Skip to main content

Cost-Sharing Enhances Cost Control

September/October 2009, Vol 2, No 6 - Perspectives
Download PDF

H.E. Frech, III, PhD

In an article published in the April/May issue of the journal (AHDB. 2009;2:70-77), Haren and McConnell report the results of a recent survey (conducted in 2008) of large employers, employee benefit consultants, and managed care executives. The survey was focused on recent and expected increases in patient cost-sharing, and the article provides the aggregated responses of those surveyed.

The survey results are fascinating, showing that most employers and health plans have increased both premiums and consumer copayments (of many types) in the past 12 months. A substantial minority of employers and health plans has also reduced copayments, strategically, on particular services, based on specific clinical considerations or chronic conditions. Of note, both groups, but especially the employers, report their high levels of satisfaction with the overall results of higher copayments. One would think that this was a story of complex organizations positively adapting to changes in their environment as healthcare delivery becomes more costly. Indeed, the survey results show the importance of health and drug benefit design.

Oddly, in their commentary on the results, the authors seem to take the trends revealed in their survey as bad news rather than good news. In part, this is the result of confounding cost-shifting (ie, raising premiums) with cost-sharing (ie, raising copayments). Cost-shifting only affects the consumer's choice of whether to purchase insurance or which policy to purchase. It does not affect incentives to use more or less healthcare services. Cost-sharing, however, strongly affects incentives to use more or less healthcare. Indeed, this effect, called "moral hazard," is probably the best established empirical regularity in the entire field of health economics.

The concept of moral hazard originated in the insurance industry, where insurance that led to higher losses was generally viewed as a moral issue. (Dramatic examples, such as insurance-induced arson, fit this view fairly well.) That moral hazard in healthcare, which was simply a demand response to lower out-of-pocket prices, was first clearly explained byMark V. Pauly in 1968,1 and has since become a staple of health economics textbooks.2-4

The existence of moral hazard, of course, implies that more cost-sharing (higher copayments) will reduce the demand for healthcare services. As Pauly clearly showed, this is simply a result of movement up an ordinary demand curve. This concept lies at the heart of theoretical and empirical studies on demand for healthcare, which is probably the most carefully and completely studied demand topic in the entire field of economics, not merely in health economics. A recent survey done for the US Department of Defense lists 70 references on this topic, mostly empirical demand studies.5

The most famous research on healthcare demand sets the gold standard.6 This study, the RAND Health Insurance Experiment, which spanned the late 1970s through the early 1980s, randomly assigned consumers to different health plans, neatly (if expensively) side-stepping the statistical problem caused by the possibility of adverse selection (where sicker consumers choose more complete health insurance).6

The RAND study is one of the largest social science experiments ever conducted. It showed large and longlasting cost-savings from higher consumer copayments (and also from health maintenance organization utilization controls). Much of the savings, regardless of exactly how the copayments were designed and defined, occurred in hospitalization costs. For example, coinsurance imposed only on outpatient services reduced inpatient hospitalization expenses, as well as outpatient expenses. The effects on health status were nonexistent for most measures, small for correctable vision and periodontal health among the lowest-income consumers.6 Similar and consistent results are found in other studies, observational and natural experiments, in many countries, but no similar massive social experiments have been done.

This moral hazard, or demand effect, is well-known to insurers and actuaries. In a letter to Anthem Blue Cross/ Blue Shield of Maine, a Milliman actuary predicted the effect of various levels of copayments on utilization, ranging from 0.997 for the lowest cost-sharing to 0.730 for the highest cost-sharing—a 27% decrease from the least to the most cost-sharing.7 Even before the RAND study, actuaries and insurers took account of the costcontrolling effects of consumer copayments.8

There are only a few ways in which health insurers can control or influence costs: price bargaining with providers, utilization controls, and consumer copayments. It is well-established that employers and insurers believe that raising copayments reduces costs. The US healthcare system has well-known problems, such as the tax subsidy to employee health insurance and the high percentage of the uninsured. But, one advantage of that system is that the degree of utilization control, aggressive price bargaining, and benefit design are determined by the competitive interaction of many well-informed buyers and competing insurers. Thus, the system is subject to experimentation, innovation, evolution, and consumer choice.

[Online first, July 8, 2009.]


  1. Pauly MV. The economics of moral hazard: comment. Am Econ Rev. 1968;58:531-537.
  2. Phelps C. Health Economics. 3rd ed. Boston, MA: Addison-Wesley; 2002:100-172.
  3. Morrisey MA. Health Insurance. Chicago, IL: Health Administration Press; 2007:93-130.
  4. Zweifel P, Manning WG. Moral hazard and consumer incentives in health care. In: Culyer AJ, Newhouse JP, eds. Handbook of Health Economics. Amsterdam: Elsevier; 2000:410-459.
  5. Ringel JS, Hosek SD, Vollaard BA, Mahnovski S. The Elasticity of Demand for Health Care: A Review of the Literature and Its Application to the Military Health System. Santa Monica, CA: RAND Corp; 2002.
  6. Newhouse JP. Free for All? Lessons from the RAND Health Insurance Experiment. Cambridge, MA: Harvard University Press; 1996.
  7. Jack P. Burke. "Coverage Utilization Adjustments." [Letter to Bill Whitmore.] August 15, 2005. Accessed June 22, 2009.
  8. Phelps CE, Newhouse JP. Coinsurance, the price of time, and the demand for medical services. Rev Econ Stat. 1974; 56:334-342.
Related Items
Need for Expansion of Coverage for Narrowband UVB Phototherapy in Mycosis Fungoides and Sézary Syndrome
Agnes Kim, MS, Amelia Insley, BA, Lindsey West, MD, Heather Woodworth Goff, MD
December 2021 Vol 14, No 4 published on December 29, 2021 in Perspectives
Medication Optimization: Integration of Comprehensive Medication Management into Practice
September 2021 Vol 14, No 3 published on September 14, 2021 in Perspectives
Concerns Related to Benefits and Harms of Real-World Shared Decision-Making for Lung Cancer Screening
James L. Mulshine, MD, David F. Yankelevitz, MD, Bruce S. Pyenson, FSA, MAAA
Web Exclusives published on June 21, 2021 in Perspectives
Leading Population Health Efforts: The Power of Vowels
Richard G. Stefanacci, DO, MGH, MBA, AGSF, Anindita Banerjee, BSc, MBA
June 2021 Vol 14, No 2 published on April 5, 2021 in Perspectives
A Call for Innovation: We Need to Go Off Road to Ensure Medication Adherence Amid COVID-19
Javier Gonzalez, PharmD
June 2021 Vol 14, No 2 published on April 5, 2021 in Perspectives
Last modified: August 30, 2021