Content

Published and working papers

 

 

 

Working papers

  • The colour of finance words (joint with Xiaowen Hu and Maximilian Rohrer)

    We study a standard machine learning algorithm (Taddy, 2013) to measure sentiment in financial documents. Our empirical approach relies on stock price reactions to colour words, providing as output dictionaries of positive and negative words. In head-to-head comparisons, our dictionaries outperform the standard bag-of-words approach (Loughran and McDonald, 2011) when predicting stock price movements out-of-sample. By comparing their composition, word-by-word, our method refines and expands the sentiment dictionaries in the literature. The breadth of our dictionaries and their ability to disambiguate words using bigrams both help to colour finance discourse better.

  • Friends during Hard Times: Evidence from the Great Depression (joint with Tania Babina and Geoff Tate)

    We test whether network connections to other firms through executives and directors increase value by exploiting differences in survival rates in response to a common negative shock. We analyze a novel dataset of over 3500 public and private firms from 1928. We find that firms that had more connections on the eve of the 1929 financial market crash have higher 10-year survival rates during the Great Depression. Consistent with a financing channel, we find that the results are particularly strong for small firms, private firms, cash-poor firms, and firms located in counties with high bank suspension rates during the crisis. Moreover, connections to cash-rich firms are stronger predictors of survival, overall and among financially constrained firms. Because of the greater segmentation of markets in the 1920s and 1930s than in modern data samples, we can mitigate the potential endogeneity of network connections at the time of the shock by exploiting variation in the local demand for directors’ services. We also find evidence that the information that flows through network links increases the odds that a firm will be acquired.

  • The kinks of financial journalism

    This paper studies the content of financial news as a function of past market returns. As a proxy for media content we use positive and negative word counts from general financial news columns from the Wall Street Journal and the New York Times. Our empirical analysis allows us to discriminate between theories that predict hyping good stock performance to those that emphasize negative news. The evidence is conclusive: negative market returns taint the ink of typewriters, while positive returns barely do. Given how pervasive our estimates are across multiple time periods, subject to dierent competitive pressures in the market for news, we interpret our results as driven by demand preferences of investors.

Published papers

  • Informed trading, indexing, and welfare (joint with Philip Bond), forthcoming in the Review of Financial Studies.

    We develop a benchmark model to study the equilibrium consequences of indexing in a standard rational expectations setting. Individuals incur costs to participate in financial markets, and these costs are lower for individuals who restrict themselves to indexing. A decline in indexing costs directly increases the prevalence of index- ing, thereby reducing the price efficiency of the index and augmenting relative price efficiency. In equilibrium, these changes in price efficiency in turn further increase indexing, and raise the welfare of uninformed traders. For well-informed traders, the share of trading gains stemming from market timing increases relative to stock selection trades.

  • Asymmetric information, security design, and the pecking (dis)order (joint with Paolo Fulghieri and Dirk Hackbarth), Review of Finance, 2020, 24(5), 961-996.

    We study a security design problem under asymmetric information, in the spirit of Myers and Majluf (1984). We introduce a new condition on the right tail of the firm-value distribution that determines the optimality of debt versus equity-like securities. When asymmetric information has a small impact on the right-tail, risky debt is preferred for low capital needs, but convertible debt is optimal for larger capital needs. In addition, we show that warrants are the optimal financing instruments when the firm has already pre-existing debt in its capital structure. Finally, we provide conditions that generate reversals of the standard pecking order.

  • Optimal contracts with privately informed agents and active principals Journal of Corporate Finance 2014, 29, 695-709.

    This paper considers an optimal contracting problem between an informed risk-averse agent and a principal, when the agent needs to perform multiple tasks, and the principal is active, namely she can influence some aspect of the agency relationship on top of the contract itself (i.e. capital budgets, task assignments). The paper shows how asymmetric information makes incentives and investment decisions substitutes for the principal. This result yields novel implications for contracting models with moral hazard and asymmetric information, i.e., capital budgeting or external capital raising games.

  • Noise and aggregation of information in large markets (joint with Branko Urosevic), Journal of Financial Markets, 2013, 16(3), 526-549.

