The latest statistics on poverty and income were released in the 2023 Current Population Survey’s (CPS) Annual Social and Economic Supplement (ASEC) on September 10, 2024. The ASEC supplement to the nation’s primary household labor force survey—the CPS—provides essential information on trends in poverty, income, and health insurance coverage and measures the financial stability of families across the country as well as the effectiveness of government programs that aim to improve these measures. The survey data, collected in partnership between the U.S. Bureau of Labor Statistics (BLS) and U.S. Census Bureau, are used to help policymakers determine the effectiveness of programs by tracking changes in poverty levels of different demographic groups over time and determine unmet need.
Unfortunately, funding constraints at the BLS could soon require a cut to the sample size, reducing the accuracy of estimates made using these data and all other statistics that use the CPS. They could also hamper efforts to modernize the survey to address declining response rates, which have affected government surveys for years.
The reports delivered by CPS data are so influential they can cause swings in the stock market. To ensure that it can continue to accurately reflect key economic experiences of the American people and serve as a reliable resource to guide relevant policy decisions, particularly those surrounding the labor market and inflation, Congress must at least provide the BLS with enough funds to match the request totaling $812 million for fiscal year (FY) 2025 made in a joint letter by two former BLS commissioners appointed by presidents of different parties.* This level of funding should stave off cuts to the sample size and allow BLS to move forward with modernization plans for the CPS while improving data collection or gathering new cohorts for the agency’s other surveys, including the Job Openings and Labor Turnover Survey, Consumer Expenditure Survey, Enhanced Unemployment Insurance Records, and the National Longitudinal Survey of Youth.
The Bureau of Labor Statistics faces a shrinking budget
Funding constraints for the CPS are an issue that BLS and the Census Bureau, which jointly sponsor the survey, are all too familiar with. According to Center for American Progress analysis, funding for BLS has dropped by more than 13 percent compared with FY 2001 in inflation-adjusted dollars, the year in which the sample size was increased to the roughly 70,000 households it remains at today, of which about 60,000 are eligible to take the survey. When compared with 2010, which marked a high point in BLS funding, appropriations have been cut by more than 20 percent. (see Figure 1) While both agencies have faced declining inflation-adjusted funding in recent years due primarily to caps on nondefense discretionary funding, BLS has reached a breaking point.
The need for cost cutting has become so significant that, without any remaining major sources of savings left to turn to, BLS Commissioner Erika McEnterfer stated that budgetary pressures may require a 5,000-household cut to the current CPS monthly sample. The reliability of overall numbers such as the national unemployment rate would not be jeopardized, but it would make unemployment rates among subpopulations—for example, by race or ethnicity, age, and geography—less reliable. Simultaneously, the costs of data collection have increased over the years, forcing BLS to cut costs by reducing in-person visits, nonresponse follow-ups, field training, and travel costs. Because data collection has become significantly more expensive, keeping up with inflation alone would be insufficient.
Currently, the FY 2025 appropriations proposals for BLS from both the U.S. House of Representatives and the Senate do not meet the agency’s needs and will not stave off a cut to the CPS sample size. (see Figure 2) In fact, they both propose a cut from FY 2024 levels after adjusting for inflation. Even the agency’s funding request is likely an insufficient amount, hamstrung by the debt ceiling deal for FY 2024 and FY 2025 that limited spending growth. For example, the BLS request for FY 2025 is more than 11.5 percent smaller in inflation-adjusted terms than their FY 2023 request.
Congress should look to the bipartisan budget agreement organized by Sens. Patty Murray (D-WA) and Susan Collins (R-ME) as a source of funding to reverse the erosion of CPS resources. Even just an extra $182 million—0.9 percent of the total funds from that agreement—and the budget increases allowed under the existing deal combined would provide BLS with enough funds to meet the various needs outlined by former BLS Commissioners Erica Groshen and William Beach in Friends of BLS’ May letter to the appropriations committees. This funding is comparable to prior year levels and is essential for supporting many government surveys, including the CPS, which the commissioners describe as being “on budgetary life support.”
How have response rates changed over time?
A 5,000-household cut to the sample size would put CPS data quality at even greater risk amid falling response rates. The CPS, like many other government surveys, has suffered from declining response rates in recent years, falling from around 90 percent to 70 percent in just the past decade. (see Figure 3) A variety of factors have likely contributed to this trend, including respondent views that surveys are burdensome, survey fatigue or confusion, and greater distrust of unknown callers leading to more reluctance to answer the phone. The end result is that the CPS is receiving only around 42,000 monthly responses from its sample of roughly 60,000, fewer than at any point in recent history excluding the COVID-19 pandemic and even less than the period after the last sample size cut due to funding constraints in 1996, when the response rate was still greater than 90 percent. Even when responses are collected, an increase in skipped questions has required the use of more data estimation techniques to produce values for missing responses. Therefore, funding needs to stretch further for nonresponse, in addition to the cost of inflation.
