INCENTIVES AND SUPPORT SYSTEMS TO FOSTER PRIVATE SECTOR INNOVATION Jerry Sheehan Introduction Governments in many countries are devoting increased attention to bolstering business innovation capabilities. This interest reflects growing recognition of both the importance of innovation in driving economic growth and the dominant role of the business sector in that process. In developed economies, business not only outspends the public sector on R&D (by a factor of two in OECD countries), but is the main source of the new products, processes and services that innovation brings to the marketplace. Designing effective incentives and support systems for private sector innovation is a challenging task. While many of the standard policy instruments government grants, tax incentives, loans, partnerships and the like are well-known and widely used, their effectiveness depends on their specific design and degree of adaptation to local (national) needs. In addition, policy makers must consider the overall mix of policies used to stimulate innovation. They must determine, for example, how best to balance broad-based horizontal policies against targeted measures that support specific sets of firms, technologies or industry sectors. They must determine how much weight to give to direct versus indirect forms of intervention, and to financial supports versus other types of policy measures that are necessary to business innovation. This paper provides a brief overview of policy instruments to support business innovation with an eye toward identifying the tradeoffs inherent in policy design and the development of coherent policy mixes. It focuses primarily on instruments used to finance business R&D, but considers complementary policy measures as well, including support to venture capital and to public-private collaboration. While the paper draws primarily from the experience of OECD economies, it aims to identify lessons that can be applied to a more diverse set of economies. Given the complexity of the issue, the paper aims less to provide definitive answers to these questions than to seed discussion among participants in the I Hemispheric Meeting of the Science, Technology and Innovation Network of the IADB s Regional Policy Dialog. Financial incentives for business innovation It is widely recognized that the social returns to investments in innovation far exceed the private returns and, hence, that firms tend to under-invest in R&D and innovation. As a result, governments in virtually all developed countries provide financial support to business R&D. This support can take many forms, including various types of grants, loans and incentives, but the two most widely used are tax breaks for R&D spending and direct R&D subsidies. Tax incentives finance business R&D by reducing the tax burden on firms in proportion to their level (volume) of expenditure on R&D or their incremental increase in R&D expenditure over a defined period of time (or a combination of the two). Typically, the reduction is taken as a credit or allowance against corporate taxes owed. As of 2006, nineteen of the thirty OECD member countries used tax incentives of one form or another to stimulate business R&D, up from twelve countries a decade earlier. Tax incentives can typically be used by all firms that engage in R&D, although special provisions may be required to ensure that they benefit young firms with little or no taxable income. Tax incentives leave to firms the decisions regarding the content and objectives of R&D projects, thereby reducing concerns about governments picking winners. Tax incentives also entail minimal administrative costs for governments, limited to efforts taken (if any) to validate amounts of the incentive claimed by firms. 1
Compared to tax incentives, direct funding allows for more targeted support of R&D, providing governments with the opportunity to direct resources to areas believed to yield the greatest social returns. Funding can be directed toward industries or technologies deemed important to national needs or toward research programs with longer term paybacks. While many programs fund the full cost of the proposed R&D, a growing number require some degree of cost-sharing with industry. Requirements can also be put in place to mandate collaboration among multiple firms, enhancing knowledge spillovers and strengthening inter-firm networks. In general, however, funding is awarded to a smaller number of firms that would be expected to claim tax incentives, raising concerns about favoritism in making awards. By its nature, direct funding entails a significant role for governments in program administration: budgets must be set, proposals solicited from industry and evaluated according to a specified set of criteria, and awards made and monitored. The costs to government of operating direct funding programs are necessarily higher than for tax incentives and are typically borne by different government ministries (e.g., ministries of industry or research instead of ministries of finance. Considerable cross-country differences exist in the use of tax incentives and direct subsidies for private R&D. Within the OECD, countries with the most generous tax incentive systems tend to be those with the lowest levels of business R&D intensity and with little direct government funding of R&D, including Spain, Mexico, Portugal and Canada. Tax incentives are also used, however, in several countries with high levels of business R&D, including Korea, Japan, France and the United States, although the incentives tend to be less generous. Direct government funding of business R&D totaled almost US$2.9 billion or 7.6% of business R&D in the thirty OECD member countries in 2004. Relative levels of funding vary considerably, ranging from a 0.12% of GDP or more in Sweden, the United States and Korea to less than 0.02% of GDP in Japan and Portugal. Countries at the higher end of the scale allocate much of their funding to defense and health R&D, which may have commercial spill-overs, even if it is