Jim Marton's Papers - updated 1/23/19

Publications (* denotes graduate student coauthor):


32. The Three Year Impact of the Affordable Care Act on Disparities in Insurance Coverage - Health Services Research, 54 (S1), January 2019, pp. 307-316 (with Chuck Courtemanche, Ben Ukert, Aaron Yelowitz, Daniela Zapata, and Ishtiaque Fazlul*)
Abstract: In this paper we estimate the impact of the major components of the ACA (Medicaid expansion, subsidized Marketplace plans, and insurance market reforms) on disparities in insurance coverage after three years. We use data from the 2011–2016 waves of the American Community Survey (ACS), with the sample restricted to nonelderly adults, to estimate difference-in-difference-in-differences models to separately identify the effects of the nationwide and Medicaid expansion portions of the ACA using the methodology developed in Courtemanche et al. (2017). The differences come from time, state Medicaid expansion status, and local area pre-ACA uninsured rates. In order to focus on access disparities, we stratify our sample separately by income, race/ethnicity, marital status, age, gender, and geography. After three years, the fully implemented ACA eliminated 43 percent of the coverage gap across income groups, with the Medicaid expansion accounting for this entire reduction. The ACA also reduced coverage disparities across racial groups by 23 percent, across marital status by 46 percent, and across age groups by 36 percent, with these changes being partly attributable to both the Medicaid expansion and nationwide components of the law. Thus the fully implemented ACA has been successful in reducing coverage disparities across multiple groups.
31. Effects of the Affordable Care Act on Health Behaviors after Three Years - Eastern Economic Journal, 45 (1), January 2019, pp. 7-33. (with Chuck Courtemanche, Aaron Yelowitz, Ben Ukert, and Daniela Zapata)
(early version available as NBER working paper 24511)  
Abstract: The purpose of this paper is to estimate the impacts of the ACA on health-related behaviors, which we define as either investments (utilization of preventive services such as well-patient checkups, flu shots, dental visits, clinical breast exams, pap tests, mammograms, and HIV tests) or disinvestments (risky health behaviors such as smoking and drinking) that are purchased in the present but influence health in the future. Our primary innovation is to use data from the 2011-2016 Behavioral Risk Factor Surveillance System (BRFSS), which gives us a longer post-treatment period than prior studies and therefore a greater ability to detect effects. Our preliminary results suggest no statistically significant effects of the private portion of the ACA, Medicaid expansion, or overall ACA on any of our risky health behavior outcomes when aggregating the three post-reform years of 2014-2016 together. The signs of our estimated effects are mixed, with the full ACA reducing BMI but increasing smoking and drinking. These findings with three years of post-reform data are consistent with both the Courtemanche et al. (2017b) analysis of the full ACA and the Simon, Soni, and Cawley (2017) analysis of the Medicaid expansion. Both of these previous papers used two years of post-reform data, so the addition of a third year does not lead to statistically significant changes in risky health behaviors over the combined post period. However, when we disaggregate by post-reform year, we find evidence of a statistically significant increase in smoking (1.8 percentage points) among residents in non-expansion states in 2016.
30. Effects of the Affordable Care Act on Health Care Access and Self-Assessed Health after Three Years - Inquiry, 55, September 2018, pp. 1-10. (with Chuck Courtemanche, Ben Ukert, Aaron Yelowitz, and Daniela Zapata)
Abstract: Using data from the Behavioral Risk Factor Surveillance System, we examine the causal impact of the Affordable Care Act on health-related outcomes after three years. We estimate difference-in-difference-in-differences models that exploit variation in treatment intensity from two sources: i) local area pre-reform uninsured rates from 2013 and ii) state participation in the Medicaid expansion. Including the third post-reform year leads to two important insights. First, gains in health insurance coverage and access to care from the policy continued to increase in the third year. Second, an improvement in the probability of reporting excellent health emerged in the third year, with the effect being largely driven by the non-Medicaid expansions components of the policy.
29. Early Effects of the Affordable Care Act on Health Care Access, Risky Health Behaviors, and Self-Assessed Health - Southern Economic Journal, 84(3), January 2018, pp. 660-691. (with Chuck Courtemanche, Ben Ukert, Aaron Yelowitz, and Daniela Zapata)
(online appendix available here)
(early version available as NBER working paper 23269)

Abstract: The goal of the Affordable Care Act (ACA) was to achieve nearly universal health insurance coverage through a combination of mandates, subsidies, marketplaces, and Medicaid expansions, most of which took effect in 2014. We use data from the Behavioral Risk Factor Surveillance System to examine the impacts of the ACA on health care access, risky health behaviors, and self-assessed health after two years. We estimate difference-in-difference-in-differences models that exploit variation in treatment intensity from state participation in the Medicaid expansion and pre-ACA uninsured rates. Results suggest that the ACA led to sizeable improvements in access to health care in both Medicaid expansion and non-expansion states, with the gains being larger in expansion states along some dimensions. However, we find no evidence of effects on  risky behaviors or self-assessed health.

