Abstract Scholars working in the evolutionary economics of innovation are familiar with the concept of the technological paradigm and, indirectly, with that of the scientific paradigm. The former derives from the latter, and was developed in recognition of the relative autonomy of technology from science. In this work, I reconsider the relationship between science and technology, arguing that, in a few important cases, technology can no longer be meaningfully distinguished from science. In such cases, science and technology converge to form a sci-tech agglomerate, hence the notion of a sci-tech paradigm. The fundamental role played by economic incentives in promoting this convergence is taken into account. Acceptance of this notion implies the need to revise policy frameworks that rely on indicators calibrated for conditions that no longer exist.
The paper develops a theoretical framework with some skill heterogeneity to investigate the way the direction of process innovation, labor-saving versus time-saving, shapes the evolution of the industrial structure, firm dynamics, and macroeconomic performance in a digitally transforming economy. We find that labor-saving innovations deliver higher capital deepening but engender structural fragility. Conversely, when firms focus on time-saving innovations, the economy grows more slowly but exhibits greater resilience: lower downtime supports higher firm survival, reduces volatility, and sustains a more competitive and inclusive market structure. These dynamics are emphasized when labor shortage constrains production capacity. We also discuss the design of more targeted and effective innovation policies, capable of steering digital transformation in directions that are economically robust and socially inclusive.
Drawing on institutional theory and resource dependence theory, this paper examines how China's National Big Data Comprehensive Pilot Zone (CBDPZ) policy influences corporate carbon performance and explores key moderating contingencies. Using a staggered difference-in-differences (DID) approach on panel data from Chinese A-share listed firms (2011-2023), we find that the CBDPZ policy significantly improves corporate carbon performance. Mechanism analysis reveals that the policy promotes carbon reduction through three pathways: digital transformation, green innovation, and human capital upgrading. Furthermore, we find these effects are amplified in firms led by environmentally aware executives and those receiving government environmental subsidies. Heterogeneity analysis indicates more pronounced effects among non-high-tech firms, state-owned firms, eastern-region firms, and those in high-regulation, high public-concern, or heavily polluting sectors. This study contributes to the literature by illuminating how non-environmental digital policies generate environmental co-benefits and offering empirical evidence for evaluating the sustainability implications of digital economy policies in emerging economies.
Longitudinal studies examining dynamic effects of research, development, and innovation (RDI) subsidies on firm-level outcomes-particularly their role in promoting significant innovations-remain limited. This has left policymakers at a disadvantage in coordinating complementary innovation policy instruments. Using a literature-based innovation output measure, we address these shortcomings by means of a nationally representative panel study of Finnish enterprises' significant innovation outcomes observed over 22 years. Relative to confounding factors, we find that the average dynamic effects of RDI grant subsidies on companies' probability to commercialize a significant innovation are modest but mostly positive, persisting, and dynamically increasing, peaking around five to eight years after first receiving a grant. Robustness tests, including results based on counts of granted patents, generally support the main findings of the study. Based on our empirical insights, we provide an economic interpretation of our results, suggesting that a consistent long-term innovation subsidy policy plays a meaningful role in advancing innovation output.
This study represents one of the first attempts to develop a global typology of regional innovation systems (RIS) across both advanced and catching-up economies, identifying distinct pathways for regional technological catch-up. Utilizing United States Patent and Trademark Office (USPTO) patent citation data from 33 regions worldwide over the period 2000-2017, we construct several RIS indicators, including intra-regional, inter-regional, and international knowledge sourcing, alongside domestic ownership of innovation. Through cluster analysis, we identify four primary RIS configurations and examine their relationship with economic performance: Group 1: large and mature RIS, characterized by long-cycle technology specialization and high domestic ownership; Group 2: mixed RIS, featuring long-cycle specialization but low domestic ownership; Group 3: strong catch-up RIS, defined by short-cycle technology specialization and high domestic ownership; Group 4: weak catch-up RIS, involving short-cycle specialization with low domestic ownership. While Groups 3 and 4 consist exclusively of emerging economies specializing in similar short-cycle technology sectors, they exhibit divergent levels of economic performance. Ultimately, the analysis identifies two empirically distinct catch-up trajectories: a "stronger" pathway driven by increasing domestic ownership and the localization of knowledge creation and a "weaker" pathway characterized by persistent dependence on foreign knowledge sources. Our findings suggest that domestic ownership and the emergence of large-scale enterprises are critical drivers of localized knowledge sourcing and sustained catch-up.
