Skin in the Game and Moral Hazard

Gilles Chemla

Description

Niveau: Supérieur, Doctorat, Bac+8
Skin in the Game and Moral Hazard Gilles Chemla Imperial College, CNRS Paris-Dauphine, and CEPR. Christopher A. Hennessy London Business School, CEPR, and ECGI. October 2011 Abstract Should skin in the game be regulated in ABS markets? If so, why and how? To address these questions we develop a noisy rational expectations model where originators can exert e?ort ex ante and exploit interim private information in choosing retentions and tranching. In all unreg- ulated market equilibria, asymmetric information reduces di?erences in type-contingent interim payo?s, discouraging e?ort. Further, originators do not internalize bene?ts from signaling in facilitating e¢ cient risk-sharing across investors. E?ort can be induced by mandating high (interim-ine¢ cient) retentions, punishing low types. In the optimal separating regulation in- ducing e?ort, originators choose from a menu of junior tranche retention sizes. In the optimal pooling regulation inducing e?ort, all originators retain a single junior claim, with size depend- ing on price informativeness. The separating (pooling) regulation generally maximizes welfare if e¢ cient risk-sharing (bank investment) is the dominant concern. Mandated opacity is only optimal amongst mechanisms providing zero e?ort incentive. JEL Codes: G32, G28. Corresponding author (Hennessy): Regent?s Park, London, NW1 4SA, U.

• optimal separating


• optimal amongst


• cient risk-sharing


• ante e?ort


• vately optimal


• pooling regulation


• such unregulated equilibria


• interim payo?

Sujets

• Université de Franche-Comté
• University of Zurich
• Pouget
• Atyrau
• Rochet
• Jean Tirole

Informations

• Publié par: profil-zyak-2012
• Nombre de lectures: 15
• Langue: English

Extrait

Skin in the Game and Moral Hazard 

Gilles Chemla Christopher A. Hennessy Imperial College, CNRS Paris-Dauphine, and CEPR. London Business School, CEPR, and ECGI. October 2011

Abstract Should skin in the game be regulated in ABS markets? If so, why and how? To address these questions we develop a noisy rational expectations model where originators can exert e¤ort ex ante and exploit interim private information in choosing retentions and tranching. In all unreg- ulated market equilibria, asymmetric information reduces di¤erences in type-contingent interim payo¤s, discouraging e¤ort. Further, originators do not internalize benets from signaling in facilitating e¢ cient risk-sharing across investors. E¤ort can be induced by mandating high (interim-ine¢ cient) retentions, punishing low types. In the optimal separating regulation in- ducing e¤ort, originators choose from a menu of junior tranche retention sizes. In the optimal pooling regulation inducing e¤ort, all originators retain a single junior claim, with size depend- ing on price informativeness. The separating (pooling) regulation generally maximizes welfare if e¢ cient risk-sharing (bank investment) is the dominant concern. Mandated opacity is only optimal amongst mechanisms providing zero e¤ort incentive. JEL Codes: G32, G28.

 Corresponding author (Hennessy): Regents Park, London, NW1 4SA, U.K.; [e-mail]; 44(0)2070008285. We thank Sudipto Bhattacharya, Sergei Guriev, Michel Habib, Yolande Hiriart, David Martimort, Sebastien Pouget, Jean Charles Rochet, and Jean Tirole for helpful feedback. We also thank seminar participants at the Institut Bachelier, IDC (Israel), London Business School, University of Zurich, Université Paris Dauphine, Université de Franche-Comté (Besançon), Queen Mary, and New Economic School (Moscow).

In the wake of the recent credit crisis, empirical researchers have sifted through the wreckage to locate root causes of sharp declines in the value of various classes of asset-backed securities (ABS). Securitization and the originate-to-distribute (OTD) business model have Þ gured prominently in the list of causal factors. For example, there is now a compelling body of empirical evidence establishing a positive relationship between securitization rates and mortgage defaults (see e.g. Mian and Su Þ (2009), Keys et al. (2010) and Keys, Seru and Vig (2011)). Such empirical evidence demonstrating a link between moral hazard and securitization raises a natural question that has not been addressed: What, if anything, is to be done? Implicit in recent legislation in the United States is the view that government intervention in ABS markets will increase social welfare. In particular, the Dodd-Frank Act, recently codi Þ ed as Section 15G of the Securities Exchange Act, charges six federal agencies (the Federal Reserve, Treasury, FDIC, SEC, FHA, and HUD) with setting mandatory retention standards for ABS securitizers. Unfortunately, the formulation of optimal regulation of ABS is hindered by the absence of a coherent theoretical framework allowing one to answer some fundamental questions. First, what are the market failures and can a regulator possibly improve upon unregulated market outcomes? Second, how do mandated retentions a ff ect determinants of social welfare such as e ff ort incentives, investment levels and risk-sharing? Third, given the trade-o ff s what are the policy options and conditions under which each dominates? This paper develops a tractable mechanism and security design framework to address these questions. Although the primary focus is ABS markets, the setting and issues are more general: An agent (originator below) is considering exerting costly hidden e ff ort ex ante, anticipating sub- sequent marketing of claims to investors who will not know the assets true type while the agent will know its type. Securitization proceeds fund a scalable investment with positive NPV. Ex post, ABS are purchased in competitive markets by an endogenously informed speculator and rational uninformed investors with hedging motives. We make three important departures from the canon- ical setting considered by Myers and Majluf (1984). First, we follow DeMarzo and Du ffi e (1999)

