Periodic Reporting for period 3 - ISEProD (Industrial Structure and the European Productivity Growth Divergence)
Berichtszeitraum: 2023-01-01 bis 2024-06-30
Zusammenfassung vom Kontext und den Gesamtzielen des Projekts
The last two decades have witnessed substantial progress in the measurement of productivity. As a consequence of these progresses, we are now aware of the enormous cross-firm dispersion in productivity as well as of the resulting losses of aggregate productivity due to resource misallocation. Against this progress on the measurement side, our understanding of the deep determinants of productivity growth is still limited. In fact, economists have not been able to offer a convincing explanation for the productivity slowdown that has characterized the US economy since the mid 2000's. The slowdown has been even stronger in Europe. Equally important, there is a clear divide between Southern European (SE) countries, where output per hour worked has been almost flat since the mid-nineties, and Northern Europe (NE), where it has trailed the US, although at a distance. Understanding the causes of such a divide is crucial, as a common currency area is unlikely to survive with divergent productivity performances of its members: absent the exchange rate, such divergences would have to be accommodated by changes in relative prices and wages, something extremely costly to manage in a low-inflation environment.
While there is evidence that product market reforms contributed to productivity growth in the nineties through increased competition, other factors might be at play. In 20 years of research, I investigated several determinants of firm productivity growth, such as labor market regulation, skills, trade, agglomeration economies, firm size distribution. My research indicates that industrial structure (IS), defined in terms of ownership (who owns the majority of shares of a firm), control (who runs it) and finance is key and that, due to technological change, its relevance has grown over time. Specifically, certain control structures, such as family control, are more reluctant to open up to external managerial skills. As such, they are less able to keep up with an increasingly complex economic environment. And given that such control structures are more common in SE, this can explain (part of) the productivity divergence.
I merge insights and state-of-the-art techniques from industrial organization and corporate finance to assess the role of IS for productivity growth. My ambition is not only to study the effects of different ISs, but also what determines the prevalence of some structures over others in different countries. These are difficult questions to answer, as they face identification issues -- the endogeneity of the ownership and control - that plague the empirical literature. Methodologically, I will use a combination of unique data sources (credit register, matched employer-employee data, firm input-output relationships), quasi-experimental design (changes in the banking regulation, executives’ mortality shocks in local labor markets) as well as structural techniques (selection models and Bayesian learning models) to address them. I also collect original data to study succession in family firms.
The ambition is to produce a set of results that, taken together, will generate a discontinuous change in our understanding of the relationship between corporate governance and productivity growth. On the way, I expect to contribute to the key debate on the diverging productivity performance of European economies.
While there is evidence that product market reforms contributed to productivity growth in the nineties through increased competition, other factors might be at play. In 20 years of research, I investigated several determinants of firm productivity growth, such as labor market regulation, skills, trade, agglomeration economies, firm size distribution. My research indicates that industrial structure (IS), defined in terms of ownership (who owns the majority of shares of a firm), control (who runs it) and finance is key and that, due to technological change, its relevance has grown over time. Specifically, certain control structures, such as family control, are more reluctant to open up to external managerial skills. As such, they are less able to keep up with an increasingly complex economic environment. And given that such control structures are more common in SE, this can explain (part of) the productivity divergence.
I merge insights and state-of-the-art techniques from industrial organization and corporate finance to assess the role of IS for productivity growth. My ambition is not only to study the effects of different ISs, but also what determines the prevalence of some structures over others in different countries. These are difficult questions to answer, as they face identification issues -- the endogeneity of the ownership and control - that plague the empirical literature. Methodologically, I will use a combination of unique data sources (credit register, matched employer-employee data, firm input-output relationships), quasi-experimental design (changes in the banking regulation, executives’ mortality shocks in local labor markets) as well as structural techniques (selection models and Bayesian learning models) to address them. I also collect original data to study succession in family firms.
The ambition is to produce a set of results that, taken together, will generate a discontinuous change in our understanding of the relationship between corporate governance and productivity growth. On the way, I expect to contribute to the key debate on the diverging productivity performance of European economies.
