Chapter

1

The Indian Context and Enabling Environment
Mark A. Dutz and Carl Dahlman

This book focuses on how to foster increased innovative activities in India to meet the twin challenges of sustained growth and pro-poor development. India is an extreme “dual” economy.1 At one extreme, it is the world’s fourth-largest economy in purchasing power parity (PPP) terms, it is a nuclear and space power, and it is increasingly becoming a top global innovation player in certain key economic sectors––such as biotechnology, pharmaceuticals, automotive components, information technology (IT), software, and IT-enabled services (ITES). At the opposite extreme, India largely remains a subsistence economy. With an average per capita income of $720 in 2005, India is still a low-income and mainly rural, agrarian economy. About a quarter of its population lives below the national poverty line, with significant spatial variance across and within states.2 Roughly 70 percent of its population is rural, and 60 percent of the workforce is engaged in agriculture. Illiteracy rates are 46 percent for women and 25 percent for men.3 Given this dual economy, it is natural to ask what can be done both to strengthen the likelihood of sustained high growth rates and to address the unmet needs of the informal sector and the poor. To sustain growth and reduce poverty, India must leverage and improve its innovation potential. Innovation can be a critical driver of increased productivity and competitiveness and, ultimately, poverty alleviation.4 India’s recent acceleration in growth has been impressive. Over the 2004–06 period, real GDP has grown by over 8 percent a year. Growth has been driven by a jump in export-oriented, skill-intensive manufacturing (pharmaceuticals, petrochemicals, auto parts and assembly) and services (IT, business services, finance). These have been accompanied by a jump in innovative activities. Higher productivity and economic growth have raised living standards and reduced the number of poor people.5 However, tremendous dispersion in productivity levels remains, both within and across economic sectors. Most workers
23

1

Mark A. Dutz and Carl Dahlman

in the informal sector—which accounts for roughly 90 percent of the workforce— are underemployed in low-skill, low-productivity, low-income activities. India’s dual economic structure and wide dispersion in productivity levels call for a broader interpretation of innovation. Innovation is defined to include both “new to the world” creation and commercialization activities and “new to the market” diffusion and absorption activities––the first use of existing knowledge in new market contexts to help underperforming enterprises come closer to the global frontier of knowledge. Innovative activities include products, processes, and business and organizational models new to the local environment. India has even more to gain from economywide productivity increases from diffusion and absorption of existing knowledge than from creation and commercialization of new knowledge. The global technological frontier is moving quickly, simultaneously opening opportunities and posing threats. India needs to tap into this rapidly moving frontier while expanding its comparative advantages. India can and must do more to take advantage of its critical capabilities to enable creation of new global knowledge. Just as important, if not more so, India must develop policies, institutions, and capabilities to diffuse local and foreign knowledge more effectively throughout its economy. From an economic viewpoint, India stands to gain more from catching up to the world frontier than from pushing out the frontier. Thus, the challenge of innovation in India combines a drive to move the global technological frontier, an effort to increase the speed at which global innovations enter the country, and the most pressing need—to improve prevailing practices across the entire economy.

Structural Features of the Indian Economy
The use of knowledge for productive economic purposes varies as countries develop. Knowledge needs are related to a country’s economic structure, and this structure changes as a country develops. In developing countries, such as India, agriculture is the economic sector absorbing the most labor. As countries increase agricultural productivity through new technologies, the workers released by agriculture are absorbed by the industrial sector—particularly manufacturing. In addition, the service sectors expand. Initially, commerce and construction expand supported by low skills. But eventually, these and other service sectors, such as tourism and health services, develop and become more knowledge intensive.

Economic Dualism
India’s economic dualism is stark, with less than 3 percent of the workforce employed in the formal private sector and the bulk of the workforce in the informal sector. Formal versus informal employment is a very imperfect proxy for India’s dual economic structure.6 According to available data, the formal sector accounts for just 11 percent of a workforce of roughly 460 million; 89 percent of workers are
24

The Indian Context and Enabling Environment

1

Figure 1.1 India’s Dualistic Economic Structure manufacturing (14% of informal sector) informal sector (89%)

services (19% of informal sector) formal sector (11%) agriculture (67% of informal sector)

services (66% of formal sector)

agriculture (6% of formal sector) manufacturing (28% of formal sector)

Source: World Bank 2006f.

in the informal sector.7 In agriculture only 1 percent of employment is formal, and even in manufacturing the share is just 19 percent. By far, most formal employment (66 percent) is in the tertiary or services sector, in which government accounts for the majority. Roughly 50 percent of workers are self-employed. The bulk of selfemployment is in low-productivity subsistence agriculture and services. Figure 1.1 summarizes this dualistic structure of the Indian economy. India needs to absorb workers out of agriculture, into manufacturing and services. The bulk of the Indian workforce is engaged in agriculture. Although the share of workers in agriculture has been declining, the decline has not been as rapid as might have been expected relative to other developing countries. By 1999–2000, roughly 60 percent of the overall workforce remained in agriculture. Part of the problem has been that manufacturing employment has not increased much as a share of total employment. As a result, most people leaving agriculture have gone into construction and the service sector—especially trade, hotels, and restaurants; personal, business, and community services; and transport, storage, and communications. The rest of this book does not focus on agriculture because the innovation challenges facing the sector are covered in detail elsewhere.8 This volume also focuses largely on central government support programs for industrial innovation—not state programs, which focus more on setting up enterprises.

Productivity Dispersion
Productivity has increased in manufacturing and services, but at very different rates. The productivity of all sectors relative to agriculture (other than construction) has increased significantly over the past 20 years (table 1.1). Most notable have been the
25

1

Mark A. Dutz and Carl Dahlman

Table 1.1 Changes in Labor Productivity Relative to Agriculture, by Economic Sector Based on Principal Status of Workers, 1983–2000
Sector
Agriculture and allied activities Mining and quarrying Manufacturing Electricity, gas, and water Construction Trade, hotels, and restaurants Transport, storage, and communications Financial, insurance, and real estate businesses Personal, business, and community services Prime-age workforce (thousands)
Source: World Bank 2006f.

1983
100 615 243 912 367 312 376 1,673 221 171,029

1988
100 641 272 1,101 253 319 424 1,825 261 184,626

1995
100 628 293 1,186 294 311 411 2,211 231 259,820

2000
100 971 352 1,797 258 321 453 2,276 358 303,895

relative near doubling of worker productivity in electricity, gas, and water (mostly government monopolies); the almost two-thirds increase in personal, business, and community services (no doubt driven by the business services); the 45 percent increase in manufacturing; and the more than one-third increase in the highly productive finance, insurance, and real estate businesses. The average productivity of this last group is nearly 23 times that of agriculture. However, it accounts for only 1.3 percent of total employment, whereas agriculture accounts for 60 percent. It is discomfiting that the relative productivity of construction—the sector that usually absorbs agricultural labor—has actually fallen; and that of trade, hotels, and restaurants—another high-absorption sector—has hardly increased. In sum, there are widening gaps between the productivity of the enormous agricultural sector and that of more knowledge-intensive sectors. Strong growth in selected service sectors, complemented by recent growth in manufacturing, suggests strong productivity increases in some sectors—though overall productivity growth remains relatively low. In the period 1993–2004, India’s GDP grew by an average of 4.6 percent per year, increasing to an average of over 8 percent for 2004–06 (figure 1.2). Although India’s total factor productivity (TFP) growth increased from 0.2 percent a year in the 1960s and 1970s to 2.3 percent in 1993–2004, this still compares unfavorably with China’s annual rate of 4 percent. A breakdown across broad sectors reveals India’s weakness in manufacturing relative to services: since 1993, while TFP growth in services has been 3.9 percent a year in India (relative to 0.9 percent in China), TFP growth in manufacturing has been only 1.1 percent in India (relative to 6.2 percent in China).9 Although TFP growth in India has probably picked up with the increase in GDP growth since 2004 (the figures are not yet available), a key question is the extent to which recent growth is sustainable.
26

The Indian Context and Enabling Environment

1

Figure 1.2 Contributions to Real GDP Growth, by Industrial Category, 2001–06
13

10

change from previous year’s GDP (%)

7

4

1

2

5 2001 Q2 2001 Q4 2002 Q2 2002 Q4 2003 Q2 2003 Q4 2004 Q2 2004 Q4 2005 Q2 2005 Q4 2006 Q2 year and quarter agriculture manufacturing services real GDP

Source: Central Statistical Organization, Government of India.

