A review of the health and economic implications of patent protection, with a specific focus on Thailand

Background Although it has been two decades since the Thai Patent Act was amended to comply with the Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), there has been little emphasis given to assessing the implications of this amendment. The purpose of this review is to summarize the health and economic impact of patent protection, with a focus on the experience of Thailand. Methods A review of national and international empirical evidence on the health and economic implications of patents from 1980 to 2009 was undertaken. Results The findings illustrate the role of patent protection in four areas: price, present access, future access, and international trade and investment. Forty-three empirical studies were found, three of which were from Thai databases. Patenting does increase price, although the size of effect differs according to the methodology and country. Although weakening patent rights could increase present access, evidence suggests that strengthening patenting may benefit future access; although this is based on complex assumptions and estimations. Moreover, while patent protection appears to have a positive impact on trade flow, the implication for foreign direct investment (FDI) is equivocal. Conclusions Empirical studies in Thailand, and other similar countries, are rare, compromising the robustness and generalizability of conclusions. However, evidence does suggest that patenting presents a significant inter-temporal challenge in balancing aspects of current versus future access to technologies. This underlines the urgent need to prioritize health research resources to assess the wider implications of patent protection.


Introduction
Access to information that is generated through research and development (R&D) is a public good [1]. Because it is impossible to exclude people from using it, a price that reflects the actual cost of production cannot be charged. To address this, patents present a legal system that provides short-term exclusivity (or monopoly rights) over the production and sale of a specific product resulting from R&D. This allows the firm to sell at a price higher than would otherwise be the case, compensating the costs of R&D. However, there is some concern that the patent price is used to achieve 'super-normal' profits (profits in excess of those required to recoup R&D costs) at the detriment of wider access to patented products [2].
The implications of patenting spread further, as innovation, technology, and knowledge development are crucial drivers of economic development and technology transfer resulting from international trade and investment. The topic of patenting has found its way onto the global agenda due to the World Trade Organization (WTO)'s Agreement on Trade-Related Aspects of Intellectual Property Rights (TRIPS), which expanded the Western tradition of patenting to all members of the WTO. Under this Agreement, patent protection must be available for at least 20 years, must be without discrimination against the place of invention or origin of product, and must apply to both products and processes [3].
This has generated especially heated debate within the health community concerning the impact that patents may have on the price of and access to medicines, affecting both availability and affordability. However, arguments concerning patenting tend to take one of two sides: that patenting should be continually strengthened to encourage greater trade and investment, or that patenting should be weakened to ensure that medicines are as cheap as possible in the belief that this will ensure the greatest access for those in need. There is seldom, if ever, consideration of both sides when informing policy makers how to strike a balance between affordable medicines, both now and in the future, and trade and investment. For instance, whilst continually strengthening patenting will likely lead to higher prices, thus further reducing access, weakening patenting may stifle longterm access since pharmaceutical companies might be reluctant to introduce new medicines into the market, and foreign investors may look to invest in other countries where there is better protection of their products. In order to determine the appropriate balance in policy (such as the use of TRIPS-flexibilities), it is important to establish: (i) the impact that patent protection has on price; (ii) the impact that price has on current and future access; and (iii) the impact that patents have on innovation in national and international settings. This paper addresses these issues through a review of the literature concerning these areas, with a specific focus on Thailand.

