Do Medicaid Expansions Affect Where Doctors Choose to Practice?

This study asks whether Medicaid eligibility expansions for pregnant women in the 1980s and 1990s influenced physicians’ practice location decisions, particularly the supply of OB/GYNs. Using county-level data from the American Medical Association Physician Masterfile combined with Medicaid eligibility measures from 1982 to 1992, the authors employ an event-study framework. They find that OB/GYN counts per 100,000 women rose significantly in expansion states, with an average post-expansion increase of 6.6 percent and annual effects ranging from 1.9 to 10.1 percent. The growth was concentrated among early-career physicians and in densely populated or poorer counties

Full Citation and Link to Article

Huh, J., & Lin, J. (2025). More doctors in town now? Evidence from Medicaid expansions. Journal of Policy Analysis and Management, 44(3), 869–892. https://doi.org/10.1002/pam.22611

Expanded Summary

Central Research Question

The article by Jason Huh and Jianjing Lin investigates how Medicaid eligibility expansions for pregnant women in the 1980s and 1990s influenced the supply and geographic distribution of obstetricians and gynecologists (OB/GYNs). The central research question is whether Medicaid expansions affected physicians’ practice location choices and increased the number of OB/GYNs at the county level. Specifically, the authors ask: do large-scale public health insurance expansions not only affect care on the intensive margin (e.g., participation rates and time spent per patient) but also on the extensive margin by changing the total number of physicians in practice? The paper focuses on this relatively understudied dimension of physician labor supply.

Previous Literature

A large body of research has examined the effects of Medicaid expansions and public health insurance reforms, but most of the focus has been on demand-side outcomes such as access to care, utilization, and health outcomes. For example, studies by Currie and Gruber (1996) showed that expansions improved prenatal care and birth outcomes, while Buchmueller et al. (2015) summarized generally positive but mixed effects of Medicaid across many dimensions. On the supply side, research has often focused on the intensive margin—how existing physicians adjust their participation in Medicaid or alter the amount of care provided. Baker and Royalty (2000), for instance, found that expansions increased access mainly from public physicians in hospitals and clinics. Garthwaite (2012) studied the State Children’s Health Insurance Program and found changes in pediatrician participation but reductions in time per patient. Research on dentists (Buchmueller et al., 2016; Huh, 2021) found expansions increased supply and participation in Medicaid, often facilitated by practice-level adjustments.

What remains relatively understudied is the extensive margin of physician supply—the number and distribution of practicing doctors. Previous literature on physicians’ location choices highlights the importance of financial incentives, such as student loan forgiveness or subsidies for practice in shortage areas (Falcettoni, 2018; Zhou, 2017). Yet, whether Medicaid expansions themselves serve as an incentive for physicians to move to or remain in particular areas has not been systematically tested. The present study fills this gap by examining whether Medicaid expansions targeting pregnant women altered OB/GYN supply across counties, providing the first evidence on this extensive margin for this type of policy change.

Data

The primary data source is the American Medical Association (AMA) Physician Masterfile, which provides county-level counts of OB/GYNs for the years 1982 to 1992 (excluding 1984 and 1990 due to missing data). These counts include OB/GYNs in private practice, hospitals, and clinics, and they also contain information on year of graduation, which allows categorization of physicians into early-career (≤10 years in practice), mid-career (10–30 years), and late-career (>30 years).

Medicaid eligibility data are taken from East et al. (2023), based on the Current Population Survey (CPS) and state-specific eligibility rules. These data provide both actual and simulated eligibility measures for women aged 15–44. The simulated measure is crucial, as it helps separate policy-driven expansions from endogenous state-level changes. Eligibility rates rose nationally from about 12 percent in 1982 to 34 percent in 1992.

The authors supplement these data with demographic and socioeconomic measures from the Area Health Resources Files (AHRF) and the SEER Program, including per capita income, unemployment rate, female population shares, racial composition, and population density. Poverty rates at the county level are drawn from the 1980 and 1990 Census. Medicaid-to-private fee ratios for obstetric care (Currie et al., 1995) are also included as controls.

