Behavioral Finance

April 12, 2014
Amit Seru and Amir Sufi, University of Chicago, Organizers

Kenneth Ahern, University of Southern California, and Denis Sosyura, University of Michigan

Rumor Has It: Sensationalism in Financial Media

The media have an incentive to publish sensational news. Ahern and Sosyura study how this incentive affects the accuracy of media coverage in the context of merger rumors. Using a novel dataset they find that accuracy is predicted by a journalist's experience, specialized education, and industry expertise. Conversely, less accurate stories contain ambiguous language and feature well-known firms with broad readership appeal. Investors do not fully account for the predictive power of these characteristics, leading to an initial target price overreaction and a subsequent reversal consistent with limited attention. Overall, the authors provide novel evidence on the determinants of media accuracy and its effect on asset prices.


Asaf Manela, Washington University in St. Louis, and Alan Moreira, Yale University

News Implied Volatility and Disaster Concerns

Manela and Moreira construct a text-based measure of uncertainty starting in 1890 using front page articles of the Wall Street Journal. News implied volatility (NVIX) captures well the disaster concerns of the average investor. NVIX peaks during stock market crashes, times of policy-related uncertainty, world wars, and financial crises. The authors find that periods when people are more concerned about a rare disaster, as proxied by news, are either followed by periods of above average stock returns, or followed by periods of large economic disasters. Concerns related to wars and government policy explain 54 percent and 21 percent of the time variation in risk premia the authors' measure identifies. These findings suggest that time variation in rare disaster risk is an important source of aggregate asset prices fluctuations. The authors provide parameter values of interest to macro-finance, such as the persistence and volatility of the disaster probability process.


Alexander Ljungqvist, New York University and NBER, and Wenlan Qian, National University of Singapore

How Constraining Are Limits to Arbitrage? Evidence from a Recent Financial Innovation

Limits to arbitrage play a central role in both asset pricing and behavioral finance. Without them, asset prices could not deviate from true values for extended periods of time. Ljungqvist and Qian describe an innovative new arbitrage strategy that sidesteps limits to arbitrage. Their evidence shows how even small, price-taking "arbitrageurs" can profitably correct overpricing despite costly information discovery, significant noise-trader risk, and severe short-sale constraints. The innovation that allows the arbitrageurs to sidestep these limits to arbitrage involves credibly revealing their information to the market in an effort to induce long investors to sell so that prices fall. This simple but apparently effective way around the limits suggests that going forward, limits to arbitrage may not be as constraining as traditionally assumed.


Steve Foerster, University of Western Ontario; Juhani Linnainmaa, University of Chicago and NBER; Brian Melzer, Northwestern University; and Alessandro Previtero, University of Western Ontario

The Costs and Benefits of Financial Advice

Foerster, Linnainmaa, Melzer, and Previtero assess the value that financial advisers provide to clients using a unique panel dataset on the Canadian financial advisory industry. They find that advisers influence investors' trading choices, but they do not add value through their investment recommendations when judged relative to passive investment benchmarks. The value-weighted client portfolio lags passive benchmarks by more than 2.5 percent per year net of fees, and even the best-performing advisers fail to produce returns that reliably cover their fees. The authors show that differences in clients' financial knowledge cannot account for the cross-sectional variation in fees, which implies that lack of financial sophistication is not the driving force behind the high fees. However, advisers influence client saving behavior, risky asset holdings, and trading activity, which suggests that benefits related to financial planning may account for investors' willingness to accept high fees on investment advice.

Kelly Shue, University of Chicago, and Richard Townsend, Dartmouth College

Growth through Rigidity: Understanding Recent Trends in Executive Compensation

Shue and Townsend explore a rigidity-based explanation of the dramatic and off-trend growth in U.S. executive compensation during the 1990s and early 2000s. They show that executive option and stock grants are rigid in the number of shares granted. In addition, salary and bonus exhibit downward nominal rigidity. Rigidity implies that the value of executive pay will grow with firm equity returns, which averaged 30 percent annually during the tech boom. Rigidity can also explain the increased dispersion in pay, the difference in growth rates between the United States and other countries, and the increased correlation between pay and firm-specific equity returns. Regulatory changes requiring the disclosure of the value of option grants help explain the moderation in executive pay in the late 2000s. Finally, the authors find suggestive evidence that number-rigidity in executive pay is generated by money illusion and rule-of-thumb decision-making.


Benjamin Keys, University of Chicago, and Jialan Wang, Consumer Financial Protection Bureau

Perverse Nudges: Minimum Payments and Debt Paydown in Consumer Credit Cards

What factors affect how much consumers repay on their credit cards each month? Keys and Wang examine the drivers of payment behavior using the Consumer Financial Protection Bureau (CFPB) credit card database which includes the monthly account activity of a large fraction of U.S. consumers from 2008-12. They find that consumers' payment behavior is consistent and strongly bimodal. Most accounts are either paid in full or paid near the minimum amount each month, with very few intermediate payment amounts. The authors then evaluate the impact of two types of policy changes: 1, changes in the minimum payment formulas implemented by individual issuers; and 2, new payment disclosures mandated by the Credit Card Accountability Responsibility and Disclosure (CARD) Act of 2009. The policy changes led to small increases in the payments made by consumers previously paying the minimum. On average, the CARD Act disclosures increased consumer payments by $19 per month from February 2010 to December 2012. However, both the formula changes and the CARD Act's three-year payment disclosure had the perverse effect of decreasing the fraction of accounts paid in full by 1 percent. These findings are difficult to reconcile with rational economic models and imply that setting suggested payments at low amounts leads some consumers to reduce their overall debt payments. The authors' results suggest that anchoring and the salience of minimum payments play an important role in the credit card market.


Cindy Soo, University of Michigan

Quantifying Animal Spirits: News Media and Sentiment in the Housing Market

Sentiment or "animal spirits" has long been posited as an important determinant of asset prices, but measures of sentiment are particularly difficult to construct for the housing market. Soo develops the first measures of sentiment across local housing markets by quantifying the positive and negative tone of housing news in local newspaper articles. She uses these measures to test the role of sentiment in the run-up and crash of housing prices that instigated the great financial crisis of 2008. She finds that her housing sentiment index forecasts the boom and bust pattern of house prices at a two-year lead, and can predict more than 70 percent of the variation in aggregate house price growth. Consistent with theories of investor sentiment, the author finds that her sentiment index predicts not only price variation but also patterns in trading volume. Estimated effects of sentiment are robust to an extensive list of observed controls including lagged fundamentals, lagged price growth, subprime lending patterns, and news content over typically unobserved variables. To address potential bias from latent fundamentals, the author develops alternate sentiment measures from a subset of weekend and narrative articles that newspapers use to cater to sentiment but which are plausibly exogenous to news on fundamentals. Estimates remain robust, suggesting bias from unobserved fundamentals is minimal.