28. Medicaid Program Choice, Inertia, and Adverse Selection - Journal of Health Economics, 56, December 2017, pp. 292-316. (with Aaron Yelowitz and Jeff Talbert)
Abstract: In 2012, Kentucky implemented Medicaid managed care statewide, auto-assigned enrollees to three plans, and allowed switching. Using administrative data, we find that the state's auto-assignment algorithm most heavily weighted cost-minimization and plan balancing, and placed little weight on the quality of the enrollee-plan match. Immobility - apparently driven by health plan inertia - contributed to the success of the cost-minimization strategy, as more than half of enrollees autoassigned to even the lowest quality plans did not opt-out. High-cost enrollees were more likely to opt-out of their auto-assigned plan, creating adverse selection. The plan with arguably the highest quality incurred the largest initial profit margin reduction due to adverse selection prior to risk adjustment, as it attracted a disproportionate share of high-cost enrollees. The presence of such selection, caused by differential degrees of mobility, raises concerns about the long run viability of the Medicaid managed care market without such risk adjustment.
27. Do High Fidelity Wraparound Services for Youth with Serious Emotional Disturbances Save Money in the Long Run? - Journal of Mental Health Policy and Economics, 20(4), November 2017, pp. 167-175. (with Angie Snyder, Susan McLaren, Bo Feng*, and Mei Zhou)
Abstract: High Fidelity Wraparound (Wrap) is an evidence-informed practice that sustains community-based placements for youth with serious emotional disturbances through the use of intensive care coordination among parents, multiple child-serving agencies, and providers. While there is growing evidence of the benefits, few studies have examined health care spending associated with Wrap participation. When overall health care spending and changes in spending for youth who transitioned from institutional care into Wrap and youth starting institutional care at the same time, but not receiving Wrap are compared, findings suggest that youth participating in Wrap had much higher average monthly baseline costs. Wrap participation reduced monthly health care spending by an additional $1,130 in the post-period, as compared to controls. This was a result of decreases in mental health inpatient spending and general outpatient spending. Further analysis is needed to assess how these spending changes impacted health outcomes.
26. Medicaid Managed Care and the Health Care Utilization of Foster Children - Inquiry, 54, March 2017, pp. 1-9. (with Makayla Palmer*, Aaron Yelowitz, and Jeff Talbert)
Abstract: The purpose of this paper is to evaluate the differential impact of the transition of foster children from fee-for-service (FFS) Medicaid coverage to Medicaid managed care (MMC) coverage on outpatient health care utilization. Linked administrative data from the Kentucky Cabinet for Health and Family Services was used to construct a dataset of 4,315 children continuously, jointly enrolled in foster care and Medicaid for 12 months. We then applied retrospective difference-in-differences analysis to compare the outpatient utilization of foster children transitioned to MMC in one region of the state to foster children in the rest of the state that remained in FFS. We find that the transition to MMC led to a 3-4 percentage point reduction in the probability of having any monthly outpatient utilization. We also estimate that MMC leads to a reduction in outpatient spending. Thus the transition of foster children into MMC coverage led to a reduction in physical health care services received in the first seven months following the transition. While such reductions in services have positive budget implications, they may lead to a reduction in subsequent health outcomes among this vulnerable population.
25. Impacts of the Affordable Care Act on Health Insurance Coverage in Medicaid Expansion and Non-Expansion States - Journal of Policy Analysis and Management, 26(1), Winter 2017, pp. 178-210 (with Chuck Courtemanche, Ben Ukert*, Aaron Yelowitz, and Daniela Zapata)
(early version also available as NBER working paper 22182)
