How much alike are SBIR-like programmes?

Abstract

Promoting innovation in small- and medium-sized enterprises through direct grants for research and development is the main objective of Small Business Innovation Research (SBIR), a programme that has inspired many countries and has grown in importance as part of their current innovation policies. In this article, we conducted a literature review to identify programmes similar to SBIR in other countries and compare them, considering key aspects of their operation, institutional setup, scope, and outcome assessments. Analysing the similarities and differences between the different programmes in eight countries addresses a gap in the literature, which has focused on the isolated analysis of programmes. In this methodological approach, we show how SBIR-like programmes, despite sharing the same main objective, undergo changes resulting from differences in the institutional and political framework, suggesting a process of translation rather than simple diffusion. Furthermore, the study illustrates how SBIR and SBIR-like programmes modify to adapt to changes in innovation policies over time.

1. Introduction

Recognition of the role of small businesses in fostering technological innovations has been spreading for over 30 years, at least since the publication of two influential studies by Acs and Audretsch (1988, 1990). Partly in response to this, as well as evidence having been found, policies stimulating innovation focusing on small businesses have also proliferated (EC—European Commission 2009; Howoldt 2024).¹ Among these policies, one has particularly attracted the attention of both economics of innovation researchers and policymakers: the Small Business Innovation Research (SBIR) programme implemented in the USA in 1982. Two compelling figures underscore the importance of the SBIR programme. By 2021, the programme had allocated almost $64 billion, in current prices, to support 195,000 projects. The SBIR programme is considered not only the main innovation promotion programme for small firms in the USA (Lerner 1999; Wessner 2008; Link and Scott 2018) but also a critical element in reshaping the US innovation system and in the emergence of the network economy (Keller and Block 2013).

The international diffusion of SBIR-like programmes is recognized in the literature (Siegel et al. 2003; GFHR—Global Forum for Health Research 2011; Llanto 2015; Howell 2017). Indeed, the literature review underpinning this article has identified references to SBIR-like programmes in at least twenty other countries: China, South Korea, Japan, India, Germany, the UK, Brazil, Canada, Belgium, Finland, the Netherlands, Sweden, Russia, Australia, Singapore, Israel, Malaysia, Thailand, Taiwan, and Chile. There is also evidence that some of these programmes have achieved a scale comparable to the original American model. As will be seen later, although created much more recently, the Chinese, South Korean, and German programmes have jointly supported over 100,000 projects.

Although there is a vast body of literature on the SBIR programme and numerous studies have been published on some SBIR-like programmes, the literature review did not identify any studies that gather information on these programmes and analyse them from a comparative perspective. The main aim of this study is to address this gap in the literature. While the research collected information on programmes from other countries, the article will focus on the experiences of this kind in the eight countries with the largest economies, according to Gross Domestic Product (GDP) converted by purchasing power parity criteria. Therefore, in addition to the original one, the programmes of the first seven countries listed above will be covered.

Methodologically, the work is primarily based on a review of academic and technical literature concerning the object of study. Regarding the American SBIR and its British counterpart, a systematic literature review was conducted, and from the initial 1,050 documents found, the selection was narrowed to the most relevant studies.² However, attempts to access information through a more general procedure for programmes in other countries yielded poor results. This required an unavoidable case-by-case search across numerous sources such as official websites, consultancy reports, and the much sparser academic literature that addressed these programmes for evaluation purposes.

After this brief introduction, the article proceeds with a concise overview of the literature on the evolution of innovation policies for small- and medium-sized enterprises (SMEs). The following section describes the evolution and key characteristics of the American SBIR. In the next one, the programmes from the other seven countries are discussed. This section highlights the key aspects of their operation, institutional setup, scope, and, wherever feasible, outcome assessments. Afterwards, the preceding information on these programmes undergoes a comparative analysis demonstrating that, despite similarities and the declared inspiration from the original SBIR, there are significant discrepancies in issues such as the organization in charge of the operations, the definition of target audience, amount of resources allocated to the programme and each project, and the reach of their results. Finally, the conclusion combines well-established findings from the broader literature on innovation policies for SMEs with the review of international experience with SBIR-like programmes to reflect on an aspect of their institutional design—integration with mechanisms of public procurement policies—which appears to correlate with the breadth of the results.

2. Support policies for SME innovation

During the 1970s and 1980s, several factors led to shifts in discussions about industrial policies. Alongside probing questions about their effectiveness, heightened by the rise of economic liberalism, the emphasis on the importance of innovation and its systemic nature has grown remarkably (Chang and Andreoni 2016). Policies aimed at improving national innovation systems became widespread in developed countries affected by deindustrialization processes, as well as in developing countries striving to accelerate their economic growth (Soete 2007).

In line with the evolving understanding of the systemic nature of innovative processes (Lundvall 1992; Nelson 1993; Freeman 1995), industrial policies increasingly integrated with science and technology policies, thereby expanding the scope of industrial policies and giving rise to what are referred to as innovation policies. Although these policies vary in terms of names, objectives, scope, instruments, and public resources involved, innovation policies are now prevalent in many countries (Soete 2007). Indeed, according to Edler and Fagerberg (2017: 5), ‘system-oriented policies […] focus, as the term suggests, on system-level features, such as the degree of interaction between different parts of the system; the extent to which some vital component of the system is in need of improvement; or the capabilities of the actors that take part’.

According to Schot and Steinmueller (2018), systemic policies shaped the second frame of innovation policies, following those of the first frame established after World War II, which primarily aimed at expanding research and development (R&D) expenditures at firm and national levels. Systemic policies emerged in the 1980s within a context of heightened globalization. In this environment, policies increasingly focused on building and strengthening interactions among different agents within innovation systems to stimulate the generation of new learning, knowledge, and innovations.

In recent years, industrial and innovation policies have incorporated other themes, particularly issues related to social inequalities (Zehavi and Breznitz 2017) and environmental sustainability (Mathews 2020). Thus, ‘a third frame linked to contemporary social and environmental challenges such as the Sustainable Development Goals and calling for transformative change is identified and distinguished from the two earlier frames’ (Schot and Steinmueller 2018: 1554). The so-called transformative innovation policies (Schot and Steinmueller 2018) or mission-oriented innovation policies (Mazzucato and Kattel 2020) expand the scope and challenges of innovation policies, as their objectives surpass those previously pursued in the first and second frames.

