statistics suggestions for further improvement.
The April jobs report, released on Friday, should put to rest concerns about the strength of the U.S. labor market. Nonfarm payrolls grew by 172,000 in the month, while payroll growth for February and March was revised upward by 93,000.
Similarly, a newly issued report by the Government Accountability Office ought to quell concerns about the quality of employment data produced by the Bureau of Labor Statistics, which assembles the monthly jobs report. The GAO report found that the integrity of the data is intact, something that President Donald Trump questioned last August when he fired then–BLS Commissioner Erika McEntarfer after the July payrolls report was much weaker than expected. Trump charged, without evidence, that the report had been manipulated for political reasons.
The GAO’s findings come at an opportune time for McEntarfer’s would-be successor, Brett Matsumoto, whose Senate confirmation hearing is scheduled for June 10. Trump initially nominated conservative economist and longtime BLS critic E.J. Antoni to head the agency, but the nomination was withdrawn after it became apparent that some Republicans opposed the pick. William Wiatrowski, a career BLS official, has been acting commissioner of the BLS since last August.
The GAO, tasked with reviewing the jobs data after McEntarfer’s termination, found that the BLS generally succeeds in providing accurate, useful, and timely information. “We found that BLS met its annual accuracy goals for the survey estimates in the jobs report from fiscal years 2020 through 2025 amid data collection challenges,” its report stated.
Those challenges could loom large for Matsumoto, a longtime BLS employee who has been on loan for the past year as a senior economist to the Council of Economic Advisers. If confirmed, he would take the helm of an agency that has lost about 20% of its staff over the past 18 months and experienced significant budget cuts. He will be tasked with not only restoring public trust and bolstering data quality, but also defending the agency against further political interference.
The GAO’s report cited future risks to data quality due to declining survey response rates. But it gave the BLS credit for taking steps to help mitigate this risk by developing an online response method for the household survey, which informs the unemployment rate. That method will be put in place next year.
The GAO also called out the BLS’ efforts to improve its net birth-death model; updates were rolled out in February. The model is used to estimate the net change in monthly jobs created by the birth of new businesses and the closure of existing businesses. An overstatement in net job growth stemming from birth-death model estimates was cited as a driver of large revisions to reported payrolls in recent years.
The updates seem to be effective. The quarterly census of employment and wages administrative records released this past week differed little from the nonfarm payroll data covering the same period. The new birth-death model may have helped to better align the two data sets, said Dan Pan, an economist at Standard Chartered.
Pan thinks the annual preliminary benchmark revision to payroll data, to be released in September, may validate the nonfarm payroll estimates, something that hasn’t happened in recent years.
The GAO staff also explored the viability of other data-quality enhancements, including the use of blended data from multiple information sources, including surveys, administrative records, and alternative data, to calculate employment rates. It considered mandatory participation in federal surveys, and floated the idea of offering compensation to businesses and individuals participating in BLS surveys. Most of these initiatives, however, were considered too resource-intensive, and posed separate challenges.
The GAO staff outlined three recommendations for the BLS, including charging the BLS commissioner with obtaining more input on changes to the jobs report and being more transparent with the public on the effects of declining response rates. It also urged the BLS to publish an assessment of possible bias stemming from lower response rates, which is already under way.
The BLS informed the GAO that it had tapped an “outside entity” to conduct a new study on the effects of lower survey response rates for the payroll survey data, and identify improvements. The study is expected to be completed later this year.
The GAO report is an important step in confirming the validity of BLS data and providing Matsumoto with a framework for further improvements.
Write to Megan Leonhardt at megan.leonhardt@barrons.com