Agent-based models (ABMs) rely on calibrated inputs to reproduce observed economic and social phenomena, making input validation an important component of the modeling process. This paper revisits the calibration of an ABM of university enrolment decisions in Italy by reassessing the income distributions used to represent skilled and unskilled individuals. Using data from the Survey on Household Income and Wealth (SHIW) for the period 2002–2016, I estimate four alternative specifications: log-normal, Singh–Maddala, Dagum, and generalized beta of the second kind (GB2). The GB2 consistently provides the best statistical fit according to likelihood-ratio tests and information criteria. The four distributions are subsequently used as alternative inputs in the ABM while all other behavioral rules remain unchanged. Despite differences in statistical fit, simulation outcomes are remarkably similar across specifications. Average enrolment rates, enrolment by parental educational background, and educational mobility indicators display only minor variation. Effect-size measures based on Cohen’s d indicate that these differences are negligible in practical terms. The results suggest that the model’s aggregate outcomes are robust to alternative empirically grounded specifications of the income distribution. More broadly, the paper illustrates how input validation can be used to evaluate whether improvements in statistical fit translate into meaningful differences in simulation outcomes.
In a global context marked by liberalization, geopolitical tensions, and ecological urgency, firms must continuously reconfigure internal operations and external engagements to ensure long-term viability. This paper proposes an institutionalist-enriched business model framework that integrates insights from Commonsian economics, evolutionary economics, and French regulation theory to analyze how organizations adapt to economic and ecological uncertainty. By extending conventional business model approaches to incorporate institutional embeddedness, learning dynamics, and the political dimensions of enterprise strategy, our framework captures both intra-territorial interactions among enterprises and the co-evolution of enterprise trajectories and sectoral transformations, understood through competition regimes. We apply this approach to a comparative study of 14 agricultural enterprises embedded in contrasting regional contexts in France. The empirical focus on dairy farms illustrates not only the co-evolution of sectoral regulation and business models but also the resulting coexistence of different business models (BMs). Liberalization and environmental awareness have weakened the dominant family-run, productivist farm model, strengthened two BMs that resisted productivism, and given rise to three archetypal adaptation strategies. The coexistence of these BMs is shaped by regionally specific resources, historical legacies, and institutional arrangements. From an environmental perspective, the results indicate that ecological resistance, associated with traditional production methods, deserves as much attention as ecological transition. Beyond agriculture, the framework helps explain how enterprises across sectors navigate uncertainty by mobilizing both internal capabilities and external alliances. The results call for policy frameworks that support the coexistence of diverse production models, recognize the political and institutional foundations of ecological transitions, and emphasize the need for greater regional political autonomy to tailor strategies to specific territories.
Sustainable energy trajectories evolve through processes of knowledge accumulation and recombination within the same trajectory and hybridization between several of them. This paper investigates the emergence of a green hydrogen trajectory by applying a structural main path analysis to both scientific publications and patents related to electrolyzed hydrogen production. Beyond identifying trajectories, we analyze distinct time phases to capture the evolving architecture of the knowledge network. Our findings reveal a clear structural evolution: from an initial phase of competing, parallel paths, through a consolidation phase marked by the dominance of proton exchange membrane (PEM) technologies, to a current phase of hybridization where this core trajectory branches out to integrate with other green technologies (solar, wind, CO2-to-methanol). This structural shift reveals empirical signals of an emerging constellation of interrelated sustainable technologies, challenging conventional views on green innovation pathways.
This article applies network analysis techniques to examine the evolution of the consumer price structure in Italy between 1996 and 2024. By representing the CPI components as nodes in a network and linking them, the study explores the underlying structure of inflation diffusion based on joint price movements. Using methodologies such as the threshold minimum dominant set algorithm (T-MDS) and dynamic stochastic block models (DSBM), we identify key price categories and clusters, analyze the evolution of clusters, and analyze inflation diffusion patterns. Our findings reveal that periods of high inflation correspond to denser networks with lower modularity, indicating widespread inflation. In contrast, low-inflation periods show greater separation between communities. Policy implications include using network connectivity as an early warning indicator for inflationary diffusion and selectively stabilizing key price nodes to mitigate inflationary contagion.
This paper develops an asset-based dynamic model of innovation that links governance structure to the co-evolution of complementary and innovation assets. By representing these assets as interdependent dynamic stocks, the framework provides a formal account of how ownership and appropriation choices may affect the rate and direction of capability accumulation. Governance influences innovation not only through incentive alignment but by altering two structural parameters—an inhibitory effect that stabilizes complementary assets and an absorptive effect that enhances innovation productivity. The model highlights four main implications: high appropriation may induce technological lock-in; moderate openness allows innovation assets to cross a threshold that activates cross-domain spillovers; intermediate governance configurations may support higher long-run performance under certain conditions; and small governance adjustments may generate threshold-driven shifts in technological trajectories. These results integrate core mechanisms from transaction-cost economics, evolutionary theory, and dynamic capabilities by showing how technological trajectories emerge from the dynamic interaction of asset stocks. The model is broadly consistent with empirical patterns in which firms combining selective openness with targeted control outperform organizations that rely on extreme integration or extreme openness. The framework provides a tractable basis for analyzing how governance structures shape long-run technological change and offers several avenues for theoretical and empirical extension.
