``Aggregate Representations of Aggregate Games," with Lars Stole.
Abstract. An aggregate game is a normal-form game with the property that each player's payoff is a function only of his own strategy and an aggregate function of the strategy profile of all players. Aggregate games possess a set of purely algebraic properties that can often provide simple characterizations of equilibrium aggregates without first requiring that one solves for the equilibrium strategy profile. We demonstrate that the defining nature of payoffs in an aggregate game allows one to embed the strategic analysis into the aggregate-strategy space, converting an n-player game to a simpler object -- a self-generating single-person maximization program. We apply these techniques to a number of economic settings including competition in supply functions or competitive agency games with nonlinear transfer functions.
``Delegation and R&D Spending: Evidence from Italy" with Jakub Kastl and Salvatore Piccolo (This version September 2010). (2)
``Delegation and R&D Spending: Evidence from Italy" with Jakub Kastl and Salvatore Piccolo (This version September 2010).
Abstract.Delegation of decision-making has important implications for firms' performance. We build a simple model predicting that decentralized firms are more involved in R&D, and use data from the Italian manufacturing industry to document a positive correlation between delegation and R&D spending. This result is robust to controlling for the determinants of R&D such as human capital, capital intensity, and sectoral or regional effects. To unveil the causal effect, we investigate the determinants of delegation in our sample in a search for potential instruments. Variables found as important determinants of delegation in Acemoglu et al. (2007) in the context of French and British industries do not appear significant in our sample. Instead, we use ownership concentration as an instrument for delegation and estimate that delegation increases R&D spending by about 5% per year.
``The Public Management of Risk: Separating Ex Ante and Ex Post Monitors'', with Yolande Hiriart and Jérôme Pouyet, Journal of Public Economics.
Abstract: When a firm undertakes risky activities, the conflict between social and private incentives to implement safety care requires public intervention which can take the form of both monetary incentives but also ex ante or ex post monitoring, i.e., before or after an accident occurs. We delineate the optimal scope of monitoring depending on whether public monitors are benevolent or corruptible. We show that separating the ex ante and the ex post monitors increases the likelihood of ex post investigation, helps prevent capture and improves welfare.
The Cost of Contract Renegotiation: Evidence from the Local Public Sector" with Philippe Gagnepain and Marc Ivaldi (This version September 2010) (2)
``The Cost of Contract Renegotiation: Evidence from the Local Public Sector" with Philippe Gagnepain and Marc Ivaldi (This version September 2010).
Abstract.The renegotiation of regulatory contracts is known to prevent regulators from achieving the full commitment efficient outcome in dynamic contexts. However, assessing the cost of such renegotiation remains an open issue from an empirical viewpoint. To address this question, we fit a structural principal-agent model with renegotiation on a set of urban transport service contracts. The model captures two important features of the industry. First, only two types of contracts are used in practice (fixed-price and cost-plus). Second, subsidies increase over time. We compare a scenario with renegotiation and a hypothetical situation with full commitment. We conclude that the welfare gains from improving commitment would be significant but would accrue mostly to operators.
``Mechanism Design with Bilateral Contracting"with Vianney Dequiedt. (This version August 2010.) (2)
``Mechanism Design with Bilateral Contracting"with Vianney Dequiedt. (This version August 2010.)
Abstract. We consider an organization run by bilateral contracts in a private-values context
where agents have private information on their cost parameters. The principal acts opportunistically and chooses the agents' outputs once they have already reported their types. A Revelation Principle with bilateral contracting characterizes the set of implementable allocations by means of simple non-manipulability constraints. Introducing those constraints simplifies contracts and restores a rent/efficiency trade-off even with correlated information.