Markets in Health Care


The first question one should ask when addressing healthcare markets is whether health care is any different from other goods. If it is not, economic theory states that, as with apples and pears, the unfettered competitive market will lead to efficient outcomes and that equity can be reached by appropriately redistributing purchasing power ex ante. This brings the question of why in some countries healthcare markets do not even exist or are severely restricted. In national health services (NHSs) like those in the UK or Spain, a single door provides access to most health goods and services (pharmaceutical products usually being an exception), and market forces may disappear or be relegated to the stage where the health authority subcontracts with providers (doctors and hospitals). Less extremely, why is regulation of healthcare markets desirable?

Before answering these questions, let the reader be first warned that a broad interpretation of what a market is has been taken here. For instance, in the market for prepaid health plans, the individual may not pay a price either when enrolling a plan of her choice nor when she uses the services this plan provides. Even farther away from the usual idea of a market, in a NHS, the health authorities may base the remuneration of the hospitals they own or subcontract with on relative performance evaluations (RPEs). In this case, a hospital’s revenue depends on how its performance compares to the average. For those who want to read more, their web search should include the terms ‘yardstick competition’ and ‘contests.’ Even an individual’s choice between seeking treatment in his or her NHS or resort to a private hospital can be seen as a market. There, the ‘price’ in the public provider may be the time the individual has to wait. All the examples given have one thing in common: as a provider, your revenues fall if your performance (the combination of price and quality) falls as compared with your rivals or to any other existing outside option (e.g., an alternative treatment). The broad term ‘market forces’ is used to refer to this effect.

Now, the questions posited above can be readily answered. Health care is either publicly provided or its market severely regulated because society does not believe that free markets do such a good job. The differences between health and other goods, as well as the differences between health care and other services, can be traced back to the work of Arrow (1963). Since then, the role of markets in the allocation of resources in the health sector has been scrutinized from many angles. The role of ethics and societal judgments was, and is, widely discussed. The societal value of health care is not necessarily restricted to the standard notion of economic welfare (measured by the difference between an agent’s willingness to pay for a service and the costs the agent bears, the so called ‘agents’ surplus’). Other considerations like happiness, freedom to choose, and absence of pain, for example, are often included as relevant for assessment of resources allocation, in addition to the mere utility from consumption of health and health care. Health in itself, by its nature, does not have a market for transaction. The role of markets in determining welfare (and the proper meaning of welfare) is distinct in the health sector. For an introduction to the discussion, see the essays contained in Cookson and Claxton (2012).

Still, markets are one mechanism that allocates resources in the health care in many ways. There are several reasons why markets may behave differently in this sector, irrespective of the social judgment one makes about the adequateness of the resulting allocation. The focus is on the particular features of the market mechanism (without discussing here the welfare reasoning associated with each of those features) and the resulting reasons for government intervention.

The first reason – and the one that is best understood – is that some healthcare goods have external effects. That is, their consumption affects (positively or negatively) other individuals. The first example that comes into mind is vaccines. It is well known that markets for vaccines will not work well because individuals often ignore their beneficial external effect on everybody else’s well-being. A free market will lead to under vaccination. Note that this does not directly apply to a hip replacement. One could say that taking care at home of a family member who can hardly walk puts lots of strain on the other family members. However, this is a type of externality that is likely to be internalized by the individuals themselves, because they probably care as much about the well-being of their family as for themselves.

The second reason – also well understood for a long time – is that providers could hold some degree of the so-called ‘market power.’ The idea is that if very few providers are available then they will be able to restrict supply in a way that price will be too high as compared to its optimum value: the cost of providing the unit that is least valued by society. This brings two effects. One is on equity, as consumers are worse-off whereas suppliers are better-off. The other is on efficiency; too few units are enjoyed by society. Why are healthcare goods prone to such a situation? The obvious reason is the existence of patents in some areas, which restrict the number of providers to one. Now patents are there for other reason that would take it too far. The other is that there exist some treatments that imply extremely large fixed costs. It does not make any sense to have two CT-scan machines in a small village. The existence of insurance coverage by making the patient (buyer) less sensitive to price induces the providers (sellers) to increase the price at the expense of the third-party payer.