Abstract: The Affordable Care Act (ACA) aimed to achieve nearly universal health insurance coverage in the United States through a combination of insurance market reforms, mandates, subsidies, health insurance exchanges, and Medicaid expansions, most of which took effect in 2014. This paper estimates the causal effects of the ACA on health insurance coverage using data from the American Community Survey. We utilize difference-in-difference-in-differences models that exploit cross-sectional variation in the intensity of treatment arising from state participation in the Medicaid expansion and local area pre-ACA uninsured rates. This strategy allows us to identify the effects of the ACA in both Medicaid expansion and non-expansion states. Our preferred specification suggests that, at the average pre-treatment uninsured rate, the full ACA increased the proportion of residents with insurance by 5.9 percentage points compared to 3.0 percentage points in states that did not expand Medicaid. Private insurance expansions from the ACA were due to increases in both employer-provided and non-group coverage. The coverage gains from the full ACA were largest for those with incomes below the Medicaid eligibility threshold, non-whites, young adults, and unmarried individuals. We find some evidence that the Medicaid expansion partially crowded out private coverage among low-income individuals.
24. Does Medicaid Managed Care Help Equalize Racial and Ethnic Disparities in Utilization? - Health Services Research, 51(3), June 2016, pp. 872-891 (with Aaron Yelowitz, Meredith Shores*, and Jeff Talbert)
Abstract: In this paper we estimate the impact of different forms of Medicaid managed care (MMC) delivery on racial and ethnic disparities in utilization. Our data is longitudinal, administrative data on 101,649 children in Kentucky continuously enrolled in Medicaid between January 1997 and June 1999. Outcomes considered are monthly professional, outpatient, and inpatient utilization. We apply an intent-to-treat, instrumental variables analysis using the staggered geographic implementation of MMC to create treatment and control groups of children. We find that the implementation of MMC reduced monthly professional visits by a smaller degree for non-whites than whites (3.8 percentage points versus 6.2 percentage points), thereby helping to equalize the initial racial/ethnic disparity in utilization. The Passport MMC program in the Louisville-centered region statistically significantly reduced disparities for professional visits (closing the gap by 8.0 percentage points), while the Kentucky Health Select MMC program in the Lexington-centered region did not. No substantive impact on disparities was found for either outpatient or inpatient utilization in either program. In conclusion, we find evidence that MMC has the possibility to reduce racial/ethnic disparities in professional utilization. More work is needed in order to determine which managed care program characteristics drive this result.
23. Enhanced Citizenship Verification and Children's Medicaid Coverage - Economic Inquiry, 54(3), July 2016, pp. 1670-1683 (with Angie Snyder and Mei Zhou)
(early version available as Andrew Young School of Policy Studies Research Paper Series No. 15-03)
Abstract: This paper examines a potential unintended consequence of the mandated Medicaid citizenship verification requirements of the 2005 Deficit Reduction Act.  We investigate whether or not these new rules led to an increase in the Medicaid exit rate among enrollees using state administrative data from Georgia.  We do this by comparing the exit rate for children enrolled in Medicaid whose first coverage recertification occurs just after implementation of the DRA (which we refer to as a “high impact” first recertification) with those whose first recertification occurs just prior (which we refer to as a “low impact” first recertification).  Our analysis suggests that children in the high impact first recertification group were about two percentage points more likely to exit Medicaid than those in the low impact group.  Furthermore, these additional exits occurred in racial and ethnic groups more likely to be citizens than noncitizens and pre-reform estimates suggest that there were very few (roughly .10 percent) noncitizen Medicaid enrollees to begin with.  Taken together, our results suggest that the DRA enhanced citizenship verification rules led to an increase in Medicaid disenrollment, and thus a reduction in coverage, among citizens.
22. Who Gained Insurance Coverage in 2014, the First Year of Full ACA Implementation? - Health Economics, 25, April 2016, pp. 778–784 (with Chuck Courtemanche and Aaron Yelowitz)
Abstract: The most significant pieces of the Affordable Care Act (exchanges, subsidies, Medicaid expansion, individual mandate), implemented in 2014, were associated with sizable gains in coverage nationally that were divided equally between gains in Medicaid and private Marketplace coverage. These national trends mask heterogeneity in gains by state Medicaid expansion status, age, income level, and source of coverage.

