Health Care, Insurance, and the Contract Choice Effect
Dusansky, Richard, Koc, Cagatay, Economic Inquiry
It is well known that consumers anticipate some aspects of their medical care consumption--examples include obstetrical services, dental procedures, vision care, outpatient mental health services, and minor elective surgery--so that medical care demand anticipation influences the choice of health insurance coverage. In turn, the characteristics of the chosen plan set the degree of cost sharing and therefore the net price that the consumer faces, which is given by the percentage of cost sharing times the gross price of medical care. This net price influences the consumer's actual demand for the broad range of medical services. (1) In this way, insurance choice is influenced by particular anticipated utilization, and actual general utilization is influenced by the selected insurance plan; health insurance choice and medical service demand are interdependent.
Although this interdependence has been noted in the literature, few researchers have attempted to model it, (2) and none have explicated its implications for consumer demand behavior and the ensuing econometric analysis of medical care demand. We do so in this article. Using an intertemporal model that highlights the role of expectations in the insurance choice-medical care demand interaction, under uncertainty, we show that the medical care demand response to a change in its gross price consists of two separate components, the standard neoclassical demand effect and an insurance contract choice effect. When the gross price of medical care changes, the net price to the consumer is altered, for a fixed degree of cost sharing. However, a change in the gross price can prompt a change in the intended choice of health plan, causing a change in the degree of cost sharing. For example, if the gross price of medical care rises, our consumer may choose to pay a higher premium for an insurance policy that is more generous, one that has a lower cost-sharing percentage. In this case, the net price, which first increased exogenously, would now decrease endogenously.
The health insurance contract choice effect can play an important role in determining the consumer's observable demand response to a change in the gross price of medical care. The "standard" medical care demand curve could be less responsive because of it and econometric studies of the price elasticity of medical care that ignore it, which is typically the case in the literature, would produce estimates that are erroneous. Furthermore, although somewhat remote, the contract choice effect opens the door to a "perverse" outcome: a medical care demand curve that is upward sloping. This could occur when the contract choice effect is very strong.
II. THE MODEL
We consider an individual consumer who faces a single period two-stage planning problem. In the first stage, the consumer makes an ex ante choice of insurance plan. At the time this choice is made, he knows his income for the period, all prices, including the gross price of medical care, and his current health status. However, he is uncertain about his future health status. In the second stage, during which health status uncertainty is resolved, the consumer makes ex post decisions about purchases of medical care and other goods. In the first stage, the consumer chooses from a variety of health insurance policies with varying characteristics. The simplest has a deductible of some amount (e.g., $250 per person, $750 per family), a percentage cost share (e.g., consumer pays 20%, the insurance provider 80%, of each episode) and a cap on total dollar outlay, beyond which the insurer pays 100% of remaining expenses. Other policies place more emphasis on co-payments (e.g., consumer pays $20 to a physician for an office visit, the insurance provider pays the balance of the negotiated cost) and some policies combine all of these elements: co-pays for office visits, 20-80% cost shares for hospital stays, different mixes of percentage cost shares depending on services that are out-of-network, in-network but not preferred provider, and so on. …