"Producing Health” treats supply, prices, and patient effort as economic inputs. But a health system patients cannot afford, reach, or navigate cannot turn capacity into better health.
HHS Chief Economist and Regulatory Officer Casey Mulligan, previously affiliated with the conservative health policy think tank Paragon Health Institute, recently released a paper on the HHS website titled “Producing Health: Capability, Capacity, and Competitive Discovery in the Health Economy.” The paper opens with a challenge worth seriously considering - health policy ought to prioritize health outcomes rather than focusing exclusively on spending or insurance enrollment. However, Mulligan’s arguments are simplistic and ignore the unique economics of health care and patients' actual experiences.
WHERE THE PAPER IS RIGHT
Mulligan correctly states the importance of supply and capacity. Insurance coverage is of limited value if patients cannot access the resources required for health: physicians, home health workers, pharmacies, or hospital beds. Administrative burdens additionally divert clinical time from patient care. Licensing, scope-of-practice, and site-of-care regulations can restrict entry into the health care workforce. We should address these characteristics of our health care system with solutions calibrated to improve patient outcomes while considering economic realities. Both past and emerging innovation models address these challenges and include value-based payment systems, alternative staffing models, improved scheduling, telehealth, home-based acute care, and new sites of service.
Mulligan is also correct in recognizing patients and families as contributors to health production.
However, I disagree on several points.
WHERE MULLIGAN’S ARGUMENTS FAIL
1. Coverage belongs in the health production function.
Mulligan argues that insurance functions primarily as a method of payment rather than an endpoint; however, this view overlooks that health insurance coverage determines whether most patients can access the inputs necessary for health. While insurance coverage alone does not produce health, I contend that Mulligan dismisses its significance too easily. For a significant proportion of people, but especially individuals with low incomes, disabilities, chronic illnesses, or limited savings, insurance is a critical determinant of access to care. Insurance coverage influences whether care is financially attainable, whether prescriptions can be filled, and whether illness results in medical debt.
In the New England Journal of Medicine, Baiker et al. found that the Oregon Medicaid experiment showed that those who gained coverage received more preventive care, more detection and treatment of diabetes, experienced lower depression, and saw near elimination of catastrophic out-of-pocket medical spending. A later Quarterly Journal of Economics study by Miller et al. found that ACA Medicaid expansion was associated with a 9.4 percent decline in annual mortality among near-elderly adults most likely to gain coverage. These studies provide strong evidence that insurance coverage can affect health by changing access.
2. “Patient capability” cannot be separated from poverty, disability, and the conditions under which people live.
The paper identifies transportation, adherence, digital access, family support, and navigation skills as productive inputs. I agree with this assessment, but we must also consider the resources patients have to provide these inputs.
Many people face trade-offs when making health care decisions, often because of barriers embedded in our health care system. An hourly worker risks losing wages to make a medical visit. A person with a physical disability may face an inaccessible exam table or unreliable transportation. A patient choosing between rent, food, and a prescription faces a financial dilemma. HHS’s Healthy People 2030 identifies insurance, transportation, income, time off work, geography, and disability among the barriers shaping access to care. CDC data also documents substantial access barriers among adults with disabilities.
Any economic model that incorporates patient and family effort must also account for the resources necessary to enable that effort.
3. Health care prices frequently reflect market power as well as scarcity.
Mulligan assigns significant importance to price signals in guiding market entry, investment, and capacity. This is largely untrue for health care, especially necessary and life-saving services. It applies well to discretionary care such as cosmetic procedures, alternative and complementary medicine treatments, wellness products, and upgraded medical services, where consumers can take time to compare prices and decide whether to purchase them. This logic applies best when competition is robust, and purchasers are well informed. The health care sector, in general, meets neither condition.