The unit of analysis is the county-year, with the primary outcome being OB/GYNs per 100,000 women aged 15–44.

Methods

The study uses an event-study framework to identify the effect of Medicaid expansions on OB/GYN supply. States are divided into treatment and control groups based on the simulated eligibility measure. Treatment states are those that experienced sharp increases in Medicaid eligibility between 1983 and 1989, while control states had smoother, more gradual increases.

The event-study specification compares changes in OB/GYN supply across these groups relative to the year before a state’s expansion. This design allows the authors to test for parallel pre-trends and to trace dynamic effects over time. The regressions include county fixed effects, year fixed effects, state-specific time trends, and controls for county characteristics and Medicaid fee ratios. Standard errors are clustered at the state level.

Heterogeneity analyses are conducted to examine differential effects by population density, poverty levels, and existing supply of OB/GYNs, as well as by physician career stage. Placebo tests are also performed on other specialties—dermatology, plastic surgery, emergency medicine, and cardiology—to confirm that the observed effects are specific to OB/GYNs.

The approach follows and extends the methodology in East et al. (2023) and builds on earlier work by Currie and Gruber (1996).

Findings/Size Effects

The main finding is that Medicaid eligibility expansions for pregnant women significantly increased the supply of OB/GYNs in expansion states relative to control states. The increase was modest but persistent, with effects strengthening over time:

• OB/GYNs per 100,000 women increased by an average of 6.6 percent across post-expansion years.

• Annual effects ranged from 1.9 percent to 10.1 percent, depending on the number of years since expansion.

• There was no evidence of pre-trends, strengthening the causal interpretation.

Heterogeneity analyses reveal that these effects were not uniform:

1. Population density: Significant increases occurred only in densely populated counties. Urban areas saw average growth of 5.3 percent post-expansion, while low- and medium-density counties showed no statistically significant changes.

2. Poverty: Moderate- and high-poverty counties experienced increases in OB/GYN supply (2.0–17.6 percent depending on year), while low-poverty counties did not. The largest effects appeared in moderately poor counties, suggesting that extremely high poverty may still deter physicians despite Medicaid demand.

3. Existing supply: Counties with moderate pre-existing OB/GYN supply experienced the largest gains. Counties with very low or very high supply saw little change, possibly due to barriers to establishing practices or excessive competition.

4. Career stage: Early-career physicians drove the observed increases. More established physicians were less likely to move or relocate in response to Medicaid expansions.

Placebo tests confirmed that other specialties unrelated to pregnancy care showed no significant supply response, strengthening the claim that the expansions specifically induced OB/GYN growth.

These findings suggest that Medicaid expansions not only increased demand for prenatal care but also altered the supply side of the market, drawing new or relocating OB/GYNs into areas where coverage increased most.

Conclusion

The study provides the first direct evidence that Medicaid expansions targeting pregnant women during the 1980s and 1990s affected the extensive margin of physician supply. By increasing eligibility, states induced measurable increases in OB/GYN counts, particularly among early-career physicians and in urban or poor counties. The magnitude of the effects—around a 6.6 percent average increase in OB/GYN supply—shows that large-scale insurance expansions can influence physicians’ location decisions.

The results have several policy implications. First, public health insurance programs do not only shape patient demand but also alter the supply and distribution of health professionals. Medicaid expansions can attract new physicians to areas of greater need, suggesting that coverage rules themselves function as indirect incentives for location choice. Second, the concentration of effects in urban and moderately poor areas underscores persistent challenges in addressing shortages in rural and extremely poor counties. Policymakers may need to combine Medicaid expansions with more direct financial incentives or infrastructure investments to encourage broader geographic distribution. Third, the fact that early-career physicians were most responsive suggests that interventions targeting new graduates may be especially effective.

Overall, the study highlights the importance of considering both supply and demand effects when evaluating health insurance expansions. By showing that Medicaid expansions increased OB/GYN supply in targeted areas, it expands our understanding of how public insurance policies can reshape the health care labor market and, in the long run, improve access to care for vulnerable populations.