The drug bargaining game: pharmaceutical regulation in Australia

https://doi.org/10.1016/j.jhealeco.2003.11.003Get rights and content

Abstract

Many countries, including Australia, regulate the price consumers pay for pharmaceuticals. In this paper, the Australian Pharmaceutical Benefits Scheme (PBS) is modelled as a multi-stage game played between the regulator and pharmaceutical firms. Conditions are derived under which vertically differentiated firms are regulated and a number of issues are discussed. These include efficiency, regulated firm profitability, leakage, and price discrimination. An extension examines the introduction of new drugs and concludes that if all the benefits of a new drug are to be realised, then existing agreements and transfers (per-unit subsidies) need to be renegotiated.

Introduction

To ensure consumers equity of access, many countries regulate the price consumers pay for pharmaceuticals. The regulated price is normally well below the market price. Therefore, to induce participation by pharmaceutical firms in the regulatory regime, transfers are given by the government to the firms. These transfers are often implemented through a negotiated agreed price for producers. Willison et al. (2001) document that Australia, the Netherlands, New Zealand, and the United Kingdom set a fixed consumer price with the difference between this price and the agreed price being the implied per-unit subsidy. In France and Sweden consumers pay a fixed proportion of the agreed price.

The literature on pharmaceutical regulation is mainly empirical with emphasis placed on measuring international price differences and seeing if they can be explained by the regulatory environment. Danzon and Chao (2000a) find that countries with strict price regulation (France, Italy, and Japan) have lower prices than the less regulated markets of the United States and the United Kingdom. However, Berndt (2000), provides a number of caveats about their interpretation of the data. In a related paper, Danzon and Chao (2000b), examine whether the extent of price competition between producers of generic drugs is affected by the regulatory environment in which they operate. They find that price competition is significant in less regulated markets (United States, Canada), but not in more regulated markets (France, Italy, and Japan).

Despite a substantial empirical literature, the theoretical literature on pharmaceutical regulation is rather scant. Johnston and Zeckhauser (1991) model Australian regulation as marginal cost pricing with a per-unit subsidy paid to monopolist pharmaceutical firms to guarantee these firms monopoly profit. Anis and Wen (1998) develop a theoretical model of pharmaceutical regulation in Canada where pharmaceutical firms are either multi-product monopolists or sell in two international markets. Regulation is modelled as a constraint on the ability of firms to set prices independently for each product or in each market. In both of these papers, strategic interactions between firms are ignored as is any bargaining between firms and the regulator over the per-unit subsidy or the form of the price constraint.

This paper endeavors to correct this situation by building a theoretical model of the Australian Pharmaceutical Benefits Scheme (PBS) in which strategic interactions between firms and bargaining between firms and the regulator play a central role. The goal is to discover the implications of the PBS’s design and suggest possible improvements. Although it is based on the Australian system, the model has wider appeal because similar schemes are in place in many European countries as well as Canada and New Zealand. In fact, one of the paper’s main contributions is to provide an analytical framework which can easily be amended to examine pharmaceutical regulatory regimes other than that of Australia.

The PBS is modelled as a five stage game. In the first stage, pharmaceutical firms choose whether to enter the regulation process. In the second, the quality of the drug is determined and in the third, given the regulated price, the regulator chooses which firm/s to regulate. In the fourth stage, the regulator and the regulated firm bargain over a transfer which can be implemented via an agreed producer price and finally, in the fifth stage, pharmaceutical firms, with different quality drugs, compete with each other in the drug market.

The main results are summarised in Proposition 1, Proposition 2. Together they state that as long as the regulated price is less than the unregulated price of the high quality firm, then the high quality firm always enters the regulation process and is regulated. The negotiated agreed price is less than the unregulated price of the high quality firm. In some circumstances the low quality firm also enters the regulation process and is regulated. Since the regulated price is the same for high and low quality firms, a regulated low quality firm makes no sales. Essentially, the low quality firm is regulated to stop the low quality firm stealing consumers away from the high quality firm.

Once the model is outlined, a number of implications are drawn. The first concerns efficiency. It is shown that there is a range of regulated prices which achieve efficiency and that this range varies between drugs. Therefore, having an identical regulated price for all drugs, as the PBS scheme does, can lead to inefficiency if this regulated price falls outside the required range. By explicitly modelling the regulator as a surplus maximiser, efficiency considerations, which are often ignored in the literature, are brought to the fore. A lower regulated price for some drug classes might not only improve equity of access, but also lead to efficiency if the lower regulated price moves into the efficient range. The policy of having an identical regulated price for all drug classes needs to be re-examined.

The second concerns firm profitability. Although the agreed price is below the unregulated price of the high quality firm, this does not mean the regulated high quality firm is worse off under regulation than without regulation. In fact, the bargaining process ensures it can not be made worse off. The reason why the lower agreed price is consistent with higher firm profit is that firm sales depend not on the agreed price, but on the regulated price. The agreed price does not affect any consumption or production decisions, it is just a device used to make transfers to the firm in return for the firm being regulated. Although this point is simple, it is not appreciated by the literature nor the regulator. In fact, the regulator seems to completely miss the point by explicitly considering foreign prices when negotiating the agreed price when all it should consider is the additional surplus regulation generates and the quantity sold at the regulated price.

