The Effect of Pharmaceutical Patent Term Length on R&D and Drug Expenditures in Canada

Abstract

The authors make a useful and original contribution to evaluating the impact of strengthened patent protection of pharmaceuticals on Canadian healthcare expenditures. This commentary argues that their second task – measuring the impact of strengthened patent protection on R&D spending in Canada – faces an insurmountable conceptual problem: Corporate decisions on the location of R&D activities bear no theoretical relationship to the location of sources of R&D funding and are guided by other factors, including political considerations.

The objective of Grootendorst and Di Matteo’s study is twofold: first, to evaluate the contribution of the strengthened Canadian patent protection of pharmaceutical innovation to R&D spending in Canada, and second, to determine its impact on drug prices and thus on healthcare spending. This commentary argues that while the latter is a useful and worthwhile exercise, the former is an impossible task.

The tenuous link between domestic patent protection and domestic R&D spending

Stronger patent protection increases the profits available to the innovator for increased R&D spending anywhere in the world. The total contribution to global R&D effort made in the form of increased profits (patent rents) collected from Canadian users of prescription drugs can, in principle, be greater or smaller than the extra R&D spending in Canada. The authors explicitly recognize that the multinationals have “discretion over the location of R&D” and that, therefore, the relevant objective would be to assess the impact on global R&D. Yet they assert that they “estimated the effect … on … expenditure on Canadian pharmaceutical R&D.”

Global pharmaceutical innovation is financed primarily from the extra profits (rents) generated by patent holders, and to a lesser extent by public funds. The contribution each country makes to the financing of global innovation is therefore largely a function of the strength of its patent system. Until the TRIPS agreement came into force (on January 1, 1995), the strength of patent protection was determined by national authorities and varied across countries.¹ The TRIPS agreement made a minimum standard of contribution mandatory; those countries that previously contributed less joined TRIPS because it was a condition of membership in the Word Trade Organization.²

The location of R&D activity (or R&D spending), as distinct from the location of the sources of financing, is a function of private decision-making by the management of the (multinational) drug companies. For example, a typical Swiss drug corporation locates much of its R&D activity in the home country, but collects well over half of its rents in the US market, while most of the rest comes from countries other than Switzerland.³ Indeed, the Swiss patent legislation was historically relatively weak; for example, until 1978 only pharmaceutical process, but not product, could be patented (Schiff 1971; Zutshi 1998: 41). The determinants of location of pharmaceutical R&D activity listed in the literature typically include historical factors, availability and cost of skilled personnel, proximity to centres of academic research and to clinical testing facilities, and government incentives, such as the tax treatment of R&D expenditures.

Controls on prescription drug spending

With patent protection more or less uniform in the post-TRIPS world,⁴ attention has focused on policies and practices seeking to control prices or influence the selection of prescribed drugs and thus reduce the burden of prescription drug spending. Specifics include encouragement of generic prescribing; promulgation of formularies that restrict reimbursement for certain (high-priced) drugs, making the percentage of a drug’s reimbursement conditional on independent evaluation of its therapeutic efficiency; and influencing physician prescribing behaviour (Grootendorst and Di Matteo briefly list the Canadian programs in their paper; those of other countries are reviewed in Jacobzone 2000).

The authors controlled for the impact of these policies by reference to drug price inflation and per capita drug expenditure in the United States, on the grounds that the United States “has not introduced drug price regulation.” It should be noted, however, that a similar role is performed by private sector players, such as the health insurance industry or pharmaceutical benefits management companies. US drug prices and per capita drug expenditures have therefore not increased as much as would be the case with “unrestricted” patent protection.

R&D spending trends

The R&D-to-sales ratio in the Canadian pharmaceutical industry, both during the latter part of the period covered by the study and during the post-study period, has been declining, although the stronger patent protection remains in place. It reached its peak of 11.7% in 1995 and declined to a low of 8.3% in 2004 before rising (marginally) to 8.7% in 2005. The absolute value of R&D expenditures (in current dollars) increased in all years during the period since 1988, except for 2003 and 2004, when it dropped by 0.4% and 2%, respectively (PMPRB 2006: 39).

In contrast, the R&D-to-sales ratios of the US-based companies, members of the Pharmaceutical Research and Manufacturers of America (PhRMA), have been much more stable: the ratio of domestic (US) R&D to global sales was 13.1% in 1995, rose to 14.0% in 1997 and dropped to a minimum of 12.2% in 2004 before rising to an estimated 12.6% in 2005. The ratio of R&D spending abroad to global sales over the same period fluctuated within the range of 2.8% to 3.6%, with no apparent trend (calculated from data in PhRMA 2006: 52, 57).

Conclusion

The authors have made a sophisticated attempt at evaluating the extra contribution Canadian consumers are making to global pharmaceutical innovation as a result of the 1987 and 2003 modifications of the Patent Act. Given the data problems, their results have to be qualified; however, theirs is an imaginative and useful effort. On the other hand, while the strength of a national patent system determines the contribution of a country’s consumers to global innovation, it has no necessary relationship to the size of the R&D activity on its territory. The “bulge” in the pharmaceutical R&D-to-sales ratio in Canada, which occurred after 1987 and peaked in 1995, has to be explained by other factors. The allusion the authors make to the “discretion” the multinationals have, to the “negotiations that resulted in patent terms being extended” and to the political commitment made by the Pharmaceutical Manufacturers Association of Canada should have been given more prominence than they received in the paper. (Incidentally, the PMAC commitment to a 10% R&D-to-sales ratio has not been kept since 2001.)