Broadly Based Innovation and Productivity
Innovation is broadly defined to include both creation and commercialization of state-of-the-art knowledge as well as diffusion and absorption of existing knowledge. Available measures of formal innovation inputs—both creation (R&D spending) and absorption (technology acquisition)—are closely and jointly associated with innovation outputs (developing new product lines). Informal efforts to create and absorb knowledge are also associated with innovation outputs. In turn, innovation outputs are strongly associated with enterprise productivity. Important in this context, absorbing existing technology has a stronger association with productivity than does spending on R&D.

Definition and Indicators
Innovation is often defined as the invention and commercialization of new products. This book defines innovation more broadly, to include both “new to the world” creation and commercialization of knowledge activities and “new to the relevant market” diffusion and absorption of knowledge activities (box 1.1). Creation of new knowledge and its commercialization require research and experimentation as well as incentives, skills, and institutional support to bring the knowledge to market. While knowledge creation traditionally focuses on formal research
27

1
Box 1.1 Broadly Based Innovation Activities

Mark A. Dutz and Carl Dahlman

Creation and commercialization Shifting out the global frontier of knowledge

Product innovation

Process innovation

Diffusion and absorption Moving toward the global frontier of knowledge

Organizational innovation

Better and cheaper products in global, national, and local markets in response to consumer needs

Creation and commercialization Creation is inventive activity, often the result of formal research and development (R&D) conducted by scientists and engineers. The key institutions involved in formal knowledge creation are public R&D laboratories, universities, private R&D centers, and enterprises. But not all knowledge creation is the result of formal R&D. It is often market- or application-based, driven by an understanding by the innovator of what consumers want. Sometimes the invention comes from experiences with production or from informal trial and error. Sometimes it comes from serendipitous insight. Its origin raises a measurement problem because not all R&D results in inventions, and not all inventions come from formal R&D. Typically, two phases follow the transformation of a basic idea into an initial proposal format: proof of concept or initial prototyping, and pilot demonstration. For some products (such as software), a first or “alpha” prototype is developed at a business customer location that agrees to be an alpha site because of the perceived value in being involved in testing rather than waiting to buy the product on the market. During this phase, the technical merit and commercial feasibility of the new idea or technology are explored. It is often at this phase that early-stage technology development (ESTD) grants and incubators are most helpful. Although there is no consensus definition of ESTD, it can be broadly defined as the stage at which “the technology is reduced to industrial practice, when a production process is defined from which costs can be estimated, and a market appropriate to the demonstrated performance specifications is identified and quantified” (Auerswald and Branscomb 2003). With proof of concept in hand, pilots are demonstrated and tested, with the development of second or “beta” versions of the prototype for early adopters to work out bugs and evaluate the idea’s commercialization potential. It is often at this second, or pilot, phase that seed and early-stage venture capital may start to become interested. Commercialization is the process of bringing new inventions to market—that is, the marketbased scaling up of production from pilot to mass market that transforms new knowledge to wealth. Products are typically monetized either by licensing or selling the intellectual property or by marketing and selling the product. (continued)

28

The Indian Context and Enabling Environment

1

Box 1.1

continued

The move from basic and applied research to prototype development and pilot demonstration, followed by monetization, is typically complex and nonlinear, with overlaps and feedback between phases and different innovation pathways (Dosi and others 1988). It is not uncommon for unanticipated applications to arise during the process. An important implication of the varied pathways and unpredictability of creation efforts is the desirability of fostering increased collaboration by bringing different players together. Diffusion and absorption The main means of diffusion of knowledge or technology transfer are trade (technology embodied in capital goods, components, or products imported or purchased locally, as well as through interaction with foreign sellers and buyers), foreign direct investment, licensing, technical assistance, expansion of enterprises that have developed specific knowledge, copying and reverse engineering, foreign study, technical information in printed or electronic form (including what can be accessed on the Internet), twinning, and training arrangements. Proprietary technology is usually sold or transferred on a contractual basis. Nevertheless, even proprietary technology may leak out depending on the strength of the regime for intellectual property rights and its enforcement, and the capability of users. But a lot of relevant technology is in the public domain or owned by governments that could put it in the public domain. There also are specialized entities, such as productivity organizations and consulting firms, that focus on helping to disseminate technologies. These efforts usually involve training, demonstration projects, or technical assistance on how to use the technology. Industrial technologies often must be adapted to local conditions—including local raw materials, special characteristics, or to other idiosyncrasies such as local standards, climate, or power sources. It is important to have appropriate mechanisms to educate potential users in the benefits of the technology. Education often involves more than providing technical information. Moreover, use of new technologies usually requires literacy and specialized training. Finally, beyond the specific skills, using new technology often requires access to complementary inputs and supporting industries, and access to finance to purchase new equipment or inputs— or even to buy the technology license.
Source: Authors.

and development (R&D), it importantly includes other market-based and applicationbased experimentation efforts based on an understanding of what consumers (enterprises, individuals, or groups) want. Creation and commercialization activities can and do take place in both formal and informal (including rural) settings. Diffusion and absorption of knowledge activities comprise the dissemination, acquisition, adaptation, and use of existing knowledge by enterprises and institutions. For absorption of knowledge by an enterprise to be considered innovative, the technology transfer should be the first use in the enterprise’s relevant market. For example, the first leather producer in a village to import and use an international
29