On the other hand, over the past four decades, studies on innovation across various domains have underlined the importance of SMEs, which is the focus of this work, in innovation activities (Acs and Audretsch 1990; De Jong and Marsili 2006; Hölzl 2009). These studies, using different methodologies and applied to several countries, highlighted the importance of small businesses in innovative activities, which can only be identified satisfactorily when these activities are no longer measured solely by R&D expenditure or patents.³ This is the case of the Community Innovation Survey and others with methodologies inspired by the Oslo Manual, dating back to the 2000s. The following studies helped identify the specificities of innovative activities undertaken by small businesses, which tend to be less concentrated in R&D expenditure than large companies (‘non-R&D innovation’). Empirical studies have identified different obstacles to innovation, such as costs, market conditions, lack of qualified human resources, difficulty in establishing external partnerships, and lack of institutional support for innovation. However, SMEs are more affected by the high costs of innovation, compared to large companies (Pellegrino and Savona 2017; de-Oliveira and Rodil-Marzabal 2019; Arza and López 2021).

Therefore, the contribution of SMEs to innovation tends to occur in less concentrated sectors with high technological opportunities, the so-called Schumpeter Mark I pattern, which emphasizes the importance of entrepreneurship and new firms for economic dynamics, found by Schumpeter (1911). In this pattern, there is great relevance of new innovators, related to low degrees of cumulativeness and appropriability, and high importance of applied sciences and external sources of knowledge. Conversely, the so-called Schumpeter Mark II pattern, which refers to the systemic view of innovation, disseminated by Schumpeter (1942), focuses on the importance of R&D laboratories and large companies for innovation (Breschi, Malerba and Orsenigo 2000).

The extent of innovation developed by small firms has had major implications for industrial and innovation policies, particularly in high-tech sectors. In this context, innovation policies have increasingly encompassed support for SMEs as a cornerstone (EC—European Commission 2009; Howoldt 2024). In addition to more traditional actions aimed at providing or facilitating credit for SMEs, new instruments have emerged, especially those that foster interaction between these firms and other entities within innovation systems, such as universities and research centres.

One noteworthy aspect is the recognition of heterogeneity that characterizes the SMEs segment. As well as their significant participation in more traditional sectors with low entry barriers, small businesses in high-tech sectors, especially those experiencing rapid growth, stand out for their high impact on job and income generation. Based on empirical evidence for European countries, Malerba and McKelvey (2020) show that the so-called knowledge-intensive innovative entrepreneurial firms are important phenomena, found in all sectors of activity and in all European countries.

Therefore, the topic of innovative entrepreneurship has gained prominence in innovation policies. Despite the enormous challenges concerning support for these businesses, given that just a few overcome the difficulties of the initial development stages (liabilities of newness and smallness) and thrive, support policies focused on entrepreneurship have spread rapidly over the past 20 years (Audretsch et al. 2020).

When designing current innovation policies from a systemic perspective, the choice of instruments is crucial since ‘rarely are innovation policy instruments ready or “prêt- à-porter” for the task at hand. Most of the time, if not always, policy instruments must be designed, re-designed, and adapted to the specific problems in the innovation systems and their uses. Instrument design can change over time according to changing preferences, changing objectives, and changing problems in the innovation system’ (Borrás and Edquist 2013: 1522).

Policymakers have a broad set of instruments to stimulate SMEs innovation activities, considering the specificities and heterogeneity of this group of firms. These instruments include tax incentives, subsidies and government grants, workforce training programmes, support for partnerships with universities and research centres, business management support, bureaucratic simplification, programmes aimed at forming clusters and networks among firms, as well as consolidating incubators and science parks (Bloom, Van Reenen and Williams 2019; Audretsch et al. 2020; Hottenrott and Richstein 2020; Nguyen et al. 2023). In addition to the mentioned instruments, Nguyen et al. (2023) also consider that government support for the technological development of firms can be provided by introducing quality assurance programmes, enhancing their capacity to introduce process innovation. According to Ferraz, Kupfer and Marques (2014: 294), in the context of open economies, it is essential for public policy to have a wide range of instruments, i.e. ‘it is a strategic requirement to pursue public policies that make effective and efficient use of all available tools—horizontal, selective, and other policy instruments—to induce industrial transformation’.

The Organization for Economic Co-operation and Development (OECD) (OCDE 2022) proposed a taxonomy of industrial policy instruments, categorizing them into supply-side and demand-side instruments, highlighting the potential complementarities between them. Supply-side instruments encompass those that affect firm performance (such as tax and financial incentives) and those that impact industrial dynamics (such as entrepreneurship policy, intellectual property, and technical standardization policy, among others). Demand-side instruments, in turn, include those that affect demand, such as product regulation and public procurement. Furthermore, the proposal includes a transversal aspect called governance, which interrelates with both supply-side and demand-side instruments and involves agent coordination.

Based on a systematic analysis of experiences from various countries, Howoldt (2024) confirms the existence and use of a wide range of innovation policy instruments. Drawing on research conducted by the OECD from 2001 to 2020 and data from 4,710 policies, a significant heterogeneity of experiences was identified. Policymakers select instruments with two main goals: addressing weaknesses in the innovation system and leveraging the strengths of the country’s innovation framework.

According to Hottenrott and Richstein (2020), despite the diversity of available instruments, the most widely used internationally are financial incentives: government subsidies and grants. By and large, opting for these instruments improves the risk-return profile for SMEs, especially for startups. While they are the most used, in line with Ben-Ari and Vonortas (2007) and Audretsch et al. (2020), the authors identified that these financial instruments have a wide range of formats.

Bloom, Van Reenen and Williams (2019) highlight the importance of innovation policies aimed at SMEs while underscoring some potential issues they may pose. Firstly, they point out that these policies might inadvertently discourage business growth, as firms become ineligible once they reach a certain size. Secondly, they argue that firms most in need of support are not necessarily small ones, but rather young ones. Young, small-sized, and technologically intensive firms—startups—often face significant financial constraints that influence their investment decisions and hinder their ability to engage in R&D activities. The authors stress the need to implement public support programmes specifically tailored to startups, particularly in the early stages of their lifecycle. Furthermore, regarding the effectiveness of these diverse policy instruments, Bloom, Van Reenen and Williams (2019) state that, based on experiences in various countries, the most effective programmes for SMEs are financial incentives such as government subsidies.

In addition to the importance attributed to government incentives and the volume of resources involved in the programmes, studies also point to encouraging the formation of networks and alliances as a critical factor for the success of the programmes. Access to external knowledge is of utmost importance for SMEs, which generally have financial constraints to develop more ambitious R&D programmes (Audretsch et al. 2020; OECD 2023). Current programmes aimed at stimulating entrepreneurship in cutting-edge technology sectors aim precisely to help overcome these obstacles by subsidizing interactions between universities or research institutes and firms. These interactions stimulate the absorption capacity of firms, especially of scientific knowledge provided by academic institutions, and the innovation processes in SMEs (Kurdve, Brid and Laage-Hellman 2020; Bucaioni, Murgia and Taalbi 2024). The design of the instruments that enable these interactions are, in turn, strongly conditioned by regulatory, political, economic, and social factors, that is, by factors linked to the characteristics of national innovation systems. Thus, the programmes present variations in objectives (broader or narrower), institutionality, target audience, and sectors covered, as will be seen in the following sections.

