François GEEROLF - Academia.edu (original) (raw)
Papers by François GEEROLF
Le Centre pour la Communication Scientifique Directe - HAL - Diderot, Feb 24, 2022
L'Europe, de réelles avancées mais des choix à assumer OFCE Policy brief ■ 101 ■ 24 février 2022 ... more L'Europe, de réelles avancées mais des choix à assumer OFCE Policy brief ■ 101 ■ 24 février 2022 2 Directeur de la publication Xavier Ragot Rédactrice en chef du blog et des Policy briefs Christine Rifflart Réalisation Najette Moummi (OFCE).
Why we can't afford the rich
A model of knowledge-based production hierarchies based on Garicano (2000) generates Zipf’s Law f... more A model of knowledge-based production hierarchies based on Garicano (2000) generates Zipf’s Law for firm sizes in the upper tail, with very minimal assumptions on the underlying distribution of agents’ skills. When the density distribution of higher skills is bounded away from zero, the span of control of managers follows a Pareto distribution of endogenous tail coecient equal to two in the upper tail between adjacent levels of hierarchical organization, four thirds between two of these levels, and in general 2 L =(2 L 1) between L levels, converging to one, or Zipf’s law, when the number of levels increases. These ancillary predictions for span of control of intermediary managers are verified in the French matched employer-employee data, with a very good degree of precision. The model attributes firms’ size unbounded heterogeneity as well as high labor income inequality to a potentially arbitrarily small heterogeneity in agents’ skills, without any functional form assumption. The m...
We study the macroeconomic effects of aggregate tax changes using more than 100 property tax chan... more We study the macroeconomic effects of aggregate tax changes using more than 100 property tax changes in advanced economies identified through the narrative record, and a structural VAR approach. Both methodologies lead to very similar estimates of tax multipliers that are higher than 2. The motivation behind using property taxes is threefold. First, property taxes are in theory the least distortive of all taxes, which allows to interpret our tax multipliers in terms of disposable income effects, and not in terms of supply or incentive effects. Second, the base for property taxes is not contemporaneously affected by GDP, unlike other major tax revenues, which considerably eases inference both in the narrative and in the structural VAR approaches. Third, the effects of property tax changes inform more broadly on the consequences of policies shifting the user cost of owner-occupied housing (including monetary policy). It casts a new light on a growing literature investigating the links...
In a seminal paper, Abel et al. (1989) argue that the United States and 6 other major advanced ec... more In a seminal paper, Abel et al. (1989) argue that the United States and 6 other major advanced economies are dynamically ecient. Updating data on mixed income and land rents, I find in contrast that the criterion for dynamic eciency is not verified for any advanced economy ; and that Japan and South Korea have unambiguously over-accumulated capital. This world "savings glut" can potentially explain otherwise hard-to-understand macroeconomic stylized facts (low interest rates, cash holding by firms, financial bubbles). Subject to some caveats, an increase of public debt, or a generalization of pay-as-you-go systems could therefore be Pareto-improving.
I build a model of the cross section of leverage ratios for borrowers based on heterogenous belie... more I build a model of the cross section of leverage ratios for borrowers based on heterogenous beliefs about future asset returns and endogenous collateral constraints. In equilibrium, borrowers and lenders are matched in an assortative way according to their relative optimism through hedonic interest rates, which are disconnected from risk aversion or expected default probabilities. Under minimal assumptions on the underlying distribution of beliefs, the leverage ratio distribution of borrowers is Pareto in the upper tail, a prediction verified in micro-level administrative data for homeowners, publicly available data for entrepreneurs and in the TASS database for hedge funds. Expected and realized returns to levered portfolios are very skewed and fat tailed, even when heterogeneity in beliefs vanishes. The market features a high degree of customization and fragmentation, as many real world financial markets which are organized over-the-counter. Pyramiding lending arrangements (equiva...
I build a model of the cross section of leverage ratios for borrowers - for example, traders or e... more I build a model of the cross section of leverage ratios for borrowers - for example, traders or entrepreneurs - assuming heterogenous beliefs about future asset returns and endogenous collateral constraints a la Geanakoplos (1997). Under minimal assumptions on the underlying distribution of beliefs, the leverage ratios of borrowers follow a Pareto distribution in the upper tail, with an endogenous tail coecient equal to two, a prediction precisely verified for hedge funds in the TASS database. This Pareto distribution can lead to similar Pareto distributions for the returns of borrowers, be they investment bankers or entrepreneurs, with empirically plausible Pareto-Lorenz coecients between one and three, providing a new intuition for the skewness of the income distribution. In the model, borrowers and lenders are matched assortatively according to their relative levels of optimism, which can help explain the Over-The-Counter structure of many collateralized asset markets : borrowers...
