INTRODUCTION
The impact of intellectual property rules and practices on the health of poor people in developing countries has generated substantial controversy in recent years. Although this predated TRIPS,[1] and featured prominently in the TRIPS negotiations, impetus has been added by the coming into force of TRIPS, and the dramatic rise in the incidence of HIV/AIDS, particularly in developing countries. For the developed countries, the pharmaceutical industry was one of the main lobbyists for the global extension of IP rights.[2] For developing countries, a major concern was how the adoption of intellectual property regimes would affect their efforts to improve public health, and economic and technological development more generally, particularly if the effect of introducing patent protection was to increase the price and decrease the choice of sources of pharmaceuticals.
We are aware of the importance of effective patent protection for the industry most directly involved in discovering and developing new pharmaceuticals. Indeed, without the incentive of patents it is doubtful the private sector would have invested so much in the discovery or development of medicines, many of which are currently in use both in developed and developing countries. The pharmaceutical industry in developed countries is more strongly dependent on the patent system than most other industrial sectors to recoup its past R&D costs, to generate profits, and to fund R&D for future products. Successive surveys have shown that the pharmaceutical companies, more than any other sector, think patent protection to be very important in maintaining their R&D expenditures and technological innovation.[3] The industry understandably takes a close interest in the global application of IPRs, and generally resists the contention that they constitute a major barrier to access or a deterrent to development in developing countries. For instance, Sir Richard Sykes, the former Chairman of GSK, said in March this year:
“Few would argue with the need for IP protection in
the developed world, but some question whether it is appropriate to extend its
coverage to the developing world, which the TRIPS agreement is gradually
doing. As I have said, IP protection is
not the cause of the present lack of access to medicines in developing
countries. At Doha last November, WTO
members agreed to defer TRIPS implementation for the least developed countries
until 2016. I do not believe that TRIPS
will prevent other developing countries like Brazil and India from obtaining
access to the medicines they need. On
the other hand, I firmly believe that these countries have the capacity to
nurture research-based pharmaceutical industries of their own, as well as other
innovative industries, but this will only happen when they provide the IP
protection that is enshrined in TRIPS.
TRIPS needs to be recognised as an important industrial development tool
for developing countries.”[4]
That said, we are also fully aware of the concerns expressed by, and on behalf of, developing countries about the impact that such rights may have in those countries, particularly on prices of pharmaceuticals. If prices are raised, this will fall especially hard upon poor people, particularly in the absence of widespread provision for public health as exists in most developed countries. Thus others from many developing countries, and the NGO community, have argued the opposite:
“Why do developing countries object so strongly to
TRIPS? Its essential flaw is to oblige all countries, rich and poor, to grant
at least 20 years' patent protection for new medicines, thereby delaying
production of the inexpensive generic substitutes upon which developing-country
health services and poor people depend. And there is no upside: the increased
profits harvested by international drug firms from developing-world markets
will not be ploughed back into extra research into poor people's diseases - a
fact some companies will in private admit.”[5]
Our starting point in this analysis is that healthcare considerations must be the main objective in determining what IP regime should apply to healthcare products. IP rights are not conferred to deliver profits to industry except so that these can be used to deliver better healthcare in the long term. Such rights must therefore be closely monitored to ensure that they do actually promote healthcare objectives and, above all, are not responsible for preventing poor people in developing countries from obtaining healthcare.
A spur to much of the recent debate has been the HIV/AIDS pandemic, although the issue of access to medicines in developing countries goes much wider. It is particularly important not to allow the debate in this area to be influenced unduly by the HIV/AIDS experience, dramatic though it is. Apart from HIV/AIDS, which is the biggest single cause of mortality in developing countries, TB and malaria claim almost as many lives. Together all three diseases claimed nearly six million lives last year, and led to debilitating illness for millions more.[6] In addition, there are a number of less common diseases which are collectively important. These include, for instance, measles, sleeping sickness, leishmaniasis and Chagas disease.[7]
Each group of diseases presents different problems in respect of the development of cures and treatments, and the economics of the R&D process. For diseases prevalent in the developed world as well as developing countries, such as HIV/AIDS, cancer or diabetes, research in the private or public sector in the developed world may produce treatments that are also appropriate to the developing world. For these diseases, one would expect that the promise of strong IP protection in the developed world would act as a major incentive for investment in R&D. But it should be noted that some strains of HIV/AIDS in Africa, for example, are different from those in developed countries, so different treatments may need to be devised.
