Search Results
Working Paper
Total Recall? Evaluating the Macroeconomic Knowledge of Large Language Models
We evaluate the ability of large language models (LLMs) to estimate historical macroeconomic variables and data release dates. We find that LLMs have precise knowledge of some recent statistics, but performance degrades as we go farther back in history. We highlight two particularly important kinds of recall errors: mixing together first print data with subsequent revisions (i.e., smoothing across vintages) and mixing data for past and future reference periods (i.e., smoothing within vintages). We also find that LLMs can often recall individual data release dates accurately, but aggregating ...
Working Paper
LLM on a Budget: Active Knowledge Distillation for Efficient Classification of Large Text Corpora
Large Language Models (LLMs) are highly accurate in classification tasks, however, substantial computational and financial costs hinder their large-scale deployment in dynamic environments. Knowledge Distillation (KD) where a LLM ""teacher"" trains a smaller and more efficient ""student"" model, offers a promising solution to this problem. However, the distillation process itself often remains costly for large datasets, since it requires the teacher to label a vast number of samples while incurring significant token consumption. To alleviate this challenge, in this work we explore the ...
Working Paper
AI and Coder Employment: Compiling the Evidence
We evaluate whether LLMs have had any discernible impact on the aggregate labor market so far. We focus on occupations that are computer programming-intensive, motivated by data showing that coding is one of the most LLM-exposed tasks. Linking O*NET to CPS we find that aggregate employment of coders has decelerated sharply since the introduction of ChatGPT. Using a novel control variable for industry-level shocks we show that the deceleration is not attributable to the exposure of coders to slowing industries, suggesting instead that coders experienced an occupation-specific shock around ...
Working Paper
Business Dynamics in the National Establishment Time Series (NETS)/Leland Crane, Ryan Decker
Business microdata have proven useful in a number of fields, but the main sources of comprehensive microdata are subject to significant confidentiality restrictions. A growing number of papers instead use a private data source seeking to cover the universe of U.S. business establishments, the National Establishment Time Series (NETS). Previous research documents the representativeness of NETS in terms of the distribution of employment and establishment counts across industry, geography, and establishment size. But there exists considerable need among researchers for microdata suitable for ...
Working Paper
Using Payroll Processor Microdata to Measure Aggregate Labor Market Activity
We show that high-frequency private payroll microdata can help forecast labor market conditions. Payroll employment is perhaps the most reliable real-time indicator of the business cycle and is therefore closely followed by policymakers, academia, and financial markets. Government statistical agencies have long served as the primary suppliers of information on the labor market and will continue to do so for the foreseeable future. That said, sources of ?big data? are becoming increasingly available through collaborations with private businesses engaged in commercial activities that record ...
Working Paper
Improving the Accuracy of Economic Measurement with Multiple Data Sources: The Case of Payroll Employment Data
This paper combines information from two sources of U.S. private payroll employment to increase the accuracy of real-time measurement of the labor market. The sources are the Current Employment Statistics (CES) from BLS and microdata from the payroll processing firm ADP. We briefly describe the ADP-derived data series, compare it to the BLS data, and describe an exercise that benchmarks the data series to an employment census. The CES and the ADP employment data are each derived from roughly equal-sized samples. We argue that combining CES and ADP data series reduces the measurement error ...
Working Paper
An Assessment of the National Establishment Time Series (NETS) Database
The National Establishment Time Series (NETS) is a private sector source of U.S. business microdata. Researchers have used state-specific NETS extracts for many years, but relatively little is known about the accuracy and representativeness of the nationwide NETS sample. We explore the properties of NETS as compared to official U.S. data on business activity: The Census Bureau's County Business Patterns (CBP) and Nonemployer Statistics (NES) and the Bureau of Labor Statistics' Quarterly Census of Employment and Wages (QCEW). We find that the NETS universe does not cover the entirety of the ...
Working Paper
Dynamic Beveridge Curve Accounting
We develop a dynamic decomposition of the empirical Beveridge curve, i.e., the level of vacancies conditional on unemployment. Using a standard model, we show that three factors can shift the Beveridge curve: reduced-form matching efficiency, changes in the job separation rate, and out-of-steady-state dynamics. We find that the shift in the Beveridge curve during and after the Great Recession was due to all three factors, and each factor taken separately had a large effect. Comparing the pre-2010 period to the post-2010 period, a fall in matching efficiency and out-of-steady-state dynamics ...
Working Paper
Tracking Labor Market Developments during the COVID-19 Pandemic: A Preliminary Assessment
Many traditional official statistics are not suitable for measuring high-frequency developments that evolve over the course of weeks, not months. In this paper, we track the labor market effects of the COVID-19 pandemic with weekly payroll employment series based on microdata from ADP. These data are available essentially in real-time, and allow us to track both aggregate and industry effects. Cumulative losses in paid employment through April 4 are currently estimated at 18 million; just during the two weeks between March 14 and March 28 the U.S. economy lost about 13 million paid jobs. ...
Working Paper
Business Exit During the COVID-19 Pandemic: Non-Traditional Measures in Historical Context
Lags in official data releases have forced economists and policymakers to leverage "alternative" or "non-traditional" data to measure business exit resulting from the COVID- 19 pandemic. We first review official data on business exit in recent decades to place the alternative measures of exit within historical context. For the U.S., business exit is countercyclical and fairly common, with about 7.5 percent of firms exiting annually in recent years. Both the high level and the cyclicality of exit are driven by very small firms and establishments. We then explore a range of alternative measures ...