A reduced sample would hamper policymakers’ ability to interpret economic data
The proposed sample cut would likely increase the unemployment rate’s 90 percent confidence interval, meaning that it would take slightly larger swings in data for changes to be considered statistically significant. Headline statistics such as the national unemployment rate would still be reliable, but it would make understanding what happens to key subpopulations less reliable. In fact, the Local Area Unemployment Statistics (LAUS) data, which rely in part on CPS, are already unable to produce unemployment rates for Black, Asian, and Hispanic populations in multiple states due to imprecise estimates, with more states failing precision requirements as these populations are further divided by sex or age.
Additionally, funding may not be the only problem the CPS faces if the House Appropriations Committee’s Commerce, Justice, Science, and Related Agencies bill were to be enacted. Section 621 would prevent the Census Bureau from asking households to participate in any of their surveys more than twice. This would significantly reduce responses and accuracy across all Census Bureau surveys, including the decennial census, the American Community Survey, and the CPS. For example, the CPS typically requires around 2 1/2 to 3 follow-up contacts after the initial request to get a response.
Efforts to restrict funding and limit follow-ups come as BLS and the Census Bureau plan to modernize the CPS by incorporating an internet self-response mode to improve response rates. This could help reach groups less likely to participate in interviews, such as households with young children and those who speak a language other than English at home. The ASEC supplement to the CPS is also modernizing with an internet option, changes to its questionnaire, and efforts to incorporate more administrative data to lessen the burden on survey respondents. Importantly, the $812 million request from Friends of BLS includes dedicated funding to account for CPS modernization.
The Current Population Survey is critical for measuring employment, inflation, and household finances
The CPS serves an important role in measuring how well Americans are doing financially. As a survey responsible for providing some of the most essential month-to-month data on labor force activities, accompanied by supplemental surveys that often dive deeper into personal finances, the CPS produces data that are crucial for developing and tracking a variety of statistics used to measure the financial health of Americans.
It is also used in other federal surveys to help develop important metrics, including consumer expenditures used to calculate inflation trends. These data help guide the decisions of policymakers, including those at the Federal Reserve who are responsible for adjusting monetary policy to promote maximum employment and stable prices. The cost of properly funding BLS is a small price to pay to ensure that decisions driving the trajectory of the economy are based on accurate information. Below are just a handful of the many important statistics to either come from the CPS directly or rely on CPS data:
- Monthly unemployment rates are used by the Federal Reserve to help inform where the federal funds interest rate should be set.
- The LAUS program uses CPS data, among other sources, to calculate unemployment rates for states and localities.
- Labor force participation rates and employment-population ratios are important indicators for measuring the strength of the labor market along with the unemployment rate.
- The Consumer Expenditures Survey studies how consumers spend their money and uses CPS population estimates to account for undercoverage in its sampling frame. Results from the survey are used in the calculation of the Consumer Price Index, which tracks inflation.
- National poverty rates (ASEC supplement) for both the official and supplemental poverty measure, as well as the number of people pulled out of poverty by various programs, are used to measure the success of government assistance at reducing financial hardship among low-income Americans.
- Income data (ASEC supplement) are used to calculate wage gaps by gender as well as race and ethnicity.
- Health insurance coverage rates (ASEC supplement) are used to track trends in health insurance access, along with how many people were on a private or public plan and how many were uninsured.
- Union membership rates are essential for tracking the prevalence of unions in the workforce as well as the impact that union membership has on wages.
- Food security (food security supplement) data are used by the U.S. Department of Agriculture to produce reports on how many households suffer from lacking access to adequate food.
- American Time Use Survey, which measures how people spend their time, samples households that have already completed the CPS and even carries over some demographic information. Researchers use this survey to gain insight into various health and safety indicators, work-life balance, and to measure the value of unpaid work.
The CPS is responsible for producing a wide variety of economic data
Research organizations, including CAP, use the CPS to measure economic well-being in a range of policy areas. Policymakers use the same measures, from price indices to the gender wage gap to poverty data, to inform the legislation they create to serve everyday Americans. Some examples of the variety of measures that would be affected are expanded upon in the following:
Conclusion
In order for policymakers to meet the needs of the American people and ensure the U.S. economy works for everyone, researchers, policymakers, and advocates must have a clear picture of the reality they face to identify what policies work or where improvements need to be made. This cannot be done without reliable data, which is why efforts to modernize the CPS are so important and why Congress must at least appropriate enough funds for the BLS to carry out this important work without limiting the ability to follow up with households in order to collect more responses.
*Authors’ note: The $812 million request includes an anticipated $68 million that BLS could transfer from the Unemployment Trust Fund, which remains at the same level BLS was allowed to transfer in FY 2024.