To position their ventures, entrepreneurs often need to map their competitive landscape. We develop a theory of cognitive cartography to explain how geography structures entrepreneurs' competitive perceptions through the fit between a venture and a location's industrial identity, and through a competitor's typicality within its location. Using a behavioral simulation, we find that while typical firms tend to attract attention at lower levels of congruence, this preference reverses in high-congruence locations. Within these locations, a "pop-out" effect emerges: locally atypical firms capture significantly more attention than their typical counterparts. This pattern suggests that entrepreneurial search balances the need for legitimacy with the detection of differentiated competitors, with implications for how ventures develop. Our findings reveal that geography functions as a distinct categorization system alongside industry classifications, with its own attentional dynamics that extend organizational theories of categories to geographic contexts and offer a geographically informed view of pre-entry competitor identification.
Digital platforms provide arenas in which autonomous participants produce and consume content, and the design of their governance constitutes a critical source of firm performance. This study focuses on output control strategies in user-generated content platforms and introduces a novel theoretical framework that goes beyond the traditional quality-based approach by incorporating category-based mechanisms. Specifically, we classify output control strategies along two domains-quality and category-and introduce a fourfold typology consisting of open, quality-based, category-based, and dual control strategies. Furthermore, we examine two boundary conditions-user composition and revenue models-and conduct an agent-based simulation to compare the relative effectiveness of each strategy. This study makes two theoretical contributions. First, by incorporating participants' within-category preferences, we establish the theoretical validity of category-based control and refine the conventional open/closed governance dichotomy. Second, we demonstrate that dual control strategies can simultaneously pursue both participant numbers and utility, offering a new perspective of compatibility that challenges the traditional view of governance strategies as a trade-off between the two. Through these contributions, this research extends the theoretical scope of platform governance studies.
This article examines the evolution of the Korean nuclear industry from the dynamic sectoral innovation system perspective, which integrates the sectoral innovation system (SIS) perspective and industrial dynamics. We analyze the Korean nuclear sector and conclude that it has developed dynamically in three stages: Imitation, Specialization, and Stabilization. During the Imitation stage, key public innovation actors emerged to absorb imported technologies, although interactive learning among them remained limited. In the Specialization stage, enough actors emerged to facilitate close collaboration and begin accumulating technological capabilities. In the Stabilization stage, Korea implemented a competent nuclear SIS, producing sufficient interactive learning among actors and attaining world-level technological capabilities.
This paper uses machine learning techniques to classify patent claims as product or process innovations and documents that the overall process share of innovation in the United States was on a secular decline from 1980 to 2015. The reallocation of patenting activity from the chemical to the computer & communication category explains much of the declining process share from 1980 to 2000. This paper rules out the rise of China as an explanation for the declining process share from 2000 to 2015. Instead, this paper finds evidence that a large contributor to the decline in the process share from 2000 to 2015 is patents in the semiconductor and electrical circuit topics becoming less process focused. This paper finds that the process share is low at the beginning of a firm's product life cycle, peaks in the middle, before plateauing at an intermediate level at the end of the life cycle. Firm size is positively associated with the process share of innovation. These relationships between life cycle, size, and the process share of innovation are not present for alternative classifications of product and process patents.
Circular digital platforms can contribute to addressing the challenges of natural resource overexploitation and material waste accumulation. Circular digital platforms incorporate diverse, complementary, and even competitive actors; therefore, circular platform orchestration is a crucial phenomenon yet unstudied. This study examines the conditions of circular platform orchestration in diverse actor settings leading to collective industry benefits, and changes in competitive dynamics among industry actors. We utilize a multiple-case study on circular digital B2B platforms orchestrated by public and private actors in Finland, Italy, and Lithuania. The findings identify three orchestration conditions of B2B circular platforms: incentives and motivation to participate in the platform, control and operational rules, and homogeneity/heterogeneity of actors. The findings also show that these conditions manifest differently under private versus public orchestrators, with the former more motivated by economic goals and the latter more by public good goals; yet, with industry outcomes being relatively similar. Our study provides implications to the literature of circular economy business and platform orchestration by demonstrating how circular platforms help to 'raise all boats' in the industry while reconfiguring some competitive dynamics between primary and secondary markets.