3. The American SBIR

Building on initiatives by the National Science Foundation (NSF) in 1977 and the Department of Defense (DoD) in 1980, the SBIR was established by the Small Business Innovation Development Act in 1982 and was then extended in 1986, 1992, 2000, 2011, 2016, and 2022, with short-term reauthorizations between 2008 and 2011. The programme was created in an environment marked by the competitive challenge Japan posed to the American economy, and it emerged amidst recognition of the role of small businesses in innovation dynamics (Allen, Layson and Link 2012). It was based on the belief that US scientific and technological prowess could leverage its economic competitiveness (Keller and Block 2013). On the other hand, Siegel et al. (2003) contextualize SBIR implementation within a framework of initiatives aimed at fostering technology transfer from universities and public research laboratories: the Bayh-Dole Act (1980), the National Cooperative Research Act (1984), and the Small Business Technology Transfer Act (1992).

The initial programme purposes were: (a) stimulate technological innovation, (b) use small businesses to meet federal R&D needs, (c) foster participation by minorities and disadvantaged persons, and (d) increase private sector innovations derived from Federal R&D. In 1992, the legislation included objectives to increase women’s participation and reinforced private-sector commercialization goals (Onken et al. 2019). Supported businesses must have no more than 500 employees, and the majority control must be held by US residents, although since 2011, control by venture capital (VC) funds has also been allowed.

The SBIR programme is decentralized and operated by 11 federal agencies engaged in R&D activities. The five largest agencies typically account for 96%–97% of the expenditures (Allen, Layson and Link 2012; Sargent 2015). The main ones among these are the DoD, responsible for 51% of the total, and the National Institutes of Health (NIH), contributing 30%. Following in importance are the National Aeronautics and Space Administration (NASA), the Department of Energy, and NSF programmes. While these agencies adhere to policy guidelines set by the Small Business Administration, which coordinates, monitors, and oversees the programme, they prioritize projects aligned with their respective missions (Sargent 2015). This arrangement fosters flexibility but also poses challenges, such as the potential for duplicate funding of the same project. Keller and Block (2012: 641) indicate that ‘when the SBIR programme began, government agencies other than NSF were initially hostile. (…) However, attitudes in many agencies shifted [once] administrators came to see that small firms were often able to deliver new capacities more quickly than large, established contractors’.

A significant development of the SBIR was the creation of the Small Business Technology Transfer Program (STTR) in 1992. STTR specifically aims to support projects conducted cooperatively between businesses and researchers affiliated with research institutions. Only the five agencies with external R&D expenditures exceeding $1 billion per year are required to implement STTR programmes.

Successive renewals of the SBIR expanded its funding allocation, which comprises a portion of the extramural research expenses of federal institutions whose research programmes exceed $100 million annually. Initially at 0.2%, the set-aside portion was increased by the 1992 legislation to 1.25% and later to 2.5%. After remaining unchanged for 15 years, the 2011 legislation gradually increased the allocation to 3.2% by 2017. The initial allocation for the STTR was 0.3%, but starting in 2012, this percentage gradually increased to reach 0.45% by 2016. Since the onset, the programme has structured projects into three phases. The first phase, lasting 6 months, focuses on evaluating scientific and technological merit and the proposal feasibility. The second phase, lasting up to 2 years, addresses actual research development. The third phase involves completing development. While funding for Phases 1 and 2 is provided by SBIR, firms are expected to secure private resources or a supply contract for Phase 3 with the agency itself (Wessner 2008).

The grant limits were $50,000 for Phase 1 and $500,000 for Phase 2, but these were readjusted over time. In 2021, the amounts reached $275,000 and $1,838,000, respectively. Agencies are also legally authorized, at their discretion, to exceed these caps by up to 50% and to go even beyond that limit (Link and Wright 2015; Audretsch and Link 2019). Furthermore, agencies can award a new grant following Phase 2 to continue R&D work. This is exemplified by the NSF, which offers Phase 2B grants of $50,000 and $500,000, provided the applicant can demonstrate having obtained $2 of third-party investment for each dollar granted by the agency (Wessner 2008). Therefore, the funds obtained for a project in Phase 2 can reach $3 million or more. Additionally, ‘some agencies may use non-SBIR funds for Phase 3 funding to support R&D or contracts for products, processes, or services intended for use by the federal government’ (Sargent 2015: 7–8).

Apart from the availability of additional funding mechanisms, there are other significant differences between the programmes of each agency. Some of them, such as the NIH and NSF, define the topics of their calls more broadly, while others (DoD and NASA) are much more specific. These various particularities reflect the agencies’ objectives and the characteristics of the technologies they support. Some agencies engage in extensive procurement of products and services, while others do not. In ‘procurement’ agencies such as the DoD and NASA, the SBIR programme is ‘focused primarily on developing technologies for the agency’s own use’ (Wessner 2008: 45).

Despite technology commercialization being the primary objective of the SBIR from the onset, there has been recurring concern about reinforcing it. In the 2011 reauthorization, this translated into measures to monitor firms’ success in progressing through phases, including restrictions on project submissions from companies that did not meet minimum performance standards. The aim was to curb the operations of so-called ‘SBIR mills’, companies that secure successive programme grants but do not demonstrate genuine commitment or capability to commercialize their research outcomes.

On the other hand, agencies were authorized to establish commercialization readiness pilot programmes. Allocating up to 10% of their SBIR and STTR programme resources, agencies would support ‘technology development, testing, evaluation, and commercialization assistance [of] Phase 2 technologies [or] the progress of R&D and commercialization to Phase 3’ (Sargent 2015: 28). Adopting Commercialization Assistance Programs was also encouraged, contracting agents to provide counselling, mentoring, and networking activities aimed at commercialization (Chavez 2015).

According to Sargent (2015), by 2011, SBIR and STTR had already allocated $33.7 billion in support for over 133,000 projects. A decade later, the cumulative number of projects neared 195,000, with disbursements totalling $63.7 billion. In real 2021 US dollars, this amount would exceed $81 billion. Although numerous projects have been funded, there is evidence that the programme’s selection process is rigorous. From 1994 to 2005, the success rate for Phase 1 project submissions hovered around 15%, and only 40% of Phase 1 grant recipients progressed to the next phase (Wessner 2008).

