This sample Health Insurance Research Paper is published for educational and informational purposes only. Like other free research paper examples it is not a custom research paper. If you need help with writing your assignment, please use research paper writing services and buy a research paper on any topic.
Insurance can be defined as a contract where the subscriber makes regular payments to the insurer in exchange of the promise of indemnity against specified risks. In health insurance, the specific risks covered are the financial costs of the treatments needed after an illness or an accident. The economics of health insurance is well developed and identifies a series of challenges, which can seriously jeopardize the proper functioning of health insurance markets based on private contracts. It is indeed rare to observe health systems that do not rely on some degree of government involvement. In this research paper, the theoretical and empirical challenges to health insurance markets will be described and discussed. Subsequently, we will provide a short description of the characteristics of the most significant health systems around the world, linking these with the issues emerging from the economic analysis.
- The Market for Health Insurance: The Basic Framework
- Health Coverage and Insurance Markets
- Preexisting Conditions and Risk Selection
- Asymmetric Information
- Adverse and Advantageous Selection
- Moral Hazard
- Supply Side
- Health Insurance around the Globe
Insurance can be defined as a contract where the subscriber makes regular payments to the insurer in exchange of the promise of indemnity against specified risks. In health insurance, the specific risks covered are the financial costs of the treatments needed after an illness or an accident. The health insurance contract can be signed by either the individual subscriber or a sponsor, for example, the employer or a community organization. The insurer can be a private for-profit organization, a private not-for-profit organization, or the government (Claxton, 2002).
A health insurance agreement usually specifies the value and the form of the following:
- Premium: the monthly or annual regular payment to be made by the subscriber;
- Coverage: the type and the limits of the medical expenses that can be reimbursed;
- Deductibles: the limit below which medical expenses are not reimbursed. The insurance pays the difference between the medical bill and the deductible;
- Copayment or Coinsurance: the out-of-pocket payment to access an insured medical service. In copayments, this amount is fixed; in coinsurance, it is a percentage of the medical bill; and
- Network providers: the list of accredited providers (for example, hospitals, clinics, or general practitioners) selected by the insurer. In order to get coverage, medical services must be provided by the institutions included in the network.
The economics of health insurance is well developed and identifies a series of challenges, which can seriously jeopardize the proper functioning of health insurance markets based on private contracts. It is indeed rare to observe health systems that do not rely on some degree of government involvement. In the following, the theoretical and empirical challenges to health insurance markets will be described and discussed. Subsequently, we will provide a short description of the characteristics of the most significant health systems around the world, linking these with the issues emerging from the economic analysis.
The Market for Health Insurance: The Basic Framework
Why do people buy health insurance? When it comes to important issues like health and health-related expenditure, people are mainly risk averse (Arrow, 1970). Formally, a rational individual is risk averse if she prefers to gain X for sure rather than participating in a lottery whose expected value is X. In other words, people tend not to like risks and are willing to give up part of a certain income in order to avoid it.
The definition of risk aversion applies directly to the demand for health insurance. Imagining the future health situation of a subject as a lottery, health insurance requires that a person should be willing to give up part of her own income (insurance premium) in the present in exchange of a (partial) insulation from the risk of getting ill and incur monetary, physical, and psychological losses in the future. This can be a chosen option only if the person is risk averse.
In the baseline economic model of health insurance markets, risk averse individuals are assumed to be perfectly informed about the parameters driving their own choice. They know about their probability of getting sick and they know what losses they will incur in that case. Indeed, the assumption of perfect information of the insurance subscriber is common in many studies (Phelps, 2010). Similarly, the baseline model implicitly assumes that risks are correctly perceived. Not only do people have information about the risks, but they also interpret it correctly. In other words, in the baseline framework, problems of too much or too little confidence or any other source of misperception are ruled out a priori.
Although rational individuals choose to insure against health risks, market outcomes and equilibria depend also on the supply side. Economic theory has clarified that three main assumptions are needed for the provision of insurance through private markets to be efficient. First, administrative costs or any other cost outside the agreed reimbursements (these costs are also known as loading factors) should be small. Second, firms should be risk neutral. Last but not least, firms should compete on premiums (Einav and Finkelstein, 2011). This in turn assures that premiums are actuarially fair, i.e., they reflect the actual risk profiles of the insured subject. In a world where all these three assumptions hold true, firms cannot charge unnecessarily high (unfair) premiums, because other firms would undercut them, gaining market shares and profits. At the same time, even in competitive markets, too many loading factors might force firms to charge unfair premiums, which could turn out to be too high for many potential subscribers.