1

Mark A. Dutz and Carl Dahlman

best practice machine and adapt it to local needs would constitute innovative absorption, while the fifth producer in the village to acquire and use the same machine would no longer be engaging in an innovative act. Innovative activities are not restricted to new products (goods and services of all types)—they also include innovations in processes (production on the shop floor, design, marketing and distribution, financing) as well as innovations in business and organizational models. Indicators of India’s capacity for innovation highlight its promising innovation potential. Most studies on innovation focus on the inputs and immediate outputs of formal R&D, because there is generally more such data on these efforts. Evidence on where India stands comes from quantitative data on formal R&D—such as the stock of scientists and engineers engaged in R&D and R&D spending on the input side, and scientific and technical publications and patents on the output side.10 The positive perception of India’s innovation potential by the international business press provides a useful context for the more quantitative assessment that follows.11 For example, in 2006 the Massachusetts Institute of Technology Technology Review (MIT 2006) ranked six Indian-Americans among the top global innovators under the age of 35. Based on respondents from Asia and the Pacific, the Indian company Infosys is among the world’s 10 most innovative companies, with only the Republic of Korea’s Samsung also included among comparator countries—Brazil, the Russian Federation, China, Korea, and Mexico.12 Underpinning India’s innovation potential on the input side, its stock of scientists and engineers engaged in R&D is among the largest in the world. India’s historically sizable investments in higher education focused on science and engineering. Its concentrated stock of high-caliber human capital is a top reason it is potentially very good at innovation, though this historic competitive advantage is being eroded through insufficient recent and current investment in education and skills.13 Estimates of India’s number of scientists and engineers that are researchers in R&D vary widely. One estimate is 117,528, compared with 810,525 for China and 487,477 for Russia (table 1.2). But another estimate puts India’s number of scientists and engineers at 300,000.14 Of these, about a third are conducting R&D, a third are performing auxiliary services, and a third are providing administrative and support services (Bhojwani 2006: 9). India has more than 12 million science and engineering graduates—of which 2 million are postgraduates and 100,000 are PhDs (NCAER 2005). Though there is no doubt that India has an impressive stock of skilled talent based on past investments in higher education, the more pressing question is the availability of qualified talent, measured by both quantity and quality (see chapter 5). Another critical innovation input is aggregate domestic R&D spending, which in India has never exceeded 1 percent of GDP. Over the past 20 years, India’s domestic R&D expenditures as a share of GDP have fluctuated between 0.71 and 0.91 percent, with the highest share recorded in 1987 (fiscal year 1987–88). In 2004, the last year with comparable data for India and its comparator countries, India’s share stood at 0.85 percent (Department of Science and Technology 2006: 3). These numbers compare unfavorably with comparator countries except Mexico—and then partly because Mexico is relatively integrated with the R&D system in the United States,
30

The Indian Context and Enabling Environment

1
Russian Fed.
477,647 3,319 6.8 1.17 15,782 431 109.1 173 39.3

Table 1.2 Formal Innovation Inputs and Outputs in Various Countries, 2003–04
Korea, Rep. of
151,254 3,187 17.9 2.65 13,746 1,332 287.5 4,671 3.8

Indicator
Researchers in R&D, 2003 R&D researchers per million population, 2004 Spending on R&D ($ billions), 2004 Spending on R&D (percentage of GDP), 2004 Scientific and technical journal articles, 2003 R&D spending ($ thousands) per scientific and technical articlea Scientific and technical journal articles per million population, 2003 Patents granted by U.S. Patent Office, 2004 R&D spending ($ millions) per patent granteda Patent applications granted by U.S. Patent Office per million population, 2004

Brazil
59,838 344 5.9 0.98 8,684 682 47.9 161 376.7

India
117,528 119 5.9 0.85 12,774 460 12.0 376 15.6

China
926,252 708 27.8 1.44 29,186 953 22.7 597 46.6

Mexico
26,800 268 2.7 0.43 3,747 722 37.1 102 26.9

0.90

1.21

0.35

0.46

97.03

0.98

Source: Compiled from data in World Bank (2006g, 2006h). a. Calculated by dividing estimated R&D spending in 2004 by number of articles or patents.

with more of its R&D occurring there. China’s share in 2004 was 1.4 percent and is expected to reach 2.0 percent by 2010, despite rapid growth in GDP. The average for developed countries is roughly 2.5 percent. India’s 10th five-year plan, launched in 2002, indicated that R&D spending was to reach 2.0 percent of GDP by 2007. However, given current levels, that goal is unrealistic.15 Although at nominal exchange rates, India’s domestic R&D spending was just $5.4 billion in 2004, in PPP terms it was $26.9 billion. The conversion of R&D spending to PPP terms made India the world’s ninth largest spender on R&D in 2004 (figure 1.3). This rank reflects the lower local costs of India’s R&D spending relative to those of Organisation for Economic Co-operation and Development (OECD) countries.16 But in PPP terms, China was the third-largest global spender at $94 billion, just after Japan. Given China’s ambitious expansion of investments in R&D and its higher GDP growth, by the end of 2006, China was already the world’s secondlargest R&D spender in PPP terms, at just over $136 billion, after the United States at almost $340 billion (OECD 2006).17 Domestic R&D spending is dominated by the public sector. India is still at a typical early innovation stage with regard to the distribution of domestic R&D efforts: about 75–80 percent of domestic R&D is conducted by the public sector, 20–25 percent by private enterprises, and just 3 percent by universities. In contrast, average R&D expenditures in OECD countries are 69 percent by enterprises, 18 percent by
31

1
Figure 1.3 R&D Effort in Various Countries, 2004

Mark A. Dutz and Carl Dahlman

(total gross domestic expenditure on R&D in purchasing power parity dollars)
5,800
researchers per million population

4,800 3,800 2,800

United States Russian Fed. France

Japan

Korea, Rep. of Germany United Kingdom 1,800 China 800 India 200 0 0.5 1.0 1.5 2.0 2.5 R&D expenditure as a percentage of GDP 3.0 3.5 Brazil

Source: Calculated from R&D as percentage of GDP and GDP in purchasing power parity (PPP) terms, data from World Bank (2006h). Note: The size of the bubbles represents the gross amount spent on R&D in 2004 in PPP terms.

32

universities, 10 percent by government R&D labs, and 3 percent by private nonprofit institutions (OECD 2005). In China, more than 65 percent of expenditures are undertaken by enterprises. However, with the significant increase in R&D by multinational corporations (MNCs) in India since 2002, total private R&D investment is estimated to have risen from $0.8 billion in 2002 to $4.1 billion in 2005. This led to a corresponding increase in total R&D spending from $4 billion in 2002 (where total private spending was only 20 percent) to $8.5 billion in 2005 (where total private spending, including MNCs, is estimated to have risen to 48 percent) (figure 1.4).18 On innovation output indicators, India has a strong record in producing basic knowledge, as proxied by internationally refereed scientific and technical publications. In 2003, the number of Indian scientific and technical articles published in internationally recognized journals tracked by the U.S. National Science Foundation was 12,774, compared with 8,684 from Brazil, 13,746 from Korea, and 3,747 from Mexico. But India is lagging China (27,816) and Russia (15,782) (see table 1.3). The largest share of India’s scientific publishing is done by the Council of Scientific and Industrial Research (CSIR), followed by the seven Indian Institutes of Technology and the Atomic Energy Research Institute. The number and frequency of citations of Indian institutes’ work in other publications are rising, indicating improved output and quality. There has also been a significant increase in patent applications filed in India. The largest one-year jump occurred when India joined the World Trade Organization and committed to harmonizing its system with international standards. Between 1975 and 1995, patent applications in India totaled roughly 1,000 Indian filings and 2,000–2,500

The Indian Context and Enabling Environment

1
1.10
8,500

Figure 1.4 R&D Expenditure in India, 1990–2005
9,000 8,000 national R&D expenditure (US$ million)

1.05 1.00

7,000 6,000 5,000 4,000 3,000
2,772 3,600 3,786 3,199 2,358 1,981 883 1,003 1,112 1,350 1,472 1,663 4,000 5,100 4,384 6,500

0.95 0.90 0.85 0.80 0.75 0.70 0.65 0.60 0.55 0.50 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 year government R&D number Evalueserve R&D estimate R&D/GDP
R&D/GDP (%)
614 1,225 1,155 1,324 919 937 885 1,524 1,147

2,000 1,000 0

Source: Department of Science and Technology 2006; Evalueserve 2006.

Table 1.3 Indian Patent Applications and Grants, 1975–2005
Applications Fiscal year
1975 to 1984 (annual average) 1985 to 1994 (annual average) 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Source: Gupta 2006: 39. Note: PCT Patent Cooperation Treaty, which India did not join until 1999. n.a Not applicable.