Arbeit, die ab Beginn des Projekts bis zum Ende des durch den Bericht erfassten Berichtszeitraums geleistet wurde, und die wichtigsten bis dahin erzielten Ergebnisse
One project investigates the determinants of succession in family firms, in particular, what determines the choice of internal succession (the new CEO is a family member), external succession (the new CEO is not a family member, but the family retains ownership of the firm) or outright sale of the firm. The aim is to understand to what extent this choice is dependent on the skills and aspirations of the former CEO’s children, and if there are cross- country differences in this sensitivity. The project entails an ambitious data collection effort through a questionnaire for the former CEOs. To prepare the questionnaire, we organized a series of focus groups with entrepreneurs close to succession. The questionnaire is completed. We decided to start with running the survey in Italy . On June 5,2022 we concluded the selection procedure of the firms that will conduct the pilot survey, which will go out after the summer.
Family ownership can affect performance independently from control, by determining the risk-return profile of the projects undertaken by the firm. In particular, entrepreneurs tend to be undiversified, as they typically hold a large share of their total wealth in the family firm. The second project tests the effects of the lack of diversification on firm performance using Danish administrative data. The project is well under way. We find that there is a clear trade-off between risk and returns: firms that grow more in terms of sales and assets also witness a higher variability of growth. This fact has never been documented. Moreover, we do find that entrepreneurs that are more leveraged in the family business tend to choose a lower risk-return combination.
The second work package is “Skills supply: Education, entrepreneurs and managers”.
In this work package we study the effects of skill supply on firm performance. A first project studies the effect of the supply of managerial skills at the local level on firm performance. Using exhaustive administrative data on Italian social security records, we construct measures of local labor market thickness for executives that vary by industry and location. We then show that firm performance is negatively affected by executive death, but only in thin local labor markets. The new executives hired after death events in thin local labor markets have lower education levels and are more likely to be replaced shortly after being hired, a fact that we interpreted as indicating that finding a good match between the firm and the executive is more difficult in thin markets. Taken together, these findings suggest that the scarcity of managerial skills is an important dimension in explaining differences in firm performance across industries and regions. From a policy perspective, they suggest that local policies aiming at boosting growth should take into consideration the supply of executive skills.
The outburst of Covid-19 raised new fundamental questions and research opportunities related to my research project. We used the Covid-19 shock to investigate if structured managerial practices (SMPs) help or hinder the firm's capacity to adapt to a changing environment. While the relationship between SMPs and performance is well-established for “business as usual'”, little is known about whether such practices help or hinder the response of firms to large shocks. From a theoretical perspective, there can be both positive and negative effects of SMPs on the firm capacity to adapt to a shock. On one hand, organizational practices centered around monitoring, targets and incentives provide firms with tools and information to sense opportunities and threats and address them by re-configuring and transforming firm organization. On the other hand, these practices may impose excessive structure and constraints on processes inside the firm, hindering flexibility and managerial discretion during a time in which these are particularly valuable. We use the spread of COVID-19 in Italy, the first Western country hit by the pandemic, to investigate the role of structured management practices in responding to a large shock. We find a sizable, positive effect of structured management practices on firm performance. Evidence points to the fact that firms with more structured practices were more likely to implement a comprehensive set of changes, such as demand, supply chain and labor management, including more intense use of remote work. This result is important, as it indicates that SMPs are useful not only in normal times, but also when facing large, unexpected shocks, an issue on which there is very little research.
The third work package is Industrial structure and IT adoption
We are investigating the effects of the diffusion of the broadband connection on firm innovation in Italy. Preliminary results show that when the broadband connection reaches a municipality, innovation output, measured by patents, increases. Moreover, firms are more likely to co-patent with other firms that are distant in space. We are extending the analysis to other proxies of innovation, such as total factor productivity, intangible capital, high-tech startups.
