Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, ECONOMICS AND FINANCE ( (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 21 November 2019

Price Regulation and Pharmaceuticals

Summary and Keywords

Pharmaceutical expenditure accounts for approximately 20% of healthcare expenditure across the Organisation for Economic Cooperation and Development (OECD) countries. Pharmaceutical products are regulated in all major global markets primarily to ensure product quality but also to regulate the reimbursed prices of insurance companies and central purchasing authorities that dominate this sector. Price regulation is justified as patent protection, which acts as an incentive to invest in R&D given the difficulties in appropriating the returns to such activity, creates monopoly rights to suppliers. Price regulation does itself reduce the ability of producers’ to recapture the substantial R&D investment costs incurred. Traditional price regulation through Ramsey pricing and yardstick competition is not efficient given the distortionary impact of insurance holdings, which are extensive in this sector and the inherent uncertainties that characterize Research and Development (R&D) activity. A range of other pricing regulations aimed at establishing pharmaceutical reimbursement that covers both dynamic efficiency (tied to R&D incentives) and static efficiency (tied to reducing monopoly rents) have been suggested. These range from cost-plus pricing, to internal and external reference pricing, rate-of-return pricing and, most recently value-based (essential health benefit maximization) pricing. Reimbursed prices reflecting value based pricing are, in some countries, associated with clinical treatment guidelines and cost-effectiveness analysis. Some countries are also requiring or allowing post-launch price regulation thorough a range of patient access agreements based on predefined population health targets and/or financial incentives. There is no simple, single solution to the determination of dynamic and static efficiency in this sector given the uncertainty associated with innovation, the large monopoly interests in the area, the distortionary impact of health insurance and the informational asymmetries that exist across providers and purchasers.

Keywords: pharmaceuticals, pricing and reimbursement regulation, R&D incentives, patents, monopoly power, health economics


Pharmaceutical treatments from a significant part of healthcare expenditure in most countries. Across the OECD countries pharmaceutical expenditure accounts for approximately 20% of total healthcare spending, with consumption increasing recently largely due to population aging and an increasing prevalence of chronic diseases; although this varies greatly with around 12% of total health spending in the larger economies to over 20% of spending in the smaller countries. Pharmaceutical prices have largely stabilized recently due in part to increased regulation. At least a third of OECD countries have introduced policies aimed at reducing pharmaceutical reimbursement prices since 2008, in response to worsening economic conditions feeding through public expenditure cuts into restrictions on healthcare expenditure growth. These measures have included explicit cuts in ex-factory prices, a reduction in distribution margins, changes in VAT rates imposed on pharmaceuticals, increased user charges, reforms of individual pricing systems, and the increasing use of generic products.

Generic use has also increased as a result of patents for a number of important products expiring. The impact has been that pharmaceutical expenditure has grown at a lower rate than other healthcare sector inputs since the mid-2000s. However, specialty drugs (particularly in oncology) and biologics are seeing large increases in prices, which has not curtailed consumption: so the expectation is that spending on pharmaceuticals will grow (OECD, 2016). This growth may be unevenly spread across diseases as well as countries. Expenditure on pharmaceuticals for the treatment of oncology is predicted to see the largest growth rates, even though the cost of new cancer drugs increased by 10% per annum in real terms between 1995 and 2013 (OECD, 2016). The United States continues to be the largest pharmaceutical market, accounting for over 40% of global revenue, and is expected to grow by 6% per annum between 2016 and 2024 due to the coverage expansion associated with the Affordable Care Act and population aging (Keehan et al., 2015). While it is anticipated that the largest European markets will either see zero to negative growth (France and Spain), or modest 2% per annum growth rates (United Kingdom and Germany) (Institute for Healthcare Informatics, 2014) over the period 2017 to 2020.

Pharmaceutical products are explicitly regulated in all of the major global markets primarily to ensure product quality but also to regulate price. Price regulation is justified as patent protection, which acts as an incentive to invest in R&D given the difficulties in appropriating the returns to such activity, creates monopoly rights to suppliers. Price regulation, of course, may itself reduce the ability of producers to recapture the substantial R&D investment costs incurred. It is estimated that only one in 10,000 compounds are marketed successfully. While drug development costs have risen nearly 600% over the past 30 years, such costs may rise at a greater pace as R&D addresses more intractable diseases such as cancer and neurological disease. According to DiMasi et al. (2016) it costs $2.5 billion (2013 prices) and 11 years (median time) to bring an individual product to market, given that companies are dealing with the difficulties of managing an innovative pipeline that incorporates a portfolio of products, many doomed to failure, as well as increasing regulatory costs. Although there is some optimism to be gained from the increase in New Molecular Entities (NMEs) before the FDA and EMA in the past, success rates associated with NMEs have fallen from 1 in 5 in the 1980s to 1 in 10 in the 2000s.

Vernon (2005) and Civa and Maloney (2007) both argue that while price regulation is justified given the potential to pursue rent-seeking activity—as it is difficult to define the optimal length and breadth of patent protection—levels of price regulation will also have a direct impact on R&D activity, and the degree of price regulation that is the fundamental driver of R&D within this sector. Neither defines the optimal level of R&D. Their empirical analysis finds that the number of drugs in a therapeutic category depends significantly on the established prices in that same category. The latter study also found that drug development was elastic to prices at values of 28–49%.

It is clear that there is an interaction between the design of price regulation aimed at optimizing current welfare, and patent protection aimed at providing incentives to produce new products, which determine future welfare levels. Current welfare is associated with static efficiency, where concern is with the distortionary impact of patent protection on pricing and access; while future welfare concerns, so-called dynamic efficiency, aim to ensure that the distortionary impact of price regulation on innovation is minimized. Jena and Philipson (2008) illustrate that the interaction between static and dynamic efficiency is not straightforward. They show that product price decreases may increase static efficiency through increasing utilization, but the effect on dynamic efficiency is indeterminate. Essentially this is because the price decreases influence utilization, and subsequently the marginal benefit of treatment, which in turn has an impact on the incentive to innovate.

Moreover, static efficiency concerns may dominate in any given country, as R&D facilities are concentrated within a small number of countries. Currently R&D is concentrated mainly within the United States, United Kingdom, France, Germany, and Switzerland, although there is some evidence of increasing concentration within the United States (EC, 2009). Smaller markets with little or no R&D interests may concentrate on the static efficiency concerns associated with price regulation, effectively free-riding on R&D incentives put in place elsewhere. The interaction between individual country revenue returns and the financing of R&D investments further complicates price setting and regulation. Companies are aware of the substantial interactions of country markets as they set prices for different countries as they respond to country specific pricing regulations, including the explicit use of international reference pricing by a number of countries. Moreover, pricing regulations may have a deterrent effect on product market entry and patient access to new products.

Given the global pricing perspective of the pharmaceutical companies some form of price discrimination is likely to operate across individual country markets. Szymanski and Valletti (2005) show, with a model of product quality determined by endogenous R&D, that price discrimination may indeed lead to higher R&D levels than uniform pricing. It is difficult to determine how such price discrimination might be enacted. An obvious means of implementing price discriminating across countries would be with respect to levels of GDP per capita; wealthier countries ought to be able to afford to pay higher prices. Danzon and Furukawa (2011) and Morel et al. (2011) found no relationship between price and GDP per capita levels across a range of high- and middle-income countries, however. If price discrimination is practiced systematically it remains unclear what aspect is used to identify the practice.