Patents granted Total
3,037 3,772 7,036 8,562 10,155 8,954 5,076 8,503 10,592 11,465 12,613 17,466

Indian
1,119 1,164 1,606 1,661 1,926 2,247 2,206 2,179 2,371 2,693 3,218 3,630

Foreign
1,918 2,608 5,430 6,901 8,229 6,707 2,601 2,160 1,870 1,723 1,678 3,165

PCT
n.a. n.a. n.a. n.a. n.a. n.a. 269 4,164 6,351 7,049 7,717 10,671

Total
1,758 1,939 1,533 907 1,844 1,800 1,881 1,318 1,591 1,379 2,469 1,911

Indian
459 498 415 293 619 645 557 399 654 494 945 764

Foreign
1,299 1,441 1,118

33

1

Mark A. Dutz and Carl Dahlman

foreign filings a year (table 1.3). Applications increased significantly when India joined the World Trade Organization in 1995, rising to more than 1,600 Indian and 5,400 foreign filings a year. In 1999, India joined the Patent Cooperation Treaty (PCT), an international organization that eases the filing of patents. After that, many applicants––overwhelmingly but not exclusively foreigners—began using this route. As India revised its patent laws through amendments to the Patent Act in 1999, 2002, 2003, and 2005, applications by both Indians and foreigners increased even more significantly. By fiscal year 2004–05, there were more than 3,600 Indian applications, more than 3,100 foreign applications, and more than 10,000 PCT filings directed at India (see table 1.3).19 Statistics on patents granted are harder to determine because different applications move through the application process at different speeds. Patent examinations take arbitrarily long and can entail multiple rounds of correspondence, depending on the application. As a result, patents granted in any given year necessarily result from applications first submitted in earlier years, and so lag behind changes in intellectual property rules and in the national culture, and broader changes in the innovation system. The share of Indian patent applications in the United States is small but has risen significantly in recent years––led by MNCs and CSIR. The share of Indian patent applications in the United States rose from 0.04 percent of the worldwide total in 1995 to 0.37 percent in 2004.20 A ranking of patents granted in the United States between 1995 and 2004 showed India in 24th place worldwide. Most of the top countries were OECD members. But seven non-OECD economies placed ahead of India: Taiwan (China), Korea, Israel, Singapore, Hong Kong (China), China, and Russia, in that order. The last two may have been expected, China because of its size and Russia because of its technological legacy. It is telling, however, that the five other economies are quite small. Israel is also a special case, but the high rankings of Taiwan, Korea, Singapore, and Hong Kong indicate the importance that these small economies place on competing in the global market based on innovation. It is also telling to examine who applied for patents in India over 1995–2005. Of the top 50 applicants, 44 were foreign firms operating in India—only 6 were Indian organizations. Three of these were public institutions (CSIR, Indian Institute of Technology, Ministry of Defense), and one was a public corporation (Steel Authority of India). Only two were private Indian firms (Ranbaxy and Dr. Reddy’s Lab), both in the generic drug industry that got its start under India’s former, more nationalist, patent regime. Overall, India appears better at producing basic knowledge than commercializable knowledge. Even so, R&D spending appears more efficient in India than in comparator countries. Using the publication of scientific and technical journal articles as a proxy for basic knowledge outputs and patents granted in the United States as a proxy for commercializable outputs, India is relatively stronger in the production of basic than market-driven knowledge. However, based on efficiency of R&D spending, as measured by the relative costs of a scientific and technical publication or a U.S. patent, India does better than all its comparator countries except Russia for journals and Korea for patents (see table 1.3).21 India has perhaps the
34

The Indian Context and Enabling Environment

1

lowest R&D costs among these countries as a result of its lower pay to scientists and engineers—the main cost component in R&D spending.

Innovation in Indian Manufacturing: Outputs, Inputs, and Productivity
Based on a 2006 survey of manufacturing enterprises, the extent of India’s innovation outputs is not too dissimilar from those of comparator countries, with India in the mid-range. The India 2006 Enterprise Survey (World Bank 2006c) provides a basis for within- and cross-country analyses. It allows both links between innovation outputs and inputs and enterprise characteristics, and links between innovation activities and enterprise productivity. Figure 1.5 shows the percentage of surveyed firms that responded positively to these questions on innovation outputs, relative to comparator countries.22 In India, 40 percent of firms had developed a major new product, while 62 percent had upgraded an existing product line. These criteria suggest that Indian firms have more innovation outputs than firms in China, but less than those in Brazil, Korea, and Russia. It is somewhat surprising that China scores so low on both measures and Brazil scores so high. The low scores for China may indicate that the Chinese are more active in copying than developing new products,23 or are culturally more reluctant to consider novelty in their evolving products. In any event, India tends to rank in the lower-middle range relative to other countries. Ultimately, innovation is not an end in itself but a means to productivity growth and higher living standards. Any discussion of how to enhance India’s environment for innovation must be grounded in an appreciation of the role that innovation can play in promoting productivity and so national competitiveness, growth, and,

Figure 1.5 Innovation Outputs in Various Countries, 2003–06
Brazil (2003) 93.4 64.1 64.1 45.9 59.2 39.3 42.6 23.5 60.5 44.2 0 10 20 30 40 50 60 percentage of firms 70 80 90 100

Russian Fed. (2005)

India (2006)

China (2003)

Korea, Rep. (2005)

upgraded a product
Source: World Bank Enterprise Surveys.

introduced new product

35

1

Mark A. Dutz and Carl Dahlman

ultimately, poverty reduction. For most enterprises in a developing country at India’s level, the acquisition of global knowledge is expected to be more important for productivity than is the creation of domestic knowledge—because there is still so much to gain by drawing on the global knowledge frontier. Only for enterprises producing at the cutting edge of the global frontier in their sectors is creation the key source of competitiveness and growth. For all others not at the frontier, competitiveness by absorbing existing but better production methods not used locally is typically the best approach. In addition, absorption of knowledge (whether generated domestically or acquired from abroad) will do more to raise economic productivity than will creation of knowledge. Finally, knowledge contributes to productivity increases only
Table 1.4 Key Innovation-Related Findings of the India 2006 Enterprise Survey
Findings
Enterprises engaged in creation are likely to be larger, export-oriented, foreign-owned, ISO-certified. Creation-oriented enterprises are concentrated in drugs and pharmaceuticals, auto components, and garments, though innovative firms exist in all sectors. Enterprises that absorb knowledge are more likely to be larger, export-oriented, foreign-owned, ISO-certified. Garment enterprises are most likely to acquire new technology. Drug and pharmaceutical firms are most likely to pay royalties, use e-mail and computers, and subcontract R&D. The most important channel for absorbing existing knowledge is through use of new machinery and equipment, followed by hiring of key personnel.

Support
Firm size, exporting, and having ISO certification correlate strongly (1 percent level of significance) with the development of important new product lines. Sectors with the largest shares as listed Sectors with the lowest shares of firms developing new product lines are paper and wood, mining, mineral prospecting, metals, and leather. Listed firm characteristics correlate strongly with enterprises that acquired new technology, or paid royalties or licensing fees. Sectors with the highest frequency of acquiring new technology, paying royalties, using e-mail, using computers, and subcontracting R&D as listed.

Some 70 percent of enterprises cite new machinery and equipment as the main source of absorbing technology. Those citing hiring key personnel as the main source total 10 percent. The third most common source for acquiring new technology, cited by nearly 10 percent of firms, is by developing or adapting it within the firm. This figure jumps to 12 percent when development with an equipment or machinery supplier is added.