Where appropriate treatments already exist, access to them depends on affordability, and the availability of the health service infrastructure to support delivery. We regard the cost of pharmaceutical products as an important concern in developing countries because most poor people in developing countries pay for their own drugs, and state provision is normally selective and resource-constrained. This is generally not the case in the developed world where costs are mainly met by the state or through insurance schemes. Even so the cost of drugs is a controversial political issue in developed countries, for governments and for patients not covered by effective state or insurance schemes.[8] In developing countries, inadequacy of the infrastructure is an important problem, and may mean that even inexpensive medicines are not used, or that they may be misused and contribute to the emergence of drug resistant pathogens or a virus.
Again, HIV/AIDS provides a helpful illustration of the issues. The treatment of HIV with anti-retrovirals (ARVs), or drugs to treat opportunistic infections associated with the disease, raises the affordability issue acutely. The minimum annual costs of ARV therapies, even at deeply discounted or generic prices which do not cover R&D costs, far exceed the annual health expenditure per capita of most developing countries. Current per capita health expenditures in low income developing countries average $23 per year, but the most inexpensive ARV triple therapies are now just over $200 per year.[9] Thus, without extra funding for medicines and health delivery services, treatment for all those requiring it will remain unaffordable even at the cheapest generic prices. The World Health Organisation (WHO) estimates that fewer than 5% of those who require treatment for HIV/AIDS are receiving ARVs. Only about 230,000 of the 6 million estimated to be in need of such treatment in the developing world actually receive it, and nearly half of these people live in Brazil.[10]
Similar questions about affordability arise for treatments of other diseases. For example, TB and malaria are for the most part prevalent in developing countries, although there is a resurgence of TB in the developed world. It also needs to be remembered that TB is the leading cause of death among HIV-infected people, and about one third of them are co-infected with TB.[11] For these diseases, and for diseases exclusive to the developing world, the issue is both how to mobilise resources for R&D from the private and public sectors for new medicines, and having developed them to ensure access for those that need them.
The latter point is one of the most crucial questions concerning healthcare in developing countries. How can the resources necessary to develop new drugs and vaccines for diseases that predominantly affect developing, rather than developed, countries be generated when the ability to pay for them is so limited? Even when there is a developed country market from which these resources can be recovered through high prices, how can the affordability of these drugs in developing countries be secured? How can conflicts between the two objectives – covering R&D costs and minimising consumer costs – be resolved? As with technological development more generally, does the IP system have a role to play in stimulating the capacity of developing countries themselves to develop and produce drugs that they or other developing countries need?
This is the context in which we need to consider the role that IPRs could play in helping to address these dilemmas. It is not for us to consider in any depth the wide range of factors that affect the health of poor people or the quality of health services in developing countries. These have been discussed at some length in the recent report of the WHO Commission on Macroeconomics and Health (CMH).[12] The CMH concluded that a large injection of additional public funds into health services, infrastructure and research was required to address the health needs of developing countries. It took the view that patent protection offered little incentive for research on developing country diseases, in the absence of a significant market.[13] As regards access to medicines, it favoured coordinated action to establish a system of differential pricing[14] in favour of developing countries backed up, if necessary, by the more extensive use of compulsory licensing.[15]
Those conclusions are relevant for our current task. It is our role to indicate in greater detail how changes in intellectual property rules and practices could contribute to better health for poor people, while being fully aware that such changes have to be complemented by the range of actions suggested by the CMH.
We do this by considering three main questions:
· How does the intellectual property system contribute to the development of drugs and vaccines that are needed by poor people?
· How does the intellectual property system affect the access of poor people to drugs and availability?
· What does this imply for intellectual property rules and practices?