Patenting and Thailand
Thailand is a lower-middle-income economy with a 2007 per capita Gross National Income of US$3,400 and a total population of 63.3 million [4]. Health services are provided by both public and private insurance schemes, with public insurance schemes covering 97% of the population. In 2007, 72% of national health expenditure was financed by public expenditure [5]. The prices of medicines in Thailand are set mainly by market competition, with no policy related to price regulation [6]. In 1979, Thailand's Patent Act (B.E.2522) established the first legal protection for inventions in the country, although only process patents for pharmaceuticals were originally covered. This Act was revised in 1992 to introduce product patent protection for pharmaceutical products (13 years ahead of TRIPS compulsory compliance) [7].
The justification to amend the Thai patent law was to interest multinational companies to invest in Thailand. The other expected benefit of restricted patent protection in medicines is that this could increase domestic capability and strengthen the local pharmaceutical industry by the transfer of new technologies into the country [8], as the aim of a patent is to encourage technology transfer. Although technology diffusion can take place through a variety of channels that involve the transmission of ideas and new technologies-such as the imports of high-technology products, adoption of foreign technology and acquisition of human capital through various means-Foreign Direct Investment (FDI) was claimed as the most important channel for technology transfer [9,10].
This revision coincided with the rise of HIV/AIDS as a major health problem together with concern over the rising costs of anti-retrovirals (ARV). For instance, 200 mg (100 capsules) of original efavirenz sold in 2006 at 3,192 baht per bottle, while a generic equivalent was available at 1,292 baht [11]. A similar concern was expressed over other new medicines, which were not covered by the National Health Insurance system due to their high price [12]. The Sub-committee on selecting essential medicines under the National Health Insurance schemes therefore proposed compulsory licensing for seven patented medicines during 2006-2008 [13].
There were reactions from pharmaceutical companies. For instance, Abbott Laboratories withdrew its registration application for 10 new medicines in protest of the government use license on its product. These reactions were not confined to the pharmaceutical industry. In 2007 the Office of the United States Trade Representative elevated Thailand's ranking from the Watch List to the Priority Watch List, indicating concerns over deficiencies in Intellectual Property Rights (IPR) protection and enforcement [14], and announcing that privileges under the Generalized System of Preferences would be removed for three Thai products: gold accessories jewellery, polyethylene terephthalate, and flat screen television sets [15].
From the experience of pharmaceutical patent protection and associated policies during 1992-2008, it is apparent that patent protection has both health and economic consequences. A conceptual framework illustrating the broad implications of patent protection is illustrated in Figure 1. Patent protection affects the price of pharmaceuticals whereby price is a component in determining affordability and therefore access to existing medicines and industry investment in introducing or developing new medicines. A higher price would hinder access, but stimulate the development of new medicines through a higher R&D budget enabling patients to benefit from access to new medicines in the future. Patent protection is also accompanied by foreign investment in domestic facilities for the production of pharmaceuticals. Finally, as indicated above, there are wider trade relationships that may be affected by patent decisions, and which are not related to medicines at all.

Methods
To ensure a manageable review, inclusion criteria covered the study scope and the type of study.

Study scope
Studies which assessed the implications resulting from patent protection for pharmaceutical products and/or processes, especially: the benefit and cost of patent protection for the pharmaceutical industry the interplay between patents and the affordability and availability of medicines the relationship between patents and new drug development the implications of patent protection on the wider economy, i.e. trade or foreign direct investment.   were set so as to ensure the inclusion of all work conducted when the requirement of global standard patent protection was needed as, since the 1980s, intellectual property has became an important business tool, and new internationally-agreed trade rules for intellectual property rights were seen as a way to cope with the international economic tension [16].

Types of studies
The keywords employed in search queries covered five areas of interest: (i) patent policy, (ii) health, (iii) price, (iv) access, and (v) economy. In each area, the related keywords are identified in Table 1.

Results
Initial searches resulted in 4,012 abstracts, of which 61 were from Thai databases. Of these, 43 passed the inclusion criteria, including only three from the Thai databases. Papers were categorised into: (i) role of patent on price, (ii) role of patented price on present access, (iii) role of patented price on future access, and (iv) role of patent on international trade and investment. Tables 2, 3, 4, 5, and 6 provide a brief summary of these empirical studies.

Role of patent on price
Twelve studies, including two Thai studies, looked at the effect of patents on price. Most of these studies focused on the patent expiration effect in the USA. Patent protection appears to increase price by around 26%-277% depending on which of the three estimation approaches is used.    The first approach uses elasticity of demand to calculate price. This has been used to estimate the likely effects of patents on the price of medicines not currently under patent protection, and to extrapolate to a situation of those medicines being under TRIPS obligations. Using this methodology, Watal (2000) showed that all patentable medicine prices in India would increase by a mean of 26% with linear demand and 242% with  How patent rights and price regulation affect whether new medicines are marketed in a country, and how quickly Probit model Using probit models of the probability that a new medicine is launched in a given country within either two years or ten years of the medicine's first appearance in the global market and a log-logistic hazard model of the time path of country launches constant price elasticity of demand [17]. Similarly, by accounting for different products through trademarks and advertising, Fink (2004) used this approach to estimate that prices would increase by 30-277% if these medicines came under patent protection [18]. This approach would be useful if the price elasticity of demand is known and correct. For the pharmaceutical market, the consumption decision commonly involves participation by a physician and a third party payer (government or hospital committee). The consumer may or may not play some part in the price payment, depending on a country's specific regulatory and reimbursement regimes. The pharmaceutical market's demand function is thus often distorted, and a model based on price elasticity of demand might not present a real world situation of the complexity of the pharmaceutical market.
Second, the observation of price before and after patent expiration is used to infer the price effect of patent protection. Boersma et.al. (2005) illustrated that-when there is no patent protection-prices generally fall by 50-70% [19]. Suh et al. (2000) showed a decline to approximately 30% of the original price three years after patent expiration [20]. Also, Magazzini et al. (2004) showed that the price index decreased three years after patent expiration by approximately 20% in Germany and France while the UK price index was stable [21]. Conversely, two US studies by Grabowski and Vernon (1992) and Griliches and Cockburn (1994) showed that-following generic entry-an original product can have an increase in price of 7% and 11% after one year and two years respectively [22]. Another study showed a 60% price increase three years after expiration, while the generic price decreased by 30% [23]. These increases may reflect increased advertising intensity once the market protection of patenting has expired.
However, the effect in each country will differ since each nation has a different health system in terms of medical tradition, policy for financing and supporting generic entry, and brand royalty of physicians, pharmacists and customers. The marketing strategy also differs among pharmaceutical companies, who often spend more heavily on the intensity of advertising once the patent has expired, which could explain at least some of the post-patent price increase. It appears that medicine prices, in general, depend on several supply and demand factors. For example, therapeutic advantage and number of substitutes are both significant price determinants; as the number of substitutes increased in one study from one to two, there was an average 38% decline in the ratio of the new drug price to the average existing market price [24]. Kanavos and Vandoros (2011) also found that product age has a significant and negative effect on prices [25].
Third, studies perform regression analysis of factors influencing medicine prices, of which patent is one such factor. Borrell (2007) estimated patent impact on price      The impact of IPR protection on economic growth (GDP growth)