Dominant innovation policy paradigms are undergoing significant shifts as emerging sustainability objectives challenge traditional techno-economic approaches, necessitating reflection on the foundations of existing policies. However, translating such high-level ambitions into practice is complicated with tensions arising from different interpretations of policy goals and required interventions. We explore a paradigm shift in Finland, where sustainability goals have been introduced as major priorities in the national innovation policy, though the new paradigm is still in transition and not yet fully integrated across innovation contexts. We conduct a case study analyzing three distinct problem-solution constellations: the battery value chain, food packaging, and AI in public services. We analyze the underlying views of policy agendas and understandings of the innovation process and identify tensions arising from the conflict between the traditional innovation policy paradigm and the emerging sustainability-oriented one in each case. We develop reflexive questions that can guide innovation actors in resolving these tensions, enabling second-order policy learning. Our study contributes to the literature on the normative turn in innovation policy by showing how tensions during paradigm shifts, and the reflection they require, vary by context, highlighting the need for deliberate mechanisms to surface and address fundamental differences in innovation policy framings. This contextual variety also creates opportunities for cross-domain policy learning.
This study examines regional disparities in both the allocation and impact of a French financial innovation support program for SMEs, assessing its effects on firm-level Total Factor Productivity, labor productivity, and intangibles-to-assets ratio as indicators of firm efficiency and innovativeness. Using a quasi-experimental design, our analysis reveals that firms in the Paris region experience significantly greater benefits from innovation support compared to those in other regions. This disparity in policy effectiveness is strongly influenced by localization and urbanization economies, as well as substantial knowledge spillovers, which are more prevalent in highly agglomerated regions like Paris. These factors amplify the effectiveness of public support, suggesting that densely networked innovation ecosystems enhance firms' ability to leverage public funding for productivity and innovation gains. Our findings underscore the importance of considering regional agglomeration effects in the design of innovation policies to address existing heterogeneity in policy impact.
A major challenge for innovation scholars is the measurement of capabilities. We develop and apply a novel methodology for quantifying the capability development of firms, where capabilities are measured in terms of expenditures on specific activities. These activities can be sorted in a hierarchy of increasing difficulty, and firms can be sorted according to their positions on the capabilities ladder. Our nestedness algorithm, inspired by biology and network science, defines an activity as complex if it is performed by only a few firms at the upper rungs of the ladder. We analyze company annual reports of almost 45,000 Indian firms for the time period 2000-2020, and observe significant nestedness. Lower rungs of the capabilities ladder correspond to basic managerial and production activities. Mid-level rungs correspond to internationalization and acquiring absorptive capacity. Higher-level rungs are more related to Mergers & Acquisitions (M&A) and innovation. Information and Communication Technology (ICT) capabilities seem to have become more fundamental lower-level rungs on the capabilities ladder in recent years. There is a close relationship between size and capabilities for small firms, but this weakens for larger firms. Some heterogeneity is observed across industrial sectors and across regions. We find that capability ranking can explain future growth patterns and survival probability of firms, summing up in one number their future potential trajectories.
We apply quantile regression coefficient modeling (QRCM) to investigate the firm growth process. QRCM imposes a parametric structure to the conditional quantile function and allows to estimate all quantiles at once by minimizing an integrated loss. To handle the presence of repeated measures, we fit a two-level model in which both the level-1 and level-2 parts of the distribution depend on predictors according to a quantile regression (QR) structure. Compared with standard QR, in which different quantiles are estimated one at a time, QRCM improves statistical efficiency, mitigates quantile crossing, simplifies estimation of extremes, and allows to incorporate identifying assumptions. We investigate growth in a panel of UK manufacturing firms. Our analysis accounts for variance-size scaling and allows to disentangle the location effect of firm size on growth from the scale effect. We propose alternative parametrizations of the QR coefficients: a flexible model based on Legendre polynomials, and a variety of more structured models that rely on known quantile functions, such as the Gaussian, logistic, and asymmetric logistic distributions, that differ in their tail behavior. Our results indicate that fat-tailed models, such as the asymmetric logistic distribution, provide a better fit than the normal distribution. We are able to detect a positive location effect and to obtain efficient estimates of the extreme quantiles.