The optimal insurance coverage is what emerges from a bargaining process where all the above-mentioned assumptions hold true (Phelps, 2010). It is thus possible to show that risk averse people, perfectly informed about their private risks, will choose full coverage (i.e., a reimbursement of all the medical expenses they will incur if they get ill) in exchange of the fair premium offered by perfectly informed firms working in a competitive environment. In this baseline model, markets are efficient and people get their fair deal.
Health Coverage and Insurance Markets
The baseline model is a useful benchmark. It provides a precise definition of the assumptions needed for health insurance markets to be efficient. However, it is realistic to expect that in many real world situations some of these assumptions would fail. Since the 1970s, the economic literature has gained substantial understanding on the functioning of health insurance markets when more realistic frameworks are used.
This literature has become very influential in the last decade and has set the basis for many policy interventions in the field of health systems design and health insurance regulation. In the following, the most important issues emerging from this literature are briefly explained.
Preexisting Conditions and Risk Selection
One of the most pervasive problems related to a voluntary market for health insurance comes from the simple observation that the premium should reflect the individual risk. Age-related illnesses or chronic diseases (preexisting conditions) can become very difficult to cover in an insurance market, since the premium required to insure against an almost sure event (i.e., needing some sort of health treatment) is too high. This situation can create two related types of problems (Cutler and Zeckhauser, 2000).
First, the market can fail altogether. In the most extreme cases, the probability of getting ill is so close to one that the fair premium is equal to the reimbursement. This means that the insurance contract does not exist. In practice, this failure translates into the exclusion of people with preexisting conditions from the market for health insurance. In this sense, some have suggested that the choice of insurance is mainly intertemporal and that people should buy health for their lifetime (always before a health condition appears) instead of annually (Cochrane, 1995). However, these types of contract are not easily found in real world markets. Second, it should be noted that the insurance choice is always limited by resources availability. A poor is less likely to buy health insurance and access health care than a person with high income (e.g., Clark et al., 2013; Hsiao, 1995; Preker et al., 2007). Moreover, budget constraints are more likely to ‘bite,’ excluding from the market the high-risk people, which are also poor.
It is important to stress that the high risks are the people that most need access to health care and health treatments. Hence, the system fails exactly those it is supposed to serve. The problems raised by preexisting conditions and aging in health insurance markets are then particularly important and are well-recognized in most health systems around the world.
The baseline model assumes that agents have perfect information about virtually everything. However, in reality one party can have more information than the other, i.e., information can be asymmetric. The literature on asymmetric information, started in the 1970s, has developed a great body of theoretical and empirical evidence. Here we define the most common typologies of asymmetric information problems commonly associated to health insurance markets.
Adverse and Advantageous Selection
In health insurance, adverse selection occurs whenever the asymmetric distribution of information causes markets to screen out the good risks, selecting only the ‘bad’ ones. Adverse selection is a phenomenon intrinsic to any insurance system. It can happen mainly for two reasons: the insured knows something that the insurance does not or they both have the same information but the insurance company is not allowed to use it in setting the premium, due to antidiscrimination laws or to regulatory limits to the rating system.
The problem of adverse selection was firstly identified in markets with asymmetric information on products’ quality (Akerlof, 1970). Subsequently, the analysis was extended to the health sector in a very influential article that set up the basic formalization of the issue (Rothschild and Stiglitz, 1976), which proceed as follows: imagine that there are now two types of persons, one high and one low risk, but that the insurer cannot distinguish one from the other. For example, subscribers can be either smokers or nonsmokers, with the insurer unable to observe their true type. In this case, the problem is that if the insurer offers two different contracts, one for the low risk (nonsmokers) and the other for the high risks (smokers), the high risk can pretend to be low risk and get a better deal. Since the insurer cannot tell one from the other, this situation is not sustainable.