    We study a novel class of noisy rational expectations equilibria in markets with large number of agents. We show that, as long as noise (liquidity traders, endowment shocks) increases with the number of agents in the economy, the limiting competitive equilibrium is well-defined and leads to non-trivial information acquisition, and partially revealing prices, even if per-capita noise tends to zero. We find that in such equilibrium risk sharing and price revelation play different roles than in the standard limiting economy in which per-capita noise is not negligible. We apply our model to study information sales by a monopolist, information acquisition in multi-asset markets, and derivatives trading, and show that our model leads to qualitatively different results with respect to those in the existing literature. Our notion of large noise is shown to be necessary and sufficient to have limiting economies with perfectly competitive behavior.

  • Sentiment during recessions, Journal of Finance, 2013, 68(3), 1267-1300.

    This paper studies the effect of sentiment on asset prices during the 20th century (1905-2005). As a proxy for sentiment, we use the fraction of positive and negative words in two columns of financial news from the New York Times. The main contribution of the paper is to show that, controlling for other well-known time-series patterns, the predictability of stock returns using news' content is concentrated in recessions. A one standard deviation shock to our news measure during recessions changes the conditional average return on the DJIA by twelve basis points over one day.

    Technical appendix accompanying the paper - gives more details on the data set and studies different Econometric specifications.

  • Geographic dispersion and stock returns (joint with Oyvind Norli), Journal of Financial Economics, 2012, 106(3), 547-565..

    This paper shows that stocks of truly local firms have returns that exceed the return on stocks of geographically dispersed firms by 70 basis points per month. By extracting state name counts from annual reports filed with the SEC on form 10-K, we distinguish firms with business operations in only a few states from firms with operations in multiple states. Our findings are consistent with the view that lower investor recognition for local firms results in higher stock returns to compensate investors for insufficient diversification.

  • Journalists and the Stock Market (joint with Casey Dougal, Joseph Engelberg and Christopher Parsons), Review of Financial Studies, 2012, 25(4), 639-679.

    We use exogenous scheduling of Wall Street Journal columnists to identify a causal relation between financial reporting and stock market performance. To measure the media's unconditional effect, we add columnist fixed effects to a daily regression of excess Dow Jones Industrial Average returns. Relative to standard control variables, these fixed effects increase the R-squared by about 35 percent, indicating each columnist's average persistent "bullishness" or "bearishness." To measure the media's conditional effect, we interact columnist fixed effects with lagged returns. This increases explanatory power by yet another one-third, and identifies amplification or attenuation of prevailing sentiment as a tool used by financial journalists.

  • Crawling EDGAR (joint with Oyvind Norli), The Spanish Review of Financial Economics, 2012, 10(1), 1-10.

    While the title may lead you to think that this paper is about spiders, it is about firms in the United States reporting relevant business information to the Securities and Exchange Commission (SEC). The paper is meant to serve as a primer for economists in the computing details of searching for information on the Internet. One important goal of the paper is to show how simple open-source computer scripts can be generated to access financial data on firms that interact with regulators in the United States.

  • Information sales and strategic trading (joint with Francesco Sangiorgi), Review of Financial Studies, 2011, 24(9), 3069-3104.

    We study information sales in financial markets with strategic risk-averse traders. Our main result establishes that the optimal selling mechanism is one of the following two: (i) sell to as many agents as possible very imprecise information; (ii) sell to a single agent a signal as precise as possible. As noise trading per unit of risk-tolerance becomes large, the "newsletters" or "rumors" associated with (i) dominate the "exclusivity" contract in (ii). The optimal information sales contracts share similar properties in market-orders and limit-orders markets, while models in which competitive behavior is assumed yield qualitatively different equilibria. The endogeneity of the information allocation implies a ranking reversal of the informational efficiency of prices across markets and models. Equilibrium prices become more informative in market-orders than in limit-orders markets, and the model with imperfect competition yields more informative prices than its competitive counterpart. These results are driven by the seller of information offering more precise signals when the externality in the valuation of information is relatively less intense.

    Technical appendix accompanying the paper - gives more details on the derivation of the equilibria (standard and tedious, but necessary).