In a seminal paper, Abel et al. (1989) argue that the United States and six other major advanced ... more In a seminal paper, Abel et al. (1989) argue that the United States and six other major advanced economies are dynamically efficient. Updating data on mixed income and land rents, I find in contrast that the criterion for dynamic efficiency is not verified for any advanced economy, and that Japan and South Korea have unambiguously over-accumulated capital. This world “savings glut” can potentially explain otherwise hard-to-understand macroeconomic stylized facts low interest rates, cash holding by firms, financial bubbles. It is also the macroeconomic counterpart of the microeconomic observation that average firms’ return on investment is lower than their measured cost of capital. Subject to some caveats, an increase of public debt, or a generalization of pay-as-you-go systems could therefore be Pareto-improving.
We investigate the causal effect of house price movements on unemployment dynamics. Using a datas... more We investigate the causal effect of house price movements on unemployment dynamics. Using a dataset of 34 countries over the last 40 years, we show the large and significant impact of house prices on unemployment fluctuations using property taxes as an instrument for house prices. A 10% (instrumented) appreciation in house prices yields to a 3.4% decrease in the unemployment rate. These results are very robust to the inclusion of the variables commonly used to explain unemployment rate developments. If house prices directly impact employment in construction, job volatility in this sector resulting in large employment fluctuations, they impact also total employment through their effects on non-residential investment and consumption, two determinants of labour demand. Housing booms have a specific effect on employment in the tradable sector as they lead to real exchange rate appreciations that affect manufacturing activity.
Quels sont les effets d'une baisse ou d'une hausse d'impot sur l’economie ? Tous les ... more Quels sont les effets d'une baisse ou d'une hausse d'impot sur l’economie ? Tous les impots ont-ils le meme effet ? Ces questions font toujours debat. Elles sont pourtant cruciales pour les choix de politique economique. Par exemple, le FMI a reconnu apres la crise des dettes souveraines avoir sous-estime l'impact negatif des hausses d'impots sur l'activite – ce qui a pu contribuer a plonger plusieurs economies europeennes dans la recession. De meme, savoir s'il faut privilegier des baisses d'impots pour les menages, cibler les riches ou les pauvres, ou s’il faut accorder des baisses de charges aux entreprises suscite questions et controverses. La reforme fiscale aux Etats-Unis, celle du CICE, de l’ISF ou de la taxe d’habitation en France ont ete l’occasion d’en debattre. Cette Lettre presente une etude du CEPII (Geerolf et Grjebine, 2018)1, qui apporte un eclairage nouveau sur ces questions en mesurant les effets sur l’economie de changements de taxe...
I study a standard growth model with a superelliptic production function, a generalization of the... more I study a standard growth model with a superelliptic production function, a generalization of the Cobb-Douglas with satiation of capital. When saving is scarce or investment demand is high, the growth model has a neoclassical regime, where output is determined only by supply factors, more saving fosters capital accumulation and growth, and labor is fully employed. When saving is abundant or investment demand is low, the model has a secular stagnation regime, where output is determined by aggregate demand in the long-run, even though prices are fully flexible. Through the lens of this model, secular stagnation is interpreted as an advanced state of dynamic inefficiency. There is a paradox of thrift: efforts to save more are self-defeating, and lead to lower saving, lower capital accumulation and growth, and underemployment. Deficit-financed government expenditures, as well as permanent tax cuts increase output, employment, consumption and investment permanently. Quantitatively, the m...