The SBIR is considered the main programme supporting innovation in small businesses in the USA, as noted by official sources (Wessner 2008; NAS 2015) and academic experts (Lerner 1999). Keller and Block (2013: 649) even characterize it as ‘a critical catalyst for the growth of the small firms at the heart of the emerging network economy’. The programme has supported companies that would later become technology giants such as Apple, Intel, and Qualcomm (Audretsch, Weigand and Weigand 2002), and compelling evidence of significant outcomes has led various countries to attempt to emulate the SBIR experience (Lanahan and Feldman 2015; Howell 2017).

SBIR evaluation is the subject of a vast body of literature, including comprehensive studies commissioned by US government agencies and numerous academic works that have leveraged the wealth of data gathered by the programme to analyse not only the programme itself but also various characteristics of its predominantly technology-based clientele. It is not within the scope of this article to review this literature in depth, but it is important to highlight that the most comprehensive evaluation concluded that ‘the SBIR program is sound in concept, effective in practice, [and] is delivering results that meet most of the congressional objectives’ (Wessner 2008: 3).

4. SBIR-like programmes in other countries

This topic addresses SBIR-like programmes created worldwide. Regarding the purposes of this article, public policy programmes are considered similar to SBIR if they encompass three characteristics: (1) targeting SMEs; (2) aiming to promote innovation by carrying out R&D projects; and (3) employing support instruments similar to grants with non-repayable funds.

We found SBIR-like programmes in at least twenty countries (Siegel et al. 2003; GFHR—Global Forum for Health Research 2011; Llanto 2015; Howell 2017) and obtained robust information on SBIR-like programmes from twelve of these countries, collectively representing over 40% of global GDP in 2020. Including the USA, this accounts for two-thirds of the world’s GDP. However, due to space constraints, only the seven countries with the largest GDPs will be considered. Detailed reviews will be conducted in specific sections for China, Japan, South Korea, India, Germany, the UK, and Brazil.

It should be noted that we do not claim that, despite being the result of our best efforts, the following review is exhaustive. The information we have gotten to gather varies significantly across countries and programmes. In some cases, detailed operational information about the programme is available from institutional sources, but no evaluation of the programme was found in the literature in English. In other cases, particularly in East Asian countries, official information obtained is very limited, but there are well-conducted evaluation studies available. Space constraints also require making choices regarding the reach of the review of these programmes.

4.1 China⁴

In May 1999, the Chinese central government established the Innofund (Innovation Fund for Technology-Based Firms), a programme aiming to ‘encourage the innovation activities of small and medium technology-based enterprises and commercialization of research by way of financing, trying to bring along and attract outside financing for [their] corporate R&D investment’⁵ (Guo, Guo and Jiang 2016). The programme targets enterprises with up to 500 employees and requires compliance with 3 additional criteria: an annual investment in R&D of at least 3% of revenue, and a workforce where 30% or more have higher education and 10% or more of employees are engaged in R&D (Guo, Guo and Jiang 2016). Wang, Li and Furman (2017) pointed out an additional criterion that at least 50% of the company’s capital must be controlled by Chinese citizens, with a stricter threshold of 5% for R&D intensity. Projects supported by Innofund must demonstrate high potential for competitiveness, as well as economic and social benefits.

According to Guo, Guo and Jiang (2016), the resources allocated by Innofund typically range between RMB 1 million and RMB 2 million ($160,000 to $320,000 at the exchange rate on 30 November 2021). Innofund often utilizes other support mechanisms in addition to grants, such as interest-free loans and equity participation. Grants are the preferred method for supporting academic spin-offs and subsidizing the development of new products.⁶ Interest-free loans are primarily used for scaling up the production of innovative projects, while VC investment, limited to 20% of the firm’s share capital, ‘is reserved for projects that use advanced technology, have high innovation capacity, and have market potential in emerging industries’ (Guo, Guo and Jiang 2016: 1131).

A survey of the 30,357 projects supported between 1999 and 2011 revealed that 27,498 received grants, 2,880 obtained interest-free loans, and 1,159 received other forms of support, including equity participation and credit insurance. Data presented by Wang, Li and Furman (2017) suggest that between 2006 and 2012 the programme’s budget multiplied by six, the number of supported projects quadrupled, and the success rate—the ratio of approved projects to submitted ones—more than doubled from 30% to 62%. The total budget of Innofund from 1999 to 2012 amounted to RMB 22.09 billion, equivalent to approximately $3.18 billion, based on average exchange rates for each year.

The agency responsible for operating the programme, the Innofund Administration Centre, is subordinate to the Ministry of Science and Technology. The selection process involves an expert panel comprising at least four technical specialists and one financial expert. After individual evaluations, the experts provide a unified score for each project in both assessment areas. To mitigate financial risk, a score below 60 in this criterion results in disqualification. Following evaluation, projects are ranked based on scores and supported within budgetary limits (Wang, Li and Furman 2017).

Programme management underwent significant changes in 2005 seeking simplification and decentralization. According to Guo, Guo and Jiang (2016), these changes have enhanced the decision-making process by harnessing local officials’ greater familiarity with companies and incentivizing provincial governments to improve project selection and monitoring.

Official evaluations of the programme are highly favourable. Each yuan spent on the programme would have catalysed another eleven from local governments, banks, and VC funds. From its onset until 2011, the programme is said to have generated 450,000 jobs, RMB 209 billion in revenue, and RMB 22.5 billion in taxes. Perhaps more significantly, 82 out of 273 companies listed on China’s stock exchange for smaller enterprises had been supported by Innofund, including the prominent Huawei (Guo, Guo and Jiang 2016).

Independent evaluations, however, yield divergent results. On one hand, Guo, Guo and Jiang (2016), using matching techniques to form a control group, argue that supported companies not only outperform their counterparts in terms of revenue from new products, exports, and patents obtained but also improve their own performance in each of these aspects after receiving programme support. On the other hand, Wang, Li and Furman (2017: 1142) acknowledge that ‘firms receiving high project evaluation scores and Innofund grants perform better than those that do not receive grants and have lower scores’ but ultimately attribute this discrepancy to selection bias and doubt that it was generated by the programme itself.

4.2 Japan⁷

As part of the initiatives to address the decline in dynamism in Japan’s economy, the Temporary Law Concerning Measures for the Promotion of the Creative Business Activities of Small and Medium Enterprises was established in 1995, providing a legal basis for supporting innovative SMEs. To a certain extent, it foreshadowed the reformulation in 1999 of the General Law for Small and Medium Enterprises, which shifted Japan’s policy orientation in this field ‘from protecting disadvantaged SMEs to assisting innovative self-motivated business’ (Eshima 2003: 86).

