The effects of patent-law changes on innovation: The caseof India's pharmaceutical industry☆

George T. Haley a,⁎, Usha C.V. Haley b,1

a Center for International Industry Competitiveness, College of Business, University of New Haven, 300 Boston Post Road, West Haven, CT 06516, USAb School of Management, Massey University, South Highway 17, Gate 1, Private Bag 102904, North Shore City 0745, Auckland, New Zealand

a r t i c l e i n f o a b s t r a c t

Article history:Received 22 August 2010Received in revised form 15 May 2011Accepted 31 May 2011Available online 28 July 2011

Recent patent-law changes in India's pharmaceutical industry provide opportunities to studychanges of institutional and regulatory environments on innovation and social welfare in low-income markets. From 1972 to 2004 under its process-patent regime, India's pharmaceuticalindustry grew to become the world's fourth largest. Indian companies were becoming globallycompetitive in generics and clinical testing, and moving into product R&D. Researchers havedebated the effects of India's new product-patent laws' effects on these trends. The authorscover the domestic characteristics and global competitiveness of India's pharmaceuticalindustry. They contrast data (from 2001 to 2004) on patents in India's process-patent regimewith preliminary data (from 2005 to 2008) on patents in the country's new product-patentregime. They argue that Indian pharmaceutical companies have changed their decision-makingin response to changed patent laws by moving from process to product research. However, thepreliminary results indicate that these changes may have hurt domestic innovation. Theyconclude with strategic implications for the Indian pharmaceutical industry and highlight theneed for research and public policy to establish optimal social returns from product-patentregimes.

© 2011 Elsevier Inc. All rights reserved.

Keywords:InnovationPharmaceutical industryIndiaPatentsRegulationProcess patentsProduct patentsCorporate strategy

1. Introduction

Political and legal considerations regarding greater social returns triggered changes to the Indian pharmaceutical industry'sregulatory and institutional environments for innovation. In the 1960s, the British company, ICI Pharmaceuticals developed thehigh-blood-pressure medication, propranolol, the first beta blocker that inhibits fight-or-flight hormones such as adrenaline. But,many Indians found the drug expensive. Yusuf Hamied, head of R&D at Indian company Cipla, and the CEO's son, startedmanufacturing a cheaper version for the Indian market. ICI protested to the Indian government, and Hamied used prior legalopinions to justify his actions as corresponding with national interests to then Prime Minister Indira Gandhi. In 1959, JusticeRajagopala Ayyangar had issued a report urging that for national interests a partial process-patent regime become the law in India[1,2]2 and Hamied advocated for the Justice's recommendations. “Should millions of Indians be denied the use of a lifesaving drugjust because the originator doesn't like the color of our skin?” he asked the Prime Minister. The legal and political arguments

Technological Forecasting & Social Change 79 (2012) 607–619

☆ An earlier version of this paper was presented at the “Global Challenge of Intellectual Property Rights”, a CIBER conference, University of Connecticut at Storrs.The authors thank JayashreeWatal of theWorld Trade Organization and other conference participants for their comments. The authors also thank the participantsof the Asia Fellows seminar at the Harvard Kennedy School, Ash Institute for Democratic Governance and Innovation for their helpful comments.⁎ Corresponding author. Tel.: +1 203 931 6004; fax: +1 212 208 2468.

E-mail addresses: gthaley@sbcglobal.net, ghaley@newhaven.edu (G.T. Haley), u.haley@massey.ac.nz, uhaley@asia-pacific.com (U.C.V. Haley).1 Tel.: +64 9 4140800x9338; fax: +64 9 4418109.2 For a historical overview of Indian patent law see Damodaran [2].

0040-1625/$ – see front matter © 2011 Elsevier Inc. All rights reserved.doi:10.1016/j.techfore.2011.05.012

Contents lists available at ScienceDirect

Technological Forecasting & Social Change

appeared to persuade Indira Gandhi. In 1970, she urged parliament to change the laws governing drug patents, applying the lawsnot to the chemical compounds themselves but to the processes used to manufacture them [3]. Until 2005, despite India's benefitsas a cheap production base, the country's backward engineering of new, high-cost drugs associated with HIV/AIDS, aging, heartdisease and cancer posed a significant cost for Western companies. We provide a strategic framework to understand and to testinstitutional and regulatory changes in India's pharmaceutical industry and propose possible effects on social welfare.

Until December 31, 2004, India's regulatory and institutional environments in pharmaceuticals and agro-chemicals limitedpatent protection to providing exclusive rights only to processes through which the products were produced, rather than to theproducts. These environments allowed India's innovative, high-quality and low-cost pharmaceutical industry to develop, toproduce and to sell legally in developing countries, low-cost Indian versions of high-costWestern pharmaceuticals without patentinfringement. Indian companies had the legal rights to produce and to sell drugs if their processes to produce the drugs differedsufficiently from the original patent holders' processes [4].

On January 1, 2005, to meet the requirements for membership in the World Trade Organization (WTO), India transitioned to aproduct-patent regime. This regulatory and institutional change provides an opportunity for an early test of the premise thatproduct-patent regimes promote innovation. Unlikemany other Asian countries, India's institutional and regulatory environmentsshould promote product patents: India has a long legal history for both its legal system and patent law, as it inherited both fromthe British. India adopted its first patent law in 1856 [5]. Also, unlike other Asian former-British colonies, upon gainingindependence in 1947, India adopted, and has maintained, a secular, democratic, political system. Other countries with recenthistories of both rule of law and patent law, such as Poland, have also recently switched from process-patent to product-patentregimes. Yet, unlike India, most of these other countries had communist governments which did not recognize Western patents;and, again, unlike India, most of their governments controlled and directed pharmaceutical research, production and distribution.Hence, India provides a close-to-ideal legal jurisdiction to test previously untested claims that pharmaceutical product-patentregimes result in greater rates of innovation, which in turn can improve social health and can contribute to greater social welfareand returns.