21. Health Insurance generosity and Conditional Coverage: Evidence from Medicaid Managed Care in Kentucky - Southern Economic Journal 82(2), October 2015, pp. 535-555 (with Aaron Yelowitz)

(early version available as Andrew Young School of Policy Studies Research Paper Series No. 14-09)

Abstract: This paper estimates the impact of the introduction of Medicaid managed care (MMC) on the formal Medicaid participation of children. We employ a quasi-experimental approach exploiting the location-specific timing of MMC implementation in Kentucky. Using data from the March Current Population Survey from 1995-2003, our findings suggest that the introduction of MMC increases the likelihood of being uninsured and decreases formal Medicaid participation. This finding is consistent with an increase in “conditional coverage” – waiting until medical care is needed to sign up or re-enroll in Medicaid. These effects are concentrated among low-income children and absent for high-income children. We find no evidence of “crowd-in” – substituting private coverage for Medicaid. These results are robust to multiple placebo tests and imply the potential for less formal participation (i.e. more conditional coverage) among the ACA Medicaid expansion population (which is likely to be primarily covered under MMC) than is typically predicted.

20. Does More Public Health Spending Buy Better Health? - Health Services Research & Managerial Epidemiology 2, April 2015, pp. 1-9 (with Jaesang Sung* and Peggy Honore)

Abstract: In this paper we attempt to address a persistent question in the health policy literature: Does more public health spending buy better health?  This is a difficult question to answer due to unobserved differences in public health across regions as well as the potential for an endogenous relationship between public health spending and public health outcomes.  We take advantage of the unique way in which public health is funded in Georgia to avoid this endogeneity problem.  Using a twelve year panel dataset of Georgia county public health expenditures and outcomes in order to address the “unobservables” problem, we find that increases in public health spending lead to increases in mortality by several different causes, including early deaths and heart disease deaths.  We also find that increases in such spending leads to increases in morbidity from heart disease.

19. Estimating Premium Sensitivity for Public Health Insurance Coverage: Selection but no Death Spiral - Health Services Research 50(2), April 2015, pp. 579-598 (with Pat Ketsche, Kathleen Adams, Angie Snyder, and Mei Zhou)

Abstract: The purpose of this paper is to estimate the duration of CHIP enrollment as a function of premiums that vary within and between income groups and over time. A secondary objective is to simulate changes in enrollment and calculate price elasticities controlling for health status and age.  We use administrative CHIP application and eligibility data to estimate the duration of enrollment as a function of the effective premium.  We control for family characteristics (family income, family size, number of adults in the home, parental age, location (rural, urban, other), race/ethnicity, prior participation in public coverage) and county characteristics such as unemployment rate, share uninsured, and poverty rate. We also control for health status of the child by matching claims to eligibility data to identify children with chronic conditions and high dollar claims.  We conduct sensitivity analysis including children starting their spell with zero premium in the first month of enrollment.  Consistent with prior research, we find that increasing premiums results in shorter duration of enrollment.  Price elasticity for the cohort as a whole is -0.20, consistent with price elasticities identified in the private sector for moderate income families. However, among children facing zero premium at the start of the spell, premium changes result in a significantly larger response (-0.33) than among children facing a positive premium at the start of the spell (-0.15).  High utilization and the presence of chronic conditions reduce price elasticity.

18. Examining the Potential of Information Technology to Improve Public Insurance Application Processes: Enrollee Assessments from a Concurrent Mixed Method Analysis - Journal of the American Medical Informatics Association 21, November 2014, pp.1045-1052. (with Angie Synder, Abhay Mishra, Pat Ketsche, and Susan McLaren)

Abstract: The objective of this paper is to assess the perceived readiness of Medicaid and Children's Health Insurance Program (CHIP) enrollees to use information technologies (IT) to facilitate improvements in the application processes for these public insurance programs.  We conducted a concurrent mixed method study of Medicaid and CHIP enrollees in a southern state.  We conducted six Medicaid (56 participants) and two CHIP (17 participants) recipient focus groups to identify the concerns of enrollees regarding the current application process and their IT proficiency.  Additionally, we surveyed beneficiaries via telephone about their access to and use of the Internet, and willingness to adopt IT-enabled processes. Stratified random sampling was used to be representative of program enrollees and the distinct regions of the state. A total of 2,013 households completed the survey.  We used chi-square analysis for comparison across different groups of respondents.  Our data suggest that a majority of enrollees will embrace IT-enabled enrollment, but a small yet significant group continues to lag access to such technology. Moreover, a segment of those enrolled in the two programs continues to place a high value on personal interactions with program caseworkers.

