The Dwindling Survey Response Problem

“Jobs Day” – typically the first Friday of each month – is a fixture on the economic calendar as it marks the release of the U.S. jobs report from the Bureau of Labor Statistics (BLS). Economists and policymakers feverishly parse through the data in search of trends and inflection points to gauge the economy’s trajectory – and potentially respond with policy adjustments reflective of the evidence.

Yet there is growing concern that the accuracy of the employment report, and other government survey data, is diminishing. Many reports have been plagued by dwindling response rates that can introduce or exacerbate biases in the data and subject them to significant revisions.

Over the past decade, the response rate for both surveys that make up the employment report has fallen by nearly 20 percentage points. The response rate to the Current Employment Statistics (CES) survey, also called the establishment survey and used to estimate the monthly change in jobs, dropped from 63 percent in April 2014 to 43.5 percent in March 2024. The response rate to the Current Population Survey (CPS), better known as the household survey and used to produce the unemployment rate, fell from 88.8 percent in April 2014 to 69.7 percent in April 2024.

The decline in response rates isn’t limited to the CES and CPS. The response rate to the Job Openings and Labor Turnover Survey, or JOLTS, which produces data on job openings, hires, and separations, fell by more than half over the past decade—from 68 percent in April 2014 to 33.2 percent in March 2024. The response rate to the employment cost index survey, measuring hourly labor costs to employers, slumped from 73.4 percent to 46.3 percent over the same period. Response rates for various inflation surveys, including CPI, have also seen steady drop-offs.

The problem is seemingly ubiquitous.

In nearly every speech, Federal Reserve Chair Jerome Powell emphasizes that the Fed’s policy decisions are “data dependent,” elevating the need for reliable and timely data. If the data are anything less than reliable, the Fed and other policymakers are likely to miss inflection points and put policy adjustments behind the curve.

To combat the problem, BLS is adjusting data collection methods and, when possible, supplementing responses with other timely data to fill the gaps. Yet these efforts alone are unlikely to reverse lagging response trends and adequately replace data best captured by surveys. Considering the importance of these data, Congress must not only ensure that government data collectors have sufficient resources to study alternative methods of data gathering, but also implement necessary changes including sufficiently expanded survey samples to help offset the drop in response rates.

Freddy’s Forecast: June Jobs

Payroll employment regained strength in May as employers added 272,000 workers. The unemployment rate ticked higher to 4.0 percent, snapping its streak of 27 consecutive months below that mark. Continuing the recent trend, hiring was led by the health care and social assistance, leisure and hospitality, and government sectors. Wage growth came in at 4.1 percent from a year ago.

Since the last report, data from ADP showed private-sector payrolls expanded by 150,000 in June, led by gains in leisure and hospitality and construction. Small business hiring was weak, up just 5,000 as establishments with 20–49 employees shed 8,000 workers.

The JOLTS report showed that the balance between supply and demand continued to normalize as the number of vacancies for every unemployed person remained unchanged at 1.22. The ratio was well off the high of 2.0 reached in March 2022 and more in line with the 1.19 average in 2019.

Reports from the Institute for Supply Management were weak as both the manufacturing and services sectors contracted in June. The employment indicator in each report showed hiring in the manufacturing sector slowed at a faster clip than the prior month while hiring in the services sector went from growing to contracting.

Initial claims remained relatively benign, but have been inching higher since January, reaching 238,000 for the week ending June 29. An uptick in the trend of continuing claims, which reached 1.858 million, suggested that those who are currently unemployed are having a more difficult time finding work. The latest reading was the highest since November 2021.

The jobs market is likely to keep chugging along at a healthy clip. Expect payrolls to expand 192,000, with the unemployment rate holding steady at 4.0 percent. Growth in hourly earnings sticks to the trend, up 10 cents or 0.3 percent.