برآورد تمایل به پرداخت برای درمان با استفاده از مدل چرخه زندگی پویای تقاضا برای بیمه سلامت

Medicare is the largest health insurance program in the US. This paper uses a dynamic random utility model of demand for health insurance in a life-cycle human capital framework with endogenous production of health to calculate the individual willingness to pay (WTP) for Medicare. The model accounts for the feature that the demand for health insurance is derived through the demand for health, which is jointly determined with the production of health over the life-cycle. The WTP measure incorporates the effects of Medicare insurance on aggregate consumption through effects on medical expenditures and mortality, and consumption utility of health. The model is estimated using panel data from the Health and Retirement Study. The average WTP or change in lifetime expected utility resulting from delaying the age of eligibility to 67 is found to be $ 24,947 in 1991 dollars ($ 39,435 in 2008 dollars). However, there is considerable variation in the WTP, e.g., in 1991 dollars the WTP of individuals who have less than a high school education and are white is $ 28,347 ($ 44,810 in 2008 dollars), while the WTP of those with at least a college degree and who are neither white nor black is $ 15,584 ($ 24,635 in 2008 dollars). More generally, the less educated have a higher WTP to avoid a policy change that delays availability of Medicare benefits. Additional model simulations imply that the primary benefits of Medicare are insurance against medical expenditures with relatively smaller benefits in terms of improved health status and longevity. Medicare also leads to large increases in medical utilization due to deferring of medical care prior to eligibility.

Medicare is the largest health insurance program in the US1 According to the Congressional Budget Office the expenditures on Medicare benefits were $ 212 million in 1999, which was 12.4% of federal outlays or 2.4% of GDP. In 2000 Medicare covered about 39.9 million or 13.9% of the US population of which about 34.4 million were aged 65 or older. It has been documented that in the absence of health insurance access to medical care, and medical utilization can be severely constrained for some individuals (e.g., (Card et al., 2004 and Benitez-Silva et al., 2005)). Hence, Medicare plays an important role in insuring the elderly against medical expenditure risk and improving their access to medical care. This paper uses a dynamic random utility model of demand for health insurance in a life-cycle human capital framework with endogenous production of health to calculate the individual willingness to pay (WTP) for Medicare. The model accounts for the feature that the demand for health insurance is derived through the demand for health. Moreover, individual decisions about health insurance, medical utilization and health related behaviors, and the consequent health outcomes (Grossman, 1972 and Phelps, 1973) are inter-related over the life-cycle. Hence the welfare effects of Medicare need to be considered in a dynamic life-cycle framework. The WTP measure computed in this paper incorporates the effects of Medicare insurance on aggregate consumption through the effects on medical expenditures and mortality, and the consumption utility of health. A life-cycle human capital model of endogenous decisions about health insurance, medical utilization, alcohol consumption, smoking and exercise is estimated using panel data from the Health and Retirement Study (HRS). In order to understand the effects of Medicare on individuals and their willingness to pay for its benefits, simulations from the model are used to infer its effects on medical utilization, out of pocket medical expenditures, and health outcomes by comparing outcomes under its coverage to those in a counter-factual situation in which the age of eligibility is delayed to 67. Given the adverse financial implications of imminent demographic changes2 this is one particular reform that has been proposed.3 There are four important reasons for analyzing the effects of Medicare in a life-cycle framework. First, such a framework helps evaluate the dynamic impact of Medicare. Medicare is a “mortality contingent claim” (Philipson and Becker, 1998) because individuals are entitled to its benefits conditional on survival to age 65.4 Thus it may alter survival incentives for those under 65 as it alters employment incentives (Rust and Phelan, 1997). Better coverage and thus potentially improved health and higher utility in old age, might induce individuals younger than 65 to on the margin increase behaviors (e.g., exercise) that decrease mortality risk (Philipson and Becker, 1998). Individuals may also defer medical care until they are eligible for Medicare at age 65, e.g., delay expensive treatments like a coronary artery bypass graft. Similarly the anticipated availability of generous coverage for the elderly may induce individuals younger than 65 to on the margin increase behaviors (e.g., smoking) that raise the risk of future medical expenditures and adverse health events. Second, given the life-cycle nature of health production, changes in individual health incentives and behaviors at younger ages may in turn affect health behaviors and outcomes after age 65. Third, the life-cycle nature of health production also implies that there will be dynamic selection (e.g., (Rosenzweig and Wolpin, 1995 and Cameron and Heckman, 2001)) in medical utilization, i.e., individuals whose past behaviors raise their current and future health risks will consume relatively more medical care over the life-cycle. In particular mortality will be endogenous to past behaviors causing selection through survivorship. Employing a life-cycle model provides a means to correct for dynamic selection in the empirical analysis in a manner similar to the method proposed by Heckman (1979). Fourth, such an analysis allows for an evaluation of the lifetime welfare effects of the program which for the reasons just outlined is important. Medicare is administered by the federal government and implemented almost uniformly across the US with little changes in its coverage since its inception. Medicare used to be called a state of the art insurance program for 1965 until December 2003 when the largest expansion of Medicare (Medicare Part D) was enacted. This was done primarily to provide prescription drug benefits as part of the Medicare Prescription Drug, Improvement, and Modernization Act (MMA) that went into effect on January 1, 2006. Using the variation generated by this expansion it is possible to study via reduced form methods the effects of this change in Medicare policy on behaviors and outcomes (e.g., (Duggan and Scott Morton, 2008)). However, such an analysis cannot be used to compute WTP for changes in the Medicare program.5 Using methods similar to those in this paper, McClellan and Skinner (2006) examine the value of Medicare in completing the missing market for health insurance for the elderly. They compute the parameters of a dynamic model using micro data. Using simulations they evaluate the value of Medicare insurance to its beneficiaries, and also Medicare reforms like progressive premiums and government vouchers. However, they do not analyze its dynamic life-cycle implications. Simulation based methods have also been used to examine the effects of Social Security benefits, e.g., French (2005), and van der Klaauw and Wolpin (2008). The benefits of Medicare are also analyzed by Lakdawalla and Bhattacharya (2005), who conclude in contrast to McClellan and Skinner (2006) that it is a very progressive program. A small recent literature examines the effects of Medicare on health behaviors and outcomes. Skinner et al. (2001) estimate the effect of intensity of Medicare expenditures on survival.6Dow (2001) and Lichtenberg (2002) examine the effects on utilization, morbidity and mortality. Card et al. (2004) analyze the effect on medical utilization, self-reported health, mortality, smoking, exercise and obesity. Decker (2005) examines effects on medical utilization among women, especially mammography and the chances of early detection of breast cancer. Finkelstein and McKnight (2005) examine the effects on mortality and medical expenditures. Though insightful, this literature is limited by its reliance on aggregate or repeated cross section data, e.g., it cannot account for individual-specific unobserved heterogeneity in estimating the effects of Medicare. This paper extends current research in various ways. In a unified framework using a single source of individual level panel data (i) it accounts for the inter-related life-cycle nature of medical utilization, health related decisions and health outcomes (in particular endogenous mortality) in assessing the effects of Medicare, (ii) it controls for individual-specific unobserved heterogeneity in the analysis, (iii) it examines the role of Medicare in insuring against medical expenditure risk, and (iv) it calculates the lifetime willingness to pay for Medicare. The model is estimated using panel data from the Health and Retirement Study. The WTP or change in lifetime expected utility resulting from delaying the age of eligibility to 67 is found to be $ 25,539 (in 1991 dollars) on average. However, there is considerable variation in the WTP, e.g., the WTP of individuals who have less than a high school education and are white is $ 27,252, while the WTP of those with at least a college degree and whose race is neither white or black is $ 20,012. More generally, the less educated have a higher WTP to avoid a policy change that delays availability of Medicare benefits. Additional model simulations imply that the primary benefits of Medicare are insurance against medical expenditures with relatively smaller benefits in terms of improved health status and longevity. Medicare also leads to large increases in medical utilization due to ‘stockpiling’ of medical care prior to eligibility. The rest of the paper is organized as follows. Section 2 describes the model and Section 3 the data. Section 4 discusses the estimation procedure. Section 5 presents estimation results and assesses model fit. Section 6 evaluates the WTP for Medicare benefits, and its impact on health related decisions and health outcomes, and Section 7 concludes.