17. A Tale of Two Cities? The Heterogeneous Impact of Medicaid Managed Care in Kentucky - Journal of Health Economics 36, July 2014, pp. 47-68. (with Aaron Yelowitz and Jeff Talbert)

(earlier version available as Andrew Young School of Policy Studies working paper 11-28)
Abstract: Evaluating Accountable Care Organizations is difficult because there is a great deal of heterogeneity in terms of their reimbursement incentives and other programmatic features. We examine how variation in reimbursement incentives and administration among two Medicaid managed care plans impacts utilization and spending. We use a quasi-experimental approach exploiting the timing and county-specific implementation of Medicaid managed care mandates in two contiguous regions of Kentucky. We find large differences in the relative success of each plan in reducing utilization and spending that are likely driven by important differences in plan design. The plan that capitated primary care physicians and contracted out many administrative responsibilities to an experienced managed care organization achieved significant reductions in outpatient and professional utilization. The plan that opted for a fee-for-service reimbursement scheme with a group withhold and handled administration internally saw a much more modest reduction in outpatient utilization and an increase in professional utilization.

16. Employer Provided Health Insurance and the Adverse Selection Problem - Public Finance Review 41(1), January 2013, pp. 3-36. (with Marco Castaneda)

Abstract: The role of employers in the health insurance market in the United States is substantial, but there is little formal analysis investigating how the introduction of employers affects the structure of equilibrium contracts in the health insurance market. In this article, the authors present and analyze a model of employer-provided health insurance and show that if the labor market is competitive, the equilibrium must be a separating equilibrium, as in the standard model. The key to this result is that insurance companies and employers have similar incentives because health insurance is an important component of total compensation. In addition, the authors show that if workers bargain collectively for employment contracts, the equilibrium can be a separating or a pooling equilibrium, but in any equilibrium, the low risks obtain more insurance than in the standard model and subsidize the insurance premiums of the high risks.

15. Retiree Health Benefits as Deferred Compensation: Evidence from the Health and Retirement Study - Public Finance Review 41(1), January 2013, pp. 64-91. (with Steve Woodbury)

Abstract: Are early retiree health benefits (RHBs) a form of deferred compensation binding workers to an employer? Most employers who offer RHBs offer them only to workers who have ten or more years of tenure and have reached age fifty-five. Accordingly, workers in firms offering RHBs have an  incentive to stay with a firm in the years before they attain eligibility for RHBs, and a greater incentive than otherwise to retire thereafter. The authors test for such a pattern of incentives by examining the age-specific relationship between workers’ eligibility for RHBs and retirement. The  findings suggest that workers in RHB-offering  firms are less likely to retire at ages fifty and fifty-one than similar RHB-ineligible workers. Also, RHB-eligible workers aged sixty and sixty-one are more likely to retire than similar RHB-ineligible workers. These results are consistent with RHBs acting as part of a delayed payment contract of the kind described by Lazear.

14. The Effects of Medicaid Policy Changes on Adults’ Service Use Patterns in Kentucky and Idaho - Medicare & Medicaid Research Review 2(4), 2012, pp. E1-E22. (with Jenny Kenney, Jennifer Pelletier, Ariel Klein, and Jeff Talbert)

Abstract: In 2006, Idaho and Kentucky became two of the first states in the nation to implement changes to their Medicaid programs under authority granted by the 2005 Deficit Reduction Act (DRA) (PL 109-171).  Concerns about Medicaid spending growth in both states led to a desire for Medicaid reform to make the programs sustainable for future generations and to encourage the greater use of preventive care and the adoption of healthier lifestyles.  The purpose of this paper is to examine the impact of the key reforms in each state on the utilization rates among non-elderly adult Medicaid recipients for a variety of services, including physician visits, dental visits, emergency room visits, inpatient stays, and prescription drugs.  While the majority of adults in Medicaid in both states received primary medical care in the past year, rates of preventive care and dental care utilization are very low.  The introduction of service limits on brand name prescriptions in Kentucky led to a shift towards generics.  New copayments for a variety of services in Kentucky did not appear to have a dramatic impact on utilization.  The addition of an adult wellness exam to the Medicaid benefit package in Idaho had modest effects on service use.  The adoption of managed care for dental provision for nondisabled adults increased both the use of overall and preventive dental care.