The third reason involves the presence of privileged information (or asymmetric information) in one or both sides of the market. Health care is plagued with examples of this. On the consumer side, he or she may be more knowledgeable about his or her own health risks and needs (say intensity of pain). On the supplier side, healthcare goods are usually expert goods, meaning that providers are more informed of the true benefits (or long run side effects) of certain treatments. Again, markets perform quite poorly under these circumstances.

The fourth reason is the presence of risk. However, risk per se is not really problem for markets. It only implies that the notion of a good becomes a little more sophisticated. Given a single service, say a hip replacement, there are two or more goods, one for each possible circumstance that the individual may encounter. A hip replacement when the individual is perfectly healthy is not the same good as a hip replacement when the individual is under excruciating arthritis pain. The solution is to create an insurance market. In such a market firms (now called insurers) offer to exchange goods (in this case money and hip replacement) between these circumstances. The provider offers a hip replacement for free in case of severe arthritis pain (and only in that case) for free in exchange for x monetary units ex ante. One refers to x as the insurance premium. The market for insurance should perform well if none of the aforementioned problems is present. However, what does ex ante mean? What if an individual has privileged information of the likelihood of needing a hip replacement? Individuals who expect to need a hip replacement are more willing to pay the insurance premium. This leads to all sorts of problems. One of the solutions to these problems is quite drastic: get rid of insurance markets altogether and have a monolithic (vertically integrated) public-health system, perhaps financed with taxes or the like.

The fifth reason, intimately linked with the previous one, is the presence of moral hazard in either or both the consumer and the provider’s side. On the demand side, the idea is that the individual, once insured, bears the consequences of a bad event less than fully and keeps a behavior that is detrimental to his/her health. Let this be illustrated with two very simple and very practical examples. If an insured individual becomes severely arthritic, he or she will obtain a hip replacement at a low price, perhaps even for free. Hence he or she will keep practicing vigorous sports. The same goes for stomach reduction and eating in a disorderly manner. These are of course rather extreme examples, but there seems to be some evidence that such behavioral decisions are present. Note that a real issue only exists if these behaviors cannot be contracted on because they are not publicly observable. Hence some researchers include moral hazard under the umbrella of asymmetric information, but they are not the same thing and require different cures. In the supply side, moral hazard refers to situations where the actual actions of the provider (be it a doctor’s diagnose effort or care in treating a patient, be it the manager of a hospital who fails to contain costs) are unobservable. Again the idea is that the decision maker (doctor, manager, and nurse) may not bear the full impact of his or her actions. This is the case for example if the doctor receives a fixed monthly remuneration (independent of health outcomes) and medical errors go unpunished. It is worth mentioning here that the term moral hazard is also used to describe the phenomenon by which an individual who faces a low or zero price due to insurance may overuse the services. To distinguish both phenomena the literature (not unanimously) uses the terms ex ante moral hazard, which refers to the change of behavior that may lead to greater needs, and ex post moral hazard, which refers to the increase in service usage for a given need. It is easy to see that empirically identifying which is the source of higher usage is a complex question.

The last reason is the presence of the so-called bounded rationality. In this concept, misperceptions of one’s own health status, lack of the mathematical capability to appraise extreme probabilities, intertemporally inconsistent behavior, or purely irrational behavior are included (perhaps too broadly). Since individuals are unable to make proper choices, the government (or a delegate of the government) makes them for them.

These are not the only reasons for government intervention. For example, social valuation of market outcomes may lead to government action as well. Generally speaking, dominance of one resource allocation mechanism over the other cannot be presumed (market vs. state). How the markets operate is of concern here.

Having said this, the next question is whether regulation or public provision do indeed palliate any of these problems present in healthcare markets. Section Introducing Market Forces in a NHS briefly addresses the introduction of competitive forces in a national health system. Section Market Regulation addresses the regulation of private health insurance and provision markets, Section Duplicate Systems discusses the interaction between public and private insurance. Section A Closer Look at the Provision of Goods and Services in Health takes a closer look at the delivery of health goods and services. Section A Teachers’ Guide offers a teacher’s guide to these issues. Section Concluding Remarks offers some concluding remarks.

