Clients and readers have asked about the recent Philly Fed research paper discussing the missing one million workers. (See "Early Benchmark Revisions of State Payroll Employment.") We haven’t seen it moving markets. To add expertise to a discussion of the pros and cons of the report’s findings, we asked Cumberland’s Chief US Economist, David Berson, for an analysis and explanation. Below is David’s response.
David R. Kotok
Chairman & Chief Investment Officer
Email | Bio
David Berson wrote:
The gold standard for estimates of the number of employed workers in the US is the nonfarm payroll survey from the Bureau of Labor Statistics (BLS). Each month, the BLS conducts a survey of approximately 131,000 business establishments and government agencies. As additional survey responses are received, the estimate for monthly nonfarm payrolls is revised in each of the following two months. The BLS then does an annual benchmark revision based on state unemployment insurance tax records, which provides a comprehensive look at actual employment (not a survey). According to the BLS, the annual benchmark revisions have averaged plus or minus one-tenth of one percent of total nonfarm employment over the past ten years, suggesting that the monthly establishment survey figures are fairly accurate.
But much of the data on which the annual benchmarks are based are available quarterly in the Census of Employment and Wages (also from BLS). The Federal Reserve Bank of Philadelphia has created an estimate of nonfarm payroll employment for each state (and the total US as the sum of the states), combining the monthly payroll figures with the available quarterly data. This is a recent endeavor on the part of the Philly Fed, and it has the potential to provide estimates of nonfarm payrolls that would have smaller benchmark revisions.
The nonfarm payroll estimates from the BLS and the Philly Fed were reasonably close over the fourth quarter of 2021 (increases of 2.02 million from the Philly Fed vs. 1.91 million from the BLS) and the first quarter of 2022 (1.70 million from the Philly Fed vs. 1.62 million from the BLS). But there was a huge difference in the second quarter of 2022, with the Philly Fed’s estimating a gain of 10,500 nonfarm payrolls vs. 1.05 million from the BLS.
Why the big difference, and does one source provide a more accurate reading of the job market than the other? Clearly, if the Philly Fed estimate is close to the mark, the job market slowed significantly in the second quarter, while if the BLS estimates are closer, the job market remained solid.
The BLS also publishes monthly data on employment from the household survey (from which the unemployment rate is derived). This survey is much smaller than the establishment survey and so tends to be more volatile. But because the establishment survey simply can’t easily survey small firms on a monthly basis (it’s easy for the BLS to find Amazon, not so easy to find Joe’s Pizza), there is a trend figure plugged into the establishment survey to account for the undersampling of small and startup firms. This results in less volatility in the establishment survey (and perhaps a better measure of the underlying trend), while the household survey may be a better leading indicator of changes in the trend (at the cost of volatility that can mask the trend). What do the household data show? In the fourth quarter of 2021, employment growth was 2.17 million; in the first quarter of 2022, employment growth was 2.48 million; and in the second quarter of 2022, there was a decline of 347,000. But importantly, there was an increase of 825,000 in the third quarter, so the first-quarter drop may not have been the start of a dangerous trend. Moreover, if the household survey data are adjusted to payroll definitions, then the weakness in this measure disappears altogether.
While weekly unemployment claims don’t gauge employment by any measure, it would be very unlikely that low claims would correspond with a significant drop in payroll gains. Instead, they tend to move together. And the 4-week average of claims dropped to the lowest levels since 1969 in the first quarter of this year.
There are also private data on the job market. The ADP National Employment Report attempts to project the monthly nonfarm payroll figures using their data as the nation’s largest provider of payroll services. For the fourth quarter of 2021, the ADP data showed a gain of 1.17 million jobs. For the first quarter of 2022, there was also an increase of 1.17 million, while for the second quarter there was a gain of 1.19 million.
The National Federation of Independent Business does a monthly survey of small businesses that includes a number of employment-related questions. The responses to almost all of these questions (among them questions regarding plans to increase employment, average number of workers per firm, and difficulty in finding qualified applicants for job openings) have indicated a very tight labor market for most of the post-Covid-recession period.
The vast majority of the employment data showed job strength in the first quarter, rather than the significant slowing that the Philly Fed estimates. Even in the household survey employment estimates, the drop reversed course, and jobs strengthened again in the second quarter.
The Philly Fed methodology is new and does not cover much history, so it’s difficult to claim that it’s either better or worse than the payroll survey estimates from the establishment survey. Note also that the Philly Fed uses their own quarterly seasonal adjustments of the state employment data, while the benchmark revisions to the payroll data use non-seasonally adjusted annual data. Contrasting methodologies could create much of the difference between the BLS payroll and Philly Fed measures. Because of the importance of correct seasonal adjustment, several years’ worth of data may be needed to determine if this is being captured correctly.
Is the job market slowing? Probably yes. But the very low estimate of 2Q22 payroll gains from the Philly Fed seems like an outlier at this time.
David W. Berson, Ph.D.
Chief US Economist
Email | Bio
Links to other websites or electronic media controlled or offered by Third-Parties (non-affiliates of Cumberland Advisors) are provided only as a reference and courtesy to our users. Cumberland Advisors has no control over such websites, does not recommend or endorse any opinions, ideas, products, information, or content of such sites, and makes no warranties as to the accuracy, completeness, reliability or suitability of their content. Cumberland Advisors hereby disclaims liability for any information, materials, products or services posted or offered at any of the Third-Party websites. The Third-Party may have a privacy and/or security policy different from that of Cumberland Advisors. Therefore, please refer to the specific privacy and security policies of the Third-Party when accessing their websites.
Cumberland Advisors Market Commentaries offer insights and analysis on upcoming, important economic issues that potentially impact global financial markets. Our team shares their thinking on global economic developments, market news and other factors that often influence investment opportunities and strategies.