Fig. 1 frames generally accepted axioms and tacit theories on normative relationships between institutional environments,patent regimes, innovation and social welfare. Historically, pharmaceutical drugs have had weaker patent protection in manycountries than electrical or mechanical innovations [6]. Yet, normative understandings of ideal patent environments have shapedmany regulatory developments, especially those spear headed by industrialized countries, including the WTO's Trade RelatedAspects of Intellectual Property Rights (TRIPS) agreement [7].3 Our study concentrates on patent regimes in India's pharmaceuticalindustry; additionally, we make operational Innovation as product and process patents. Reflecting the generally acceptedperspectives that have shaped TRIPS, the framework portrays product-patent regimes as affecting innovation positively throughprotecting intellectual property rights (IPR); consequently, these regimes also affect social welfare and returns positively asinnovators have greater incentives. Conversely, process-patent regimes affect innovation, social welfare and returns negatively asthe benefits of investing in intellectual property do not accrue to the innovators.

First, we discuss the evolution of India's institutional and regulatory environments. Next, we review existing theories andresearch on the effects of patent regimes on social welfare and returns. In the ensuing section, we focus on innovation in India'spharmaceutical industry, including structural characteristics, R&D, and its global positioning. Then, we review general measures ofsuccess in innovation. We contrast patents in the process-patent regime governing India's pharmaceutical industry (from 2001 to2004), with patents in the product-process regime (from 2005 to 2008). We conclude with strategic implications for the Indianpharmaceutical industry, and emphasize the need for empirical research on optimal social returns of product-patent regimes.

2. Regulatory and institutional changes in India's pharmaceutical industry

Fig. 2 provides a timeline for regulatory and institutional changes in India's pharmaceutical industry. India adopted its firstpatent laws in 1856 [5]. The first major revision came in 1859. In 1872, India undertook a major overhaul of patent laws with thePatents and Designs Act.

For our study, significant patent legislation and agreements include the 1970 enactment of the Indian Patents Act; India's 1994signings of the Uruguay Round of the General Agreement on Tariffs and Trade (GATT) and the TRIPS agreement. The Patents Act of1970 established a process-patent regime in India's pharmaceutical and agro-chemical sectors. The 1994 treaties accepted India'seventual move in January 2005, to a full product-patent regime in all Indian economic sectors, including pharmaceuticals andagro-chemicals.

The Patent Act of 1970 replaced a product-patent regime which had constrained the Indian pharmaceutical industry'sparticipation in its domestic home market [8]. By establishing a process-patent regime in the Indian pharmaceutical and agro-chemical industries, the Act opposed policies on intellectual property favored by the pharmaceutical industries in industrializedmarkets. The Act provided seven-year process patents from application time, or five-year process patents from sealing time (thedate for the official granting of the patent), whichever was shorter. In addition, if the Indian Patent Office determined that thepatents did not provide social returns, which it interpreted as the Indian public's interests, it could require the patent holders tolicense the processes at reasonable fees to individuals or companies seeking those technologies. After three years, if the Indian

3 For an exposition of the philosophical and theoretical underpinnings of the TRIPS agreement, and pressure from developed countries, see Correa [7].

608 G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

patent holders appeared to demand unreasonable licensing fees, the Indian Patent Office could stipulate what it viewed as fair feesand require the patent holders to grant the licenses.

After the Patent Act's passage in 1970, the Indian pharmaceutical industry grew dramatically, appearing to challenge theargument that product-patent regimes constitute necessary conditions for innovation, growth and social returns. Fig. 3 presentsthe growth in the number of companies over the late 20th century and under the process-patent regime. The number of Indianpharmaceutical companies burgeoned from 2257 in 1970 to over 23,000 in 2004–2005 [8]. Similar growth occurred in domestic-market share and exports of pharmaceutical drugs developed and made in India, and in bulk drugs (used in pharmaceutical drugsand compounds) made in India for domestic markets and exports.

1994 proved awatershed year for the Indian pharmaceutical industry as the Indian government signed agreements for plannedchanges to the product-patent regime beginning January 2005. The new product-patent regime in Indian pharmaceuticals wouldprovide for twenty-year patents; define products as ‘new chemical entities’; and, continue to provide for compulsory licensing.The ten-year lag between the signing of TRIPS and the change-over to product patents in pharmaceuticals and agro-chemicalscomplied with the maximum time allowed by the agreement. Some Western companies complained that India's product-patentregime would fail to guarantee data secrecy.

In 1999, India passed patent-reform amendments to the Patent Act of 1970. The 1999 amendments, retroactive to 1995,provided for a mailbox system of patent applications and a system of Exclusive Marketing Rights (EMR) for products developedafter January 1, 1995. Patent mailbox systems allow companies to file patents for possible approval at future dates. India's mailboxsystem conformed to global standards. It provided pharmaceutical and agro-chemical companies with the ability to file, ahead oftime, for their patents' approval upon implementation of the 2005 product-patent regime. The EMR addressed the needs ofpharmaceutical companies that had developed products after January 1, 1995. It required that the filing company had obtainedpatents for their product from Patent Conventionmember countries, and had obtained rights tomarket the products in the patent-granting country. The EMR granted these companies the exclusive rights to market the products in India for either five years oruntil the Indian patent office granted or rejected the companies' patents, whichever came earlier. Consideration of companies'Indian product-patent applications would not begin until after December 31, 2004 [9].

In March 2006, India granted the first product patent under the new regime for the Roche drug Pegasys. Roche filed for thepatent through a 1997 mailbox application, introduced the drug to the Indian market in 2003, and the Indian patent for the drugran through May 15, 2007 [10]. The next section reviews theories on the relative benefits of process vs. product-patent regimes.

Regulatory &Institutional

Environments

Innovation

Social Returns &Welfare

Product-PatentRegimes

Process-PatentRegimes

+ -

++

+

-

Fig. 1. An institutional framework for patent innovation.

609G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

3. Theoretical rationales for social returns of patents

The concept of patents originated over three centuries ago from John Locke's economic philosophy. In his Second Treatise ofGovernment [11], Locke argued that just as people own their bodies, they also own the ‘fruits of their labor.’ Following Locke,several authors [12,13] attempted to define ‘the fruits of labor’. If fruits of labor referred to the technologies developed to produceproducts, it covered process patents; if the phrase referred to the products that resulted from the technologies, it covered productpatents. Usually, countries have recognized and enforced patents when they coincided with economic welfare. Today, the USAserves as a champion of IPR; yet, at its founding in 1776, the USA became the first country in history to establish the theft of foreignintellectual property as an official government policy [14–16].

