Sophisticated Intermediation and Aggregate Volatility (Job Market Paper)
I consider an economy where investors delegate their investment decisions to financial institutions that choose across multiple investment opportunities featuring different levels of idiosyncratic risk and different degrees of correlation with the aggregate of the economy. Investors solve an optimal contracting problem to induce financial institutions to allocate their investment optimally. I then study how investment decisions are affected when financial securities are introduced that allow agents to trade their risks. Investors do not have the necessary information to understand these securities, but give incentives to financial institutions to hedge certain risks. I show that hedging idiosyncratic risks ameliorates the agency problem between investors and financial institutions and reduces aggregate volatility. On the contrary, when aggregate risk can be hedged the agency problem worsens and aggregate volatility increases. Finally, I study the efficiency properties of the equilibrium and the potential role for financial regulation.
Cycles, Gaps, and the Social Value of Information
(with George-Marios Angeletos and Jennifer La’O) - revise and resubmit, American Economic Review
What are the welfare effects of the information contained in macroeconomic statistics, central-bank communications, or news in the media? We address this question in a business-cycle framework that nests the neoclassical core of modern DSGE models. Earlier lessons that were based on "beauty contests" (Morris and Shin, 2002) are found to be inapplicable. Instead, the social value of information is shown to hinge on essentially the same conditions as the optimality of output stabilization policies. More precise information is unambiguously welfare-improving as long as the business cycle is driven primarily by technology and preference shocks---but can be detrimental when shocks to markups and wedges cause sufficient volatility in "output gaps". A numerical exploration suggests that the first scenario is more plausible.
Liquidity Insurance with Market Information
This paper studies how market signals---such as stock prices---can help alleviate the severity of the asymmetric information problem in credit and liquidity management. Asymmetric information hinders the ability of borrowers (firms, investment banks, etc.) to undertake profitable investment opportunities and to insure themselves against liquidity shocks. I show that on the equilibrium path creditors do not learn anything from market signals because they can use a menu of contracts to screen the different types of borrowers. However, by conditioning liquidity insurance on ex post price signals, creditors are able to provide the borrowers with better incentives for truth-telling. At the same time, prices depend on the liquidity the creditors offer to the borrowers. This two-way feedback impacts the design of the optimal contract and potentially generates multiple equilibria in financial markets.
Social Insurance with Persistent Types
(with Mikhail Golosov)
In this paper we study the problem of optimal social insurance when agents are privately informed and the government cannot commit. Contrary to the previous literature, we assume that agents' types are persistent. When the government lacks commitment and types are persistent, it is not optimal for the agents to fully reveal their information to the government. We characterize the worst equilibrium for this economy and use it sustain the best equilibrium in which the government never fully learns the types of the agents.
Financial Advice and the Costs of Information Acquisition
(with Marco Di Maggio)
In the market for financial services, investors heavily rely on advisers' or sellers' information to make their investing decisions. This is especially true for off-exchange and over-the-counter markets, where all the transactions take place bilaterally and no public price aggregates information. We focus on three sources of distortions that these markets typically display. First, the adviser acquires information about the financial products that better match the investors' characteristics. Second, the adviser's ability and the investor's characteristics are private information. Finally, the adviser's outside option is affected by the information acquired, which generates endogenous adverse selection. We characterize the optimal contract which features distortions in the information acquisition and trading decisions. The paper also contributes to the debate on licensing access to markets for complex products by showing that it can be optimal to prevent investors from participating when information acquisition generates adverse selection.
Asymmetric Information and Overexposure to Risk in the CDS Market
(with Stefano Giglio)