Seoul Center for Finance and Innovation The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS © 2024 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The find- ings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522- 2625; e-mail: pubrights@worldbank.org. World Bank Group Finance, Competitiveness and Innovation Global Practice The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS TIPS: Tech Incubator Program for Startup Seoul Center for Finance and Innovation january 2024 Acknowledgments The report is a collaboration between the World Bank Seoul Center for Finance and Innovation and the Korea Advanced Institute of Science and Technology (KAIST) Innovation Strategy and Policy Institute (ISPI). The research team was led by Anwar Aridi (Senior Private Sector Specialist, World Bank) and Wonjoon Kim (Professor, KAIST), which included Kibum Kim (Private Sector Specialist, World Bank), Kyeyoung Shin (Consultant, WB), Joo Sueb Lee (Senior Economist, WB), and Dae- hyun Kim, Taekyun Kim, and Hyunsung Park from KAIST. The note benefited from the guidance of the World Bank Management, Zafer Mustafaoglu (Practice Manager) and Jason Allford (Country Manager), and from feedback and comments provided by Xavier Cirera (Senior Economist) and Nelson Gray (Consultant). This knowledge note was made possible by a grant from the Korean Ministry of Economy and Fi- nance, provided through the Seoul Center for Finance and Innovation and the World Bank Group Korea Office. An earlier version of the data analysis was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Edu- cation (2021R1A6A1A14045741). 4 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Contents Acknowledgments..........................................................................................................4 Abbreviations and Acronyms...........................................................................................6 Executive Summary.........................................................................................................7 01. Introduction.............................................................................................................9 02. Development of Korea’s Entrepreneurial Ecosystem....................................................15 03. Overview of TIPS.....................................................................................................20 04. Data and Methods................................................................................................... 24 05. Results...................................................................................................................30 06. Conclusion and Policy Implications........................................................................... 39 References................................................................................................................... 45 Appendices..................................................................................................................50 A. Overall Process of the Tech Incubator Program for Startup...........................................51 B. Types of Tech Incubator Program for Startup Support ................................................ 53 C. Eligible Tech Incubator Program for Startup Expenditures........................................... 55 D. Comparison of public-private matching grant programs for technology startups............ 56 E. Deeptech Tech Incubator Program for Startup........................................................... 58 5 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Abbreviations and Acronyms GDP Gross domestic product IPO Initial public offering KAIST Korea Advanced Institute of Science and Technology KBAN Korea Business Angels Association KISED Korea Institute of Startup & Entrepreneurship Development KRW Korean won KoDATA Korea Rating and Data KODIT Korea Credit Guarantee Fund KOSDAQ Korean Securities Dealers Automated Quotations KOTEC Korea Technology Finance Corporation KVIC Korea Venture Investment Corporation M&A Merger and acquisition MSS Ministry of SMEs and Startups OLS Ordinary Least Squares R&D Research and development SMEs Small and Medium-sized Enterprises TIP Technological Incubators Program TIPS Tech Incubator Program for Startup VC Venture capital 6 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Executive Summary Although governments around the world have introduced policies to support startups and foster entrepreneurial ecosystems, research findings on the effectiveness of such public support are in- conclusive, suggesting that the design and implementation of startup support programs may play a role in determining program effectiveness. While private firms are often involved in designing and implementing these programs, research into how this affects program effectiveness is limited. The objective of this study is to provide empirical evidence of the effectiveness and additionality of a public-private matching grant program and identify key design features that may be replicable in other countries. The target audience of this note includes policymakers and researchers inter- ested in the design and implementation of entrepreneurship policy. This study investigates the case of a Korean public-private matching grant program called the Tech Incubator Program for Startup (TIPS). Launched in 2013 and benchmarked after Israel’s Techno- logical Incubators Program, TIPS leverages the capabilities of its implementation partners from the private sector to select and support innovative early-stage technology startups. TIPS partners, which include accelerators, venture capital firms, and other organizations certified by the Kore- an government, invest in startups and then recommend the startups to the public program for matching grants. With the aim of growing innovative startups into globally competitive firms, TIPS provides a comprehensive package of support to selected startups, including funds for research and development (R&D) and mentorship, for up to three years. After 10 years in operation, TIPS is particularly well suited to answer the question of whether public funding can help startups inno- vate and subsequently improve their performance. To assess the impact of TIPS on select entrepreneurial outcomes, we employed a sharp regression discontinuity design using a dataset that included 1,650 startups that applied for TIPS between 2013 and 2020. This empirical study leveraged a large dataset of program beneficiaries to reveal the impact of TIPS on startup performance. The regression results show that TIPS led to better startup performance one year after selection in terms of innovation input (R&D intensity) and output (patenting), although a statistically significant impact of TIPS on beneficiaries’ revenue and R&D collaboration activities was not found. TIPS beneficiaries exhibited a higher propensity to conduct R&D activities, file patents, and increase staffing than firms that applied but were not selected. TIPS beneficiaries also showed better performance in terms of subsequent investments than non-recipients. The fact that the program had no significant impact on revenues could imply that the matching grants enabled firms to hire more staff to work on product development, allow- ing them to focus more on exploratory innovations. 7 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS The findings have several policy implications. First, a well-designed coordination mechanism be- tween private investors and public funds may serve as a viable public-private partnership model for fostering innovative startups. TIPS shows that when the private sector’s selection capabilities are coupled with the public sector’s resources and long-term risk tolerance, entrepreneurship policy can deliver on its objectives. Second, appropriate implementation of a public-private co-in- vestment model can have a multiplier effect on private investment by reducing the risk of invest- ment in early-stage startups. Although the government invests five to seven times more than the private investors throughout the program, recent data show that the TIPS beneficiaries crowded in more follow-on investments from private investors. The cumulative amount of follow-on in- vestments was about 10.4 times the amount of public spending on TIPS. Third, the effectiveness of an entrepreneurship program may depend on complementary support programs, which could target different stages of the startup lifecycle. This is illustrated by the complementary programs Pre-TIPS and Post-TIPS targeting early-stage entrepreneurs and scaleups graduating from TIPS re- spectively. Fourth, patient capital and continuity in entrepreneurial policy with a long-term view are key to nurturing a vibrant startup ecosystem. Fifth, constant engagement with beneficiaries through data collection and monitoring enables the development of a dynamic monitoring and evaluation mechanism. In addition to discussing policy implications, we suggest several policy considerations for TIPS-like program replicability in developing economies. 8 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 01 Introduction 9 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 1. Introduction Supporting startups and fostering dynamic entrepreneurial ecosystems have become major policy agendas globally (Bai et al. 2021; Berger and Hottenrott 2021). Startups are widely recog- nized as drivers of innovation and knowledge spillover (Acs et al. 2009), and numerous studies have highlighted their impact on economic growth (e.g., Audretsch et al. 2006; Gries and Naudé 2008), innovation (Spender et al. 2017), and productivity (Aghion et al. 2009). Thus, policymak- ers are increasingly involved in promoting startups through grants, subsidies, and tax incentives, favoring those with high potential for innovation. Such public support programs have two main objectives: reduce the consequences of financial market frictions that frequently occur during the early stages of company development and, by providing seed funding, alleviate specific risks tied to R&D that might deter private investment. The unique financial challenges that startups face and the potential societal gains from their innovation activities (R&D externalities) have led many countries to design public support programs that target early-stage innovative companies (Brown and Earle 2017; Hellmann and Thiele 2019; Storey and Tether 1998). Government funding for startups has become widespread worldwide and is increasing. A re- cent comprehensive study of 755 national government startup funding programs in 66 countries found that they allocated a cumulative budget of USD 156 billion between 2010 and 2019 (Bai et al. 2021). On average, USD 1.85 billion was spent per year on financing programs, and each coun- try had an average of 11.4 such programs with an average program period of 11 years between 1995 and 2019. This phenomenon was not only limited to developed countries, but also present in developing economies, indicating that public funding support towards entrepreneurial activi- ty is prevalent globally. Furthermore, government spending on these programs has continuously increased over time worldwide, from approximately USD 50 billion in 1995 to more than USD 170 billion in 2019 (Figures 1 and 2). 10 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 1. Number of Startup Funding Policies per Country, 1995–2019 (30,45] (20,30] (17,20] (15,17] (10,15] (8,10] (4,8] (3,4] (2,3] [1,2] no programs found Source: Bai et al. 2021. Figure 2. Aggregate Spending on Startup Funding Programs, 1995–2019 180 160 140 120 100 80 60 Billion USD 40 20 0 Source: Bai et al. 2021. 