Researchers have long debated the effectiveness of patents vis-à-vis othermechanisms, such as secrecy, to protect technologiesand to promote innovation. Substantial evidence indicates that patents fail to protect technology. Mansfield, Shwartz andWagner[17] questioned the effectiveness of patent protection when they found that 60% of patented products are imitated within fouryears. In addition, Mansfield [18] and Co, List and Qui [19] found that industries where imitation occurs relatively easily also valuepatentsmost highly; but, even in these industries, themajority of patentable products are not patented. Thus, while patents should

2257

5156

16000

20053

0

5000

10000

15000

20000

25000

1970 1980 1990 2000

Number of Pharmaceutical Companies

Fig. 3. Growth of India's pharmaceutical industry, 1970–2000.Source: Pharmaceutical and Drug Manufacturers.

1856• First Indian patent laws

1859• First major revisions of Indian patent laws

1872• Indian Patents & Designs Protection Act

1888• Indian Inventions & Designs Act

1911• Indian Patents & Designs Act

1912-1969

• Myriad minor amendments to Indian patent laws

1970

• Indian Patents Act• India Signatory to 1883 Paris Convention & Patent Co-operation Agreement

1994

• India Signatory to General Agreement on Tariffs & Trade (GATT)• India Signatory to Trade Related Aspects of Intellectual Property Rights (TRIPS) Agreement

1999

• Indian Mailbox amendment (retroactive to 1995)• Indian Exclusive Marketing Rights amendment (retroactive to 1995)

2005• India fulfils TRIPS requirements

Fig. 2. Timeline of regulatory and institutional changes in India's pharmaceutical industry.

610 G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

protect technology, research shows that patents often appear unsuccessful in providing protection, and companies often perceivepatents as providing inadequate protection for innovation.

Mansfield [18] also found that strengthening patents had, at best, mixed effects on foreign direct investment (FDI). In asubsequent study, Lee andMansfield [20] established that strengthening patents in developing economies had the desired effect ofincreasing inward FDI. Pfister and Defains [21] concluded that French companies tended to reduce investment in countries thatstrengthen their patent protection and have high GDP and low research intensity. Though they appear to contradict one another,Lee and Mansfield's [20] and Pfister and Defains' [21] studies form logical complements. High GDP countries generally havedeveloped, not developing economies. While no conflict exists between the two studies' findings, both questioned if enhancedproduct-patent protection increases FDI.

Qian's [22] findings appear more troublesome. In a study that covered 26 countries that established their latest pharmaceuticalpatent laws between 1978 and 2002, he found that pharmaceutical patents alone do not stimulate innovation; however, patentprotection combined with higher levels of economic development, education and economic freedom does. Qian [22] alsodiscovered optimal levels of patent protection; strengthening protection beyond the optimal levels reduces domestic innovativeactivities. Thumm [23] (p. 536) summarized the findings on effective patenting policy: “The provision of intellectual propertyrights is a matter of finding the right balance between an innovation spurring effect and a certain limitation to the diffusion oftechnological knowledge. For practical policy making it is hard to find the adequate level of protection in order to arrive atmaximum wealth creation.” Jensen and Webster [24] arrived at a similar conclusion.

Even with strong patent-protection regimes, secrecy plays an important role in IPR protection. Anton and Yao [25] found thatthe more important an innovation, the more likely that the developing companies emphasize secrecy over patenting as the majorform of IPR protection. Cohen, Nelson andWalsh [26] concluded that companies viewed secrecy asmore important than patents inprotecting IPR. Thus, the strengthening of patent-protection laws by less-developed countries' governments does seem to result inincreased FDI. However, the more crucial a company regards a technology, the less likely that company will seek to protect itstechnology through patents.

Other researchers have studied the social returns of patents within the scope of industry characteristics. For example, Co, Listand Qui [19] studied strengthening patent protection in industries where copying technology occurs relatively easy, such as inpharmaceuticals and chemicals. They concluded that stronger patent protection had a greater effect on inward FDI in industriessuch as pharmaceuticals, than in industries where imitation is more difficult. Lanjouw [27] focused on enhancing patent protectionexclusively within the pharmaceutical industry. She argued that patent-protection policies should differ based upon therapeuticmarkets. She also contended that insufficient research existed on diseases endemic to poor and middle-income countries; hence,pharmaceutical companies should receive world-wide patents for these drugs to promote research in the areas. She argued againstpatents that limit access to drugs for diseases that strike both prosperous and poor or middle-income countries. Lanjouw [26]concluded that in the latter case, pharmaceutical companies should receive patent protection in thewealthier countries; but, drugsshould sell free of patent protection when their projected sales in poorer countries constitute less than two percent of total worldsales.

Less-developed countries, such as India, have argued against strong product-patent protection in the pharmaceutical industryfor political reasons [6]. Policy makers and theorists have contended that these countries suffer inequitably from the incomeredistributions and deadweight losses arising from price increases and strengthened product patents. The local losses in a lessdeveloped country frommonopoly prices almost always exceed the local gains from an enhanced supply of general-purpose drugs[28]. Assuming straight-line demand functions and constant marginal costs, Deardorff [28] showed that the short-run consumers'surplus gain from an additional monopolized drug approximates one-third the loss of consumers' surplus from monopolisticpricing of a comparable drug otherwise available at competitive prices. To keep less developed countries' consumers equally well-off, the number of new drugs developed and sold through strengthened patent laws at home would have to be three times thenumber of drugs available without strengthening—a highly unlikely outcome. Payments for imported drugs and remittances ondrugs produced locally by multinational companies also appropriate hard currencies that these countries need for economicdevelopment [29]. Representatives of less-developed countries have additionally accused the USA of hypocrisy: besides itsfounding history, on at least six occasions during the 1950s and 1960s, the country ordered compulsory licensing of more than 100drug patents at home under antitrust case settlements [25].

