Due to the COVID-19 crisis, the transition into subscription mode of the Oxford Research Encyclopedia of Economics and Finance has been postponed to May 28th. Please watch this space for updates as we work toward launching in the near future. Visit About to learn more, meet the editorial board, or learn how to subscribe.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ECONOMICS AND FINANCE (oxfordre.com/economics). (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 28 May 2020

Competition and Quality in Healthcare

Summary and Keywords

Quality competition between alternative providers is an increasingly important topic in the health economics literature. This literature includes theoretical and empirical studies that have been developed in parallel to 21st-century policies to increase competition between doctors or hospitals. Theoretical studies have clarified how competitive markets can give healthcare providers the incentive to improve quality. Broadly speaking, if providers have an incentive to attract more patients and patients value quality, providers will raise quality until the costs of raising quality are equal to the additional revenue from patients attracted by the rise in quality. The theoretical literature has also investigated how institutional and policy parameters determine quality levels in equilibrium. Important parameters in models of quality competition include the degree of horizontal differentiation, the level of information about provider quality, the costs of switching between providers, and the time-horizon of quality investment decisions.

Empirical studies have focused on the prerequisites of quality competition (e.g., do patients choose higher quality providers?) and the impact of pro-competition policies on quality levels. The most influential studies have used modern econometric approaches, including difference-in differences and instrumental variables, to identify plausibly causal effects. The evidence suggests that in most contexts, quality is a determinant of patient choice of provider, especially after greater patient choice is made available or information is published about provider quality.

The evidence that increases in competition improve quality in healthcare is less clear cut. Perhaps reflecting the economic theory of quality competition, showing that different parameter combinations or assumptions can produce different outcomes, empirical results are also mixed. While a series of high-quality studies in the United Kingdom appear to show strong improvements in quality in more competitive areas following pro-competition reforms introducing more choice and competition, other studies showed that these quality improvements do not extend to all types of healthcare or alternative measures of quality.

The most promising areas for future research include investigating the “black box” of quality improvement under competition, and behavioral studies investigating financial and nonfinancial motivations for quality improvements in competitive markets.

Keywords: healthcare, quality, competition, choice, hospitals, primary care, markets, health economics


This article discusses how competition between alternative providers affects the quality of care provided in healthcare markets. We outline the literature in economics on this topic including both theoretical and empirical studies, which interacts closely with recent government policy changes across the developed world.

In particular, the focus of this article on quality, as opposed to price or other performance measures, limits the scope of our analysis but allows us to provide a deeper and more focused study. We choose to concentrate on key studies in the literature rather than provide a complete review. We use these example studies to explain the development of the literature and how it interacts with policy and other institutional factors.

We develop a theoretical framework for presenting the relationship between competition and healthcare quality to guide interpretation of policies and existing research. This framework develops a canonical model of quality competition and shows how the theoretical literature has extended this model by considering the role of information, endogenous provider entry, and quality investment over time.

Next, we review the key empirical approaches and findings that have closely followed policy development. For example, in the United Kingdom where the government introduced a series of pro-competition reforms to the hospital sector in the mid-2000s, a series of high-quality empirical studies followed over the following decade.

The debate on the appropriate role of market forces in healthcare is contentious. Economists (Gaynor, Moreno-Serra, & Propper, 2013) and health services researchers (Mays, 2011) have disagreed on the interpretation of data showing competition may improve quality. This article describes the research findings with a view to understanding what factors may affect the relationship between competition and quality across contexts including information, funding, and the type of healthcare (e.g., primary versus secondary, emergency versus planned).

Why Quality?

What is the motivation to improve or increase the quality of healthcare in developed countries? Perhaps one early source can be found in the literature on unexplained variations in process measures of healthcare use and practice (Wennberg & Gittelsohn, 1973, McPherson, Wennberg, Hovind, & Clifford, 1982; Wennberg, 2002). This literature documents wide variations between providers in the provision of common surgical procedures such as tonsillectomy or hernia repair across small areas within the United States (Wennberg & Gittelsohn, 1973) or across developed countries such as the United States, United Kingdom, and Norway (McPherson et al., 1982) with no explanation from observable factors. If patients can receive such varied treatment with no obvious explanation, one possible interpretation is that some patients are receiving inadequate (or excessive) care. This pattern of variations in healthcare use has been interpreted as evidence that some providers must be providing low-quality care (Wennberg, 2011).

A second motivation comes more recently from the increasing availability of data on the variation in measures of health outcomes between healthcare providers (Marshall, Shekelle, Davies, & Smith, 2003). Due to improvements in administrative data collection, IT systems, and the Internet, the construction of measures quality based on patient outcomes has developed over time: from early experience with publishing mortality rates (Dranove, Kessler, McClellan, & Satterthwaite, 2003) to recent experiments with patient-reported outcome measure for hip-replacement patients (Gutacker, Siciliani, Moscelli, & Gravelle, 2016). The availability of this data to easily make comparisons across providers has prompted calls for action to improve quality at underperforming providers.

Why Competition?

Drawing comparisons with other sectors of the economy where consumer sovereignty, price comparison, and quality comparison are deeply engrained the concept of increasing consumer choice in healthcare and other public services has become appealing to policymakers (Le Grand, 1991). Concurrently, the increasing use of “prospective” funding mechanisms such as payments for patient episodes based on “diagnosis related groups” put the emphasis on funding being proportional to the quantity of patients treated. This mechanism incentivizes providers to attract more patients rather than provide more care per patient.

This institutional framework of choice of provider, coupled with prospective funding then provides the foundation for quality competition. If providers have an incentive to attract more patients, they will raise quality (which patients value) until the costs of raising quality are equal to the additional revenue from patients attracted by the rise in quality (see, e.g., Brekke, Nuscheler, & Straume, 2006). Intuitively, this mechanism leads to higher quality levels: as a result, the competition is fiercer between providers.

Compared to most other quality-improvement initiatives that involve cost increases—or analogously, large investments by public health systems—increasing competition between providers does not in principle require higher levels of public spending. Pro-competition reforms can be seen as something like a “silver bullet” by some policymakers, especially in cash-strapped public healthcare systems where there is a lack of funding to increase quality by other means.