The impact that country-specific price regulation has on time to market launch, has been analyzed by a number of studies and found to have significant impact. Cockburn et al. (2016), Danzon and Epstein (2008), Kyle (2006, 2007), Lanjouw (2005), and Costa-Font et al. (2015) have all found that price regulation is associated with launch delays and in some cases even acted as a deterrent to launch in some countries. Cockburn et al. (2016) do find that more extensive patent rights tend to offset this effect of pricing regulation and accelerate product market launches. All these studies are empirical and do not consider or forward conceptual frameworks to analyze aggregate welfare effects.

Indeed, generally, there has been little written on optimal patent protection in the pharmaceutical market specifically. There is a literature on the impact of patents generally on markets and welfare, although the conclusions are weak. Tirole (1988) summarizes the general literature noting that given the uncertain returns from the investment, defining optimal levels of R&D is inherently difficult and that the welfare effects of patent protection depend on patent strength. Weak patent protection tends to lead to under-investment in R&D, while strong patent protection tends to lead to duplication and over-investment in R&D. The seminal work by Scherer (1972) and Nordhaus (1969) shows conceptually that welfare gains and losses due to changes in patent duration depend on price elasticities of demand, highlighting the interaction between patent protection and price.

Turning to studies that have explicitly considered the pharmaceutical industry, Arora et al. (2003) found a 10% patent premium on pharmaceutical products compared to the return earned by other sectors, implying a degree of over-protection. Although Berndt et al. (2015) recently found that while the net return on R&D investment was positive through the period 1995–2004, it has peaked and from 2005–2009 the return on R&D investment within the pharmaceutical sector was negative. They argue that if this state of affairs persists historical rates of biochemical R&D will be difficult to sustain.

Hughes et al. (2002) compared a patent scenario with that of immediate generic competition, illustrating that while a no-patent strategy would yield larger surpluses in terms of present consumer welfare, future consumers would lose three times the gained amount in present value terms. This dynamic relationship was analyzed by Horowitz and Lai (1996) also, with findings that the patent length that maximizes the rate of innovation exceeded the consumer surplus’ maximizing patent length.

Beyond the static versus dynamic efficiency relationship, a further complication affecting the price regulation of pharmaceutical products is that most healthcare purchases, including pharmaceuticals, are associated with the holding of (private or publicly provided) insurance. The presence of health insurance has an impact on consumption through well-known moral hazard issues arising from the distortion of price elasticity. In general, optimal insurance with moral hazard gives rise to partial coverage even if insurance premiums are actuarially fair and coverage is complete.

To counter moral hazard, some financial risk in the form of cost sharing may be shifted to the consumer. The optimal level of cost sharing is related to the price elasticity of the specific product. As the elasticity of demand for pharmaceuticals is held to be inelastic, generally cost-sharing will have to be significant to counter moral hazard. This may lead to further welfare distortion if individuals avoid the purchase of pharmaceuticals with the subsequent substitution of more expensive forms of healthcare as their health state worsens. However, the interaction between insurance, pharmaceutical pricing, and purchase is complex, and the impact and definition of moral hazard is itself problematic. By changing the relative price of medical care generally, insurance incorporates an income effect that, as noted by Nyman (2003), the pooling effect that underlies insurance allows individuals to purchase healthcare that they otherwise could not have afforded. This insurance (income) benefit should be accounted for when defining levels of moral hazard.

Moreover, as Grootendorst (2012) has also noted, patients make consumption decisions in tandem with their physician; therefore the demand-side impact of financial cost sharing will be weakened given that the physician bears no cost in this risk bearing by the consumer. Ideally pricing regulation should affect those making consumption decisions, even where demand is inelastic. Consumers play a limited role, and reliance on co-payment levels to regulate prices will have limited effect for the simple reason that it is physicians who make the main consumption (treatment) decisions, but they are largely unresponsive to changes in prices. There is a long-standing literature stating that physicians are price sensitive only to a limited degree, as they are not the insurance or financial risk holders (Danzon & Chou, 2000). There are then few clear-cut conclusions with respect to the degree of risk holding that cost sharing should impose on patients to define optimal expenditure (for further discussion see Pauly, 2011).

Physicians, in fact, play a multiple agency role; they are acting on behalf of their patients but also, through the resource allocation issues associated with their treatment prescriptions, making budgetary decisions for the insurance/funding bodies. These latter bodies hope that in making prescription decisions physicians weigh up the therapeutic benefits derived with the budgetary implications. Even under the cost-constrained conditions facing most healthcare systems today, physicians are liable to give more weight to therapeutic benefits rather than costs. Regardless of the way in which physicians are paid, this reflects their years of training and their professional peer-group pressures.

Nonetheless, some studies have used the insurance setting to argue that an optimal means of pricing pharmaceuticals is through setting consumer/patient cost sharing to the marginal costs ex post of the sunk costs of R&D expenditure, with an additional insurer/payer top-up to cover R&D expenses (Garber et al., 2006; Lakdawalla & Sood, 2013). The rationale is that the patient co-payments, through reflecting the marginal cost of the marketed product, would determine the appropriate quantity of demand and address issues of static efficiency. While the insurer/payer top-up would address issues of dynamic efficiency through ensuring the producer was adequately reimbursed for the R&D costs entailed in bringing the product to market. Of course, a number of practical concerns exist including the determination of the appropriate marginal cost, the appropriate level of insurer/payer top-ups (which itself relates to the optimal length and breadth of patent protection), the fact that co-payments may have to be large given the inelastic demand curves in the sector and that consumers/patients may rely on agents/physicians to make treatment decisions. Equity issues are important here as prices for specialty drugs may be extremely high, such that the co-pays arising from marginal cost pricing may themselves be high and/or require subsidy.

To summarize, one way of characterizing the aim of pharmaceutical price regulation is that it promotes the pursuit of productive efficiency and allocative efficiency, presuming technical efficiency has been achieved through establishment of some therapeutic outcome established through a risk-(therapeutic) benefit calculation. Productive efficiency is aimed at establishing optimal static and dynamic efficiency while promoting cost-effective decisions. For countries with little concern with dynamic efficiency, and therefore R&D decisions, productive efficiency may be limited to cost-effective prescribing. For countries that presume to have some impact on pharmaceutical R&D portfolios, either through the size or influence of their domestic market or through the importance of pharmaceutical R&D activities to the country, concerns with productive efficiency will also have to address dynamic efficiency concerns. Allocative efficiency is aimed at ensuring the optimal mix of prescribing output, based on maximizing therapeutic benefit versus budget impact, and must be defined in a manner appropriate to the individual country’s needs.

Reliance purely on price regulation to achieve productive and allocative efficiency is difficult under such circumstances, as there is one instrument (price) attempting to achieve cost-effective prescribing, with the appropriate mix of prescribing as based on therapeutic benefit, as well as providing incentives to promote long-term R&D needs. Unless therapeutic benefit can be internalized to promote R&D decisions, reimbursement and prescribing practice (as determined through price alone) provides the insurer/funder limited instruments to hit at least two (productive efficiency—with or without dynamic efficiency concerns—and allocative efficiency) targets. Price regulation is the primary instrument; and, while often supplemented by others (including co-payment and generic prescription, for example), these supplementary instruments are weakened through the distortions introduced by the purchase of insurance and the decision-making capabilities of the physician supplier.