  • Relative wealth concerns and complementarities in information acquisition (joint with Gunter Strobl), Review of Financial Studies, 2011, 24(1), 169-207.

    This paper studies how relative consumption effects, in which a person's satisfaction with their own consumption depends on how much others are consuming, affect investors' incentives to acquire information. We find that such consumption externalities can generate complementarities in information acquisition within the standard rational expectations paradigm. When agents are sensitive to the wealth of others, they herd on the same information, trying to mimic each other's trading strategies. We show that there can be multiple herding equilibria in which some assets receive considerable attention while others with similar characteristics are ignored. Further, different communities of agents may specialize in different assets. This multiplicity of equilibria also generates jumps in asset prices: an infinitesimal shift in fundamentals can lead to a discrete price movement.

  • Information acquisition and mutual funds (joint with Joel Vanden), Journal of Economic Theory, 2009, 144(5), 1965-1995.

    We study the size and the existence of the mutual fund industry by generalizing the standard competitive noisy rational expectations framework with endogenous information acquisition. Since informed agents optimally choose to open mutual funds in order to sell their private information, mutual funds are an endogenous feature of our equilibrium. Our model yields novel predictions on price informativeness, optimal fund fees, the equilibrium risk premium, and the size and competitiveness of the mutual fund industry. In particular, we show that a sufficiently competitive mutual fund sector yields more informative prices and a lower equity risk premium. Thus, the paper explicitly links the existence of mutual funds to equilibrium asset prices.

    Supplement accompanying the paper - gives details on the proofs of the paper.

  • Sports sentiment and stock returns (joint with Alex Edmans and Oyvind Norli) , Journal of Finance, 2007, 62(4), 1967-1998.

    This paper investigates the stock market reaction to sudden changes in investor mood. Motivated by psychological evidence of a strong link between soccer outcomes and mood, we use international soccer results as our primary mood variable. We find a significant market decline after soccer losses. For example, a loss in the World Cup elimination stage leads to a next-day abnormal stock return of -38 basis points. This loss effect is stronger in small stocks and in more important games, and is robust to methodological changes. We also document a loss effect after international cricket, rugby, and basketball games.

  • Overconfidence and market efficiency with heterogeneous agents (joint with Francesco Sangiorgi and Branko Urosevic), Economic Theory, 2007, 30(2), 313-336.

    We study financial markets in which both rational and overconfident agents coexist and make endogenous information acquisition decisions. We demonstrate the following irrelevance result: when a positive fraction of rational agents (endogenously) decides to become informed in equilibrium, prices are set as if all investors were rational, and as a consequence the overconfidence bias does not affect informational efficiency, price volatility, rational traders' expected profits or their welfare. Intuitively, as overconfidence goes up, so does price informativeness, which makes rational agents cut their information acquisition activities, effectively undoing the standard effect of more aggressive trading by the overconfident. The main intuition of the paper, if not the irrelevance result, is shown to be robust to different model specifications.

  • Monotonicity in direct revelation mechanisms, Economics Letters, 2005, 88(1), 21-26.

    This paper studies a standard screening problem where the principal's allocation rule is multi-dimensional, and the agent's private information is a one-dimensional continuous variable. Under standard assumptions, that guarantee monotonicity of the allocation rule in one-dimensional mechanisms, it is shown that the optimal allocation will be non-monotonic in a (weakly) generic sense once the principal can use all screening variables. The paper further gives conditions on the model's parameters that guarantee that non-monotonic allocation rules will be optimal.

  • Convergence and biases of Monte Carlo estimates of American option prices using a parametric exercise rule, Journal of Economic Dynamics & Control, 2003, 27(10), 1855- 1879.

    This paper presents an algorithm for pricing American options using Monte Carlo simulation. The method is based on using a parametric representation of the exercise boundary. It is shown that, as long as this parametric representation subsumes all relevant stopping times, error bounds can be constructed using two different estimates, one which is biased low and one which is biased high. Both are consistent and asymptotically unbiased estimators of the true option value. Results for high-dimensional American options confirm the viability of the numerical procedure. The convergence results of the paper shed light into the biases present in other algorithms proposed in the literature.