Quality Competition in Practice

While the potential benefits of quality competition seem clear in principle, there are some complications in implementing a quality competition framework in practice:

  • What do we mean by quality?

  • How can providers increase their level of quality?

  • How can providers have an incentive to compete for more patients if they are capacity constrained?

Firstly, healthcare quality is difficult to define and measure. While sophisticated quality of life measures are commonly used in clinical studies, the most prominent studies on competition and quality use fairly crude failure rates (e.g., AMI mortality rates in Gaynor et al. 2013). Only recently have papers begun to use patient reported health outcomes to measure the effects of competition (Longo, Siciliani, Gravelle, & Santos, 2017). All measures of quality using health outcomes rely on appropriate risk adjustment to account for differences in casemix between providers. Overall, the best approach to measuring quality of healthcare providers is still an open question.

Secondly, even if measures of quality can be agreed on, it is often hard to see exactly what levers are available to providers to improve their performance on those measures. With notable exceptions (Bloom, Propper, Seiler, & Van Reenen, 2015), the literature has treated the quality production of healthcare providers as a black box. Theoretical literature has argued that quality production may have a “public good” interpretation, where quality production is unrelated to the marginal cost of treating patients (Brekke et al., 2006; Gravelle & Sivey, 2010). These models suggest providers invest in quality at an institutional level: for example, through training, facilities, and the policies and protocols decided by management.

Thirdly, capacity constraints provide a major potential disconnect between theory on quality competition and the real-world experience of public healthcare systems. Providers will only have an incentive to compete if they are below capacity—if they can accommodate any extra patients they attract by raising their quality levels. In public health systems this can seem implausible for many parts of the market where waiting times are high (Siciliani, Moran, & Borowitz, 2014). Theoretical literature has tried to incorporate assumptions under which hospitals have an incentive to compete even when they are capacity constrained. These approaches show that hospitals may still have an incentive to compete while capacity constrained if there are competitive and non-competitive market segments and providers are partially altruistic with respect to patients (Brekke, Siciliani, & Straume, 2008), or if providers are only capacity constrained in the short term and may still have an incentive to attract more patients over a longer period (Chen, Meinecke, & Sivey, 2016).

The remainder of this article is split into three sections: The second section develops a theoretical framework with which to summarize some key theoretical approaches to quality competition in healthcare. The third section focuses on key empirical studies related to competition and quality. The fourth section provides some concluding remarks.

A Theoretical Framework for Quality Competition in Healthcare

Base Model

Consider a market that can be described as a linear city as in Hotelling (1929). It is a line segment with length l. At each end of the city, there is a healthcare provider, indexed by i{1,2}. Assume the location of provider 1 is at 0 , and provider 2 at l. Each provider can set its quality level ri0, for which it incurs the cost αri², with α>0.

The patients are uniformly distributed along the market with mass l. Each patient can choose one provider to receive a healthcare service. A provider receives a regulated payment p for each patient it treats, and its cost for treating a patient is constant at c. Assume that patients are covered by a public nationwide healthcare program such as the United Kingdom’s National Health Service, and therefore they do not need to pay the fee. If a patient, indexed by j, chooses a provider i, her utility from receiving is treatment is


where dij is the distance between the patient and the provider, and t>0 measures the consumer’s marginal travel cost. For ease of analysis, we assume that if a patient does not go to any provider, her outside utility u0j will be sufficiently low, such that u0j<tl. This assumption ensures that in equilibrium the market will be fully covered, that is, every patient will go to one provider.

While primitive, the set-up incorporates both horizontal quality differentiation and vertical quality differentiation. The former refers to the situation where the same product is of different value to different consumers. This arises from patient heterogeneity, which is reflected in patients’ differentiated locations. Consequently, the parameter t can be interpreted as measuring the degree to which the two providers are horizontally differentiated. Vertical quality differentiation, which is at the center of this article, is captured by ri. We assume the providers are profit maximizing. Suppose a provider receives patients with size mi, then its profit is


We are interested in deriving the quality levels the providers will set in equilibrium. The solution concept here is a Nash equilibrium (r1*,r2*) such that given the other provider’s quality level ri*, provider i finds ri* maximizes its profit.

First, consider a general pair of quality levels (r1,r2). For a patient whose distance from provider 1 is d, she will choose provider 1 if



Consequently, all patients whose locations are below d(1/(2t))(tl+r1r2) will choose provider 1, and those located above (1/(2t))(tl+r1r2) will choose provider 2. Therefore, the demand for provider 1 is m1=(1/(2t))(tl+r1r2) and the demand for provider 2 is m=l(1/(2t))(tl+rr), as shown in Figure 1.

Competition and Quality in Healthcare

Figure 1. Quality levels and demands for each provider.

To see the providers’ incentive to compete in quality, note that each provider’s demand increases with its quality but decreases with the opponent’s quality. Figure 2 illustrates the scenario where as provider 1’s quality increases, it becomes more attractive to the patients, and consequently its demand increases and the opponent’s demand drops.

Competition and Quality in Healthcare

Figure 2. Increase of provider 1’s quality level leads to an increase in demand.

Given provider 2s quality choice r, provider 1 solves its profit maximization problem:



Similarly, we can set up provider 2’s profit maximization problem. First order conditions lead to the providers’ equilibrium quality


Intuitively, the equilibrium quality level is determined by the provider’s consideration on the marginal benefit and the marginal cost of increasing quality. The marginal benefit comes from the increased volume of patients that a higher quality will attract, which leads to a higher revenue. On the other hand, the strict convexity assumption on the cost of quality implies that the marginal cost of quality increases with the quality level, and this dampens a provider’s incentive to attract more patients by further increasing quality.

There are several channels through which we can examine how competition impacts on the providers’ quality. First, let us consider a benchmark case where competition is absent. In particular, assume that patients located lower than (l/2) can only choose provider 1, while those residing above (l/2) can only go to provider 2. This assumption is consistent with historical national healthcare systems in some countries with Beveridge healthcare systems. For example, prior to 1999, in the United Kingdom’s NHS system a patient’s street address (postcode) largely determined which healthcare provider she could go to (Pearson & Rawlins, 2005). Since we assume a patient’s utility from not going to any provider is sufficiently low (lower than tl), this implies that even if the providers set the quality at 0, all patients will seek their service. Because it is costly to improve quality, it is then a straightforward result that in the absence of competition and/or patient choice, profit-maximizing providers will set their quality at its minimum (i.e., 0). Therefore, introducing competition helps improve both providers’ quality by (1/4)((pc)/(αt)).