Perhaps not surprisingly under these circumstances, price regulation tends to be augmented by patent protection (and extensions), budgetary impositions, industrial policies aimed at optimizing within country R&D operations, and various individual prescribing policies aimed at promoting changing medical needs. Given these complexities, there are a plethora of different pricing regulatory regimes that have evolved across a range of countries over time.

Regulation of Which Price?

A further major general concern within this regulation environment is the actual definition of product price. As the literature on price indices has noted, definition of the pharmaceutical product is itself an important factor here (Morel et al., 2011).

The price index literature tends to focus on issues of therapeutic class, dosage, strength, pack size, among other factors. The significance of such definitional issues to price regulation cannot be overstated. For example, as the number of therapeutic classes increase, the products within class decrease, and the degree of potential price competition necessarily lowers. In this manner the definition of therapeutic class helps define degrees of complementarity and substitutability, the basket of reference prices across which drugs may be assessed and even the formularies across which drugs are allocated. It is even important for patent protection, as generics may challenge incumbent products on the expiry of patent protection through establishing bioequivalence based on therapeutic equivalence or therapeutic substitutability.

More generally, product prices in the pharmaceutical sector are normally tied to at least two listed prices. The wholesale acquisition cost, which is the price at which brand manufacturers sell product, delimited in terms of active therapeutic dose or pack size to wholesalers. While the average wholesale price, which is the price wholesalers sell goods to retailers (hospital pharmacists, prescribing practices, independent retail pharmacists, etc.), again delimited in terms of active therapeutic dose or pack size. Generally, as to be expected, wholesale acquisition cost is higher (by up to 20% it is estimated) than the average wholesale price. It has been generally argued that the wholesale acquisition cost margins have been falling over time, with subsequent impact on the average wholesale price. In a sense, these are posted prices, and the prices actually paid by the user (the insurer/provider) are different and subject to wide-ranging discounts. In the United States, for example, Medicare receives a rebate equivalent of either the “best” price paid to a private purchaser (defined as the price paid by pharmaceutical companies after discounts to retailers) or 23.1% of the average manufacture’s price, whichever is the lower. Most discounts are applied at the individual healthcare provider level, and (including the Medicare rebate) are not open to public scrutiny. Most regulators take the average wholesale price, related to average therapeutic dose as the benchmark price. Given wide-ranging discounts it is clear that this price does not reflect the market price and either price discrimination and/or bargaining practices are widespread both across and within countries.

That said, even across countries attempts may be made to narrow price discrimination. Notably the EU countries promote parallel trade of pharmaceutical products as a means of inducing price convergence. However, Kanavos and Costa-i-Font (2005) have found little evidence of the success of this cross-country policy. Pharmaceutical companies tend to impose supply restrictions in an attempt to dampen parallel trade, but even where such restrictions are in place the capture of rents through arbitrage tends to limit the passing on of benefits to funders and/or consumer/patients. The EU is currently trying to strengthen the common regulatory environment in the European pharmaceutical sector, with respect to some dimensions of regulation; however, such initiatives remain under discussion (EC, 2009).

Models of Pharmaceutical Price Regulation

An optimal regulatory structure would seek to achieve an incentive structure that changed relevant actors’ behavior to achieve efficient outcomes. Given the complex range of healthcare environments, with their many players (pharmaceutical companies, funders, physicians, and consumer/patients) it is no surprise that different countries have adopted different regulatory pricing schemes. No individual regulatory scheme appears optimal, although some may achieve better trade-offs between productive and allocative efficiency and between static and dynamic efficiency. Additional regulatory concerns pursuing issues of budgetary control and access to treatments generally, as well as access to specific medicines (e.g., orphan diseases and/or end-of-life treatments), add to the complexity of the regulatory landscape.

In a number of cases pricing regulation is imposed after price setting by the company. The company is said to be “free” to set prices but subject to rules on reimbursement imposed by the insurer/funder. Under a variety of reimbursement rules the pharmaceutical provider can set a price, recognizing that if the price is lower than or equal to the reimbursement level then the consumer/patient will face no co-payment. But if the price is higher, then the consumer/patient will be charged a co-payment to cover the difference between the set price and the reimbursement level. While such demand-side risk-sharing agreements are common under many regulatory systems, some systems operate free pricing with regulation on the supply side, including “free” price-setting with profits regulation and claw-back agreements; where prices are set for a delimited time and if utilization breaks a given predetermined level, revenue is claw-back by the regulator, and prices are subsequently reduced.

Few countries operate under completely free pricing regimes for pharmaceuticals. Certain parts of the U.S. healthcare system are said to have free pricing; however, generally speaking the majority of third-party funding bodies in the United States, including Medicare, Medicaid, the Veterans Association and a growing number of private insurance companies have either explicit reimbursement rules or negotiate prices with individual companies. Discounting of prices is widespread in the United States. Germany, until 2011, was one of the few EU countries to allow companies to freely set prices. After 2011 cost-containment measures were put in place, which had a flattening effect on pharmaceutical prices. As might be expected, some of the highest pharmaceutical prices in the world can be found in both the United States and (pre-2011) Germany.

Various pricing regulatory regimes have been proposed based on the arguments given above, which are now reviewed below.

Ramsey Pricing

In an unregulated world pharmaceutical companies would set a monopoly price in each national market. The standard economic model for regulating prices in such markets relates such monopoly pricing to conditions of profit maximization. Price (p) is set to marginal cost (MC) to replicate perfectly competitive markets, where free entry and exit pushes the equilibrium quantity for each firm to be determined by p=MC. Or, with decreasing average costs (AC) arising due to high fixed costs of production existing, deterring entry but allowing an incumbent to spread this cost over the levels of demand it is faced with, price setting is associated with a monopoly inverse elasticity rule.

Patent drugs tend to have inelastic demands and face little competition, particularly if they are first in their therapeutic class. While this may in reality be mediated by the rulings introduced by various drug plans, this general position has given rise to Ramsey regulatory pricing rules, which derive from the assumption that a profit-maximizing monopolist seeks the markup of price over cost as it relates to the inverse of their demand elasticity; pMC/p=1/ε (where MC is the marginal cost of production and ε‎ is the elasticity of demand for the product). The pricing rule, derives from profit maximizing occurring at the point where additional revenue meets marginal cost, (MR = MC) and implies that the monopolist sets lower prices as the demand for the product becomes more elastic. Faced with such a monopoly position a regulator seeks to lower prices through imposing a breakeven constraint, such that price (average revenue) is set equal to average cost (AC) (i.e., p = AC).

Working through the inverse elasticity condition Ramsey pricing regulation imposes the condition that pMC/p=(λ/1λ).1/ε, where the inverse pricing rule is extended to incorporate the term (λ/1λ), which lowers the mark-up over long-run cost to the level that allows a firm merely to breakeven. The term (λ/1λ) can be thought of as the level of taxes imposed on a firm to ensure breakeven. Here price regulation will lower prices, through the Ramsey term, to average cost (AC); a pricing rule compatible with the perfectly competitive model (as MC=AC in this latter case).

