-
AI Exposure and Adoption Among U.S. Firms
September 2026
Authors:
John Haltiwanger,
Lucia Foster,
Martha Stinson,
Emin Dinlersoz,
Cheryl Grim,
Zoltan Wolf,
Sabrina Wulff Pabilonia,
Matthew Dey,
Sean Wang,
Aditya Pande,
Peter B. Meyer
Working Paper Number:
CES-26-61
Measures of exposure to artificial intelligence (AI) are widely used to study where AI is likely to affect firm and worker outcomes, yet limited evidence exists on how closely exposure relates to realized AI adoption at the firm level. We examine this relationship by linking 14 firm-level exposure measures, constructed from occupational exposure estimates in the literature and occupational employment shares from the Bureau of Labor Statistics Occupational Employment and Wage Statistics program, to direct measures of firm AI adoption from the Census Bureau's Business Trends and Outlook Survey. Exposure is positively and statistically significantly associated with adoption, but explains only a modest share of its variation. The strength of the relationship varies considerably across standardized exposure measures: a one-standard-deviation increase in firm-level exposure is associated with a 4'11 percentage point higher firm adoption probability, falling to 1'8 percentage points after controlling for year and sub-sector fixed effects. The exposure'adoption relationship is also heterogeneous across firm-size classes and sectors. Exposure is thus an informative but incomplete signal of adoption. More broadly, the results have implications for the interpretation of exposure-based measures in studies of the economic effects of AI.
View Full
Paper PDF
-
You're (not) Hired: Artificial Intelligence and Early Career Hiring in the Quarterly Workforce Indicators
April 2026
Working Paper Number:
CES-26-27
Using detailed tabulations from matched employer-employee administrative data, I document evidence of an immediate, sizable, and persistent decrease in the level of early career (22-24 year old) hires following introduction of ChatGPT within the industry-state cells that are most exposed to AI. The decline in hires is the primary cause of large observed declines in employment over the subsequent period. Regressionadjusted employment of early career workers in the most AI-exposed quintile of industry-state cells declined by 12% over the 10 quarters following the introduction of ChatGPT, even as employment in lessexposed industries has remained stable. The rate of hiring largely recovered by early 2025, attributable to a smaller employment base. Earnings growth of early career workers in the most exposed industries slowed slightly relative to those in less exposed industries. Although the most AI-exposed quintile of detailed industries is dominated by a handful of industry sectors, I find that the association of higher AI exposure with reduced early career employment and fewer hires is observed across most sectors of the economy. Timing of effects in event studies is consistent with an immediate effect on hiring following introduction of ChatGPT. However, triple difference estimates provide some evidence of earlier trend shifts on employment, hiring, and separations around the onset of the COVID pandemic. I discuss potential explanations, including the increase in remote work and increased educational attainment among workers in AI-exposed occupations. Nonetheless, job gains to early career workers and backfill hires show evidence of discontinuous decline at the time of ChatGPT's release in comparison to older workers in the same industries. A local projections analysis at the NAICS industry group level shows that industries with high AI exposure are not particularly sensitive to unexpected fluctuations in monetary policy on average relative to other industries in employment, hiring, or separations. A historical decomposition suggests that up to one quarter of relative early career employment declines through 2025q2 may be attributable to monetary policy shocks through 2023, but the analysis does not find evidence that these shocks can explain the rapid decline in hires at the most AI-exposed firms in comparison to others.
View Full
Paper PDF
-
The Microstructure of AI Diffusion: Evidence From Firms, Business Functions, and Worker Tasks
April 2026
Working Paper Number:
CES-26-25
Using novel, nationally representative data from the 2026 AI supplement to the U.S. Census Bureau's Business Trends and Outlook Survey (BTOS), we characterize AI diffusion across three interconnected layers: overall firm use, deployment across business functions, and worker-task use. This multi-layered approach provides a nuanced picture of business AI adoption. During the supplement reference period (Nov 2025-Jan 2026), 18% of firms used AI in a business function, rising to 32% on an employment-weighted basis; adoption is expected to reach 22% within six months. AI use is substantially higher in large firms and knowledge-intensive sectors, with use rates reaching 50%-60% (60%-70%, employment-weighted) for very large firms in the Information, Professional Services, and Finance sectors. Among adopting firms, the scope of use remains limited: 57% of users integrate AI in three or fewer business functions, most commonly Sales and Marketing (52%), Strategy and Business Development (45%), and IT (41%). In 23% (41%, employment-weighted) of firms, workers use AI in work-related tasks. Writing, document analysis, and information search are the leading Generative AI use in tasks, though 65% of firms limit use to three or fewer tasks. The evidence points to both top-down and bottom-up diffusion channels: worker task use sometimes occurs without formal firm-level adoption, and firm-level adoption sometimes occurs without worker task use. Most users (66%) rely on AI solely to augment tasks, while AI-related employment decreases are rare, occurring in only 2% of firms. Regression analysis shows a robust positive correlation between firm commercial performance and the breadth of AI integration, including functional deployment, task-level use, and operational investment. A distinct divergence emerges, however, with respect to labor outcomes. Functional breadth and operational investment are positively associated with employment decreases, whereas worker-task integration shows no significant link to headcount reduction once functional integration and operational investment are taken into account.