To summarize, empirical research reveals that industrialized countries' policy makers and companies often champion patentsto protect technology and innovation, despite criticisms from less-developed countries. On balance, little evidence indicates thatpatents successfully protect technology, promote innovation or provide substantial social returns, especially to low-incomecountries. Our review shows that the more important a technological innovation, the less likely that a company will patent it, andthe more likely that the company will depend on secrecy. Consequently, if countries do not adequately protect secrecy, as somemultinational companies have voiced on India, regulatory changes may not lead to increased patents or innovation. The nextsection describes the structure of India's pharmaceutical industry.

4. Structure, strategy and innovation in India's pharmaceutical industry

India has the fourth-largest pharmaceutical industry in the world by volume but only the thirteenth largest in sales. Valued atUS$5.3 billion in 2005, the Indian pharmaceutical industry represents slightly less than 1% of theworldmarket. Yet, as Fig. 4 shows,India's pharmaceutical industry has enjoyed enormous growth. In 2006, Indian pharmaceutical companies supplied 95% of its totaldomestic pharmaceutical market, whereas in 1970 they supplied only about 20%. Simultaneously, in 2006, Indian companies

611G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

supplied 75% of themedicinal market. Since 1970, it has transformed itself from primarily a producer of bulk pharmaceuticals usedfor formulations, to an industry comprised of about 45% bulk pharmaceuticals and 55% formulations [4].

India's pharmaceutical industry has also developed into a global player. Over the past decade the Indian pharmaceuticalindustry has enjoyed a 22.7% compounded annual growth rate in exports. In 2005, exports approximated US$3.7 billion andrepresented about 61% of total industry production. In 2007, Ranbaxy, India's largest pharmaceutical company, exported about80% of its production, and though relatively unknown among US consumers, had the USA as is its largest market [30].

Yet, the Indian pharmaceutical industry remains fragmentedwith over 23,000 companies [31]. Despite industry fragmentation,several large Indian pharmaceutical companies compete globally. Concentration occurs at the top-end of the industry: the tenlargest Indian pharmaceutical companies control 36% of the market. The industry provides direct employment for about 5,000,000Indians and indirect employment for 24,000,000. Fig. 4 summarizes the industry's structural characteristics.

Fig. 5 highlights the Indian pharmaceutical industry's strategic cost-based opportunities for innovation and drug production.Analysts expect Indian companies to capture 30% market share of the drugs losing their US and European patent protection [32].Pharmaceutical industries that emerge from less-developed countries' process-patent regimes often possess advantages ingenerics. India too has excelled at the production of bulk, generic chemicals. In this sector, India's production costs approximate50% of the USA's and the European Union's (EU's) production costs [32]. Consequently, since 2005, the Indian industry hasresponded to potential losses in sales from the TRIPS agreement by focusing on exports of generic drugs to the US and Europeanmarkets and acquisition of generic capacity in those markets.

Fig. 5 also reveals that the USA's and other industrialized countries' advantages in high value-added products and innovationseem to have withered in the pharmaceutical industry. India has excelled at the higher value-added medicinal preparationsmarket. In 1970, India imported virtually all medicinal preparations; today India's pharmaceutical companies supply over 75% ofits domestic market [32]. Some estimates have placed the figure at over 85%. These results contradict researchers and policymakers who argue that process-patent regimes discourage the development of significant research functions.

In part, the Indian pharmaceutical industry's strategic success has stemmed from reverse engineering. Indian researchers havenot concentrated on developing new drugs, but rather on new processes to produce already established drugs. Reverseengineering has proven more cost effective than developing new drugs from scratch. Though they conducted their research on a

Fig. 5. Cost-based opportunities for India's pharmaceutical industry vs. the USA's and EU's.

Fig. 4. Structural characteristics of India's pharmaceutical industry, 2007.Sources: Intec.net; Third World Network, Organization of Pharmaceutical Producers of India.

612 G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

diverse group of industries, Mansfield, Shwartz and Wagner [17] found that in 85.7% of cases, companies that reverse engineeredinnovations also developed their products more cheaply than the original innovators.

Relative costs for R&D provide important factors for India's success in higher value-added innovation. For basic research, thehighest value-added innovation activity, India's costs fall betweenone-third and one-seventh of theUSA's or EU's. For clinical researchtrials, India's cost advantages loom even higher and approximate between one-fifth and one-tenth of the USA's or EU's [32,33].Consequently, even though the Indian pharmaceutical industry spends only 1.6% [32] of its sales on R&D, as compared to 15% in theWest, the Indian companies enjoy significant cost advantages and have engaged in more innovation for their investments.

Human resources also contribute to the Indian pharmaceutical industry's advantages in R&D. These resources include India'stechnically-skilled workforce, fluent in English, because of its English-language education system.4 India's workforce contributesto an environment conducive to joint-venture projects with, and contract research for, multinational companies. Its researchersalso have greater access to English-language research reports and papers, thereby contributing to free-flowing communicationthat promotes innovation. Simultaneously, foreign executives and researchers enjoy higher comfort levels in India's Englishlanguage environments. Combined with the Indian market's high-growth potential, their executives' comfort levels promote awillingness among foreign companies to invest in Indian research. Favorable manufacturing environments enhance favorableresearch environments in India. India has 75 US Food and Drug Administration (FDA) approved facilities [33], more than any othercountry outside the USA. India's manufacturing environments contribute especially to its edge in clinical-trial research: costefficiencies derive from both production and research, and companies can manufacture drugs for clinical trials on site.

India's competitive advantages in R&D compensate in part for the Indian pharmaceutical industry's fragmentation. R&Dinvestments in this Indian industry approximate 1.6%, but most small and micro-sized pharmaceutical companies engage in nobasic research. Many of the small pharmaceutical companies focus instead on producing India's traditional folk and Ayurvedicmedications; others produce generics and bulk drugs. Table 1 shows that the largest Indian companies' R&D as a percent of salesremains about one third that of Western companies. Yet, companies in India spend about three times less for comparable researchactivities than companies spend in the West. Due to the Indian pharmaceutical companies' smaller sizes, they also spendsubstantially less on total R&D than their Western counterparts.