Alternatively, one can assume that the regulator has set a minimum quality level r, below which the quality can be interpreted as medical malpractice. Then in equilibrium, the providers will set quality at max{r,(1/4)((pc)/(αt))}. Consequently, introducing competition helps improve both providers’ quality by max{(1/4)((pc)/(αt))r,0}. In other words, if the regulated minimum quality r is already higher than (1/4)((pc)/(αt)), then introducing competition will not help improve quality.

There is another way to examine the impact of increasing competition on quality. As Brekke, Siciliani, and Straume (2011) and Katz (2013) pointed out, more competition can be interpreted as increasing the number of providers. For example, suppose the benchmark case is a monopoly with only one provider located at point 0 of the Hotelling line. If we maintain the assumption that the patients’ outside utility is sufficiently low so that they have to seek treatment from a provider (so that the market is fully covered), then the monopolist will set the quality at the minimum level r. Consequently, introducing competition (duopoly) helps improve quality by max{(1/4)((pc)/(αt))r,0}. However, if we amend the assumption about patients’ outside utility such that, say, u0j=0, where the market is not fully covered, then it can be shown that the monopolist will set the quality at (1/2)((pc)/(αt)). Here, the monopoly facing a market that is not fully covered has higher quality than the duopoly with a fully covered market. The intuition is as follows. A higher outside patient utility means that the monopolist has to increase quality in order to attract patients. Moreover, as Katz (2013) suggests, the monopolist now faces a more quality-elastic demand than a duopolist. This is because the monopolist takes “share” away from outside good that is valued by all consumers equally. In contrast, at the margin it is harder for a duopolist to take market share from the rival than for the monopolist competing against the outside good, because a duopolist has to appeal at the margin to patients with increasingly high valuations of its rival’s services and increasingly low valuations of its own services.

There is a third channel to examine the effect of competition on quality. As explained above, the parameter t can be interpreted as measuring the degree to which the two providers are horizontally differentiated. In other words, the smaller the t, the more competitive the market. As we can see, the equilibrium quality level increases as t goes down. The intuition is as follows. As t decreases the providers become less horizontally differentiated, or equivalently, the consumers are more homogeneous. As the travel cost weighs less in the consumers’ decision, they care more about the provider’s (vertical) quality. That is, for a lower t, the same increase of ri will attract more patients. Therefore the providers have more incentive to compete on quality when the market is less horizontally differentiated.

Endogenous Number of Providers and Endogenous Price

A shortcoming of the Hotelling-line setting is that it is hard to accommodate more than two providers, because the symmetry of the equilibrium breaks down as it goes beyond duopoly. Gravelle (1999) adopted the circular-city setting in Salop (1979) to model the behavior of GP firms. Although aimed at the market for GPs, the model is also applicable to competition between other geographically differentiated healthcare providers competing on quality. Instead of being a line segment, the market is assumed to be a circle with a fixed circumference. The circular setting allows the researcher to endogenize the number of providers.

Specifically, the market has free entry to the providers. Each provider can set its price as well as quality. Given the number of providers, the equilibrium will be a symmetric one where the providers are equidistantly distributed along the market with the same price and quality.

There can be three types of equilibrium for a given number of providers, each corresponding to a different extent to which patients are covered. First, it can be a monopoly equilibrium. When the market is sufficiently large and the number of providers is relatively small, every pair of neighboring providers are sufficiently apart. Consequently, each provider will set its quality and price as if it is the only provider in the market. Some patients will not seek treatment from any provider due to their long travel distance. Each provider’s demand is not affected other providers’ price and quality. Second, there can be a competitive equilibrium in which every patient joins the list of a provider. In this case a provider can attract more patients from adjacent competitors by lowering its price and/or increasing its quality. Third, in between the monopoly equilibrium and the competitive equilibrium, there can be a kink equilibrium where the market is just covered.

Gravelle (1999) showed that strong conditions must be imposed on preferences and costs for quality to be efficiently supplied. For example, for any given number of firms, the market equilibrium has welfare maximizing quality for all values of the welfare weight on consumers if and only if (1) consumer preferences are weakly separable in distance, and consumers have zero income elasticity of demand for quality; and (2) the cost function is linear in quantity.

The framework allows one to examine the efficiency of the equilibrium number of providers. In a monopoly equilibrium an additional firm enables extra consumers to be served. The marginal firm, however, only considers its profit and not the benefit to the extra consumers. Consequently, there are too few providers in equilibrium. In contrast, in the competitive equilibria, price and quality vary with the number of firms. Because provider profit exceeds the marginal social value of an additional provider and so conveys the wrong entry signal, it can be shown that under certain restrictions on the welfare weight on consumers, the competitive equilibria has too many firms.

Moreover, consistent with the base model, the equilibrium quality level increases with the regulated payment. Raising the fee makes marginal patients more valuable and induces the providers to compete for them by raising quality. As Gravelle (1999) pointed out, this result implies that the regulated payment is an effective instrument for the regulator to control quality even if quality is not observable by her. Without monitoring the quality level of individual providers, the regulator can achieve the outcome of an unregulated private market to induce optimal quality by setting the regulated payment equal to the private market equilibrium fee.

Imperfect Information About Quality and Switching Costs

In the base model, we assume that the providers’ quality is commonly known to patients before they choose the providers. In reality, however, healthcare services can be a good experience because patients may not know the actual quality of a provider prior to joining its list or receiving treatment. Moreover, upon finding out the true quality of a provider, a patient may want to switch to another provider. The switching can be costly. For example, the patient may need to register with the new provider, meet new doctors and medical staff, and go through new medical tests.

Gravelle and Masiero (2000) examined the extent to which switching costs and imperfect patient information about provider quality affect providers’ incentives for quality. They consider a Hotelling linear market similar to the one in the base model. But they consider two periods. At the beginning of the first period, the patients are young. At the end of the first period, the patients become old, and a proportion of them are replaced by a new generation of young patients. At the end of the second period, all patients leave.