View Full
Paper PDF
-
'Class of Customer' Question from the US Economic Census
September 2025
Working Paper Number:
CES-25-66
The Economic Census (EC) collects detailed information on the class of customers served by establishments'for example, the share of an establishment's sales to other businesses or to government entities'for a subset of sectors in the economy. In this paper, we evaluate the data from the 'Class of Customer' question from the EC, with a particular focus on sales to the government. These data have seldom been used in empirical research and are unique in that they enable researchers to link establishment-level Census data with information on government procurement.
We compile and analyze large volumes of publicly available tabulated data about the class of customer question over time. Using these data, we document three main findings. First, total sales to government from establishments covered by the class of customer question account for approximately 4 percent of GDP'just under half of total government procurement as measured in the national accounts. Second, the sectoral distribution of government expenditure is significantly different from that of private sector spending. Certain industries, such as Construction and Professional, Scientific, and Technical Services, account for a much larger share of government expenditure relative to private sector expenditure. Third, sales to the government make up a substantial portion of total sales in several sectors'for instance, 70 percent in Facilities Support Services, 30 percent in Waste Treatment and Disposal, and 17 percent in Construction. Finally, we use the microdata to examine nonresponse rates to the class of customer question across establishments based on the number of employees.
View Full
Paper PDF
-
Startup Dynamics: Transitioning from Nonemployer Firms to Employer Firms, Survival, and Job Creation
April 2025
Working Paper Number:
CES-25-26
Understanding the dynamics of startup businesses' growth, exit, and survival is crucial for fostering entrepreneurship. Among the nearly 30 million registered businesses in the United States, fewer than six million have employees beyond the business owners. This research addresses the gap in understanding which companies transition to employer businesses and the mechanisms behind this process. Job creation remains a critical concern for policymakers, researchers, and advocacy groups. This study aims to illuminate the transition from non-employer businesses to employer businesses and explore job creation by new startups. Leveraging newly available microdata from the U.S. Census Bureau, we seek to gain deeper insights into firm survival, job creation by startups, and the transition from non-employer to employer status.
View Full
Paper PDF
-
Size Matters: Matching Externalities and the Advantages of Large Labor Markets
April 2025
Working Paper Number:
CES-25-22
Economists have long hypothesized that large and thick labor markets facilitate the matching between workers and firms. We use administrative data from the LEHD to compare the job search outcomes of workers originally in large and small markets who lost their jobs due to a firm closure. We define a labor market as the Commuting Zone'industry pair in the quarter before the closure. To account for the possible sorting of high-quality workers into larger markets, the effect of market size is identified by comparing workers in large and small markets within the same CZ, conditional on workers fixed effects. In the six quarters before their firm's closure, workers in small and large markets have a similar probability of employment and quarterly earnings. Following the closure, workers in larger markets experience significantly shorter non-employment spells and smaller earning losses than workers in smaller markets, indicating that larger markets partially insure workers against idiosyncratic employment shocks. A 1 percent increase in market size results in a 0.015 and 0.023 percentage points increase in the 1-year re-employment probability of high school and college graduates, respectively. Displaced workers in larger markets also experience a significantly lower need for relocation to a different CZ. Conditional on finding a new job, the quality of the new worker-firm match is higher in larger markets, as proxied by a higher probability that the new match lasts more than one year; the new industry is the same as the old one; and the new industry is a 'good fit' for the worker's college major. Consistent with the notion that market size should be particularly consequential for more specialized workers, we find that the effects are larger in industries where human capital is more specialized and less portable. Our findings may help explain the geographical agglomeration of industries'especially those that make intensive use of highly specialized workers'and validate one of the mechanisms that urban economists have proposed for the existence of agglomeration economies.
