Investor Learning Research Articles

What Do Investors Learn from Advertisements?

Individual investors use advertisements to make investment decisions. We test how advertising content affects investor’s knowledgeability and evaluation of an equity offering. The results show that persuasive content (e.g. images) increases investor knowledgeability about fundamental characteristics of securities offerings while salient disclosure of risk factors only increases risk factor knowledge. Persuasive content also increases average investment amounts by 16 percent, while salient risk factor disclosure reduces the tendency to consult additional information. Our results are robust against differences in financial literacy and investor experience and provide insights for regulating investment marketing.

Reporting System, Investor Learning, and Stock Return Regularities

We propose a simple framework for understanding accounting-based stock return regularities. A firm's accounting reports provide noisy information about hidden economic states that evolve according to a Markov process. In response to the accounting reports, a representative Bayesian investor forms beliefs about the underlying state and hence the value of the firm. For a population of such firms, the model provides predictions consistent with two sets of well-documented regularities: (i) the market reaction to an earnings announcement that ends a string of consecutive earnings increases and (ii) the return predictabilities based on accruals and book-tax differences. The model also yields novel cross-sectional predictions about the distinct roles of economic persistence and earnings informativeness. We confirm these predictions through empirical tests.

Jurisdiction Ratione Temporis in Successive International Investment Agreements: What Can Chinese Investors Learn from the Ping An Case?

Jurisdiction Ratione Temporis in Successive International Investment Agreements: What Can Chinese Investors Learn from the Ping An Case?

Disclosure Dynamics and Investor Learning

This paper examines whether investor learning about profitability (i.e., the mean of earnings distribution) leads to persistence in disclosure decisions. A repeated single-period model shows that persistent investor beliefs about profitability lead to persistent disclosure decisions. Using earnings forecast data, I structurally estimate the model and perform several counterfactual analyses. I find that, when investors are assumed to know profitability, the persistence of management forecast decisions significantly declines by 17% to 27%, and 24% of firms would have disclosed differently, resulting in 3.9% net change in the amount of information (i.e., posterior variance) provided to the capital market. Collectively, the results indicate the importance of learning profitability in understanding disclosure decisions and the capital market consequences of disclosures.

Information and investment under uncertainty

In a model of mutual fund investor learning about managerial ability, we investigate how investment decisions depend on both past returns and non-return information. We show that non-return information affects investors’ reliance on past returns as well as their decision to participate. Our results have important consequences for the relationship between flows of money and past fund performance.

Information and Investment Under Uncertainty

In a model of mutual fund investor learning about managerial ability, we investigate how investment decisions depend on both past returns and non-return information. We show that non-return information affects investors' reliance on past returns as well as their decision to participate. Our results have important consequences for the relationship between flows of money and past fund performance.

Long-Term Rewarded Equity Factors: What Can Investors Learn from Academic Research?

This article analyzes what academic research has to say on equity factors. The objective is to understand what lessons can be learned from such research on designing and evaluating factor indexes. When analyzing academic publications on equity factor investing, five important lessons emerge, which provide useful perspective on practical questions about factor indexes. This article looks at the empirical analysis that is required to identify rewarded factors. It then turns to the economic rationale behind factors and looks into the role of diversification for a given factor tilt. Moreover, it discusses the issue of implementation costs and addresses the question of crowding risks. Finally, the article discusses how popular practical implementations relate to the academic grounding.

Tests of investor learning models using earnings innovations and implied volatilities

This paper investigates alternative models of learning to explain changes in uncertainty surrounding earnings innovations. As a proxy for investor uncertainty, we use model-free implied volatilities; as a proxy for earnings innovations, representing signals of firm performance likely to drive investor perceptions of uncertainty, we use quarterly unexpected earnings benchmarked to the consensus forecast. We document that uncertainty declines on average after the release of quarterly earnings announcements and this decline is attenuated by the magnitude of the earnings innovation. This latter result is consistent with models that incorporate signal magnitude as a factor driving changes in uncertainty. Most important, we document that signals deviating sufficiently from expectations lead to net increases in uncertainty. Critically, this result suggests that models allowing for posterior variance to be greater than prior variance even after signal revelation [e.g., regime shifts in Pastor and Veronesi (Annu Rev Financ Econ 1:361–381, 2009)] better describe how investors incorporate new information.

