Sean Wang

Update: Measuring Intangible Capital with Market Prices 

• Forthcoming at Management Science

  • TL;DR - Our capitalized intangible stocks, created from parameters estimated by a model that uses market prices, have quality that is better than stocks created with status-quo BEA parameters. Researchers are welcome to download our parameters and stocks in the links provided below.

• Highlights

  • Existing accounting standards expense R&D, advertising and SG&A. 

  • These expenditures build intangible capital, but are missing from a firm's balance sheet, resulting in a downward bias of reported assets. 

  • To characterize off-balance sheet intangible assets, we use market prices to estimate this missing intangible capital. 

  • EPW adjustments to the balance sheet and income statement create market-to-book and average ROE's that now converge to their expected theoretical values.

  • EPW adjustments create higher quality capital stocks than those made with parameters currently used by the Bureau of Economic Analysis. 

• To facilitate new research on intangible capital, we now provide both the parameters and the capitalzed intangible stocks from EPW (2024) for public use.  

  • • code to construct knowledge and organization capital stocks (plus data) 

    • Mapping of SDC deal numbers to "gvkeys"

    • MD&A text from 10-Ks

Update: Do Sell-side Analysts React Too Pessimistically to Bad News for Minority-led Firms? Evidence from Target Price Valuations

• Accepted at Journal of Accounting and Economics

  • TL;DR - When bad news hits, analysts' valuations are 57% more negative for Non-White CEOs, revealing a subconscious race-related bias that diminishes with increased racial awareness and familiarity, suggesting DEI efforts can promote equality in capital markets.

• Highlights

  • Bad news has a 57% larger adverse impact on analysts' valuations when the CEO is Non-White, resulting in more pessimistic valuations compared to White CEOs.

  • Non-White CEO firms are more likely to surpass analysts' valuation targets in the subsequent 12 months, suggesting this racial gap lacks economic justification.

  • Evidence of racial bias is provided through:

    • A controlled experiment corroborating the empirical findings

    • Analysts' valuation disparities towards Non-White CEOs becoming larger when race relations are worse

  • Increases in racial awareness and CEO familiarity attenuate these disparities, suggesting the bias is subconscious.

  • Findings suggest resources allocated towards Diversity, Equity and Inclusion (DEI) regarding racial stereotypes may promote equality within capital markets.

Update: Firm-specific Information Processing and the Delayed Discovery of Macroeconomic News: Evidence from Earnings Announcement Returns

• Revise and Resubmit at Review of Accounting Studies

  • TL;DR - Investors have processing constraints. Too much focus on processing firm-specific news can lead to neglect of macro news when extreme earnings are announced. As a result, prices reflect macro-news with delay and are predictable in the subsequent days after the earnings date.

• Highlights

• Intensive efforts to acquire and process firm-specific information can delay the incorporation of macroeconomic information into stock prices.

• Extreme earnings surprises lead to the neglect of macro news embedded in the aggregate return, resulting in predictable returns over the subsequent 3-day [t+1, t+3] period  after the earnings announcement (EA) date when conditioned on the magnitude of the aggregate return on the EA date.

• Macro news delays and return predictability are larger when:

  • Firm-specific information acquisition efforts are highest (SEC Edgar downloads)

  • Investors have limited resources (retail trade intensity)

  • The firm's economics are more strongly tied to the macroeconomy (beta)

• While prior research shows that information acquisition aids price discovery, our study finds that limited investor attention and a focus on firm-specific news can delay the processing of macroeconomic information.

Latest research: Analyst Forecast Bundling Intensity and Earnings Surprise

• TL;DR - Analysts' EPS forecasts are biased, and  information is conveyed through another channel--forecast bundling. Adjusting the consensus for forecast bundling over a firm quarter results in massive improvements to the "earnings kink" and the distribution of analysts' EPS surprise. 

• Highlights

  • Analysts who bundle their EPS revisions with recommendation and target price revisions of consistent sign convey private information to capital markets about future EPS without directly revising their EPS forecasts.

  • Analysts signal information via bundled EPS reports more frequently when the cost of being bold and incorrect is high, e.g. during periods of high macroeconomic uncertainty (e.g. during COVID-19 and Great Financial Crisis).

  • Aggregating bundled EPS reports at the firm-quarter level creates a consensus measure of implicitly signaled information, BF_Score, and reveals a robust predictor of quarterly analyst-based EPS surprise, Analyst_SUE.

  • The correlation between (BF_Score_Rank and Analyst_SUE)  = 0.70, and a 1 SD increase in BF_Score  ≈  $0.03 cents of positive EPS surprise.

• Analysts also bundle (positive) private EPS news, effectively suppressing their EPS targets to help firms more easily meet/beat their expectations:

• Most importantly, accounting for the information embedded in BF_Score improves the distributional properties of Analyst_SUE by 55.9% (towards a normal distribution). These improvements come from a reduction in abnormal levels of: 

  • Kurtosis (48% improvement), negative skewness (95% improvement), kink asymmetry around zero EPS surprise (66% improvement).

• Our study suggests that implicit signals relayed through analysts’ bundling intensity should be accounted for in research involving the consensus EPS forecast & Analyst_SUE.