The third implication concerns what is known as leakage. The model suggests that in the bargaining process all high quality uses of the drug should be specified and its subsidised use restricted to these uses. Failure to do so results in the subsidised use of the drug leaking out into low quality uses and non-negotiated high quality uses. Although this increases consumer surplus, it can reduce the regulator’s payoff if the induced unnegotiated transfers are large enough. This leakage problem is discussed in the informal literature, Johannesson (1992), where it is realised that administering complex use restrictions is problematic. This paper makes it clear that the source of the leakage problem is the method the regulator chooses to make transfers to the pharmaceutical firms, namely, per-unit subsidies. If lump-sum transfers or price-volume contracts were used leakage would not result and problematic use restrictions would not be needed.

A feature of the Australian PBS is that there are two regulated prices. Concessional patients face a lower regulated price than general patients. Amending the analysis to incorporate a high and a low regulated price leads to the fourth implication, namely, that having two regulated prices can increase the regulator’s payoff if in the presence of one regulated price (i) some consumers purchased the low quality drug or (ii) both high and low quality firms are regulated. If a single regulated price was chosen efficiently by the government, neither of these two cases would arise. Therefore, it is the arbitrariness of the setting of the regulated price that introduces situations in which having two regulated prices leads to greater regulator payoffs.

Finally, the model is amended to take exogenous innovation into account. This leads to the fifth implication which concerns the renegotiation of agreements in the presence of new drugs. First, a new lowest quality drug is introduced. It is shown that this can increase the payoff of the regulator even if the firm producing the new drug makes no sales. This follows because the presence of the new drug alters the disagreement payoffs in the absence of regulation in such a way that a smaller transfer is paid to the high quality regulated firm. It is also shown that no regulation might maximise the regulators payoff. In either of these cases, for all the benefits of the new drug to be realised, it is necessary for existing regulatory agreements to be renegotiated. This may entail drugs that were initially regulated being removed from regulation. Next, a new highest quality drug is introduced. The message is similar, to realise all the benefits from a new drug requires existing regulatory agreements to be renegotiated. Once again, this point seems to be missed by the literature and the regulator.

Section snippets

Australian pharmaceutical regulation—institutional detail and procedures

Pharmaceutical patents provide their holders with monopoly power which allows them to charge monopoly prices. These prices can be such that an individual whose health outcome would be improved by taking the drug cannot afford to do so. To ensure equity of access to drugs, the Australian government has implemented a system of regulated prices and subsidies known as the Pharmaceutical Benefits Scheme (PBS).

Game structure

Pharmaceutical regulation in Australia can be modelled as a stage game. In the first stage, a foreign owned pharmaceutical firm, at some cost, chooses whether or not to go through the drug evaluation, bargaining, and regulation process for a particular drug.

Efficient price regulation

Pharmaceutical firms are foreign owned, so a government solely concerned with efficiency would choose the regulated price to maximise S1+S2L.

Proposition 3

A regulated price is efficient if and only if p̄≤α=c+(s2−s1)θ.

Proof

In the Appendix A

The intuition is clear. Any regulated price in this interval has an equilibrium in which only the high quality firm is regulated and all consumers purchase the high quality drug. Therefore, any price in this interval maximises the total surplus, net of disagreement payoffs,

Two regulated prices

The analysis to date has been based on there being one regulated price. However, the Australian government sets two regulated prices, one significantly lower than the other. Consumers with a certain characteristic, e.g. a welfare recipient, are regarded as concessional patients and can purchase at the low price, other consumers are regarded as general patients and purchase at the high price. Resale is stopped by having consumers who are eligible for the low price present documentation to this

Exogenous innovation and regulation

To date it has been implicitly assumed that the stage game outlined in Section 3 is repeated every period. If nothing in the environment changes, then the equilibrium of the game does not change either. In this section, it is assumed that a new drug exogenously becomes available. This changes the environment of the stage game and so a new equilibrium arises with disagreement payoffs given by the new unregulated Nash equilibrium payoffs. In this new equilibrium, both the drug that is regulated

Conclusion

Although the model of pharmaceutical regulation developed in this paper is relatively simple it captures the essence of the Australian system and allows current policy debates to be analysed in a coherent framework that till now does not appear in the literature. The model has many implications some of which extend beyond the Australian setting. The first is that although the regulated price is chosen to ensure equity of access, it also has efficiency implications that should be considered when

Acknowledgements

This work was undertaken while visiting CHERE at the University of Technology, Sydney. I would like to thank Rosalie Viney, Tony Harris, and participants at a seminar presentation at CHERE for helpful comments.

References (17)

There are more references available in the full text version of this article.

Cited by (15)

View all citing articles on Scopus
View full text