Third-party purchase introduces a further distortion into pricing. Assuming some form of demand-side risk sharing exists through co-payments (α), it is reasonable to assume that price markups will be higher the lower are the co-payments. Ramsey pricing could be adjusted to incorporate this distortion; the simplest way is to assume a monopolist can raise prices when the co-payment is low, given the product price elasticity. So then pMC/p=(λ/1λ).1/ε+(1α).1/ε.

If the pharmaceutical manufacturer is dealing directly with the treatment provider this distortion will be further weighted by the degree of monopsony purchasing power (1γ) held by the treatment provider; so pMC/p=(λ/1λ).1/ε+(1)γ[ (1α).1/ε ]. The regulated Ramsey price would take account of the monopoly mark-up (λ/1λ), the bargaining power, (1γ) and the insurance distortion (1α), and could vary for pharmaceuticals with different elasticities. In other words, any tax imposed on the monopoly price, as identified through the inverse elasticity relationship, would attempt to adjust for monopoly rents, including the distortion in such rents arising from consumer co-pays and monopsony bargaining rents.

The markup could conceptually, through a lowering of this tax rate, incorporate some return for R&D activities to address issues of dynamic efficiency. However, it is difficult to judge the appropriate level of tax lowering to provide efficient dynamic incentives. Indeed if the defined marginal cost (MC) incorporates R&D costs the definition of the appropriate level of tax lowering is further complicated.

The lead time to market for any given product is an additional complication for Ramsey pricing. The consequences of a long lead time to market combine with the fundamental uncertainties associated with R&D and result in pharmaceutical firms facing considerable risk arising from product development. As recognized since Arrow (1962) while there are institutional devices that can deal with probabilistic outcomes, the type of uncertainty that characterizes basic research is difficult to deal with. Generally speaking risk spreading or portfolio diversity applied to potential product development will be seen. Historically this has typified large pharma’s R&D approach and may also explain recent merger and acquisition activity to counter declining productivity with respect to R&D returns in the sector. A further approach has been to shift risk out to outside companies through the increasing use of Clinical Research Organisations (CROs), with contracts drawn up on an outcome performance basis limiting the risk holding of the pharmaceutical company. Regardless of the specific time-to-market incurred by a given product, the resultant long-run MC is liable to be substantial and spread over a long period. Moreover, if the long-run MC is related to R&D activities, then identification through yardstick comparison (such as the R&D costs of other companies) will not work given all the uncertainties and indivisibilities incurred with any individual company’s R&D process.

For these all reasons, specifying adequately the monopoly markup, identifying the insurance distortion on this markup, the monopsony bargaining power and the relevant level of tax deduction for R&D, Ramsey pricing is difficult to apply to the pharmaceutical sector.

Simple applications of Ramsey pricing include those by Danzon (1997) and Danzon and Towse (2003), who argue that close to Ramsey prices may evolve with profit-maximizing pharmaceutical producers under cross-national pricing, where the pricing markup is inversely related to per capita income. However, even if this is the case, regulation of these prices is difficult without a cross-national regulator. Barros and Martinez-Giralt (2012) show that insurance coverage will distort Ramsey prices upward but can still be subject to a break-even constraint. Although, as they discuss, this will be difficult to implement as the constraint has to operate across countries. Perhaps not surprisingly, given the complexities and the need for cross-country incentives, individual countries have in practice invoked a range of simpler pricing and reimbursement rules that are now discussed.

Cost-Plus Pricing

While a common pricing regulation in other sectors cost-plus pricing, where the reimbursement price is set after accounting for production costs, marketing costs, a reasonable rate-of-return on capital and charges and profit margins incurred in the supply chain, is not widely used in the pharmaceutical sector. It has been used in a number of low- and middle-income countries including Bangladesh, China, and Indonesia. However, as the sector is dominated by a number of multinational companies that set prices globally and tend to operate transfer-pricing arrangements, it is difficult to implement within local markets. Asymmetry of information over the definition of the appropriate cost base fundamentally weakens this form of regulation.

Rate-of-Return Regulation

A related form of regulation based on defining the appropriate rate-of-return for an individual pharmaceutical company is essentially an indirect price control, with the manufacturer’s rate-of-return encompassing return for R&D activity. The United Kingdom has operated this form of regulation on a voluntary basis since 1957 as the Pharmaceutical Pricing Regulation Scheme (PPRS), although it is now complemented by therapeutic value regulations operated through National Institute for Health and Care Excellence (NICE; see section on therapeutic pricing below). Essentially the PPRS initially set a target rate of return within a band of 17–21%, with each individual firm’s return defined within this band as reflecting each company’s degree of innovation. There was some tolerance around this return. In defining the rate-of-return R&D expenditure of up to approximately 20% of sales and marketing expenses of up to 9% of sales was allowed as offsets. Over time, however, the PPRS has also moved toward setting limits on the growth of the overall cost of branded drugs such that by 2014 overall expenditure growth was limited to an agreed per annum percentage growth rate (just under 2% per annum) with retrospective payments to the UK Department of Health if expenditure exceeds certain limits.

While individual products can be “freely” priced within this scheme, it was criticized as companies could modulate prices by setting older products at low prices and newer products at high prices while still maintaining their overall rate of return constraint. Moreover, the lack of transparency in the scheme means that it is not clear how any individual product is actually priced. It is also unclear how the R&D and marketing costs are set. Indeed, there is criticism that rate-of-return regulation leads to over-capitalization generally, as the return is based on a predefined capital base that provides an incentive to over-capitalize. In the case of pharmaceuticals (as the base includes levels of R&D and marketing) this implies that rate-of-return regulation will lead to R&D and/or marketing activities that are higher than optimal levels. Finally, there is increasing criticism that the PPRS does not complement other aspects of pharmaceutical regulation within the UK, particularly the use of cost effectiveness by NICE.

Reference Pricing

A historically common form of pharmaceutical price regulation has been through reference pricing, where a bundle of products are defined to produce a benchmark price for a specific pharmaceutical product. The bundle may be chosen as internal to the country or based on external (to the country) prices. The bundle may be based on products with the same compound, essentially determined through generic reference prices, or based on products with different compounds but with similar modes of action or applied to the same disease indication. The latter are referred to as therapeutic reference prices. Regardless of how defined benchmarked products are, generally all are reimbursed at the same price per daily dose. In a number of cases, reference pricing is used to supplement other regulatory controls making it difficult to identify the impact of reference pricing alone in the regulatory environment.

External reference pricing generally uses a benchmark bundle as based on the price of the same drug across a predefined set of comparator countries. External reference pricing can be thought of as price control rather than regulation per se. It seeks merely to reduce prices to the lowest of some external benchmark without regard to the opportunity costs of pharmaceutical treatments within the healthcare system within the country adopting this form of benchmarking. Indeed, as a control mechanism, it is weak and dependent upon the regulations in force beyond the national borders of the regulator adopting the practice. It therefore cannot be thought of as a true regulatory mechanism. It might be better thought of as a crude form of indirect budgetary control aimed at restricting pharmaceutical budgets. External reference pricing certainly has little (efficiency) incentive structure attached to it. Moreover, as the benchmarking is based on other countries posted prices, such as the average wholesale price across the comparator countries, it may not even have a significant influence on internal discounting practices. This will be especially true if international companies game the system by setting initial prices higher than they otherwise would be if external referencing had not existed (Anis & Wen, 1998).