One possible solution could be to offer some form of average premium to everyone (pooled premium). However, for low risks this premium is unfair. They are thus less likely to accept the deal and some of them will exit the market. Since the remaining pool will include relatively more high risks than before, the insurance will be forced to increase further the average premium in order to balance the exits. This will force some other low risk types to turn down the offer. The process could potentially go on until only high risks remain in the market, i.e., adverse selection. This phenomenon is also known as the ‘death spiral’ in health insurance markets (Cutler and Zeckhauser, 1998).
Under certain conditions, the failure of pooling contracts could be overcome by offering contracts through which buyers self-select themselves (i.e., separating equilibrium). That is, the insurer can offer a combination of premium-coverage that is the best option only for the high risk and another that is optimal only for the low risk. By picking one option or the other, subscribers reveal their true risk profile. This could potentially represent an equilibrium. It turns out that these types of contract imply offering to the low risks less than the full coverage in exchange of a fair premium (Rothschild and Stiglitz, 1976). That is, low risk profiles agree to give up part of the coverage in exchange of a fair premium, while the high risks prefer to pay a higher premium in order to be fully reimbursed. With respect to the full information scenario, the low risks would be worse off, since they pay the same but they get a lower coverage. This is the loss associated to adverse selection when the market does not fail. In addition, however, it is also possible that if the proportion of low risk buyers is higher than a certain threshold, no separating equilibrium exists either, so that only the high-risk profiles remain in the market.
The existing empirical evidence on adverse selection is contradictory (Cohen and Siegelman, 2010). According to the method and the data analyzed, some studies find a positive relation between risk profiles and insurance coverage (Sapelli and Vial, 2003; Spenkuch, 2012), others do not observe such a correlation (Cardon and Hendel, 2001; Fang et al., 2008; Finkelstein and McGarry, 2006). Evidence of separating equilibria through underprovision of insurance to low risks has also been found (e.g., Buchmueller and DiNardo, 2002).
There are different responses to the challenges posed by adverse selection. Two are worth mentioning here. First, if the insurance company can acquire the missing information after the claim has been filled, then an ex-post control can be imposed and defined in the contract. In this case, however, the company must be allowed to use the information about the individual claimant. This applies, for example, to smoking. Second, the regulator can introduce an individual mandate on health insurance. This is the approach taken by many countries such as the Netherlands or Switzerland and is one of the main pillars of the Obama health care reform known as the Affordable Care Act (ACA) in the US (US Congress, 2010). In this case, since all the residents are obliged to subscribe to some form of insurance, good risks are automatically included into the system and average premiums might reduce (e.g., Hackmann et al., 2013).
In recent years a literature has emerged pointing out an opposite phenomenon to adverse selection: advantageous selection (De Meza and Webb, 2001; Hemenway, 1990). Advantageous selection arises whenever those who are more risk averse both buy more insurance coverage and have lower risks. In other words, when cautious and risk averse individuals are also more likely to sign up for insurance. Advantageous selection shifts the attention from the actual individual risk profiles to their risk aversion, which is obviously not known to the insurer. From the point of view of the insurance company, however, this type of selection is profitable. Although the empirical evidence so far seems to confirm advantageous selection in the US, for example, in long-term care insurance (Finkelstein and McGarry, 2006) or in the Medigap (Fang et al., 2008), it is important to note that adverse and advantageous selection can, and probably do, coexist in these markets (Einav and Finkelstein, 2011).
Moral hazard occurs whenever people take more risks or demand more health care just because they signed up for health insurance (Arrow, 1963; Folland et al., 2006). Broadly speaking, moral hazard can arise before (ex-ante) or after (expost) a claim is filed. In the former case, people would reduce their preventive effort (for example, paying less attention or doing less physical activity) because they feel ‘safe,’ i.e., insulated from the losses they would face in case the negative event materialized. In the ex-post moral hazard, the claimants would increase their demand for health care services beyond what they would be willing to buy if they were not reimbursed. In other words, they demand excessive medical services just because they are (partially) free. Note that moral hazard is based on the assumption that the insurer cannot observe the insured’s actions. This is why it is often referred to as hidden action.
Ex-post moral hazard is considered a serious problem for voluntary private insurance (Pauly, 1968; Zeckhauser, 1970). Indeed, when confronted with an excessive demand of reimbursements, insurance companies might raise their premiums above what is considered fair by some subscribers. Government-based health systems also suffer ex-post moral hazard, often in the form of too high demand for pharmaceuticals and medical diagnostics. Although ex-post moral hazard is a pervasive phenomenon in health insurance markets, solutions do exist and are widely applied. Since the problem originates in the free access to health care services, the solution is simply to make the patient share part of the medical costs. This is usually done through deductibles, copayments, and coinsurance, as defined above (Folland et al., 2006).
