Working Papers
[Link to my SSRN research page] [Link to my research keywords]
JOB MARKET PAPER: Information Percolation Driving Volatility
First version: September 2011. This version: December 2011.
[download pdf] [BibTeX citation]
Sudden big price changes are followed by periods of high and persistent volatility. I develop a tractable dynamic rational expectations model consistent with this observation. An infinity of agents possess dispersed information about future dividends and trade in centralized markets. Information is processed, transmitted, and aggregated in two ways: (i) agents meet randomly and exchange information through word-of-mouth communication, and (ii) the price aggregates information through the trading process. Both mechanisms operate simultaneously to generate high and persistent volatility. The resulting information flow drives both returns and volume. The short-term asset, defined as the claim to immediate future dividends, becomes more volatile. The pronounced heterogeneity in investors’ information endowments induces patterns of trade consistent with empirical findings. These results serve as a road sign indicating the central role played by word-of-mouth communication in financial markets.
Conference & Seminar Presentations:
Princeton-Lausanne Workshop in Quantitative Finance, Lausanne, Switzerland, May 2011; Brown Bag Seminar at HEC, Lausanne, Switzerland, April 2011; "Les Doctoriales de l'UNIL", Lausanne, Switzerland, March 2011; "Asset Pricing Workhop", Lausanne, September 2011; Geneva Finance Research Institute, November 2011; Department of Banking and Finance at the University of Zurich, November 2011; Finance Research Seminar at the University of Lausanne, December 2011
Investors' Attention and Stock Market Volatility (with Michael Hasler)
First version: December 2010. This version: November 2011.
[download pdf] [slides] [BibTeX citation]
Days with big stock price changes are hardly related to single important economic news. Two potential explanations exist. The first argues that price movements are due to the confluence of multiple small pieces of information. The second argues that the attention of market participants is time varying and may produce large price fluctuations without apparent large movements in fundamentals. We integrate these two arguments in a general equilibrium model. We build a continuous-time pure exchange economy with fluctuating attention. The model features excess and countercylical volatility, driven by investor's attention. The model predicts a quadratic relationship between attention and volatility. We confirm empirically this relationship. Finally, most of the excess volatility is concentrated in the short-term asset, defined as the claim to recent future dividends.
Conference & Seminar Presentations:
Gerzensee Swiss Doctoral Workshop in Finance, June 2011*; Mathematical Finance Days, Montréal, Canada, May 2011; Princeton-Lausanne Workshop in Quantitative Finance, Lausanne, Switzerland, May 2011*
Volatility Clustering with Learning and Model Heterogeneity (with Michael Hasler)
First version: May 2010. This version: August 2011.
[download pdf] [slides] [poster] [BibTeX citation]
We consider a standard Lucas economy with a single consumption tree and two agents. Agents do not observe the drift of the dividend stream, which is assumed to follow a persistent process. Consequently, agents have to filter it and update their beliefs under the main assumption that they have different parameter estimates for the process of the expected growth rate: different long term means, mean reversion speeds and volatilities. We explain both the level of the stock market volatility and its dynamics in close connection with the state of the economy. We principally show that the stock market volatility is significantly higher than the consumption volatility and strongly countercyclical. Stock market volatility is high when the expected dividend growth rate is low and vice versa. We highlight the endogenous mechanism that generates our results, namely persistent difference of beliefs. Finally, using Malliavin calculus we show that the clustering effect comes exclusively from the difference of opinion process.
Conference & Seminar Presentations:
The Financial Risks International Forum on "Long Term Risks", Paris, France, March 2011; Gerzensee Swiss Doctoral Workshop in Finance, June 2010
Information Percolation in Centralized Markets (with Julien Cujean)
First version: September 2010. This version: January 2011.
[download pdf] [slides] [BibTeX citation]
We explore the effects of information propagation in a centralized financial market. Specifically, we embed search frictions within the Grossman and Stiglitz (1980) framework, relying on information percolation as modeled in Duffie, Malamud, and Manso (2009). First, we show that information percolation produces a positive autocorrelation in stock returns. Second, we introduce a rumor among the population of investors which we let to propagate during two periods. As trading approaches its final round and that the fundamental value is close to be revealed, we let the rumor die out and show that it produces a strong price reversal toward the fundamental value. Importantly, information percolation introduces heterogeneity in individual precision among agents. This leads to drastically different investment strategies: agents who have been more efficient at gathering signals will tend to act as market makers. Whereas, agents who collected a lesser amount of signals will tend to be trend followers.
Conference & Seminar Presentations:
Brown Bag Seminar at Kellogg School of Management, Evanston, USA, December 2010; Seminar at MIT Sloan, Boston, USA, March 2011*; Seminar at Boston University, Boston, USA, March 2011*; Princeton-Lausanne Workshop in Quantitative Finance, Lausanne, Switzerland, May 2011*
International Portfolio Choice and Relative Wealth Concerns
First version: May 2010. This version: July 2010.
[download pdf] [slides] [BibTeX citation]
In a standard information based model à la Admati [1985] with private information, it is shown how a plausible social interaction between investors, namely the relative wealth concerns, might amplify an almost insignificant informational advantage and produce sizable home bias. The model considers a quantitative informational advantage, which is different of what has been postulated so far in the literature. The solution is found in closed form and is in line with the economic intuition. The home bias results are analyzed not only at country level, but as well at investor’s level. It is shown that there is a cross-sectional variation within countries in the level of home bias.
Conference & Seminar Presentations:
Gerzensee Swiss Doctoral Workshop in Finance, June 2010
Global Public Signals, Heterogeneous Beliefs and Stock Markets Comovement (with Julien Cujean)
First version: May 2009. This Version: February 2010
[download pdf] [slides] [BibTeX citation]
We build an information-based two-country general equilibrium model. There are two dividend processes with correlated growth rates. Agents observe a global public signal informative about both growth rates. We first let agents rationally process information, and then we allow for reasonable departures from rationality. That is, agents are overconfident with respect to the signal, and thus have heterogeneous beliefs. We report a significant increase in comovement between stock returns. Moreover, we find that a small amount of misinterpretation of the global signal is sufficient to generate sizable comovement, as compared to the benchmark case of rational expectations. As an additional implication of overconfidence, we show that our model is able to produce a substantial home equity bias.
Conference & Seminar Presentations:
Gerzensee Swiss Doctoral Workshop in Finance, June 2009; The Australasian Finance & Banking Conference, Sydney, Australia, December 2009
Trade Costs, Heterogeneous Firms and International Portfolio Choice
First version: May 2008.
[download pdf] [slides] [BibTeX citation]
This paper builds on a two-country dynamic stochastic general equilibrium (DSGE) model in which households can invest in home and foreign equities. First, a benchmark model is created to obtain several previous results of the existing literature. A closed form solution is derived for the optimal portfolio holdings. In this benchmark model it is shown why the hedging of both real exchange rate risk and labor income risk will counterfactually generate foreign equity bias instead of home equity bias. Then, new features are introduced, such as non-traded goods, heterogeneous firms and fixed export costs. The interactions between the fixed export costs, the non-tradadable sector, and the firm heterogeneity helps to explain the home equity bias documented in many empirical studies.
Conference & Seminar Presentations:
Gerzensee Swiss Doctoral Workshop in Finance, June 2008
(* indicates presentation by co-author)