Shifting Catastrophic Health Costs to Medicaid: Implications for Private Insurance Premiums
A Simulation of Medicaid as a Catastrophic Payer in Private Insurance
Abstract
This paper evaluates a hybrid model in which Medicaid assumes responsibility for catastrophic health expenses exceeding $50,000 per person per year, while private insurers cover lower-cost claims. Using recent Medical Expenditure Panel Survey (MEPS) data, the analysis finds that shifting this high-cost layer to Medicaid could reduce private insurance premiums by 15–25 percent for the working-age, privately insured population, at an estimated public cost of $130–$190 billion annually under full implementation.
Health care spending in the United States is heavily concentrated, with a small share of patients accounting for most costs; this concentration explains why reinsurance can substantially reduce private premiums by removing the extreme tail of medical risk. The results align with an earlier simulation published in The Geneva Papers on Risk and Insurance (2010), which modeled a 20 percent public reinsurance share and produced similar premium reductions. Together, the studies confirm that reinsurance can stabilize premiums, though its precise magnitude warrants further empirical study.
1. Introduction
Health care spending in the United States is highly concentrated: a very small share of patients accounts for most total costs. This concentration creates challenges for private insurance markets, where premiums must be set high enough to cover rare but extremely expensive claims.
This paper examines a hybrid model in which the public sector (Medicaid) assumes responsibility for catastrophic costs above $50,000 per person per year, while private insurers cover routine and moderate expenses. We estimate the effect of this shift on private insurance premiums for the working-age U.S. population using data from the Medical Expenditure Panel Survey (MEPS) and related sources.
This work builds directly on the author’s earlier contribution published in The Geneva Papers on Risk and Insurance (2010), which modeled a smaller public reinsurance share (20 percent of costs above $50,000). Both papers considered the impact on private premiums from a policy of shifting costs exceeding a $50,000 threshold to a government entity. This paper proposes automatic access to Medicaid once premiums hit particular thresholds.
One month free:
https://bernsteinbook1958.substack.com/fb965b7d
50 percent off annual membership ($30 total.)
https://bernsteinbook1958.substack.com/4d9daaf9
The analysis in the first paper involved the construction of a complex statistical model built from survey data obtained from the Medical Expenditures Panel published by AHRQ. The analysis presented here is based on information obtained and models constructed with the assistance of CHAT GPT. Estimates on the benefits of reinsurance from the two models are consistent, though the 2010 program covered only 20 percent of costs above $50 000.
The results suggest that shifting the $50,000+ portion of health spending to Medicaid could reduce private insurance premiums by roughly 15–25 percent, depending on population mix and spending distribution.
Additional analysis, presented in the appendix of the paper, indicates that the reform could be instituted at relatively modest cost to taxpayers.
2. Policy Background
The U.S. health financing system combines private insurance, employer coverage, and public programs such as Medicaid and Medicare. For working-age adults, private insurance dominates coverage, but premiums continue to rise due in large part to a small fraction of extremely high-cost enrollees.
Private insurers use large risk pools and reinsurance to manage volatility, but high-end medical claims still exert upward pressure on premiums. Medicaid, meanwhile, already serves as a financial safety net for low-income or medically impoverished individuals.
A hybrid structure—where Medicaid covers costs above $50,000—would make Medicaid function as a catastrophic reinsurer, leaving private insurers responsible for the first layer of expenses (for example, $5,000–$50,000 per enrollee).
Here we have assumed a $5,000 deductible and no coinsurance over the deductible. A more complex cost-sharing arrangement could also be considered.
3. Analytical Framework
Let X denote total annual health expenditures per person, and d the deductible ($5,000).
Under a pure private plan, the insurer’s expected annual cost per person is: E[(X − d)+].
Under the hybrid plan, the insurer’s expected annual cost is: E[(min(X, 50,000) − d)+].
The ratio of these two expectations determines the relative premium under the hybrid plan:
This equation shows that the premium reduction equals the share of total post-deductible costs that occur above $50,000 — the “catastrophic tail.”
4. Data and Empirical Inputs
The analysis draws primarily on MEPS, which provides nationally representative data on health spending.
Key 2022 MEPS findings
The top 1 % of individuals account for 22–24 % of all spending.
The top 5 % account for 50 % of spending.
The 99th percentile spending threshold is $80,990; the 95th percentile is $30,206.
Average spending among the top 1 % is $147,071, implying an average of $97,000 in spending above $50,000.
These figures imply that only about 1–3 % of the population exceeds $50,000 in annual costs, with somewhat lower rates among the working-age (18–64) population.
5. Estimating the Premium Impact
We estimate the private insurer’s expected cost for the two plans as follows:
Pure private plan: E(X − 5,000)+ ≈ $5,000–$6,500 per person per year
Hybrid plan tail shifted to Medicaid: E(X − 50,000)+ ≈ $800–$1,000 per person per year (working-age estimate)
Substituting these values into the equation:
Premium Ratio = 1 − (800 ÷ 5,000) = 0.84
This result implies that the private premium under the hybrid plan would be roughly 16 percent lower than under the pure private plan. Across plausible assumptions, the reduction ranges from 15 to 25 percent.
6. Fiscal and Distributional Implications
Transferring catastrophic costs to Medicaid would shift an estimated $800–$1,200 per person per year in spending from the private to the public sector.
The total cost would depend on the size of the population receiving the new benefit.
If the benefit were applied to the entire working-age population, the total cost would be approximately 160 million privately insured individuals, around $130–$190 billion annually in gross public expenditure.