This paper estimates the WTP for Medicare benefits using a dynamic random utility model in a life-cycle human capital framework with endogenous production of health. Medicare is the largest health insurance program in the US, and with the imminent demographic transition is likely to become even more important. Given these expected demographic changes and the consequent implications for the financial viability of the Medicare program, such a calculation is imperative to understanding its value to society, and in redesigning the program to fix its existing weaknesses. The model accounts for the feature that the demand for health insurance is derived through the demand for health, which is jointly determined with the production of health. The WTP measure incorporates the effects of Medicare insurance on medical expenditures, medical utilization and health outcomes. The model is estimated using panel data from the Health and Retirement Study. The change in lifetime expected utility resulting from delaying the age of eligibility to 67 is found to be $ 24,947 (in 1991 dollars) on average. However, there is considerable variation in the WTP. More generally, the less educated have a higher WTP to avoid a policy change that delays availability of Medicare benefits. Additional model simulations imply that Medicare mainly benefits the elderly primarily through insurance against medical expenditures with relatively smaller benefits in terms of improved health status and longevity. It also leads to large increases in medical utilization due to ‘stockpiling’ of medical care prior to eligibility. The results in this paper warrant a caveat that the model ignores the cost to society of subsidizing Medicare, i.e., the financing of medical services. However this would require a general equilibrium analysis that is beyond the scope of the paper. Inclusion of such costs may reduce the computed welfare benefits of Medicare. The model abstracts from savings or retirement decisions for reasons of computational tractability. On the other hand, inclusion of such decisions in the model would likely lead to larger estimated welfare benefits of Medicare. Additionally, the effects of Medicare on innovation in medical technology (Weisbrod, 1991 and Newhouse, 1992), which are not the focus of this research, are also potentially welfare enhancing.73 Future research that examines the role of these factors in assessing the impact of Medicare is warranted. Nevertheless, on the basis of this examination it can be concluded that Medicare meets its intended goals of insuring against medical expenditure risk and providing access to medical care for the elderly.