Fig. 6 highlights R&D in the Indian pharmaceutical industry. R&D investments increased substantially from 2000 to 2006, withthe greatest growth from 2003 to 2004, the year before the new product-patent regime took effect. Indian companies probablyraced to accumulate patents through backward engineering before they could no longer file process patents. Fig. 6 also shows thatthe rate of growth in R&D spending fell precipitously in 2005–2006 to below levels in 2001. This drop occurred despite increasingnumbers of joint ventures on innovation between Indian and Western pharmaceutical companies. Data on the 25 largest Indianpharmaceutical companies add to that concern. In 2008–2009, the largest Indian pharmaceutical companies increased their R&Dspending by only 8.4% in Indian Rupees—whichwhen converted to dollars, reflects a decline of 8% [34]. The next section examinespatents as measures of success in innovation and R&D.

5. Measuring innovation in the Indian pharmaceutical industry

Several methods exist to measure success in innovation generally and R&D specifically. Yet, a research survey [35] found thatno more than 10% of the companies used any single measure of success. Patents generally provide the most widely acceptedmeasure of innovation [36], through:

• Providing objective measures• Drawing on both self and external reports• Clearly indicating research value• Correlating strongly with measures such as research publications and a United Nations Organization's national ratio oftechnology products to exports.

Yet, patents differ greatly in the importance of the innovations they patent, as well as in their applications to product families orsingle products. Companies and researchers also have difficulties determining a priori if patented technologies represent majorinnovative breakthroughs [34,35]. For instance, almost no one predicted that shipping containers, simple metal boxes, would

4 India's 2001 Census indicated that the number of people speaking English ranged from 9% to 33% of the population or from 100 million to 350 million people,making it the second-largest English speaking country in the world after the USA.

Table 1R&D as percent of sales for India's pharmaceutical industry, 2004–2005.Source: Authors' calculations.

Companies in pharmaceutical industry Percent

Indian pharmaceutical industry 1.6Western pharmaceutical industry 15.0India's top 10 pharmaceutical companies 5.7Dr. Reddy's 14.7R&D growth of India's top 5 pharmaceutical companies 47.0

613G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

revolutionize their industry. Some authors have suggested that shipping containers, along with the container ships and ports thatthey spawned, facilitated globalization through their enhanced logistical efficiencies [37].

To correct some flaws in raw-patent data, the Organization for Economic Cooperation and Development (OECD) developed ameasure of triadic patents, or patents filed in each of the triad economies, the EU, Japan and the USA. By focusing solely on patentsfiled in all three markets, the OECD sought to ensure the use of patents of the greatest and most comparable values. The OECDreasoned that companies or individuals will only patent separately, and in all three markets, technologies that they view as mostvaluable — despite the Paris Convention's stipulation that all Convention members must recognize patents filed in any member'scountry. The OECD gathered national, not industrial data; yet, the data demonstrate increased innovation across countries. Fig. 7presents the OECD's triadic data. In 2001, under its old process-patent regime, India enjoyed a 21.9% rate of growth in triadicpatents, the fourth highest in the world, and substantially above the average growth rate of 12.2% for all countries.

Fig. 6. R&D (millions of dollars) in India's pharmaceutical industry, 2000–2006.Source: The Associated Chambers of Commerce and Industry of India.

Fig. 7. Innovation by countries in triadic patent families, 2001.Source: OECD Science, Technology and Industry Scoreboard.

614 G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

Table 2 shows available patent data on the Indian pharmaceutical industry prior to 2005 and in the process-patent regime.Significant growth in patents occurred from 2001 to 2004, despite a substantial decline in patents from the Government's primaryresearch labs, the Council for Science and Industrial Research (CSIR), from 2003 to 2004. Ranbaxy's patent productivity over thesame two-year period helped to offset CSIR's decline. Despite ranking among India's five largest pharmaceutical companies, Ciplahad the lowest percentage of sales invested in R&D, indicating its reliance on process patents through backward engineering.Missing data in Table 2 reflects poor data availability on Indian patents prior to 2005.

Table 3 presents a partial list of India's pharmaceutical patents by company from the start of the product-patent regime in2005–2008. The new patent regime aimed to foster innovation and to increase patents; yet, the number of patents issued in eachyear fluctuated wildly. Patents for 2006 fell to 315 from 468 in 2005, skyrocketed up to 750 in 2007, but then plummeted backdown to 484 in 2008.

As shown in Table 3, we conducted a signs test for annual changes in patents by major Indian pharmaceutical companies, andthen on the difference in patents between 2005 and 2008. The signs test seeks to determine significant movement in values overtime by testing for significantly more or fewer numbers of minus or plus signs than expected by the null hypothesis. In a signs test,the null hypothesis proposes that half our sample's population will generate pluses and half minuses. In this instance, if theinnovation rate as measured by patents were to increase, there should be a significant increase in the number of minuses in eachcolumn. However, only from 2006 to 2007 do the comparisons produce a significant degree of change in the desired direction.Additionally, the change in patents from 2007 to 2008 reverses the preceding year's increase with a significant reduction in thenumber of patents. Finally, a test for change in the number of patents for the entire 2005–2008 period showed no significantchange in the number of patents in either direction. Consequently, patent data from India's pharmaceutical industry do notsupport the proposition that product-patent regimes promote innovation. Because of graphic missing data prior to 2005, wecannot compare the number of all India patents for 2004 (Table 2) with the data in Table 3.

Fig. 8 presents seven companies that we can compare across the process and product-patent regimes. The number of patentsper year for five of the seven companies, as well as the overall total number of patents per year across all companies, dropped. Aftera significant drop between 2004 and 2005, Ranbaxy's patents held constant in 2006, spiked in 2007, only to plummet in 2008.India's second-largest pharmaceutical company, Cipla, dropped in number of patents in each of the five years. Matrix Labs andWockhardt bucked this downward trend by following differing routes to success.Wockhardtmoved strongly into bio-genetics andproduct patents, through both acquisition and research, increasing its patents every year. Conversely, Matrix emphasized genericsand process research, to plateau in 2008, at a higher level than in 2004. These results indicate that product-patent regimesmay notpromote greater rates of innovation, and process-patent regimes may not stunt innovation.