In each period, a patient can choose to join the list of a provider. Before joining the list, a young patient has imperfect information about the quality of both providers and observes the quality of each provider with an error comprising an additive component and a component that multiplies the quality level.

In contrast, an old patient has perfect knowledge of the quality provided by his first-period provider. The old patient’s information about the quality of the other provider remains the same as when he was young. At the second-period provider, an old patient can pay a switching cost and switch to the other provider.

Gravelle and Masiero (2000) found that the additive error component of the imperfect information and the switching cost have no effect on equilibrium quality. This is because the additive error and the switching cost enter into the demand functions of both providers and are equal and offsetting. This implies that they do not affect the marginal revenue from quality changes. As a result, they have no effect on the providers’ quality choice. In contrast, in equilibrium the providers’ quality is decreasing in the multiplicative error. The intuition is as follows. The multiplicative error means that higher quality leads to greater errors. The greater the error, the more likely old patients are to switch to the other provider. Consequently, the benefit to the providers from increasing quality is reduced, and thus they respond to a higher multiplicative error with lower quality. Following this line of argument, it is further shown that the socially optimal quality is also reduced by the multiplicative error. If there is a multiplicative error, the optimal quality becomes lower as the switching cost increases and the travel cost decreases.

Gravelle and Masiero (2000) showed that if the capitation fee (the fee per patient enrolled) is the regulator’s sole instrument, then it will not lead providers to choose simultaneously the socially optimal levels of quality and effort to reduce patient errors. Nor will the capitation fee lead the providers to choose simultaneously the socially optimal levels of quality and switching costs. When patient information about provider quality and switching costs can be endogenously set by providers, the optimal capitation fee will be a second-best compromise balancing its effects on quality, information, and effort to reduce switching costs. This suggests the scope for additional policy instruments to influence quality, information provision, and switching costs.

Gravelle and Sivey (2010) also consider a set-up with quality competition between two providers and imprecise information about provider quality. They consider policies that change the precision of information about provider quality (hospital “report cards”), which affects the degree of competition and quality levels in equilibrium. In this model, increases in information precision increase equilibrium quality levels apart from when one provider has much higher costs of increasing quality than the other provider.

Quality as a Stock and Quality Investment Over Time

So far, we have focused on static settings where quality is assumed to be determined by a “one-shot” investment and remains constant. In reality, however, quality may change over time. For example, surgery facilities may depreciate as they age, and medical knowledge may become obsolete. In a dynamic setting, Brekke et al. (2010) adopted a differential-game approach and model quality as a stock within the Hotelling framework. Being a stock means that quality can deteriorate. On the other hand, quality can be increased if investment in it is higher than its deterioration.

Brekke et al. (2010) compared the equilibrium quality in two solution concepts. First, if each provider only knows the initial quality of the other providers but not their quality in the following periods, then a provider has to decide its optimal dynamic plan of quality investment at the beginning and adhere to it for the duration. The corresponding solution concept is called an open-loop solution. Second, if a provider not only knows the initial quality of other providers but also observes their quality at all the subsequent periods, then the provider can adjust its quality investment in later periods according to the observed quality (the state). Specifically, if each provider’s quality investment is only based on the current states, then the solution concept is called a (feedback) closed-loop solution.

If the marginal cost of treating patients is constant, the open-loop and closed-loop solutions coincide: in the steady-state, where values do not change over time, investment and quality are identical under the two solution concepts. The intuition is as follows. With a fixed regulated fee per patient and a constant marginal cost of treatment, it can be shown that the marginal revenue gain of quality investments is constant. This implies that the optimal investment rule for each provider is independent of the other providers’ quality level. That is, the strategic interaction among the providers in respect of quality investment is absent. The key insight of this result is that when the marginal cost of treatment is constant, the solution within a dynamic approach is qualitatively similar to the ones obtained within a static approach. Thus, we can say that the static analysis is reasonably robust in this particular case.

However, if the marginal cost of treating patients is increasing, then the steady-state investment and quality are lower under the closed-loop solution than under the open-loop solution. When the marginal cost of treatment is increasing, closed-loop investment rules imply that lower quality investment by one provider will induce a future reduction in quality investment by the other providers. For each provider the instantaneous demand loss due to lower quality investments can be outweighed by the future gains following the strategic response—lower investment—by the competitor. As a result, the providers’ incentive for quality investments is reduced, and, compared with the open-loop solution or with the solution of a static game, they obtain a more collusive outcome.

When the marginal cost of treating patients is increasing, the difference between the closed-loop and the open-loop solutions implies that the beneficial effects from competition in terms of higher quality can be lower than expected from previous theoretical literature that is focused on static settings. Henceforth, it provides an explanation to some contrasting results in empirical studies, which are reviewed in the following section.

Empirical Studies on Competition and Quality

The 21st-century literature has shown a proliferation of empirical studies of quality competition in healthcare. In this section we split the empirical analysis of competition into two parts. Firstly, we discuss studies that estimate the degree to which patients choose higher quality hospitals. Quality influencing choice of provider is a prerequisite for quality competition. Secondly, we review studies that directly estimate the effects of increased competition on quality levels of healthcare providers. Some of the studies reviewed here use “hospital report cards” reforms to identify their effects. These studies estimate the effects of publishing quality information on patients’ choice of provider or on provider quality itself. These studies reinforce the importance of information and switching costs in determining the amount of competition in the market.

While we try and give an international perspective, many of the most significant empirical studies on quality and competition use data from the U.K. National Health Service (NHS). Several reasons explain this feature of the recent literature. Firstly, the United Kingdom is a large single-payer health system with few complications from private health insurance or price competition. Data on healthcare services (especially on hospital care) is relatively rich and easily available. Finally, the United Kingdom went through a series of reforms in the mid-2000s which explicitly focused on introducing more “choice and competition” into the healthcare system. We make sure to also review key studies from the United States and Europe to compare and contrast results across alternative institutional contexts.

Do Patients Choose Higher Quality Providers?

A prerequisite for quality competition is that patients (or their surrogates such as primary care physicians) choose healthcare providers on the basis of quality. A substantial strand of research has been directed toward answering this question: Do patients choose higher-quality hospitals if information and choice is available to them?