13. The Effects of Medicaid and CHIP Policy Changes on Receipt of Preventive Care among Children in Kentucky and Idaho - Health Services Research 46 (1p2), February 2011, pp. 298-318. (with Jenny Kenney, Jennifer Pelletier, Ariel Klein, and Jeff Talbert)

Abstract: The objective of this paper is to examine children’s receipt of preventive care in Medicaid/CHIP in two states that adopted policy changes aimed at promoting greater provision of preventive care to children using Medicaid/CHIP claims and enrollment data from Idaho and Kentucky from 2004 to 2008.  Linear probability and hazard models are used to estimate policy effects with either a pre-post model or a pre-post model with a comparison group. Models control for age, gender, race/ethnicity, and eligibility category.   The effects of the increased reimbursement for well-child care were small, suggesting at most an increase in the probability of having any annual well-child visit of 1 to 3 percentage points.  The introduction of the Wellness PHA benefit in Idaho was associated with between 11 and 29 percentage point increases in well-child receipt among the three key target groups and with quicker receipt of well-child care following CHIP enrollment.  Children were 1 to 6 percentage points more likely to receive preventive dental care and received preventive dental care more quickly after the policy changes.  Policy changes such as fee increases, incentives, and delivery system changes can lead to increases in preventive care use among children. However, additional policy changes would be required to address the persistent shortfalls in preventive care receipt for children.

12. Disparate Effects of CHIP Premiums on Disenrollment for Minorities - Current Issues in Health Economics, Volume 290, edited by D. Slottje and R. Tchernis, December 2010, United Kingdom: Emerald Group. (with Cynthia Searcy and Jennifer Ghandi*)

Abstract: This paper examines whether or not the introduction of a new $20 family premium in Kentucky’s CHIP program in late 2003 had a differential impact on the enrollment duration of children in different demographic groups, with a special focus on any potential differences by                     race/ethnicity.  A competing risk hazard model is estimated in order to differentiate between children exiting CHIP via a transfer to  Medicaid and children that exited public coverage completely.  We find that non-white children are generally more likely to exit than white children.  This                  general white / non-white difference increases immediately following the introduction of the $20 premium.


11.
Funding Trauma Care in Georgia - Journal of the Medical Association of Georgia 99(4), December 2010, pp. 19-20. (with Pat Ketsche)

Abstract: On November 2, Georgia voters rejected legislatively-referred constitutional amendment number 2 that would have provided funds to support the statewide trauma care network, with 53% of votes against the measure.  This amendment was placed on the November ballot by the Georgia legislature and it imposed a $10 annual trauma charge on each motor vehicle designed to carry ten or fewer persons.  These funds were to be placed in a designated trauma trust fund and would have raised an estimated $80 million per year in a manner that is proportional to citizens engagement in an activity related to trauma care, driving. Potential opposition to the amendment may have been due to some distaste for any new taxes or fees during a time of economic uncertainty and some unwillingness from parts of the state with greater trauma care coverage to subsidize coverage for other parts of the state.  The purpose of this paper is to discuss the background related to trauma care funding in Georgia and general challenges associated with funding trauma care.

10. CHIP Premiums, Health Status, and the Insurance Coverage of Children - Inquiry, 47(3), Fall 2010, pp. 199-214. (with Jeff Talbert)

(also available as an ERIU working paper)
Abstract: This study uses the introduction of premiums into Kentucky's Children's Health Insurance Program (KCHIP) to examine whether the enrollment impact of new premiums varies by child health type. We also examine the extent to which children find alternative coverage after premium nonpayment. Public insurance claims data suggest that those with chronic health conditions are less likely to leave public coverage. We find little evidence of a differential impact of premiums on enrollment among the chronically ill. Our survey of nonpayers shows that 56% of responding families found alternative private or public health coverage for their children after losing CHIP.

9. SCHIP Premiums, Enrollment, and Expenditures: A Two State, Competing Hazard Analysis - Health Economics, 19(7), July 2010, pp. 772-791. (with Pat Ketsche and Mei Zhou)

Abstract: Policymakers may impose or increase premiums for SCHIP recipients under an assumption that family income is exogenous, implying no spill-over effects on other public health insurance such as Medicaid.  Using state administrative data on the introduction of a new premium in Kentucky’s SCHIP program and an increase in existing premiums in Georgia’s SCHIP program, we test this assumption by estimating competing risk models that differentiate between different exit routes from SCHIP coverage.  The theoretical motivation for this analysis is a model of family budget constraints in which changes in SCHIP premiums create notches in the budget constraint that may cause families to re-evaluate their labor / leisure trade-off.  We find the premium changes in each state lead to an increase in the likelihood of children leaving the SCHIP program for no public health insurance coverage in the months immediately following the premium changes.  This short run disenrollment effect in each state is offset to some extent by a short run increase in the likelihood that children transfer to lower income eligibility / lower premium categories of SCHIP.  In Kentucky we also find a short run increase in the likelihood that children exit SCHIP by transferring to Medicaid coverage.  SCHIP covered children in Georgia are found to be less likely to transfer into Medicaid immediately following the premium increase.