Regression
Creating an IPR index and estimating a system of equations to identify the effect of IPR protection and other national characteristics on economic growth such as R&D activity, investment, and education Table 6 Empirical studies of the implications of patenting on economic growth and/or foreign direct investment (Continued) Athukorala  Providing the trend of FDI for industry in general and specifically for the chemical industry in Thailand in 14 ARV molecules in 34 low-and middle-income countries which have different patent regimes, where patent was eligible or ineligible between 1995 and mid-2000. This showed that combination therapy containing at least one patented medicine is on average priced 70% higher than combination therapy containing only generic alternatives. Combination therapy containing at least one original medicine is priced 16% higher than local copies even when it is introduced in to no-patent regimes [26]. In Thailand, the introduction of product patent protection in1992 seems to have had no effect on the price of patentable medicines that were in the market before 1992 [27]. Rather, the effect appears to concern only those patented medicines which were introduced subsequently. For example, one study compared the price of patented and generic HIV/AIDS medicines from 2001-4 and found that the patented price was approximately 1.5-3 times higher than the generic price in 2001 [28].
The experiences in Canada and Italy reflect the situation in Thailand. A study of the impact of the 1987 Canadian Patent Act, which extended the period of protection from seven to ten years while also allowing the generic industry to implement compulsory licensing, found that after 1987, medicine prices increased relative to pre-1987 prices [29]. Similarly, after a patent law in Italy came into effect in 1978, new medicine prices were 163% higher than new drug prices before 1978 [30].

Role of patented price on present access
Empirical evidence directly linking patented price and access is rare. Most studies describe how patenting increases prices (as above) and then assume that price affects access, but there is a lack of direct association between the extent of price increase and the extent of changes in access, controlling for other influence factors. As shown in Table 3, there were four studies found concerning this issue, two of which are in the Thai setting. Akaleephan et al. (2009) examined the effect of patent life extension from a TRIPS-Plus proposal on access to medicines. They illustrated the drawbacks of extending the period of protection by showing that the availability of generics would help to save 105% of actual government expenditure, and accessibility would increase by 54% [31]. In addition, after compulsory license introduction in Thailand, the price of generic medicines was about 3-38% of their original price. As a result, there were approximately 8,000 extra patients utilizing EFV, and it is estimated that the increased number of patients with access to EFV will be 17,959 in five years [32].
Conversely, Attaran (2004) suggested that the main obstacles are associated with the country's socio-economic status, such as the lack of manufacturing capacity or poor health care systems [33]. His survey results show that only 17 from a total of 319 medicines on the WHO Essential Medicines List are protected by patents. In addition, Borrell and Watal (2003) showed that switching all medicines under a patent regime to a no patent regime globally would have increased the percentage of AIDS patients with access to new medicines from 0.88% to 1.18% [34]. However, with reference to individual countries, the findings suggested different magnitudes of impact. For example, in Thailand where most of the relevant medicines were under patent, it was estimated that around 10,000 additional prescriptions would be prescribed if all patents were waived, generating an increase in access of some 50%.
Role of patented price on future access Incentive to introduce medicine to market Pharmaceutical companies may refuse to market new medicines in response to weak national patent policy [35,36]. Mansfield (1986) estimated that 65% of products would not have been introduced if patent protection could not have been obtained [37]. While patents make local markets more attractive, multinationals may delay or avoid launching medicines in lower-priced countries because they are concerned about the implications for pricing in other markets [38]. For instance, Lanjouw (2005) determined the effects of patent policy and price control policy on market entry, and showed that extensive price control and process-only patent protection lowers the probability of having a new medicine in lower-income countries by 30% [39]. A brief summary of these two studies is shown in Table 4.
The model employed in these two studies was multicountry, from high-income to low-income countries. Although the results are more generalized, they sometimes mislead. Under some circumstances and model assumptions, patent protection has a positive effect for some countries, while under other circumstances it has a negative effect. Single country studies are particularly effective at maximizing their explanatory leverage by exploiting the availability of comparable units of analysis, whether over market or medicine characteristic variations within a country [40].