Several studies from the UK use a set of “choice and competition” reforms in the mid-2000s as a natural experiment to aid their econometric identification strategies. The reforms involved gave all NHS planned (elective) care patients an explicit choice of provider for planned procedures from January 1, 2006. A new website was rolled out to facilitate this choice (“Choose and Book”) and referring general practitioners (GPs) were trained in how to advise patients in how to use the system. These choice reforms were coupled with reform to hospital reimbursement (“Payment by Results”), which involved switching to DRG-style prospective reimbursement, thus giving hospitals a financial incentive to attract more patients.

Gaynor, Propper, and Sieler (2016) used mortality rates as the primary measure of hospital quality and take a structural approach to modeling choices available to patients through “consideration sets.” Their results show the 2006 reforms in the United Kingdom lead to coronary-artery bypass graft (CABG) patients choosing higher-quality hospitals, with their quality-elasticity of demand increasing substantially following the reform.

Other evidence from the United Kingdom has focused on choice of hospital for hip replacements (Beckert, Christensen, & Collyer, 2012; Gutacker, Siciliani, Moscelli, & Gravelle, 2016; Moscelli, Siciliani, Gutacker, & Gravelle, 2016). All three of these studies confirm that patients tend to choose hospitals that perform better on measures of quality including lower risk-adjusted mortality and MRSA rates (Beckert et al., 2012) lower emergency readmission, revision and mortality rates (Moscelli et al., 2016), and better patient-reported outcomes (Oxford Hip Score) (Gutacker et al., 2016). Moscelli et al. (2016) used data for a 10-year period from 2002 to 2011 and showed that the size of the effect of quality on choice of provider appeared to increase through the period of the “choice and competition” reforms. Gutacker et al. (2016) compared their results using patient-reported measures of quality with more traditional measures (mortality and readmission) and found the quality elasticity of demand is larger when using patient-reported quality measures.

In summary, the literature on quality and choice of provider in the United Kingdom seems to give a convincing picture that patients do choose higher-quality hospitals for planned procedures, especially after the 2003–2004 “choice and competition” reforms.

Evidence from outside the United Kingdom has also confirmed that patients’ choice of hospital is positively associated with measures of quality. In the Netherlands, studies have shown that choice of hospital is associated with measures of readmission and reputation for angioplasty patients (Varkevisser & van der Geest, & Schut, 2012) and with a composite hospital quality measure for hip-replacement patients (Beukers, Kemp, & Vakevisser, 2014). Although the Netherlands healthcare system makes extensive use of private insurance, prices for providers (i.e., hospitals) are regulated, and patients are encouraged to choose hospitals based on publicly reported quality measures.

In contrast, some studies have suggested that in countries such as Italy where hospital quality measures are collected but not publicly reported, there is no evidence of association between quality measures and hospital choice (Moscone, Tosetti, & Vittadini, 2012).

The literature in the United States has also found associations between hospital choice and quality measures. Notable studies include Tay (2003), which analyzes choice of hospital emergency department for acute myocardial infarction patients and finds that patients are more likely to choose hospitals with lower mortality rates and better treatment facilities. Jung, Feldman, and Scanlon (2011) used a stated preference approach to compare patients individual stated preferences over hospital attributes with their rankings of nearby hospitals. They find consumers’ perceptions of reputation and medical services as well as objective quality scores are determinants of their rankings of nearby hospitals, even without any public reporting of quality information.

The high-profile paper Chandra, Finkelstein, Sacarny, and Syverson (2016) provided a big-picture theoretical and empirical study analyzing patient choice of hospital using U.S. Medicare inpatient data on common emergency and planned procedures. They show that higher-quality hospitals are likely to have higher market shares and a larger growth in market share over time. They also show among choices made by patients with the most capacity to exercise consumer choice—for example, patients transferring from another hospital—patients are more likely to flow toward high-quality hospitals than among patients who have little opportunity for choice (e.g., patients visiting the emergency department).

A distinguishing feature of the U.S. literature is a particular focus “report cards,” the information provided to patients about hospital or health plan quality. Most of these studies compare patients’ choices before and after the introduction of a new report cards scheme to see whether highly rated hospitals/plans increase their market share (Dafny & Dranove, 2008; Dranove & Sfekas, 2008; Pope, 2009; Wang, Hockenberry, Chou, & Yang, 2011). These studies show positive effects of quality measures on market share (or patient choices), although sometimes with modest effect sizes and sensitive to model specification (Dranove & Sfekas, 2008). The effects of report cards on patient choice seems to be higher when they are more easily understood by patients: for example, in the case of the “America’s Best Hospitals” rankings (Pope, 2009).

While studies on choice of hospital are relatively numerous, there is less known about patients’ choice of doctor for primary care or other out-of-hospital care. Santos, Gravelle, and Propper (2017) analyzed U.K. data in 2010–2011 and showed that patients’ choices of GP practices were associated with quality as measured by Quality and Outcomes Framework (QoF) points in 2006. Crucially, these QoF quality measures are published by a government website, NHS Choices, which also contains quality information about hospitals. Other studies, for example in Norway, have modeled GP choice as a function of distance and GP characteristics such as gender or education but have not included explicit measures of quality (Godager, 2012).

The Effect of Competition on Quality

Next, we move on to discuss studies that directly estimate the effects of increased competition on quality levels of healthcare providers. Empirical studies in this area have increasingly employed modern econometric approaches to establish causality. Many studies have used a difference-in-differences approach to evaluate policy reforms that introduced a greater degree of competition into healthcare markets. Another key empirical strategy has been using instrumental variables to isolate variation in “competition” that can plausibly be considered to be exogenous.

The most notable strand of literature especially focused on quality of care as an outcome is a series of studies evaluating the effects of the 2006 “choice and competition” reforms in the U.K. National Health Service (Cooper, Gibbons, Jones, & McGuire, 2011; Gaynor, Moreno-Serra, & Propper, 2013). These studies showed a fall in mortality from acute myocardial infarction (AMI) (Cooper, Gibbons, Jones, & McGuire, 2011), as well as all-cause mortality (Gaynor et al., 2013) for hospitals in more competitive areas compared to those in less competitive areas over the period of the reform. Other measures of hospital performance, such as waiting times and hospital-acquired infections, appeared to be unaffected by competition during the period of the reform.

