Principal investigator: Bruno Benevit
Original title: Does Public Insurance Crowd Out Private Insurance
Authors: David M. Cutler and Jonathan Gruber
Location of the Intervention: United States
Sample Size: 266.421 children, 549.472 adults
Sector: Health Insurance
Primary Variable of Interest: Health Insurance Coverage
Type of Intervention: Eligibility; Insurance coverage amount
Methodology: Instrumental Variable
Summary
The broad coverage of health services is the target of great demand and public interest. At the same time, public managers face the challenge of identifying the negative consequences of applying subsidies in this market. The objective of this article was to analyze the impacts of the policy of expanding eligibility for the program. Medicaid Regarding the program's coverage, the study examines private health insurance coverage. It also verifies how employer-provided insurance was affected. Using an instrumental variables methodology, the authors identify that the policy of expanding eligibility led to an increase in program utilization and total coverage; however, this movement was accompanied by a reduction in coverage and uptake of employer-provisioned health insurance.
- Policy Problem
The provision of free public insurance services has consequences across the market as it alters its supply structure. Theoretically, the Medicaid It is a very valuable policy, offering broad coverage with little or no fees charged to beneficiaries. However, the program's cost is subsidized and therefore lower than that of private insurance. Due to the low reimbursement rates, service providers are often reluctant to treat patients through it. Medicaid (Currie et al., 1995). The value of Medicaid Adherence to [the private sector] may be low due to difficulties transitioning to the private sector in cases of pre-existing medical conditions. Medicaid It can also be impacted due to the stigma associated with public programs or registration difficulties related to bureaucracy.
Public subsidized health insurance policies provide a homogeneous basket of services to their beneficiaries. Therefore, individuals who opt for Medicaid They forgo the option of supplementing their access to health services. This dynamic implies that individuals who place little value on this type of service join the program, while individuals willing to pay more for the quality and quantity of these services opt for private insurance.
In this sense, the challenge for policymakers lies in verifying the effects caused by the expansion of the program. Medicaid in the population's health insurance coverage, and extends to identifying the mechanisms that explain these potential impacts on the sector.
- Implementation and Evaluation Context
The expansion of the program Medicaid The expansion aimed to broaden coverage for the poorest population. Prior to the expansion, eligibility criteria were excessively restrictive, only including individuals in extreme poverty and/or single-parent families. In 1987, the percentage of children and women of reproductive age eligible for the program was 17,8% and 21,1%, respectively. Five years after the program's expansion began, these numbers reached 27% and 44,9%, respectively. Furthermore, almost 90% of private health insurance in the U.S. was provided by employers.
The objective of this article was to analyze how the expansion of Medicaid The study examined the impact on health insurance coverage for the population in the US. Observing the differences in services and incentives between the public and private health insurance sectors, the study verifies possible repercussions in both sectors, as well as checking if there was an increase in the number of uninsured individuals. Additionally, the authors perform an analysis of how the expansion of eligibility for... Medicaid It affected the supply and uptake of private insurance policies offered by employers.
- Policy/Program Details
The program Medicaid Eligibility for pregnant women and children has historically been linked to participation in the program. Aid for Families with Dependent Children – AFDC, a policy to assist families with dependent children, focusing on single-parent families. This linkage resulted in restrictions on access to the program due to the AFDC program's eligibility criteria. Simultaneously, several other state programs existed aimed at relaxing the socioeconomic eligibility criteria for the program. Medicaid. However, the family structure criteria still limited the policy to individuals in situations of extreme poverty.
With the goal of expanding insurance coverage for the low-income population, the link between AFDC coverage and Medicaid eligibility was gradually weakened for pregnant women and children. According to Currie & Gruber (1994), these measures promoted the expansion of eligibility for the program. MedicaidThis substantially increased the maximum income limit for families to qualify for the program and eliminated restrictions for non-single-parent families. Children born after September 30, 1983, into poverty were fully covered, regardless of family composition. Subsequently, new measures in 1992 expanded the coverage obligation to pregnant women and children under 6 years of age and relaxed the poverty criteria adopted by the states.
- Method
This article used the Current Population Surveys database (Current Population Surveys – CPS) from March 1988 to 1993. The CPS is the largest nationally representative annual survey of demographic data relating to insurance coverage and program eligibility information. MedicaidThe sample used contains 815.893 observations, of which 266.421 are children (between 0 and 18 years old), 194.139 are women of reproductive age (between 15 and 44 years old), and 355.333 are from the remaining group of adults. The study also employed the Employee Benefits supplement (Employee Benefits Supplement) for the CPS data from May 1988 and April 1993.
Controlling for potential biases in estimations is essential and requires certain precautions regarding the variables used in the model. Children need to be categorized by age given the differences in demand and eligibility between newborns and other children. Furthermore, using the direct eligibility variable is not appropriate because the program's eligibility criteria involve family demographic and socioeconomic characteristics, private coverage for low-income workers, state-by-state differences in expansion, and differences in the frequency of granting eligibility in relation to the variables used.
