Abstract:

We build a simple international trade model to study how the interaction between technological learning and imperfect market selection shapes the dynamics of firms, industries and export flows. The model features two countries populated by firms heterogeneous in productivity and size. Market selection in each country is driven by a finite pairwise Pólya urn process, which embodies dynamic increasing returns in firm size. We show that, with a static distribution of firm productivity, market selection leads either to an international or to a national monopoly depending on the entity of trade barriers. In presence of firm learning and entry-exit of firms, the firm productivity distribution changes over time and the model generates non-monopolistic marke structures, whose properties depend on trade openness and market selection intensity. Finally, we show that the the model is able to jointly reproduce a wide ensemble of stylized facts concerning intra-industry trade, industry and firm dynamics as well as realistic dynamics of productivity and export leadership at the country level. 

Abstract:

We provide a survey of the main mechanisms of market selection used in economics. We categorise existing theories into three broad classes -- evolutionary selection, reduced form selection, and rational equilibrium -- based on their adopted selection mechanisms. Each paradigm is explored in terms of underlying laws of selection, searching for elements of convergence and divergence in epistemological approaches, hypotheses and results. The comparison of these paradigms reveals convergences in research directions, particularly in replicating empirical patterns related to firm heterogeneity and acknowledging the role of increasing returns. However, these paradigms diverge in key assumptions and results, including emphasis on model outcomes, sources of increasing returns, mechanisms generating firm heterogeneity, and assumptions regarding firm rationality. The discussion highlights that these differences stem from the epistemological foundations of paradigms. The survey contributes to a nuanced understanding of market selection mechanisms within diverse theoretical frameworks, emphasising both areas of convergence and divergence among them.

Abstract:

This report analyses the use of artificial intelligence (AI) in firms across 11 countries. Based on harmonised statistical code (AI diffuse) applied to official firm-level surveys, it finds that the use of AI is prevalent in ICT and Professional Services and more widespread across large – and to some extent across young – firms. AI users tend to be more productive, especially the largest ones. Complementary assets, including ICT skills, high-speed digital infrastructure, and the use of other digital technologies, which are significantly related to the use of AI, appear to play a critical role in the productivity advantages of AI users.

Abstract:

 In this work we characterise French firms using artificial intelligence (AI) and explore the link between AI use and productivity. We relevantly distinguish AI users that source AI from external providers (AI buyers) from those developing their own AI systems (AI developers). AI buyers tend to be larger than other firms, while AI developers are also younger. The share of firms using AI is highest in the ICT sector, which exhibits a particularly high share of developers. Complementary assets, including digital skills, capabilities and infrastructure, play a key role for AI use, with AI buyers and developers leveraging different types of human capital. Overall, AI users tend to be more productive, however this appears largely related to the self-selection of more productive and digital-intensive firms into AI use. This is not the case for AI developers, for which the positive link between AI use and productivity remains evident beyond selection, suggesting a positive effect of AI on their productivity.

Abstract:

We use decomposition techniques on data from EXIOBASE 3 (Stadler et al., 2018) between 1995 and 2011 to study the main drivers of the increase in CO2 emissions and of the decrease in CO2

intensities at the global, sectoral and national level. To this end, we modify the FHK decomposition approach (see Foster et al., 2001), which allows to study the impact of the key drivers while

accounting for their covariance, thus controlling for contemporaneous variations. Our findings show that emission intensity at the global level has reduced over time, as a result of technological

improvements. Yet, the role of technological improvements has not been constant, and has slowed down over time. Conversely, the upward pressure resulting from increased production

is continuing, resulting in higher aggregate emission levels. We show that within each sector, production tended to relocate towards countries with higher emission intensities. Conversely,

we illustrate that some countries successfully moved away from emission intensive sectors and promoted overall emission reductions. We discuss the implications of these results for informing

climate and technology policies in the context of the transition towards carbon neutrality, with a particular focus on the role of demand-side interventions.

Abstract:

In this work we investigate the firm growth rates volatility-size relation and its determinants in a comprehensive dataset of French manufacturing firms between 1994 and 2018. Differently from previous contributions, we study the relation using sales data for firms at both the aggregate and sectoral level. First, we show that the relation deviates from the linear approximation found in previous studies. It is indeed J-shaped, very steep for small firms and flat for large ones. Second, we explain this new empirical finding via a tractable model of imperfect selection encompassing firms competing on the basis of both size and productivity. Our contribution suggests that large firms are Gibrat's and that the empirical shape of the firms' growth rates variance-size relation can be explained by imperfect selection mechanisms whose outcomes are mediated by both the strength of shares reallocation and firms'  joint heterogeneity in size and productivity.

Abstract:

In this study, we investigate the relationship between firms' adoption of artificial intelligence (AI) and risk, measured as the firm-level volatility of productivity growth rates. Our findings support a positive association between AI usage and risk. To address potential endogeneity issues, we employ a Coarsened Exact Matching design, providing evidence that the observed relationship is causal. Furthermore, we explore potential mediating factors between AI and risk and find that AI users with larger shares of ICT workers are associated with lower risk. Additionally, we distinguish between AI buyers (i.e., firms employing AI technologies developed by external providers) and AI developers (i.e., firms utilizing internally created AI systems). Our results indicate that the positive relationship between AI use and risk is primarily driven by AI buyers, with no significant association found for AI developers. In conclusion, our work suggests that the use of AI increases the risk for firms when appropriate complementary assets are lacking.

Abstract:

We exploit a uniquely comprehensive combination of data sources to explore the complementarities between human capital and AI in French firms. We show that firms using AI leverage the presence of ICT engineers, with highly specialised ICT occupations, notably ICT R&D engineers and ICT managers, playing a key role. The presence of ICT engineers was significant also in 2011, when AI use by firms was highly unlikely, and when causal estimation strategies are considered, suggesting that ICT human capital is crucial for adopting AI. Our analysis suggests that significant investments in ICT human capital development are necessary to foster the diffusion of AI among firms. Finally, we show that AI-human capital complementarities are heterogeneous across AI users and sectors. Firms developing their own AI systems leverage a broader set of higher-level intellectual occupations beyond the ICT domain with respect to firms buying AI from external providers. AI-human capital relations are also heterogeneous across economic sectors, with ICT occupations being more relevant in services while non-ICT technical occupations in manufacturing.

Abstract:

We investigate the impact of the purchase of cloud services on the size growth rate of French firms. Our findings indicate that cloud services positively impact firm growth rates, with smaller firms experiencing more significant benefits compared to larger firms. When distinguishing different types of cloud services, we find that ones associated to the use of software and the purchase of computational power induce larger returns for smaller firms than cloud services providing virtual storage space. This evidence suggests that the diffusion of cloud technologies may help reduce concentration by favoring the digitalisation and growth of smaller firms, especially when the cloud services provided are more advanced.