Incentive to invent new medicine
One implication of removing patent protection to gain increased current access is that this might result in patients foregoing the opportunity to receive a new medicine in the future, as it would not be discovered or developed [41]. There were four studies looking at this possibility, as shown in Table 5. Grootendorst (2007) illustrated that this clearly generates trade-offs between benefits now and in the future [42]. Indeed, heavily depending on assumptions, Hughes et al. (2002) estimated that for every dollar in consumer benefit realized from providing greater access to current medicines, future consumers would be harmed at a rate of three dollars in present value terms from reduced future innovation [43]. Giaccotto C. et al. (2005) investigated the role of price control on new medicine development, showing that price control policy in the USA during the 1980s resulted in forgone R&D investment of US$264-293 million, translating into 330-365 fewer new medicines, which is equal to one-third of all actual new medicines launched on the global market during that time period [44]. However, such studies lack a direct link between profitability and actual investment in R&D. They illustrate the effect of patents on profit and assume that this translates directly into R&D. Conversely, an observation of the innovation activities of pharmaceutical companies affected by compulsory licensing found that there was no uniform decline in the rate of medicine patenting and other measures of inventive activity by companies affected by compulsory licenses [45] Table 5.

Role of patent protection on international trade and investment
With respect to the broader impact, in a Free Trade Agreement (FTA) between Thailand and the US, one of the 23 negotiation issues was TRIPS-Plus, which requires a higher level of intellectual property protection than existed in the TRIPS agreement [46]. Five studies focusing on the impact of patents on trade were found. Based on a Computable General Equilibrium model, it was estimated that the FTA would increase the export and import levels of Thailand by 3.4% and 4.7%, respectively [47]. This study also found, at the international level, that IPR protection had a positive influence on overall trade flows for both small and large developing economies [48][49][50]. These results are in line with other findings which show that patent protection had a positive impact on Indian pharmaceutical exports [51].