Process-patent and product-patent regimes differ in the kinds of innovation they emphasize. For example, process-patentregimes tend to encourage more investment and effort into reverse-engineering known products; conversely, product-patentregimes should encourage investments and efforts into creating new products. As we discussed, some patents reflect greaterinnovation than others. Citation analysis provides one indicator of value by identifying heavily-cited, and therefore, valuablepatents. Yet, in this study, India had fewer than four years under the new product-patent regime, and so we could not engage inlegitimate citation comparisons. Generally, research has found that process innovations generate greater returns on investmentfor companies than product innovations [36].

Fig. 9 corroborates that in the new regime, the Indian pharmaceutical industry appears to be engaging in less processinnovation, although some individual companies are not. Specifically Fig. 9 shows a downward trend in the percent of process-oriented patents filed by the largest Indian pharmaceutical companies, and especially by the two companies at the forefront ofIndian pharmaceutical research, Ranbaxy and Wockhardt. Sun Pharma's and Cadila's process patents, as well as the overall totalpercent, have also fallen. The Indian pharmaceutical industry's innovation appears in transition from a process to a productorientation. The next section discusses the implications of our study for managerial practice, policy and research.

Table 2Pharmaceutical patents in India's process-patent regime, 2001–2004.Source: Indian Patent Office; Independent research.

Organization Number of patents

2001 2002 2003 2004

All Pharma 295 784Ranbaxy 66 121CSIR 124 69Cipla⁎ 32Jubilant Organosys 16Vaman Technologis 12Matrix Labs 12Hetero 10Wockhardt⁎⁎ 10

The italics merely differentiate between statistics for the industry as a whole and statistics for individual companies.

⁎ Cipla invested b5% of sales on R&D, the lowest among India's top 5.⁎⁎ Especially significant because of its Biotech operations.

615G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

6. Implications

The relatively short period of study, and its exploratory nature, do not allow for the identification of long-term trends.Additionally, our exploratory study used data varying over different periods e.g. data on number of pharmaceutical companies till2000, and total number of patents till 2007–2008. This can affect our overall analysis and interpretation of results. However, thepreliminary results do indicate that significant cost advantagesmay have helped Indian companies to offset smaller investments inR&D to compete in a global environment. Our study shows that Indian pharmaceutical companies have been using their costadvantages to refocus their research from process to product innovation. This strategy has not had sufficient time to prove its

Fig. 8. Total patents for seven Indian pharmaceutical companies, 2004–2008.Source: India Big Patent Database; Independent research.

Table 3Pharmaceutical patents in India's product-patent regime, 2005–2008.Source: India Big Patent Database; Independent research.

Signs test

Company 2005 2006 2007 2008 05–06 06–07 07–08 05–08

Alembic 9 12 27 8 − − + +Aurobindo 6 20 54 28 − − + −Biocon 2 11 21 4 − − + −Cadila 41 32 61 35 + − + +Cipla 44 22 36 20 + − + +Dr. Reddy's 107 8 100 51 + − + +Hetero Drugs 15 6 4 3 + + + −Jubilant Organosys 2 9 16 16 − − / −Lupin 18 27 35 16 − − + +Matrix Labs 2 15 23 21 − − + −Natco 29 18 34 14 + − + +Nicholas Piramal 4 5 8 5 − − + −Orchid 7 2 31 10 + − + −Ranbaxy 57 67 221 133 − − + −Reliance Life Sci. 10 12 12 13 − / − −Sun Pharma 84 14 25 6 + − + +Themis 3 6 2 3 − + − /Torrent 20 16 7 11 + + − +Wockhardt 8 13 33 87 − − − −Total 468 315 750 484 + − + −

Number of +'s 9 3 15 8Z score −.447 −3.131 2.236 −.894Probability .326 .000 .013 .187Number of –'s 11 16 4 11Z score .447 2.683 −2.683 .447Probability .326 .004 .004 .326

616 G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

success. Our study also shows that the growth of innovation has declined in the product-patent regime that governs India'spharmaceutical industry. Several implications exist for managers, policymakers and researchers.

6.1. For managerial practice and policy

India's pharmaceutical companies are leveraging their cost advantage to become major developers of intellectual property forthemselves and for Western pharmaceuticals through contract research [32,33]. As contract researchers, the Indianpharmaceutical companies do not hold title to the products they develop, but retain rights to use technologies in their domesticmarket, one of the world's fastest growing. As contract researchers, the Indian pharmaceutical companies also obtain foreign fundsto build their own product-research cultures and competitive capabilities. The sharing of risk in conducting research providesadditional welfare benefits that Indian companies share with their Western counterparts through contract R&D agreements.

Contract research provides the Indian pharmaceutical industrywith substantial immediate returns, butmay also increase socialcosts in themedium to long terms. Though Indian pharmaceutical companies retain the rights to produce and to sell drugs in India,they surrender these rights in foreign markets. Consequently, Indian companies may expend some of their best research talent ondrugs that generate limited financial returns. Indian patent law defines patentable, new pharmaceutical drugs as entirely newproducts and does not recognize modifications and enhancements of existing drugs as patentable. Therefore, foreign companies,from jurisdictions that define patents more broadly, protect their intellectual property in India from stronger positions than doIndian companies within their home markets. As India's domestic markets mature, Indian pharmaceutical companies also havelimited abilities to protect and to extend domestic IPR, or to develop the drugs they need to promote future competitiveness abroad.

Our study, though preliminary, shows that the rate of growth in innovation, as measured by investments in R&D, has fallen inIndia's product-patent regime. Consequently, solutions that lead to greater global social returns through efficient production andeffective co-ordination of global resources, may not necessarily lead to greater social welfare nationally or to greater financialreturns to Indian pharmaceutical companies. India and other low-incomemarkets need policy tools to developmore Pareto optimalsolutions.5 With Pareto movement, the gains for one party, such as multinationals who can source more efficiently incorporatingIndia's cost advantages, would not lead to the losses of other parties. For instance, Indian society would not relinquish the socialreturns that can accrue from Indian companies inventing andmanufacturing drugs as independent entities rather than as contractmanufacturers. Pareto optimal policies would alsomitigate the costs to small andmedium-sized Indian pharmaceutical companiesthat cannot efficiently set up as contract manufacturers of generic drugs for multinational companies. This study's findings havesignificance for other countries such as China, that also balance global pressures on IPR with social-welfare concerns.