La crise du Covid-19 a mis en lumière la dépendance de la France aux importations et à la product... more La crise du Covid-19 a mis en lumière la dépendance de la France aux importations et à la production industrielle étrangère, notamment pour les médicaments, les tests, les masques ou les respirateurs. Le problème n’est pas, d’un point de vue économique, que la France ne produise pas des biens à faible valeur ajoutée comme les masques, dans lesquels elle n’a pas intérêt à se spécialiser. Le problème n’est pas non plus réductible au fonctionnement des chaînes de valeur mondiales qui impliquent un processus de production fragmenté au niveau international. Ce que révèle plutôt au grand jour cette crise, c’est la désindustrialisation accélérée que connaît la France depuis vingt ans et ses déséquilibres commerciaux persistants. Pourtant, pour beaucoup d’économistes, la taille du secteur industriel et les défi cits commerciaux ne sont pas des problèmes en soi. Ils s’accordent généralement pour critiquer le « fétichisme industriel » des politiques qu’ils attribuent à un colbertisme d’un aut...
Alors que les réformes fiscales constituent un élément central des politiques publiques, leurs co... more Alors que les réformes fiscales constituent un élément central des politiques publiques, leurs conséquences macroéconomiques font toujours l’objet d’intenses controverses. Pourtant, il est tout à fait essentiel d’en connaître avec précision les effets sur les ménages et sur l’activité des entreprises pour la bonne gestion macroéconomique, ou pour utiliser au mieux la politique fiscale à l’occasion des crises. Par exemple, savoir s’il faut relancer la consommation, ou s’il vaut mieux accorder des baisses de charges afin de relancer l’investissement et les embauches sont des questions qui ne cessent d’être débattues, comme en témoignent les interrogations récentes autour des réformes du CICE, de l’ISF ou de la taxe d’habitation en France. En outre, les erreurs qui peuvent être commises en matière de politique fiscale, faute de bien en apprécier les effets, ont des conséquences lourdes et qui peuvent être durables. C’est ainsi que le Fonds monétaire international a reconnu après la cri...
I propose a competitive asset pricing model based on the scarcity and heterogeneity of lenders. E... more I propose a competitive asset pricing model based on the scarcity and heterogeneity of lenders. Expected excess returns on a given investment do not just depend on the investment itself and the identity of the investor, but on aggregate factors such as the whole distribution of investors’ wealth and borrowers’ wealth, investors’ heterogeneity and borrowers’ heterogeneity. The model implies that expected returns have a common factor which cannot be traced to characteristics of asset, or the investor holding this asset, unlike in both conventional asset pricing and intermediary-based asset pricing. These asset pricing implications are obtained by microfounding lenders’ technologies explicitly, in line with textbook models of corporate finance and financial frictions, and allowing heterogeneity among them; it is conceptually different from existing theories, in that it does not rely on risk aversion or liquidity. Applications of the model to syndicated loans, private equity, and ventur...
We use variations in property taxes in a panel of more than 20 countries, and over 40 years, to i... more We use variations in property taxes in a panel of more than 20 countries, and over 40 years, to investigate one source of joint determination between house prices and output, employment, consumer spending, investment and the trade balance. We use a narrative approach to identify changes in property taxes that are automatic or politically motivated, and thus exogenous to the business cycle. We also use the easiness with which property tax increases can be passed on to renters depending on landlord-tenants regulations across countries as a source of overidentification. In preliminary results, we find large non-ricardian effects. Property tax increases are highly contractionary and the economic effect is large, with a multiplier strictly higher than one. Regressions of housing on macroeconomic aggregates, plagued by the omission of the joint effect of property taxes, would suggest a marginal propensity to consume out of housing wealth of about 9 cents, and a marginal propensity to inve...
A strong empirical regularity is that firm size and top incomes follow a Pareto distribution. A l... more A strong empirical regularity is that firm size and top incomes follow a Pareto distribution. A large literature explains this regularity by appealing to the distribution of primitives, or by using dynamic “random growth” models. In contrast, I demonstrate that Pareto distributions can arise from production functions in static assignment models with complementarities, such as Garicano’s (2000) knowledgebased production hierarchies model. Under very limited assumptions on the distribution of agents’ abilities, these models generate Pareto distributions for the span of control of CEOs and intermediary managers, and Zipf’s law for firm size. I confirm this prediction using French matched employer-employee administrative data. This novel justification of Pareto distributions sheds new light on why firm size and labor income are so heterogeneous despite small observable dierences. In the model, Pareto distributions are the benchmark distributions that arise in the case of perfect homogen...