Empirically, moral hazard has been tested in one of the first large-scale randomized study in social science, the RAND Health Insurance Experiment (Brook et al., 1983; Manning et al., 1987). Between 1974 and 1982 a total of 5809 people were randomly assigned to different insurance plans: free health care, individual deductible plan (family pays 95% of costs under $150), a series of intermediate coinsurance plans (coinsurance of either 25, 50, or 95% until the threshold limit of $1000 per family). These subjects were then followed for the whole observation period. Results showed that free-care insured were more likely to go to the doctor and to be hospitalized. At the same time, health consequences were negligible. This can be seen as a strong support for the theory of moral hazard.
These results have been very influential throughout the 1990s in the health policy debate. Serious criticisms directed toward both the method and the interpretation of the results have, however, been advanced in the literature. Two such criticisms are of particular importance.
First of all, what might look like opportunistic behavior (demanding more medical care when it is freely accessible) could actually be the consequence of the income transfer from the healthy to the sick implicit in any health insurance mechanism (Nyman, 2003, 2001, 1999). In other words, one person might ask for new medical treatments not because medical care is free or cheaper, but simply because without insurance the same treatment would not be affordable. Indeed, increasing accessibility by making treatment affordable should be one of the main purposes of health insurance. Also, medical treatments are always related to personal costs that are increasing with the complexity and the invasiveness of the intervention. In these cases, people would not ask for medical treatments if they did not actually need them. It is thus important to distinguish between different types of services. For important and expensive treatments, an increase in usage following insurance coverage might signal an improvement in accessibility to health care rather than opportunistic behavior.
A second criticism is related to health care access for preventive reasons (Ellis and Manning, 2007; Seog, 2012). Consistent with the evidence, higher copayments imply lower demand for diagnostic and preventive medical services. This means that people might refrain from seeing the doctor or run diagnostic tests if the cost is too high. Studies generally focus on the savings for the insurance from this contraction in demand and interpret this phenomenon as a reduction of moral hazard. However, for patients who have a condition, this would represent just a postponement of treatment. In general, high copayments might inhibit early detection of illness. Since typically it is cheaper and safer to treat conditions earlier rather than later, what might appear as a saving might turn out to be an increase in the overall expenditure. For example, analyzing 172 Medicare plans in the US, 18 of which raised copayments for ambulatory care during the observation period, a study found an average reduction (between 2001 and 2006) in the number of ambulatory visits for the group whose copayment increased (Trivedi et al., 2010). This is consistent with moral hazard. On the other hand, for the same group the hospital admissions increased by 2.2%. Overall, although copayment saved $7150 per capita in reduced demand for ambulatory care, it increased by $24 000 the expenditure for inpatient care.
Overall, moral hazard is unavoidable in health insurance. It affects both private and public systems. In theory, the problem can be ameliorated by using deductibles and copayments. However, caution should be taken in interpreting any ‘excess’ demand as a sign of opportunistic behavior, possibly avoiding overreactions and extreme policy implications.
The main assumption for the supply side in the baseline model is that firms compete on prices. In other words, they try to get people to subscribe to their policies by offering the best price for any given coverage. This assumption is particularly problematic in reality and often requires some sort of government intervention at various levels.
The first issue to be considered is risk selection, also known in the economic literature as cream skimming: the selection by the insurance providers of those consumers expected to be more profitable given their risk profiles (Barros, 2003; Ellis, 1998). Within the same premium-risk group, profitability is inversely related to the risk profile of the subscribers: good risks are profitable and firms will try to attract them. Even when they are not allowed to directly deny access to insurance to the bad risks, for example, because they are obliged by law to accept subscribers independently from their risk profile, they still can set up contracts and offer services that are typically more appealing for the good risk profiles. Competition in this case can be problematic to manage. One solution is risk adjustment (Shen and Ellis, 2002; Van de ven and Ellis, 2000). Information can be used to subsidize firms for each high risk in the pool of the enrollees. Hence, pools with proportionally more high risks will receive more subsidies. Subsidization can come from either the government or other firms with advantageous risk profiles among their pool. If correctly set, these subsidies can significantly reduce the incentive for risk selection.