not the primary objective. Among the countries that use both direct and indirect mechanisms for financing business R&D, the balance between these two instruments differs considerably. In Australia, Canada and the Netherlands, for example, support provided via tax incentives is greater than that provided via direct funding by as much as a factor of five in Canada. In France, Japan, and the United States, in contrast, direct funding exceeds the cost of tax incentives, often by a wide margin. Differences in the mix of instruments used in these countries reflect different perceptions of the types of failures to be addressed (finance, risk, etc.) as well as different industry and institutional structures. Sectoral targeting While many R&D grant programs are sector-neutral in that they fund projects in all industry sectors, the tendency is toward greater sectoral and technological targeting of programs. Many countries use foresight or other procedures to identify key scientific and technological priorities and channel funds preferentially into those sectors. While broad-based enabling technologies such as information technology, biotechnology and nanotechnology populate many of these lists, more individualized national interests appear as well. Iceland has placed high priority on its fisheries and raw materials industries, Norway on its oil and gas industries, and Korea on digital television, wireless communications and new generation vehicles. Such targeted programs illustrate the role government R&D funding can play in encouraging the development and growth of industries. To date, R&D tax incentives have not been used in this manner, due in large part to their comparative advantage in supporting a broad base of firms in all industry sectors. Of particular interest in recent discussions of sectoral targeting is the service sector. Among OECD countries, services account for more than 70% of total value and are the source of most job creation. While it is often though that service sector firms are not innovative, recent surveys show that in business 2
services, telecommunications and financial intermediation, the share of innovative firms is higher than in manufacturing. Service sector innovation differs from that in manufacturing, however, in that it, to date, it has been driven less by internal R&D and more by the acquisition and adaptation of external knowledge and equipment and by training and education. Many countries are beginning to develop policies to stimulate service sector innovation. In some, such as Denmark, the focus is on the deployment and diffusion of information technology; in others, such as Finland and the United States, new types of R&D programs are being formed that aim to contribute to services innovation, often in conjunction with information technology. Targeting small firms Incentives for business innovation are also tailored to support small firms, both traditional SMEs and entrepreneurial startups. Both types of firms are widely acknowledge to play an important role in national innovation systems, but face serious challenges stemming from limited financial and human resources. For the most part, standard policy instruments for financing business R&D need to be tailored to the needs of small firms and additional policy instruments are needed to provide a policy mix supportive of entrepreneurship. Financing R&D in small firms Several steps can be taken to enable direct funding programs to benefit small firms. Because small firms are often at a disadvantage in competing for government R&D contracts, the United Kingdom and United States have mandated that a specified percentage (approximately 2.5%) of business R&D funding be allocated to small firms. The Netherlands has also put in place a pilot program to test out this approach. In Finnish and German R&D grant programs, special consideration is given to small firms, and as a result, over half of all funding is awarded to them. Many other countries, including Canada, Korea and New Zealand have established funding programs specifically targeted at small firms, and funding has increased for many such programs, even as funding for large firms has stagnated. As a result, small and mediumsized enterprises (those with 250 or fewer employees) account for a disproportionate share of government R&D funds in most OECD countries. Given that tax incentives can potentially be used by all firms that conduct R&D, they are seen as an especially effective way of financing R&D in small firms. Several OECD countries, including Canada, Italy, Netherlands, Norway and the United Kingdom, offer more generous tax incentives to small firms than large ones. Caps on the total amount of incentive that an individual firm may claim further skew these benefits of these incentives toward small firms. One limitation to such incentives is that they do not necessarily benefit early stage startups that have little or no tax burden against which to deduct a credit. To correct this problem, countries have added provisions allowing firms to carry-forward their tax credits to future years (in which they are expected to have a tax burden) or to receive a direct payment from the government equivalent to the value of the credit. In the Netherlands, Belgium and Spain, tax incentives have been designed to offer a reduction not in corporate taxes, but in social charges (payroll taxes) related to R&D workers. Firms benefit from the incentives during each pay period regardless of their profitability. Venture capital support A complementary approach to funding R&D in small firms is to stimulate the development of venture capital markets. Venture capital does not finance R&D per se but provides a source of general financing for firms that tend to be innovative and R&D-intensive. Early and expansion-stage venture capital, in particular tends to finance the activities of small, growth-oriented companies that are active in high- 3