Introducing Market Forces In A NHS

Market forces in NHSs have been implemented basically trough two different policies. On one hand, in some national health systems like that in England, Denmark, Sweden, and Norway, patients are allowed to choose between a set of hospitals in case of needing specialized care. On the other hand, some national health systems are remunerating hospitals according to the results of RPE.

In the first case, the idea is that increasing patients’ choice where patients face no copayment would foster competition in quality, as long as quality is observable by patients (or by doctors, but then patients and doctors incentives should be aligned). However, this depends in turn on how the chosen hospital is remunerated. If the remuneration is per episode and is fixed (like in a diagnosis-related group system) and quality is observable, theory predicts an increase in quality, since hospitals offering lower qualities would lose market share. However, it is not so clear that this assumption is satisfied to a sufficient degree. Moreover, even if a specific episode is reimbursed at a fixed fee, a hospital could be in financial trouble if it faces a catchment area where individuals bring higher costs for the same ailment. Things are even worse if hospitals are allowed to set their fees, since in that case they could raise a given service fee without compromising demand (recall that the patient obtains the service for free).

As for the effects of RPE-based remuneration systems under public provision, issues related to obtaining information on quality of care and on hospital performance at large. The main problem is the asymmetry of information between patients and payers, on one hand, and providers, on the other hand, on provider’s effort to deliver the adequate amount of care at the right cost. Asking the providers information faces evident problems related to truth telling and monitoring. Comparing across providers is, in this setting, a natural way to obtain information as long as performance of different providers is correlated. Fichera et al. discuss the instruments for quality comparison. One important decision is to which type of quality measurement is more informative, and whether attention should be put in quality of outcomes, in quality of processes, or in quality of inputs. It is not surprising that instead of a single quality indicator, a set is used. Using a set of indicators, however, poses the question of how to aggregate these indicators into a single variable. Without such aggregation it may be impossible to obtain a clear ordering of hospitals according to quality, as some hospitals may fare above the average in some services and worse in others.

Market Regulation

Every market is characterized by demand, supply, and a ‘mechanism’ that connects both to each other. In the standard textbook treatment of markets, that mechanism is the price of the good. Markets included in the health sector often deviate from this simple framework. Market analysis applied to health care or health insurance has to adjust for particular features involved.

First of all, the health sector involves several types of markets. Three of them are highlighted: the market for labor inputs, the market for goods and services, and the market for health insurance. Each of them has their own specific set of issues.

Take a healthcare good or service. As mentioned above, the uncertainty about the moment and intensity of need for health care leads to the existence of insurance mechanisms (public, private, or both). Such insurance implies markets for healthcare goods and services have a third agent, the insurer, who decouples the price received by the provider from the price paid by the consumer. This third agent may take a passive role, as in traditional reimbursement models, or may take an active role. The active role can range from establishing conditions under which demand can exert (eventually limited) choice of providers, to contracting and paying directly providers or even integrating vertically insurance and provision. Of course, if a single enterprise implements such integration it is got back to NHS (Figure 1).

Markets in Health Care Figure 1

The Private Health Insurance Market

Even in a basic private health insurance system, like the one present in Switzerland or in the US for those not covered by Medicare or Medicaid, the market incorporates a vertical dimension that is depicted in Figure 2. In the final stage of the vertical relationship, doctors and hospitals are contracted by insurers in order to provide healthcare goods and services. In the intermediate stage, consumers seek insurance contracts from insurers. Each of these stages is in itself a market, and often researchers have concentrated attention in one of them either by taking the outcomes in the other as given or by assuming that the other performs efficiently. In other words, in studying the insurance market, all the problems listed above are often assumed away in the relation between insurer and provider. Conversely, the insurers’ revenue is taken as given when one studies the contracting phase between providers and insurers.