A second issue arising from competitive health insurance markets is related to the presence of administrative costs and other loading factors. In these cases, firms would need to increase premiums above the fair level in order to recover the additional costs. This would alter significantly the picture emerging from the baseline model (Einav and Finkelstein, 2011). Administrative costs in competitive markets imply a duplication of resources and procedures, which reduces the efficiency of a free market health insurance system.
Competition might even be difficult to achieve. The literature in industrial economics is full of examples of markets where firms either do not compete with each other because they can collude or do not find it profitable to start price wars (Martin, 1994; Tirole, 1988). Much depends on how easily new and small firms could enter the market and gain market shares by offering low premiums. For example, the presence of switching costs, i.e., fixed and psychological costs people have to pay when they change insurer, do indeed represent an obstacle for competition in health insurance markets (Farrell and Klemperer, 2007; Frank and Lamiraud, 2009).
Health Insurance around the Globe
Both the theory and the empirical observations have clarified that universal health coverage, i.e., universal access to health care, cannot be achieved through market mechanisms only. Information and economic problems are too pervasive to the system. Government interventions, whether in the form of regulation, redistribution, or direct provision of insurance and health care, characterize the organization of health systems everywhere around the world.
Countries generally have developed different and complex systems to provide health care to their citizens. Health care systems are often classified in many different ways depending on the types and numbers of factors researchers might find relevant to consider (Beckfield et al., 2013). A health system, however, is determined by the interaction of different actors, such as insurers, patients, and medical providers. Hence, a precise taxonomy of health systems goes beyond the purpose of the present research paper, which focuses on insurance only. In this sense, a common classification of health systems from the point of view of the insurance dates back to the early 1970s (Anderson, 1972) and identifies three general approaches: the Bismarck model, the Beveridge model, and the voluntary/ private insurance model.
The voluntary health insurance model is essentially the one described so far, where people decide whether to buy health insurance and companies supply a variety of contracts suited for all the different risk profiles. It thus sees health insurance as a good to be exchanged in the market, as any other good. To the extent that universal health coverage is a concern within a country, government interventions in this model usually take the form of the introduction of some form of public health insurance for those high risks for which the system might fail. The typical examples in this sense are Medicare (public insurance for those over 65) and Medicaid (public health insurance for the poor) programs in the US. In addition, more recent reforms, such as the ACA, have tried to tackle adverse selection and the exclusion of preexisting conditions by imposing an individual mandate for health insurance for every US citizen and by requiring insurance providers to compulsorily accept high risks (Austin and Wetle, 2012).
Interestingly, politicians have grasped the weaknesses implicit in the voluntary health insurance mechanisms long before the economic theory could formally explain them. In particular, the two most common names quoted when describing the alternative health systems are Bismarck and Beveridge (Lassey et al., 1997).
Otto von Bismarck (1815–1898), a Prussian statesman and Chancellor, first understood the importance of welfare programs for maintaining order and status by gaining the support of parts of the working class. Often described as a conservative and nationalistic politician, he introduced, with the support of the German industry, old age pensions, unemployment insurance, and health insurance. In Bismarck model, health coverage is distributed through employers related to Social Health Insurances (SHIs) (Burau and Blank, 2006). Hence, health coverage can change substantially according to the position in the labor market. SHIs are financed through payroll contributions levied on employers and employees and are in competition with each other. The private sector can thus have an important role within the system, although coordination among different players is a key for success. In particular, competition is usually heavily regulated and the government can keep a strong redistributive role. In general, workers and patients can maintain the power to choose among different insurance providers.
William Beveridge (1879–1963) was a British economist and politician who, in 1942, wrote a very influential report on welfare and unemployment. The so-called ‘Beveridge report’ set the scene for the establishment of the National Health Services under the Labor government of Clement Atlee in 1945. According to Beveridge, health and social insurance in general should be financed through a contribution by all working people in the country, something that now is referred to as general taxation. Originally, in these systems health care was meant to be provided entirely by public hospitals. Indeed, sometimes in the international literature the Beveridge model is viewed in this purest version. However, as clarified by the quasi-market reforms introduced by Margaret Thatcher at the end of the 1980s, Beveridge systems can separate the purchasing authority, which must remain the taxed-based public insurance, from the providers, which can be either private or public.
