Executive Summary

1. Introduction

2. Model

3. Data and Calibration

4. Assessing Debt Sustainability in the APEC Economies

5. Conclusion

References

Appendix
A. OLG Model and Calibration
B. Dynamics of the Debt-to-GDP Ratio
C. Additional Tables

In this paper, we estimate additional government expenditure used to reduce AI’s existential risk and assess public debt sustainability. Our most important policy-relevant finding is that even under the conservative assumption that government expenditure equals the maximum amount society is willing to sacrifice to mitigate AI risk, and that all such expenditure is financed by issuing sovereign bonds rather than raising taxes, government debt does not necessarily become explosive. Instead, in our benchmark scenario—where the growth-enhancing effect of AI is calibrated to the average of prior studies—the debt ratio remains sustainable for most APEC economies. We also find that, except for Russia, an additional AI-driven growth effect of 3.4–6.1% would suffice for debt financing to remain sustainable for many APEC economies. In particular, for the United States, the required effect is 3.8% or 4.6%, depending on parameter assumptions. For faster growing economies such as China and Korea, the required additional effect is even smaller than for the United States.

Executive Summary

1. Introduction

2. Informal Economy and Composition of ODA in the Philippines

3. Model

4. Macroeconomic Equilibrium

5. Results

6. Conclusion

References

We analyze how the size and composition of official development assistance (ODA) shape aggregate performance and informality in the Philippines using a small open-economy dynamic general equilibrium model with a formal and informal sector. Two main scenarios are considered: (i) an increase in total ODA and (ii) a higher share of tied aid, given a fixed amount of ODA. Both scenarios raise capital, output, and consumption in steady state, but through distinct mechanisms. The first scenario primarily increases demand and appreciates the relative price of informal goods, expanding informality. By contrast, the second scenario expands public capital, crowds in formal investment, lowers the relative price of informal good, and shifts resources toward the formal sector, despite short-run reallocation costs.

Executive Summary

1. Introduction

2. Theoretical Framework

3. Data

4. Methodology

5. Results

6. Discussion

7. Conclusion

References

Appendix

This paper investigates how environmental, social, and governance (ESG) conditions, together with economic factors, shape fertility dynamics in APEC economies. Using World Development Indicators from 1996 to 2021 and assembling more than 1,400 indicators, we predict annual changes in the crude birth rate. Models are trained on non-APEC economies and tested out of sample on APEC economies. Random Forest achieves the lowest RMSE at 0.397, and models that combine ESG and economic variables outperform those relying on economic indicators alone, with RMSE values of 0.271 and 0.298 respectively. SHapley Additive Explanations (SHAP) reveal that environmental factors are the most influential predictors of fertility in APEC, followed by governance, while social factors are smaller in magnitude but show increasing importance over time. Lag analysis indicates short-run effects for social and governance variables at one-year lags and medium- term cumulative effects for environmental variables, peaking around four years. Country-level profiles highlight clear heterogeneity: environmental drivers dominate in China and Russia, governance factors are most important in Korea and the United States, and social influences are stronger in Canada and Japan. The study provides a comprehensive, externally validated, and interpretable framework for fertility prediction, while emphasizing that the analysis remains predictive and associational rather than causal. We regard this as a first step toward more rigorous causal evaluation of the highlighted drivers.

The importance of tariff measures in recent international trade has been highlighted again with the universal tariffs and reciprocal tariffs of the second Trump administration in the United States, as well as the retaliatory tariffs from trade partners. Interest in their economic effects is growing day by day. To properly analyze these economic effects, it is essential to accurately estimate the tariff elasticity of trade. This study comprehensively analyzes the main content of previous researches on the tariff elasticity of trade through a literature review, presenting various theoretical grounds and analytical methodologies. We applied these to Korea’s 2014 import clearance data to empirically estimate tariff elasticity. The main contents and analysis results of this study are as follows.

First, this study introduces the basic theoretical model of Fontagné et al. (2022) and derives an equation to estimate tariff elasticity. According to this, the tariff elasticity of trade can be estimated by the elasticity of substitution between items. If a consumer can easily substitute another product when the tariff of a specific product rises, the import of that product is expected to decrease more actively.

Second, this study suggests the possibility of developing a more realistic model based on the basic theoretical model of Fontagné et al. (2022). Particularly, the development of theoretical models and empirical analysis strategies that appropriately reflect other factors affecting tariff elasticity in addition to the elasticity of substitution between items is necessary. These suggestions can be categorized into 1) the level of hierarchy in the CES utility function, 2) consideration of the export supply function, 3) the possibility of partial pass-through, 4) consideration of other trade costs, 5) resolution of endogeneity issues, and 6) consideration of the customs valuation.

Third, this study investigated several empirical analysis studies on the tariff elasticity of trade. The results showed that the estimated value of tariff elasticity of trade ranged from an absolute value of 2.5 to 5.1 on average, but the variance was quite large depending on the sample, data characteristics, and analysis methodology. By country characteristics, it was confirmed that the greater the market dominance of the importing country, and when the exporting country is a developed country, the smaller the tariff elasticity of trade. Also, tariff elasticity was relatively large when there was a bilateral trade agreement, but smaller in cases of past colonial relationships, large price gaps, or great geographical distances. By industry and item characteristics, homogeneous and standardized products such as agricultural and mineral products had higher tariff elasticity, while heterogeneous products like machinery and textiles had lower tariff elasticity. In terms of industrial structure, the higher the use and diversity of intermediate goods, the greater the productivity differences between items within the industry, and the more products traded on regular exchanges, the higher the tariff elasticity of trade. Regarding market competitiveness and structure, the more intense the market competitiveness, the higher the tariff elasticity, but items with larger profit margins had lower tariff elasticity. It was also confirmed that various non-price competitive factors such as technological advantages and brand awareness were important. By firm characteristics, tariff elasticity of trade was greater when strategic complementarity between firms was high but smaller when firm productivity was high. Regarding data characteristics, the finer the aggregation level of item classification, the greater the tariff elasticity appeared. It was also confirmed that the analysis period and scope, data source, and consideration of proxy variables could also affect changes in estimated tariff elasticity.

Fourth, this study applied these discussions to the quarterly HS 6-digit import clearance data of Korea in 2014 to estimate tariff elasticity. The empirical analysis methodology mainly considered the Poisson Pseudo- Maximum Likelihood (PPML) estimation based on the structural gravity equation. The empirical analysis results estimated the absolute value of tariff elasticity to be about 5 to 10, confirming it to be similar to existing previous studies. However, it was found that the estimated tariff elasticity significantly differed statistically depending on the aggregation unit of the product, and the pattern of difference varied according to the setting of fixed effects. By product characteristics, the tariff elasticity of consumer goods was the highest, followed by intermediate goods and capital goods. By product group, the tariff elasticity of homogeneous items such as mineral products was estimated to be very large with an absolute value of 10 or more, but statistically significant elasticity was not observed in some product groups such as stone and ceramics. To verify the robustness of the main analysis results, both the log-linear equation and asymptotic bias-corrected PPML estimation method were applied, but the analysis results were qualitatively similar and statistically significant consistently.

Based on these analysis results, this study draws the following implications. First, the government should manage and pay close attention to the elasticity of substitution between similar products when establishing tariff policies. In particular, since the elasticity of substitution varies by industry and item, tailored tariff policies that consider the characteristics and substitutability of each industry and item are necessary. Especially for homogeneous products like agricultural and mineral products, meticulous policy design is required as they are sensitive to tariff changes. Second, the government should consider various factors in addition to the elasticity of substitution when establishing tariff policies, including economic and political relations with trade partners, strategic behavior of related firms, and market competition structure. Comprehensive consideration of these variables becomes a prerequisite for the government to design more effective and sustainable tariff and response measures. Third, when estimating tariff elasticity as a basis for policy, the aggregation method of data is crucial. In particular, it is important to divide the data in detail by item rather than total data to design the correct policy. Fourth, in analysis methodology, appropriate fixed effects should be selected by fully considering the analysis purpose and data characteristics. Fifth, when establishing and evaluating trade policies, careful consideration of other trade costs, such as freight insurance rates, in addition to tariffs is necessary. Lastly, regarding empirical analysis techniques for estimating tariff elasticity of trade, it is proposed that the general PPML estimation method is a sufficiently reliable and valid approach for analyzing tariff elasticity.

Since the inauguration of Trump’s second term, U.S. trade policy has triggered profound changes in the global trading order. The United States, prioritizing the prevention of illegal immigration and drug trafficking as matters of national security, has implemented stringent trade measures such as the imposition of high tariffs and the renegotiation of bilateral agreements. These policies have amplified uncertainty across the external environment of Latin America, including Mexico and Brazil. The recent U.S. decision to impose a 50 percent retaliatory tariff on Brazilian products starkly illustrates the vulnerability of Latin American trade to policy volatility. Against this backdrop, the upcoming USMCA renegotiation in July 2026 is likely to become another critical variable for the region’s external relations and supply chain configuration.

Despite the rising geopolitical and geoeconomic importance of Latin America, the region remains relatively marginal within Korea’s trade strategy. Yet, the ascent of the Global South, the restructuring of supply chains, and the diversification of trade partners underscore Latin America’s growing strategic value and the urgency for Korea to pursue proactive and long-term policy responses.

In this context, Japan’s experience in Latin America provides meaningful insights for Korea. Japan has consolidated its presence by focusing on traditional manufacturing sectors such as automobiles, machinery, and chemicals, establishing dual hubs in Mexico and Brazil while simultaneously diversifying into markets such as Argentina and Chile. Japanese firms have strengthened localization strategies by responding proactively to policy changes, including Brazil’s “Mover” program and Mexico’s environmental regulations. In the resource sector, Japan has sought stable access to critical minerals such as lithium and copper through close government–business collaboration, supported by financial and policy instruments that helped mitigate risks.

This study investigates Japan’s industry- and period-specific entry cases and government support policies to derive policy implications for Korea’s Latin America strategy. The key recommendations are as follows:
First, strengthen government–business linkages to institutionalize supply chain restructuring support.

Second, emphasize reference-building at the initial stage of entry to establish a foundation for long-term growth.

Third, respond proactively to environmental and regulatory changes to build trust with local governments and turn compliance into a source of competitive advantage.

Fourth, maintain a Brazil–Mexico-centered strategy while expanding into Argentina, Chile, and Colombia to diversify regional risks.

Fifth, explore opportunities for joint ventures with Japan, particularly in strategic sectors such as minerals.

In sum, Japan’s experience highlights essential lessons for Korea to achieve stable and sustainable outcomes in Latin America under conditions of global trade uncertainty. Rather than merely replicating Japan’s approach, Korea should design tailored strategies that reflect the specific characteristics of Korean firms and the diverse demands of Latin American economies.

This study comprehensively analyzes cross-border investment and spillover effects of artificial intelligence (AI) technology in the era of the 21st century’s paradigm shift toward an intangible asset-centered economy, and proposes policy directions for enhancing Korea’s AI industrial competitiveness. The research particularly focuses on exploring effective response strategies that technology-follower countries like Korea can adopt in a situation where global AI investment is extremely concentrated in the United States and China.

Examining the research background, the proportion of intangible assets in the market value of major global companies has surged from 17% in 1975 to 90% in 2020, with global AI investment reaching approximately $252.3 billion in 2024, a 17-fold increase from 2013. However, while the United States accounts for over 60% of global AI investment, Korea represents merely 1.5-2.0%, revealing an extreme polarization of technological hegemony with 93% of AI-related patents concentrated in the US and China. In this context, although Korea possesses successful technology catch-up experience in manufacturing sectors from the past, the AI era’s technological competition presents fundamentally different challenges due to the characteristics of intangible assets—high initial development costs and near-zero marginal production costs.

Starting from this problem awareness, this study systematically conducted historical experience analysis, global AI investment status surveys, theoretical model construction, and empirical analysis using the GVAR model. Analysis of Korea’s past technology spillover experiences revealed that Samsung Electronics’ case of overcoming a 5.5-year technology gap in the memory semiconductor sector demonstrated the importance of systematic technology partnerships and continuous R&D investment. In the IT hardware sector, economic benefits related to productivity improvement increased approximately 3.7-fold from 76.4 trillion won to 286.4 trillion won during the 2000-2005 period.

Through theoretical model analysis, the study confirmed that when intangible asset investment spillovers occur from technology-leading countries to technology-adopting countries, the adopting countries can achieve higher utility while investing less. Furthermore, under an appropriate spillover royalty system, benefits can accrue to both leading and adopting countries; however, the findings suggest that adopting countries need to maintain continuous intangible asset investment to prepare for potential spillover cessation by leading countries due to narrowing technology gaps.

The empirical analysis using the GVAR model yielded particularly noteworthy results. When Korea experienced an AI-related FDI outflow shock, domestic real GDP increased by 0.33%, while FDI inflow shocks resulted in only a 0.19% increase, confirming that technology acquisition through overseas investment may show stronger economic correlations. Stock indices showed a positive response of 0.75% to FDI outflow shocks and a negative response of -1.61% to FDI inflow shocks, indicating that markets tend to positively evaluate overseas AI investment from the perspective of securing future competitiveness.

Synthesizing these analytical results, this study presents the following policy implications. First, considering that overseas AI investment shows stronger economic correlations than domestic investment attraction, it is necessary to establish an institutional foundation that supports Korean companies’ investments and joint ventures with global AI companies and facilitates the return of acquired technology and talent to Korea. Second, given the 83-fold investment gap with the United States, rather than competing in all AI fields, a strategy of selection and concentration in specific areas such as AI hardware-software convergence that leverages Korea’s strengths in manufacturing and IT hardware capabilities is required. Third, while effectively utilizing technology spillovers, Korea must maintain an appropriate level of independent intangible asset investment to prepare for potential technology blockades by leading countries. Fourth, Korea should pursue a balance between international cooperation and independent innovation by strengthening partnerships with the United States while also participating in the establishment of AI technology cooperation networks within the Northeast Asian region.

In conclusion, this study suggests that for Korea to bridge the technology gap in the AI era and secure sustainable competitiveness, it needs to reinterpret past success patterns to fit the characteristics of the AI era and adopt a differentiated approach through selection and concentration. Particularly, an integrated strategy linking technology acquisition through overseas investment with the construction of a domestic innovation ecosystem is required, along with the establishment of a flexible policy framework capable of responding to the rapidly changing global AI technology environment.

Executive Summary

1. Introduction

2. Data

3. The Model

4. Qualitative properties

5. Numerical Simulation

6. Conclusion

Appendix

References

Digital platforms have become central to modern economies, as consumers increasingly spend leisure time online and platform firms transform user engagement into economic value. This study develops a multi-country, multi-industry open economy general equilibrium model to analyze how web traffic—measured as visits to platform domains—can be converted into “data capital” through big data and AI algorithms and then used as a non-rival input for production and innovation.

To ground the analysis, we construct a new dataset that links domain-level web traffic to corporate ownership across countries. The evidence reveals sharp cross country differences in the geographic origin of users. U.S. platforms attract a strongly international audience, with domestic users accounting for less than one-third of visits on average. Chinese platforms are more domestically concentrated, though TikTok stands out as a global exception with a highly international user base. Korean platforms, by contrast, rely overwhelmingly on domestic users, with less than ten percent of traffic originating abroad. Comparing these patterns with international service trade statistics, we find that countries with higher shares of foreign users are systematically more active in exporting online services.

Qualitative analysis provides further insight into user behavior. Consumers allocate more time to a platform when they experience greater content satisfaction, which depends on both technological quality and country-specific content preferences. When the quality of foreign platform content improves, consumers reallocate time away from domestic platforms toward foreign ones. The extent of this shift is influenced by the elasticity of leisure demand, the substitutability across platforms, and the share of time already devoted to foreign platforms.

The general equilibrium model allows us to simulate counterfactual scenarios and evaluate policy-relevant outcomes. Reducing cross-border frictions to foreign content raises welfare in all countries, primarily through the expansion of content variety and the reallocation of consumer time, while leaving the broader industrial structure largely unchanged. Lowering barriers faced by foreign platforms, such as discriminatory digital services taxes, increases platform revenues, encourages consumers to spend more leisure time online, and strengthens investment incentives for domestic platforms. Although the immediate welfare gains are modest, the longer-run effects on intangible investment and innovation are more significant. Enhancing the productivity of platform R&D generates asymmetric effects by expanding platform activity in the innovating country and reducing it in the non innovating country, yet global welfare increases because users everywhere benefit from improved content quality and diversity.

Taken together, the findings highlight that the accumulation of web-traffic-based data capital is a central driver of platform growth. Policies that improve access to foreign content, the taxation of digital services, and foster innovation in platform R&D directly influence how consumer time is allocated and how platform firms invest in data-driven growth. For policymakers, the results suggest that lowering barriers to cross-border online activity can deliver substantial welfare gains, that digital taxation should be carefully designed to balance fiscal and innovation objectives, and that investment in platform R&D is essential for competitiveness in the global digital economy.

In sum, this study demonstrates that digital platforms thrive not only on technological progress but also on the ability to accumulate and deploy data capital derived from web traffic. Recognizing the economic role of data capital is crucial for designing effective policies in the digital era, where market access, taxation, and innovation shape the international competitiveness of platform industries.

Executive Summary 1. Introduction 2. The model 3. Quantitative analysis 4. Conclusion Appendix References

Recognizing changes in the global economic order and its severe dependency on foreign resources and technology, Europe is greatly invested in strengthening its economic security and further developing its industries. Consecutively, European governments have introduced policies to enhance the competitiveness of their advanced industries. This study identifies the status, strengths and weaknesses of European advanced industries, by examining the present conditions of advanced industries in Europe in terms of trade, R&D expenditure, and human resources, and also comparing the competitiveness in advanced industries with the US and China. Furthermore, it identifies the policy direction of the EU and the UK for advanced industries from the perspectives of promoting innovation and bridging the technological gap, expanding investment, and fostering a level playing environment. This study also reviews the key support policies and regulations for specific sectors in advanced industries, i.e., batteries, semiconductors, AI & digital industry, health & biotech, clean technology, and aerospace.

The following points were confirmed through this research. First, both the EU and the UK recorded trade deficits in items classified as high-tech industries, indicating extensive dependence on foreign sources in these fields. The EU’s R&D expenditure rate also turned out to be less than that of the US and China. While the number of jobs in advanced industries are likely to increase, the EU’s low projection for working-age population in the future indicates that this vacancy is not expected to be filled, unlike in the US and China. These statistics reaffirm the necessity and legitimacy of European policies aimed at enhancing competitiveness in the advanced industry. Second, the EU and the UK are pursuing multifaceted policies to strengthen competitiveness in advanced industries under the overarching goal of climate neutrality. The EU aims to meet more than 40% of its demand for climate-neutral technologies internally by 2030, seeking to reinforce clean-tech production capacity. Initiatives such as Horizon Europe, the Strategic Technologies for Europe Platform (STEP), and the EU Blue Card are being employed to address key challenges and labor and skill shortages. In accordance with its Modern Industrial Strategy (2025), the UK is planning to actively invest in key sectors - automobiles, aerospace, biotech, and clean energy - to stabilize supply chains. In terms of utilization of various funds, the EU is actively promoting Public-Private Partnerships (PPP) projects and investment guarantees via InvestEU, while the UK is significantly increasing investment to strengthen its industrial ecosystem. Furthermore, both the EU and the UK are working to foster fair competition and to expand international cooperation while building internal supply chains and pursuing supply chain diversification. Third, in key strategic industries, both the EU and the UK are simultaneously pursuing large-scale investments and regulatory reforms to drive green and digital transitions along with technological sovereignty. In the batteries sector, the EU is continuously investing in battery manufacturing within Europe, while advancing battery R&D and recycling systems. Meanwhile, the UK is focusing more on technological development and design cooperation. As for the semiconductors sector, the EU is aiming to engage across the entire value chain to address manufacturing vulnerabilities. Considering how the sector is so far dominated by Asian countries, establishing a strong position in batteries manufacturing remains challenging. However, the UK is maintaining a niche market strategy focusing on R&D, design, and IP. In the digital industries sector, such as AI and quantum technology, both the EU and the UK are pioneering regulatory and normative frameworks to challenge US-China technological leadership. Their policies target technological sovereignty and competitiveness through R&D and infrastructure development, with accompanying efforts to implement policies addressing carbon neutrality in the energy-intensive infrastructures. In the health and biotech sectors, the EU policy centers on supply chain stabilization, enhancing internal manufacturing capacities, regulatory streamlining, expanding clinical trial infrastructure, and strengthening the security of critical pharmaceutical supply chains. Clean energy technologies are being promoted through the Net-Zero Industry Act and the Clean Industrial Deal, aimed at strengthening local manufacturing capabilities for eight strategic technologies - including hydrogen, batteries, carbon capture and storage (CCS) - while reducing reliance on non-European countries. Large-scale joint investments and infrastructure development under the IPCEI initiative are actively in process in the hydrogen segment. Meanwhile, the EU maintains a strong position in the aerospace industry sector, mainly thanks to Airbus, while concurrently advancing research on sustainable aviation fuels (SAF) and improving aircraft efficiency. In space, major projects like Galileo and Copernicus are underway, and the EU Space Act has been proposed to secure global standard-setting leadership.

The key implications for Korea’s strategic engagement with the EU and the UK’s advanced industry policies can be summarized in three main points. First, in terms of investment, Korea should strengthen R&D and equipment investment in key sectors such as batteries, semiconductors, AI and quantum technology, while actively leveraging its associate membership in Horizon Europe to enhance networks and market access opportunities. Second, to improve the investment environment, Korea needs to introduce converged regulatory sandboxes and streamline approval procedures, as demonstrated by the EU’s One-Stop Shop model, to foster a competitive industrial ecosystem. Third, for broader cooperation, Korea may consider solidifying its global position in positive technology exposure and standardization processes by participating in public procurement projects and the Korea-EU digital, green, and security partnerships. It should also mitigate supply chain risks by easing upstream dependencies and strategically practicing mid- and downstream network collaboration. Furthermore, it is crucial to proactively respond to ESG regulations (such as the CSDDD) and solidify the foundation for long-term cooperation with Europe through talent exchange and joint research.

Carbon neutrality, as a long-term goal of the international community, can only be achieved through innovation in climate technologies. The scale of climate finance and clean energy investments has continued to grow, with new markets recently emerging through the convergence of climate technologies with other advanced technologies such as artificial intelligence (AI). In this context, this study focuses on the growth and global expansion of startups, which are key economic players capable of driving innovation in climate technologies. In the study, a climate tech startup is defined as an unlisted company established within the past 10 years that has innovative ideas or business models related to technologies contributing to greenhouse gas reduction or climate change adaptation. Based on this definition, the study examines not only the roles of government support in leading countries but also success stories of promising climate tech startups. It also analyzes the ecosystem for climate tech startups and the effectiveness of R&D support programs in Korea, ultimately deriving policy implications for Korea.

Chapter 2 analyzes the key strategies of major countries and promising climate tech startups. Across all the countries reviewed, governments are increasing financial support, encouraging private investment, and supporting networking hubs to connect climate tech startups with each other and other partner companies. In particular, Germany operates large-scale and long-term funds such as the DeepTech & Climate Fund (DTCF), targeting startups at the growth stage rather than those in the early phase. Japan offers incentives, including tax deductions, to encourage collaboration between startups and large corporations as well as investment from large corporations. On the other hand, the UK and the US, where the climate tech markets are largely driven by the private sector, operate specialized institutions for high-risk, high-reward technologies such as ARIA and ARPA-E. The UK promotes the commercialization of climate technologies through regional clusters and innovation networks, including Catapults and Living Labs. The US supports climate tech startups through initiatives like the Energy Program for Innovation Clusters (EPIC), which funds organizations within regional innovation ecosystems. Finland, where both the public and private sector are actively engaged, encourages startups to expand overseas from the early stages of their establishment.

Promising climate tech startups around the world are growing and expanding their businesses into global markets by leveraging various government support programs, attracting private investment, engaging in technology collaborations and partnerships, and capitalizing on the capabilities of their founders. For example, companies such as TBM and Sila Nanotechnologies have successfully utilized a range of government programs from the R&D stage to commercialization for their growth and international expansion. Climeworks and Ascend Elements have secured some of the largest investments in the industry by demonstrating operational success with facilities applying their own technologies and by leveraging government matching funds. Sunfire, Clean Planet, and Coolbrook have achieved successful demonstration and commercialization through strategic partnerships with corporations, universities, and research institutions. Meanwhile, Carbon Clean and 44.01 have developed distinctive business models driven by the leadership and technical expertise of their founders.

Chapter 3 explores the ecosystem and enabling environment for climate tech startups in Korea, along with an analysis of the effectiveness of government R&D programs. Climate tech startups in Korea account for only approximately 5% in terms of both number and total investment scale as of the 2015-2024 period. Over 70% of the total investment was concentrated in the early stages of investment (Series A and below), and the pace of investment attraction remains relatively slow. Notably, government support has led the growth of startups, while the investment shares of venture capital (VC) firms and corporate venture capital (CVC) entities remain comparatively low. Policies and institutional support for climate tech startups primarily focus on financial assistance and the development of a startup ecosystem. Through financial mechanisms such as the Climate Technology Fund, the government aims to stimulate private investment while strengthening the roles of technology demonstration platforms and regional innovation clusters including Green Convergence Clusters. In addition, programs supporting global expansion, such as an initiative to link the Creative Technology Solution (CTS) and Tech Incubator Program for Startup (TIPS), are recently being implemented.

In order to evaluate the effectiveness of the Korean government’s support for climate technology development, the study analyzed the five-year outcomes of R&D support programs implemented between 2016 and 2018. The analysis focused on major climate technologies, including renewable energy, energy efficiency, and hydrogen and ammonia utilization technologies, selected based on their potential for emission reduction and trends in national R&D investment. An AI-based deep learning classification model was used to identify participants in major climate technology R&D support programs. Propensity Score Matching (PSM) and Difference-in-Differences (DID) methods were then used to estimate the impact. The results indicate that the R&D support programs had a positive effect on the financial performance of both startups and small and medium-sized enterprises (SMEs), with a notably stronger and more sustained impact observed in startups compared to SMEs. However, innovation outcomes showed a temporary increase only among SMEs, whereas social outcomes such as job creation were not statistically significant.

Based on these findings, the study suggests policy directions necessary to effectively support climate tech startups in Korea. These include general strategic directions and stage-specific support strategies tailored to the R&D, demonstration, and growth/scale-up phases. Detailed implications for each are discussed below.

Above all, it is crucial to enhance the understanding of climate technologies while ensuring the continuity and consistency of policy frameworks. Achieving this objective requires adopting a balanced and comprehensive perspective on climate technologies. Moreover, it is important to recognize that the perceptions of firms and investors are likely to shift only when there is a credible assurance regarding the long-term sustainability of government policies. The positive impacts of major climate technology R&D programs identified in this study further underscore the importance of sustained support for climate tech startups.

By stage of technology development, the first priority in the R&D phase is to assess whether technology support adequately addresses the integration of different technologies and market demand, and to strengthen the role of universities. Evaluating whether existing policies sufficiently foster technological innovation through convergent approaches is crucial. From a market demand perspective, a detailed analysis should be conducted to identify which technologies are likely to attract significant investment incentives at specific points in time, and this information should be effectively utilized. In addition, measures to enhance incentives for university-based startup activities and collaborative research with external partners should be considered.

Second, in the demonstration phase, it is important to encourage early-stage startups to pursue global expansion and to support regional networks. Given that the Korean economy has high export dependency and a small climate technology market, overcoming these limitations requires strategies that promote consideration of overseas market entry from the initial stage of idea development, similar to the Finnish model. Furthermore, strengthening the role of regional living labs and facilitating the adoption of technologies validated within the region at relatively low cost should also be prioritized.

Finally, in the growth and scale-up phase, it is necessary to expand investment incentives for climate technologies and to support the strengthening of partnerships between firms. It is recommended that tax incentives for climate technology be introduced or enhanced to encourage participation from a diverse range of private investors. In addition, policy mechanisms should be developed to attract investment from well-capitalized firms and corporate venture capital (CVC) entities. For startups seeking global expansion, efforts to facilitate strategic partnerships with local companies in target markets should be further strengthened.