Markets in Health Care Figure 2

This is not to say that the lessons learned in studying one market are not useful in studying the other. Indeed, in both markets, a major issue is how to set the payments given that each firm (be it an insurer or a final provider) faces a heterogeneous set of consumers. Indeed, an individual may have a high or a low probability of falling ill, which matters for the insurer, and the same individual, once ill and hence requiring a specific treatment, may bring high or low costs, which matters for the service provider. In the first instance and if the insurer is able to choose and collect premia from the individual, the issue is whether and how does the insurer charge (or is allowed to charge) different premia to individuals presenting different characteristics (age or gender for instance). This practice is usually termed risk classification or risk categorization (see Section Risk classification). If the insurer instead receives the premium from a third-party payer (like in Medicare or in the Netherlands), the same issue is termed risk equalization or risk adjustment. In the relationship between the insurer and the provider, one speaks of designing patient classification systems and their use to set risk-adjusted payments.

The overall idea is that premium should fit the expected cost for every individual and that health treatment price should fit its cost. Otherwise, the supplier (again, insurer, or provider) will have an interest in attracting the individuals or serving the treatments where payment exceeds cost (cream skimming or cherry picking) and avoid the individuals or treatments where the inequality is reversed (dumping or skimping). Both opportunistic behaviors fall under the term risk selection.

Risk Classification

Important issues arise from information asymmetries between agents. A natural intervention is the demand for further information to be incorporated in decisions. In this vein, risk classification aims at reducing informational asymmetries between health insurers and individuals.

Under perfect risk classification, an insurer conditions its contracts on so many individual characteristics that the individual and the insurer have the same information (and therefore the same expectations) about future costs. In this case an individualized premium fitting these expected costs can be set. Such elimination of asymmetric information will lead to an efficient market allocation. However, high risks will face higher premia than low risks. The government can then set appropriate taxes and transfers to improve the welfare of the former at the expense of the latter. Voluntary participation by the low risks may put a limit to such cross subsidization, unless the government makes purchasing insurance mandatory. Incidentally, equalizing premia by the government is not to be confused with some naive ‘community rating’ where the insurer is not allowed to set different premia to different individuals. Such a policy might either lead to risk selection or to the self-exclusion of the low risks (the so called ‘spiral of death’).

The problem is that risk classification can only be performed on the basis of observable characteristics of the population. Moreover, collecting data on such characteristics may be quite costly. In any case, only a small set of variables is actually used to design contracts, and this implies that these variables fall very short of being perfect predictors of health risk. As a consequence, the market usually reacts by implementing self-sorting menus of contracts (be at the industry or at the firm level). By this it is meant that individuals, who now have privileged information on their true health risks, reveal this information by choosing one contract instead of another. Such self-sorting menus can only be constructed, however, by reducing the coverage and premia of the contracts aimed at attracting the low risks. Better risk classification could reduce these distortions according to some authors.

Note that the importance of improving – or limiting – risk classification depends on the extent of asymmetric information existing at the outset. This needs to be tested empirically. Current empirical work has progressed but is still far from a definite answer to the question of how significant and pervasive are the asymmetric information problems. Some studies found effects of relevant magnitude whereas others found less impressive implications. A well-recognized problem in testing for asymmetric information is that individuals may have privileged information on dimensions other than risk, and that differences in one dimension could be countervailed by differences in another. For instance, more risk-averse individuals (in principle more willing to pay for coverage) may at the same time have safer habits or lifestyles, which reduces their willingness to pay for coverage.

Switching Costs

Whenever more than one possibility of health insurance coverage exists, patients will typically face trade-offs in choosing one health insurer over the other, and issues of switching across health insurers cannot be neglected. Taking the Swiss long-lasting experience with health plans’ competition, a review of it is of interest to the countries promoting choice of insurance contract. As for consumers’ choice and consequently for switching behavior, several issues are particularly relevant: choice overload, the resistance to change (status quo bias); and the existence of risk selection.

Preferred Providers

It is stated above that the vertical structure Individuals– insurers–providers is seldom studied as a whole. Some exceptions exist, however. An important issue is with which provider each insurer decides to work. The selection of providers by insurers will also determine how demand for health services is directed toward providers. The main issue is how insurers define the size of the network of providers they use, and how that size depends on the specific rules used to define which providers belong to the network. Selecting a subset of market providers as preferential providers (insiders henceforth) changes the strategic incentives of providers to compete in the market. More specifically, by being included in the network of a health insurance plan (public or private), insiders gain a competitive advantage vis-a` -vis the outsiders. To see this, notice that patients will pay less (or even not pay at all) when choosing to be treated by an insider than when selecting an outsider. Hence insiders will face a demand for health care with lower price elasticity, bringing higher prices in equilibrium. This harms the insurer since it must bear higher prices herself. Providers will compete to become members of the preferred network of providers in order to obtain this competitive advantage. Equilibrium may have most or even all providers as preferred ones.