Table 1 Bismarck and Beveridge approaches to health insurance
In Table 1 the main characteristics of the Bismarck and Beveridge models are summarized. Examples of countries are provided. Note that although in principle the distinction between the two types of systems is clear, in most Bismarckian countries the government actually intervenes in various ways to assure universal health coverage and sometimes to fix directly the reimbursement rates of medical services. Indeed, according to some scholars, the two systems are meant to converge in many different dimensions (e.g., Stevens, 2009).
- Akerlof, G.A., 1970. The market for “Lemons”: quality uncertainty and the market mechanism. Quarterly Journal of Economics 84, 488–500.
- Anderson, O.W., 1972. Health Care: Can There Be Equity?; the United States, Sweden, and England. John Wiley & Sons Inc., New York.
- Arrow, 1963. Uncertainty and the welfare economics of health care. American Economic Review 53, 947–973.
- Arrow, K.J., 1970. Essays in the Theory of Risk-bearing. North-Holland, Amsterdam.
- Austin, A., Wetle, V.L., 2012. The United States Health Care System: Combining Business, Health, and Delivery. Pearson, Upper Saddle River.
- Barros, P.P., 2003. Cream-skimming, incentives for efficiency and payment system. Journal of Health Economics 22, 419–443.
- Beckfield, J., Olafsdottir, S., Sosnaud, B., 2013. Healthcare systems in Comparative perspective: classification, convergence, institutions, inequalities, and five missed turns. Annual Review of Sociology 39, 127–146.
- Brook, R.H., Ware, J.E., Rogers, W.H., Keeler, E.B., Davies, A.R., Donald, C.A., Goldberg, G.A., Lohr, K.N., Masthay, P.C., Newhouse, J.P., 1983. Does free care improve adults’ health? New England Journal of Medicine 309, 1426–1434.
- Buchmueller, T., DiNardo, J., 2002. Did community rating induce an adverse selection death spiral? Evidence from New York, Pennsylvania, and Connecticut. American Economic Review 92, 280–294.
- Burau, V., Blank, R.H., 2006. Comparing health policy: an assessment of typologies of health systems. Journal of Comparative Policy Analysis 8, 63–76.
- Cardon, J.H., Hendel, I., 2001. Asymmetric information in health insurance: evidence from the national medical expenditure survey. Rand Journal of Economics 32, 408–427.
- Clark, C.R., Ommerborn, M.J., Coull, B.A., Pham, D.Q., Haas, J., 2013. State medicaid eligibility and care delayed because of cost. New England Journal of Medicine 368, 1263–1265.
- Claxton, G., 2002. How Private Insurance Works: A Primer. Kaiser Foundation, Georgetown City University Institution for Health Care Research and Policy, Washington, DC.
- Cochrane, J.H., 1995. Time-consistent health insurance. Journal of Political Economy 103, 445–473.
- Cohen, A., Siegelman, P., 2010. Testing for adverse selection in insurance markets. Journal of Risk and Insurance 77, 39–84.
- Cutler, D.M., Zeckhauser, R.J., 1998. Adverse selection in health insurance. National Bureau of Economic Research, 1–32.
- Cutler, D.M., Zeckhauser, R.J., 2000. The anatomy of health insurance. In: Culyer, A.J., Newhouse, J.P. (Eds.), Handbook of Health Economics. Elsevier, Amsterdam, pp. 563–643.
- De Meza, D., Webb, D.C., 2001. Advantageous selection in insurance markets. Rand Journal of Economics 32, 249–262.
- Einav, L., Finkelstein, A., 2011. Selection in insurance markets: theory and empirics in pictures. Journal of Economic Perspectives 25, 115–138.
- Ellis, R.P., 1998. Creaming, skimping and dumping: provider competition on the intensive and extensive margins. Journal of Health Economics 17, 537–555.
- Ellis, R.P., Manning, W.G., 2007. Optimal health insurance for prevention and treatment. Journal of Health Economics 26, 1128–1150.
- Fang, H., Keane, M.P., Silverman, D., 2008. Sources of advantageous selection: evidence from the medigap insurance market. Journal of Political Economy 116, 303–350.