The Almost Extinction of Stock Dividends

We argue that investors could have wrong initial beliefs about the information content of stock dividends. They gradually update their beliefs after learning from firm performance following stock dividend distributions. In support of the learning hypothesis, we document several new findings. We show that operating performance significantly declines after stock dividend distributions. Investors overreact to a firm's early stock dividend announcements and drastically lower the announcement return as the firm announces more and more follow-on stock dividends. Investor learning also reduces demands for stock dividends. The fraction of stock dividend payers sharply decreases from 14.0% in 1950s to merely 0.2% in 2010s. Institutional investor ownership is able to explain 92.5% of the temporal variation in the fraction of stock dividend payers.

A learning-augmented approach to pricing risk in South Africa

Through application of state-space modelling, the asset pricing model is re-explored. The result is an asset pricing model which tracks the evolution of investor probability beliefs and learning through a Kalman filter. This behaviourally inspired model shows marked improvement over a traditional asset pricing model, with pricing errors being reduced by as much as 41 % over a 16 year period using South African equities data. We find that investors tend to price long-run risk whilst being notably influenced by exposure to lagged market performance. Together, these findings lend support to the hypothesis that investors tend to price risk as a dynamic learning process in an emerging market.

Show Us Your Shorts?

How does greater public disclosure of arbitrage activity and informed trading affect informational efficiency? To answer this, we exploit rule amendments in U.S. securities markets, which increased the frequency of public disclosure of short positions. Higher public disclosure can potentially improve or deteriorate informational efficiency. We find that with more frequent disclosure, short-sellers' information is incorporated into prices faster, improving informational efficiency. In support of the mechanism driving this result, we document significant market reactions to short interest announcements, suggesting investor learning, and furthermore, we find increases in short-selling activity and reductions in short-sellers' holding periods with the rule amendments.

What Does Reporting Reputation Buy in the Options Market? The Effect of Reporting Streaks on Ex-Ante Uncertainty

This paper predicts and finds that investor ex ante uncertainty is decreasing in a firm’s reporting reputation. Our two primary proxies for investor uncertainty are model-free implied volatilities and the variance risk premium, both with maturities that surround a key disclosure event: an impending quarterly earnings announcement. Our experimental construct of reporting reputation reflects manager/firm ability to achieve expected earnings targets, empirically measured as the number of consecutive quarters the firm meets or beats the consensus analyst forecast (i.e., “reporting streak”). We document that both measures of uncertainty are decreasing in the length of the reporting streak. We interpret this as consistent with investor learning, under which investors appear to condition the expected price effects (and related compensation to provide price protection) surrounding a major future signal release upon the firm’s historical tendency to deliver anticipated performance.

A Tale of Two Uncertainties

Consistent with the predictions from a stylized Bayesian learning model, this study shows the opposite effects of two types of uncertainty— market uncertainty and firm-specific uncertainty — on investors’ learning from new information. I provide novel evidence that investor learning increases with the level of market uncertainty, and decreases with firm-specific signal uncertainty. The stock price responses to earnings announcements increase when market volatility is high, whereas the price responses decrease when a firm’s cash flow volatility is high. I do not find evidence of alternative explanations such as liquidity effects, increased firm risk, or differences in the information environment under market uncertainty.

Tests of Investor Learning Models Using Earnings Innovations and Implied Volatilities

This paper investigates alternative models of learning to explain changes in uncertainty surrounding earnings innovations. As a proxy for investor uncertainty, we use model-free implied volatilities; as a proxy for earnings innovations, representing signals of firm performance likely to drive investor perceptions of uncertainty, we use quarterly unexpected earnings benchmarked to the consensus forecast. First, we document ― consistent with most Bayesian models of learning and prior research ― that uncertainty declines on average after the release of quarterly earnings announcements. Second, we show that this decline is attenuated by the size of the signal ― that is, by the magnitude of the earnings innovation. This latter result is inconsistent with some models of Bayesian learning, and consistent with more sophisticated models, which incorporate signal magnitude as a factor driving changes in uncertainty. Third and most important, we document that signals deviating sufficiently from expectations ― that is, sufficiently large earnings innovations ― lead to net increases in uncertainty. Critically, this latter result suggests that models allowing for posterior variance to be greater than prior variance even after signal revelation (such as “regime shifts” in Pastor and Veronesi 2009) appear more descriptive of how investors incorporate new information.