Research Incentives
It is estimated that less than 5% of the money spent worldwide on pharmaceutical R&D is for diseases that predominantly affect developing countries.[16] Pharmaceutical research by the private sector is driven by commercial considerations and if the effective demand in terms of market size is small, even for the most common diseases such as TB and malaria, it is often not commercially worthwhile to devote significant resources to addressing the needs. In 2002, the world drug market is valued at $406 billion, of which the developing world accounts for 20%, and low income developing countries very much less.[17] In many pharmaceutical companies, research objectives are set by reference to threshold returns. We were given to understand that the large pharmaceutical companies are unwilling to pursue a line of research unless the potential outcome is a product with annual sales of the order of $1 billion. Given that private companies have to be primarily responsible to their shareholders, this necessarily leads to a research agenda led by the market demand in the markets of the developed world, rather than by the needs of poor people in the developing world, and thus a focus mainly on non-communicable disease.
Regardless of the intellectual property regime prevailing in developing countries, in reality there is little commercial incentive for the private sector to undertake research of specific relevance to the majority of poor people living in low income countries. Accordingly, little such work is done by the private sector. Total pharmaceutical R&D in the private sector has more than doubled in the last decade to an estimated $44 billion in 2000.[18] Exactly what proportion of this is directed to diseases afflicting mainly developing countries is difficult to determine. However it has been estimated that of 1393 drugs approved between 1975 and 1999, only 13 were specifically indicated for tropical diseases.[19] Where diseases are common to both developed and developing countries, the picture is different. Thus, there is significant private sector R&D on HIV/AIDS. This contrasts with the limited work on tuberculosis and malaria, and virtually none on diseases such as sleeping sickness.[20] As regards HIV/AIDS, there are now 64 approved drugs in the US for treatment of the disease and opportunistic infections, and 103 in development.[21]
In the case of the public sector, such as the National Institutes of Health (NIH) in the US or Medical Research Councils (MRCs) in other developed countries, the situation is little different because their research priorities are principally determined by domestic considerations. Public sector spending on health research was estimated to be $37 billion in 1998, of which $2.5 billion was spent in low and middle income developing countries.[22] In 2001 the US National Institutes of Health (NIH) alone accounted for over $20 billion. In addition, charitable foundations are estimated to have spent $6 billion.[23] The WHO’s Special Programme for Research and Training in Tropical diseases (known as TDR) receives only about $30 million annually. The exact proportion of public sector spending on diseases relevant to developing countries has not been authoritatively estimated, but seems unlikely to be higher than 10%.[24] This situation is now being addressed through the WHO, the Global Forum for Health Research, the initiative of Médecins Sans Frontières (MSF) on drugs for neglected diseases, additional funding by foundations and the development of several public-private partnerships to address specific diseases.[25] But the overall level of funding for these new efforts is still very modest in relation to the scale of the problem and global R&D expenditure of about $75 billion, and the outcome uncertain.
So what role does IP protection play in stimulating R&D on diseases prevalent in developing countries? All the evidence we have examined suggests that it hardly plays any role at all, except for those diseases where there is a large market in the developed world (for example, diabetes or heart disease). There is some weak evidence related to an increase in indicators of research activity in malaria since TRIPS was agreed, but the relation between cause and effect is not at all clear.[26] The heart of the problem is the lack of market demand sufficient to induce the private sector to commit resources to R&D. Therefore, we believe that presence or absence of IP protection in developing countries is of at best secondary importance in generating incentives for research directed to diseases prevalent in developing countries.
Thus this research may be inadequate in quantity because of inadequate effective demand from developing countries where the disease is heavily concentrated. Moreover research, particularly on vaccines, may require tackling characteristics of diseases specific to developing countries, where the solution for the developed world may not address the problem in the developing world. For example, the majority of HIV vaccines are being developed for genetic profiles of subtype B, prevalent in developed countries, but most AIDS sufferers in developing countries are types A and C. Vaccine research for HIV is also particularly scientifically challenging because of the way the virus evades the body’s natural immune responses, and the way it mutates.[27] Malaria vaccine research is also challenging, because of the size and diversity of the malaria parasite, and the complexity of its mutations.[28] Thus, for the private sector, vaccine research is a high risk/low return investment, particularly in relation to disease types prevalent in developing countries. The market tends to undervalue the social returns from vaccines, more than is the case for treatments.[29] In the case of malaria, the market demand is dominated by prophylaxis for travellers from developed countries, rather than vaccines which would be of greater relevance to sufferers in the developing world.