I propose a theory of Zipf's Law for firm sizes and Pareto distributions for top labor income... more I propose a theory of Zipf's Law for firm sizes and Pareto distributions for top labor incomes based on arbitrarily small differences in skills between workers, and no functional form assumption. A version of Garicano (2000)'s knowledge-based production hierarchies model is shown to generate Pareto distributions for high span of control of intermediary managers with coefficients equal to 2L/(2L-1) between L levels of management, under very limited assumptions on the underlying distribution of agents' skills. This breakdown of the aggregate firm size distribution receives considerable empirical support in the French matched employer-employee administrative data. The model provides a method to structurally estimate the relative importance of different factors in the recent rise in labor income inequality: the tail index of the Pareto top labor income distribution depends on a notion of race between education and technology, while the scale parameter depends on communicatio...
The Phillips Curve is a negative relationship between inflation and unemployment but has been rep... more The Phillips Curve is a negative relationship between inflation and unemployment but has been repeatedly challenged: missing deflation in 1933-1941 and 2007-2009, stagflation in the 1970s, missing inflation in the late 1990s and 2010s, contrasting with always strong regional Phillips Curves across states, cities and counties. This paper reconciles these inconsistencies by distinguishing between fixed and flexible exchange rates and hypothesizing a real exchange rate Phillips Curve: a correlation between real exchange rate growth and unemployment. Inflation correlates with unemployment only with fixed exchange rates, when real exchange rate growth equals inflation, explaining Phillips’ original correlation. According to this interpretation, the 1970s stagflation was not due to expectations changing, but to the end of the Bretton-Woods system of fixed exchange rates. The real exchange rate Phillips Curve is consistent with a demand side secular stagnation model based on an excess of s...
Politique étrangère, 2021
Les inquietudes souvent avancees sur la dette publique reposent sur nombre d’idees recues. Le rat... more Les inquietudes souvent avancees sur la dette publique reposent sur nombre d’idees recues. Le ratio de dette sur PIB est une mesure peu signifiante, et la dette publique absorbe l’epargne privee en exces. Le deficit public ne nuit pas au financement de l’investissement prive et n’est pas un fardeau pour les generations futures. Mais la dette publique doit etre geree serieusement, mise au service de projets permettant de satisfaire les besoins et objectifs societaux, des lors que l’epargne privee ne les finance pas.
Le Centre pour la Communication Scientifique Directe - HAL - Diderot, Feb 24, 2022
L'Europe, de réelles avancées mais des choix à assumer OFCE Policy brief ■ 101 ■ 24 février 2022 ... more L'Europe, de réelles avancées mais des choix à assumer OFCE Policy brief ■ 101 ■ 24 février 2022 2 Directeur de la publication Xavier Ragot Rédactrice en chef du blog et des Policy briefs Christine Rifflart Réalisation Najette Moummi (OFCE).
Why we can't afford the rich
A model of knowledge-based production hierarchies based on Garicano (2000) generates Zipf’s Law f... more A model of knowledge-based production hierarchies based on Garicano (2000) generates Zipf’s Law for firm sizes in the upper tail, with very minimal assumptions on the underlying distribution of agents’ skills. When the density distribution of higher skills is bounded away from zero, the span of control of managers follows a Pareto distribution of endogenous tail coecient equal to two in the upper tail between adjacent levels of hierarchical organization, four thirds between two of these levels, and in general 2 L =(2 L 1) between L levels, converging to one, or Zipf’s law, when the number of levels increases. These ancillary predictions for span of control of intermediary managers are verified in the French matched employer-employee data, with a very good degree of precision. The model attributes firms’ size unbounded heterogeneity as well as high labor income inequality to a potentially arbitrarily small heterogeneity in agents’ skills, without any functional form assumption. The m...
We study the macroeconomic effects of aggregate tax changes using more than 100 property tax chan... more We study the macroeconomic effects of aggregate tax changes using more than 100 property tax changes in advanced economies identified through the narrative record, and a structural VAR approach. Both methodologies lead to very similar estimates of tax multipliers that are higher than 2. The motivation behind using property taxes is threefold. First, property taxes are in theory the least distortive of all taxes, which allows to interpret our tax multipliers in terms of disposable income effects, and not in terms of supply or incentive effects. Second, the base for property taxes is not contemporaneously affected by GDP, unlike other major tax revenues, which considerably eases inference both in the narrative and in the structural VAR approaches. Third, the effects of property tax changes inform more broadly on the consequences of policies shifting the user cost of owner-occupied housing (including monetary policy). It casts a new light on a growing literature investigating the links...