Indian researchers are gaining experience in product discovery through both contract research and proprietary research; and,India's pharmaceutical companies, especially its largest companies, are growing rapidly through industry consolidation, domesticand international mergers and acquisitions, and organic growth [4,33]. Yet, in a global economy, political failures to improveinstitutional infrastructure, and to remove regulatory obstacles on perceived data security, appear as large threats to the Indianpharmaceutical industry's continued growth and success.

5 Named after Vilfredo Pareto, Pareto optimality constitutes one measure of efficiency. Not all efficient outcomes are Pareto optimal. An outcome is Paretooptimal if no other outcome exists that makes every player at least as well off and at least one player strictly better off. That is, a Pareto optimal outcome cannotbe improved upon without hurting at least one player. Often, a Nash equilibrium is not Pareto optimal implying that the players' payoffs can all be increased.

0 10 20 30 40 50 60 70 80 90

Ranbaxy

Cipla

Lupin

Cadila

Sun Pharma

Wockhardt

Dr. Reddy's

N Piramal

Aurobindo

Totals

Percent of Process Patents

Co

mp

any

2007 2006 2005

Fig. 9. Percent of process patents in India's pharmaceutical industry, 2005–2007.Source: India Big Patents Database; Independent research.

617G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

6.2. For future research

This study indicates that product-patent regimes do not necessarily generate greater rates of innovation than process-patentregimes, and may reduce innovation. Our preliminary findings, together with those that we reviewed earlier, suggest that weshould question our beliefs that strong product-patent regimes promote innovation. Indeed, substantial research needs toestablish the true value of product-patent regimes to innovation, especially in low-income markets where distribution of benefitsassumes more importance.

In addition, as China has proven, alternative reasons besides effective IPR protection exist to invest in a market, including rapidgrowth [38,39]. Given the size and growth of the Indian pharmaceutical market and the rate of India's economic growth, FDI inIndia's pharmaceutical industry may have occurred regardless of patent regime. Further research could enquire into the importantcharacteristics that prompt domestic and foreign investment, including innovativeness, rapid growth and patent regimes. Futureresearch could also look at the influence of data security on investment in pharmaceuticals.

Finally, research could explore whether product-patent regimes can detrimentally affect technology development in less-developed countries as Indian Justice Rajagopala Ayyangar's 1959 report had proposed [1]. Future large-scale research can build onour exploratory findings through techniques such as logistic-regression modeling, instead of sign tests, to observe trends ofinnovations in different regimes. Less-developed countries have generally begun the move to near industrialization or to fullindustrialization by engaging in theft of intellectual property. Future research could shed light on the questions raised by less-developed countries' policy makers on the social-welfare effects of product-patent regimes on India, and to the other emergingand low-income markets that depend on India's pharmaceutical industry.

Appendix A. Sample and methods

Sample

Our sample consisted of 19 of the largest Indian pharmaceutical companies from 2005 to 2008. We chose the companies in oursample through reviewing media reports, industry analyses, interviews with managers and information from industry websites.For patents in the product-patent regime beginning 2005 to 2008, we used the India Big Patents database, jointly developed by XBLabs and the ColumbiaMailman School of Public Health. The database only tracked Indian patents from January 1, 2005. Therefore,we also tracked available patents for the companies in our sample in the process-patent regime, from 2000 to 2004, throughcompany-reported filings at the Indian Patent Office and content analysis of media reports. In both periods, we screened data bycomparing patent numbers to eliminate duplicate listings of patents, and to classify the patent as product or process. Very highinter-coder reliability indicated the relative objectivity of our classifications. For 2008, we did not classify patents as either processor product as full patent descriptions did not yet exist at the time of data collection.

Methods

We chose patents to measure innovation as a) patents provide objective measures of innovation, b) patents are measurable,and c) researchers and analysts generally view patents as the best overall measure of innovation [36].

The small number of years of data constrained our ability to establish a statistical trend. In such situations, and whenidentifiable and differentiable entities such as companies or individuals generate periodic data, a signs test provides an excellentstatistical tool [40]. For the signs test, we collected numeric totals for each company and for each year of data in the database. Foreach year, we subtracted a company's total patents from the immediately preceding year's total for that company (2005–2006,2006–2007, 2007–2008), to generate one of the three possible results for each year: a minus sign (for a Y-o-Y increase in patents),a plus sign (for a Y-o-Y decrease), or if no change, a slash. Because of apparent volatility in year-to-year results, we subtracted thefinal year from the first year (2005–2008) to determine measurable change between the first and last year of data.

References

[1] Ayyangar, Justice N. Rajagopala (1959), Report on the Revision of the Patent Law, New Delhi, India.[2] A.D. Damodaran, Indian patent law in the post-TRIPS decade: S&T policy appraisal, J Intellect. Property Rights 13 (September 2008) 414–423.[3] E. Check, The Treasure of Mumbai', Wired, Issue 14.12, December 2006.[4] W. Greene, The emergence of India's pharmaceutical industry and implications for the U. S. generic drug market, Office of Economics Working Paper,

No. 2007-05-A, United States International Trade Commission, May 2007.[5] N. Zacharias, S. Ferias, Patents and the Indian pharmaceutical industry, Business Briefing Pharmagenerics 2003, Business Briefings, Ltd., London, 2003.[6] F.M. Scherer, Pricing, profits and technological progress in the pharmaceutical industry, J. Econ. Perspect. 14 (3) (1993) 97–115.[7] C.M. Correa, Intellectual property rights, the WTO and developing countries, The TRIPS Agreement and Policy Options, Zed Books and Third World Network,

London, New York, 2000.[8] Pharmaceutical & Drug Manufacturers, India pharmaceutical industry: an overview, http://www.pharmaceutical-drug-manufacturers.com/pharmaceutical-

industry. 2007. downloaded April 2, 2007.[9] Indian Patents Office, Exclusive Marketing Rights, http://www.patentoffice.nic.in/ipr/patent/emr.htm.

[10] P.T.J. Datta, Roche Gets Product Patent on Hepatitis C, The Hindu Online, 2006 posted March 3.[11] J. Locke, Second Treatise of Government, Hackett Publishing Company, Indianapolis, IN, 1764/1980.[12] S.A. Bell, J.F. Henry, L.R. Wray, A chartalist critique of John Locke's theory of property, accumulation and money: or is it moral to trade your nuts for gold, Rev.

