How and Why the Federal Jobs Report Is Revised

The Jobs Report and the Debate Over Its Accuracy

The Bureau of Labor Statistics (BLS) has become a focal point of political controversy after recent revisions to its jobs report. These changes, which reduced employment numbers for May and June by 258,000 jobs, have drawn sharp criticism from President Trump and his allies. They accused the agency of manipulating data for political reasons. However, most economists argue that these revisions are a normal part of the process and not politically motivated.

The BLS revises its numbers to incorporate more comprehensive data from its nationwide surveys. Despite the backlash, the agency is still considered the gold standard for macroeconomic data in the United States. Supporters of the BLS suggest that the agency could benefit from modernizing its data collection methods, especially for its population survey, which is one of two key surveys used to compile the jobs report.

A group of former BLS directors has urged Congress to allocate at least $770 million for the upcoming fiscal year to support the agency. Kyle Ross, a fellow at the left-leaning Center for American Progress, emphasized that ensuring the BLS has sufficient resources would be the best way to restore public confidence in its work.

How the Jobs Report Gets Revised

Each month, the BLS conducts a survey of over 120,000 employers via email and phone, gathering data on wages, total employment, and other factors. Based on this information, the agency publishes an initial estimate of job growth. It also conducts a separate household survey to track employment status and take-home wages across the country.

In the following two months, the bureau updates its estimates by incorporating additional responses and adjusting for seasonal variations. While the August revisions caught many economists off guard, they were not the first time the BLS made significant changes. During the pandemic, the agency had to revise many of its estimates, including a reduction of 626,000 jobs for June to September 2021.

Over the past two decades, response rates for key BLS surveys have declined. The Federal Reserve Bank of San Francisco estimates that response rates to the employment survey are around 45%, down from about 60% before the pandemic. However, the bank noted that these lower response rates did not appear to impact the size of BLS revisions after 2022.

According to analysis by Ernie Tedeschi, an economist at the Yale Budget Lab, the BLS’s initial job estimates have gradually improved in accuracy over more than 60 years of data collection.

Concerns Over Other BLS Metrics

While claims of political bias against the BLS are widely dismissed, advocates argue that the agency could use additional funding to modernize its operations, particularly for the Current Population Survey, which collects data from households rather than businesses.

Friends of BLS, an advocacy group that includes former commissioners William Beach and Erica Groshen, called on Congress in May to provide at least $770 million for the next fiscal year. In a letter to appropriators, the group stated that increased funding would allow the BLS to move forward with planned updates to its data collection and methods.

One of the modernization efforts involves implementing an online response model for the Current Population Survey. Additional funding, according to Beach and Groshen, would also help the BLS maintain detailed data for critical statistics like the consumer price index, which tracks inflation.

The BLS relies on field workers who travel across the country to monitor prices of goods and services. For example, a field worker might compare the price of a jar of Pringles from one month to the next, noting if the quantity has decreased while the price remains the same, indicating an actual price increase.

Last summer, the BLS considered reducing the sample size of the population survey by 5,000 households due to budget constraints. This decision sparked concerns among advocates about the potential impact on the quality of data collected.