Internal reference pricing tends to benchmark on-patent products with similar products, including off-patent products. If the internal reference price is calibrated to a generic compound this will obviously affect the patent protection enjoyed by new products. This has direct dampening effects on R&D and dynamic efficiency. The impact is greater the broader the referencing category is and the greater the role of generic competition in the market involved. This may not matter if the country involved, as is generally the case, plays little role in the international pricing decisions of pharmaceutical companies and their R&D objectives due to the globally insignificant size of its market.

There has been a small theoretical literature that analyzes optimal pricing strategies under internal reference price regulation, where generics are already in existence (Merino, 2003; Brekke et al., 2007; Miraldo, 2009; Ghislandi, 2011). The theoretical models are particularly concerned with the competitive impact of reference pricing, emphasizing the endogenous nature of this form of regulation. All these models consider the impact of various pricing strategies, given the endogenous nature of reference pricing, on competition among branded and generic pharmaceutical products. The conclusions reflect the assumptions made in the various papers concerning the competitive environment imposed within the model. Although generally, if reference pricing is based on minimum price benchmarking, consumers’ and producers’ surpluses are reduced on the introduction of the regulation; the former through the indirect effect on dynamic efficiency. The general conclusion is that external reference pricing has a negative effect on both branded and generic prices, with a subsequent dampening of innovation (see Brekke, Grasdal, & Holmas, 2009; Danzon & Epstein, 2008; Filson, 2012). The associated empirical literature tends to be largely descriptive focusing on how the schemes actually operate or on the complexity of definitions of the reference set. The empirical models study whether branded prices are reduced where reference pricing exists but tend not to present fully developed conceptual mechanisms to underpin their results (Lopez-Casasnovas & Puig-Junoy, 2000). Although Herr and Suppliet (2017) argue that when price tiers and copayments are considered alongside the German reference price system generic prices decreased by 4%, but branded drug prices increased by 5%, as brand companies concentrated on the price-insensitive segments of the market—emphasizing the impact that complementary policies may have on the general reference price conclusions.

Therapeutic Benefit Regulation

While external reference pricing has been a popular form of pricing regulation in smaller markets, in the larger European markets referencing is increasingly tied to therapeutic benefit. France, Germany, the Netherlands, and Sweden have all operated some form of reference pricing explicitly tied to therapeutic benefit. In France and Germany product reimbursement reflects therapeutic benefit as reflecting both directly derived health benefit and innovative value, where innovative value is aligned with the therapeutic benefit achieved (Bridges et al., 2009; Mossialos & Oliver, 2005). Although, even in these cases incremental cost effectiveness (discussed below) can play an additional role. For example, the Netherlands and Sweden operate positive reimbursement lists incorporating health benefit as determined (respectively) by therapeutic or chemical equivalence but supplemented through implementation of cost-effectiveness analysis. Here the value, and the reimbursed price of the product is determined through its impact on treatment cost as explicitly linked to the incremental health benefit, as discussed in more detail below (OFT, 2007; Moise & Docteur, 2007). A panel of medical experts normally defines therapeutic benefit: for example, the AMNOG in Germany, and the “Transparency Commission” in France.

It is argued that therapeutic benefit, reference pricing effectively erodes patent protection, with a subsequent dampening effect on R&D activity. There is little empirical evidence to support this conclusion, although, as noted above, there is a substantial body of empirical evidence to support the case that regulation generally has a negative impact on price and launch times. Moreover, therapeutic reference pricing can be consistent with substantive reimbursement initiatives for first-in-class therapies. For example, in France the Transparency Commission relies on the SMR (Service Medical Rendu) to advise on absolute therapeutic benefit as well as the ASMR (Amelioration de Service medical Rendu) to assess the innovative quality of new products. The higher the innovative assessment the higher the potential reimbursement level for any given product.

If physicians do not respond to prices, therapeutic reference pricing has the potential to lead to optimal consumption patterns. If therapeutic benefit is appropriately assessed product prices in that category will converge. Although this conclusion will be affected by the consumption distortions introduced by consumer co-payments. If therapeutic value is used (directly or indirectly) as a basis for reimbursement, then under a fixed budget system the price establishes the opportunity cost of the health benefit foregone in using each new technology within that therapeutic area.

To the extent that the health gain is calibrated against standard therapy in any given therapeutic area, therapeutic benefit, and therefore consumers’ surplus, will differ across treatment areas; the therapeutic value added will be different for any given treatment area, even though the absolute health gain is the same as produced by another intervention in another therapeutic area. If reimbursement is attached to relative effectiveness within therapeutic areas, under a fixed budget system, the opportunity cost of resource use will differ across different therapeutic areas, implying that some treatment areas are “valued” higher than others. Unless there is firm empirical evidence to support this differential valuation of different therapeutic areas, such an approach will lead to inconsistency.

Cost Effectiveness and Reimbursement

An obvious extension of therapeutic reference pricing is to consider the use of cost effectiveness, based on estimating the societal value of a unit of health provision. These approaches are related to the so-called value-based pricing (VBP) approaches to the extent that these VBP approaches tend to be based on cost-effectiveness ratios. As practiced in England, Sweden, and other countries the healthcare sector has commonly valued health benefits in terms of Quality Adjusted Life-Years (QALYs) gained, or even simply as some measure of therapeutic benefit. Two different conceptual bases are given for the valuation of the QALY in the cost-effectiveness literature. Traditionally it was based upon the societal willingness-to-pay for a unit of health (a QALY). More recently, it has been associated with an estimate of the opportunity cost, under a budget-constrained system, to provide a unit of health (a QALY).

The QALY is a widely recognized valuation instrument that attempts to combine dimensions of morbidity and mortality into a single commensurate measure of health state. The QALY has been used extensively for two main reasons: it arguably values health outcomes in a more acceptable metric than money does; and it feeds more easily into the wider medical decision-making process. Whether the QALY is reflective of health state preferences or health states per se, it has given rise to a long rather fruitless literature (see Broome, 1993).

In most healthcare systems QALYs appear to be taken as measures of health states per se with an additional valuation on society’s WTP for a given additional QALY taken as representing the societal value of a QALY. Danzon, Mulcahy, and Towse (2013) show that this approach can conceptually be designed to be consistent with Ramsey price differentiation based on income elasticity differences across countries and addresses concerns of dynamic efficiency as well. Even if agreement is reached over this normative approach, the actual calculation of a QALY relies on the measurement of social preference for different health states. In other words, the use of QALYs in health resource allocations moves us from the normative to the positive, where the decision rule based on QALY maximization under resource constraint necessitates some measure of the societal WTP for additional QALY gains (Claxton, 2007).