Policymakers have increasingly supported collaborative R&D, recognizing its role in enhancing the innovation performance and scientific reputation of organizations. This paper proposes a three-step procedure to examine collaborative patterns and participation dynamics over consecutive research programs by integrating Social Network Analysis and statistical methods. First, participants in R&D projects are ranked based on their centrality in the collaborative network. Second, transition probabilities between classes of centrality across consecutive programs are estimated. Third, the Markovian nature of collaborative patterns, which is controversial across methodological, theoretical, and empirical contributions, is tested. Using data from the first eight European Union Framework Programmes (EU FPs), we analyze the relationship between EU funding mechanisms and the micro-level behaviors of participants. Our findings provide an integrated understanding of co-evolving micro- and macro-level dynamics, reflected in the emergence of core-periphery structures, and emphasize the role of EU-funded projects in strengthening organizations' popularity. However, these dynamics also highlight the risk of "oligopolistic" behaviors that may limit the European Research Area. Exogenous mechanisms, such as R&D policies and funding, emerge as crucial mechanisms in shaping participation dynamics and organizational positions, emphasizing the need to promote openness and democratization in research funding.
This paper looks at how industrial strategy can use conditionalities to make sure that public-private partnerships are goal-oriented with conditionalities that serve public purpose. Conditionality can be a key tool to shape markets and foster inclusive and sustainable economic growth. We develop a taxonomy to understand the range of conditionalities that governments and policymakers can consider when structuring calls for proposals, funding agreements, partnership contracts, tax incentives, regulatory frameworks, and other policies aimed at shaping the economy for the common good. The paper first provides an analytical framework for exploring the role that conditionality can play in modern industrial strategies. It then highlights a range of global case studies (lessons learned, both positive and negative) to explore the different dimensions of conditionalities and what they can achieve in practice.
How to effectively stimulate enterprise digital innovation has long been a critical research question. This study posits that government industrial policy may serve as a key driver of such innovation. Focusing on enterprises recognized as "Single Champions" between 2014 and 2021, we employ a multi-period propensity score matching combined with difference-in-differences (PSM-DID) approach to empirically examine the impact of the Single Champions policy on corporate digital innovation. The findings reveal that being designated as a Single Champion exerts a significant positive incentive effect on firms' digital innovation. Further analysis identifies market status and network centrality as important moderators that strengthen this relationship. This study contributes to the literature by demonstrating that government industrial policy can effectively incentivize and propel enterprise digital innovation.
This paper investigates whether foreign direct investment (FDI) significantly facilitates the access of firms in developing countries to digital transformation. It also examines if the level of access varies across firms of different geographical regions and operating sectors. Empirical analysis on a sample of more than 8000 firms in Vietnam during 2019 indicates that FDI plays an important role in helping firms cope with challenges raised by Industry 4.0. In addition, it lends support to the hypothesis that this effect is not homogenous. The results are robust to alternative econometric specifications and sample sizes. These findings convey important implications for developing countries' FDI strategies and policies that aim to promote technological development for digital transformation. Thus, they are of special interest to researchers, policymakers, and industrial practitioners.
To more closely align theories of paradigm-shifting discoveries and their empirical quantification, we propose a novel measure that incorporates a discovery's impact, novelty, and tendency to break with the past into a single, coherent measure. Calibration using the National Inventor Hall of Fame data reveals that the three dimensions are strict complements, meaning, for example, that greater impact cannot substitute for moderate novelty. We illustrate how the measure works and validate it in several complementary ways using data on the United States Patent and Trademark Office patents from 1982 to 2015. High values of the measure are strongly predictive of discoveries that are simultaneously exceptional in terms of their impact, novelty, and disruptiveness.
We examine the effect of changes in a firm's horizontal scope on its position moves, disentangling the effect of scope economies, exploration, inertia, and crowding. We argue that scope economies emanate from the contemporaneous breadth of a firm's scope while exploration and inertia relate to the sequence of historical choices a firm makes about whether to change or maintain its scope. Our theory predicts that exploration routines (that promote position moves) and inertial tendencies (that deter them) evolve over time and that their effects are moderated by competitive crowding. Analysis of all US auto firms between 1895 and 1981 confirms these patterns, thereby reconciling existing findings from prior research and highlighting the advantages of an evolutionary approach through the lens of organizational sociology.