8. A Model of Commodity Differentiation with Indivisibilities and Production - Economic Theory, 34 (1), January 2008, pp. 85-106. (with Marco Castaneda)

(earlier version also available here)

Abstract:
This paper presents an existence theorem in a general equilibrium model of a production economy with commodity differentiation and indivisibilities.  The model is motivated by the existence of markets with indivisible commodities, such as the markets for automobiles, health insurance, and computers.  As is standard in the literature, the space of commodity characteristics is described by a compact metric space and a commodity vector is described by an integer-valued Borel measure on the space of commodity characteristics.  An atomless measure space of producers and consumers is assumed to overcome the problem of non-convexity of the production and consumption sets induced by indivisibilities.

7. Employee Choice of Flexible Spending Account Participation and Health Plan - Health Economics, 17(7), July 2008, pp. 793-813. (with Bart Hamilton)

(earlier version also available here)

Abstract:
Despite the fact that flexible spending accounts (FSAs) are becoming an increasingly popular employer-provided health benefit, there has been very little empirical study of FSA use among employees at the individual level.  This study contributes to the literature of FSAs through the use of a unique dataset that provides three years of employee level matched benefits data.  We examine the determinants of FSA participation and contribution levels using yearly order probit models and a random effects ordered probit model.  FSA participation and health plan choice decisions are also modeled jointly in each year using conditional logit models.  We find that, even after controlling for a number of other demographic characteristics, non-whites are less likely to participate in the FSA program than whites.  We also find evidence that choosing health plans with more expected out of pocket expenses is correlated with use of the FSA program.

6. Assessing Potential Enrollment and Budgetary Effects of SCHIP Premiums: Findings from Arizona and Kentucky - Health Services Research42 (6p2), December 2007, pp. 2354-2372. (with Jenny Kenney, Julia Costich,  Josh McFeeters)

(earlier version also available here)

Abstract:
In this study
we use administrative data from two states, Arizona and Kentucky, which introduced new premiums for certain income categories in their SCHIP programs in 2004 and 2003, respectively.  We use multivariate hazard models to study rates of disenrollment and re-enrollment for the recipients who had been enrolled in the categories of SCHIP in which the new premiums were implemented.  Competing hazard models are used to determine if recipients leaving SCHIP following the introduction of the premium were obtaining other public coverage or exiting public insurance entirely at higher rates.  We also used time-series models to measure the effect of premiums on changes in caseloads in premium-paying SCHIP and other categories of public coverage and we assessed the budgetary implications of imposing premiums.

5. Responses to Declining Retiree Health Benefit Coverage - in Government Spending and the Elderly, edited by D. B. Papadimitriou, 2007, New York: Palgrave Macmillan. (with Steve Woodbury)

(also available as Upjohn Institute staff working paper No. 06-128)
Abstract: Employer-provided health benefit coverage for workers who retire before age 65 has fallen over the last decade. We examine a cohort of male workers from the Health and Retirement Survey to examine questions about the dynamics of retiree health benefits and the relationship between retiree health benefit and retirement behavior. Having a better understanding of this relationship is important for the policy debate over the best way to increase health coverage for older Americans without reducing the incentive to work. On dynamics, we find that between 1992 and 1996, 24 percent of full-time workers who had retiree health benefits lost their coverage, while 15 percent of full-time workers who lacked coverage gained it. Also, of the full-time employed men who were covered by retiree health benefits in 1992 and had retired by 1996, 3 percent were uninsured, and 15 percent were covered by health insurance other than employer-provided insurance. On the relationship between retiree health benefits and retirement, we find that workers with retiree benefits were 29 to 55 percent more likely to retire than those without. We also find that workers who are eligible for retiree health benefits tend to take advantage of them when they are relatively young.

4. The Effects of Premium Increases on Enrollment in SCHIP Programs: Findings from Three States - Inquiry, 43 (4), Winter 2006/2007, pp.378-392.
(with Jenny Kenney, Andy Allison, Julia Costich,  Josh McFeeters)

Abstract: This study examines the effects of new and higher premiums on SCHIP enrollment in Kansas, Kentucky, and New Hampshire—three states that implemented premium changes in 2003. We used state administrative enrollment records from 2001 to 2004–2005 to track changes in total caseloads, new enrollments, and disenrollment timing in premium-paying categories of SCHIP before and after the premium changes were implemented. Premium hikes were associated with lower caseloads in all three states and with earlier disenrollment in Kentucky and New Hampshire. Premium increases appeared to have greater disenrollment effects for lower-income children in New Hampshire and for nonwhite children in Kentucky.
3. The Impact of the Introduction of Premiums into a SCHIP Program - Journal of  Policy Analysis and Management, 26 (2), Spring 2007, pp. 237-255.