If the benefit were applied only to the portion of the working population currently receiving insurance from state exchanges or the small-group market, not subject to the employer mandate, the total cost would be approximately 25 percent of the full-population estimate.
A more in-depth analysis of taxpayer costs from the proposed reform would examine eight factors: (1) the incidence of high-cost cases, (2) positive skewness of health expenditures, (3) the lower compensation rate of the Medicaid program, (4) impact of premium reduction on the premium-tax-credit subsidy, (5) increases in utilization, (6) enrollment growth, (7) reduction in uncompensated care from fewer uninsured, and (8) disenrollment from traditional Medicaid.
Other outcomes from the reform include potentially higher wages from the lower direct cost to employers, stress on state and federal budgets, and more stable insurance outcomes for low-income households or households experiencing a decrease in income from a change in employment circumstances.
Distributional effects include:
Employers and workers: lower premiums and potentially higher wages.
State and federal budgets: increased responsibility for catastrophic costs.
Low-income households: reduced financial-risk exposure and more stable coverage.
7. Discussion
Because catastrophic spending is so concentrated, relatively small coverage changes can produce significant premium shifts. The hybrid model effectively turns Medicaid into a public reinsurance mechanism, similar in concept to ACA Section 1332 waiver programs but targeted specifically at high-end costs.
The policy could stabilize private premiums and expand affordability, but it also raises issues of coordination, cost tracking, and possible incentives for cost-shifting between private insurers and Medicaid.
8. Limitations
Data limits: MEPS underrepresents the very highest-cost cases due to top-coding; commercial datasets (HCCI, Marketscan) suggest even heavier tails.
Static analysis: the results do not incorporate behavioral responses by insurers or providers.
Payment rates: Medicaid typically reimburses at lower rates than private insurers, so real fiscal transfers could differ.
9. Comparisons to Previous Research
A related earlier study by the author—“Health Care Reinsurance and Insurance Reform in the United States: A Simulation Model” (Bernstein, The Geneva Papers on Risk and Insurance, 2010)—developed a simulation of a publicly subsidized reinsurance system covering 20 percent of private medical costs above $50 000. That work used 1999–2006 MEPS microdata to estimate how partial government reinsurance could reduce insurer cost variance and premiums in the individual and small-group markets, focusing exclusively on households with private health insurance.
The present analysis differs from that earlier work in several key respects:
• Coverage structure: The 2010 model simulated partial reinsurance (government paying 20 percent above $50 k), whereas this paper evaluates a full public takeover of catastrophic spending through automatic Medicaid eligibility above $50 k.
• Analytical method: The earlier study relied on a micro-simulation algorithm applied to household-level MEPS records; the current analysis uses results from a simple expectation-based model constructed with the assistance of CHAT GPT.
• Scope and population: The prior paper examined only the privately insured population in the individual and small-group markets; the present study applies to the entire working-age, privately insured U.S. population.
• Data vintage: The original analysis used 1999–2006 MEPS data (inflated to 2006 dollars), while this paper uses 2018–2022 MEPS data and contemporary spending-concentration measures.
• Data adjustment: The 2010 paper increased all MEPS expenditure figures by 27 percent to correct for MEPS’s underestimation of aggregate national spending (as documented by Sing et al., 2006). The current analysis does not apply this adjustment, relying directly on published MEPS means. This difference partly accounts for lower overall cost estimates in the current study.
• Plan design differences: The Geneva simulation used distinct deductibles and coinsurance rates for individual and family policies, while the present study assumes a uniform $5 000 deductible and no coinsurance. This difference led to lower insurer payments above the threshold in the Geneva model and thus a larger percentage premium reduction under reinsurance relative to baseline.
• Policy framing: The 2010 paper conceptualized reinsurance as a subsidy mechanism supporting market regulation; this paper treats Medicaid as a structural catastrophic payer integrated into the national insurance system.
It is clear that reinsurance subsidies reduce private insurance premiums, but additional research is needed to determine the precise magnitude of the effect.
10. Conclusion
This analysis indicates that if Medicaid assumed responsibility for all spending above $50,000 per enrollee, private insurance premiums could decline by roughly one-fifth for the working-age population.
Such a policy would meaningfully lower private-premium costs and improve affordability, though at the expense of additional public outlays. Policymakers considering this approach must weigh the fiscal burden on Medicaid against potential gains in economic efficiency and coverage stability.
Future research should replicate these results using commercial claims data, model dynamic effects (such as trend growth in specialty drugs), and explore alternative thresholds (e.g., $40,000 or $75,000) to optimize cost-sharing between private and public sectors.
References
Agency for Healthcare Research and Quality (AHRQ). Medical Expenditure Panel Survey (MEPS) Statistical Brief #556: Concentration of Health Expenditures and Selected Characteristics of High Spenders, 2022.
Bernstein, D. B. (2010). “Catastrophic Health Reinsurance: A Market-Based Approach.” The Geneva Papers on Risk and Insurance — Issues and Practice, 35(3), 450–472.
Commonwealth Fund. Catastrophic Health Spending Among the Privately Insured.
Health Care Cost Institute (HCCI). 2018–2022 Health Care Cost and Utilization Reports.
Kaiser Family Foundation (KFF). Health Costs: How Concentrated Are Health Care Expenditures?
Milliman. U.S. Stop-Loss Market Survey Reports.
Sing, M., Banthin, J. S., Selden, T. M., Cowan, C. A., & Keehan, S. P. (2006). Reconciling Medical Expenditure Estimates from the MEPS and the National Health Accounts. Health Care Financing Review, 28(1), 1–23.