In this context, the payment rules set to providers gain importance as a way to induce competitive pressure. Indeed, the third-party payer can reintroduce competitive pressure by having the patient pay less for the outsider treatment the higher the price of the insider treatment is. In other words, if the insider sets a higher price, the reimbursement received by the patient when choosing an outsider is also larger. Some demand is then diverted from the insider to the outsider. Provider networks are also a form of competition between insurers as well. Patients may opt for one or the other networks based on the list of providers in each network. Competitive forces will be present both across providers and across insurers.

Demand-Side Issues

The demand side of health care is also characterized by information problems. Patients are not fully knowledgeable about their own health condition, and they are not completely informed about treatment options. Patients rely on physicians to guide them through the healthcare system in order to restore their health condition. This agency relationship where the interaction of the referencing of patients to other providers (hospitals and specialists) and incentives is discussed. Gatekeeping receives particular attention. In itself, gatekeeping is a constraint on freedom of choice by patients, in the context of a trade-off between free choice and more informed choice. The advantage of using a gatekeeping organization does depend on the incentives of physicians acting as gatekeepers, which raises the issue of incentives faced by gatekeepers to perform their role. Primary care concentrates several roles (health promotion and prevention, diagnosis and treatment, referral, and long-term care). By restricting freedom of choice, gatekeeping is expected to be associated with lower patient satisfaction but also with lower (unnecessary) use of health services and lower expenditures. The empirical evidence on this trade-off has not yet produced results that account for confounding factors introduced by financial incentives faced by physicians (systems with freedom of choice use, generally, fee for service payments, whereas systems with gatekeeping use capitation).

The Role Of Risk Adjustment In Market Competition Among Health Plans

Countries with competition in health insurance want to ensure, at the same time, affordability of health care to all citizens and nondiscrimination of contribution based on individual risk. Health insurers, however, for the same value of contribution, prefer to contract with the better risks. This led to a role for risk adjustment in market competition among health plans. The solution adopted was to set a two-steps system (Figure 3). First, contributions not based on individual risk are used to build a pool. Second, risk-adjusted payments from the pool to health insurers (or health plans) aim at the double objective of providing enough funds and avoiding incentives for selection of good risks. In most cases these payments are made on a capitative basis, that is, per enrollee in the health plan, hence the term capitation rates. Risk adjustment is an essential element of market competition but its accurate definition is a difficult task. This system is in place, with some variations, in the Medicare sector in the US, the Dutch, Belgian, and German systems, as well as the system in place for public servants in Spain. The European approach to it has been mainly data driven (hence term statistical risk adjustment), attempting to find the more adequate system of risk adjustment based on observables like age, gender, and even prior use of healthcare services.

Markets in Health Care Figure 3

Although statistical risk adjustment takes it for granted that an insurer will not engage in risk selection if expected costs (calculated ex ante) are close enough to the capitation rate, the latter entry admits the possibility that even in this case individuals may make use of their privileged information on their true health risks. Equivalently, there may be observables that are correlated with expected costs but cannot be used in the risk adjustment formula (either due to non discriminatory laws or inherent uncontractability). This implies that both individuals’ choices and insurers’ behavior must be taken into account when designing the capitation system. This leads these authors to seek an adequate distortion in weights of the risk adjustment model to provide the correct incentives to providers. This approach requires empirical work in understanding how providers (or health insurers) react to risk adjustment rules in order to design these rules taking into consideration such reactions. The empirical challenge is not to find the best statistical fit, but to measure reactions in behavior of health plans (or providers).