In a seminal paper, Abel et al. (1989) argue that the United States and 6 other major advanced ec... more In a seminal paper, Abel et al. (1989) argue that the United States and 6 other major advanced economies are dynamically ecient. Updating data on mixed income and land rents, I find in contrast that the criterion for dynamic eciency is not verified for any advanced economy ; and that Japan and South Korea have unambiguously over-accumulated capital. This world "savings glut" can potentially explain otherwise hard-to-understand macroeconomic stylized facts (low interest rates, cash holding by firms, financial bubbles). Subject to some caveats, an increase of public debt, or a generalization of pay-as-you-go systems could therefore be Pareto-improving.
I build a model of the cross section of leverage ratios for borrowers based on heterogenous belie... more I build a model of the cross section of leverage ratios for borrowers based on heterogenous beliefs about future asset returns and endogenous collateral constraints. In equilibrium, borrowers and lenders are matched in an assortative way according to their relative optimism through hedonic interest rates, which are disconnected from risk aversion or expected default probabilities. Under minimal assumptions on the underlying distribution of beliefs, the leverage ratio distribution of borrowers is Pareto in the upper tail, a prediction verified in micro-level administrative data for homeowners, publicly available data for entrepreneurs and in the TASS database for hedge funds. Expected and realized returns to levered portfolios are very skewed and fat tailed, even when heterogeneity in beliefs vanishes. The market features a high degree of customization and fragmentation, as many real world financial markets which are organized over-the-counter. Pyramiding lending arrangements (equiva...
I build a model of the cross section of leverage ratios for borrowers - for example, traders or e... more I build a model of the cross section of leverage ratios for borrowers - for example, traders or entrepreneurs - assuming heterogenous beliefs about future asset returns and endogenous collateral constraints a la Geanakoplos (1997). Under minimal assumptions on the underlying distribution of beliefs, the leverage ratios of borrowers follow a Pareto distribution in the upper tail, with an endogenous tail coecient equal to two, a prediction precisely verified for hedge funds in the TASS database. This Pareto distribution can lead to similar Pareto distributions for the returns of borrowers, be they investment bankers or entrepreneurs, with empirically plausible Pareto-Lorenz coecients between one and three, providing a new intuition for the skewness of the income distribution. In the model, borrowers and lenders are matched assortatively according to their relative levels of optimism, which can help explain the Over-The-Counter structure of many collateralized asset markets : borrowers...
In a seminal paper, Abel et al. (1989) argue that the United States and six other major advanced ... more In a seminal paper, Abel et al. (1989) argue that the United States and six other major advanced economies are dynamically efficient. Updating data on mixed income and land rents, I find in contrast that the criterion for dynamic efficiency is not verified for any advanced economy, and that Japan and South Korea have unambiguously over-accumulated capital. This world “savings glut” can potentially explain otherwise hard-to-understand macroeconomic stylized facts low interest rates, cash holding by firms, financial bubbles. It is also the macroeconomic counterpart of the microeconomic observation that average firms’ return on investment is lower than their measured cost of capital. Subject to some caveats, an increase of public debt, or a generalization of pay-as-you-go systems could therefore be Pareto-improving.
We investigate the causal effect of house price movements on unemployment dynamics. Using a datas... more We investigate the causal effect of house price movements on unemployment dynamics. Using a dataset of 34 countries over the last 40 years, we show the large and significant impact of house prices on unemployment fluctuations using property taxes as an instrument for house prices. A 10% (instrumented) appreciation in house prices yields to a 3.4% decrease in the unemployment rate. These results are very robust to the inclusion of the variables commonly used to explain unemployment rate developments. If house prices directly impact employment in construction, job volatility in this sector resulting in large employment fluctuations, they impact also total employment through their effects on non-residential investment and consumption, two determinants of labour demand. Housing booms have a specific effect on employment in the tradable sector as they lead to real exchange rate appreciations that affect manufacturing activity.