2

in considering optimal securities written on assets-in-place. Second, speculator orders drive prices closer to fundamentals in the event of interim-stage pooling, with endogenous trading in noisy ra- tional expectations equilibria. Finally, and most importantly, there is an ex ante stage in which the originator can increase the probability of becoming a high type by exerting hidden costly e ff ort. In order to understand the merits, if any, of mandating retentions, we Þ rst describe potential equilibria in unregulated ABS markets. One possible equilibrium is that a low type securitizes the entire asset while a high type signals by retaining the minimal junior tranche needed to deter mimicry. This is the familiar least-cost separating equilibrium (LCSE). Ex post, the LCSE has the attractive feature that e ffi cient risk-sharing is achieved across investors since they are symmetrically informed. However, in terms of ex ante e ff ort incentives the LCSE is problematic since a low type receives his Þ rst-best interim payo ff while a high type receives less than his Þ rst-best. This diminishes the originators willingness to exert e ff ort ex ante with the hope of becoming a high type. If the NPV of the new investment is very high, the set of interim equilibria in unregulated markets also includes the low types preferred pooling equilibrium where both types sell the entire asset in an opaque form that prevents information acquisition by the speculator, resulting in prices being uncorrelated with true value. This equilibrium is attractive from an ex post perspective since there is socially e ffi cient risk-sharing given the absence of informational asymmetries across investors. However, this equilibrium destroys ex ante e ff ort incentives since the two types get the same interim payo ff . Under less stringent conditions on NPV than those discussed in the preceding paragraph, the set of unregulated market equilibria also includes the high types preferred pooling equilibrium featuring full securitization (OTD below) cum tranched securities designed to stimulate informed trading. Here the high type sells a safe senior claim and a mezzanine tranche tailored to the preferences of uninformed hedging investors. This increases uninformed buying volume and raises the gain to informed speculation, driving prices closer to fundamentals. Ex post, risk-sharing is imperfect since uninformed investors reduce purchases of mezzanine debt when facing informed trading. In

3

this pooling equilibrium with zero retentions, market prices provide discipline and ex ante e ff ort incentives can actually be stronger than in the LCSE featuring voluntary retentions. Moreover, tranching of ABS increases price informativeness, strengthening e ff ort incentives. Thus, we argue the root cause of moral hazard is not OTD or tranching, but OTD combined with uninformative prices. Our baseline analysis reveals two valid welfare arguments for mandated retentions. First, pri- vately optimal retentions can be socially suboptimal since originators do not internalize the bene Þ t of improved investor risk-sharing resulting from signaling. Second, the equilibrium payo ff to ex ante e ff ort in unregulated markets may be insu ffi cient to induce e ff ort. Curbing moral hazard requires punishing originators who produce low value assets. But if retentions are not mandated, a low type can always achieve his Þ rst-best payo ff , if not more, by admitting he is a low type and proceeding to securitize the entire asset. The formal basis for the preceding two arguments favoring regulation is as follows. The equi- librium set at the interim-stage continuation game consists of all allocations Pareto dominating the LCSE from the perspective of originators. 1 The Þ rst problem with such unregulated equilibria from a social perspective is that Pareto optimality is only evaluated from the originator perspective. Thus, investor-level bene Þ ts of signaling are ignored. Second, a low type always receives at least his Þ rst-best interim payo ff , in opposition to the punishment needed to encourage e ff ort ex ante. A socially optimal mandatory retention scheme increases e ff ort incentives by increasing the spread between interim payo ff s to high and low types, while accounting for costs imposed on investors as well as originators. There are two kinds of skin-in-the-game regulatory schemes featuring di ff ering bene Þ ts and costs: separating schemes inducing originators to reveal the true asset type and pooling schemes that do not. In an optimal separating scheme, originators are forced to choose from a menu of strictly positive junior tranche retentions. The optimality of using junior tranche retention size as the basis for sorting is a consequence of single-crossing, as distinct from a standard moral hazard argument calling 1 This is an application of a result of Maskin and Tirole (1992).

4