|
on Innovation |
| By: | Jingni Zhang; David Popp |
| Abstract: | Electric vehicles (EVs) are crucial for cutting transportation emissions, yet the policy drivers of EV innovation remain underexplored. This study analyzes firm-level panel data on EV and battery patents, covering more than 4, 000 firms across 19 countries from 2010 to 2021, to assess how these policy tools and their interactions in different time horizons influence innovative activity. We test the effects of individual policy instruments that either raise demand for EVs or support the development of EV technologies. Stringent fuel-economy standards, financial incentives, adoption targets, and public R&D investments each significantly increase patenting in EV and battery technologies. Moreover, long-term EV targets amplify the innovative impact of public R&D and standards while diminishing the marginal effect of short-term price signals. The results suggest that governments can accelerate clean automotive innovation by combining long-term adoption commitments with sustained R&D investment or strong performance standards, and by managing these instruments as a coordinated policy portfolio rather than as separate tools. The study contributes cross-country, firm-level evidence that links policy design to the direction of clean technology innovation. |
| Keywords: | electric vehicle, technological innovation, policy horizons |
| JEL: | O31 Q55 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12421 |
| By: | Andrea Guccione; Pau Roldan-Blanco |
| Abstract: | Firms' innovation outcomes depend on their ability to attract and retain talented inventors. What market frictions prevent the sorting between firms with high innovation potential and high-productivity inventors? How does this sorting impact aggregate innovation, growth and welfare? We address these questions both empirically and theoretically. Empirically, we show that firms facing strong competition in the product market employ more productive inventors, while less productive inventors tend to be allocated in concentrated industries. Theoretically, we embed a frictional labor market for inventors into an endogenous-growth model of strategic innovation. In line with the data, the model predicts that high-productivity inventors are disproportionately employed in firms that operate in competitive industries. We then use the model to quantify the growth and welfare implications of this inventor sorting. Our results show that matching frictions in the market for inventors impede the allocation of high- productivity inventors to firms with high implementation intensity, and are responsible for a 32% loss in economic growth. Industrial policies that subsidize R&D spending relax these frictions by boosting inventor productivity, helping high-quality inventors reallocate to firms with high implementation incentives. Under optimal subsidies, growth increases as much as 74 basis points, closing most of the gap in missing growth caused by frictions in the market for inventors. |
| Keywords: | innovation, inventors, R&D productivity, search |
| JEL: | L16 J6 O3 O4 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:bge:wpaper:1562 |
| By: | Ville Korpela (Turku School of Economics, University of Turku, Finland); Eero Mäkynen (Turku School of Economics, University of Turku, Finland) |
| Abstract: | Business dynamism has been slowing globally over the last several decades. In a recent study, Akcigit and Ates (2023) examine the relative importance of different channels behind this development and highlight weakened knowledge diffusion from the technology frontier to followers as a dominant force.1 Their study also suggests that diffusion may weaken endogenously as the technology gap widens and market power accumulates, raising the question of how innovation policy can strengthen diffusion without reducing welfare. In this paper we study leader-to-follower licensing as a policy-relevant diffusion margin, and evaluate licensing subsidies relative to direct R&D subsidies. We develop an endogenous-growth general equilibrium model in which firms compete in prices and invest in R&D; the technology leader endogenously chooses whether to license to the follower, trading off higher static profits against faster follower catch-up through knowledge diffusion. We calibrate the model to Finnish data from 2014–2019. Our first exercise evaluates whether allowing licensing is desirable by shutting down the licensing channel in the calibrated economy. In the Finnish benchmark, shutting down licensing lowers growth but increases consumption-equivalent welfare, because the level effects of reduced concentration dominate the diffusion benefits of licensing. We then vary the diffusion rate through licensing and product substitutability to characterize when licensing becomes welfare-improving. In that region, solving the policymaker’s problem shows a non-trivial interaction: higher R&D subsidies can reduce equilibrium licensing by moving leaders more quickly into the monopoly-pricing states where licensing is privately unattractive, so the optimal policy mix augments R&D support with a non-negligible licensing subsidy to sustain diffusion. |
| Keywords: | Antitrust Policy, Business Dynamism, Endogenous Growth, Innovation Policy, Licensing, Technology Diffusion |
| JEL: | E22 L10 L41 O33 O34 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:tkk:dpaper:dp174 |
| By: | Srikanth Pai ((Corresponding author), Madras School of Economics, Gandhi Mandapam Road, Behind Government Data Centre, Kotturpuram, Chennai, 600025, India.); Akila Hariharan (Madras School of Economics, Gandhi Mandapam Road, Behind Government Data Centre, Kotturpuram, Chennai, 600025); Naveen Srinivasan (Madras School of Economics, Gandhi Mandapam Road, Behind Government Data Centre, Kotturpuram, Chennai, 600025) |