Duplicate Systems

In countries with a NHS, the main insurance protection is Government provided. Private (voluntary) health insurance has then a duplicate role of coverage (see Figure 4). Duplicate health insurance coverage means private health insurance to cover for the same risks as the NHS. The reasons suggested include promotion of population health, containing health expenditures, increased population choice, and health system ‘responsiveness,’ whenever the NHS fails to deliver health care to the extent desired by the population. The empirical support for the reasons behind duplicate private health insurance is not fully conclusive.

Markets in Health Care Figure 4

A different set of questions is related to the impact of double coverage on use of services, and whether it adds, or substitutes for, NHS expenditures. On this aspect, no conclusive evidence is available. Double coverage seems to be associated with higher use of healthcare services, though some of it can be diverted from the NHS. Overall, there is no evidence or theory justifying a strong presumption of more efficient (less costly) healthcare provision through the duplicate coverage than under the NHS. That the private and public systems implicitly compete for patients brings two other issues that are discussed in Sections Waiting Lists and Specialists: the possibility that a physician works for both sectors and the role of waiting as a rationing device.

A Closer Look At The Provision Of Goods And Services In Health

On the supply side of healthcare markets, many different providers exist, according to the particular good or service. The main ones are hospitals, primary-care services, imaging services, and pharmacies. Firms operating in other markets have several instruments to compete and attract demand: price, quality, and advertising are the main ones. Their use in health care is often restricted by regulation. Three nonprice competition variables are addressed below, namely, advertising, waiting time, and quality; and two specific providers, Pharmacies and Specialists.


Advertising from healthcare providers is often subject to strong restrictions if not banned at all. Advertising has long-been recognized has having two different roles, information and persuasion. Both types of advertising direct demand to the product or service being advertised. Although the informative advertising is usually taken to produce positive effects (as it reduces information asymmetries), persuasive advertising is considered socially wasteful, as it distorts preference. An important aspect is the ambiguous impact of advertising on unobserved quality of health care, both in theory and in empirical evidence. Advertising seems to matter for competition between healthcare providers, as when bans on advertising are lifted, the later increases. But again the nature of advertising matters, as persuasive advertising may soften competition (and lead to high prices) whereas informative advertising increases price competition.

Waiting Lists

In some healthcare markets, price is not the only instrument to match supply and demand. Owing to random demand and random treatment times, setting demand to equal supply in each moment in time leads to excess capacity and idle resources. In private markets, these random elements are diversified across the several existing providers. In the presence of NHSs (or integration of insurance and provision), the diversification role of random arrivals for treatment and random treatment times cannot be done by choosing available providers. Instead, waiting lists and waiting times are used as an alternative mechanism to balance the system. Determining access to health care based on the price paid is often considered unfair and undesirable, and using time is preferred. Discriminating time to access based on clinical need (prioritization) is acceptable whereas doing it on the basis of ability to pay the price usually is not. Waiting times are common in NHSs. Waiting times perform several roles. Waiting time works as a variable that balances demand and supply as a substitute to price, as health insurance protection and equity considerations entail prices having a much smaller role. Waiting times lead patients to make a trade-off between faster treatment and price paid when a private sector having no waiting times is available. The use of waiting times and waiting lists can also be seen as an alternative device to redistribute resources, as patients resorting to the private sector to skip waiting lists pay twice.


In many health systems, prices of health of health care are regulated, either by decision of a NHS or by agreement set with health insurers. Providers, nonetheless, would like to guide demand toward their own services and products. When price is not available, other competition instruments have to be found. One of these instruments is quality, as long as quality is observable by the key choice maker (which in some cases is the patient and in other cases is the medical doctor, acting as agent of the patient). A main concern is whether competition will lead to lower quality, as providers attempt to save costs, or in higher quality, as providers look at ways to increase demand.


Pharmaceuticals are probably the most diffused type of good. Pharmaceutical products can be used by patients during treatment episodes, during admission for treatment (e.g., at hospitals), but are also widely used in ambulatory care. Many chronic conditions are treated on a daily basis with pharmaceutical products. Physicians prescribe the treatment and patients will buy the product from specialized retail outlets, pharmacies. The retail distribution of pharmaceutical products is, therefore, one more aspect of competition in the health sector. There are often constraints on price (regulation of pharmaceutical prices) and margins (distribution margins may be regulated), as well as constraints on entry. In some countries only pharmacists can own pharmacies. In some countries, a new pharmacy can only open on authorization from a regulatory body. In some countries, opening of a new pharmacy needs to obey rules related to population size and distribution as well as distance to competing pharmacies.  Price regulation and entry regulation interact strategically and getting it right requires careful analysis.