11 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Researchers have identified three rationales for providing public funding support to innova- tive startups: reducing market failures that limit commercialization of research, stimulating private investment, and fostering an environment that enables growth of startups and de- velopment of private venture finance. First, public funding for innovative startups can mitigate uncertainty as to whether R&D projects will lead to usable results or outputs (Hottenrott and Richstein 2020). Innovative startups face substantial incomplete appropriability problems; often, they cannot fully appropriate returns from their early-stage R&D investments and do not possess complementary assets such as a well-branded reputation, systemized distribution channel, or installed base (Anton and Yao 1994; Gans and Stern 2003; Teece 1987). Public support can ad- dress such problems by investing in areas with positive externalities. Second, government funding can act as a catalyst, stimulating private investment in early-stage startups. Government invest- ment in startups can serve as validation for external investors, potentially prompting subsequent private investments (Lerner 1999). Particularly when technological and market uncertainties are substantial in the context of early-stage startups, reducing uncertainties through government intervention can make investment opportunities more appealing to private investors (Toole and Turvey 2009). Third, government funding can facilitate establishment of an institutional environ- ment that enables the growth of the private investors (Salmenkaita and Salo 2002). In the absence of external stimuli, achieving a new configuration in the innovation system becomes challenging (Metcalfe 1995). A prime example is Israel in the 1990s, when only two private venture capital (VC) funds existed. The government invested USD 100 million in a VC fund dedicated to high-technol- ogy startups in 1993, and the number of private VC funds specializing in technology startups grew to 40 by 1996, at which point the government divested its stake. This public intervention, aimed at mitigating barriers to commercialization for startups, helped shape an institutional environment that fosters expansion of the private VC industry. Although the rationale for government funding is well recognized, whether it achieves the in- tended benefits remains inconclusive (Bianchi, Murtinu, and Scalera 2019; Chen et al. 2018; Gao et al. 2021; Islam, Fremeth and Marcus 2018; Szczygielski et al. 2017; Vanino, Roper, and Becker 2019). Researchers have found a positive correlation between public support and startup perfor- mance. Howell (2017) found that firms benefiting from the Small Business Innovation Research program in the United States tend to receive more subsequent VC investments and perform better in revenue and patenting. Similarly, public financial support allows startups to focus on explorato- ry innovations (Gao et al. 2021). Cantner and Kösters (2015), using a dataset of innovative startups in the German state of Thuringia, showed that subsidized startups have 2.8 times as many patents as nonrecipients and 66 percent higher growth in employment. Giraudo et al. (2019) showed sim- ilar effects for Italian startups. Nevertheless, some research indicates that subsidizing startups through public funding crowds out private investors (Dimos and Pugh 2016). There are instances in which government funding led to adverse effects such as rent-seeking and low investment efficien- cy (Luo et al. 2021). Empirical evidence of the effect of public support on startups is mixed, and determining whether government funding can help startups innovate and subsequently improve their performance remains a challenging, unanswered empirical question. The mixed findings on the effect of public support on startups suggest that design and im- plementation of startup support programs may play important roles in determining program effectiveness. Conti (2018) confirmed that the design features of an Israeli support program af- fected its efficacy; when limitations were placed on sharing knowledge outside of a particular 12 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS area, the advantages of participating in the program were reduced, but when these limitations were removed, the program helped startups survive, attract external investment, and improve innovation performance. In their study of high-technology startups in China, Luo et al. (2021) identified the configurations in which government subsidies led to performance improvements under multiple scenarios and suggested that support program design can have different conse- quences for the effectiveness of startup support programs. Although these studies indicate the importance of program design, the question remains as to which features of program design are most important for effectiveness. Research into how collaboration between public and private actors could improve the de- livery and effectiveness of startup support programs has been limited. Startups often lack publicly visible information on themselves, creating information asymmetry between them and potential financiers. This asymmetry is particularly pronounced in knowledge-intensive sectors (Carpenter and Petersen 2002), where the complexity and novelty of products and services mean that financiers typically lack readily available information for accurate assessment. Securing fi- nancing for such startups requires considerable expertise (Kaplan and Stromberg 2003), making it challenging for public officials without such expertise to allocate capital effectively. Research suggests that private financiers, such as VC firms and angel investors, who have better knowledge of startups and monitoring capabilities (Norton and Tenenbaum 1993), can compensate for the government’s limitations. Consequently, most public funding support programs involve the pri- vate sector in program design, implementation, and evaluation.1 However, limited research has been undertaken to examine how such programs work and how effective they are. An impact evaluation of Korea’s Tech Incubator Program for Startup (TIPS) may bridge the research gap and also serve as a practical guidance for policymakers. TIPS is a public-private matching grant program that leverages the selection capabilities of private sector implementation partners, known as TIPS partners (e.g., accelerators, VC firms, large enterprises; details available in Section 3) to compensate for the public sector’s lack of the skills, know-how, and expertise required to support technology startups. TIPS partners invest approximately 100 million Korean won (KRW; USD 77,000) in a promising technology startup and recommend the firm to the TIPS management agency, which is then evaluated by the TIPS committee. The selected beneficiaries receive additional financial support that can be used for R&D (maximum KRW 500 million; USD 387,000), commercialization (maximum KRW 100 million; USD 77,000), and global marketing (maximum KRW 100 million; USD 77,000). In addition to financial support, TIPS provides access to technological experts and business incubation space, infrastructure, and mentoring services. Since its launch in 2013 with the goal of identifying new growth engines and creating job opportu- nities for young talents, the program has become one of the country’s flagship government pro- grams for supporting entrepreneurship. A small number of studies looked into the effectiveness of TIPS and the evidence remains in- conclusive due to limitations in examining causality and accessing data. For example, a study by the Korea Development Institute (Koo 2018) found that mentoring and ties with investors pos- 1 Private investors participated in 85 percent of the 755 worldwide government programs on which Bai et al. (2021) collected data. 13 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS itively influenced follow-on investment in startups. Using network analysis of participating firms and investors, Kim and Kim (2018) also found that the stronger the investor‘s network connectivity, the more likely it is that the investor’ s startup succeeds in attracting subsequent investment. Although these studies provide valuable insights, assessing the effectiveness of TIPS, and gov- ernment funding programs for startups in general, requires more investigation and supporting evidence. The objective of this evaluation is to provide empirical evidence of the effectiveness and ad- ditionality of TIPS and identify the key design features that may be replicable for other coun- tries. Specifically, we analyzed whether TIPS beneficiary startups achieved higher revenue growth, obtained more subsequent investments, hired more staff, and participated in more research and collaboration than startups that applied for the program but did not receive support. This research used an original granular dataset that the Korean Ministry of SMEs and Startups (MSS) provided and other firm-level financial and patent data for the empirical analysis. The findings suggest that TIPS had positive effects on several but not all the entrepreneurial outcomes of focus. Based on the findings, the research concludes with several policy lessons that could inform policymakers in developing and developed economies. Finally, we suggest several policy considerations for TIPS- like program replicability in developing economies. 14 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 02 Development of Korea’s Entrepreneurial Ecosystem 15 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 2. Development of Korea’s Entrepreneurial Ecosystem The burst of the dot-com bubble in 2000 caused the Korean government to recalibrate its en- trepreneurship policies (Doh and Kim 2014). The government implemented several measures to increase transparency and stability in venture investments, such as introducing the Venture Com- pany Stabilization Plan in 2002 and tightening requirements for exit from the Korean Securities Dealers Automated Quotations (KOSDAQ) market in 2003. To reinvigorate startup activities that had stagnated after the dot-com bubble burst, a variety of support policies were implemented in the mid-2000s. For example, in 2005, the government launched its first fund of funds to provide stable, sustainable capital for investment in startups. From 2008 to 2012, the government focused on streamlining administrative procedures, eliminating minimum capital requirements, and de- veloping online startup registration systems to support venture firms (Kim and Kang 2021). It es- tablished the Angel Investment Support Center in 2011 and has launched several angel investment matching funds (Yang and Park 2020). These measures were implemented with the primary objective of promoting private venture investment and stimulating entrepreneurial activities in the country. Budget allocations to the MSS, which is the government body mainly responsible for the entrepreneurship agenda, in- creased substantially, reaching KRW 849.2 billion (USD 658 million) in 2020, a 5.9-fold increase in a decade (Figure 3).2 The MSS’s budget for the entrepreneurship agenda was more than 0.04 percent of gross domestic product (GDP) in 2020. Different administrations over the last 20 years showed a strong commitment to the promotion of innovative entrepreneurship. Major policy ac- tions and initiatives (Table 1) were correlated with consistent growth in the number of venture firms throughout the mid-2000s (Figure 4). 2 When including all budgets for supporting entrepreneurship administered by line ministries and local governments, the total budget was KRW 1.45 trillion (USD 1.12 billion) in 2020 (MSS 2022). 16 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 3. MSS Budget for Startup Support and Share of GDP 900 0.05 800 700 0.04 600 0.03 500 400 0.02 300 KRW billion 200 % of gdp 0.01 100 0 0.00 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 mss budget for startup support mss budget for startup support as share of gdp Source: MSS budget data, MSS 2021; GDP data, Statistics Korea (https://kosis.kr/index/index.do). Note: The MSS budget for startup support includes allocations for providing incubation space, fostering startup clusters, ex- panding infrastructure for entrepreneurial activities, and supporting commercialization, as defined by the MSS, and excludes allocations for research and development, policy loans, and investments. Table 1. Major Policy Actions and Events to Support Startups Administration year Policy actions and events The 1980s through 1986 Support for Small and Medium Enterprise Establishment Act, Financial Assistance to the early 2000s New Technology Businesses Act 1996 The Small and Medium Business Administration established; the Korean Securities Dealers Automated Quotations (KOSDAQ) opened 1997 Act on Special Measures for the Promotion of Venture Businesses 2002 The venture firm registration system redesigned Roh Mu-hyun 2003 Enhancing transparency of venture capital (2003-08) 2004 Measures to promote venture firms 2005 Rolling out a fund-of-funds 2006 Measures to promote venture capital (for example, on private fund-of-funds) 2007 Promoting new-technology spinoffs from universities and research institutions; renewing the Act on Special Measures for the Promotion of Venture Businesses Lee Myung-bak 2008 Measures to promote technology startups (2008-13) 2011 Measures to promote startups by youths Park Geun-hye 2013 Streamlining startup and venture capital promotion programs; starting the Tech (2013-17) Incubator Program for Startups Moon Jae-In 2017 Measures to further develop the innovation startup ecosystem (2017-22) 2018 Measures to promote scaling-up Source: WBG 2023. 17 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Despite rapid growth in the number of venture firms, growth in the number of VC-backed firms stagnated in the 2000s (Figure 4).3 Moreover, the number of VC firms fell, and the volume of new investments increased only modestly in the 2000s (Figure 5). The amount of angel invest- ment and the number of firms that angel investors backed also stagnated during this period (SMBA 2016). Against this backdrop, the Korean government introduced TIPS as a support mechanism for innovative early-stage ventures and to enhance the domestic investment ecosystem. Figure 4. Number of Venture Companies and Venture Capital (VC)-Backed Companies 45,000 3,000 number of vc-backed companies number of Venture Companies 40,000 2,500 35,000 30,000 2,000 25,000 1,500 20,000 15,000 1,000 10,000 500 5,000 0 0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2020 2011 2012 2013 2014 2015 2016 2017 2018 2019 2021 year number of Venture Companies number of vc-backed companies Source: Korea Venture Capital Association (KVCA), Statistics Korea (https://www.index.go.kr/unity/potal/main/EachDtlPage- Detail.do;jsessionid=64Xf3-GjaVd1Ofr0ZYI9aRnRrp1hhvbxNGaaPcyf.node11?idx_cd=1195) Figure 5. Number of Venture Capital (VC) Firms and Investment Amounts by VC Firms number of Venture capital firms 250 9,000 8,000 200 7,000 new investment amount 6,000 150 5,000 4,000 100 3,000 (billion won) 50 2,000 1,000 0 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2020 2011 2012 2013 2014 2015 2016 2017 2018 2019 2021 year number of vc firms new investment by vc firms Source: Statistics Korea (https://www.index.go.kr/unity/potal/main/EachDtlPageDetail.do?idx_cd=1196) 3 Venture firms refers to small and medium-sized businesses as prescribed in the provisions of Article 2 of the Act on Special Measures for the Promotion of Venture Businesses, and VC-based firms refers to firms that use venture capital to fund their businesses. 