States are innovating on labor market data. The feds could, too

Over the past few years, debates over the state of the U.S. labor market have exposed a mismatch between the questions policymakers are asking and the available statistics to answer those questions. For example, existing federal surveys have helped assess the impact of economic shocks and evaluate how much job churn could be driven by AI. At the same time, these national data sets are not always able to track key trends at the occupational or worker level that would improve our understanding of how work is changing over time.

State employment records are already central to federal labor market statistics. The Bureau of Labor Statistics relies on state unemployment insurance (UI) wage records to produce the Quarterly Census of Employment and Wages (QCEW) and to benchmark the Current Employment Statistics (CES) survey; state Labor Market Information (LMI) partners help produce Local Area Unemployment Statistics (LAUS) estimates; and the Census Bureau’s Longitudinal Employer-Household Dynamics (LEHD) program uses state wage records to track worker and firm dynamics over time.

As federal lawmakers consider new approaches to measuring labor market outcomes, they could learn valuable lessons from states that have taken steps to enrich these records further. In this post, we summarize findings from two recent studies—a Niskanen Center report and an Urban Institute analysis—on what those state innovations can teach us.

Lessons from states using enhanced wage records

A growing number of states across the country have begun collecting more detailed employment data from employers as part of the unemployment insurance wage records already submitted each quarter. In general, employers in every state must supply records that include the name of each covered employee, their Social Security number, and the amount of wages earned in the calendar quarter. Workforce agencies use these administrative records to assess eligibility for unemployment benefits and analyze the state of the labor market.

State wage records also serve as a key input for federal data sets such as the QCEW and the CES. However, the quarterly UI wage records have typically lacked basic information on job title, hours worked, and job location.

To see firsthand how “enhanced” wage records (EWRs) have helped improve labor data overall, a recent Niskanen Center study engaged workforce and unemployment agency officials in ten states comprehensively collecting one or more of those additional data categories. Their experiences offer insight into how better wage data can improve labor market analysis and program implementation.

State officials emphasized different benefits depending on the type of data requested from employers. By gathering occupational data, Alaska and South Carolina have been able to provide more tailored recommendations to job-seekers by assessing which industries have growing or shrinking job supply gaps and the wage premiums received from educational programs. Washington state officials discussed how documenting the total number of hours worked each quarter allowed the state to more accurately gauge pay rates and job contractions or expansions. Meanwhile, officials in states such as Indiana argued that job location data helped identify which state regions were experiencing job growth or contractions and determine where to target investments. Agency staff in other states echoed these reflections.

State officials were equally candid about the challenges of implementing enhanced wage records. Mandatory submissions were not necessarily possible if EWRs were to be collected with administrative funding from the federal government, and sufficient outreach to employers was critical to make the collections worthwhile. The challenges did not end once state agencies were able to add the new requested fields on the wage record forms. If anything, the biggest hurdles were ensuring that employers understood the value of EWRs, responded at high rates, and accurately submitted the information. State officials mentioned “top-coding” as a problem when adding an occupational field, where employers would list every employee under the same job code. Meanwhile, agencies collecting job location details found that employers might incorrectly list all employees as working at business headquarters rather than at various branch locations. Correcting and preventing these errors required time and money—resources that unemployment agencies often lack.

Many officials noted that these implementation challenges were compounded by how Congress finances unemployment insurance administration. Over the last two decades, the base administrative allocations that UI agencies plan their fiscal years around have eroded by over 30% when adjusting for inflation. The congressional allocations provided to states year-to-year are also quite variable, making it difficult for more states to commit to a long-term funding stream for EWRs. Funding available in a given year could be wiped away the next.

A successful federal effort to improve administrative labor market data could require fixing how state agency work is financed while simultaneously establishing national EWR standards. Individual states have had successful EWR rollouts, but the policy details have varied in each case. There’s no guarantee that states’ data fields and formats are compatible. An increasingly complicated patchwork of state wage reporting requirements could place burdens on multi-state employers and third-party administrators (TPAs) and could hamper data quality. Federal-level standardization could help alleviate these issues.

Learning more about job quality

As another illustration of how EWRs could fill important gaps in current employment statistics, consider the issue of job quality: A central goal for federal employment data and statistics is not just counting the jobs that Americans hold but characterizing them. The unemployment rate, monthly payroll numbers, and other key labor force indicators provide essential information on the health of the economy, employment, and business trends. But policymakers, researchers, and workers themselves also need to know about the quality of jobs—what they pay, what benefits they provide, their schedules, working conditions, and other key features.

Current employment statistics and data, primarily from survey sources such as the Current Employment Statistics (CES), Current Population Survey (CPS), and National Compensation Survey (NCS), do provide key information on wages, hours, and some benefits. But policymakers have increasingly shown an interest in extending the breadth and depth of information collected on the quality of jobs, and researchers have investigated the dimensions along which additional survey data and measurement would be valuable.

One promising direction is the role that state-level administrative data sources, such as unemployment insurance wage records, might play in improving our understanding of job quality. Administrative data increasingly play an important role in employment statistics, and the UI wage records collected by states are an especially rich, accurate, and timely source: They capture quarterly employment and earnings information on the vast majority of the workforce, in every state (including Washington, D.C.).

While wage record data already provide important information on employment, the level of detail they have historically provided on job quality remains relatively narrow. The primary piece of information that all states consistently collect is quarterly earnings, along with some additional characteristics of the job such as the industry of employment.

Here, however, is where the growing interest in richer data for characterizing job quality meets the similarly growing interest—among state officials, federal policymakers, statistical agencies, labor market researchers, and employers—in enhancing UI wage records to collect additional information from employers about the terms of work and the jobs filled by their workers.

With wage data expanded to include additional fields, the potential for improving our understanding of job quality and the experiences of workers in modern labor markets is substantial. Wage records with information on worker hours—as some states already collect—would allow these data to better characterize pay rates. Beyond wages, any additional information from employers on other components of compensation such as employer-sponsored health insurance, workplace retirement benefits, or paid leave would substantially enrich our understanding of job quality.

Including information on occupation—as another small set of states already does—would also be especially valuable. This is in part because it would allow information from these data to be analyzed at the occupational level, which is important in itself, and would enable analysis in conjunction with occupational information from other labor market data sources. It would also allow analysts to take advantage of another key feature of these data—that by following workers over time they allow for observing labor market dynamics—to study how workers move between occupations, how those moves are shaped by economic forces, and the implications for job quality.

What more can be done?

Opportunities for—and in some instances, progress toward—enhancement of wage data to meet these needs are promising in part because of the capacity the UI wage record system allows for state-led innovation. At the same time, enhanced data will ultimately best meet these needs if it is coordinated across states and can be analyzed nationally, which is where a federal role becomes essential. Innovation in this space can leverage the comparative advantages of both the federal government and its state partners. Continued innovation could build on efforts like the JedX initiative, currently working with states, and the BLS wage records program that compiles state data.

As policymakers grapple with the labor market implications of AI, economic volatility, and structural change, the need for more detailed and timely labor market data will only grow. Enhanced UI wage records offer a promising path forward, but state experiences show that success depends on stable federal administrative funding, common guidelines, and time for iterative implementation. Without those elements, efforts to modernize national labor market data risk adding complexity without delivering the insights policymakers and researchers are seeking.

Federal and state policymakers should act now to capitalize on the momentum already building in states. Congress can start by stabilizing UI administrative funding, while the Department of Labor could develop common EWR standards. State legislators and workforce officials, for their part, can expand enhanced wage record collection and share what they learn with peer states. The tools to build a richer, more accurate picture of the American labor market already exist—what’s needed is the political will to put them to work.