Similar to its American counterpart, the Japanese programme focuses on supporting R&D in SMEs and it is structured in phases. Its implementation is also decentralized, involving ministries and government agencies in areas such as science, health, social welfare, defense, transportation, agriculture, and economy⁸ (Inoue and Yamaguchi 2017). The programme targets companies with up to 300 employees or a maximum capital of ¥300 million ($2.66 million at the exchange rate of 1 December 2021).

Its governance features additional specific characteristics. Phase 1 projects must be submitted to local governments and cover a business plan and an R&D project. Selection is jointly made by local authorities and the Small and Medium Enterprise Agency, which is the central government agency responsible for the programme. In Phase 2, various support instruments can be mobilized, including R&D grants, subsidies for upgrading production facilities, tax reductions for fixed capital investments, low-interest loans, credit insurance, and public VC funding. A survey covering the programme’s initial 3 years (1995–7) shows that while R&D subsidies were more often used (29% of projects and 19% of the amount), VC investments (9% of projects and 23% of the amount) were more substantial. Nonetheless, credit insurance surpassed both mechanisms: it was employed in 38% of cases and involved 49% of the resources (Eshima 2003). In a more recent study, Inoue and Yamaguchi (2017) added patent registration fee exemptions and opportunities to access public procurement bids to this list.

A different programme entitled the ‘Strategic Basic Technology Advancement Support Project’ intends to facilitate cooperation between SMEs and universities, research institutes, and their industrial clients for R&D in manufacturing technologies, prototype development, and market development. The research execution period is 2–3 years, and the programme funds up to two-thirds of the expenses, capped at ¥97.5 million over 3 years (approximately $865,000 at the 1 December 2021 exchange rate).⁹

According to a study by OECD (2010), in 2007 there were 1,365 supported projects, slightly more than the 1,242 supported in 2003, but fewer than the 1,494 reported by Eshima (2003) for 1997. However, expenditures on projects grew by a significant 42% between 2003 and 2007, and another 22% from there to 2011, reaching ¥45.1 billion ($565 million at the average exchange rate of the year) (Charumilin 2012). This figure implies that the programme in 2011 was four times smaller than its US counterpart.

As in the case of the Chinese programme, independent evaluations have reached conflicting conclusions. Eshima (2003) used matching methods to compare firms that received support from the programme in 1995 and 1996 with similar non-beneficiary firms and identified higher growth rates in employment and revenue among the supported ones. Nonetheless, using similar methods for companies with approved projects between 2004 and 2006, Inoue and Yamaguchi (2017) concluded that the performance of supported companies was weaker in terms of employment growth, revenue, and even the number of patents. They attribute the inferior results compared to the US programme to several relative deficiencies of the Japanese programme and innovation system: weaknesses in the mechanisms for promoting commercialization, VC fragility, and lower capabilities of Japanese universities.

4.3 South Korea¹⁰

The Korea Small Business Innovation Research (KOSBIR) was established in 1998. Similar to the US SBIR programme, the initiative draws its funding from public agencies’ budgets. Twelve government agencies and six state-owned enterprises were mandated to allocate at least 5% of their R&D budget to support SMEs, but in practice they allocate a much higher proportion. During the triennium 2009–11, the share of these entities’ R&D budgets dedicated to the programme collectively ranged between 9.6% and 10.5% (Llanto 2015).

Since the programme started, its budget has been considerably expanded. Lee and Jo (2018: 49) stated that ‘the expenditure for SME-operated government R&D projects reached 2,897 billion won in 2016, equivalent to 15.2% of the government’s total R&D investment amount and similar to the US SBIR’s total grant amount’. Indeed, this amount equals $2.495 billion, slightly below the $2.675 billion that the US SBIR rules would have mandated as ‘obligatory’ for that year. Between 1998 and 2013, expenditure on the programme increased five-fold annually, totalling $13.2 billion (Hong 2015). Considering such a significant volume of resources, the programme supported 30,448 projects from 2010 to 2014 (Lee and Jo 2018).

The programme has been operated since July 2017 by the Ministry of SMEs and Startups. It supports SMEs, which in South Korea are defined specifically based on their operation sector. According to Lee and Jo (2018), although the programme ostensibly involves a two-phase structure similar to its American counterpart, there is a strong tendency to omit the feasibility demonstration stage and proceed directly to project execution support. Indeed, 80% of the projects received more than KRW 100 million ($85,000 at the exchange rate on 2 December 2021), and half of them received more than double that amount.

Apart from KOSBIR, support for innovative SMEs involved several other instruments, such as business incubators, credit guarantee programmes, public investment in VC funds, preference for government procurement, and the promotion of cooperative projects with universities and research institutes (Kim 2007; OECD 2010; Llanto 2015). For instance, the SME Technology Innovation Development Program sought to increase the competitiveness of SMEs with already established R&D capabilities by developing new products or processes. Lilischkis (2011: 54) emphasizes the programme’s focus on ‘pragmatic developments which can be commercialized within three years’, while Llanto (2015) indicates that it covers 75% of the costs.

Lee and Jo (2018) carried out the only KOSBIR evaluation found. The authors showed that supported companies performed better in terms of R&D investment and intellectual property indicators. However, they did not show a better performance in the indicator that summarizes economic outcome, the value added by the firm. Concerning this, the performance of these companies was, on the contrary, inferior to that of their peers. The main explanatory factors for this adverse performance were ‘an ineffective recipient selection system that relies solely on qualitative assessments by technology experts’ and an ex-post assessment process that emphasizes indicators of technological effort, patenting, and scientific publications (Lee and Jo 2018: 47, 62).

4.4 India

The Small Business Innovation Research Initiative (SBIRI) is a programme of the Biotechnology Industry Research Assistance Council (BIRAC), a public enterprise linked to the Department of Biotechnology of the Ministry of Science and Technology in India. Launched in 2005, this was the country’s first public–private partnership initiative of its kind. Focusing on biotechnology innovation, SBIRI aims are strengthening companies whose product development is based on internal R&D, encouraging SMEs to enhance their R&D capabilities, creating opportunities for new businesses by scientists, and stimulating technological innovation. To achieve these objectives, the programme provides support for early-stage business research, prior to proof of concept, aiming to forge appropriate links between entrepreneurs, academia, and government.¹¹

To be eligible for the SBIRI, the control of the majority of the firm’s shares must be held by Indian citizens holding Indian passports, excluding citizens residing abroad. The company must have facilities suitable for project implementation. Project submissions occur during three calls each year. The evaluation process involves some steps: (1) assessment by referees specializing in the knowledge area, (2) review by a committee comprising eminent scientists and financial analysts, and (3) final decision by a committee of qualified technical experts from various government bodies. Evaluation criteria include technical quality, clarity of strategy, potential for creating and commercializing new technology, social and national relevance, project team competence, and the company’s previous success in other projects and technology commercialization (BIRAC 2020).