In respect of TB, where there are an estimated eight million people in developing countries that have the disease, no new class of anti-TB drug has been developed for over 30 years. Current treatments require drug courses of 6 months or more. A drug that produced the same effect in two months could have a dramatic impact in helping to control the disease globally. The scientific challenge of producing such a medicine is significant because of the characteristics of the disease.[30] A recent report by the Global Alliance for TB Drug Development has estimated that based on market demand (both private and public, including from developed countries) there might in fact be a respectable financial rate of return on the estimated cost of developing a new and improved drug. Nevertheless it is still not considered that IP protection, and favourable economics, will induce investment without considerable public sector involvement.[31] The current business model of the research-based pharmaceutical companies is such that research expenditure and profit generation are dependent on the sales of a few “blockbuster” drugs (normally with sales in excess of $1 billion per annum), which help finance the high percentage of failures in the R&D process.[32] But these companies have the freedom to pursue promising avenues wherever they may lead (for example, treatment for a disease or condition not previously envisaged). The economics of research for a specific treatment for a particular disease have to be very favourable to induce significant research effort.
Some, such as Sir Richard Sykes above, have argued that providing IP protection in developing countries with significant scientific and technical skills will help to increase the amount of research devoted to developing country diseases. Evidence on this is lacking because most of the relevant countries have only just introduced TRIPS-compliant laws, or are yet to do so. But we see no reason why firms with research capability in developing countries should respond to global IP and market incentives significantly differently from those based in developed countries. There is some evidence for this behaviour from firms in countries such as India.[33] The reality is that private companies will devote resources to areas where an optimal return can be made. Moreover, widely supported moves to establish differential pricing would reduce margins to reward R&D in developing countries, further undermining any incentive for additional research on developing country diseases.
In short we do not think that the globalisation of IP protection will make a significant contribution to increasing R&D expenditure by the private sector relevant to the treatment of diseases that particularly affect developing countries. The only feasible way to do this is by increasing the quantity of international aid resources devoted to such R&D. The CMH recommended an additional $3 billion annually to be spent on R&D through a new Global Health Research Fund, existing mechanisms and public-private partnerships.[34]
How increased publicly funded research should be directed requires careful consideration. It should not act as a form of subsidy to the existing pharmaceutical industry, although the industry certainly has an important part to play. The opportunity should be taken to build up the capacity of developing countries themselves to undertake R&D on treatments for those diseases which particularly affect them. In the technologically more advanced developing countries, such research can be highly cost-effective. For instance, General Electric has established its second largest R&D Centre in the world in India, employing about 1000 PhDs and 27 other global firms set up R&D centres in India between 1997 and 1999.[35] Thus research could be conducted with the active participation of selected research institutions and companies in developing countries, taking advantage of the human resources available in such countries and lower R&D costs. The institutional structure of for such funding also needs thought. The CGIAR[36] network of agricultural research institutes (which we discuss in Chapter 3) is one model. More promising in this context might be a network of public-private partnerships in developing countries, taking advantage of the concentration of research resources in public sector institutions but also the opportunity to build research capacity in the private sector. In particular the arrangements for intellectual property arising from such research need to be such that access by the poor to the products of research is ensured as much as possible.
Public funding for research
on health problems in developing countries should be increased. This additional funding should seek to
exploit and develop existing capacities in developing countries for this kind
of research, and promote new capacity, both in the public and private
sectors.
Although IP may not have much to contribute in generating additional research relevant to poor people, it is clear to us that there are important issues about the impact of the patent system on the research process. While patent protection provides an incentive for R&D, the patenting of intermediate technologies (particularly gene-based ones) required in the research process may actually create disincentives for researchers in terms of accessing, or unwittingly infringing patents on, technologies they need.[37] This is an area where patent practices in the developed world can impinge directly on what research is done for people in the developing world, and there are implications for the type of patent regimes that developing countries adopt. The IP arrangements in public-private partnerships also give rise to important questions of managing IP to benefit poor people. We consider these questions in Chapter 6.