Intellectual Property Protection and investment
Nineteen studies, including one Thai study, looked at the effect of patents on FDI. Most of these studies used regression to analyze the effect of IPRs on FDI. Additional variables are included in the regression to control the differences in country specific factors. Their main similarity is in comparing IPR risk along with economic risk and/or political risk. Proxy indicators were used to represent economic or political risk. Most of these focused on the role of the national patent protection policy to attract US investors.
Six empirical studies found that patent rights protection does not influence the location choices of foreign investors. Some regression analyses of FDI in the 1980s based on research by Ferrantino (1993), Markus and Penubarti (1995), Kondo (1995), and Braga and Fink (1999) found no significant link between IPR protection and FDI [48,52,53]. Using FDI data from the 1990s, the above results were again confirmed in the study of Pfister and Deffains (2005) who investigated the role of patent protection on the location choices of French firms investing in 17 developing countries [54]. In addition, the Fosfuri's (2004)'s study, which focused specifically on the chemical industry and accounted for the differences of country characteristics, did not find that IPR protection played a significant role in fostering international activity [55].
However, some studies revealed that the volume of FDI in a country tends to be inversely related to the weakness of IPR protection. Five studies looking at the FDI determinants of US Multinational Enterprises (MNEs) were found. The results show that the strength of IPR protection has a significant positive impact on the U.S. FDI [56]. For example, a one percent rise in the perceived weakness of IPR protection would decrease the U.S. FDI in that country by 14%. In a sample of chemical firms, it was found that firms are likely to allocate their investment to sales and distributions and simple production activities rather than to manufacturing final products or to R&D facilities [57]. This is confirmed in another study [58], in which it was suggested that a one percent rise in the extent of patent protection would increase the U.S. investment in that country by 0.45% [59]. Javorick (2004) indicated that weak protection of intellectual property rights deters foreign investors in four technology-intensive sectors: (1) drugs, cosmetics and health care products; (2) chemicals; (3) machinery and equipment; and (4) electrical equipment. In addition, foreign investors, in all industries, tend to set up distribution facilities rather than engaging in local production in a country with weak IPR protection [60]. Four studies confirmed that MNEs prefer investing in the regions that have better intellectual property rights protection [61][62][63]. For example, a one point increase in IPR index would boost FDI by $1.5 billion [64].
Three studies using regression analysis yielded inconclusive results when analyzed in subgroups. Two studies showed the positive impact of strong national patent laws in developed countries, but showed a negative impact in developing countries [65,66] while another study revealed the converse results [67].
In terms of the pharmaceutical industry specifically, a strong patent system was found to have caused a considerable flow of investment into the American pharmaceutical industry [68]. However, some studies show a negative correlation between the levels of protection and foreign investment. This is supported by conclusions elsewhere that the exclusion of pharmaceuticals from patent protection was a significant factor leading Italy to become a base for the export-oriented production of generic medicines [69]. Supakankunti (2001) showed that in Thailand there has been little foreign investment in the pharmaceutical sector since the introduction of the strengthened patent law in 1992 [70]. It has been suggested that this is because foreign investors consider Thailand an unsuitable destination due to the insufficiency of well-trained human resources, technology, and equipment, as well as the inadequacy of the registration system for new medicines [8] Table 6.
In conclusion, there are a number of empirical investigations pointing to an uncertain relationship between patent and FDI distribution, which depends on the political and business risks of the country included, FDI sources, data from opinion surveys or secondary data, and the approach used to calculate the level of patent protection scale. The question of just how important patent protection is for FDI is still unsettled. Some evidence indicates that patents have had a positive impact on FDI overall and the pharmaceutical industry in particular, while other evidence suggests that weak patent protection of pharmaceuticals was the main factor in making the country a manufacturing base for these pharmaceutical companies. As both countryspecific and regional factors influence the effect of patents on FDI, more regional and country-specific studies should be conducted in order to validate the findings of this study. As noted by Lesser (2002), the effect of IPR on FDI may only be possible on a countryby-country basis.

Conclusions
The empirical literature provides answers to some important questions related to the impact of patents, although evidence remains largely inconclusive. With respect to patent impact on price, both Thai and international evidence confirm that patenting shifts prices up and has an effect on the price of the new registration of medicines. In terms of present access, international empirical evidence demonstrates that patent protection does not always impede access, whereas a Thai study suggested that implementing a limited patent life may actually increase access. As for future access, evidence suggests that strengthening patent policy in a given nation may speed up the time required for entry into the pharmaceutical market. Empirical models estimate that higher profits, from patents, would increase the number of new medicines to market through higher R&D budgets, enabling patients to benefit from access to new medicines in the future. Conversely, one observational study revealed that withdrawing exclusive rights by compulsory licensing might not have an effect on innovation in the future.
The evidence found from this review confirms that policy stimulating patent protection does have a positive impact on trade flows. In terms of FDI, evidence provides inconclusive results, both generally and specific to the pharmaceutical industry. Figure 2 summarizes existing evidence and the remaining gap, with an indication of the relationship found from this review.
The review revealed that little empirical research has been undertaken on the extent to which patent rights affect health and economic factors. With respect to health, the settings of the studies are very mixed across therapeutic areas and medicines. The literature generally shows that the size of impact varies wildly, depending on which methods are employed in the studies. Current evidence therefore makes it difficult for a country, such as Thailand, to come to a conclusion on advice to national policy makers who are to make decisions which trade off health or access impacts with wider economic issues. The high price of medicines may not be related to patent rights. Furthermore, price may not be related to access, either.

Recommendations
The trade-offs between patent protection, current and future access to medicines, and related aspects of trade and investment, are still subject to debate, since empirical studies are relatively rare, especially in countries such as Thailand. This underlines the urgent need to prioritize health research resources to assess the implications of patent protection.
It is clear that evidence on the role of patent/price on access, especially with respect to non-communicable diseases, is scare, inconclusive, and problematic. This suggests a more holistic assessment is required which takes into account a country's socio-economic status and health care system when estimating patent impact on access to medicine. The estimation of patent impact on technology transfer through FDI should be conducted on a country basis. In order to try and assess whether, on balance, a country is better off with patent policy related to health or not will require evaluating the implications for current and future access to medicines, and the wider national economy.