Although there is no cap on the project value, the SBIRI grant fully covers project costs only up to 5 million rupees ($67,000 at the exchange rate on 1 December 2021). Exceeding this amount, the programme covers half of the costs. A unique aspect of the programme is that the company must pay BIRAC a 5% royalty on the net sales of the developed product, capped at the amount of the grant received from the programme. Additionally, the programme stipulates BIRAC’s right to compulsory licensing in specific cases of ‘nationally important projects’. As of 2020, there were 42 project calls, totalling 1,938 proposals received, 295 projects supported, and 2.72 billion rupees in funding. Moreover, 56 products were developed, and 35 intellectual properties were generated (BIRAC 2021).

BIRAC operates more than a dozen other innovation support programmes that cover the entire technology development cycle, from exploratory research by students to VC for startup consolidation. This includes a well-established programme promoting entrepreneurship among scientists (BIRAC 2021).

4.5 Germany

The Central Innovation Program for Small and Medium-sized Enterprises (ZIM—Zentrales Innovationsprogramm Mittelstand) is an initiative of the Ministry for Economic Affairs and Energy aimed at enhancing the innovation capabilities of SMEs. According to its official website, it is the largest innovation programme for this type of firm in the country. Created in 2018 by consolidating various existing programmes, ZIM provides non-repayable grants to companies seeking to ‘develop new or significantly improve existing products, processes or technical services, [funding projects which are] highly innovative, market-oriented and entail a substantial technological risk’.¹² In the 10 years from the original initiative to June 2018, the programmes merged to form ZIM allocated €5.5 billion in public resources to approximately 40,500 projects with a total budget of €10.3 billion (Kaufmann et al. 2019).

In addition to (1) feasibility studies and (2) research projects supported in most countries that have implemented SBIR-like programmes, ZIM also covers (3) cooperative projects and (4) innovation networks. The operational rules and access conditions vary for each of these four modalities of the programme. Feasibility studies can extend up to 8 months, and a company can have two approved studies per 12 months. Eligibility criteria include having fewer than 500 employees, a qualified team for project execution, and restrictions on recurring support. Therefore, except for micro-enterprises (fewer than ten employees) and young companies (founded less than 10 years ago), firms cannot apply for support if they have received funding from ZIM or any other public funding programme in the previous 3 years. The overall funding cap is €100,000 for individual projects and €200,000 for cooperative projects, with an additional restriction on ‘non-technical’ items of €30,000 and €40,000, respectively. Cost coverage ranges from 60% for firms with 50–250 employees to 70% for firms with fewer than 50 employees.

Feasibility studies can precede individual or cooperative R&D projects but are not a prerequisite for submitting these projects. Individual R&D projects can be submitted only by companies with fewer than 500 employees. In this context, the funding cap reaches €550,000, with cost coverage ranging from 25% to 45%, depending inversely on the size of the company. The programme also finances the development and commercial launch of the resulting product from the project. With limits of 50% of expenses and €60,000, this covers participation in trade fairs, advisory marketing services, product design, innovation support services (market research, testing, and certification), and protection of intangible assets, among other items.

Cooperative R&D projects must involve at least two companies or one company and one research institution in a partnership where all parties provide innovative services. Firms with 500–1,000 employees are eligible, provided they are in projects that include firms with fewer than 250 employees. The grant is subject to a limit of €2.3 million, with each company eligible to receive a maximum of €450,000 and research institutions €220,000. Once again, the portion of expenses covered by ZIM decreases with the size of the company, ranging in this case from 55% to 30%. Moreover, in cooperative projects, there are resources for commercial development.

ZIM also includes support for innovation networks, which must involve a minimum of six companies and one organization responsible for network management. This line is subject to supporting limits similar to those applied to cooperative projects but with a longer-term perspective structured in two phases. Most of the programme is managed by independent companies acting as administrators.

Among the other programmes promoting innovation in SMEs in Germany, KMU-Innovativ should be mentioned. Operated by the Ministry of Education and Research since 2007, it targets the medium-sized firms characteristic of the German economy (Mittelstand). The programme allocated €1.7 billion in its first 10 years to 2,380 individual or joint projects, involving 3,880 companies. Unlike ZIM, projects supported by KMU-Innovativ are guided by a list of priority technological fields.

4.6 The UK¹³

The British SBRI aims to (1) stimulate innovation by supporting companies in developing new products based on science and technology, and (2) address government agencies’ needs for new solutions to their operational challenges. While its early initiatives date back to 2001, the programme gained momentum after being revamped in 2008, when the agency ‘Innovate UK’ took over its coordination, and especially after the setting of resource allocation goals in 2013 by the departments responsible for its execution (Connel 2017; Steer-ED 2022). From October 2008 to October 2016, 22 government departments issued project calls under SBRI, with five of them accounting for nearly 80% of the resources (Connel 2017; MIIR—Manchester Institute of Innovation Research (MIIR) 2017). In addition to Innovate UK, which contributes to funding in some cases, key areas include national defense, health, energy, and climate change.

Similar to its American counterpart, the SBRI programme is divided into phases. The first phase aims to establish project feasibility (lasting 6 months) and receives an investment ranging from £50,000 to £100,000. In Phase 2, which focuses on developing, testing, and demonstrating a prototype, the maximum duration is 2 years with funding capped at £1 million (MIIR—Manchester Institute of Innovation Research (MIIR) 2017). In both cases, SBRI covers 100% of the project budget. The third phase, commercialization-driven, is expected to be carried out without programme funding. However, as in other countries, commercialization often encounters challenges that some departments, such as the NHS—Health, face with their own programmes, separate from SBRI.

An important specificity of SBRI is that, despite its name, the programme is not limited to small businesses. MIIR—Manchester Institute of Innovation Research (MIIR) (2017) estimates that 62% of its resources between 2008 and 2016 were allocated to firms with up to 250 employees, but a significant 25% were allocated to larger ones. The remainder went to non-profit organizations, especially universities. Moreover, mature companies prevail in the calls for proposals and, even more so, among those receiving support from the programme. The proportion of firms over 20 years (among participants in the calls) was more than twice that of those under 5 years old (MIIR—Manchester Institute of Innovation Research (MIIR) 2017).

The SBRI is characterized as a public procurement programme, specifically a pre-commercial procurement programme. Connel (2017) emphasizes that the SBRI awards contracts rather than grants to the supported organizations. Its operation is based on calls made by public agencies, which define the technological problems to be addressed. The calls are highly competitive, with a success rate that averaged 26% during 2008–20 (Steer-ED 2022).