- Farrell, J., Klemperer, P., 2007. Coordination and lock-in: competition with switching costs and network effects. In: Armstrong, M., Porter, R. (Eds.), Handbook of Industrial Organization. Elsevier, Amsterdam, pp. 1967–2072.
- Finkelstein, A., McGarry, K., 2006. Multiple dimensions of private information: evidence from the long-term care insurance market. American Economic Review 96, 938–958.
- Folland, S., Goodman, A., Stano, M., 2006. Economics of Health and Health Care, fifth ed. Prentice Hall, Upper Saddle River.
- Frank, R.G., Lamiraud, K., 2009. Choice, Price competition and complexity in markets for health insurance. Journal of Economic Behavior and Organization 71, 550–562.
- Hackmann, M.B., Kolstad, J.T., Kowalski, A.E., 2013. Adverse Selection and an Individual Mandate: When Theory Meets Practice (Working Paper No. 19149). NBER.
- Hemenway, D., 1990. Propitious selection. Quarterly Journal of Economics 105, 1063–1069.
- Hsiao, W.C., 1995. Abnormal economics in the health sector. Health Policy 32, 125–139.
- Lassey, M.L., Lassey, W.R., Jinks, M.J., 1997. Health Care Systems Around the World: Characteristics, Issues, Reforms. Prentice Hall, Upper Saddle River.
- Manning, W.G., Newhouse, J.P., Duan, N., Keeler, E.B., Leibowitz, A., 1987. Health insurance and the demand for medical care: evidence from a randomized Experiment. American Economic Review 77, 251–277.
- Martin, S., 1994. Industrial Economics: Economic Analysis and Public Policy, second ed. Macmillan, New York.
- Nyman, J.A., 1999. The economics of moral hazard revisited. Journal of Health Economics 18, 811–824.
- Nyman, J.A., 2001. The income Transfer Effect, the access value of insurance and the Rand health insurance Experiment. Journal of Health Economics 20, 295–298.
- Nyman, J.A., 2003. The Theory of Demand for Health Insurance. Stanford University Press, Stanford.
- Pauly, M.V., 1968. The economics of moral hazard: comment. American Economic Review 58, 531–537.
- Phelps, C.E., 2010. Health Economics, fourth ed. Addison-Wesley, New York.
- Preker, A.S., Scheffler, R.M., Bassett, M.C., 2007. Private Voluntary Health Insurance in Development: Friend or Foe? World Bank Publications, Washington, DC, pp. 115–146.
- Rothschild, M., Stiglitz, J., 1976. Equilibrium in competitive insurance markets: an essay on the economics of imperfect information. Quarterly Journal of Economics 90, 629–649.
- Sapelli, C., Vial, B., 2003. Self-selection and moral Hazard in Chilean health insurance. Journal of Health Economics 22, 459–476.
- Seog, S.H., 2012. Moral Hazard and health insurance when treatment is preventive. Journal of Risk and Insurance 79, 1017–1038.
- Shen, Y., Ellis, R.P., 2002. How profitable is risk selection? A comparison of four risk adjustment models. Health Economics 11, 165–174.
- Spenkuch, J.L., 2012. Moral Hazard and selection among the poor: evidence from a randomized Experiment. Journal of Health Economics 31, 72–85.
- Stevens, F., 2009. The convergence and divergence of modern health care systems. In: President, W.C.C. of S (Ed.), The New Blackwell Companion to Medical Sociology. Wiley-Blackwell, Hoboken, pp. 434–454.
- Tirole, J., 1988. The Theory of Industrial Organization. MIT Press, Cambridge, Mass. Trivedi, A.N., Moloo, H., Mor, V., 2010. Increased ambulatory care copayments and hospitalizations among the elderly. New England Journal of Medicine 362, 320–328.
- US Congress, 2010. Pub.L. 111-148. 111th US Congress, Washington, DC.
- Van de ven, W.P.M.M., Ellis, R.P., 2000. Risk adjustment in competitive health plan markets. In: Culyer, A.J., Newhouse, J.P. (Eds.), Handbook of Health Economics. Elsevier, Amsterdam, pp. 755–845.
- Zeckhauser, R.J., 1970. Medical insurance: a case study of the tradeoff between risk spreading and appropriate incentives. Journal of Economic Theory 2, 10–26.