Criticisms of these studies have focused on the use of AMI mortality as a poor measure of overall hospital quality and the inability of hospital admissions data to correctly account for casemix (Mays, 2011). Some other studies also seem to show apparently conflicting evidence. Moscelli, Gravelle, Siciliani, and Santos (2018) finds an effect with a similar specification to the previous papers but no effect on AMI mortality with a more elaborate specification accounting for changes in casemix. Another study with measures of planned surgery quality (Moscelli, Gravelle & Siciliani, 2018) provides some evidence quality did not change (or even fell) for planned patients. The authors argue the results need not be inconsistent with the earlier studies of Cooper et al. (2011) and Gaynor et al. (2013) if there are differences in the costs of improving quality between emergency and planned care. In particular, Moscelli et al. (2018) argued that the relative unprofitability of certain planned treatments (hip and knee replacements) could have contributed to the fall in quality associated with competition for these treatments. These results are not necessarily contradictory in light of the theoretical literature showing increased competition could lead to increases or reductions in equilibrium quality levels under different assumptions about institutions and technology (Gravelle & Sivey, 2010; Katz, 2013).

Another group of U.K. papers has focused on using hospital mergers to measure if less competition in a local area following a merger leads to better or worse outcomes using a variety of quality and other measures (Gaynor, Laudicella, & Propper, 2012; Bloom et al., 2015). These papers use a feature of U.K. politics: that hospitals are less likely to be subject to a merger by the government if they are in politically “marginal” areas, to find exogenous variation in merger activity across areas. The first study uses data from the “pre-competition” period (Gaynor et al., 2012) and finds either no effects of mergers, or a deterioration associated with mergers, on a variety of outcomes, including waiting times and financial performance but also clinical quality as measured by mortality and readmission rates. The later study uses cross-sectional data from a “management quality” survey of NHS hospitals in 2006, just after the “choice and competition” reforms (Bloom et al., 2015) linked with a measure of the degree of hospital competition in the area, which is predicted in a first stage by the political “marginality” of the area. The study finds that areas that have more hospital competition, due to being politically “marginal,” have better management quality as well as lower AMI mortality, higher productivity, and a lower intention of staff to leave the hospital. These results seem to support the results of the difference-in-differences studies that found positive effects on quality (Cooper et al., 2011; Gaynor et al., 2013) but with a very different cross-sectional identification approach. However, it seems possible that the lack of hospital mergers in “marginal” areas could affect quality through other channels apart from competition.

An important comparison for the contemporary literature on the United Kingdom’s 2006 “choice and competition” reforms is an earlier set of studies on the 1990s U.K. reforms. These reforms found increasing competition associated with worse quality as measured by higher mortality (Propper, Burgess, & Green, 2004; Propper, Burgess, & Gossage, 2008). This finding has been explained by reference to the lack of publicly reported quality measures in the United Kingdom in the 1990s, as well as the focus at that time on waiting times (which were publicly reported) and price negotiations in the “internal market” of the NHS. The implication of these earlier findings is therefore not necessarily that competition cannot improve quality but that the policy settings in the United Kingdom during this earlier period did not facilitate quality competition.

Robust evidence on competition and quality from elsewhere in Europe is scarce, possibly because of poor data availability but probably more because a lack of policy focus on competition as an explicit goal in the healthcare sector. Evidence from Italy, where quality outcomes are not reported to the public, shows no association between competition measures and quality (Berta, Martini, Moscone, & Vittadini, 2016). This finding could reflect the already high regulated levels of quality in Italy or that with a lack of the necessary prerequisites for quality competition there are unlikely to be any effects of competition measures.1

The literature in the United States has tended to focus on the role of “report card” policies that increase the visibility of hospital quality measures to patients. Chou, Deily, Li, and Lu (2014) uses a report cards reform in 1998 in Pennsylvania to study the role of competition in promoting quality of care. The authors find hospitals in more competitive markets increased their quality (as measured by mortality) and also spent more resources per patient as a result of the reform. Cutler, Huckman, and Landrum (2004) found that hospitals identified as poor performers (high mortality) improved their quality levels (i.e., mortality fell) following the introduction of report cards that published mortality rates, whereas quality did not change for high performers following the reform. An important early study by Dranove et al. (2003) highlights another incentive associated with report card policies: improving quality scores through risk selection of patients. This study provides an important caveat to all the evidence using quality measures such as mortality or other failure rates. If a pro-competition reform affects the incentive to perform well on these quality measures, providers have an incentive to game their quality reports through risk selection, as well as to engage in genuine quality improvement.

An innovative study from Kolstad (2013) using surgeon-level mortality reports provides more information about the mechanisms by which report cards may lead to quality improvement. As some surgeons have stronger competitive financial incentives to attract more patients than others, the study can disentangle whether financial or non-financial (intrinsic) motivations are most important in driving quality improvement. He finds that intrinsic motivation is a much more important driver of quality improvements than financial (competitive) incentives. This finding therefore doubts the importance of competition per se in the quality-improving effects of report-card policies.

Overall, compared to the literature on quality and choice of provider, the evidence that competition improves quality in healthcare is less clear. Perhaps reflecting the nuances of the theoretical literature, showing that different parameter combinations or assumptions can contradict the simple intuitive result that competition improves quality, empirical results are also mixed. While a series of quality U.K. studies appear to show strong improvements in more competitive areas following the mid-2000s reform introducing more choice and competition, later studies showed that these improvements could not be found for planned care or that the finding may be sensitive to specification. Furthermore, results in other contexts are mixed, and there is some evidence that it may not be a traditional form of competition that drives quality improvements but rather a form of “intrinsic” motivation following more focus on published quality measures.


While encouraging healthcare providers to compete to improve quality is a reasonable goal for policymakers, the underlying economics of quality competition in healthcare is complex. Many features of the market for healthcare can affect how providers compete and by how much quality can be increased: the degree of horizontal differentiation of providers, the level of prospective reimbursement for different types of patients, information about quality, switching costs, provider altruism, and the time horizon of quality investment decisions. However, most models do show that increases in competition—through freeing up patient choice, an increase in providers, or less horizontal differentiation—lead to an increase in quality as providers strive to attract more patients.