A significant amount of informal, inputtype creation activities beyond R&D occurs within enterprises.

Measures of formal innovation inputs— both creation (R&D spending) and absorption (technology acquisition)— are closely and jointly associated with innovation outputs.

Larger, export-oriented, and ISO-certified firms as well as enterprises with a higher general level of manager education are more likely to spend on R&D and acquire new technology. These measures correlate strongly with developing new product lines. These measures are both significant even in the same probit regression, suggesting that they are complements rather than substitutes.

36

(continued)

The Indian Context and Enabling Environment

1
Support
A number of enterprises that developed or upgraded product lines did not conduct R&D or acquire technology: 39 percent of firms developed new products, and 59 percent upgraded existing product lines. But only 27 percent did formal R&D, only 16 percent reported acquiring a new technology, and only 34 percent had any R&D spending, acquired new technology, or paid royalty or licensing fees. Based on ordinary least squares estimation of productivity functions, development of new products has a strong and significant association with productivity. Based on estimation of frontier production functions, separate measures of creation and absorption are significantly associated with narrowing the distance the frontier. The absorption effect is stronger and more significant, implying a bigger movement toward the frontier. When both absorption and creation are included, the latter becomes insignificant.

Table 1.4 continued
Findings
Innovation outputs are also associated with informal efforts to create and absorb knowledge.

Innovation outputs are strongly associated with enterprise productivity.

Absorbing existing technology has a stronger, more significant association both with productivity than does spending toon R&D.

Source: Authors.

to the extent that it is used—hence, an emphasis on productive use is critical. Table 1.4 highlights key empirical findings from the Enterprise Survey on innovation inputs, outputs, and productivity. 24

The Enabling Environment for Innovation
The effort made by enterprises to create and absorb knowledge depends on the enabling environment for innovation. This section introduces some of the policies, institutions, and capabilities required to support innovation. Competition as the critical incentive for spurring innovation is discussed in detail, as are desirable principles for public financial support of innovation and institutional coordination. The enabling environment for innovation should stimulate knowledge creation and absorption by enterprises to enhance competitiveness, exploit synergies between enterprises and other centers of knowledge, and provide incentives and support to supply inputs for innovation––particularly skills, information, and finance. The enabling environment for innovation comprises a country’s innovation system as well as support for essential inputs as part of the country’s broader investment climate (table 1.5). The innovation system, in turn, consists of policies, institutions, and capabilities that affect how enterprises create and absorb knowledge. Key support institutions for creating and commercializing knowledge include universities, public and private research centers, and policy think tanks. However, enterprises are at the center: if the private
37

1
Components of enabling environment
Creation and commercialization of new knowledge

Mark A. Dutz and Carl Dahlman

Table 1.5 The Enabling Environment for Innovation: Policies, Institutions, and Capabilities

Policies
• Policies to promote more private R&D Intellectual property rights regime Matching grants Tax subsidies • Public spending on R&D National mission programs Competitive grants Peer reviews • Support for pro-poor innovations • Openness to global knowledge flows Trade Foreign direct investment Technology licensing policy Internet access • Foreign education and attracting the diaspora • • • • • • •

Institutions

Capabilities

Public labs, universities • High-level human Private R&D labs capital for R&D IPR institutions (scientists, Technology transfer offices engineers, Science and technology parks technicians) Technology incubators • TechnoResearch and education entrepreneurship networks • Specialized nongovernmental institutions • Grassroots networks • Early-stage technology development finance and venture capital

Diffusion and absorption of existing knowledge in new locations

• Technical information • Formal education services and skills • Technology upgrading • Engineering • Productivity organizations consulting firms • Metrology, standards, testing, • Business support and quality control systems services • National research and education networks • Networks at cluster level • Technology absorption finance for micro, small, and medium enterprises • • • • Efficient financial system Flexible labor market Effective courts and judiciary Market-responsive formal education institutions and lifelong learning system • Literacy • Secondary and higher education graduates • Managers • Entrepreneurs

Broader investment climate

• Competition and trade • Regulatory policies, especially toward infrastructure • Entrepreneurship support • Good rule of law • Macroeconomic stability

Source: Authors.

sector has little demand for knowledge, the innovation system cannot be effective. The main idea behind an innovation system is synergy between the major innovation players to create better, cheaper products that meet consumer needs.

Competition-Related Incentives to Improve Performance
Competition is critical for innovation. Product market competition encourages enterprises to innovate. The classic characterization of innovation dates to Joseph Schumpeter’s notion of “creative destruction” (Schumpeter 1975 [1942]). Markets reward powerful creativity with extraordinary returns, and consumer demand turns genuine innovators into powerful incumbents. Because the best way to defeat such an incumbent is to produce a far superior product, competition drives the next
38

The Indian Context and Enabling Environment

1

generation of innovators to “destroy” the incumbent’s market position by “creating” the next generation of products. In well-functioning markets, investments by new entrants can stimulate larger, established enterprises with market power into greater efforts—leading them to likely be the first to achieve routine or incremental innovations. But new entrants are also more likely to effect revolutionary or leapfrog innovations that render heritage products obsolete. The speed and vigor of each group’s investment in innovation are driven by competition from others in its group and members of the other group. Government barriers that impede fluid entry and exit––that either overly protect incumbent firms and workers, or sour the rewards of successful innovation by challenging profitable gains––stultify the forces that make innovation a self-sustaining outcome of competition. The importance of competition in stimulating knowledge creation and absorption is shown by comparing the situation in India before and after the business liberalization of the 1980s and trade liberalization of the 1990s. Before these reforms, India followed a policy of technological self-reliance. The large public R&D infrastructure was oriented toward developing technologies that supported small industries using indigenous materials, as well as large state enterprises in key sectors and in defense. Although many technologies were developed, most were not world-class. The private sector invested little in R&D and did not develop much globally competitive technology. After liberalization, Indian businesses began facing more competition and started increasing R&D investments (figure 1.6) (Rodrik 2005). Enterprise R&D
Figure 1.6 R&D Intensity of Indian Corporations in All Reporting Firms and Three Key Sectors, 1991–2004 (R&D spending as a percentage of sales)
6

5

R&D as share of sales

4

3

2

1

0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 year all firms auto pharmaceuticals software

Source: Compiled from data in Bowonder and others (2006).

39

1

Mark A. Dutz and Carl Dahlman

spending as a share of sales increased more than sevenfold, from 0.07 percent in 1991 to 0.53 percent in 2004. The sectors most open to competition––pharmaceuticals, software, auto components––have increased R&D spending the most. In addition to sharp market incentives for competition and entrepreneurship, India needs to strengthen sociocultural norms for innovation. Entrepreneurship-friendly policies include ensuring that markets are open to competition from new entrants and that bankruptcy laws facilitate rapid recovery. In addition, sociocultural norms should place high social value on commercial success and see failure as an often indispensable learning experience that can eventually lead to business success. Finally, an unfortunate legacy of India’s colonial past pertains to education. The British Raj fostered learning by memorization rather than creative problem solving, because the goal was to produce administrators who could help the British rule India more efficiently (Evalueserve 2006: 8). Recommendations to strengthen innovation-friendly sociocultural norms center on the following: • Campaigns to raise awareness of the importance of R&D for competitiveness and of commercialization of ideas for wealth creation and national welfare. Mass media (TV, movies) and champions, role models, and mentors could be used for these efforts. • Dissemination of success stories of techno-entrepreneurs and other innovators through publicity, prizes, and public recognition for cases that exemplify how knowledge has been turned into wealth or used to improve welfare. • High-profile awards for creative teachers to encourage them to inspire creativity in their students, from primary and secondary school through vocational training and university education. The most important policy change for increasing knowledge creation, commercialization, diffusion, and absorption is to sharpen competition among enterprises so that innovation becomes essential. Unencumbered entry and exit of enterprises are perhaps the most important stimuli to innovation. Limits on small firms and other barriers to entry and exit should be eased, imports further liberalized and firms pushed to export more, and opening to foreign competition expanded. Enterprise entry and expansion are also constrained by insufficient access to skilled labor, information and communication technology, finance, and other business services, especially power. These changes would also likely increase India’s ability to tap into rapidly growing global knowledge (see chapters 2 and 3). Although these are rather conventional recommendations that have been articulated many times,25 they remain crucial. Such reforms will be resisted by vested interests who would be hurt by greater competition. Shorter-term recommendations, therefore, are to continue to open the Indian economy to global competition, encourage exports, increase domestic competition as quickly as is politically feasible, and to do so in a way that raises awareness of longer-run benefits and creates more support. Eventually, India should aspire to reach levels of trade and investment openness comparable to those of OECD countries, as China is doing. Such openness
40