Fourth work package: Bank Financing: Asymmetric information and Market Structure
Bank lending remains the main source of firm finance in continental Europe. In this work package, we study the firm-bank relationship and how it can affect firm performance. In a first project, we study the effects on corporate loan rates of an unexpected change in the Italian legislation which forbade interlocking directorates between banks. Interlocking directorates indicates a situation in which the same person sits on the board of competing banks. A large literature suggests that interlocks can help corporations to sustain collusion, but no clear evidence so far exists on hypothesis. The Italian law and the structure of the market for corporate loans offer a unique opportunity to test this conjecture. We find that prohibiting interlocks decreased the interest rates of previously interlocked banks relative to other banks. Moreover, firms borrowing more from previously interlocked banks expand investment, employment and sales.This indicates that prohibiting interlocks is an effective way to foster competition and benefit the economy at large.
In another project, developed with ECB researchers, we use Anacredit, a new dataset of firm-bank credit relationships for all the Euro area countries maintained by the European Central Bank. We study how such relationships differ across firms located in different countries. In a set of preliminary results, we find that firms in Northern European countries tend to borrow from a smaller number of banks and rely more on one main bank. They also obtain credit with longer maturity. No systematic differences emerge in terms of the interest rates paid by firms. We interpret this evidence as indicating that firms in Northern European countries rely more on relationship lending, something that might help them to achieve higher productivity, as relationship lending allows the bank to finance more long-term, risky projects.
Finally, during the acute phase of the pandemic I developed a simple method to estimate firm liquidity needs in real time throughout the lockdown period. The project received considerable attention from policymakers, including the OECD, the ECB and the Italian government. The method was applied to assess if the measures put in place during the pandemic would have been sufficient to avoid massive bankruptcies.
Family ownership can affect performance independently from control, by determining the risk-return profile of the projects undertaken by the firm. In particular, entrepreneurs tend to be undiversified, as they typically hold a large share of their total wealth in the family firm. The second project tests the effects of the lack of diversification on firm performance using Danish administrative data. The project is well under way. We find that there is a clear trade-off between risk and returns: firms that grow more in terms of sales and assets also witness a higher variability of growth. This fact has never been documented. Moreover, we do find that entrepreneurs that are more leveraged in the family business tend to choose a lower risk-return combination.
The second work package is “Skills supply: Education, entrepreneurs and managers”.
In this work package we study the effects of skill supply on firm performance. A first project studies the effect of the supply of managerial skills at the local level on firm performance. Using exhaustive administrative data on Italian social security records, we construct measures of local labor market thickness for executives that vary by industry and location. We then show that firm performance is negatively affected by executive death, but only in thin local labor markets. The new executives hired after death events in thin local labor markets have lower education levels and are more likely to be replaced shortly after being hired, a fact that we interpreted as indicating that finding a good match between the firm and the executive is more difficult in thin markets. Taken together, these findings suggest that the scarcity of managerial skills is an important dimension in explaining differences in firm performance across industries and regions. From a policy perspective, they suggest that local policies aiming at boosting growth should take into consideration the supply of executive skills.
The outburst of Covid-19 raised new fundamental questions and research opportunities related to my research project. We used the Covid-19 shock to investigate if structured managerial practices (SMPs) help or hinder the firm's capacity to adapt to a changing environment. While the relationship between SMPs and performance is well-established for “business as usual'”, little is known about whether such practices help or hinder the response of firms to large shocks. From a theoretical perspective, there can be both positive and negative effects of SMPs on the firm capacity to adapt to a shock. On one hand, organizational practices centered around monitoring, targets and incentives provide firms with tools and information to sense opportunities and threats and address them by re-configuring and transforming firm organization. On the other hand, these practices may impose excessive structure and constraints on processes inside the firm, hindering flexibility and managerial discretion during a time in which these are particularly valuable. We use the spread of COVID-19 in Italy, the first Western country hit by the pandemic, to investigate the role of structured management practices in responding to a large shock. We find a sizable, positive effect of structured management practices on firm performance. Evidence points to the fact that firms with more structured practices were more likely to implement a comprehensive set of changes, such as demand, supply chain and labor management, including more intense use of remote work. This result is important, as it indicates that SMPs are useful not only in normal times, but also when facing large, unexpected shocks, an issue on which there is very little research.