On the empirical side, there is strong evidence that a prerequisite for quality competition is met. Patients do tend to choose higher-quality providers, especially after greater patient choice is made available or information is published about provider quality. As suggested by the theoretical literature, the question of whether and how much increases in competition increase quality levels is a more mixed picture. Some key studies show a positive effect of pro-competition reforms on quality levels, but others using different quality measures or different specifications find a weaker or nulling effect.

Future research on this topic has many promising potential directions. There is more work to be done testing behavioral hypotheses around what (apart from pure financial incentives) motivates providers to improve healthcare quality. Future studies may extend the initial work in unpacking the black box of quality improvement to understand how providers actually improve patient outcomes. Above all future research will continue to discover the institutional and policy settings that optimize quality competition to ensure that patients get the maximum benefits from the healthcare market.

Further Reading

Bloom, N., C. Propper, S. Seiler, & J. Van Reenen. (2015). The impact of competition on management quality: Evidence from public hospitals. The Review of Economic Studies, 82(2), 457–489.Find this resource:

Brekke, K. R., Siciliani, L., & Straume O. R. (2011). Hospital competition and quality with regulated prices. Scandinavian Journal of Economics, 113(2), 444–469.Find this resource:

Chandra, A., Finkelstein, A., Sacarny, A., & Syverson, C. (2016). Health care exceptionalism? performance and allocation in the us health care sector. American Economic Review, 106(8), 2110–2144.Find this resource:

Chou, S.-Y., Deily, M. E., Li, S., & Lu, Y. (2014). Competition and the impact of online hospital report cards. Journal of Health Economics, 34, 42–58.Find this resource:

Gaynor, M., Ho, K., & Town, R. J. (2015). The industrial organization of health-care markets. Journal of Economic Literature, 53(2), 235–284.Find this resource:

Gaynor, M., Moreno-Serra, R., & Propper, C. (2013). Death by market power: Reform, competition, and patient outcomes in the National Health Service. American Economic Journal: Economic Policy, 5(4), 134–166.Find this resource:

Gravelle, H., & Sivey, P. (2010). Imperfect information in a quality-competitive hospital market. Journal of Health Economics, 29(4), 524–535.Find this resource:

Katz, M. L. (2013). Provider competition and healthcare quality: More bang for the buck? International Journal of Industrial Organization, 31(5), 612–625.Find this resource:

Kolstad, J. T. (2013). Information and quality when motivation is intrinsic: Evidence from surgeon report cards. American Economic Review, 103(7), 2875–2910.Find this resource:

Siciliani, L., Chalkley, M., & Gravelle, H. (2017). Policies towards hospital and GP competition in five European countries. Health Policy, 121(2), 103–110.Find this resource:


Beckert, W., Christensen, M., & Collyer, K. (2012). Choice of NHS‐funded hospital services in England. The Economic Journal, 122(560), 400–417.Find this resource:

Berta, P., Martini, G., Moscone, F., & Vittadini, G. (2016). The association between asymmetric information, hospital competition and quality of healthcare: Evidence from Italy. Journal of the Royal Statistical Society: Series A (Statistics in Society), 179(4), 907–926.Find this resource:

Beukers, P. D. C., Kemp, R. G. M., & Varkevisser, M. (2014). Patient hospital choice for hip replacement: Empirical evidence from the Netherlands. The European Journal of Health Economics, 15(9), 927–936.Find this resource:

Bloom, N., Propper, C., Seiler, S., & Van Reenen, J. (2015). The impact of competition on management quality: Evidence from public hospitals. The Review of Economic Studies, 82(2), 457–489.Find this resource:

Brekke, K. R., Cellini, R., Siciliani, L., & Straume, O. R. (2010). Competition and quality in health care markets: A differential-game approach. Journal of Health Economics, 29(4), 508–523.Find this resource:

Brekke, K. R., Nuscheler, R., & Straume, O. R. (2006). Quality and location choices under price regulation. Journal of Economics & Management Strategy, 15(1), 207–227.Find this resource:

Brekke, K. R., Siciliani, L., & Straume, O. R. (2008). Competition and waiting times in hospital markets. Journal of Public Economics, 92(7), 1607–1628.Find this resource:

Brekke, K. R., Siciliani, L., & Straume, O. R. (2011). Hospital competition and quality with regulated prices. Scandinavian Journal of Economics, 113(2), 444–469.Find this resource:

Chandra, A., Finkelstein, A., Sacarny, A., & Syverson, C. (2016). Health care exceptionalism? Performance and allocation in the U.S. health care sector. American Economic Review, 106(8), 2110–2144.Find this resource:

Chen, Y., Meinecke, J., & Sivey, P. (2016). A theory of waiting time reporting and quality signaling. Health Economics, 25(11), 1355–1371.Find this resource:

Chou, S.-Y., Deily, M. E., Li, S., & Lu, Y. (2014). Competition and the impact of online hospital report cards. Journal of Health Economics, 34, 42–58.Find this resource:

Cooper, Z., Gibbons, S., Jones, S., & McGuire, A. (2011). Does hospital competition save lives? Evidence from the English NHS patient choice reforms. The Economic Journal, 121(554), F228–F260.Find this resource:

Cutler, D. M., Huckman, R. S., & Landrum, M. B. (2004). The role of information in medical markets: An analysis of publicly reported outcomes in cardiac surgery. American Economic Review 94(2), 342–346.Find this resource:

Dafny, L., & Dranove, D. (2008). Do report cards tell consumers anything they don't already know? The case of Medicare HMOs. The Rand Journal of Economics, 39(3), 790–821.Find this resource:

Dranove, D., Kessler, D., McClellan, M., & Satterthwaite, M. (2003). Is more information better? The effects of ‘report cards’ on health care providers. Journal of Political Economy 111(3), 555–588.Find this resource:

Dranove, D., & Sfekas, A. (2008). Start spreading the news: A structural estimate of the effects of New York hospital report cards. Journal of Health Economics, 27(5), 1201–1207.Find this resource:

Gaynor, M., Laudicella, M., & Propper, C. (2012). Can governments do it better? Merger mania and hospital outcomes in the English NHS. Journal of Health Economics, 31(3), 528–543.Find this resource:

Gaynor, M., Moreno-Serra, R., & Propper, C. (2013). Death by market power: Reform, competition, and patient outcomes in the National Health Service. American Economic Journal: Economic Policy, 5(4), 134–166.Find this resource:

Gaynor, M., Propper, C., & Seiler, S. (2016). Free to choose? Reform, choice, and consideration sets in the English National Health Service. American Economic Review, 106(11), 3521–3557.Find this resource:

Godager, G. (2012). Birds of a feather flock together: A study of doctor–patient matching. Journal of Health Economics, 31(1), 296–305.Find this resource:

Gravelle, H. (1999). Capitation contracts: Access and quality. Journal of Health Economics, 18(3), 315–340.Find this resource:

Gravelle, H., & Masiero, G. (2000). Quality incentives in a regulated market with imperfect information and switching costs: Capitation in general practice. Journal of Health Economics, 19(6), 1067–1088.Find this resource:

Gravelle, H., & Sivey, P. (2010). Imperfect information in a quality-competitive hospital market. Journal of Health Economics 29(4), 524–535.Find this resource:

Gutacker, N., Siciliani, L., Moscelli, G., & Gravelle, H. (2016). Choice of hospital: Which type of quality matters? Journal of Health Economics, 50, 230–246.Find this resource:

Hotelling, H. (1929). Stability in Competition. The Economic Journal, 39(153), 41–57.Find this resource:

Jung, K., Feldman, R., & Scanlon, D. (2011). Where would you go for your next hospitalization? Journal of Health Economics, 30(4), 832–841.Find this resource:

Katz, M. L. (2013). Provider competition and healthcare quality: More bang for the buck? International Journal of Industrial Organization, 31(5), 612–625.Find this resource:

Kolstad, J. T. (2013). Information and quality when motivation is intrinsic: Evidence from surgeon report cards. American Economic Review 103(7), 2875–2910.Find this resource:

Le Grand, J. (1991). Quasi-markets and social policy. The Economic Journal 101(408), 1256–1267.Find this resource:

Longo, F., Siciliani, L., Gravelle, H., & Santos, R. (2017). Do hospitals respond to rivals' quality and efficiency? A spatial panel econometric analysis. Health Economics, 26, 38–62.Find this resource:

Marshall, M. N., Shekelle, P. G., Davies, H. T., & Smith, P.C. (2003). Public reporting on quality in the United States and the United Kingdom. Health Affairs, 22(3), 134–148.Find this resource:

Mays, N. (2011). Is there evidence that competition in healthcare is a good thing? No. BMJ: British Medical Journal, 343, d4205.Find this resource:

McPherson, K., Wennberg, J. E., Hovind, O. B., & Clifford, P. (1982). Small-area variations in the use of common surgical procedures: An international comparison of New England, England, and Norway. New England Journal of Medicine, 307(21), 1310–1314.Find this resource:

Moscelli, G., Gravelle, H., & Siciliani, L. (2018). Effects of market structure and patient choice on hospital quality for planned patients. Guildford, UK: University of Surrey, School of Economics.Find this resource:

Moscelli, G., Gravelle, H., Siciliani, L., & Santos, R. (2018). Heterogeneous effects of patient choice and hospital competition on mortality. Social Science & Medicine, 216, 50–58.Find this resource:

Moscelli, G., Siciliani, L., Gutacker, N., & Gravelle, H. (2016). Location, quality and choice of hospital: Evidence from England 2002–2013. Regional Science and Urban Economics, 60, 112–124.Find this resource:

Moscone, F., Tosetti, E., & Vittadini, G. (2012). Social interaction in patients’ hospital choice: Evidence from Italy. Journal of the Royal Statistical Society: Series A (Statistics in Society) 175(2), 453–472.Find this resource:

Pearson, S. D., & Rawlins, M. D. (2005). Quality, innovation, and value for money: NICE and the British National Health Service. Jama, 294(20), 2618–2622.Find this resource:

Pope, D. G. (2009). Reacting to rankings: Evidence from ‘America's Best Hospitals’.” Journal of Health Economics, 28(6), 1154–1165.Find this resource:

Propper, C., Burgess, S., & Gossage, D. (2008). Competition and quality: Evidence from the NHS internal market 1991–9. The Economic Journal 118(525), 138–170.Find this resource:

Propper, C., Burgess, S., & Green, K. (2004). Does competition between hospitals improve the quality of care? Hospital death rates and the NHS internal market. Journal of Public Economics 88(7–8), 1247–1272.Find this resource:

Santos, R., Gravelle, H., & Propper, C. (2017). Does quality affect patients’ choice of doctor? Evidence from England. The Economic Journal, 127(600), 445–494.Find this resource:

Salop, S. C. (1979). Monopolistic competition with outside goods. The Bell Journal of Economics, 10(1), 141–156.Find this resource:

Siciliani, L., Moran, V., & Borowitz, M. (2014). Measuring and comparing health care waiting times in OECD countries. Health Policy, 118(3), 292–303.Find this resource:

Tay, A. (2003). Assessing competition in hospital care markets: The importance of accounting for quality differentiation. RAND Journal of Economics, 786–814.Find this resource:

Varkevisser, M., van der Geest, S. A., & Schut, F. T. (2012). Do patients choose hospitals with high quality ratings? Empirical evidence from the market for angioplasty in the Netherlands. Journal of Health Economics, 31(2), 371–378.Find this resource:

Wang, J., Hockenberry, J., Chou, S.-Y., & Yang, M. (2011). Do bad report cards have consequences? Impacts of publicly reported provider quality information on the CABG market in Pennsylvania. Journal of Health Economics, 30(2), 392–407.Find this resource:

Wennberg, J., & Gittelsohn, A. (1973). Small area variations in health care delivery: A population-based health information system can guide planning and regulatory decision-making. Science, 182(4117), 1102–1108.Find this resource:

Wennberg, J. E. (2002). Unwarranted variations in healthcare delivery: Implications for academic medical centres. BMJ: British Medical Journal, 325(7370), 961.Find this resource:

Wennberg, J. E. (2011). Time to tackle unwarranted variations in practice. BMJ: British Medical Journal, 342, 1413.Find this resource:


(1.) See, for example, the relatively low heart attack and stroke mortality rate in Italy compared to other developed countries on the Nuffield Trust site.