Healthcare services are intensive in labor. Several dedicated professions exist, like physicians, nurses, and pharmacists, for example. The working of labor markets is therefore of importance. This is particularly true for physicians. They have the ability to ‘guide’ demand (patients) across services and providers. Physicians direct demand, which, under public or private health insurance arrangements, is often insensitive to prices. Then, the incentives faced by physicians will be a crucial element in determining how they allocate demand to providers. Gonzalez takes aim at a particular setup for doctors’ decisions. In this setup, physicians may work in both a public and a private healthcare provider. Their decisions will define how demand splits across the two sectors. Physicians’ decisions are again sensitive to the incentives they face. Possible policies include bans to working in a second job. Theoretical treatments of dual practice provide ambiguous effects on efficiency and quality of care. This ambiguity is not solved by empirical research, leaving room for further work. Although at first inspection a ban on dual practice would be welfare enhancing as it reduces the incentives for physicians to shift patients from public to private healthcare providers, there are other effects at play. Allowing dual practice has the benefit to the public sector of a lower cost to retain highly qualified professionals. Thus, a careful analysis of each institutional setting is called for.

Concluding Remarks

Since the seminal work of Arrow (1963) the application of economic theory to health care has evolved tremendously. The application of the usual apparatus of demand and supply to healthcare markets has been questioned in many aspects including how society assesses allocations of resources. Both demand and supply side features received attention. Markets of health insurance, healthcare goods and services, and health professions were and are today explored in detail.

Since early the asymmetric information aspects of demand and supply of health insurance were explored. Inefficiencies of market allocations and the issue of existence of market equilibrium were identified. It is now understood to some extent how health insurance markets work, what motives there are for regulation, but challenges remain. Two of the main ones, deserving both theory and applied research, are risk categorization and risk adjustment and consumers’ switching behavior. Market equilibrium is often determined by freedom of choice of consumers, and competitive pressure for efficient supply comes from consumers’ exerting choice. Therefore, reducing information asymmetries and understanding how choice of consumers occurs is likely to originate further research. Regulation in health insurance markets is often intertwined with reducing impact of information asymmetries (here the risk pooling funds in certain countries, the mandates for health insurance in other countries and the existence of NHSs as mandatory insurance can be named) and with promotion of consumers’ choice of health plans (e.g., such as rules and periods of switching health plans).

The markets for provision of health care also deviate from standard textbook analysis, as patients, the final consumers, often use the services of experts (doctors) to guide their demand of health care. They often have health insurance (either public or private), which makes them less (or even totally) insensitive to price at the moment of use. As a result, consumption decisions are distorted and market prices, if left totally unregulated, too high. Consequently, third-party payers (health insurers, sickness funds, and NHSs) over time moved into a more active role, in both the demand and the supply side. Market equilibrium then becomes the result of the interaction of demand, supply, and third-party payers. The continued growth of healthcare costs leads to interest in how the market allocates resources and how such allocation can be influenced. The role of nonprice market equilibrium mechanisms also ranks high in the agenda. Waiting lists can be named, where time is the rationing device, but also supplyside management such as medical guidelines and protocols, procedure authorizations, and so on. Advertising restrictions in healthcare markets are common, and competition in quality often substitutes for price competition. Bringing together these different aspects into the analysis of market equilibrium and its properties has resulted in a large stream of research in specific topics.

The development of new products and services, innovation, is another area of interest, not the least because technology is considered one of the main drivers of rising healthcare costs. The so-called ‘bending the cost curve’ in health care will certainly be related to the rate of growth and costs of delivering innovation.

