EconBase
← All papers

Estimating Wage Disparities Using Foundation Models

Keyon Vafa, Susan Athey, David M. Blei

arXiv 15 Sep 2024 · Machine Learning · publishedProceedings of the National Academy of Sciences (2025) · 3 citations (OpenAlex)

arXiv:2409.09894 · PDF · DOI · OpenAlex · Extracted main text

Abstract

The rise of foundation models marks a paradigm shift in machine learning: instead of training specialized models from scratch, foundation models are first trained on massive datasets before being adapted or fine-tuned to make predictions on smaller datasets. Initially developed for text, foundation models have also excelled at making predictions about social science data. However, while many estimation problems in the social sciences use prediction as an intermediate step, they ultimately require different criteria for success. In this paper, we develop methods for fine-tuning foundation models to perform these estimation problems. We first characterize an omitted variable bias that can arise when a foundation model is only fine-tuned to maximize predictive accuracy. We then provide a novel set of conditions for fine-tuning under which estimates derived from a foundation model are root-n-consistent. Based on this theory, we develop new fine-tuning algorithms that empirically mitigate this omitted variable bias. To demonstrate our ideas, we study gender wage decomposition. This is a statistical estimation problem from econometrics where the goal is to decompose the gender wage gap into components that can and cannot be explained by career histories of workers. Classical methods for decomposing the wage gap employ simple predictive models of wages which condition on coarse summaries of career history that may omit factors that are important for explaining the gap. Instead, we use a custom-built foundation model to decompose the gender wage gap, which captures a richer representation of career history. Using data from the Panel Study of Income Dynamics, we find that career history explains more of the gender wage gap than standard econometric models can measure, and we identify elements of career history that are omitted by standard models but are important for explaining the wage gap.

Citation extraction

68
references
156
in-text mentions
68
distinct cited
8
self-citations
9,883
main-text words

appendix boundary found by appendix_command · 55% of the source is main text. Read the extracted text to check this.

Most heavily cited references

The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.

ReferenceIntensityMentionsSectionsMain text
1Devlin, J., Chang, M.-W., Lee, K., and Toutanova, K (2018) BERT: Pre-training of deep bidirectional transformers for language understanding1.00084100%
2Vafa, K., Palikot, E., Du, T., Kanodia, A., Athey, S., and Blei, D. M (2023) CAREER: A foundation model for labor sequence data self0.8947471%
3Chernozhukov, V., Chetverikov, D., Demirer, M., Duflo, E., Hansen, C… (2018) Double/debiased machine learning for treatment and structural parameters0.8746467%
4Chernozhukov, V., Cinelli, C., Newey, W., Sharma, A., and Syrgkanis, V (2022) Long story short: Omitted variable bias in causal machine learning0.8115280%
5Oaxaca, R (1973) Male-female wage differentials in urban labor markets0.81142100%
6Blau, F. D. and Kahn, L. M (2017) The gender wage gap: Extent, trends, and explanations0.78429748%
7Panel Study of Income Dynamics (2023) Public use dataset, produced and distributed by the Survey Research Center, Institute for Social Research, University of Michiga…0.7373367%
8Blau, F. D. and Kahn, L. M (2013) The feasibility and importance of adding measures of actual experience to cross-sectional data collection0.73732100%
9Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von… (2021) On the opportunities and risks of foundation models0.73732100%
10Kitagawa, E. M (1955) Components of a difference between two rates0.73732100%

Showing the top 10 of 68 scored citations.

Cited by, within the corpus

arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.

Citing paperIntensityMentionsSections
1LABOR-LLM: Language-Based Occupational Representations with Large Language Models0.58531
2Welfare Analysis in Dynamic Models0.51121
3Model-Agnostic Covariate-Assisted Inference on Partially Identified Causal Effects0.40511
4Program Evaluation with Remotely Sensed Outcomes0.40511
5Decoding Consumer Preferences Using Attention-Based Language Models0.40511
6Econometrics with Pre-Trained Embeddings for Unstructured Data0.40511