Bayesian Bid Updating in Experimental IPO Pricing Methods

This article investigates bid updating and subject learning in experimental initial public offers (IPO) of stock pricing methods. A clock auction stands for the Dutch auction. A sealed-bid uniform price auction represents book building. A two-stage book building emulation represents a recent innovation called competitive IPO. The literature addresses investor learning in a single method from actual bidding data in Taiwan. Investor IPO bidding is commonly private information and one needs to employ experiments to analyze investor behavior. Rational or Bayesian updating takes place when investor performance improves in successive bids. Naive reinforcement or updating means that subjects bid influenced by their success in previous biddings, with little consideration of the information available. The experimental book building led to gains more often. The evidence in the updating analysis suggests that learning occurs according to Bayesian updating and not to naive updating in “book building” and is consistent with it being the most common IPO method around the world.

A JOINT EXPERIMENTAL ANALYSIS OF INVESTOR BEHAVIOR IN IPO PRICING METHODS

This article jointly examines the differences of laboratory versions of the Dutch clock open auction, a sealed-bid auction to represent book building, and a two-stage sealed bid auction to proxy for the “competitive IPO”, a recent innovation used in a few European equity initial public offerings. We investigate pricing, seller allocation, and buyer welfare allocation efficiency and conclude that the book building emulation seems to be as price efficient as the Dutch auction, even after investor learning, whereas the competitive IPO is not price efficient, regardless of learning. The competitive IPO is the most seller allocative efficient method because it maximizes offer proceeds. The Dutch auction emerges as the most buyer welfare allocative efficient method. Underwriters are probably seeking pricing efficiency rather than seller or buyer welfare allocative efficiency and their discretionary pricing and allocation must be important since book building is prominent worldwide.

Through the looking glass: Indirect inference via simple equilibria

This paper develops an indirect inference (Gourieroux et al., 1993; Smith, 1993) estimation method for a large class of dynamic equilibria. Our approach consists of constructing econometrically tractable auxiliary equilibria, obtained by simplifying the economic primitives of the structural equilibrium model, via which estimation can proceed. We use this approach to develop an accurate estimator for the long-run risk model of Bansal and Yaron (2004). We demonstrate the method in Monte Carlo simulations and implement it on U.S. data. We also illustrate the good performance of the proposed methodology on an equilibrium model with investor learning.

The Role of Revenue-Expense Mismatching in Explaining the Tax Fundamental's Ability to Predict Earnings Growth

Prior research finds that taxable income is a more useful performance metric when book income quality is low (i.e., the “supplemental information role of taxable income”). We predict and find that taxable income’s supplemental information role for future earnings growth increases over time as book income quality declines over time. This time-series increase in the supplemental information role of taxable income is associated with temporal changes in the market pricing of book-tax differences. We find that mispricing disappears over time in firms where the supplemental information role of taxable income is less pronounced, but persists in firms where the supplemental information role of taxable income is more pronounced. Thus, the complexity of tax-based information and the time-series increase in the supplemental role of taxable income appear to be factors that limit investor learning. Investigating the increase in the supplemental information role of taxable income helps researchers and market participants better understand the time-series implications of two different performance metrics for predicting earnings growth and the equity valuation process as book income quality declines over time.

Through the Looking Glass: Indirect Inference via Simple Equilibria

This paper proposes an indirect inference (Gourieroux, Monfort and Renault, 1993; Smith, 1993) estimation method for a large class of dynamic equilibrium models. Our approach is based on the observation that the econometric structure of these systems naturally generates auxiliary equilibria that can serve as building blocks for estimation. We use this insight to develop an accurate estimator for the long-run risk model of Bansal and Yaron (2004). We demonstrate the accuracy of our method by Monte Carlo simulation and estimate the long-run risk model on U.S. data. We also illustrate the good performance of the methodology on an asset pricing model with investor learning.

Overnight price discovery and the internationalization of a currency: The case of the Korean won

This paper proposes a measure for the level of internationalization of a currency. It is based on the relative share of price discovery, termed the information share, in overnight offshore markets. I show that the overnight offshore markets play an increasingly important role in determining the KRW/USD exchange rate. The information share of the overnight offshore markets has increased from less than 10% in 1999–2000 to above 50% in 2011–2012. The variations in the overnight information share are consistent with variations in overnight market efficiency and investor learning.