The Indian Context and Enabling Environment

1

Figure 1.7 Obstacles to Starting and Closing a Business in Various Countries, 2006
a. Starting a business, time and procedures Brazil Russian Fed. India China Korea, Rep. of Mexico OECD 0 28.0 35.0 35.0 22.0 27.0 16.6 20 40 60 80 100 120 140 160 152.0 Brazil Russian Fed. India China Korea, Rep. of Mexico OECD 0 2 4 6 6.2 8 10 12 14 16 18 8.0 7.0 11.0 13.0 12.0 17.0

time (days)

number of procedures

b. Closing a business, time and recovery rate Brazil Russian Fed. India China Korea, Rep. of Mexico OECD 0 1.5 1.8 1.4 2 4 6 time (years)
Source: World Bank 2006b. Note: OECD Organisation for Economic Co-operation and Development.

4.0 3.8 10.0 2.4

Brazil Russian Fed. India China Korea, Rep. of Mexico OECD 8 10 12 0

12.1 28.7 13.0 31.5 81.8 63.2 74.0 10 20 30 40 50 60 70 80 90 recovery rate (cents on the dollar)

encourages enterprises to upgrade their productivity and competitiveness––and so provide better and cheaper products, create higher-paying jobs, and help reduce poverty. Efforts to sharpen competition that improves performance should focus on facilitating entry and easing the reallocation of capital to more productive enterprises. Figure 1.7 shows some of the barriers to competition in India by comparing its regulations for enterprise entry and exit with those of comparator countries. As noted, streamlined regulations easing entry and exit are critical to stimulating innovation and productivity: • Starting a business identifies the legal and bureaucratic hurdles that entrepreneurs must overcome to enter the market—here, the average time spent completing entry requirements and the number of procedures required to register a firm. Registration is typically critical for accessing a range of market infrastructure, including finance, physical infrastructure (electricity, water), and contract enforcement. The greater is the number of procedures, the more scope for enforcing them in uneven ways. In terms of the time required to start, India ranks second relative to benchmark countries at 35 days, tied with China and
41

1

Mark A. Dutz and Carl Dahlman

ahead of Brazil, but significantly behind the OECD average of 17 days. (India is especially behind Australia and Canada, each requiring only two procedures and no more than 3 days.) Although India has significantly liberalized foreign direct investment (FDI) since the early 1990s, some sectors still have restrictions— with FDI either completely prohibited, as in retail, or sector caps added, as in telecommunications and insurance. Moreover, 326 products in various sectors are reserved for small industries, preventing entry by larger firms and normal expansion by the smaller ones. • Closing a business tracks legal, procedural, and administrative bottlenecks in the bankruptcy process—here, the time required to complete a bankruptcy and how many cents on the dollar that claimants (creditors, tax authorities, employees) recover from insolvent firms. There is particular scope for facilitating exit in India in terms of time––10 years, the longest in the world––which is in stark contrast to China (2.4 years) and the OECD average (1.4 years). The focus on competition problems should occur at the local level, since that is where problems occur and pressures by vested interests are greatest. Indian authorities should recognize the variance across cities. Time and recovery rates are 8.3 years and 17.3 cents on the dollar in Bangalore—and 20.2 years and 5.1 cents in Kolkata (compared with an average recovery rate of 74 cents in OECD countries).26 Reducing the stigma of failure and easing the reallocation of capital to more productive enterprises are among India’s most important reforms. In India, capital is too scarce not to allow its reallocation to more productive uses as quickly as possible. Reforming exit policy reform through more efficient bankruptcy rules would help remove the stigma of failure and contribute to increased risk-taking and experimentation. The government has initiated reforms to improve the legal and regulatory framework for insolvency. The Companies (Second Amendment) Act 2002, expected to be presented in Parliament in 2007, removes a number of deficiencies. The J. J. Irani Committee reviewed the Second Amendment provisions and recommended further significant reforms to the Companies Act, including its insolvency and rehabilitation provisions. Implementation of these recommendations would significantly improve the insolvency regime. In addition, initiatives have been launched to computerize company registries and introduce information technology and case management tools in courts. Authorities also need to build capacity among all participants in the process— judges, liquidators, creditors, operating agencies, investigators, auditors, valuers, and oversight bodies. Systematic education and certification for liquidators are also needed to ensure professionalism and develop insolvency practices. Finally, modernizing the Industrial Disputes Act to reduce the bias toward adjudicating disputes and increase flexibility for employers in hiring and firing in a way that also protects worker rights would help close less productive enterprises and enable capital to flow to more productive ones. Public policy must ultimately promote both competition and collaboration. A main theme of this book is the effectiveness of market forces in inducing innovation,
42

The Indian Context and Enabling Environment

1

and the need to strengthen and extend the domain of market competition so that innovation can flourish. However, an enterprise’s ability to generate new products and processes depends on two complementary processes—analysis and interpretation.27 On the one hand, innovation requires analysis, where alternative outcomes can be clearly defined and distinguished from one another—an engineering and management–based project approach of problem solving and rational decision making. On the other hand, innovation also requires interpretation and trust, where the possible outcomes are unknown—a more open-ended process approach that requires drawing various actors together, initiating and guiding conversations to allow economic actors to understand where they have the same and where they have different interests. Both processes are necessary for successful innovation, with the latter especially critical when the problem is not yet well defined. But while analysis is spurred by competition, interpretation benefits from collaboration, a sheltered space from competition where the risk of private appropriation of information is reduced and where misunderstandings have fewer direct consequences. Accordingly, there is a need to build an effective institutional infrastructure that allows appropriate collaboration and networking. The breakdown of bureaucratic silos within the public sector, as well as the improvement of linkages between universities, public and private R&D labs, and private enterprise through consortia, is part of this institutional infrastructure.