Early work on medical economics by Kenneth Arrow showed that health care differs from ordinary markets because patients cannot predict when they will become ill, how serious an illness will be, what treatment they will need, or what that treatment will cost. Additionally, physicians typically have far more medical knowledge than patients, making it hard for patients to judge which services are necessary, whether a recommended treatment is appropriate, or whether the quality of care is good. Arrow identified this combination of uncertainty and unequal access to information as a central feature of health care markets.
Critically, MedPAC’s 2024 report to Congress describes evidence linking hospital and health system consolidation within markets to higher commercial prices.
In such markets, elevated prices may reflect bargaining leverage or excess profits or margins ("economic rent"). Lowering administered prices can reduce excess margins resulting from market power without necessarily causing the supply reductions predicted by a simple competitive model.
4. The effect of administered prices is an empirical question.
Mulligan correctly identifies an important economic reality, or one often used to justify high drug prices: reducing expected pharmaceutical revenues can reduce investment in research and development, potentially keeping some drugs from reaching the market. This potential cost warrants serious consideration. However, the magnitude of this effect matters.
The Congressional Budget Office (CBO) estimated that the drug provisions enacted in 2022 would reduce global revenues from new drugs by roughly 1 to 3 percent. Its model projected about 13 fewer drugs reaching the market over 30 years, out of roughly 1,300 expected new drugs. In other words, CBO predicts an innovation effect, but a very modest, if not insignificant, one relative to the total flow of new products.
Policymakers should incorporate that estimate into a broader and detailed analysis. Lower drug prices and reduced out-of-pocket expenses make existing medicines more affordable for individuals who might otherwise delay, ration, or forgo treatment. We can acknowledge a potential small reduction in future drug development while also seriously considering the immediate benefits of lower prices for current patients.
At this point, Mulligan’s framing is inadequate. Sure, prices influence incentives in general, but that does not mean every reduction in producer revenue leads to socially undesirable outcomes. We must objectively evaluate the magnitude of the innovation response, the types of innovation affected, and whether these losses are outweighed by the benefits of lower prices, increased affordability, and improved access.
The CBO provides a template for quantifying this tradeoff. This approach exemplifies the type of health economics analysis that policymakers should be doing.
5. Public financing and decentralized experimentation can coexist.
Mulligan worries that when one central system controls payments, a bad policy decision can affect the whole market. However, U.S. programs have clearly shown that government-funded payment systems can still allow states and local areas to try different approaches and test what works.
The CMS Innovation Center tests payment and delivery models, uses independent evaluations, modifies or ends weak models, and can expand models that meet statutory standards. For example, Maryland has combined administered hospital payments with global budgets, care redesign, and provider-level experimentation. CMS says its Total Cost of Care model was structured to generate more than $1 billion in Medicare savings through 2023 while encouraging coordination and prevention.
These programs demonstrate that public financing does not inherently preclude learning, variation, or experimentation.
6. Distribution and financial protection belong in the welfare analysis.
A $100 copayment impacts poorer patients more than those in the top 1% of income. Moreover, factors beyond income, such as disability, chronic illness, caregiving responsibilities, and accumulated medical debt, change the significance of this cost.
The RAND Health Insurance Experiment found that cost sharing reduced use of both effective and less effective care. Among the poorest and sickest participants, free care produced better outcomes on several measures, including blood pressure control for people with hypertension. More recent research has also found cost-related medication nonadherence was most common among people with fewer financial resources.
For patients with serious or chronic illnesses, delayed care can result in poor outcomes, including death. These consequences, along with the distribution of financial risk, are essential components of the economic outcome.
7. The COVID-19 discussion needs stronger causal identification.
Mulligan highlights excess non-COVID deaths during the pandemic and argues that disruptions to work, treatment, social life, family routines, and other forms of self-management may have contributed to these excess deaths. That assertion has face validity but is simplistic. The evidence he presents is largely conjecture and does not isolate the effects of pandemic policies from the many other forces operating at the same time.