 Abstract: This paper examines the introduction of premiums into the SCHIP program in Kentucky.  Kentucky introduced a $20 monthly premium for SCHIP coverage for children with family incomes between 151% and 200% of the federal poverty level in December 2003.  Administrative data between 2001 and 2004 is used to estimate a Cox proportional hazard model that predicts enrollment duration in this premium-paying category.  The results suggest that a premium reduces the length of enrollment and that the effect is much stronger in the first two months after the introduction of the premium.  Similar results are not found for the non-premium category. 

2. Medicaid Expenditures and State Budgets: Past, Present, and Future - National Tax Journal, 60 (2), June 2007, pp. 279-304. (with Dave Wildasin)

(earlier version also available as IFIR working paper 2007-04)

Abstract:
Rapid spending growth has made Medicaid a major element in state budgets; financial support from Federal matching grants is now a main component of state government revenue and of intergovernmental fiscal relations.   We discuss recent, ongoing, and prospective reforms of intergovernmental finances and regulations, including the 1996 welfare reform, the introduction of Medicare Part D, Section 1115 waivers, SCHIP reauthorization, and a shift to block grants.  Each would affect the assignment of responsibilities between the state and Federal governments, the viability of which is questionable due to current and future interstate demographic and policy variation, population aging, and Federal fiscal imbalances.

1. State Government Cash and In-Kind Benefits: Intergovernmental Fiscal Transfers and Cross-Program Substitution - Journal of  Urban Economics, 61 (1), January 2007, pp. 1-20. (with Dave Wildasin)

(earlier versions also available as IFIR working paper 2006-01)

Abstract: US states provide both cash and health insurance benefits for the poor, partially financed by fiscal transfers from the Federal government.  The 1996 welfare reform drastically reduces Federal support for cash transfers at the margin , lowering the relative price to states of providing benefits to the poor through Medicaid.  This paper analyzes the comparative-statics response of state governments to such changes in intergovernmental transfers, showing (in central cases) that they can contribute not only to reductions in state expenditures on cash benefits but to increases in expenditures on Medicaid, whether or not beneficiary populations are mobile among states.
        



Papers Under Review / Revision Requests:


1. Inequality and Health: A New Approach to Measure Income Inequality (with Jaesang Sung* and Qihua Qui*) 
(early version available as a SSRN working paper)
Abstract: The relative income hypothesis suggests that an individual’s health is impacted by the income of others. However, prior studies suffer from mixed empirical findings that could be due to a lack of annual individual income data with sufficient sample size. We apply a new methodology to calculate a variety of income inequality measures based on aggregate income and household size data from various Federal data sources. Our proposed methodology provides a way to express various income inequality measures as a function of the ratio of mean to median household income under the assumption that individual income is log-Normally distributed. This approach produces a variety of precise annual income inequality measures at different levels of geography, thus solving the sample size problem by incorporating externally calculated inequality measures. Combining the 2001-2012 editions of the U.S. Behavioral Risk Factor Surveillance System with annual regional income inequality measures derived from our methodology enables us to estimate both the contemporaneous and the lagged effect of income inequality on individual health outcomes.




Working Papers:


1. The Impacts of the Food Stamp Program on Mortality (with Chuck Courtemanche, Christian Gregory, and Jordan Jones*)
Abstract: This study examines the effect of food stamps on health.  Specifically, we use the county-level rollout of the Food Stamp Program from 1961 to 1975 as a source of variation in access to food stamps in order to examine food stamps’ single-year and multi-year effects on various county-year level mortality rates using fixed effects models.  We consider aggregate mortality rates, subgroup rates for sex, race groups, and age groups, and rates for specific causes of death to examine the mechanisms through which food stamps affect health.  We find mixed results for the entire 1969 to 1978 county sample that indicate small or zero overall effects of access to food stamps on mortality rates.  Using a subsample of poorer counties, we find that food stamps tend to reduce most mortality rates over time.   
2. Group Prenatal Care and Emergency Room Utilization (with E. Heberlein, J. Smith, A. Crockett, and S. Covington-Kolb)
Abstract: