The impact of patent protection on US pharmaceutical exports to developing countries

Abstract

This article provides evidence that patent protection can have a positive effect on trade, by analysing the impact of the implementation of intellectual property rights (IPR) in developing countries on the US exports of pharmaceutical products, following intense lobbying efforts from the US pharmaceutical industry to have the Trade-Related Aspects of Intellectual Property Rights (TRIPS) agreement included in the creation of the World Trade Organization (WTO). A gravity model using panel data from 1995 to 2010 suggests that the implementation of minimum standards of patent protection has increased US exports of pharmaceuticals to 108 nonadvanced countries.

Keywords:

• pharmaceutical trade
• intellectual property rights
• patents
• trips agreement

JEL Classification:

• F13
• F14
• L65
• O19
• O34

Notes

¹ The TRIPS agreement originally stated that IPRs were to create incentives for new ideas which could benefit the whole of society. This objective is detailed in Article 7: ‘The protection and enforcement of intellectual property rights should contribute to the promotion of technological innovation and to the transfer and dissemination of technology, to the mutual advantage of producers and users of technological knowledge and in a manner conducive to social and economic welfare, and to a balance of rights and obligations’ (WTO, 2012).

² Generic manufacturers use reverse engineering to develop generic formulations.

³ There is some evidence that pharmaceutical firms tend to set prices above purchasing power in developing countries. Firms fear that if the difference in prices between poorer and richer countries is too high, rich countries will import drugs from poorer countries (such as Mexico) to benefit from lower prices (Danzon and Furukawa, 2003).

⁴ For example, the US patents for Lipitor and Plavix, the two drugs which generated the largest sales worldwide in 2011 (IMS Health, 2011), expired in 2011.

⁵ A major health concern for developing countries is that patent protection limits the access of developing countries to inexpensive essential drugs that could be traded by generic manufacturers when no IPR protection measures are enforced (e.g. Chaudhuri et al., 2006; El‐Said and El‐Said, 2007; Akaleephan et al., 2009; Shaffer and Brenner, 2009).

⁶ Proponents of patent protection argue that access to drugs in developing countries would not be secured even if intellectual property rights did not exist, because inefficient health systems and infrastructure (transportation, electricity, clean water supply) are major impediments to sustainable health care in developing countries (Watal, 2000; PhRMA, 2003).

⁷ The use of the Ginarte and Park index can generate a problem of endogeneity, because it indicates the strength of a country’s patent laws. The index therefore does not take into account the fact that some countries might have laws generated by pressure from the US. Some papers also used the index created by Rapp and Rozek (1990), which rates countries’ legislations according to minimum standards established by the US Chamber of Commerce in 1987. This index does not cover the period studied in this article.

⁸ PEPFAR’s original 15 focus countries are Botswana, Cote d’Ivoire, Ethiopia, Guyana, Haiti, Kenya, Mozambique, Namibia, Nigeria, Rwanda, South Africa, Tanzania, Uganda, Vietnam and Zambia.

⁹ PEPFAR has been extended to a new 5-year period (2009–2013). It has benefited from increased funding to 48 billion dollars, and its scope has widened (PEPFAR, 2012).

¹⁰ The components of the actual flows variable are as follows: trade (in per cent of GDP) for 23%, FDI in per cent of GDP) for 29%, portfolio investment (in per cent of GDP) for 27%, and income payments to foreign nationals (in per cent of GDP) for 22%.

¹¹ Since is undefined, log-linearization leads to the estimation of a truncated sample. A common way of solving this problem in the literature has been to use an ad hoc correction for the presence of zeros with OLS: the dependant variable is the natural logarithm of the value of trade plus a small constant, such as 1 or 0.01 (e.g. Eichengreen and Irwin, 1995). Another solution has been the use of a Tobit estimator. However, both of these solutions generate inconsistent estimates of the parameters of interest (Silva and Tenreyro, 2006).

¹² Martin and Pham (2008) have argued that the PPML estimator is appropriate if zero-trade values are not frequent. However, Silva and Tenreyro (2010) have justified their approach even in the case of large zero-trade values.

¹³ Although Poisson and negative binomial models are originally count data models, Woolridge (2002) shows that they can be used to analyse models with nonnegative continuous variables (see Burger et al., 2009).

¹⁴ As robustness tests, the negative binomial, OLS and PPML estimators were applied to an equation in which GDPPC was replaced by GDP and the population variable was kept, and another equation in which GDPPC was replaced by GDP but the population variable was dropped. In both cases, there was no major change regarding the results of the other variables.

¹⁵ FTAs tend to be signed with countries that are candidates to becoming production platforms for US firms. These FTAs are likely to have a larger impact on US imports than exports.