Rigorous studies point to a considerably more complicated picture. These studies have found evidence of multiple mechanisms, including direct but unrecognized effects of infection, health-system strain, disruptions in care, and increases in some external causes of death such as drug overdoses, accidents, and homicide. One national analysis estimated that 84 percent of all-cause excess mortality was statistically attributable to the direct effects of infection. In contrast, indirect effects were more important among younger people and for external causes. These studies found associations between stricter interventions and some forms of indirect mortality, so those effects should not be dismissed entirely. Still, the evidence points to a more complicated set of factors.
Therefore, there is no evidence that public-health restrictions were the principal cause of excess non-COVID mortality. That causal claim requires a much more careful accounting of competing explanations than the descriptive mortality patterns alone can provide.
BOTTOM LINE
Mulligan’s paper reads like an argument for replacing centralized health policy with price signals, competition, and decentralized markets. The problem is that health care does not behave like the competitive market his framework assumes, and patients do not experience it that way. Humans are complex and live in highly variable environments that impact health. Patients are not like the widgets in economics textbooks.
Mulligan’s most compelling argument is his call for health policy to prioritize supply considerations. My disagreement centers on which factors belong in the production function and what barriers prevent patients from using available supply.
This is also a values question. Do we believe people should be able to obtain needed health care regardless of their income, health status, or circumstances? Most Americans believe the federal government has a responsibility to ensure that everyone has health coverage.
But the stakes extend beyond individual access to health care. The health of the population affects workforce participation, productivity, economic resilience, and national security, consequences that Mulligan’s paper does not adequately consider. Health is therefore not simply the product of individual choices and market transactions; it is also a matter of national interest. That reality makes a coherent national health policy, and a meaningful federal role in financing, access, public health, and system oversight, essential.
So, focusing on health outcomes must take into consideration a complex set of factors that impact patients’ ability to achieve good outcomes, such as insurance coverage, affordable medications, transportation, accessible facilities, stable housing, caregiver support, competition policy, public investment, and protection from catastrophic financial risk; all affect whether patients can access care.
Also, a health system may offer advanced technology and highly skilled clinicians and institutions, yet still fail individuals who cannot afford the basic contributors to health outcomes: prescriptions, access to facilities, paid leave, or finding providers who accept their insurance.
To be comprehensive, a health production framework must address three questions simultaneously:
Do we have enough capacity?
Can patients access care and afford it?
Who bears the health and financial risks?
SOURCES
Casey B. Mulligan, “Producing Health: Capability, Capacity, and Competitive Discovery in the Health Economy,” HHS ASPE, Aug. 3, 2026.
Kenneth J. Arrow, “Uncertainty and the Welfare Economics of Medical Care,” American Economic Review 53(5), 1963.
Katherine Baicker et al., “The Oregon Experiment: Effects of Medicaid on Clinical Outcomes,” New England Journal of Medicine 368, 2013.
Sarah Miller, Norman Johnson, and Laura R. Wherry, “Medicaid and Mortality: New Evidence From Linked Survey and Administrative Data,” Quarterly Journal of Economics 136(3), 2021.
U.S. HHS, Healthy People 2030, “Access to Health Services” and “Social Determinants of Health.” https://odphp.health.gov/healthypeople/priority-areas/social-determinants-health/literature-summaries/access-health-services
CDC, “Disability Impacts All of Us” https://www.cdc.gov/disability-and-health/articles-documents/disability-impacts-all-of-us-infographic.html
Medicare Payment Advisory Commission, March 2024 Report to the Congress, discussion of provider consolidation and commercial prices. https://www.medpac.gov/document/march-2024-report-to-the-congress-medicare-payment-policy/
CMS Innovation Center, “Evaluations” and model evaluation framework. https://www.cms.gov/priorities/innovation/key-concepts/evaluations
CMS, “Maryland Total Cost of Care Model.” https://www.cms.gov/priorities/innovation/innovation-models/md-tccm
RAND, “The Health Insurance Experiment,” research brief summarizing the RAND HIE. https://www.rand.org/pubs/research_briefs/RB9174.html