View Full
Paper PDF
-
The Composition of Firm Workforces from 2006'2022: Findings from the Business Dynamics Statistics of Human Capital Experimental Product
April 2025
Working Paper Number:
CES-25-20
We introduce the Business Dynamics Statistics of Human Capital (BDS-HC) tables, a new Census Bureau experimental product that provides public-use statistics on the workforce composition of firms and its relationship to business dynamics. We use administrative W-2 filings to combine population-level worker demographic data with longitudinal business data to estimate the demographic and educational composition of nearly all non-farm employer businesses in the United States between 2006 and 2022. We use this newly constructed data to document the evolution of employment, entry, and exit of employers based on their workforce compositions. We also provide new statistics on the interaction between firm and worker characteristics, including the composition of workers at startup firms. We find substantial changes between 2006 and 2022 in the distribution of employers along several dimensions, primarily driven by changing workforce compositions within continuing firms rather than the reallocation of employment between firms. We also highlight systematic differences in the business dynamics of firms by their workforce compositions, suggesting that different groups of workers face different economic environments due to their employers.
View Full
Paper PDF
-
Measuring the Characteristics and Employment Dynamics of U.S. Inventors
September 2022
Working Paper Number:
CES-22-43
Innovation is a key driver of long run economic growth. Studying innovation requires a clear view of the characteristics and behavior of the individuals that create new ideas. A general lack of rich, large-scale data has constrained such analyses. We address this by introducing a new dataset linking patent inventors to survey, census, and administrative microdata at the U.S. Census Bureau. We use this data to provide a first look at the demographic characteristics, employer characteristics, earnings, and employment dynamics of inventors. These linkages, which will be available to researchers with approved access, dramatically increases the scope of what can be learned about inventors and innovative activity.
View Full
Paper PDF
-
Diversity and Labor Market Outcomes in the Economics Profession
July 2022
Working Paper Number:
CES-22-26
While the lack of gender and racial diversity in economics in academia (for students and professors) is well-established, less is known about the overall placement and earnings of economists by gender and race. Understanding demand-side factors is important, as improvements in the supply side by diversifying the pipeline alone may not be enough to improve equity in the profession. Using the Survey of Earned Doctorates (SED) linked to Longitudinal Employer-Household Dynamics (LEHD) jobs data, we examine placements and earnings for economists working in the U.S. after receiving a PhD by gender and race. We find enormous dispersion in pay for economists within and across sectors that grows over time. Female PhD economists earn about 12 percent less than their male colleagues on average; Black PhD economists earn about 15 percent less than their white counterparts on average; and overall underrepresented minority PhD economists earn about 8 percent less than their white counterparts. These pay disparities are attenuated in some sectors and when controlling for rank of PhD granting institution and employer.
View Full
Paper PDF
-
High Frequency Business Dynamics in the United States During the COVID-19 Pandemic
March 2021
Working Paper Number:
CES-21-06
Existing small businesses experienced very sharp declines in activity, business sentiment, and expectations early in the pandemic. While there has been some recovery since the early days of the pandemic, small businesses continued to exhibit indicators of negative growth, business sentiment, and expectations through the first week of January 2021. These findings are from a unique high frequency, real time survey of small employer businesses, the Census Bureau's Small Business Pulse Survey (SBPS). Findings from the SBPS show substantial variation across sectors in the outcomes for small businesses. Small businesses in Accommodation and Food Services have been hit especially hard relative to those Finance and Insurance. However, even in Finance and Insurance small businesses exhibit indicators of negative growth, business sentiment, and expectations for all weeks from late April 2020 through the first week of 2021. While existing small businesses have fared poorly, after an initial decline, there has been a surge in new business applications based on the high frequency, real time Business Formation Statistics (BFS). Most of these applications are for likely nonemployers that are out of scope for the SBPS. However, there has also been a surge in new applications for likely employers. The surge in applications has been especially apparent in Retail Trade (and especially Non-store Retailers). We compare and contrast the patterns from these two new high frequency data products that provide novel insights into the distinct patterns of dynamics for existing small businesses relative to new business formations.
View Full
Paper PDF