Some have, in any case, argued that individuals may show systematic biases in attempting to measure preferences associated with their quality and length of life. Dolan and Khaneman (2008) argue that such preferences are liable to be distorted by an individual’s own experiences, and that as health states change individuals will adapt—so who to ask also becomes important. Others argue that the instruments used to measure such preferences are not well understood and may likewise impart biases. Broome (1993) has eloquently argued that notwithstanding these problems, and that although additionally the QALY may not be conceptually clear and may reflect the “goodness” of or benefit from a state of health rather than a preference, it still represents the best approach to date.1

Historically, the QALY appears to have been defended on a WTP basis and associated with a threshold value that although largely devoid of conceptual foundation, appears to be held at a similar level across a number of countries. In England, the threshold value is taken to be £20,000–£30,000 per QALY as established by NICE, which (at least historically) was similar to the value of $50,000 per QALY suggested by the U.S. cost-effectiveness panel (Gold et al., 1996). The English value is also used as an international benchmark by the Swedish HTA body, although this body also considers cost-effectiveness rates used by other Swedish public-sector bodies and is not universally bound by a formal cost-effectiveness threshold.

Devlin, Parkin, and Appleby (Written Evidence for the House of Commons Select Committee Inquiry on NICE) provide evidence that NICE effectively operates a threshold somewhat above £30,000 per QALY, and that these were justified as based on a “special considerations” argument. Special considerations presumably relate to the mandate given to NICE to consider issues of innovation, patient preferences, and political considerations as well as cost effectiveness and therapeutic benefit when assessing healthcare technologies.

Assuming the NICE threshold has some justification related to societal WTP, Laska et al. (1999) have shown that the QALY approach is equivalent to the general WTP for a statistical life approach under straightforward assumptions. Indeed, both Chilton et al. (2002) in a UK study, and Markanndya et al. (2004) in a wider European study, convert their estimates of the value of a statistical life, which lie close to the officially adopted UK figure, into the value of a statistical life year (VOSL). The conversion made by Chilton et al. (2002) is simply:


where VOLY is the value of a life year, calculated from VOSLA, the value of a statistical life at age A, divided by the expected age at death (T) determined from life tables and the current age, A.

On this basis, most reasonable VOSL for the average road traffic accident from Chilton et al. (2002), valued at the average age of what? for a traffic accident, is £27,630 per life year, while an equivalent estimate gained from Markanndya et al. (2004) for a 5 in 1000 risk change is £41,975 per life year (the median value is £22,080). Differences in estimated valuation may reflect the estimation procedures used and possibly the different mean age of the respondents in these questionnaires and the fact that the former values increases in life expectancy, while the latter values changes in the probability of death. Although such estimates derive from a different U.K. governmental office, assuming consistency in valuation is to be defended across public funded bodies, this provides the strongest (if not the only) conceptual basis for the NICE existing threshold values of £20,000–£30,000 per QALY against which new treatments, including pharmaceuticals are judged.

Moreover NICE, like some other countries, appears to operate differential values for different treatments, especially for oncology treatments. Partly to address the issue of high oncology drug prices, an “end-of-life” NICE threshold operates with implied values of QALYs set at approximately 1.5 times the norm and cost-effectiveness ratios in the region of £50,000 per QALY.

The justification for these differential values can only be that society attaches a different value to life at end of life. This is the argument used in other U.K. government departments; for example, the Health and Safety Executive apply a value of twice the standard Value of Prevented Fatality (VPF) for cancer when assessing their cost-benefit decisions. Round (2012), however, in assessing the empirical evidence on social valuations in health found little support for different valuation of end of life treatments, although it might be that severity of illness rather than shortness of life expectancy warrants a premium given the empirical evidence.

Others, including Coast et al. (2008), have suggested that the QALY is in any case not useful in valuing end-of-life care as when there is little or no life expectancy gain as the QALY collapses in one dimension (the Life Year becomes redundant). Additionally, that quality of life is much more important and the QALY is not sensitive enough to capture all quality-of-life dimensions at end of life. Moreover, that patient preferences are unstable at end of life and, in any case, that QALYs use a scale that anchors on a value of death (normally as = 0), and this is invalid if death is imminent. Finally they argue that time is valued differently at end of life compared to other stages of life. All such problems amalgamate to make the QALY a redundant measure for end-of-life care.

The problems arising from a WTP basis for estimating the cost-per-QALY threshold have led Claxton et al. (2015) to argue recently that English National Health Service (NHS) opportunity cost calculations, based on preexisting treatments provided under the NHS budget, should be used as the basis of the threshold value. They calculate that the NHS treatments provided in 2008–2009 returned an average estimate of £13,000 per QALY (estimated as £12,936 per QALY) across a range of 23 treatment programs. Although this is an agreed value, this average opportunity cost value was determined through applying the opportunity cost approach to a range of programs that had widely varying (opportunity) cost per QALY estimates. The average cost per QALY of £13,000 is a weighted average across estimates over a number of NHS programs, with the weights being a complex interaction of calculated elasticities estimated for the different treatment programs.

While the methodological debate over the estimation is ongoing, the opportunity cost approach unlike the WTP approach, does explicitly recognize health sector budget constraints. If the health sector budget is predetermined by central government expenditure rounds, once fixed the uptake of new treatments within this budget displaces other existing treatments. This displacement must take account of the opportunity cost, defined by the cost-effectiveness levels, of these existing treatments. New pharmaceutical treatments must be valued within this environment according to this approach.

Some attempts have been made to widen the concept of value beyond health effects. There has been limited impact of these arguments on pharmaceutical price and reimbursement decisions. This may reflect the ring-fenced nature of health sector budgets. For whatever reason, the discussion has rarely gone beyond the incorporation of disease severity elements, unmeet (disease) needs, and indirect (productivity) costs into a cost-effectiveness framework.

Managed Entry Agreements

Managed Entry Agreements (MEAs) are individually negotiated agreements that are normally instituted where there is uncertainty over the diffusion and uptake of a pharmaceutical product in the market. This may arise because of uncertainty over the product’s clinical value, as might occur with first-in-line drugs or where patient heterogeneity is an issue, or over cost effectiveness as it is uncertain what the true treatment cost and/or budget impact will be. Under MEAs regulatory agencies and pharmaceutical companies agree on a payment for specific performance (defined in terms of treatment benefit achieved), population coverage combined with pricing or budget agreements or discount and rebate schemes associated with a trigger level of utilization. Although generally limited to high-cost and orphan drug treatments MEAs offer a degree of flexibility associated with price setting not seen with other “across the board” regulatory frameworks. That said, the lack of transparency that arises from commercial sensitivities, the complexities associated with various schemes and the lack of evidence over their operation has meant that their use within a regulatory framework has been limited so far.

MEAs may also, however, be seen as part of a wider response to regulatory delay. Both the FDA and the EMA have recently initiated a number of fast track schemes aimed at getting promising new therapies on to the market quicker. Such schemes include Accelerated Approval, Priority Review, Fast Track Designation, and Breakthrough Designation, and since their introduction there has been increasing attempted use of these quicker routes to market. MEAs may be seen in this wider context as an attempt to quicken market access by spreading financial risks across the regulator and the companies. MEAs may also be seen as a means of providing specific evidence on treatment value once a drug is ready to access the market. However, for MEAs to become a major regulatory tool, issues of definitional consistency and clarity over modes of action need to be resolved.