Principles for Pragmatic Coordination of Innovation Support Programs
Coordinating existing and new programs that support innovation would benefit from a new approach—one that builds an effective institutional infrastructure that helps the various elements form a coherent system.28 The main idea behind a national innovation system is collaboration and synergy among enterprises, universities, research institutes, and government. Yet the reality in most countries, including India, is of ivory towers––and additional silos within each––with insufficient incentive to interact. Most innovation systems resemble a jigsaw puzzle. Many pieces of the puzzle (a vibrant private sector, solid public research institutes, a rapidly growing venture capital industry) already exist in India. The challenge is putting the pieces together in a coherent whole. In designing a new national innovation program, India should not introduce more targeted programs that, although individually useful, would worsen the jigsaw problem. For example, a typical well-intentioned solution is to coordinate the pieces top-down through an interministerial council. However, international experience suggests that such attempts largely fail: councils usually become cartels of established interests, while new ministries become efficient at creating and defending their own turf. The drive for top-down coordination is a symptom that something is wrong with the incentives of enterprises, universities, research institutes, and government entities. In fact, as economies have become more complex, top-down coordination has posed more of a problem.
43

1
Box 1.2

Mark A. Dutz and Carl Dahlman

Innovation Foresight Processes in Canada and the Netherlands

Canada. Technology road maps for industry R&D. Technology road mapping is a planning process driven by the projected needs of tomorrow’s markets. It helps companies identify, select, and develop technology alternatives to satisfy future service, product, and operational needs. Through this process, companies in a given sector can pool their resources and work with academia and governments to look 5–10 years into the future and determine what their market will require. The technology road map process is led by industry and facilitated by Industry Canada, Canada’s Ministry of Industry. The Strategic Project Grants Program of the Natural Sciences and Engineering Research Council of Canada (NSERC). This program funds project research in target areas of national importance and emerging areas of potential significance to Canada. The research is at an early stage, with the potential to lead to breakthrough discoveries. Targeted areas are identified in consultation with experts from all sectors. Identifying opportunities for leapfrogging. A new body has been created by the NSERC to advise on areas where Canada may be able to leapfrog to the front ranks of research in the natural sciences and engineering. The Netherlands. Foresight processes are conducted by a number of advisory bodies. The Royal Netherlands Academy of Arts and Science engages in foresight processes from the perspective of promising scientific developments. Several other bodies conduct or are involved in foresight processes from the perspective of knowledge demand. For instance, the Sector Councils, which cover a broad array of societal sectors, draw up research agendas based on inputs from government, scientists, and the sectors involved. A recent example of a priority-setting mechanism with a direct follow-up in investment funding is the ICES-KIS program, which involves extensive consultations with various stakeholders.
Source: Authors.

A possible solution of pragmatic coordination involves combining three coordination elements, drawing on lessons from Chile, China, and Finland, among others: • Conducting regular, independent evaluations with international benchmarking. Such evaluations link the assessed impacts of programs––relative to international benchmarks—to decisions on the allocation of budget resources. Best practice is to allocate 3–5 percent of a program’s budget to evaluation. Although there will be resistance to evaluating programs whose inefficiency is likely to be revealed, impartial and public access to results will provide a strong disciplining effect and help push for changes to improve performance. This kind of evaluation must be driven by “champions”––individuals willing to risk their reputations on the results of reforms. An informal group of champions in key innovation entities exists in India, and should include national leaders from industry, research centers, universities, and government, as well as leaders from the diaspora. For policy and programmatic ideas to be turned into actions, the champions must include key national decision makers in their deliberations.
44

The Indian Context and Enabling Environment

1

• Introducing new programs through strategic pilots. Because scaling up, not piloting, is the main goal of new programs, pilots should be selected based on their scalability and diffusion, as is done in China. This is why they are called “strategic” pilots. • Fostering innovation foresight for future societal needs to inform current decisions. Innovation foresight links a consensus view on the future needs of society at large with new tools of science and technology to address those needs. Early efforts at using innovation foresight approaches were conducted in the United Kingdom in the 1990s. The first resulted in the unexpected articulation of the widespread ramifications of an aging population and the possibilities of innovation in meeting those needs. Since then, foresight processes have been adopted elsewhere and have proven particularly useful in defining long-term needs and developing the creative synergies from which innovation emerges (box 1.2). A nationwide foresight process in India with collaborative interaction of a group of industry, government, civil society, and research community representatives could start with a focus on thematic national challenges, such as access to clean water or road transport congestion in cities.

Notes
For questions or further information, please contact Mark A. Dutz at mdutz@worldbank.org or Carl Dahlman at carldahlman@gmail.com. 1. The term “economic dualism” is used here not in the sense of Nobel laureate Arthur Lewis’ 1954 model of labor market dualism, but rather to emphasize the stark contrast between the “two Indias.” 2. See Planning Commission (2006, Table 2) where an all-India consumption poverty head count ratio of 27.8 percent is reported based on the 2004–05 National Sample Survey, with poverty in the worst state rising to 46.5 percent. 3. See World Bank (2006h) and Planning Commission (2006, Table 2): all-India female illiteracy of 46.3 percent is calculated based on information from the 2001 Census, with illiteracy in the worst state rising to 66.9 percent. 4. For economic theory and empirical bases for the linkage between competition, innovation, productivity growth, and overall economic growth, see Aghion (2006), Aghion, Bloom, and others (2005), and Aghion, Blundell, and others (2006). 5. As reported in Planning Commission (2006), the consumption poverty head count ratio has fallen from 36.0 percent in 1993–94 to 27.8 percent in 2004–05. Although there is no established causal link between innovation and poverty alleviation, it is plausible to presume that innovation can have a longer-term impact on poverty by increasing growth, as well as a more direct impact through pro-poor innovation efforts. 6. While informality is often deemed to be driven by high formal sector taxes and restrictive rules and regulations, the dualism referred to here is driven by skills gaps and other impediments to knowledge creation, absorption, and use. 7. Defining formal and informal employment is difficult. The formal sector is usually more modern, subject to more government regulation and taxation, and uses more updated production and organization techniques than the informal (also known as traditional) sector. This section uses estimates from World Bank (2006f) that compare data from five-year National Statistical Office surveys with the population censuses conducted every decade, and data from the Annual Survey of Industries for the organized manufacturing sector (see table A.1 in the technical appendix).

45

1

Mark A. Dutz and Carl Dahlman

In World Bank (2006d: 7), the workforce estimate is roughly 390 million for 2003, with only 8 million employed in the formal private sector. According to the latest National Sample Survey data for 2004–05, the size of India’s workforce was 457.4 million; although private formal employment has definitely grown since 2003, an updated figure is not available. 8. See in particular World Bank (2006e). 9. See Bosworth and Collins (2006). 10. There is much less international, comparable information on the inputs and outputs of informal creation and on knowledge diffusion and absorption by enterprises. 11. The rankings presented in the following paragraphs, based on quantitative measures, are more sober than, for instance, the subjective assessments of international business leaders on India’s innovation potential in World Economic Forum (2006). 12. Inspired by the comparison of the so-called BRICs economies of Brazil, Russia, India, and China initiated by reports from Goldman Sachs, this study uses a somewhat larger set of comparator countries wherever data are available, and reports figures in the BRICKM order: Brazil, Russia, India, China, Korea, and Mexico. See Business Week’s report (April 24, 2006) on the world’s most innovative companies, based on a global survey of 1,070 senior managers in early 2006. 13. These are 2 of the top 10 reasons India can innovate, according to Govindarajan (2006). 14. http://www.investmentcommission.in/human_capital.htm. 15. By including own estimates of domestic and especially multinational corporation spending on R&D for 2003 and beyond, Evalueserve (2006) estimates that total R&D spending in India was 1.08 percent of GDP in 2005, and forecasts that this figure will rise to 1.75 percent of GDP by 2010. 16. A dollar of R&D spending goes further in India—or, conversely, R&D in India can be conducted at roughly one-fourth of the cost in the United States. See Evalueserve (2006). 17. Figures for 2005 and 2006 are projections based on continuation of the observed 2000–04 growth. 18. See Evalueserve (2006) who have added to the Indian Department of Science and Technology figures for 2003–05 (fiscal years 2003–04 and 2005–06) rough estimates based on surveys of domestic private enterprises and MNCs in sectors such as IT services, electronics and electrical equipment, chemicals, drugs and pharmaceuticals, biotech, and automotive engineering services. 19. Based on Abramson (2007: 23). 20. To make international comparisons on patenting, it is necessary to standardize for the patent regime because of national differences in what can be patented. Thus, it is useful to examine patents submitted to the U.S. Patent Office, because the United States represented the largest global market until the European Community recently overtook it. 21. These very aggregate measures are just a rough proxy because they do not control for the importance of the articles or the patents. In addition, issue can be taken with the relevance of U.S. patents as the measure. Still, these measures provide a characterization of India’s broad performance and strengths relative to other countries. 22. The survey’s best available proxy for innovation output is whether an enterprise developed an important new product line in the past two years. A second, broader proxy for innovation is whether an enterprise upgraded an existing product line. Neither of these enables inferences about whether the products were truly new to the world or simply new to India. Most “innovative” products are probably not new to India but new only to the enterprise or to its market. The latter is more likely in cases where the relevant market is local or regional, rather than national. The extent of novelty also could vary across enterprises depending on how the question was asked and what responding managers perceived as being “new.” 23. See, for instance, Economist (2006). 24. R&D spending (or whether any R&D is done within the firm or subcontracted) is used as a proxy for creation activities, while reporting on having “acquired new technology over the past two years that either substantially changed the way that main products are produced or allowed the production of new products” is the main proxy for absorption activities (also paying royalties or licensing fees, using Web sites, and the like). See technical appendix tables A.2 and A.3 for key regression results.