| Abstract: | We build a tractable model that links institutional dynamics with the private value of innovation. Our approach differs from much of the existing literature in that an inventor does not retain a perpetual monopoly over its use, and the cash flows generated from a new idea are uncertain. In our framework the relevant dimension of institutional quality is enforcement strength. We model institutional strength as a two-state continuous-time Markov chain. This makes the cash flows from innovation stochastic and state-dependent, and hence the incentive to innovate varies with the strength of enforcement regime. Countries alternate between periods of strong and weak enforce-ment, reflecting irregular political and legal events such as reforms, leadership changes, or crises. Our model shows how institutional fragility can alter the incentive to innovate and connects institutional dynamics with cross-country differences in standard of living. |
| Keywords: | institutions, innovation, patents, continuous-time Markov chain, economic growth |
| JEL: | O31 O33 O34 O43 |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:mad:wpaper:2026-292 |
| By: | Eugenie Dugoua; Joelle Noailly |
| Abstract: | This paper examines the patterns and mechanisms of global clean technology diffusion over the last two decades. We document four stylized facts: uneven sectoral progress favoring power and light transport; China’s dominance in innovation and manufacturing; the role of modularity in driving cost declines; and limited adoption in developing economies. Through case studies of solar, electric vehicles, and hydrogen, we analyze how policy and infrastructure enable scale. Finally, we assess emerging challenges for the next phase of diffusion, including critical mineral constraints, artificial intelligence, and geopolitical fragmentation. |
| Keywords: | Clean technology diffusion, Climate change mitigation, Renewable energy, Industrial policy, Solar photovoltaics, Electric vehicles, Hydrogen |
| JEL: | O33 Q55 O25 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:wip:wpaper:95 |
| By: | Francesco Androni (Gran Sasso Science Institute); Andrea Ascani (Gran Sasso Science Institute); Alberto Marzucchi (Gran Sasso Science Institute) |
| Abstract: | The phenomenon of zombie firms has been increasing through time in the last decades. Prior re search has extensively examined the role of zombie firms in credit misallocation and weak insolvency regimes However, limited attention is paid to how the competitive environment has influenced its surge. The study aims at linking the diffusion of zombie formation with the field of industrial dynam ics. The analysis focuses on whether the intensity of competition influences the diffusion of zombie f irms, by assessing competition forces such as firm entry and innovation intensity. We use micro aggregated data at the region-sector level to analyse the diffusion of zombie firms in Italy for the years from 2014 to 2020, and identify a substantive role of reallocation forces in driving the shares of zombie firms. Competition in the form of entry and, albeit more weakly, innovation intensity reduces the diffusion of zombie firms, ultimately showing that a decrease in competition intensity is part of the phenomenon. This research contributes to understanding the relationship between zombie firms and sluggish economic activity, describing further factors that affect their formation and persistence. |
| Keywords: | Zombie firms; competition; firm entry; innovation; industrial dynamics |
| JEL: | O33 O31 L25 R11 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:ahy:wpaper:wp65 |
| By: | Ernest Miguelez; Michele Pezzoni; Fabiana Visentin; Catalina Martínez; Reinhilde Veugelers; Julio Raffo |
| Abstract: | This paper examines the evolving geography of international technological knowledge diffusion over the last four decades using multiple patent-based indicators. We first review the main mechanisms through which knowledge diffuses across borders—including trade and global value chains, foreign direct investment, skilled migration, global science, and markets for technology—highlighting their complementarities and the role of domestic capabilities. We then provide new empirical evidence based on cross-border patent citations, technological trajectories defined by IPC recombinations, patent-to-science linkages, and international patent families. The results reveal persistent asymmetries, with a small group of advanced economies remaining central knowledge hubs, alongside the rising role of emerging countries, especially China. Science-based technologies diffuse farther and faster, while capability constraints continue to limit integration for many regions. |
| Keywords: | Technological knowledge diffusion, Geography, Patents, Citations, Technological trajectories, Science, Patent families |
| JEL: | O34 O33 F14 F23 R12 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:wip:wpaper:92 |
| By: | Pinelopi Koujianou Goldberg; Reka Juhasz; Nathan Lane; Giulia Lo Forte; Jeff Thurk; Pinelopi Goldberg |
| Abstract: | The resurgence of subsidies and industrial policies has raised concerns about their potential inefficiency and alignment with multilateral principles. Critics warn that such policies may divert resources to less efficient firms and provoke retaliatory measures from other countries, leading to a wasteful “subsidy race.” However, subsidies for sectors with inherent cross-border externalities can have positive global effects. This paper examines these issues within the semiconductor industry: a key driver of economic growth and innovation with potentially significant learning-by-doing and strategic importance due to its dual-use applications. Our study aims to: (1) document and quantify recent industrial policies in the global semiconductor sector, (2) explore the rationale behind these policies, and (3) evaluate their economic impacts, particularly their cross-border effects, and compatibility with multilateral principles. We employ historical analysis, natural language processing, and a model-based approach to measure government support and its impacts. Our findings indicate that government support has been vital for the industry’s growth, with subsidies being the primary form of support. They also highlight the importance of cross-border technology transfers through FDI, business and research collaborations, and technology licensing. China, despite significant subsidies, does not stand out as an outlier compared to other countries, given its market size. Model estimates suggest the presence of learning-by-doing at the firm-product level as well as economies of scope within a firm and substantial cross-border learning spillovers. These spillovers likely reflect cross-country technology transfers and the role of fabless clients and input suppliers in disseminating knowledge globally through their interactions with foundries. Such cross-border spillovers are not merely accidental but result from deliberate actions by market participants that cannot be taken for granted. Firms may choose to share knowledge across borders or restrict access to frontier technology, thereby excluding certain countries. Future research will use model estimates to simulate the quantitative implications of subsidies and to explore the dynamics of a “subsidy race” in the semiconductor industry. |
| Keywords: | semiconductors, industrial policy, subsidies, learning-by-doing, multilaterism |
| JEL: | F13 F61 L63 N60 O38 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12495 |
| By: | Charles de Grazia; Nicholas E. Rada; Gregory Graff |
| Abstract: | Technology diffusion is central to the process of innovation, as new products or processes must be adopted for them to make meaningful contributions to societal welfare or economic growth. We focus here on the global diffusion of technology that has the potential to improve food insecurity and address challenges posed by climatic effects, genetically modified (GM) crops. We adopt a variety of sources and methods to demonstrate the reach and timing of genetically modified crop technology diffusion worldwide, relying primarily on national regulatory approval information. Specifically, we depict the international adoption of genetically modified crop technology over time and assess the rate at which GM cotton, maize, and soybeans have been adopted within countries. In addition, we examine two case studies that assess an underused information source—trademark data—to determine whether they provide an alternative measure of diffusion. The case studies focus on two different contexts: established branded technologies and nascent technologies. In addition to significant overlap with regulatory approval data for established branded technologies, trademarks appear to provide an indicator of pre-commercialization in countries where regulatory approval coverage can expand. We end with guidance on when trademarks may serve as an indicator of international technology diffusion. |
| Keywords: | Technology diffusion, Agricultural innovation, Genetically modified crops, Trademarks, Diffusion measurement |
| JEL: | O3 Q16 Q18 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:wip:wpaper:93 |
| By: | David Popp; Myriam Gregoire-Zawilski; Lizhen Liang; Daniel Acuna |
| Abstract: | Does government funding influence the choice of research topics? Novel grant-making modalities such as the Advanced Research Projects Agency-Energy (ARPA-E) program aim to encourage scientists to take on difficult-to-solve, wicked societal problems such as clean energy. Yet little causal evidence exists linking funding and research direction, with most existing studies focusing on health sciences. We provide new evidence on the effect of funding on clean energy research, addressing two questions: (1) Do scientists change the focus of their research in response to targeted government funding opportunities? (2) If so, what types of calls for funding best attract new researchers? Using data on grants from the Department of Energy and National Science Foundation, we combine text and regression analyses to compare the publication trajectories of funded scientists to a set of matched controls. After funding, the research of funded scientists becomes more similar to the grant topic than that of the matched controls. The effect is largest for ARPA-E, which explicitly aims to attract new scientists to clean energy research, suggesting that agency efforts to attract new researchers to a topic area can succeed. General calls for funding such as offered by traditional NSF directorates generate less movement. |
| JEL: | O38 Q48 Q55 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34856 |
| By: | Dugoua, Eugenie; Noailly, Joëlle |
| Abstract: | Reaching net zero emissions by 2050, a goal now endorsed by many countries, requires the rapid and massive deployment of clean energy systems. However, this transition hinges on a new form of dependence. The technologies required to decarbonise depend on a small set of critical minerals like lithium, cobalt and rare earth elements that have highly concentrated supply chains marked by heavy environmental footprints and challenges regarding labour and community rights. This paper presents a simple framework to explore how innovation can strengthen critical mineral supply chains. It examines technological, digital and organisational solutions across the full lifecycle — from mining exploration and processing, to manufacturing, reuse and recycling. The authors examine what drives or hinders innovation, including price volatility, industrial policy, environmental regulation, firm strategies (such as vertical integration), market size and rising demand from competing sectors like AI and defence. They highlight that a portfolio of innovations is essential for reconciling the growing demand for critical minerals with climate, economic and geopolitical priorities. Innovation can reduce the costs of reform, allowing governments to pursue security and sustainability simultaneously. The challenge is not simply to secure more materials but to build a system that is resilient, circular and adaptive. |