Society’s values have an impact on the way market equilibria in healthcare markets are looked at and consequently on the regulation imposed. Access to health care is a major concern. Ensuring access implies the development of networks of providers. The network can be centrally defined and built, as it is the case in NHSs, or can result from decisions of health insurers. Defining networks of providers of health care affects market equilibrium and competition both in health insurance markets and in healthcare provision markets. As third-party payers are increasingly active in managing demand and supply, this area is likely to receive further research attention.

Input markets in health care have own specific characteristics as well. Training healthcare professionals, in particular medical doctors, takes a long time and it is highly costly. Allocation of doctors to training vacancies and specialties is an important issue in many countries. Doctors are special input factors as they determine demand (when acting as agents for patients) and provide services (as suppliers of healthcare services). But also nurses and other professions have markets for their services, and the scope of health professions is changing in response to market forces. As an example of these changes there is the ability, in some places, of qualified nurses to prescribe pharmaceuticals for common health problems of the population. Or the issues associated with dual practice by doctors, especially when both public and private healthcare provision coexist (and compete). The way training and tasks of health professions evolve will potentially change market equilibrium in input markets, influencing the supply side of healthcare provision. Market equilibria will change as well in provision of healthcare services and ultimately in health insurance markets.

Many forces shape market equilibria and regulation in health care. Understanding the economics of markets in health care is an unfinished task, and future research will certainly develop issues addressed in this article and likely open new areas of research as well.


  1. Arrow, K. (1963). Uncertainty and the welfare economics of medical care. American Economic Review 53, 941–973.
  2. Cookson, R. and Claxton, K. (2012). The humble economist – Tony Culyer on health, health care and social decision making. York: Office of Health Economics and The University of York.
  3. Dranove, D. (1988). Demand inducement and the physician/patient relationship. Economic Inquiry 26, 251–298.
  4. Frank, R. G., Glazer, J. and McGuire, T. G. (2000). Measuring adverse selection in managed health care. Journal of Health Economics 19, 829–854.
  5. Garcia-Marin˜oso, B. and Jelovac, I. (2003). GP’s payment contracts and their referral practice. Journal of Health Economics 22, 617–635.
  6. Grossman, H. (1979). Adverse selection, dissembling and competitive equilibrium. Bell Journal of Economics 10, 336–343.
  7. Hoy, M. and Polborn, M. (2000). The value of genetic information in the life insurance market. Journal of Public Economics 78, 235–252.
  8. Jack, W. (2005). Purchasing health care services from providers with unknown altruism. Journal of Health Economics 24, 73–94.
  9. Ma, C. A. (1994). Health care payment systems: Cost and quality incentives. Journal of Economics & Management Strategy 3, 93–112.
  10. Rothschild, M. and Stiglitz, J. E. (1976). Equilibrium in competitive insurance markets: An essay on the economics of imperfect information. The Quarterly Journal of Economics 90, 629–649.
  11. Zeckhauser, R. (1970). Medical insurance: A case study of the tradeoff between risk spreading and appropriate incentives. Journal of Economic Theory 2, 10–26.
  12. Barros, P. P. and Siciliani, L. (2012). Public–private interface in health and health care. In Pauly, M., McGuire, T. and Barros, P. P. (eds.) Handbook of health economics. Amsterdam: Elsevier Science.
  13. Blomqvist, A˚ . (2011). Public sector health care financing. In Glied, S. and Smith, P. C. (eds.) The oxford handbook of health economics. Oxford: Oxford University Press.
  14. Dowd, B. and Feldman, R. (2012). Competition and health plan choice. In Jones, A. M. (ed.) The Elgar companion to health economics, 2nd Cheltenham: Edward Elgar Publishing Limited.
  15. Glazer, J. and McGuire, T. G. (2012). Optimal risk adjustment. In Jones, A. M. (ed.) The Elgar companion to health economics, 2nd FALTA. Northampton, MA: Edward Elgar Publishing, Inc.
  16. Iversen, T. and Siciliani, L. (2011). Non-price rationing and waiting times. In Glied, S. and Smith, P. C. (eds.) The oxford handbook of health economics. Oxford: Oxford University Press.
  17. Zweifel, P. (2011). Voluntary private health insurance. In Glied, S. and Smith, P. C. (eds.) The oxford handbook of health economics. Oxford: Oxford University Press.