Quels sont les effets d'une baisse ou d'une hausse d'impot sur l’economie ? Tous les ... more Quels sont les effets d'une baisse ou d'une hausse d'impot sur l’economie ? Tous les impots ont-ils le meme effet ? Ces questions font toujours debat. Elles sont pourtant cruciales pour les choix de politique economique. Par exemple, le FMI a reconnu apres la crise des dettes souveraines avoir sous-estime l'impact negatif des hausses d'impots sur l'activite – ce qui a pu contribuer a plonger plusieurs economies europeennes dans la recession. De meme, savoir s'il faut privilegier des baisses d'impots pour les menages, cibler les riches ou les pauvres, ou s’il faut accorder des baisses de charges aux entreprises suscite questions et controverses. La reforme fiscale aux Etats-Unis, celle du CICE, de l’ISF ou de la taxe d’habitation en France ont ete l’occasion d’en debattre. Cette Lettre presente une etude du CEPII (Geerolf et Grjebine, 2018)1, qui apporte un eclairage nouveau sur ces questions en mesurant les effets sur l’economie de changements de taxe...
I study a standard growth model with a superelliptic production function, a generalization of the... more I study a standard growth model with a superelliptic production function, a generalization of the Cobb-Douglas with satiation of capital. When saving is scarce or investment demand is high, the growth model has a neoclassical regime, where output is determined only by supply factors, more saving fosters capital accumulation and growth, and labor is fully employed. When saving is abundant or investment demand is low, the model has a secular stagnation regime, where output is determined by aggregate demand in the long-run, even though prices are fully flexible. Through the lens of this model, secular stagnation is interpreted as an advanced state of dynamic inefficiency. There is a paradox of thrift: efforts to save more are self-defeating, and lead to lower saving, lower capital accumulation and growth, and underemployment. Deficit-financed government expenditures, as well as permanent tax cuts increase output, employment, consumption and investment permanently. Quantitatively, the m...
La crise du Covid-19 a mis en lumière la dépendance de la France aux importations et à la product... more La crise du Covid-19 a mis en lumière la dépendance de la France aux importations et à la production industrielle étrangère, notamment pour les médicaments, les tests, les masques ou les respirateurs. Le problème n’est pas, d’un point de vue économique, que la France ne produise pas des biens à faible valeur ajoutée comme les masques, dans lesquels elle n’a pas intérêt à se spécialiser. Le problème n’est pas non plus réductible au fonctionnement des chaînes de valeur mondiales qui impliquent un processus de production fragmenté au niveau international. Ce que révèle plutôt au grand jour cette crise, c’est la désindustrialisation accélérée que connaît la France depuis vingt ans et ses déséquilibres commerciaux persistants. Pourtant, pour beaucoup d’économistes, la taille du secteur industriel et les défi cits commerciaux ne sont pas des problèmes en soi. Ils s’accordent généralement pour critiquer le « fétichisme industriel » des politiques qu’ils attribuent à un colbertisme d’un aut...
Alors que les réformes fiscales constituent un élément central des politiques publiques, leurs co... more Alors que les réformes fiscales constituent un élément central des politiques publiques, leurs conséquences macroéconomiques font toujours l’objet d’intenses controverses. Pourtant, il est tout à fait essentiel d’en connaître avec précision les effets sur les ménages et sur l’activité des entreprises pour la bonne gestion macroéconomique, ou pour utiliser au mieux la politique fiscale à l’occasion des crises. Par exemple, savoir s’il faut relancer la consommation, ou s’il vaut mieux accorder des baisses de charges afin de relancer l’investissement et les embauches sont des questions qui ne cessent d’être débattues, comme en témoignent les interrogations récentes autour des réformes du CICE, de l’ISF ou de la taxe d’habitation en France. En outre, les erreurs qui peuvent être commises en matière de politique fiscale, faute de bien en apprécier les effets, ont des conséquences lourdes et qui peuvent être durables. C’est ainsi que le Fonds monétaire international a reconnu après la cri...
I propose a competitive asset pricing model based on the scarcity and heterogeneity of lenders. E... more I propose a competitive asset pricing model based on the scarcity and heterogeneity of lenders. Expected excess returns on a given investment do not just depend on the investment itself and the identity of the investor, but on aggregate factors such as the whole distribution of investors’ wealth and borrowers’ wealth, investors’ heterogeneity and borrowers’ heterogeneity. The model implies that expected returns have a common factor which cannot be traced to characteristics of asset, or the investor holding this asset, unlike in both conventional asset pricing and intermediary-based asset pricing. These asset pricing implications are obtained by microfounding lenders’ technologies explicitly, in line with textbook models of corporate finance and financial frictions, and allowing heterogeneity among them; it is conceptually different from existing theories, in that it does not rely on risk aversion or liquidity. Applications of the model to syndicated loans, private equity, and ventur...