From 2008 to April 2020, the SBRI allocated £788 million to over 4,000 projects (Steer-ED 2022). Comparing these figures with data from the first 9 years of this period, as reported in MIIR—Manchester Institute of Innovation Research (MIIR) (2017) and Connel (2017), reveals an intensification in the programme’s operations in recent years. However, the forecasted expenditure for the fiscal year 2014/15 (£200 million) was never achieved, despite targets for the six main departments involved in the programme have been set (Yeow, Rigby and Li 2017). Indeed, a recent evaluation report concludes that ‘the key challenge for SBRI is […] the limited traction it has had within many parts of the public sector’ (Steer-ED 2022: 10).

The most recent evaluation of the SBRI reports that 42% and 21% of beneficiary firms surveyed believed the programme contributed to increases in employment and profit, respectively. Besides, 31% indicated that the developed product had been commercialized. An econometric analysis using a counterfactual approach confirmed positive impacts on employment but not on revenue. This analysis also formed the basis for estimating a benefit-cost ratio, which, focusing solely on the impact on firms, ranged from 1.53 to 4.07 (Steer-ED 2022).

4.7 Brazil

The most consistent and enduring initiative similar to SBIR¹⁴ in Brazil is the PIPE (Pesquisa Inovativa em Pequenas Empresas), a programme created in 1997 and conducted at a subnational scale by the São Paulo state government agency¹⁵ responsible for supporting scientific research. The programme offers non-repayable resources and follows the well-known three-phase structure. The first phase, with a maximum duration of 9 months, focuses on the technical-scientific feasibility analysis of the proposal and can receive up to R$200,000 ($36,000 at the exchange rate on 1 December 2021).¹⁶ The second develops the research, which must be executed within 24 months with programme resources limited to R$1 million ($178,000). Besides projects that have succeeded in Phase 1, the programme supports more mature projects through a ‘direct Phase 2’. The third phase, aimed at commercial scale development, ideally should be executed using resources external to the programme. Although oriented towards small businesses—a cap of 250 employees—due to statutory restrictions, the programme allocates resources to the researcher proposing the project, who could be one of the owners of the firm or someone appointed by them.

The inspiration for the programme in the American SBIR is acknowledged (FAPESP—Fundação de Amparo à Pesquisa do Estado de São Paulo 2017) and reflected in some of its developments. In 2016, an entrepreneurship training and business model enhancement scheme similar to the NSF’s I-Corps SBIR was created (FAPESP—Fundação de Amparo à Pesquisa do Estado de São Paulo 2017). On the other hand, PIPE-TC is a similar programme to the American STTR and focuses on partnerships between small businesses and academic institutions.

Operated by a research funding agency analogous to the American NSF, PIPE typically operates through project calls without restrictions on technological themes. In this context, challenges related to the commercialization phase are particularly prominent, leading to the development of several mechanisms to address them. Since 2004, these include funding opportunities for Phase 3 projects with subsidies from the federal innovation agency. Additionally, starting in 2020, PIPE-Invest has provided supplementary resources to sustain research continuity for high-potential projects that demonstrate matching funds from private investors.

Fapesp reports (FAPESP—Fundação de Amparo à Pesquisa do Estado de São Paulo 2017, 2021) show that up to 2020 PIPE supported 2,695 projects from 1,633 companies, allocating a total of $235 million (converted using the average exchange rate for each year). During the period from 2016 to 2020, 6.9% of Fapesp’s funding disbursements were directed towards the PIPE programme.

The programme underwent two comprehensive evaluations. The first covered projects completed by 2006 and involved a questionnaire answered by those responsible for 214 out of 338 projects. The main findings were that 29% filed patent applications, 40% generated revenues, and 52% obtained additional funding from other sources, proportions very much alike to those found in the American SBIR case. Furthermore, based on the relationship between project revenues up to the survey time and the funding allocated by Fapesp, a cost-benefit ratio of 5.98:1 was found (Salles-Filho et al. 2011). A second round of evaluation targeted projects initiated from 2006 to 2014. The new survey obtained responses from 185 out of 400 such projects. Adopting a counterfactual strategy, it also included respondents responsible for rejected projects, yielding 296 (11%) valid responses from 2,700 projects. Among the results, a notable finding was that while the overall growth in the total number of employees was similar in both groups of firms, the R&D personnel grew at a rate 5.7 times higher in companies whose projects were approved (Salles-Filho et al. 2019).

5. Comparative analysis of the programmes

The main results of the survey on SBIR-like programmes are summarized in Table 1. Reviewing international experience shows that since the mid-1990s, several countries, inspired by the programme implemented in the USA in the previous decade, have implemented programmes that share its fundamental traits: promoting innovation in SMEs through direct grants for R&D projects. This is very much in line with the first frame of innovation policies. SBIR evolution, particularly with STTR implementation, targeted the encouragement of projects carried out cooperatively between companies and research institutions, a typical characteristic of the second frame of innovation policies, the so-called systemic policies (Edler and Fagerberg 2017; Schot and Steinmueller 2018). Although the focus on supporting SMEs and promoting innovation is common to all these programmes, there are considerable differences among them. These differences start with additional objectives stipulated in various cases, such as providing technological solutions for governments (e.g. the UK) and achieving social objectives (e.g. India). Indeed, the SBIR in the USA itself, as discussed in Section 3, pursued other goals such as the inclusion of disadvantaged social groups, consistent with the third frame of innovation policies (Zehavi and Breznitz 2017; Schot and Steinmueller 2018).

The type of organization responsible for executing the programme is another aspect where significant heterogeneity is observed in international experience. Decentralized agencies manage the programme in the USA and the UK, but responsibility for programme execution usually lies with institutions focused on fostering innovation (China and Brazil) or promoting SMEs (Japan and South Korea). There are also cases where operation is entrusted to a sector-specific body (India) or even delegated to private enterprises (Germany). While programmes are typically managed by central government agencies, subnational government involvement in direct programme operation is substantial in Japan and China, and relevant even in the USA, where state governments often provide support for proposal development under SBIR. Furthermore, there are countries where subnational entities have taken the lead in consistently executing a SBIR-like programme (e.g. Brazil).

There are many specificities in defining the target audience of these programmes. Restrictions commonly include limiting access to companies controlled by local citizens, but, besides variations in SME status definitions, there are cases of additional restrictions based on criteria such as technological effort intensity (seen in China) and company age (applied in Germany).

Sectoral targeting of stimulus is another area with discrepancies. In addition to the Indian programme’s focus on biotechnology and implicit priorities in meeting agency needs through decentralized programmes, there are instances of explicit technological selectivity (Chinese Innofund) as well as programmes that actively pursue neutrality (NSF’s SBIR in the USA and PIPE in Brazil).