46

The Indian Context and Enabling Environment

1

25. See in particular the excellent multisector study by McKinsey Global Institute (2001), which remains as relevant today as it was upon publication. 26. For data highlighting variations in restraints across cities in specific countries, see World Bank (2006a: 28; 2007a). 27. See, for instance, Lester and Piore (2004). 28. This section draws from, among others, Kuznetsov (2006) and World Bank (2007b). Examples of how strengthening the institutional infrastructure becomes a key part of the action agenda for fostering innovation are provided in World Bank (2007b: 100–04).

References
Abramson, Bruce. 2007. “India’s Journey toward an Effective Patent System.” Policy Research Working Paper 4301, South Asia Finance and Private Sector Development, World Bank, Washington, DC. Aghion, Philippe. 2006. “A Primer on Innovation and Growth.” Bruegel Policy Brief, Issue 6, Brussels. Aghion, Philippe, N. Bloom, R. Blundell, R. Griffith, and P. Howitt. 2005. “Competition and Innovation: An Inverted U Relationship.” Quarterly Journal of Economics 120 (2): 701–28. Aghion, Philippe, R. Blundell, R. Griffith, P. Howitt, and S. Prantl. 2006. “The Effects of Entry on Incumbent Innovation and Productivity.” NBER Working Paper No. 12027, National Bureau of Economic Research, Cambridge, MA. Auerswald, Philip E., and Lewis M. Branscomb. 2003. “Valleys of Death and Darwinian Seas: Financing the Invention to Innovation Transition in the United States.” Journal of Technology Transfer 28 (3–4): 227–39. Bhojwani, H. R. 2006. “Report on the Indian Civilian Public R&D System.” Background paper commissioned by South Asia Finance and Private Sector Development, World Bank, Washington, DC. Bosworth, B., and S. Collins. 2006. “Accounting for Growth: Comparing India and China.” Paper presented at Tokyo Club Foundation for Global Studies annual conference, December 6–7. www.tcf.or.jp/data/2006120607_B_Bosworth-S_Collins.pdf. Bowonder, B., V. Kelkar, N. G. Satish, and J. K. Racherla. 2006. “Innovation in India: Recent Trends.” TMTC (Tata Management Training Center) Research Paper, Pune, India. Department of Science and Technology. 2006. Research and Development Statistics 2004–05. New Delhi: Ministry of Science and Technology. Dosi, G., C. Freeman, R. Nelson, G. Silverbert, and L. Soete. 1988. Technical Change and Economic Theory. New York: Columbia University Press. Economist. 2006. “While China’s Carmakers Copy, India’s Are Inventing.” December 16, p. 64. Evalueserve. 2006. “Arth Karicheye Vidya: From Innovation to Commercialization.” Revised Part 1 and Part 2. Prepared for Federation of Indian Chambers of Commerce and Industry, New Delhi. Govindarajan, Vijay. 2006. “Reasons Why India Can Innovate.” Business World, July 3. Gupta, R. K. 2006. “The IPR System in India.” Background paper commissioned by South Asia Finance and Private Sector Development, World Bank, Washington, DC. Kuznetsov, Y. 2006. “Radical Transformation, Step-by-Step: Inside-Out Reform of India’s Innovation System.” Background paper commissioned by South Asia Finance and Private Sector Development, World Bank, Washington, DC. Lester, Richard K., and Michael J. Piore. 2004. Innovation—The Missing Dimension. Cambridge: Harvard University Press.

47

1

Mark A. Dutz and Carl Dahlman

Lewis, W. Arthur. 1954. “Economic Development with Unlimited Supplies of Labor.” Manchester School of Economics and Social Studies 22: 139–91. McKinsey Global Institute. 2001. India: The Growth Imperative. Washington, DC: McKinsey & Company. MIT (Massachusetts Institute of Technology). 2006. Technology Review. Cambridge, MA. http://www.technologyreview.com/TR35/. NCAER (National Council of Applied Economic Research). 2005. India Science Report: Science Education, Human Resources and Public Attitudes towards Science and Technology. New Delhi: NCAER. OECD (Organisation for Economic Co-operation and Development). 2005. Science Technology and Industry Scoreboard 2005. Paris. ———. 2006. Science, Technology and Industry Outlook 2006. Paris. Planning Commission. 2006. “Towards Faster and More Inclusive Growth: An Approach to the 11th Five Year Plan.” Government of India, New Delhi. Rodrik, Dani. 2005. “From ‘Hindu Growth’ to Productivity Surge: The Mystery of the Indian Growth Transition.” IMF Staff Papers 52 (2): 193–228. Schumpeter, Joseph A. 1975 [1942]. Capitalism, Socialism and Democracy. New York: Harper and Row. World Bank. 2006a. Doing Business in South Asia in 2006. Washington, DC. ———. 2006b. Doing Business 2007: How to Reform. Washington, DC. ———. 2006c. Enterprise Surveys. Washington, DC. www.enterprisesurveys.org. ———. 2006d. “India Inclusive Growth and Service Delivery: Building on India’s Success.” Development Policy Review, Report No 34580-IN, Washington, DC. ———. 2006e. “India: National Agricultural Innovation Project.” Project Appraisal Document, Report No. 34908-IN, Washington, DC. ———. 2006f. “India’s Employment Challenge: Creating Jobs, Helping Workers.” Report No. 35772-IN Draft, PREM South Asia, Washington, DC. ———. 2006g. KAM (Knowledge Assessment Methodology) database 2006. Washington, DC. www.worldbank.org/kam. ———. 2006h. World Development Indicators 2006. Washington, DC. ———. 2007a. Doing Business in South Asia 2007. Washington, DC. ———. 2007b. Enhancing Agricultural Innovation: How to Go Beyond the Strengthening of Research Systems. Washington, DC: World Bank. World Economic Forum. 2006. Global Competitiveness Report 2006–07. Geneva, Switzerland.

48