We use variations in property taxes in a panel of more than 20 countries, and over 40 years, to i... more We use variations in property taxes in a panel of more than 20 countries, and over 40 years, to investigate one source of joint determination between house prices and output, employment, consumer spending, investment and the trade balance. We use a narrative approach to identify changes in property taxes that are automatic or politically motivated, and thus exogenous to the business cycle. We also use the easiness with which property tax increases can be passed on to renters depending on landlord-tenants regulations across countries as a source of overidentification. In preliminary results, we find large non-ricardian effects. Property tax increases are highly contractionary and the economic effect is large, with a multiplier strictly higher than one. Regressions of housing on macroeconomic aggregates, plagued by the omission of the joint effect of property taxes, would suggest a marginal propensity to consume out of housing wealth of about 9 cents, and a marginal propensity to inve...
A strong empirical regularity is that firm size and top incomes follow a Pareto distribution. A l... more A strong empirical regularity is that firm size and top incomes follow a Pareto distribution. A large literature explains this regularity by appealing to the distribution of primitives, or by using dynamic “random growth” models. In contrast, I demonstrate that Pareto distributions can arise from production functions in static assignment models with complementarities, such as Garicano’s (2000) knowledgebased production hierarchies model. Under very limited assumptions on the distribution of agents’ abilities, these models generate Pareto distributions for the span of control of CEOs and intermediary managers, and Zipf’s law for firm size. I confirm this prediction using French matched employer-employee administrative data. This novel justification of Pareto distributions sheds new light on why firm size and labor income are so heterogeneous despite small observable dierences. In the model, Pareto distributions are the benchmark distributions that arise in the case of perfect homogen...
I propose a theory of Zipf's Law for firm sizes and Pareto distributions for top labor income... more I propose a theory of Zipf's Law for firm sizes and Pareto distributions for top labor incomes based on arbitrarily small differences in skills between workers, and no functional form assumption. A version of Garicano (2000)'s knowledge-based production hierarchies model is shown to generate Pareto distributions for high span of control of intermediary managers with coefficients equal to 2L/(2L-1) between L levels of management, under very limited assumptions on the underlying distribution of agents' skills. This breakdown of the aggregate firm size distribution receives considerable empirical support in the French matched employer-employee administrative data. The model provides a method to structurally estimate the relative importance of different factors in the recent rise in labor income inequality: the tail index of the Pareto top labor income distribution depends on a notion of race between education and technology, while the scale parameter depends on communicatio...
The Phillips Curve is a negative relationship between inflation and unemployment but has been rep... more The Phillips Curve is a negative relationship between inflation and unemployment but has been repeatedly challenged: missing deflation in 1933-1941 and 2007-2009, stagflation in the 1970s, missing inflation in the late 1990s and 2010s, contrasting with always strong regional Phillips Curves across states, cities and counties. This paper reconciles these inconsistencies by distinguishing between fixed and flexible exchange rates and hypothesizing a real exchange rate Phillips Curve: a correlation between real exchange rate growth and unemployment. Inflation correlates with unemployment only with fixed exchange rates, when real exchange rate growth equals inflation, explaining Phillips’ original correlation. According to this interpretation, the 1970s stagflation was not due to expectations changing, but to the end of the Bretton-Woods system of fixed exchange rates. The real exchange rate Phillips Curve is consistent with a demand side secular stagnation model based on an excess of s...
Politique étrangère, 2021
Les inquietudes souvent avancees sur la dette publique reposent sur nombre d’idees recues. Le rat... more Les inquietudes souvent avancees sur la dette publique reposent sur nombre d’idees recues. Le ratio de dette sur PIB est une mesure peu signifiante, et la dette publique absorbe l’epargne privee en exces. Le deficit public ne nuit pas au financement de l’investissement prive et n’est pas un fardeau pour les generations futures. Mais la dette publique doit etre geree serieusement, mise au service de projets permettant de satisfaire les besoins et objectifs societaux, des lors que l’epargne privee ne les finance pas.