The third work package is Industrial structure and IT adoption
We are investigating the effects of the diffusion of the broadband connection on firm innovation in Italy. Preliminary results show that when the broadband connection reaches a municipality, innovation output, measured by patents, increases. Moreover, firms are more likely to co-patent with other firms that are distant in space. We are extending the analysis to other proxies of innovation, such as total factor productivity, intangible capital, high-tech startups.
Fourth work package: Bank Financing: Asymmetric information and Market Structure
Bank lending remains the main source of firm finance in continental Europe. In this work package, we study the firm-bank relationship and how it can affect firm performance. In a first project, we study the effects on corporate loan rates of an unexpected change in the Italian legislation which forbade interlocking directorates between banks. Interlocking directorates indicates a situation in which the same person sits on the board of competing banks. A large literature suggests that interlocks can help corporations to sustain collusion, but no clear evidence so far exists on hypothesis. The Italian law and the structure of the market for corporate loans offer a unique opportunity to test this conjecture. We find that prohibiting interlocks decreased the interest rates of previously interlocked banks relative to other banks. Moreover, firms borrowing more from previously interlocked banks expand investment, employment and sales.This indicates that prohibiting interlocks is an effective way to foster competition and benefit the economy at large.
In another project, developed with ECB researchers, we use Anacredit, a new dataset of firm-bank credit relationships for all the Euro area countries maintained by the European Central Bank. We study how such relationships differ across firms located in different countries. In a set of preliminary results, we find that firms in Northern European countries tend to borrow from a smaller number of banks and rely more on one main bank. They also obtain credit with longer maturity. No systematic differences emerge in terms of the interest rates paid by firms. We interpret this evidence as indicating that firms in Northern European countries rely more on relationship lending, something that might help them to achieve higher productivity, as relationship lending allows the bank to finance more long-term, risky projects.
Finally, during the acute phase of the pandemic I developed a simple method to estimate firm liquidity needs in real time throughout the lockdown period. The project received considerable attention from policymakers, including the OECD, the ECB and the Italian government. The method was applied to assess if the measures put in place during the pandemic would have been sufficient to avoid massive bankruptcies.
Fortschritte, die über den aktuellen Stand der Technik hinausgehen und voraussichtliche potenzielle Auswirkungen (einschließlich der bis dato erzielten sozioökonomischen Auswirkungen und weiter gefassten gesellschaftlichen Auswirkungen des Projekts)
With respect to the state of the art, the project takes a more comprehensive view of the determinants of productivity growth and uses it to assess the roots of the productivity growth divide between Northern and Southern Europe.
Specifically, the project pursues four specific objectives in four deeply integrated work packages: a) The role of firm ownership and control on skill acquisition and risk taking; b) Skills supply: the effects of the education of entrepreneurs and managers; c) Industrial structure, input/output relationships and IT adoption; d) Access to finance in imperfectly competitive financial markets. My objective is to make progress within each work package and to use the results jointly to assess the sources of productivity growth. The ambition is to generate a discontinuous change in our understanding of the relationship between corporate governance/finance and productivity growth. On the way, I expect to contribute to the key debate for the future of the European Union on the diverging productivity performance of European economies. My results could signal that the approach “one package of structural reforms fits all” is inadequate, as different countries have inherently different ISs, and therefore policy needs.
Specifically, the project pursues four specific objectives in four deeply integrated work packages: a) The role of firm ownership and control on skill acquisition and risk taking; b) Skills supply: the effects of the education of entrepreneurs and managers; c) Industrial structure, input/output relationships and IT adoption; d) Access to finance in imperfectly competitive financial markets. My objective is to make progress within each work package and to use the results jointly to assess the sources of productivity growth. The ambition is to generate a discontinuous change in our understanding of the relationship between corporate governance/finance and productivity growth. On the way, I expect to contribute to the key debate for the future of the European Union on the diverging productivity performance of European economies. My results could signal that the approach “one package of structural reforms fits all” is inadequate, as different countries have inherently different ISs, and therefore policy needs.