18 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Since its launch in 2013, TIPS has been considered the flagship policy program of the coun- try’s expanding entrepreneurship support policy mix. Benchmarked after Israel’s Technological Incubators Program (TIP), TIPS was launched to promote technology startups with an initial cohort of just 15 beneficiary startups. Since then, the program has been credited with several achieve- ments, including a growing number of beneficiary startups exiting through initial public offerings (IPOs) and mergers and acquisitions (M&As), as well as a number of beneficiaries that became unicorns.4 The government has expanded the program over the years, which coincided with the growth of venture firms, VC firms, and VC investments (Figures 4 and 5). The government launched Pre-TIPS and Post-TIPS in 2018, which are complementary programs for startups in developmental stages before and after the stage that TIPS targets, respectively. 4 Unlisted startups that are less than 10 years old and have achieved USD 1 billion in valuation. 19 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 03 Overview of TIPS 20 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 3. Overview of TIPS TIPS is a public-private matching grant program that leverages the capabilities of its private sector partners to select and support innovative technology startups. To be eligible to apply for TIPS, a startup that is less than seven years old must receive an investment of about KRW 100 million (USD 77,000) from a private entity pre-designated as a TIPS partner.5 After this initial investment, the TIPS partner recommends the startup to the program, which then evaluates the startup for a final decision. This selection process allows private investors to take the lead in in- vestment decision making knowing that they can leverage public funding to reduce risks in their portfolios. The TIPS process is described in more detail in Appendix A. TIPS provides a comprehensive package of support to selected startups, with the aim of growing them into globally competitive firms. Financial support from the government includes a maximum of KRW 500 million (USD 386,000) for R&D throughout the two-year period. If a startup applies and is selected for further support, the program offers an additional KRW 100 million (USD 77,000) for commercialization, KRW 100 million (USD 77,000) for overseas marketing activities, and KRW 200 million (USD 144,000) for another angel matching fund for another 10 months. In total, a TIPS startup can access up to KRW 900 million (USD 693,000) throughout the three-year participation period. The government also provides access to R&D facilities, office space, net- working opportunities, and mentorship. The match ratio of financial support between the private and public sectors is approximately 1:5 with the government bearing a significant portion.6 The ratio is comparable to that of Israel’s TIP which TIPS benchmarked in that TIP provides a grant of up to 85 percent of the approved budget. More details about the various types of TIPS support and eligible expenditures are available in Appendices B and C. Appendix D includes a comparison of TIPS with Israel’s TIP and the US’s Small Business Innovation Research (SBIR) program. Since its launch, TIPS has become one of the country’s flagship government programs for entrepreneurs. TIPS started in 2013 with five partners and 15 startups and grew to more than 100 5 Generally, TIPS partners are required to complete the initial investment in startups before the startup can apply for TIPS, but depending on the specifics of private contracts, it is possible to apply with a commitment letter specifying investment amount, share quantity, and payment deadline. During final selection, the commitment is verified to ensure that the investment has been made. 6 The match ratio is based on the matching grant for R&D activity and excludes the government’s additional support for non- R&D activities such as commercialization and overseas marketing that are also included in the TIPS support package. The match ratio is not fixed and may change depending on the initial investment amount a TIPS partner makes. More details on the different types of TIPS support and their conditions are available in Appendix B. 21 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS partners and 500 startups in 2022. The MSS increased the TIPS budget from KRW 26.8 billion (USD 20.8 million) in 2014 to KRW 293.4 billion (USD 227.4 million) in 2022—from 0.008 percent to 0.043 percent of total government spending (Figure 6). Cumulatively, the government had allo- cated KRW 1,144.9 billion (USD 887.3 million) to TIPS by 2022.7 From 2013 to 2022, 2,134 startups were selected and matched with TIPS partners’ angel investments of KRW 493.9 billion (USD 382.5 million), with an average of KRW 230.0 million (USD 178,000) per startup8 (Figure 7). The com- bined domestic and foreign private follow-on investments that private angel investors, VC firms, incubators, and accelerators made reached KRW 10.4 trillion (USD 8.0 billion) by the end of 2022 (KBAN 2023a)—10.4 times the government’s cumulative spending on TIPS (KRW 1.0 trillion; USD 774.5 million) and 21 times the initial private investments that TIPS partners made (KRW 493.9 billion; USD 382.5 million). In 2023, the government launched a separate program that targeted a set of deep technology startups involved in 10 future strategic technologies that Korea selected as future growth engines: system semiconductors, biotechnology and health care, future mobility, green energy, robot- ics, big data and artificial intelligence, cybersecurity and networking, aerospace and maritime, next-generation nuclear power, and quantum technology (see Appendix E). Figure 6. Tech Incubator Program for Startup (TIPS) Budget and Share of Government Spending, 2014–22 350 350.0 0.05 0.043 0.043 0.045 300 300.0 0.04 0.035 0.035 250 250.0 0.035 0.03 tips budget (krw billion) 200 0.03 200.0 0.024 0.021 0.025 150 150.0 0.02 0.015 100 100.0 0.015 0.008 0.008 0.01 percent 50 50.0 0.005 26.8 29 29.0 59 59.0 84 84.0 104.2 145.4 195.1 208 208.0 293.4 0 0.0 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 annual tips budget tips budget as share of government spending Source: National Assembly Budget Office 2023; annual TIPS budget data from the MSS. 7 The cumulative budget was calculated by adding the annual budget amounts in announcements that MSS published. 8 As of March 2023. 22 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 7. Initial Private Investments Made by TIPS Partners and Number of TIPS Participants, 2014–22 600 140 investment amount (krw billion) 500 500 120 400 100 400 300 80 number of startups 300 256 255 205 60 200 40 79 85 100 54 20 12.2 14.4 16.1 43.7 54.9 57.2 72.7 98.2 124.5 0 0 2014 2015 2016 2017 2018 2019 2020 2021 2022 yearly number of tips participants yearly amount of angel investment Source: KBAN The three key stakeholders involved in TIPS are (a) TIPS partners, (b) TIPS recipients (tech startups), and (c) public implementation and oversight bodies. Details about their role are described below. A/ TIPS partners - TIPS partners are private sector entities such as angel investors, accelerators, VC firms, large corporations, and other organizations the Korean government has certified as qualified to support TIPS beneficiaries. TIPS partners are responsible for the initial selection of high-potential tech startups and recommending the startups that they are investing in to the TIPS management agency, the Korea Business Angels Association (KBAN). Every year, the Korean government designates TIPS partners, based mainly on their investment performance, technological expertise, and the ability to provide necessary financial resources to invest in startups. As of October 2023, 111 TIPS partners were registered. B/ TIPS recipients (tech startups) - To be eligible to apply for TIPS, startups must have a TIPS partner’s initial investment, be less than seven years old, have two or more employees (includ- ing the chief executive officer), and have earned no more than KRW 2 billion (USD 1.55 million) in revenue in the previous year. To ensure accountability, the founders’ share of equity must be more than 60 percent and the private investors’ share needs to be less than 30 percent. C/ Public implementation and oversight bodies – The Ministry of SME and Startups (MSS) over- sees the program. KBAN is the TIPS management agency that operates the program, and the Korea Institute of Startup & Entrepreneurship Development (KISED) is the TIPS specialized agency that facilitates connections between startups and additional private investors and pro- vides further support. Designated to support the implementation of the program by the gov- ernment, KBAN and KISED play a major role in the beneficiary selection process. 23 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 04 Data and Methods 24 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 4. Data and Methods The research used a dataset that includes 1,650 startups that applied for TIPS between 2013 and 2020.9 The data that the Korean government provided included not only evaluation scores but also exit types (e.g., M&A, failure), investment types, VC information, firm age, number of em- ployees, location, and industry code. The research also used patent data and financial data such as revenues and R&D expenses collected from Korea Rating and Data (KoDATA)10. Subsequent investment data were collected from the Organization for Venture Enterprise Certification, which is operated by the MSS. Competitions are held every other month, and every competition cohort has a different ac- ceptance cutoff score. Most of the TIPS applicants’ scores11 were concentrated between 58 and 70. Figure 8 shows the distribution of TIPS applicants’ evaluation scores from 2013 to 2020, and Figure 9 shows a split distribution of applications according to whether they were rejected (Figure 9a) or accepted (Figure 9b). The scores of accepted applicants (1,221) mostly ranged from 64 to 70, whereas those of the rejected applications (429) ranged from 58 to 64. For most competitions, the threshold score for being accepted was 64 points where most of the TIPS startups were posi- tioned. Figure 8. Distribution of Scores of TIPS Applications in the Sample, 2013–20 400 376 340 300 261 200 185 156 frequency 99 100 86 33 36 3 16 28 19 1 2 1 6 2 0 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 scores for all applications Source: MSS 9 The original sample dataset covered 2013 to 2021 and 2,334 startups, but some dependent variable data were not available when the analysis was conducted. The adjusted period covers the years 2013 to 2020. An earlier version of the analysis by Choi et al. (2021) was published in the Academy of Management Proceedings. 10 KoDATA specializes in credit information on enterprises that the Korea Credit Guarantee Fund (KODIT) and Korea Technology Finance Corporation (KOTEC) have collected and updated. 11 Scoring criteria are explained in detail in Appendix A. 25 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 9. Scores for Applications Based on Whether They Were (a) rejected or (b) accepted, 2013–20 (a) rejected (b) accepted 400 349 336 300 200 185 158 frequency 139 99 103 100 85 33 28 36 19 31 2 1 16 274 17 62 0 1 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 score Source: MSS We used rank information for a regression discontinuity design. The independent variable is rank, which is based on evaluation scores. Table 2 illustrates how rank data are constructed and how the treatment and control groups are divided based on the evaluation score. Firms with a positive rank are the TIPS beneficiary group, and those with a negative rank are the control group. Table 2. Illustration of Construction of Rank Data Startup Score Accepted Rank A 61.7 No -3 B 62.8 No -2 C 63.2 No -1 D 64.1 Yes +1 E 65.8 Yes +2 F 67.9 Yes +3 The dependent variables of the analysis were formulated based on the main objectives of TIPS, which are to nurture innovative technology-based startups and vitalize the venture fi- nance market. To assess the additionality of the program, we examined several dependent vari- ables that may represent desirable outcomes of this intervention: whether TIPS beneficiaries re- ceived more subsequent private investments than those that were not selected and whether there was a change in the recipients’ R&D intensity and employment levels. We hypothesized that, if TIPS had targeted the correct recipients and was implemented effectively, it would have had a positive impact on the variables. We examined changes in the recipients’ revenues and their patenting and co-patenting activities to assess their performance in terms of innovation outputs and re- search collaborations. In sum, the dependent variables of interest are (1) subsequent investment (whether a startup received subsequent investment, how many times the startup received the investment, and how much subsequent investment the startup received); (2) revenue; (3) employ- 26 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS ment; (4) innovation input for which R&D intensity12 was a proxy; (5) innovation output for which the number of patent applications was a proxy; and (6) co-patenting activities. We included firm age, number of employees, revenue, and R&D and capital intensity13 at the time of TIPS application as control variables and fixed effects for industries, TIPS partners, regions, and competition time. ployees, revenue, and R&D and capital intensity 14 at the time of TIPS application as ntrol variables and Wefixed effects a employed for industries, sharp TIPS regression partners, regions, discontinuity and design to competition assess the causal impact of TIPS on e. the variables of interest. This design allowed us to estimate the local average treatment effect by e employed a sharp comparing regressionstartups that were close discontinuity to assess threshold to the acceptance design the causal based on their impact of evaluation scores. PS on the variables of interest. To ensure This of the validity design allowed the design, weus to estimate followed the(2017) local average Howell’s empirical work, which requires atment effect by comparing that treatment startups that were not influence close rank andto acceptance theaward that threshold decisions be made based on ranking process. after the ir evaluation scores. To ensure the validity of the design, we followed Howell's (2017) In the case of TIPS, awardees are selected after the ranking, and the cutoff is determined inde- pirical work, which requiresof pendently that the treatment ranking (Howellnot influence 2017; Leerank and that and Lemieux award 2010). decisions Using be the regression discontinuity de after the ranking process. In the case of TIPS, awardees are selected after the design, we were able to compare how startups around the cutoff line (rank from -2 to 2) differed nking, and the cutoff is determined independently of the ranking (Howell 2017; Lee and in terms of the selected performance outcomes after receiving TIPS support. It is difficult to com- mieux 2010). Using the regression discontinuity design, we were able to compare how pletely rule out the selection bias problem as TIPS beneficiaries could have been initially more rtups around the cutoff line (rank from -2 to 2) differed in terms of the selected rformance outcomes after startups capable receiving compared to non-selected TIPS support. startups. It is difficult However,rule to completely the regression out the discontinuity de- sign aimed to control this ection bias problem as TIPS beneficiaries could have been initially more capable and treatment (TIPS-ac- effect. The fact that the control (TIPS-rejected) rtups compared to cepted) groups were non-selected around startups. the selection However, the cutoff line indicates regression that the discontinuity baseline capabilities by the design med to control thistime effect.they The were evaluated fact that thewere similar. control (TIPS-rejected) and treatment (TIPS- cepted) groups were around the selection cutoff line indicates that the baseline We they pabilities by the time used were evaluated logistic regression were for thesimilar. binary dependent variables (whether the firm received subse- quent investment and whether e used logistic regression for the binary dependentfirm the filed a patent variables (whetherapplication) and negative binomial estima- the firm received bsequent investment for the tion and count variables whether (number the firm filed of patent a patent applications, application) and number negative of subsequent investments). Otherwise, omial estimation for the count wevariables used ordinary least squares (number of patent (OLS). applications, number of bsequent investments). Otherwise, we used ordinary least squares (OLS). e regressions are The regressions are as follows: as follows: = + (1| > 0) + + + ere is the Y where main dependent Performance is the main variable that represents dependent variable that the performance represents of the performance of startup i; i rtup i; (1| τ> (1|0) is rank a > dummy 0) is a dummy variable with variable a value with a of value 1 if of startup 1 if i startup has i a has rank a higher rank higher than the cutoff i i n the cutoff point point and 0 otherwise; and 0 otherwise; is a γX is firm-level a firm-level control control variable variable (firm (firm age, age, numberof employees, reve- number employees, revenue, nue,R&D intensity, R&D intensity, capital capital i intensity) intensity) at at the the timetime of application; of application; and λand represents the fixed effects presents the fixed effects for industries, regions, TIPS partners, and competition time. c for industries, regions, TIPS partners, and competition time. ble 3 provides descriptive statistics for the 1,650 startups in the sample, and Tables 4 and rovide descriptive statistics Table 3 providesfor accepted descriptive and rejectedfor statistics startups, the 1,650 respectively. startups in the Thesample, and Tables 4 and 5 cepted and rejected startups in the sample were of a similar provide descriptive statistics for accepted and rejected startups, respectively. age and had a similar number The accepted and employees, indicating that the size and age of the firm are not important rejected startups in the sample were of a similar age and had a similar number of employees, selection criteria TIPS. The acceptance rate within the sample was 74 percent. indicating that the size and age of the firm are not important selection criteria for TIPS. The accep- ble 3. Descriptive Statistics tance rate within (all the startups samplein wasthe74sample) percent. able Standard Minimum Maximum Observations Mean deviation ow − on investment +1 (binary) 1,650 0.267 0.442 0 1 ow − on investment +1 (#) 1,650 0.43 0.922 0 8 ow − on investment 12 (amount) +1R&D 1,641 intensity is defined as expenditures by 0.618 1.157 a firm on its R&D 0 assets. 4.796 divided by the firm’s enue +1 (log) 13 Capital intensity measures the amount of12.084 1,269 1.766 spending on assets to support a 6.399 14.43 The formula for certain level of revenue. ployees +1 (log) 1,629 calculating capital intensity 0.536 consists of dividing 1.021 tangible fixed assets by total 0 4.691 assets of a company. nt application +1 (binary) 1,650 0.512 0.5 0 1 nt application +1 (#) 1,650 1.941 3.883 0 61 earch and development intensity +1 1,519 0.313 0.666 0 10.016 Co − patenting +1 1,650 0.049 0.217 0 1 o − patenting with university +1 1,650 0.042 0.202 0 1 m age (log) The Effects of Matching Grants 1,648 0.843 0.506 0 2.072 on Technology Startups: The Case of Korea’s TIPS 27 Capital intensity measures the amount of spending on assets to support a certain level of revenue. The mula for calculating capital intensity consists of dividing tangible fixed assets by total assets of a company. Table 3. Descriptive Statistics (all startups in the sample) Variable Observations Mean St. dev. Min Max Follow-on investment t+1 (binary) 1,650 0.267 0.442 0 1 Follow-on investment t+1 (#) 1,650 0.43 0.922 0 8 Follow-on investment t+1 (amount) 1,641 0.618 1.157 0 4.796 Revenue t+1 (log) 1,269 12.084 1.766 6.399 14.43 Employees t+1 (log) 1,629 0.536 1.021 0 4.691 Patent application t+1 (binary) 1,650 0.512 0.5 0 1 Patent application t+1 (#) 1,650 1.941 3.883 0 61 R&D intensity t+1 1,519 0.313 0.666 0 10.016 Co-patenting t+1 1,650 0.049 0.217 0 1 Co-patenting with university t+1 1,650 0.042 0.202 0 1 Firm age t (log) 1,648 0.843 0.506 0 2.072 Employees t (log) 1,629 0.508 0.859 0 3.989 Table 4. Descriptive Statistics (TIPS accepted) Variable Observations Mean St. dev. Min Max Follow-on investment t+1 (binary) 1,221 0.292 0.455 0 1 Follow-on investment t+1 (#) 1,221 0.488 1.006 0 8 Follow-on investment t+1 (amount) 1,214 0.697 1.229 0 4.796 Revenue t+1 (log) 962 12.029 1.797 6.399 14.43 Employees t+1 (log) 1,219 0.607 1.076 0 4.691 Patent application t+1 (binary) 1,221 0.55 0.498 0 1 Patent application t+1 (#) 1,221 2.003 3.996 0 61 R&D intensity t+1 1,171 0.34 0.72 0 10.016 Co-patenting t+1 1,221 0.056 0.231 0 1 Co-patenting with university t+1 1,221 0.048 0.215 0 1 Firm age t (log) 1,221 0.818 0.505 0 2.072 Employees t (log) 1,219 0.53 0.873 0 3.989 28 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Table 5. Descriptive Statistics (TIPS rejected) Variable Observations Mean St. dev. Minimum Maximum Follow-on investment t+1 (binary) 429 0.196 0.397 0 1 Follow-on investment t+1 (#) 429 0.266 0.595 0 3 Follow-on investment t+1 (amount) 427 0.394 0.888 0 4.263 Revenue t+1 (log) 307 12.254 1.657 6.909 14.43 Employees t+1 (log) 410 0.322 0.796 0 4.078 Patent application t+1 (binary) 429 0.401 0.491 0 1 Patent application t+1 (#) 429 1.765 3.542 0 24 R&D intensity t+1 348 0.221 0.425 0 3.381 Co-patenting t+1 429 0.030 0.171 0 1 Co-patenting with university t+1 429 0.026 0.158 0 1 Firm age t (log) 427 0.914 0.505 0 2.05 Employees t (log) 410 0.443 0.811 0 3.497 29 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 05 Results 30 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 5. Results Tables 6 to 8 present the results of the analysis of the effects of TIPS funding on startup per- formance, including subsequent investments, revenue, employment, R&D intensity, patent applications, and co-patenting activities, one year after the decision to accept or reject was made. This research only examined firms one year after the decision. We expect that it will be pos- sible to examine the program’s medium- to long-term effects when more data can be obtained in the future. Nevertheless, analyzing the effects at the one-year mark yields valuable insights given that startups experience immediate effects of changes in resources and other forms of support. The lower survival rate of startups compared to average firms further justifies this approach.14 Startups that received TIPS funding tended to have better performance in terms of subse- quent investments one year after the funding decision was made. Table 6 reports our estimates for the effects of TIPS on startups’ subsequent investments. Specifically, we examined whether a startup received subsequent investments; how many times it received subsequent investments; and the amounts of subsequent investment in M1, M2, and M3. We used logistic regression, neg- ative binomial regression, and OLS and consistently found that the coefficients of M1 (β = 2.613, standard error (SE) = 0.734); M2 (β = 0.709, SE = 0.364), and M3 (β = 0.430, SE = 0.170) were significant and positive. In terms of the likelihood of receiving subsequent investments (M1), the results indicate that TIPS recipients were 3.7 times more likely to receive subsequent investments than non-recipients. In terms of the number of subsequent investments, TIPS recipients were 70.3 percent more likely (0.709 percentage points) to receive subsequent investments (M2) and re- ceived 48.9 percent (0.430 percentage points) more in amount than non-recipients (M3). 14 In Korea, the one-year survival rate of startups is 64.8 percent (2020). One-year survival rate refers to the portion of startups established in 2019 and still operating in 2020 (KOSTAT 2022). 31 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Table 6. Effect of Tech Incubator Program for Startup (TIPS) on Startup Future Investments (t+1) M1 M2 M3 Bandwidth Rank ≤ ± 2 Rank ≤ ± 2 Rank ≤ ± 2 Investment Investment Investment (t + 1) (t + 1) (t + 1) Binary Number Amount TIPS award (standard error) 2.613*** (0.734) 0.709* (0.364) 0.430** (0.170) Controls Yes Yes Yes Industry fixed effects Yes Yes Yes Region fixed effects Yes Yes Yes TIPS partners fixed effects Yes Yes Yes Time fixed effects Yes Yes Yes Observations 160 249 247 Pseudo R-squared 0.344 0.354 0.434 Note: This table estimates the effects of TIPS on startups’ subsequent investments. All models include control variables such as logged firm age and number of employees, revenue, R&D and capital intensity, and fixed effects for industries, region, TIPS partners, and competition time. * p < 0.1, ** p < 0.05, *** p < 0.01. The effects of TIPS on subsequent investments across ranks from -6 to 6 are plotted in Figures 10 to 12. Startups that received TIPS funding (ranks 1 to 6) had a higher level of subsequent investment than those that were rejected (ranks -6 to -1). We also compare the mean values for subsequent investments in the ranks from -2 to 2, which is our sample for the regressions. For the three dif- ferent measures of subsequent investments, the mean values for TIPS recipients are meaningfully higher than those for non-recipients (subsequent investment: 0.343 vs 0.108 [Figure 10]; number of subsequent investments: 0.414 vs 0.139 [Figure 11]; amount of subsequent investment: 3.067 vs 0.567 [Figure 12]). Figure 10. Subsequent Investment (Binary) for Ranks from -6 to +6 .5 subsequent investment or not .4 .3 .2 .1 0 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award 32 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 11. Subsequent Investment (Number) for Ranks from -6 to +6 .8 the number of subsequent .6 investment recieved .4 .2 0 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award Figure 12. Subsequent Investment (Amount) for Ranks from -6 to +6 subsequent investment amount 10 8 6 4 2 0 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award Note: The unit of measurement for the y-axis is log(KRW + 1). Although TIPS had a significant positive effect on TIPS beneficiaries’ employment growth, it did not have a significant effect on revenue generation one year after the funding decision was made. Startups that received TIPS support tended to hire more employees, as M1 in Table 7 shows. TIPS recipients had 13.2 percent (0.265 percentage points) more employees than non-re- cipients. Mean employment growth in ranks from -2 to 2 in Figure 14 also supports the findings. This is not surprising given the sizable financial support the selected startups received. Column M2 in Table 7 and the candle charts in Figure 13 show that there were no significant differences in revenue generation between accepted and rejected startups. Considering that the average age of startups applying for TIPS was 1.3 years and that most were still at the pre-commercialization stage, it can be inferred that the financial support provided incentives to hire staff for R&D rather 33 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS than pursuing short-term revenue generation. In the absence of this financial support, early-stage startups that did not have a stable revenue stream might not have been able to hire and expand their teams to further develop their products and business models. One thing to consider in inter- preting these results is that we examined the effects one year after TIPS selection; commercializa- tion and revenue generation may take longer than hiring new staff. Table 7. Effect of Tech Incubator Program for Startup (TIPS) on Startups’ Employment and revenue M1 M2 Bandwidth Rank ≤ ± 2 Rank ≤ ± 2 Employment (t + 1) Revenue (t + 1) TIPS award (standard error) 0.265* (0.157) -0.011 (0.415) Controls Yes Yes Industry fixed effects Yes Yes Region fixed effects Yes Yes TIPS partners fixed effects Yes Yes Time fixed effects Yes Yes Observations 251 197 Pseudo R-squared 0.453 0.456 Note: This table estimates the effects of TIPS on startups’ revenue and employment. All models include control variables such as logged firm age and number of employees, revenue, R&D and capital intensity, and fixed effects for industries, region, TIPS partners, and competition time. * p < 0.1. Figure 13. Employment for Ranks from -6 to +6 1.5 1 employment .5 0 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award Note: The unit of measurement for the y-axis is log(number of employees +1). 34 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 14. revenue for Ranks from -6 to +6 13 12 11 revenue 10 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award Note: The unit of measurement for the y-axis is log(KRW + 1) TIPS had a positive impact on innovation input (proxied by R&D intensity) and output (prox- ied by patent application). The R&D intensity of TIPS recipients was 37.6 percent greater than that of non-recipients (Table 8, M1). As shown in the ranks from -2 to 2 in Figure 15, the mean value for TIPS recipients (0.326) was 2.5 times higher than that for non-recipients (0.127). This is also not surprising given that the program provides sizable amounts of financial support to awardees to be spent on R&D activities. M2 indicates whether a startup applied for a patent or not, and M3 indicates the number of patents applied for. The results show that TIPS recipients are 3.13 times as likely to apply for patents as non-recipients (β = 1.142, SE = 0.447). In terms of the number of ap- plied patents, TIPS recipients were 17.2 percent more likely to apply for patents. Figures 16 and 17 show the effects of TIPS on innovation outputs (patents) across the ranks from -6 to 6. Across the ranks from -2 to 2, TIPS recipients applied for 1.5 times as many patents as non-recipients (Figure 16: 0.488 vs 0.341; Figure 17: 2.308 vs 1.484). The analysis did not find a statistically significant effect of TIPS on startups’ co-patenting ac- tivities with external partners one year after the funding decision.15 TIPS did not have a statisti- cally significant effect on startups’ collaboration with universities and public research institutions (β = 0.013, SE = 0.014; Table 8, Column M5). Given that early-stage startups tend to focus more on the development of their business model and less on connecting with external R&D collabora- tors, this outcome is expected. Figures 18 and 19 show the effects of TIPS on co-patenting across the ranks from -6 to 6; there were no significant differences between the accepted and rejected startups. 15 External partners can be categorized as universities and public institutions versus other private corporations. The number of collaborations with other private corporations was statistically limited, so only the results of collaboration with universities and public institutions are presented. 35 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Table 8. Effect of Tech Incubator Program for Startup (TIPS) on Startups’ Research and Development (R&D) Intensity and Patenting and Co-Patenting Activity M1 M2 M3 M4 M5 Bandwidth Rank ≤ ± 2 Rank ≤ ± 2 Rank ≤ ± 2 Rank ≤ ± 2 Rank ≤ ± 2 R&D (t + 1) Innovation Innovation Co-patent Co-patent (t + 1) (t + 1) all (t + 1) university Binary Number (t + 1) TIPS award 0.142* 1.142** 0.760** 0.026 0.013 (standard error) (0.066) (0.447) (0.334) (0.019) (0.014) Controls Yes Yes Yes Yes Yes Industry fixed Yes Yes Yes Yes Yes effects Region fixed effects Yes Yes Yes Yes Yes TIPS partners Yes Yes Yes Yes Yes fixed effects Time fixed effects Yes Yes Yes Yes Yes Observations 226 196 251 251 251 Pseudo R-squared 0.534 0.243 0.171 0.618 0.688 Note. This table estimates the effects of TIPS on startups’ R&D intensity, patent applications, and co-patenting activities. All models include control variables, such as logged firm age and number of employees, revenue, R&D and capital intensity. and fixed effects for industries, region, TIPS partners, and competition time. * p < 0.1, ** p < 0.05. Figure 15. Research and Development (R&D) Intensity for Ranks from -6 to +6 1 .8 .6 .4 r&D intensity .2 0 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award 36 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 16. Patent Application (Binary) for Ranks from -6 to +6 .8 patent application or not .6 .4 .2 0 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award Figure 17. Patent Applications (Number) for Ranks from -6 to +6 5 the number of patents applied 4 3 2 1 0 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award Figure 18. Co-Patenting Activities (Binary) for Ranks from -6 to +6 .2 .15 .1 .05 co-patenting 0 -.05 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award 37 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Figure 19. Co-Patenting Activities (Binary) with Universities for Ranks from -6 to +6 .2 co-patenting with university .15 .1 .05 0 -.05 -6 -5 -4 -3 -2 -1 1 2 3 4 5 6 rank around cutoff for award The three-year survival rate of TIPS recipients was higher than that of startups that did not receive TIPS support; 95.7 percent of TIPS beneficiaries survived after three years, 7 percentage points higher than for the rejected group (88.7 percent).16 This may suggest that program funding helped startups survive the valley of death17 although the high survival rate of TIPS awardees, as well as of those rejected, may suggest that the scoring and selection criteria that the program and its partners used selected more-competitive startups. The results above indicate the immediate impact of TIPS on startups one year after selection. We also examined startups’ performance two years after selection (t+2), considering that startups may not exhibit immediate performance changes within one year, and the main findings remained consistent with the findings observed at the one-year interval.18 In line with our main findings, we found that TIPS recipients received more subsequent investments and had better performance in employment, innovation inputs (R&D intensity), and outputs (patent applications). Similarly, robustness checks confirmed that the effects of TIPS were not observable in terms of revenue gen- eration and co-patenting activities, consistent with the main findings. 16 Survival rates of TIPS-accepted and rejected startups were significantly higher than the average for all startups. In Korea, the average survival for startups is 64.8 percent for one year and 33.8 percent for five years (KOSTAT 2022). 17 The valley of death refers to the crucial early phase of a startup when it is difficult to cover the negative cash flow before its product or service brings in revenue from real customers. During this period, companies often deplete their initial capital in their quest to establish the business. 18 Some firm performance data for t+2 were missing from the dataset because they were not publicly available at the time of writing. 38 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 06 Conclusion and Policy Implications 39 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 6. Conclusion and Policy Implications This research analyzes the effects of Korea’s government matching grant program for ear- ly-stage technology startups, TIPS, on recipients’ performance and offers empirical evidence to inform entrepreneurship policymaking. Although the rationale for government funding for the entrepreneurship agenda is well recognized, the effects of startup funding programs remain inconclusive. This is also the case in Korea, where evidence of effectiveness is scarce because of lack of data or reliance on descriptive and qualitative approaches without explicitly examining causality. This research addresses this knowledge gap by using a novel, granular dataset. The research examined 1,650 startups that applied for TIPS from 2013 to 2020. The compiled dataset included not only evaluation scores of TIPS applicants and their R&D intensity, but also patent and financial information. To estimate the impact of TIPS on startup performance, the research deployed a sharp regression discontinuity design to estimate the local average treatment effect around a cutoff in a rating variable. The empirical results show that TIPS positively affected startup performance one year af- ter selection in terms of innovation input and output, although it did not have a significant effect on revenue or research collaboration activities. TIPS beneficiaries were more likely to conduct R&D activities, file patents, and increase staffing than firms that were not selected. The TIPS beneficiaries also received more subsequent investments than non-recipients. Startups that received TIPS support had 13.2 percent more hires and 37.6 percent higher R&D intensity than non-recipients. TIPS recipients were 3.7 times as likely to receive subsequent private investment as non-recipients and received 48.9 percent more in investment amount than non-recipients. The fact that there was no significant difference in the increase in revenues between beneficiaries and rejected applicants could imply that recipients used the matching grant mainly for R&D activities (as intended by the program) rather than pursuing short-term revenue generation. Further anal- ysis is needed over the long term to determine whether financial support from the program may have allowed the startups to further develop their businesses and increase revenues. There are some caveats in interpreting the results. First, it is difficult to isolate the marginal ef- fects of TIPS from those of complementary policies. There are several other government programs designed to resolve obstacles that hinder growth of startups (e.g., access to finance, access to infrastructure, intellectual property rights). Further investigation is needed to identify the addi- tionality of such targeted programs empirically and analyze the policy in the broader context of an extensive entrepreneurship policy mix. Second, although the research found positive effects of TIPS on startup performance, it is difficult to determine whether this was due to the design and 40 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS implementation features of TIPS or whether the TIPS-selected firms were better performers from the beginning. Even without TIPS support, TIPS beneficiaries might have performed well. A closer counterfactual analysis would be helpful to disentangle such effects. Third, whereas this research examined the immediate impact of TIPS on startup performance (one year after selection), track- ing mid- to long-term performance, particularly in terms of growth milestones, R&D outcomes, valuation, and exit performance, would be necessary to assess the benefits of the program.19 The findings of this research should be considered in the context of Korea’s recent efforts to develop its startup ecosystem. In recent decades, the Korean government has made significant efforts to shift away from a large conglomerate-led growth model to creating a more vibrant en- trepreneurial ecosystem (WBG 2023). The Korean government has actively implemented policies to support startups and SMEs to the extent that there have been concerns that the government’s financial support of the startup (and SMEs) agenda could have created dependency and extended the life span of unproductive firms with limited innovation potential (Baek 2017; FSC 2008).20 The extensive public support programs might have helped startups manage financial difficulties, but a large, complex policy mix inevitably raises the risks of inefficient allocation of resources (SBG 2023). Based on Korea’s policy experience in designing and implementing TIPS, five lessons could be derived. First, a well-designed coordination mechanism between private investors and public funds may serve as a viable public-private partnership model for fostering innovative startups. Leveraging public resources to crowd in private participation and investment is often regarded as the norm for co-investment programs, and the TIPS case offers validation that the program design is effective in meeting its objectives. When the private sector’s selection capabili- ties are coupled with the public sector’s resources and long-term risk tolerance, entrepreneurship policy can deliver on its objectives. Private investors and funds have experience, knowledge about the industry, and the ability to identify entrepreneurs with high growth potential. The matching grant from TIPS provides incentives for investors to make high-risk initial investments. In other words, TIPS plays a bridging role, catalyzing the ecosystem of private investors and startups by providing financial and technical resources through a functioning coordination mechanism. The effectiveness of this collaboration structure corroborates evidence from another public–private partnership program in the United States, the Small Business Investment Company Program,21 which combines funds from private investors with public funds to support small businesses and is intended to support high-performing, innovative small businesses and has been considered a successful program for job creation (Paglia and Robinson 2017). Future studies could explore whether this type of public-private collaboration for funding technology startups is more effective than more-traditional co-investment schemes in which public funds are invested on the same basis as private funds. 19 Due to data limitations, the sustainability of the impact of TIPS (i.e., whether and for how long the impact of TIPS lasts compared to firms that were not selected for TIPS) is not addressed in the present study. Furthermore, the research could not examine whether TIPS’s R&D support is cost-effective compared to other Korean R&D support programs. These areas could be addressed in future research to more wholistically examine the effects of TIPS. 20 There have been concerns about duplication of support programs as well; a study found that 22 companies received financing support from the government’s start-up support fund at least 65 times each from 2011 to 2013 in Korea (Baek 2017), highlighting inefficiencies in disbursement of financial support to firms. 21 More information is available at Small Business Investment Company Overview (https://www.sba.gov/document/support-- sbic-program-overview) 41 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Second, appropriate implementation of a public-private co-investment model can have a multiplier effect on private investment by reducing the risk of investment in early-stage startups. Because of high levels of uncertainty and risk, many private investors avoid investing in early-stage firms, forgoing opportunities to help growth-oriented startups expand. Public funds (e.g., matching funds, funds of funds, grants) can play an important role in facilitating investments in early-stage firms because they can reduce the cost of investment failure. In the case of TIPS, the match ratio between the private (initial private investment by a TIPS partner) and public (govern- ment matching grant) sectors is 1:5.22 The government’s share is disproportionately higher than the initial private investment, and the risk-sharing mechanism could reduce private investors’ risk and crowd in more investments in the ecosystem over the medium to long term. For instance, as of the end of 2022, the cumulative amount of follow-on investments from private investors (KRW 10.3 trillion; USD 7.9 billion) was about 10.4 times the amount of public spending on TIPS (KRW 1 trillion; USD 774 million) (KBAN 2023).23 Third, the effectiveness of an entrepreneurship program may depend on complementary sup- port programs, which could target different stages of the startup lifecycle. TIPS support can be extended up to three years, a relatively short period if the objective is to prepare startups for successful exit or initial public offering, which can take longer than 10 years and several funding rounds. With the goal of creating more scale ups, the MSS created Post-TIPS, a complementary program that targets startups that have completed TIPS and are younger than seven years. Post- TIPS provides financial support of up to KRW 500 million (USD 386,000) over 18 months to help successful TIPS beneficiaries further expand their businesses. Another complementary program is Pre-TIPS, which was introduced in 2018 to identify and support startups that are less than three years old and are being nurtured by angel investors but are not yet qualified to apply for TIPS.24 Pre-TIPS serves as a steppingstone for early-stage entrepreneurs before they are accepted to TIPS. While Pre- and Post-TIPS are smaller in scale compared to TIPS, the Korean government has been putting emphasis on providing complementary support to startups, tailoring assistance based on their specific lifecycle stages.25 Nevertheless, the impact of these complementary programs has not been assessed rigorously. 22 The match ratio is based on the marching grant for R&D activity and excludes the government’s additional support for non- R&D activities such as commercialization and overseas marketing that are also included in the TIPS support package. The match ratio is not fixed and may change depending on the initial investment amount a TIPS partner makes. More details on the different types of TIPS support and their conditions are available in Appendix B. 23 Another example illustrating the multiplier effect is the fund of funds that the Korea Venture Investment Corporation manages. From 2017 to 2021, the Korean fund of funds accounted for an average of 20 percent of new VC funds raised in Korea (KVCA 2022). Over the past two decades, the fund of funds has mobilized private co-investment in VC funds that is 5 times the corporation’s funding, suggesting that the fund of funds has had a significant crowding in effect (Kwak 2019). 24 In addition to nurturing early-stage technology startups, another policy goal of Pre-TIPS is to support geographically balanced growth of technology entrepreneurship in Korea by allocating more than 60 percent of the budget to startups in non-capital regions. 25 In addition to Pre-TIPS and Post-TIPS, other programs in Korea support startups at different stages of development. Although firms are generally allowed to apply for and benefit from multiple programs, the Korean government introduced restrictions in eligibility criteria to prevent public support and resources from being concentrated in a small number of beneficiaries. Such measures are intended to improve efficiency of the allocated resources. In 2022, TIPS supported 500 startups whereas Pre-TIPS supported 30 and Post-TIPS 50 startups. 42 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Fourth, patient capital and continuity in entrepreneurial policy with a long-term view are key to nurturing a vibrant startup ecosystem. Successful exits, mostly through mergers or ac- quisitions, and IPO ultimately create spillovers to the ecosystem in terms of financial returns to investors and knowledge spillovers through spinoffs and serial entrepreneurs who go on to start their next venture. However, most startups fail to cross the valley of death, and the very few ven- ture-backed startups that survive and grow typically need at least 7 to 10 years to exit. Thus, policymakers should recognize that the entrepreneurial agenda requires a longer time horizon for maturity than other policy agendas. In other words, developing an entrepreneurial ecosystem requires a significant amount of capital, time, and commitment from stakeholders and policymak- ers. TIPS has been expanded over three consecutive administrations, with its budget increasing nearly 11-fold between 2014 and 2020. As a result, according to an analysis of KBAN data in Kim et al. (2023), a total of 81 TIPS beneficiaries successfully exited to date. 68 startups exited through M&A with a cumulative M&A volume of KRW 1.9 trillion (USD 1.5 billion), and 13 startups filed IPOs. On average, TIPS startups that exited took 4.5 years for M&A (3 years since TIPS selection) and 6.9 years for IPO (5.1 years since TIPS selection). Finally, constant engagement with beneficiaries through data monitoring and collection helps the program develop a dynamic monitoring and evaluation mechanism. Although the government maintains minimum control over the execution (e.g., selection, operations) of TIPS, it constantly monitors and evaluates all stakeholders involved in the program. For instance, start- ups selected for TIPS are encouraged to move into the TIPS partners’ designated incubation space and participate in regular mentorship meetings and consultations. If a startup is not located in the designated spaces, it must regularly report on its progress and performance, such as by pro- viding mentoring reports and emails to the supervisory agency. The government also periodically reviews the performance of TIPS partners and (re)designates their status as official TIPS partners. Therefore, TIPS partners have a strong incentive to monitor, evaluate, and boost the performance of the startups that they support. This monitoring and evaluation structure helps ensure effective operation of the public-private partnership and thus is highly recommended for similar programs. Box 1. Policy Considerations for Replicability in Developing Economies In addition to resource constraints, there are a number of limitations that could hamper success- ful replication of a TIPS-like program in less-developed entrepreneurial ecosystems. For the model to be successful, it must be tailored to meet local conditions and implementation capabilities and compensate for missing complementarities. Four main limitations are discussed here. First, the legal and regulatory environment might not be conducive to implementation of a pub- lic-private partnership model wherein private sector actors are entrusted with investing public money. Additionally, some countries prohibit public institutions from making investment losses and discourage bureaucrats from taking risks that could result in prosecution for mishandling 43 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS public money. Thus, it is important for governments to provide an appropriate legal environment for governance and operation of a TIPS-like scheme to be successful. Second, emerging and developing entrepreneurial ecosystems usually lack the investor base of angels and angel networks, VC and private equity funds, accelerators, entrepreneurship support organizations, and mentors who should be the key implementation partners of a TIPS-like pro- gram. Regional or international funds and a small number of local funds competing for a small number of deals usually drive investment activities in such ecosystems. Korea faced a similar challenge when TIPS was started in 2013, and it depended on foreign funds in the early stages of program implementation.26 Thus, policymakers in emerging ecosystems should recognize that a secondary objective of a TIPS-like program would be development of the local investor base and would-be implementation partners of the program. This could also be achieved by introducing complementary policy instruments that target development of new funds through a fund of funds or co-investment funds to invest in deals that other funds lead. This is what Korea did with es- tablishment of its fund of funds in 2005 and other complementary instruments that eventually resulted in a sharp increase in the number of VC funds in the 2010s. Third, emerging ecosystems face constraints beyond access to finance that can limit the flow of quality investable and growth-oriented startups, including outdated and burdensome regulato- ry environments, shortages of technical and managerial skills, low rates of technology adoption and diffusion, weak research and innovation capabilities, weak incentive frameworks for tech- nology transfer, and lack of related professional services. Thus, any TIPS-like policy intervention that targets financial constraints should be coupled with interventions and services that alleviate demand-side constraints to increase the flow of innovative startups and improve ecosystem con- ditions generally. This is evident from the combination of services that TIPS offers, which includes, in addition to financial support, access to infrastructure and R&D facilities and mentoring and networking services. Finally, although the matching grant mechanism remains viable, targeting of a TIPS-like program in an emerging ecosystem could be adjusted. Korea’s TIPS targeting of technology and R&D-based startups should be understood in the context of the government’s objective of searching for new growth industries and diversifying the economy. The program targets innovative technology start- ups that are less than seven years old and extends financing for R&D activities, hiring of high- skilled workers, adoption of technology, and the like. The technology used and business feasibility were important selection criteria during the evaluation stage. More than half of the employees of TIPS beneficiaries have master’s and doctoral degrees, and approximately 30 percent come from large firms such as Samsung, Naver, Kakao, and SK (Han 2019). In an emerging ecosystem, a TIPS- like program is more likely to be effective by targeting a wide subset of startups—not necessarily just those that are R&D and knowledge intensive, but also those that are innovative in their local context (developing products and services that are new to their respective market rather than new to the world). In that sense, the program must respond to market demand through its imple- mentation partners and their ability to identify investable deals. As the ecosystem develops and evolves, so does the targeting of the program and its ability to be selective. 