Economic regulation is a fact of life in the pharmaceutical market. Reimbursement regulation is used to reward value, while simultaneously offsetting the monopoly created through patent protection. The tradeoffs involved in regulated price are further complicated through the role of insurance and its distortion of price elasticity. Regulation has been increasingly applied to both the licensing and the pricing of drugs, which affects sales volume. This is true particularly in the predominately public-funded European healthcare systems characterized either by tax-based or social insurance financing where the vast majority of healthcare expenditure is budget-delimited.

Price and reimbursement regulation, even with coupled with patent protection, does not currently efficiently incentivize R&D and may distort R&D activity toward large markets as firms attempt to recover costs. Even so, returns to R&D appear to be falling as costs and time-to-market (with the attendant product revenue delays) increase. The growth in R&D spending on expensive biologics anticipated to be used by relatively small populations is doing nothing to ameliorate this falling return. The complex tradeoff between the pursuit of dynamic and static efficiency in the pharmaceutical sector makes price and reimbursement regulation extremely challenging.

Against this background it is not surprising that many, smaller countries have historically used external reference pricing regulation as the major support for reimbursement. New forms of pricing regulation, based on an explicit definition of treatment value (essentially through calculating incremental health benefit), are currently evolving perhaps in an attempt to reward and increase R&D returns, in areas where the greatest population health gains can be realized. There are also moves to increase price transparency across markets. While such policies are currently more prevalent in Europe, it is likely that there will be a global proliferation of these regulatory tools.


Airoldi, M. (2007). Gains in QALYs vs. DALYs averted. The troubling implications of using residual life expectancy. Health Economists’ Study Group, Summer 2007, Brunel UniversityFind this resource:

Airoldi, M., & Morton, A. (2009). Adjusting life for quality or disability: stylistic difference or substantial dispute? Health Economics, 18, 1237–1247.Find this resource:

Anis, A. H., & Wen, Q. (1998). Price Regulation of Pharmaceuticals in Canada. Journal of Health Economics, 17, 21–38.Find this resource:

Arora, A., Ceccagnoli, M., & Cohen, W. M. (2003). R&D and the patent premium. NBER Working Paper Series, WP9431, NBER, USA.Find this resource:

Arrow, K. J. (1962). Economic welfare and the allocation of resources for invention. In R. R. Nelson (Ed.), The rate and direction of inventive activity: Economic and social factors (pp. 609–625). Princeton, NJ: Princeton University Press.Find this resource:

Barros, P., & Martinez-Giralt, X. (2012). Health economics: An industrial organisation perspective. London: Routledge.Find this resource:

Berndt, E., Nass, D., Klienrock, M., & Aitken, M. (2015). Decline in economic returns from new drugs raises questions about sustaining innovations. Health Affairs, 34, 245–252.Find this resource:

Berthelot, J., Roberge, R., & Wolfson, M. C. (1993). The calculation of healthadjusted life expectancy for a Canadian province using a multi-attribute utility function: A first attempt. In J. Robine, C. D. Mathers, M. R. Bone, & I. Romieu (Eds.), Calculation of health expectancies: Harmonization, consensus achieved and future perspectives (pp. 161–172). Montpellier, France: Colloque INSERM/John Libbey Eurotext Ltd.Find this resource:

Brekke, K. R., Koningbauer, I., & Straune, O. R. (2007). Reference pricing of pharmaceuticals. Journal of Health Economics, 26(3), 613–642.Find this resource:

Brekke, K. R., Grasdal, A., & Helge, T. (2009). Regulation and pricing of pharmaceutical expenditures: theory and evidence from a natural experiment. Journal of Public Economics, 95, 624–638.Find this resource:

Bridges, J., Cohen, J., Grist, P., & Muhlbacker, A. (2009). International experience with comparative effectiveness research: Case studies for England/Wales and Germany. Advances in Health Economics and Health Services Research, 22, 29–50.Find this resource:

Broome, J. (1993). Qalys. Journal of Public Economics, 50, 149–167.Find this resource:

Chilton, S., Covey, J., Hopkins, L., Jones-Lee, M., Loomes, G., Pidgeon, N., & Spencer, A. (2002). Public perceptions of risk and preference-based values of safety. Journal of Risk and Uncertainty, 25(3), 211–232.Find this resource:

Civa, A., & Maloney, M. (2007). The effect of price on pharmaceutical R&D. The B>E Journal of Economic Analysis and Policy, 9, 15–21.Find this resource:

Claxton, K. (2007). OFT, VBP: QED? Health Economics, 16(6), 545–558.Find this resource:

Claxton, K., Martin, S., Soares, M., Rice, N., Spackman, E., Hinde, S., . . . Sculpher, M. (2015). Methods for the estimation of the NICE cost effectiveness threshold. Health Technology Assessment, 14, 1–503.Find this resource:

Coast, J., Flynn, T., Sutton, E., Al-Janabi, H., Vosper, J., Lavender, S., . . . Peters, T. (2008). Investigating choice experiments for preferences of older people (ICEPOP): evaluative spaces in health economics. Journal of Health Services Research & Policy, 13(Suppl. 3), 31–37.Find this resource:

Cockburn, I., Lanjouw, J., & Schankerman, M. (2016). Patents and the global diffusion of New Drugs. American Economics Review, 106, 136–164.Find this resource:

Costa-Font, J., McGuire, A., & Varol, N. (2015). Regulation effects on the adoption of new medicines. Empirical Economics, 49, 1101–1121.Find this resource:

Danzon, P. (1997). Price Discrimination for Pharmaceuticals: Welfare Effects in the U.S. and the E.U. International Journal of the Economics of Business, 4, 301–321.Find this resource:

Danzon, P. M., & Chao, L. W. (2000). Cross-national price differences for pharmaceuticals: How large, and why? Journal of Health Economics, 19(2), 159–196.Find this resource:

Danzon, P. M., & Epstein, A. (2008). Launch and pricing strategies of pharmaceuticals in interdependent markets. NBER Working Paper No. WP14041, NBER, USA.Find this resource:

Danzon, P., & Towse, A. (2003). Differential Pricing for Pharmaceuticals: Reconciling Access, R&D, and Patents. International Journal of Health Care Finance and Economics, 3, 183–205.Find this resource:

Danzon, P., & Furukawa, M. (2011). Prices and availability of pharmaceuticals: Evidence from nine countries. Health Affairs, 22, 521–536.Find this resource:

Danzon, P., Mulcahy, A., & Towse, A. (2013). Pharmaceutical Pricing in Emerging Markets: Effects of Income, Competition, and Procurement. Health Economics, 24, 238–252.Find this resource:

Devlin, N., Parkin, D., & Appleby, J. (2002). Written evidence for the House of Commons Select Committee Inquiry on NICE House of Commons Health Committee: National Institute for Clinical Excellence. Second Report of Session 2001–02. The Stationary Office.Find this resource:

DiMasi, J., Grabowski, H., & Hansen, R. (2016). Innovation in the pharmaceutical industry: New estimates of R&D costs. Journal of Health Economics, 47, 20–33.Find this resource:

Dolan, P., & Khaneman, D. (2008). Interpretations of utility and their implications for the valuation of health. Economic Journal, 118, 215–234.Find this resource:

Duggan, M. G., & Scott Morton, F. (2011). The medium-term impact of Medicare Part Don pharmaceutical prices. American Economic Review, 101(3), 387–392.Find this resource:

Erickson, P., Wilson, R., & Shannon, I. (1995). Years of healthy life. Statistical Notes, 7, 1–14.Find this resource:

European Commission. (2009). Pharmaceutical sector inquiry, Final Report, July 2009. Retrieved from

Filson, D. (2012). A Markov-perfect equilibrium model of the impacts of price controls on the performance of the pharmaceutical industry. RAND Journal of Economics, 43, 110–138.Find this resource:

Garber, A. M., Jones, C., & Romer, P. (2006, March). Insurance and incentives for medical innovation. Cambridge, MA: National Bureau of Economic Research, Working Paper No. W12080.Find this resource:

Ghislandi, S. (2011). Competition and the reference pricing scheme for pharmaceuticals. Journal of Health Economics, 30, 1137–1149.Find this resource:

Gold, M. R., Siegel, J. E., Russell, L. B., & Weinstein, M. C. (1996). Cost-effectiveness in health and medicine. New York: Oxford University Press.Find this resource:

Grootendorst, P. (2012). Prescription drug insurance and reimbursement. In A. Jones (Eds.), The Elgar Companion to Health Economics (2nd ed., ch. 11, pp. 114–122). Cheltenham, UK: Edward Elgar Publishing.Find this resource:

Herr, A., & Suppliet, M. (2017). Tiered copayments, pricing and demand in reference price markets for pharmaceuticals. Journal of Health Economics, 56, 19–29.Find this resource:

Horowitz, A. W., & Lai, E. L. (1996). Patent length and the rate of innovation. International Economic Review, 37(4), 785–801.Find this resource:

Hughes, J., Moore, M., & Snyder, E. (2002). ‘Napsterising’ pharmaceutical access, innovation, and consumer welfare, NBER Working Paper 9229. NBER, USA.Find this resource:

Institute for Healthcare Informatics. (2014, April). Medicine use and shifting costs of healthcare. A review of the use of medicines in the United States in 2013. Institute of Health Care Informatics, NJ, USA.Find this resource:

Jena, A. B., & Philipson, T. J. (2008). Cost-effectiveness analysis and innovation. Journal of Health Economics, 27(5), 1224–1236.Find this resource:

Kanavos, P., & Costa-i-Font, J. (2005). Pharmaceutical parallel trade in Europe: stakeholders and competition effects. Economic Policy, 20, 751–798.Find this resource:

Keehan, S. P., Cuckler, G., Sisko, A., Madison, A., Smith, S., Stone, D., . . . Lizonitz, J. (2015). National health expenditure projections, 2014–24: Spending growth faster than recent trends. Health Affairs, 34(8), 1407–1417.Find this resource:

Kyle, M. (2007). Pharmaceutical price controls and entry strategies. Review of Economics and Statistics, 89, 88–99.Find this resource:

Kyle, M. K. (2006). The role of firm characteristics in pharmaceutical product launches. RAND Journal of Economics, 37, 602–618.Find this resource:

Lakdawalla, D., & Sood, N. (2013). Health insurance as a two-part pricing contract. Journal of Public Economics, 102, 1–12.Find this resource:

Lanjouw, J. O. (2005). Patents, price controls and access to new drugs: How policy affects global market entry. WHO Paper. Available online.Find this resource:

Laska, E. M., Meisner, M., Siegel, C., & Sinnett, A. (1999). Ratio-based and net benefit-based approaches to health care resource allocation: proofs of optimality and equivalence. Health Economics, 8, 171–178.Find this resource:

Lopez-Casanovas, G., & Puig-Junoy, J. (2000). Review of the literature on reference pricing. Health Policy, 54, 87–123.Find this resource:

Markanndya, A. (2004). Metodologia para valorar los costes de los impactos del cambio climatico. Ekonomiaz, 57(3), 164–215.Find this resource:

Mehrez, A., & Gafni, A. (1989). Quality-adjusted life years, utility theory, and health-years equivalents. Medical Decision Making, 9, 142–149.Find this resource:

Merino, A. (2003). Impact of reference price system on the pharmaceutical market: A theoretical approach. Working Paper Pompeu Fabra. Available online.Find this resource:

Miraldo, M. (2009). Reference pricing and firms’ pricing strategies. Journal of Health Economics, 28, 176–197.Find this resource:

Moise, P., & Docteur, E. (2007). Pharmaceutical pricing and reimbursement policies in Sweden. Paris: OECD.Find this resource:

Morel, C., McGuire, A., & Mossialos, E. (2011). The level of income appears to have no consistent bearing on pharmaceutical prices across countries. Health Affairs, 30(8), 1545–1552.Find this resource:

Mossialos, E., & Oliver, A. (2005). An overview of pharmaceutical policy in four countries: France, Germany, the Netherlands and the UK. International Journal of Health Planning and Management, 20(4), 291–306.Find this resource:

Nordhaus, W. (1969). Invention, growth and welfare. Cambridge, MA: MIT Press.Find this resource:

Nyman, J. (2003). The Theory and Demand for Health Insurance. Stanford, CA: Stanford University Press.Find this resource:

OECD. (2016). Health indicators at a glance. Paris: OECD.Find this resource:

OFT. (2007). The pharmaceutical price regulation scheme. Office of Fair Trading, London, UK.Find this resource:

Pauly, M. V. (2011). Insurance and the Demand for Medical Care. In S. Glied & P. C. Smith (Eds.), Oxford Handbook of Health Economics (ch. 16). London: Oxford University Press.Find this resource:

Round, J. (2012). Is a QALY still a QALY at the end of life? Journal of Health Economics, 31, 521–527.Find this resource:

Sassi, F. (2006). Calculating QALYs; comparing QALY and DALY calculations. Health Policy and Planning, 21, 402–408.Find this resource:

Scherer, F. (1972). Nordhaus’ theory of optimal patent life: A geometric reinterpretation. American Economic Review, 62, 422–427.Find this resource:

Szymanski, S., & Valletti, T. (2005). Parallel trade, price discrimination, investment, and price caps. Economic Policy, 44, 705–749.Find this resource:

Tirole, J. (1988). The theory of industrial organization. New York: MIT Press.Find this resource:

Vernon, J. (2005). Examining the link between price regulation and pharmaceutical R&D investment. Health Economics, 14(1), 1–16.Find this resource:


(1.) There are other contenders in terms of health state valuation. Other instruments such as Years of Healthy Life (HYL) and health-adjusted healthy life are QALYs in all but acronym (Berthelot, Roberge, & Wolfson 1993; Erickson, Wilson, & Shannon, 1995). Mehrez and Gafni (1989) proposed values based on health profiles, where various health states are considered in different sequences of event (profiles), and individuals trade off the number of years in perfect health against the years in profile that they deem equivalent. This seems an extension of the QALY concept to incorporate time in a health state into the value. Disability Adjusted Life Years (DALYs), which estimate life expectancy lost and weight this by the number of years lived in disability, are possibly the most commonly proposed alternative. Airoldi and Morton (2009) argue that once age weighting and differences in discounting into the DALY calculation have been made and adjustments made to allow a comparison between loss in quality of life and the disability weighting in the DALY, the two valuation concepts do not differ much. Both Airoldi (2007) and Sassi (2006) found, however, that the actual estimates of health change based on the two approaches do differ systematically.