To address endogeneity issues, the authors adopt instrumental variables (IV) correlated with program eligibility. MedicaidThe first IV consists of individual eligibility (IV_eligibility) and is calculated by considering state-level eligibility averages through national random samples. The second IV is created to represent the relative dollar value of family health expenses covered by Medicaid (IV_%DollarsAdditionally, the second instrumental variable is divided into two parts, representing the individual's coverage value (IV_%Dollarsproprio) and the coverage amount for the remaining family members (IV_%Dollarsfamilia).
Instrumental variables are used to measure the impact of expanding eligibility for the Medicaid regarding the outcome variables analyzed. The first instrumental variable measures the effect of eligibility for the Medicaid, while the others measure the effect of the program's health expense coverage value. The defined outcome variables were: (i) the program's own coverage, (ii) private insurance coverage, and (iii) the uninsured rate. The models also considered covariates of demographic characteristics, including sex (children), race, marital status, number of workers in the household, sex of the head of household, and binary variables of age, state, and year.
All analyses use CPS data. The first analysis of the study estimates the effect of program eligibility. Medicaid for the group of children and women of reproductive age on the outcome variables: (i) coverage of Medicaid, (ii) coverage by private insurance plans, and (iii) lack of health insurance coverage. The second analysis incorporates data from the supplement Employee Benefits Supplement from CPS and checks the impact of the variable IV_%Dollars for all workers aged 24 to 64 on outcome variables: (i) insurance coverage offered by employers, (ii) insurance offers by employees, and (iii) employee adherence to insurance offered by employers. The third analysis observes the impact of the variable IV_%Dollars for men aged 24 to 64 on the outcome variables: (i) insurance coverage offered by employers, (ii) individual insurance coverage, and (iii) insurance coverage for dependents. The fourth analysis verifies the effect of the variables IV_%Dollarsown and IV_%Dollarsfamily regarding outcome variables: (i) coverage of Medicaid, (ii) coverage by private insurance plans, and (iii) non-coverage by health insurance. Finally, the authors present the impact on overall coverage with adjustments for conditional coverage, accounting for non-continuous use of the program. Medicaid.
- Main results
In relation to the first analysis, the results showed a significant increase in the coverage of children by the program. MedicaidHowever, this was accompanied by reductions in private coverage and a lack of coverage, both reactions in line with expectations. The results for women of reproductive age do not indicate an increase in coverage by MedicaidHowever, they demonstrate a reduction in private coverage and an increase in the absence of coverage in proportional magnitudes.
The estimates from the second analysis demonstrated that the dollar value of family coverage provided by Medicaid It significantly reduced private insurance coverage for workers between the ages of 24 and 64. Furthermore, the lack of effect on insurance offerings by employers and the reduction in uptake of private insurance plans indicated that the reduction in private insurance coverage occurred voluntarily on the part of the employees.
The results of the third analysis indicated that expanding eligibility for children led to changes in the type of private coverage demanded, with an increase in individual coverage for adult men and a reduction in coverage for dependents. The results of the fourth analysis demonstrated that increases in the dollar value of family coverage led to an increase in coverage for... Medicaid of children and women of reproductive age. However, both groups observed reductions in private insurance coverage resulting from the value of their own and family coverage.
Considering the estimates of the effect of increased adherence to Medicaid (800 people) and a reduction in the contracting of private insurance (700 people), estimates pointed to relative stagnation in total coverage for women. When considering conditional coverage of MedicaidThe authors identified that the coverage of Medicaid For women, the number of users increased by 900. The authors estimated this result by considering the annual cost to women of reproductive age for pregnancy-related services. Similarly, for children, there was an increase of 1,5 million in coverage. Medicaid Directly, 400 conditionally, and a reduction of 600 in private coverage. In total, the results indicate that there was a total increase of 3,5 million individuals covered by the program. Medicaid and a reduction of 1,7 million in private coverage.
- Lessons in Public Policy
This study analyzed the impacts of expanding the eligibility of the program. Medicaid experienced between the late 80s and early 90s. The evidence found in this study indicated that the increase in program coverage Medicaid This came at the cost of reduced private insurance coverage for all segments of the population.
The children's segment showed the greatest convergence towards coverage of Medicaid, in accordance with the focus established by the program's eligibility flexibility policies. The results regarding conditional coverage of Medicaid For women and children, this provides important information about the use and/or knowledge of eligibility, indicating that a large proportion of women only benefit from the program at the end of their pregnancy.
Although all population segments have reduced their private insurance coverage, there has been no reduction in the supply of these services by employers, indicating that this decrease occurred by choice of the employees. Overall, the estimates in this study indicate that the expansion of Medicaid It was responsible for 17% of the decline in private health insurance coverage observed between 1987 and 1992.
References
Currie, J. and Gruber, J. (1994), Saving Babies: The Efficacy and Cost of Recent Expansions of Medicaid Eligibility for Pregnant Women, No. w4644, National Bureau of Economic Research, Cambridge, MA, p. w4644.
Currie, J., Gruber, J. and Fischer, M. (1995), “Physician Payments and Infant Mortality: Evidence from Medicaid Fee Policy”, The American Economic Review, Vol. 65 No. 2, pp. 106–111.