26 For instance, the MSS and Korea Venture Investment Corp opened start-up investment centers in Silicon Valley, United States, in August 2013 and in Shanghai, China, in June 2014 to attract foreign capital. 44 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS References 45 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS References Acs, Z. J., P. Braunerhjelm, D. B. Audretsch, Bianchi, M., S. Murtinu, and V. G. Scalera. and B. Carlsson. 2009. The Knowledge 2019. R&D Subsidies as Dual Signals in Spillover Theory of Entrepreneurship. Small Technological Collaborations. Research Policy Business Economics 32: 15–30. 48 (9): 103821. Aghion, P., R. Blundell, R. Griffith, P. Howitt, Brown, J. D., and J. S. Earle. 2017. Finance and and S. Prantl. 2009. The Effects of Entry on Growth at the Firm Level: Evidence from SBA Incumbent Innovation and Productivity. Review Loans. Journal of Finance 72 (3): 1039–80. of Economics and Statistics 91(1): 20–32. Cantner, U., and S. Kösters. 2015. Public Anton, J. J., and D. A. Yao. 1994. Expropriation R&D Support for Newly Founded Firms— and Inventions: Appropriable Rents in the Effects on Patent Activity and Employment Absence of Property Rights. American Growth. Journal of Innovation Economics and Economic Review 81 (1): 190–209. Management 1: 7–37. Audretsch, D.B., Keilbach, M.C. and Lehmann, Carpenter, R. E., and B. C. Petersen. 2002. E.E., 2006. Entrepreneurship and economic Capital Market Imperfections, High-Tech growth. Oxford University Press. Investment, and New Equity Financing. Economic Journal 112 (477): F54–72. Baek, H. 2017. Establishment of an Efficient SME Support Policy System. KOSBI SME Focus Chen, J., C. S. Heng, B. C. Tan, and Z. 17–08. Seoul, Korea: Korea Small Business Lin. 2018. The Distinct Signaling Effects of Institute. [Korean] R&D Subsidy and Non-R&D Subsidy on IPO Performance of IT Entrepreneurial Firms in Bai, J., S. Bernstein, A. Dev, and J. Lerner. China. Research Policy 47 (1): 108–20. 2021. The Dance Between Government and Private Investors: Public Entrepreneurial Choi, S., Lee, Y.S., Kim, T. and Kim, W., 2021. Finance Around the Globe. National Bureau of Stay the Course? The Effects of Government Economic Research, Cambridge, MA. Matching R&D Funding on Tech Startups. Academy of Management Proceedings (Vol. Berger, M., and H. Hottenrott. 2021. Start-Up 2021, No. 1). Briarcliff Manor, NY 10510: Subsidies and the Sources of Venture Capital. Academy of Management. Journal of Business Venturing Insights 16: e00272. 46 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Conti, A. 2018. Entrepreneurial Finance and Hellmann, T., and V. Thiele. 2019. Fostering the Effects of Restrictions on Government R&D Entrepreneurship: Promoting Founding or Subsidies. Organization Science 29 (1): 134–53. Funding? Management Science 65 (6): 2502– 21. Dimos, C., and G. Pugh. 2016. The Effectiveness of R&D Subsidies: A Meta- Hottenrott, H., and R. RichsteinR. 2020. Regression Analysis of the Evaluation Start-Up Subsidies: Does the Policy Instrument Literature. Research Policy 45 (4): 797–815. Matter? Research Policy 49 (1): 103888. Doh, S., and B. Kim. 2014. Government Howell, S. T. 2017. Financing Innovation: Support for SME Innovations in the Regional Evidence from R&D Grants. American Industries: The Case of Government Financial Economic Review 107 (4): 1136–64. Support Program in South Korea. Research Policy 43 (9): 1557–69. Islam, M., A. Fremeth, and A. MarcusA. 2018. Signaling by Early Stage Startups: US FSC (Financial Services Commission). 2008. Government Research Grants and Venture Market-Friendly SME Finance Support Plans Capital Funding. Journal of Business Venturing (unofficial translation). Press release, June 11, 33 (1): 35–51. 2008. (Korean). Kaplan, S. N., and P. StrömbergP. 2003. Gans, J., and S. Stern. 2003. When Does Financial Contracting Theory Meets the Real Funding Research by Smaller Firms Bear Fruit?: World: An Empirical Analysis of Venture Evidence from the SBIR Program. Economics of Capital Contracts. Review of Economic Studies Innovation and New Technology 12 (4): 361–84. 70 (2) 281–315. Gao, Y., Y. Hu, X. Liu, and H. Zhang. 2021. KBAN (Korea Business Angels Association). Can Public R&D Subsidy Facilitate Firms’ 2023a. Newsletter. Volume 28. January. Exploratory Innovation? The Heterogeneous Effects Between Central and Local Subsidy KBAN (Korea Business Angels Association). Programs. Research Policy 50 (4): 104221. 2023b. Newsletter. Volume 29. April. Giraudo, E., G. Giudici, and L. Grilli. 2019. Kim, H., and C. K. Kim. 2018. The Effects Entrepreneurship Policy and the Financing of of Network Between Business Founder and Young Innovative Companies: Evidence from Investor on Enterprise Performance: Focused the Italian Startup Act. Research Policy 48 (9): on TIPS Support Business. Asia-Pacific Journal 103801. of Business Venturing and Entrepreneurship 13 (3): 47–57. Gries, T., and W. Naudé, 2009. Entrepreneurship and Regional Economic Kim, M., and J. W. Kang. 2021. Growth: Towards a General Theory of Start- Entrepreneurship Policy Changes from Ups. Innovation–The European Journal of the Perspective of Policy Paradigm. Asia- Social Science Research 22 (3): 309–28. Pacific Journal of Business Venturing and Entrepreneurship 16 (3): 43-58. Han, J. W. 2019. Promotion of Technology- Based Start-Ups: TIPS Policy of Korea. Asian Journal of Innovation and Policy 8 (3): 396– 416. 47 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Kim, S., Bae, S., and Kim, K. 2023. Startup NABO (National Assembly Budget Office). investment ecosystem and growth analysis II: 2023. NABO Fiscal Estimates and Tax Issues. Performance and future tasks of TIPS. STEPI Vol 23. June. Insight 314. Norton, E., and B. H. Tenenbaum. 1993. Koo, J. 2018. Does Early Incubating Improve Specialization Versus Diversification as a the Performance of Start-Ups? Evidence from Venture Capital Investment Strategy. Journal TIPS in Korea. KDI Journal of Economic Policy of Business Venturing 8 (5): 431–42. 40 (2): 75–96. Paglia, J. and D. T. Robinson. 2017. Measuring KOSTAT. 2022. Business Demography Statis- the Role of the SBIC Program in Small tics in 2021. https://kostat.go.kr/board.es?mi Business Job Creation. A Report Prepared d=a20104100000&bid=11731&act=view&li by the Federal Research Division, Library of st_no=423089 Congress. https://www.sbia.org/wp-content/ uploads/2021/04/Library-of-Congress-SBIC- KVCA (Korean Venture Capital Association). Jobs-Report-Jan-17.pdf. 2022. Statistics and Trends of Domestic Venture Capital Investments: First Quarter of Salmenkaita, J.P. and Salo, A., 2002. 2022. KVCA, Seoul, Korea. Rationales for government intervention in the commercialization of new technologies. Kwak, G. 2019. Effects of the Fund-of-Funds Technology Analysis & Strategic Management, on the Domestic Venture Capital Industry. 14(2), pp.183-200. Korea Venture Investment Corporation Market Watch, vol. 10. Spender, J.-C., V. Corvello, M. Grimaldi, and P. Rippa. 2017, Startups and Open Innovation: A Lee, D.S. and Lemieux, T., 2010. Regression Review of the Literature. European Journal of discontinuity designs in economics. Journal of Innovation Management 20 (1): 4-30. economic literature, 48(2), pp.281-355. Storey, D. J., and B. S. Tether. 1998. Public Luo, X., F. Huang, X. Tang, and J. LiJ. 2021. Policy Measures to Support New Technology- Government Subsidies and Firm Performance: Based Firms in the European Union. Research Evidence from High-Tech Start-Ups in China. Policy 26 (9): 1037–57. Emerging Markets Review 49: 100756. Szczygielski, K., W. Grabowski, M. T. Pamukcu, MSS (Ministry of SMEs and Startups). 2022. and V. S. Tandogan. 2017. Does Government 2022년 창업지원사업 통합공고 [2022 Public Support for Private Innovation Matter? Programs Supporting Entrepreneurship]. Firm-Level Evidence from Two Catching-Up Countries. Research Policy 46 (1): 219–37. ———. 2016. 보도자료 – 2015년 엔젤투자, 1,000억원 돌파. Press Release, July 26, 2016. Teece, D. J., ed. 1987. The Competitive Challenge: Strategies for Industrial Innovation ———. 2023. Announcement No. 2023-43: and Renewal. Pensacola, FL: Ballinger Integrated Announcement of the 2023 TIPS Publishing Company. (Tech Incubator Program for Startup) Startup Support Plan. 48 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Toole, A.A. and Turvey, C., 2009. How does initial public financing influence private incentives for follow-on investment in early- stage technologies? The Journal of Technology Transfer, 34, pp.43-58. Vanino, E., S. Roper, and B. Becker. 2019. Knowledge to Money: Assessing the Business Performance Effects of Publicly-Funded R&D Grants. Research Policy 48 (7): 1714–37. World Bank Group. 2023. Innovative Korea: Lessons on Leveraging Innovation and Technology for Development. World Bank, Washington, DC. Yang. H., and J. Park. 2020. Analysis of the Current Status and Development Tasks of Angel Investment in Korea. Journal of SME Finance 40 (2): 51-88. 49 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Appendices 50 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Appendix A. Overall Process of the Tech Incubator Program for Startup There are four main steps in implementation of the program. 1/ Application TIPS accepts applications from startups online year-round. One eligibility requirement is to have received an initial investment from a designated TIPS partner. TIPS partners are responsible for initial selection of TIPS beneficiaries and recommend selected firms to the Korea Business Angels Association (the TIPS management agency). To be selected for the program, TIPS partners are typ- ically required to complete the initial investments in startups, although depending on the specifics of the private contract, it is possible to apply with a commitment letter (specifying investment amount, share quantity, and payment deadline), in which case, the commitment letter is verified at the point of selection, and firms that have received a private investment can be selected for TIPS. 2/ Proposal evaluation and selection Once the application is submitted, the KBAN screens the proposal, followed by qualitative and quantitative assessments, including an interview with the startup’s founders; a review of the start- up’s business plans and financial statements; and analysis of performance metrics such as revenue growth, customer acquisition, and product development milestones. The R&D Project Evaluation Committee, which is composed of approximately seven third-party experts with backgrounds in various industries who are typically entrepreneurs, venture capitalists, angel investors, and tech- nology commercialization professionals, evaluates applications. Individuals from the MSS, KBAN, KISED, and TIPS partners do not participate on this evaluation committee to ensure that startups are selected based on the market demand. These committee members are responsible for assess- ing the performance of participating startups and providing feedback to program administrators. TIPS applicants are evaluated based on a comprehensive set of predefined criteria outlined in the TIPS general operational guidelines.27 The guidelines detail selection criteria, application process, scoring systems, and other detailed conditions related to TIPS operation. The evaluation focuses on technical merit (40 percent), business feasibility (40 percent), and organizational capabilities (20 percent). Applicants are scored on a 100-point basis, with possible extra points awarded based on whether they are headquartered in a nonmetropolitan area (2 points); their technology 27 The Ministry of SMEs and Startups updates and releases the TIPS general operational guidelines every year to provide details about the selection processes, criteria, and other major rules related to TIPS operation. The most recent guidelines are available in MSS 2023. 