ACCESS TO MEDICINES FOR
POOR PEOPLE
The purpose of patents, as we have noted, is to provide a temporary monopoly to rights holders as a stimulus to inventions and their commercialisation. However, it should also be noted that the monopoly right provided by a patent normally only excludes others from making, using or selling that particular invention. It does not prevent competition from other drugs, patented or not, that address the same medical conditions. Nevertheless, other things being equal there is a presumption that the producer of a patented product, through the ability to exclude copies, will attempt to earn a monopoly profit and charge higher prices than would otherwise be the case. That, indeed, is the basis of the system. The bargain with society is precisely that the benefits to society generated by the extra innovation induced (for example, a lifesaving drug which might not exist but for the patent system) should exceed the extra cost of the product.
Given that in developing countries most people are poor and that patent protection can increase prices, it is necessary to examine with particular care the arguments put forward by some that patents in developing countries are not likely significantly to affect access to pharmaceuticals subject to patent protection. There are two grounds on which this argument is made. First, because patents are not always sought in some – especially smaller - developing countries, they cannot be a significant problem in accessing medicines. Secondly, even if they are sought, either this is not a determining factor in pricing or there are other overriding factors that prevent access to drugs by the poor.
Prevalence of Patenting
It is true that, although patent protection for pharmaceutical products is available in most developing countries, multinational companies have not patented their products in all of them. This is normally the case for countries with small markets and limited technological capacity. Companies may take the view that it is not worth the expense of obtaining and maintaining protection when the potential market is small, and the risk of infringement low. For instance, a recent study in 53 African countries found that the extent of patenting of 15 important antiretroviral drugs was 21.6% of the possible total.[38] In 13 countries there were no patents on these medicines at all. The conclusion was drawn that, because the patenting rate was so small, patents “generally do not appear to be a substantial barrier to…treatment in Africa today”, although it was recognised that there would be an issue when TRIPS came into force for all WTO members.[39]
Although the overall prevalence of patents found in the study is relatively low in aggregate, it is perhaps surprising that it is not lower, given the very low treatment rates, small markets, and the fact that few countries are capable of producing generic copies. The prevalence of patents is very much higher in countries where there is a substantial market, and technological capacity. Thus in South Africa (which alone counts for over 17% of Africa’s HIV cases) 13 of the 15 drugs are patented. There are 6-8 patents for these drugs in Botswana, Gambia, Ghana, Kenya, Malawi, Sudan, Swaziland, Uganda, Zambia and Zimbabwe, which together account for another 31% of HIV cases in sub-Saharan Africa.[40]
The industry points out that the prevalence of patenting is very much lower, or nil, for a wide range of drugs to treat other diseases. Until the latest revision this year, less than 5% of the drugs on the WHO Essential Drugs List were patented.[41] An industry survey indicated that 94% of countries surveyed had no patents on TB and malaria drugs, and no country has patents on all the relevant drugs for these diseases. There were no patents at all on drugs for trypanosomiasis or diarrhoeal diseases.[42] The argument advanced by industry is that even where there is no patent protection, the drugs are still not available.[43] For instance, even where vaccines are available for various common diseases and cheap (for example, less that $1 for a polyvalent vaccine), WHO’s Expanded Programme of Immunisation (EPI), in spite of undoubted successes, still fails to reach many children who could benefit.
This is of course true, but it does not follow that the patent system has no adverse effects. Even if patents do not exist for particular products and countries, the patent system may still have an effect on access to medicines. Most low income developing countries have to rely on imports for their supplies. The existence of patents in potential supplier countries may allow the patentee to prevent supplies being exported to another country, particularly through controls on distribution channels. This is another reason why companies may selectively patent in countries such as South Africa because it is a potential supplier to its poorer neighbours in the rest of Southern Africa (or indeed elsewhere). At present, importing countries where there is no patent protection have the option of importing supplies from generic companies, principally in India, because India need not have pharmaceutical product protection until 2005. But thereafter, under TRIPS, new drugs and those for which patent applications were submitted after 1994 will be patentable, and the opportunity for these imports will diminish correspondingly over time. However, it should be noted that all existing drugs produced as generics in India or elsewhere will continue to be available for export provided, of course, they are not patented in the importing country. We return to this issue below in our discussion of policy options.
