U.S. Jobs Report Remains Inaccurate Despite Revisions

The so-called “jobs number” is inherently inaccurate, as it is a statistical estimate based on incomplete data, designed to provide early insight rather than absolute precision [1]. The U.S. Bureau of Labor Statistics (BLS) acknowledges this by updating its reports as more information becomes available [1]. For instance, the July jobs report, which initially stated 73,000 jobs were added, is subject to revision as more data is collected [1]. The BLS does not wait for all 121,000 surveyed employers to respond, often working with only 60% of the data at the time of the initial report [1]. Even after all data is collected, the final number remains an estimate based on various assumptions [1].

The process of compiling and reporting job data has a long history of political scrutiny. For example, in 2004, the Economic Report of the President suggested reclassifying fast-food cooks as manufacturing workers, citing the chemical transformation that occurs when a frozen patty is heated [1]. The BLS rejected the proposal, adhering to its standard definition of manufacturing, which requires a transformation into a “new product” [1]. The proposal was widely perceived as an effort to artificially inflate the manufacturing sector’s performance [1].

Historical precedents show that job data has frequently been a political flashpoint. In 1971, the Nixon administration criticized the BLS for downplaying a drop in unemployment [1]. A month later, a statistical error led to further accusations of manipulation. In 1961, Reader’s Digest accused the Kennedy administration of overstating unemployment to justify increased government intervention [1]. While investigations found no evidence of manipulation in these cases, they underscore the sensitivity of job data in political discourse [1].

The BLS has also faced internal political pressures. In 1932, Ethelbert Stewart was “involuntarily retired” for disagreeing with the Hoover administration’s optimistic view of the labor market [1]. More recently, Erika McEntarfer, the BLS commissioner, was fired under President Trump, who accused her of politically motivated miscalculations [1]. The BLS website itself highlights numerous precedents that suggest the agency’s independence has often been questioned [1].

Despite these controversies, the BLS plays a crucial role in providing early signals about the job market. As noted by statistician George Box, “All models are wrong, but some are useful” [1]. The BLS model, while imperfect, allows businesses, investors, and policymakers to react quickly to economic trends [1]. While the final accuracy of job market data will only be known once all tax reports are filed a year or so later, the initial estimates are invaluable in shaping immediate economic responses [1].

The power of numbers to influence perception is significant. Economic data can trigger self-fulfilling prophecies, where the fear of a recession turns into an actual slowdown [1]. However, with high-frequency data like non-farm payrolls, accuracy can sometimes hinder utility [1]. The goal is not to achieve perfect precision but to provide timely and actionable insights [1].

Source: [1] The Breakdown – Jobs Report Number Is Always Wrong (https://blockworks.co/news/jobs-report-number-is-always-wrong)