| Keywords: | critical minerals; innovation; clean energy transition; supply chain; mining; rare earth elements; batteries; industrial policy; circular economy |
| JEL: | O31 Q55 L72 |
| Date: | 2024–12–04 |
| URL: | https://d.repec.org/n?u=RePEc:ehl:lserod:137103 |
| By: | Carsten Fink; Maria de las Mercedes Menéndez; Julio Raffo |
| Abstract: | Technology diffusion is central to economic development. This paper examines diffusion patterns for 31 technologies for 139 countries over two centuries, extending existing databases to include recent digital technologies and renewable energy technologies. Using cross-country panel regressions, we find that while adoption lags have declined from 50 years (pre-1950) to 15 years (post-2000), adoption intensity in developing economies remains at 53% of advanced economy levels. We document diverging intensity for older technologies but emerging convergence for post-2000 technologies, suggesting digital innovations may reduce the technology gap. These findings inform policies aimed at accelerating technology diffusion to developing economies. |
| Keywords: | Technology diffusion, Digital technologies, Adoption lag, Intensity of use |
| JEL: | O33 O47 O57 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:wip:wpaper:91 |
| By: | Daron Acemoglu; David Autor; Simon Johnson |
| Abstract: | This paper defines pro-worker technologies, including Artificial Intelligence, as technologies that make human skills and expertise more valuable by expanding worker capabilities. Our conceptual framework distinguishes among five categories of technological change: labor-augmenting, capital-augmenting, automating, expertise-leveling, and new task-creating. Only the last category is unambiguously pro-worker, generating demand for novel human expertise rather than commodifying it. We illustrate these distinctions through hypothetical and real-world examples spanning aviation maintenance, electrical services, custodial work, education, patent examination, and gig delivery. While AI’s capacity to automate work is substantial, we argue that its potential to serve as a collaborator, by extending human judgment, enabling new tasks, and accelerating skill acquisition, is equally transformative and currently underexploited. We identify market failures, including misaligned firm and developer incentives, path dependence, and a pervasive pro-automation ideology, that may lead to underinvestment in pro-worker AI. We consider nine policy directions that would change incentives, including targeted investments in health care and education, tax code reform, antitrust enforcement, and intellectual property protections for worker expertise. |
| JEL: | J23 J24 J31 M50 O33 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34854 |
| By: | Egbert Amoncio; Alexander Cuntz; Carsten Fink |
| Abstract: | The paper examines how legal certainty shapes protection and appropriation of digital designs such as icons, animations, and layouts. Leveraging the 2012 Apple v. Samsung verdict as a decisive clarification of their protectability and enforceability, we analyze USPTO design patents from 2009–2015 using a matched difference-in-differences approach. We show that legal certainty reduces due diligence costs far more than monitoring costs. This asymmetry lowers the threshold for securing protection, leading to a 9 percent increase in digital design patents. At the same time, appropriation shifted away from licensing toward transfers, with the effect strongest in dense design spaces where monitoring costs remain high despite increased legal certainty. These findings extend transaction cost theory by showing that legal certainty unevenly reduces transaction costs, which in turn alters protection thresholds and shifts appropriation strategies. They also demonstrate how policy changes influence innovation when value is created “under the glass.†|
| Keywords: | Digital designs, Design patents, Appropriation, Transaction costs |
| JEL: | O31 O34 K11 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:wip:wpaper:97 |
| By: | Hemanth Asirvatham; Elliott Mokski; Andrei Shleifer |
| Abstract: | We present the GABRIEL software package, which uses GPT to quantify attributes in qualitative data (e.g. how “pro innovation” a speech is). GPT is evaluated on classification and attribute rating performance against 1000+ human annotated tasks across a range of topics and data. We find that GPT as a measurement tool is accurate across domains and generally indistinguishable from human evaluators. Our evidence indicates that labeling results do not depend on the exact prompting strategy used, and that GPT is not relying on training data contamination or inferring attributes from other attributes. We showcase the possibilities of GABRIEL by quantifying novel and granular trends in Congressional remarks, social media toxicity, and county-level school curricula. We then apply GABRIEL to study the history of tech adoption, using it to assemble a novel dataset of 37, 000 technologies. Our analysis documents a tenfold decline of time lags from invention to adoption over the industrial age, from ~50 years to ~5 years today. We quantify the increasing dominance of companies and the U.S. in innovation, alongside characteristics that explain whether a technology will be adopted slowly or speedily. |
| JEL: | C8 O3 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34834 |