Patents and Prices
The importance of prices of medicines to poor consumers in developing countries is perhaps obvious. But it is worth emphasising that if a sick person has to pay more for a pharmaceutical product as a result of a patent, it means that he or she will have less to spend on other essentials of life such as food or shelter. Alternatively, foregoing the medicine because it is unavailable or unaffordable may result in long term ill health, or death. That is why it is essential to consider the impact of the introduction of an IP regime on prices, while recognising that prices are affected by many factors. These include purchasing power, competition and market structure, responsiveness of demand to price and government price controls and regulations.
It is particularly difficult to observe directly and isolate the impact of introducing patents in developing country markets. In part we have to rely on econometric models to simulate the impact of introducing patent protection, and in part the experience of developed countries where generic producers compete with research-based ones.
There is extensive evidence from developed countries that prices fall quite steeply as soon as drugs go off patent, assuming there are generic competitors. The price fall seems to be greater the more generic competitors enter the market. Governments can encourage price reductions by facilitating the early entry of generic producers into the market. For instance, the 1984 Drug Price Competition and Patent Term Restoration Act in the US (known as the Hatch-Waxman Act) did precisely that, resulting in the share of generics in prescriptions dispensed rising from 19% in 1984 to 47% in 2000.[44] In other developed countries, such as the UK, the generic share of the market is often much higher. Pharmaceutical companies have also brought or defended expensive court actions to delay or prevent generic entry, and to protect or extend a monopoly on a best selling drug.[45] Correspondingly, we must remember that generic producers are governed by market incentives just as the research-based industry, and that it is necessary to encourage competition within the generic industry if lower drug prices are to be achieved. A recent study in the US found that prices fall when generic competition enters the market but at least five generic competitors are necessary to push prices down to a minimum.[46] The number of competitors entering the market, and the speed with which they do so, will depend on the expected profits. A crucial finding is that the full benefits of competition will only be felt at quite large market sizes – in smaller markets fewer generic firms will consider the market worth entering and prices to consumers will be higher. This is very relevant to the position of developing countries, as discussed below.
Developing countries can also limit the costs of the patent system for their population by facilitating generic entry and generic competition. But in most cases their options are severely limited by the small size of their markets and lack of indigenous technological, productive and regulatory capacity. It is this lack of capacity to create a competitive environment for both patented and generic products that makes the existence of patents more contentious than in developed markets with greater capacity to enforce a strongly pro-competitive regulatory environment.
International comparisons show that copies of drugs patented elsewhere are much cheaper in markets which do not offer patent protection. The Indian market, where there is no product protection, is the lowest priced in the world. One of our studies indicated that for 12 drugs covering a range of conditions US prices range from four to 56 times the price of equivalent formulations in India, and yet still a large number of people in India cannot obtain access to them.[47]
However, studies of multinational company pricing policies (mainly for ARVs) indicate that until recently there was remarkably little correlation between the price of the same drug and a country’s per capita income. This correlation is expected on theoretical grounds because companies should be able to make more profits by charging low prices in low income markets and high prices in high income markets (known as differential pricing), than by charging a uniform global price. But prices have appeared to vary more or less randomly between countries. Some developing countries paid more than US prices and some less. At best there was a very weak relationship between wholesale drug prices and per capita income.[48] The actual price to the patient is complicated by import duties, local tariffs, taxes and wholesaler profits.[49]
In the last two years this situation may have changed somewhat as some companies have drastically lowered prices offered in response to international pressure, principally from NGOs, and potential competition from generic manufacturers, particularly from India. For instance, between July 2000 and April 2002 the annual cost of a branded triple therapy ARV combination fell from over $10000 to just over $700 for selected groups of consumers. By then the lowest generic price for this combination had fallen to $209.[50]
But to estimate the impact of introducing patent regimes anew in developing countries, it is necessary to use econometric models. There is a small but growing literature, that relates almost entirely to lower and middle income developing countries which already have significant pharmaceutical industries. This literature demonstrates that the introduction of patent regimes into such developing countries has, or is predicted to have, the effect of raising prices. The estimates range widely depending on the drugs and countries being considered – from 12% to over 200%, but even the lower estimates imply very substantial costs for consumers.[51] The range of estimates is indicative of the degree of uncertainty about the dynamic effect of introducing patents, and suggests that the outcome will be very much determined by market structure and demand, in particular the degree of competition.