Soc. Econ. 62 (1) (2004) 51–65.[13] L. Zemer, The making of a new copyright Lockean, Harv. J. Law Public Policy 29 (3) (2006) 891–947.

618 G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619

[14] G.T. Haley, Intellectual property rights and foreign direct investment in emerging markets, Mark. Intell. Plann. 18 (5) (2000) 273–280.[15] G.T. Haley, The extent of the government's control of China's economy, and implications for the United States, Testimony before the United States–China

Economic and Security Review Committee of the US Congress, May 25, 2007.[16] A. Hamilton, Writings, Literary Classics of the United States, Inc., New York, 2001.[17] E. Mansfield, M. Shwartz, S. Wagner, Imitation costs and patents: an empirical study, Econ. J. 91 (1981) 907–918.[18] E. Mansfield, Patents and innovation: an empirical study, Manag. Sci. 78 (2) (1986) 181–186.[19] C.Y. Co, J.A. List, L.D. Qui, Intellectual property rights, environmental regulations, and foreign direct investment, Land Econ. 80 (2) (2004) 153–173.[20] J.Y. Lee, E. Mansfield, Intellectual property protection and U.S. foreign direct investment, Rev. Econ. Stat. 78 (2) (1996) 181–186.[21] E. Pfister, B. Defains, Patent protection, strategic FDI and location choices: empirical evidence from French subsidiaries' location choices in emerging

economies, Int. J. Econ. Bus. 12 (3) (2005) 329–346.[22] Yi Qian, Do national patent laws stimulate domestic innovation in a global patenting environment? A cross-country analysis of pharmaceutical patent

protection, 1978–2002, Rev. Econ. Stat. 89 (3) (2007) 436–453.[23] N. Thumm, Strategic patenting in biotechnology, Technol. Anal. Strateg. Manage. 16 (4) (2004) 529–538.[24] P.A. Jensen, E. Webster, Achieving the optimal power of property rights, Aust. Econ. Rev. 37 (4) (2004) 419–426.[25] J.J. Anton, D.A. Yao, Little patents and big secrets: managing intellectual property, RAND J. Econ. 35 (1) (2004) 1–22.[26] W.N. Cohen, R.R. Nelson, J.P. Walsh, Protecting their intellectual assets: appropriability conditions and why U. S. manufacturing firms patent (or not), NBER

Working Paper No. 7552, 2000.[27] J.O. Lanjouw, Intellectual property and the availability of pharmaceuticals in poor countries, NBER Innov. Policy Econ. 3 (1) (2003) 91–129.[28] A.V. Deardorff, Welfare effects of global patent protection, Economica 59 (233) (February 1992) 35–51.[29] F.M. Scherer, The economic effects of compulsory patent licensing, New York University, Graduate School of Business Administration, Monograph Series in

Finance and Economics, , 1977.[30] Authors' interviews[31] S.V. Ramani, Who is interested in biotech? R&D strategies, knowledge base and market sales of Indian biopharmaceutical firms, Res. Policy 31 (2002)

381–398.[32] KPMG International, The Indian Pharmaceutical Industry: Collaboration for Growth, Zurich, 2006.[33] Federation of Indian Chambers of Commerce and Industry (FICCI), Competitiveness of the Indian Pharmaceutical Industry in theNewProduct Patent Regime—

FICCI Report for the National Manufacturing Competitiveness Council, March 2005, 16 pp.[34] S. Pingle, R & D Spend: Gamble on Blockbuster Drugs Hots Up, Pharmabiz.com, March 26 2009.[35] Haley, G. T. (1989) New Product Decision Making in Major American Firms: A Study of the Chemical, Computer and Metals Industries, University of Texas at

Austin, Doctoral Dissertation.[36] Economist Intelligence Unit, Innovation: Transforming the Way Business Creates, EIU Publications, London, 2007.[37] M. Levinson, The Box: How the Shipping Container Made the World Smaller and the World Economy Bigger, Princeton University Press, Princeton, NJ, 2006.[38] G.T. Haley, U.C.V. Haley, C.T. Tan, The Chinese Tao of Business: The Logic of Successful Business Strategy, John Wiley & Sons, New York & Singapore, 2004.[39] Z. Ma, Y. Lee, C.-F.P. Chan, Booming or emerging? China's technological capability and international collaboration in patent activities, Technol. Forecast. Soc.

Chang. 76 (2009) 787–796.[40] H. Moskowitz, G.P. Wright, Statistics for Management and Economics, Charles E. Merrill Publishing Company, Columbus, OH, 1985.

George T. Haley (PhD, University of Texas at Austin) is Professor of Marketing, University of New Haven and Founding Director, Center for International IndustryCompetitiveness. He has over 100 articles, presentations and books including The Chinese Tao of Business and New Asian Emperors. He has presented seminars tomanagers and policymakers on four continents, including for the National Intelligence Council, and the United States International Trade Commission and testifiedtwice before the US–China Economic and Security Review Commission. He serves on the Boards of Directors of listed companies, manufacturing organizations andgovernment agencies. In November 2009, the American Marketing Association's Marketing News named him a Rising Academic Star and “Person toWatch” on thebasis of research, teaching and broader impact.

Usha C. V. Haley (PhD, New York University) is Professor of International Business, School of Management, Massey University, in Auckland, New Zealand, andResearch Associate at the Economic Policy Institute in Washington, DC. She has more than 150 publications, presentations and books including MultinationalCorporations in Political Environments. She has testified before the US–China Economic and Security Review Commission and the Committee onWays andMeans,110th US Congress, and presented her research before the US International Trade Commission, the US Trade Representative, and the US Department of Commerce.She serves on several corporate and government boards. In 2003, she received a Lifetime Achievement Award from the Literati Club (UK) and a panel of academics,businesspersons and policy makers for contributions to understanding business in the Asia Pacific. The Economist has named her a "Thought Leader" on High-Growth Markets.

619G.T. Haley, U.C.V. Haley / Technological Forecasting & Social Change 79 (2012) 607–619