51 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS involves one of the top-10 cutting-edge startup fields in materials, components, and equipment (1 point); and they have passed Pre-TIPS or TIPS-R28 (1 point). The R&D Project Evaluation Committee scores the startups by calculating the arithmetic mean of the total scores, excluding the highest and lowest scores. There is no absolute cutoff score, and the selection is made based on each cohort’s score distribution. The results are reported and announced to accepted applications. Rejected startups can reapply after improving their application. Evaluation Score = Sum of Evaluation Scores - (Highest Score + Lowest Score) Number of Evaluators - 2 Overall Score = Evaluation Score + Extra points (3 points max) 3/ Program implementation From application submission to selection, the process takes approximately two months. Upon selection, beneficiaries receive matching grants from the TIPS program during the two-year TIPS incubation period. If the startup applies for additional support programs (commercialization, overseas marketing), the incubation period can be extended to up to three years. TIPS partners mentor selected startups, who are sometimes required to work in working spaces with the TIPS partners. The TIPS town is a co-working space that the KISED and KBAN manage for TIPS beneficiaries. 4/ Completion After two years of participation in TIPS, awardees are required to submit a report on the outcomes, including any commercialization outcomes and economic impact, for a final evaluation. The eval- uation criteria include whether the startup adhered to the planned expenditures as outlined in their original business plan, provided all required reports and documents, and diligently executed their business operations to achieve the goals stated in the business plan. If a project is found to be successful,29 startups must pay back 10 percent of the government R&D funds as a loyalty pay- ment. The government does not seek equity in return for their investment. Startups that successfully complete the program may seek additional public support and apply for other government programs, for instance, through Post-TIPS. Although startups that fail to complete TIPS successfully do not face any penalties, TIPS partners that fail to meet performance measures could be banned from participating as partners the following year. 28 TIPS-R is a program that provides financial support to startups that have failed experience and are relaunching their businesses. 29 Startups are considered to have successfully commercialized if they completed a M&A worth more than KRW 1 billion, exceeded KRW 1 billion in annual revenue from technology commercialization, received a follow-on investment, filed an initial public offering, achieved annual export volume of USD 500,000 or more, or increased staffing by more than 20 employees. 52 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Appendix B. Types of Tech Incubator Program for Startup Support Although the core of TIPS lies in financial support, TIPS provides a range of resources and compre- hensive supports for startups, which can be categorized as follows. 1. Financial support: When applying for TIPS, a R&D expenditure plan must be submitted. If the TIPS partner completes its initial investment in a TIPS applicant of KRW 100 million (USD 77,000), the government provides R&D matching grant of up to KRW 500 million (USD 387,000). To ensure that the startup has the ownership of the company, the TIPS partner’s share of equity should not exceed 30 percent. Government matching grant can cover up to 80 percent of total R&D expen- diture. If the TIPS awardee requests additional financial support, it may receive an additional KRW 400 million (USD 309,000) for overseas marketing, technology commercialization, and matching an- gel funds. In sum, TIPS awardees can receive a maximum KRW 1 billion (USD 770,000) of private and public support during the three years of TIPS incubation period. Table B.1. Tech Incubator Program for Startup (TIPS) Funding Total R&D expenditure(A+B) Additional financial Initial investment by Government matching Share allocated support from the TIPS partner grant to TIPS startup government Period (KRW 100 mil) (Max KRW 500 mil) (A) (B) (Max KRW 400 mil) 2-3  RW 100 million •K  overnment can •G The proportion has •  • KRW 100 million (USD years (USD 77,000) support up to KRW to be greater than 77K) for technology 500 million in R&D 20% of total R&D commercialization •T  he initial investment (USD 387,000) expenditure (A+B) by the TIPS partner •KRW 100 million (USD has to be greater  he proportion •T 77K) for overseas than 20 percent has to be less than marketing assistance of the government 80% of total R&D matching grant (A) expenditure (A+B) KRW 200 million (USD • 154K) in matching angel funds Source: KBAN. 53 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS 2. Mentorship and advisory services: Participating startups gain access to a network of men- tors and industry experts who can provide guidance and support on business strategy, market research, product development, and commercialization. Although TIPS partners mainly provide these services, the Korea Business Angels Association also provides opportunities for TIPS start- ups and partners to interact and network. 3. R&D facilities: Universities and research institutions that partner with TIPS provide R&D facili- ties, equipment, and resources. 4. Co-working space: TIPS offers office space, meeting rooms, and conference halls for partic- ipating startups. Sponsored startups may use the space that their TIPS partners provide or TIPS Town, the incubating space that the Korea Institute of Startup and Entrepreneurship Development provides. 5. Networking opportunities: TIPS provides opportunities for participating startups to network with potential investors, venture capitalists, customers, and business partners through events such as demo days and industry conferences. 6. Other support: Accelerators and incubators provide startups with other support such as legal and accounting services and access to additional investment opportunities. 54 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Appendix C. Eligible Tech Incubator Program for Startup Expenditures Eligible TIPS expenditures may vary depending on the needs and stage of development of partici- pating startups. Generally, participating startups can spend TIPS funding in such areas as: Research and development: Many participating startups use their TIPS funding to cover pro- totyping, testing, and product development costs, including expenses related to designing new products, conducting user research and obtaining feedback, and refining existing products. Marketing and advertising: TIPS funding can be used in campaigns to increase brand awareness and acquire new customers, including branding, social media marketing, paid advertising, and market research. Salaries and wages: If employees are directly involved in product development and business op- erations, their salaries and wages can be covered with TIPS funding, as can hiring new staff and providing bonuses and incentives. Professional services: TIPS funding can be used to pay for legal, accounting, and consulting ser- vices, including securing patents, preparing financial statements, and seeking advice on business strategy. Travel and training: Startups may use TIPS funding to cover expenses for attending business events, conferences, and training programs, including transportation, lodging, and registration fees for events and programs aimed at building business skills and networking. Equipment and infrastructure: Equipment and software can be purchased or leased with TIPS funding, including expenses related to computers, servers, software licenses, and other infra- structure needs. 55 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Appendix D. Comparison of public-private matching grant programs for technology startups Table D.1 compares Korea’s TIPS with Israel’s Technological Incubators Program, which Korea benchmarked after, and the US’s Small Business Innovation Research (SBIR) programs. Although the specifics are different, these matching programs share a common objective of nurturing inno- vative startups by bridging the funding gap and leveraging private sector resources and expertise. Table D.1. Comparison of three government matching programs Tech Incubator Program Technological Incubators Small Business Innovation Program for Startup (TIPS) Program (TIP) Research (SBIR)2 Country Korea Israel United States Launched in 2013 1991 1982 Line ministry Ministry of Startups Ministry of Economy Small Business and SMEs (MSS) and Industry Administration (SBA) Implementation Korea Business Angel Israel Innovation Authority (IIA)1 SBA’s Office of Investment arm Network (KBAN) & Innovation Target Technology startups that Technology startups Startups and small businesses beneficiaries are less than 7 years old across technology areas (under 500 employees) Financing The match ratio of financial A government provides grant of To apply for Phase II – Bridge through support between the private up to 85% and the investment program, the applicant needs public-private and public is approximately financing from the private to receive investment from partnership 1:5. Depending on the incubators supplement of up the private investors, and additional support startups to 15 percent of the approved the government matches the receive, the matching budget. No financial investment grants with private investment ratio may go up to 1:10. is required by the startup. with 1:1 matching ratio. Amount of TIPS partner completes its Maximum NIS 3.5 million Phase I (concept development): government initial investment of KRW 100 (USD 913,000) USD 50,000 – 250,000 commitment to million (USD 77,000), and Phase II (prototype the program the government provides R&D development): USD beneficiaries matching grant of up to KRW 500,000 – 1.5M 500 million (USD 387,000). Phase II-Bridge: Maximum USD 3M of matching fund (USD 1 million for 3 years) 56 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Tech Incubator Program Technological Incubators Small Business Innovation Program for Startup (TIPS) Program (TIP) Research (SBIR)2 Government Grant Grant Grant and matching grant financing type Private 111 TIPS partners 15 incubators VC, angel investors, big firms investors Program Up to 2 years with an Up to 2 years with an Phase I: 6 months – 1 year; duration additional 1 year additional 1 year Phase II: 24 months; Phase II-Bridge: 3 years Note 1: IIA replaced the Office of the Chief Scientist in January 2016. Note 2: Public grant is awarded to the SBIR beneficiaries during Phase I (concept development) and Phase II (prototype devel- opment). The matching grant is awarded during the Phase II-Bridge stage, and it is for those that already completed both Phase I and II of the program. 57 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Appendix E. Deeptech Tech Incubator Program for Startup In 2023, the Korean government introduced a program called the Deeptech Tech Incubator Pro- gram for Startup (TIPS), which is aimed at startups involved in the top 10 technology sectors desig- nated by the Korean government. Following a process similar to that of the original TIPS, startups receive an initial investment of KRW 300 million (USD 231,000) from private investors and are then recommended for government research and development (R&D) funding of up to KRW 1.5 billion (USD 1.16 million). 10 technologies sectors for Deeptech TIPS System semiconductors Biotechnology and health care Future mobility Green energy Robotics Big data and artificial intelligence Cyber security and networking Aerospace and maritime Next-generation nuclear power Quantum technology In 2023, the government announced that the program would support 120 projects with a total budget of KRW 31.5 billion (USD 24.3 million) allocated for R&D support. Similar to TIPS, additional support such as funding for commercialization and overseas marketing (KRW 100 million; USD 77,000 each) are available through a separate evaluation process. The program also provides non-financial assistance, including incubation, mentoring, and business consultation. To be eli- gible for application, startups must be less than seven years old and have less than KRW 2 billion (USD 1.5 million) in annual sales. Startups currently participating in TIPS or Pre-TIPS are not eligible but may apply when they complete that program. 58 The Effects of Matching Grants on Technology Startups: The Case of Korea’s TIPS Seoul Center for Finance and Innovation Website: https://www.worldbank.org/en/programs/seoul-center-for-finance-and-innovation Seoul Center for Finance and Innovation