Grant limits awarded to approved projects vary widely. In most countries, caps for Phase 2 projects are around $500,000, but much higher amounts are found in the USA, South Korea, and Germany’s cooperative projects. Even more variable is the overall expenditure on the programme. While the economic size of the country certainly plays a role, it does not appear to be the only explanatory factor for the disparities in programme size. The available evidence is summarized in Table 2 and suggests that the South Korean programme is not much smaller than that of the USA, despite the large differences in country populations and GDP. Programmes in large economies such as China, Germany, and Japan often have annual budget allocations worth hundreds of millions of dollars, but the British programme is much less funded. In the two least developed countries, Brazil and India, even controlling for economic size, the programmes are smaller than the original SBIR by one or two orders of magnitude.

The original conception of the SBIR limited the use of public resources to the stages of the R&D process preceding prototype development. Even in the USA, known for its ample availability of VC, arrangements were made to address the funding gap in Phase 3. Notably, these include the DoD’s Phase 2+ and NSF’s Phase 2B programmes. Moreover, as recognized by Sargent (2015), some agencies utilize external resources outside of SBIR to support additional R&D or procure technologies for government use. Research has identified that almost all other countries mobilize policy instruments to address this issue. In countries with deep-rooted traditions in industrial and innovation policies, such as Japan, South Korea, and Germany, there is a wide range of complementary programmes to support innovative SMEs. The use of diverse instruments and their adaptation to contingencies reinforce findings from other studies employing different methodologies, such as those by Borrás and Edquist (2013), OCDE (2022), and Howoldt (2024).

Although there is a significant disparity in the availability of information regarding these programmes, which hampers a more systematic comparison, it is essential to discuss the evaluations of the results. Unlike the American SBIR, which has undergone numerous institutional and independent assessments, the literature review has not identified robust studies with this perspective for many other programmes. Nevertheless, one could argue that results similar to those of the American SBIR may only be evident in the case of the Chinese SBIR, which nevertheless is not immune to criticism. Evaluations of two other major programmes, i.e. the South Korean and Japanese ones, are mixed, with strong arguments from skeptical evaluators (Inoue and Yamaguchi 2017; Lee and Jo 2018). Particularly it appears to be generalizable the appreciation, articulated by Inoue and Yamaguchi (2017) regarding the Japanese case, that programme results depend on both effective commercial orientation and the capabilities of the national innovation system, notably its scientific knowledge generation and capital mobilization components.

6. Conclusion

Reviewing international experience with SBIR-like programmes unequivocally demonstrates their relevance. Although information for equally comprehensive periods is not available, data from just three of the largest economies (China, Germany, and South Korea) that have emulated the original programme indicate support for over 110,000 projects. This number stands strong even when compared to the total of 195,000 projects funded by the American SBIR over four decades from 1982 to 2021. With programmes of this nature implemented in at least twenty-one countries, they have become an essential component of the toolkit for promoting innovation among SMEs globally.

The original SBIR was created in an environment where there were serious concerns about the competitive performance of the American economy, and there was growing recognition of the role of small businesses in technological and innovative dynamics. However, more solid theoretical justifications for the SBIR were formulated afterwards. Wessner (2008) argues, on one hand, that large companies are typically reluctant to invest in disruptive technologies and, on the other hand, that there are obstacles to private capital investment in small technology-based firms, especially in their early stages of development. The actions of angel investors and VC tend to be confined to areas located within a short distance of them (Ben-Ari and Vonortas 2007). Furthermore, information asymmetry between entrepreneurs and investors inhibits investment in unproven projects. Finally, high transaction costs (assessment and monitoring) restrict the number of invested firms in portfolios and hamper smaller enterprises. It is the vacuum created by these situations that, even from a liberal perspective, justifies the government’s intervention through SBIR-like programmes, which ‘addresses segments of the innovation cycle that private investors often do not fund because they find it too risky or too small’ (Wessner 2008: 33). The most influential and cited work in this century on the SBIR identified yet another rationale for the programme. By alleviating financial constraints on deepening research, grants ‘reduc[e] technological uncertainty, which makes the firm a more viable investment opportunity’ (Howell 2017: 1162). Based on Yeow, Rigby and Li (2017: 116), it should be added that the design of these programmes ‘can incorporate strategic goals of socio-economic development so as to embed a mission orientation’.

SBIR-like programmes are often, though not universally (Edquist and Zabala-Iturriagagoitia 2015), classified as public procurement policies, specifically falling under pre-commercial procurement, which involves contracting R&D for products or services not yet available (Selviaridis 2021). Yeow, Rigby and Li (2017) even list SBIR and its British and Dutch counterparts as typical examples. However, the extensive review in this article suggests that such classification depends on specific characteristics of programmes in some countries, particularly those emphasizing R&D contracting for government use. This classification may not apply to many similar programmes worldwide that operate based on more open proposal calls.

When these programmes are aligned with government demands, the challenges of transitioning technological development to commercialization are naturally reduced. A sample survey conducted by Keller and Block (2013) based on SBIR projects approved in 2002 showed that 47% of companies obtained in the 6 years following at least one non-SBIR federal government contract. However, neither studies on the American SBIR nor the limited literature on its counterparts provide a clear answer as to whether the most striking cases of success concentrate on the more open-ended aspects of these policies or not.

At first glance, the international spread of programmes similar to SBIR exemplifies a case of mimetic institutional isomorphism, a process where even organizations not subject to unified command adopt practices similar to those of a model organization perceived as more legitimate or successful in their field (DiMaggio and Powell 1983). Indeed, as argued by Cao and Prakash (2010: 113), ‘countries carefully watch policies of their competitors to ensure that they are not disadvantaged’. Nevertheless, due to the recognition of how policies are altered when transplanted from one jurisdiction to another, the literature on this process has evolved towards characterizing it as a process of translation rather than simple diffusion. Discussing what the author terms ‘indigenization of policy’, Stone (2012: 489) emphasizes that the socio-economic specificities of the receiving environment effectively induce transformations in policies from the initial adoption phase, and these transformations are reinforced over time. This is because the ‘logics of appropriateness entail a gradual adjustment and modifications that lead to different outcomes than may have originally been envisaged’.

Acknowledgements

The authors are grateful to Pedro Peverari di Lallo for his outstanding research assistance. The authors also thank the São Paulo Research Foundation (Fapesp) officers in charge of Innovation Research, especially Luciana Hashiba, for discussing the research report on which this paper is based.

Conflict of interest.

None declared.

Funding

This work was supported through grants by the São Paulo Research Foundation (Fapesp) and by the Brazilian National Council for Scientific and Technological Development (CNPq) [311900/2021-6 to M.R.A.B. and 308313/2020-8 to A.L.V.T.].

Footnotes

References