technology or knowledge-intensive industries. While venture capital markets are well-developed in the United States and a handful of other developed countries, they continue to lag in many parts of the world despite the increasingly international reach of many venture investments. As with R&D financing, government support of venture capital entails a mix of direct and indirect policy instruments. Direct measures, such as those used in Iceland and Canada, involve investments of government monies in a venture capital fund, at times in association with private sector investments. Because of the risks associated with such investments and the challenges in developing sound and equitable investment criteria, a number of countries have pursued more indirect approaches to venture capital development, such as through tax incentives. Special tax considerations for monies invested in venture capital funds and for the profits earned by such investments can also stimulate development of venture capital markets, although estimates of the costs of such programs are limited. Financing of venture capital--whether directly or indirectly--addresses only some of the challenges facing entrepreneurial start-up firms. Startups also need an exit strategy in order to provide investors with a means of liquidating their investments. Some countries have attempted to establish small-capitalization stock markets to provide a source of public market funding (going public), but these have seen mixed success due, in part to turbulence in global stock markets in the early part of the decade. The acquisition of a startup firm by a larger firm is another viable exit strategy, but is affected more by policies regarding competition and foreign investment. Fostering public-private collaboration While lack of adequate financing can be a significant obstacle to business innovation, lack of scientific and technical knowledge can also impede innovative efforts. Firms cannot maintain or develop in-house all of the expertise they need to generate new products, processes and services, and inter-firm collaboration may be insufficient to enable them to expand the technological frontiers. Efforts to address this limitation increasingly center on fostering strong collaboration between business and public sector research organizations. Policy mixes for improving collaboration include the following elements: Collaborative research programs that involve researchers in public and private sector organizations. In many cases, governments finance the cost of the public participation in the project, and business is expected to contribute a similar or greater amount. Specific programs, such as those in Korea, aim at providing technical assistance to small firms that lack in-house R&D capabilities. Public-private partnerships that provide a more formalized, structured environment for research that pursues objectives jointly shared by participating public and private researchers and cofinancing of R&D programs. Many such programs are designed to involve multiple organizations and center around specific themes or subject areas (e.g., digital media, biotechnology, mechatronics) Requirements to include small firms can improve knowledge flows to such firms. Reforms to intellectual property rights regimes to promote the patenting and licensing of inventions developed in public research organizations. Most OECD countries have introduced reforms that assign ownership of intellectual property to the public research organizations, rather than the individual researchers, with the expectation that institutions are better equipped and can be better encouraged to license their inventions. Creation of spin-offs from public research organizations to commercialize inventions directly. Permitting public researchers to engage in start-up firms that commercialize their inventions can require modifications to employment policies and other statutes governing public sector workers. It often requires seed funding from government ministries or the public research institution. 4
Experience suggests that a prudent approach to improving public-private collaboration involves the exploitation of all possible channels of collaboration, as each is appropriate to different types of interaction among different types and groups of organizations. Over-reliance on any particular form of collaboration can stifle strong linkages between industry and national science systems. A final word on program evaluation Key to efforts to enhance the effectiveness of incentives and support systems for innovation is regular evaluation. Only through evaluation can policy makers learn which instruments are achieving their desired objectives and which types of reforms are needed to strengthen others. Given the multiplicity of policy instruments needed to create an effective incentive system for business innovation, evaluation tools themselves must be sophisticated enough to distinguish between their different effects on firms. When evaluating policies to finance business R&D, for example, they need to consider not only the effect on aggregate levels of business R&D spending or the numbers of new products or services or patents emerging from supported firms, but on the subtle changes in the way participating firms conduct R&D. Do participating firms tackle more challenging R&D problems? Do they collaborate more with other firms or public research organizations in doing so? Have they developed new approaches to R&D management that increase the efficiency of their R&D process? These types of behavioral changes can be important determinants of overall innovative capability. The extent that they represent learning by firms, they can influence future innovation performance beyond the duration of any particular policy measure. Inducing behavioral such changes can be and often is a specific objective of government policy. By measuring the effects that different types of policies have on business R&D behavior, governments can learn to better differentiate between similar policy instruments and better tailor policy instruments to the particular needs of the business innovation. 5