Numerous gravity applications have resorted to panel data econometric techniques over the past decade. However, with the theory of gravity being so far only static, these estimations lack solid structural dynamic foundations. As a consequence, a consensus on a unified dynamic gravity estimation approach is yet to be reached. In this paper, (i) we build the theoretical foundations for a dynamic gravity model, (ii) we provide guidance for gravity-type estimations with panel data and we consider applications, and (iii) we calibrate and simulate our model to compare its properties with those of the standard, static gravity setup. JEL classification: F10, F11Gravité dynamique : le cas où la taille du pays et l'accumulation d'actifs sont endogènes. De nombreuses applications du modèle de gravité ont eu recoursà des techniqueś econométriques de données de panel au cours de la dernière décennie. Cependant, la théorie de la gravité demeurant statique pour le moment, ces estimations manquent de fondements structurels dynamiques solides. En conséquence, on n'a pas fait consensus sur une approche unifiéeà l'estimation de modèles de gravité dynamique. Dans ce mémoire, (i) on construit les fondements théoriques d'un modèle de gravité dynamique; (ii) on suggère uen façon de procéder pour l'estimation de modèles de gravitéà l'aide de données de panel, et on examine certaines applications; et (iii) on calibre notre modèle et on le soumet a des simulations pour comparer ses propriétés avec celles obtenues au moyen du modèle statique standard.We are grateful to participants in the 2009 EEA meetings for helpful comments and productive feedback. In addition, we thank Werner Antweiler and three anonymous referees whose comments and constructive criticism improved the quality of the paper significantly. We are solely responsible for all errors in the manuscript.

No abstract

No abstract

No abstract

Recent empirical evidence shows that price-cost margins in the market for bank credit are countercyclical in the U.S. economy and that this cyclical behavior can be explained in part from the fact that switching banks is costly for customers (i.e., from a borrower hold-up effect). Our goal, in this paper, is to study the "financial accelerator" role of these countercyclical margins as a propagation mechanism of macroeconomic shocks. To do so, we apply the "deep habits" framework in Ravn, Schmitt-Grohé, and Uribe (2006) to financial markets to model this hold-up effect within a monopolistically competitive banking industry. We are able to reproduce the pattern of pricecost margins observed in the data, and to show that the real effects of aggregate total factor productivity shocks are larger the stronger the friction implied by borrower hold-up. Also, output, investment, and employment all become more volatile than in a standard model with constant margins in credit markets. An empirical contribution of our work is to provide structural estimates of the deep habits parameters for financial markets.JEL codes: E32, E44