Giving context to accounting numbers: The role of news coverage

https://doi.org/10.1016/j.dss.2010.08.025Get rights and content

Abstract

Accounting numbers such as earnings per share are an important information source that conveys the value of firms. Previous studies on the return-earnings relation have confirmed that stock prices react to the information content in accounting numbers. However, other information sources such as financial news may also contain value-relevant information and affect investors' reaction to earnings announcements. We quantify news coverage about S&P 500 companies in the Wall Street Journal (WSJ) before earnings announcements and model its interaction with the return-earnings relation. Our empirical results show that news coverage decreases the information content of unexpected earnings and thus leads to a lower earnings response coefficient (ERC) for firms with higher news coverage. Statistically significant interaction between news coverage and unexpected earnings was observed. News coverage does not impact cumulated abnormal returns directly. We further document that this finding is not driven by firm size. The results suggest that financial news may play an important role in conveying value-related information to the markets.

Introduction

Accounting numbers are an important means for management to communicate firm performance to outside investors. Through regularized financial statements, investors receive creditable and useful firm-specific information, which helps them better evaluate the true value of firms. High quality accounting numbers not only reduce the information asymmetry between managers and outside investors, but also facilitate sound investment decisions and the efficiency of security markets.

The usefulness of accounting numbers has been an important issue for accounting researchers and general investors. Using the earnings response coefficient (ERC) to measure the magnitude of the relationship between stock returns and earnings, previous studies have concluded that the ERC is significantly positive [25] and numerical earnings information indeed conveys value-relevant information to the markets. The empirical results suggest that a favorable earnings surprise induces positive abnormal stock returns, while an unfavorable earnings surprise induces negative abnormal stock returns.

Accounting numbers, nonetheless, are not the only source of information conveying the fundamental value of firms. Other sources, such as financial news, trade association publications, and reports issued by analysts and brokerage houses may also contain useful information. These alternative information sources often provide timely updates between earnings announcements and may play an important role in shaping investors' beliefs. The magnitude of the ERC, as a result, may be influenced by these information sources.

While highly circulated financial news has been shown to impact short-term market returns [32], few studies have investigated how financial news impacts the return-earnings relation. Tetlock et al. [33] quantified sentiment in news articles by counting words associated with negative outlooks. Their empirical results showed that the fraction of negative words in firm-specific news stories forecasts low firm earnings. Previous studies on ERCs have identified four important determinates: earnings persistency, firm risk, firm growth, and interest rate [7], [13], [25]. However, these studies have not examined the interaction between the information content of earnings and news articles.

Given the limitations of previous studies, our research aims at investigating how financial news coverage impacts the return-earnings relation. To the best of our knowledge, this is the first study that documents how financial news coverage affects investors' reactions to accounting earnings. We used the Wall Street Journal (WSJ) as the representative source of financial news and collected news articles discussing S&P 500 companies from August 1999 to February 2007. Our collection contains 283,457 news articles and spans more than seven years. This testbed provides a solid ground for statistical inference. Firm-level news coverage computed from our news collection facilitates our investigations on the interaction between financial news coverage and return-earnings relation.

The remainder of the paper is organized as follows. Section 2 provides a review of related literature followed by the discussion of the research objectives and hypotheses in Section 3. In Section 4, we describe our data sources and empirical models. The main findings of the study are presented in Section 5. Section 6 discusses the managerial implications of our results. We conclude with a summary and future research directions in Section 7.

Section snippets

Literature review

The seminal works of Ball and Brown [2] and Beaver [5] spawned the study of the information content of accounting numbers. Researchers study a wide range of topics via the return-earnings relation and the event study framework [14], [25]. In this section we first summarize the earnings response coefficient research and then focus on two major aspects that are directly relevant to this study: lagged performance information in accounting earnings and asymmetry in the return-earnings relation.

Hypotheses development

In this study, we aim to investigate the role of financial news coverage in determining the ERC. Financial news contains timely updates on firm value. The “price lead earnings” viewpoint suggests that investors would have incorporated the information from financial news into stock prices before the earnings announcements. The information content of earnings announcements, as a result, is reduced by news coverage. In other words, we expect a negative relationship between news coverage and ERCs:

H1

Research testbed

We used articles from the WSJ to develop our research testbed. News articles from August 1999 through February 2007 were collected from the ProQuest database. We retrieved 283,457 news articles in total for the 91-month period and developed a system for automatic analysis of firm news coverage. We focused on S&P 500 companies because financial news tends to cover large firms. The monthly S&P 500 companies list was obtained from the Center for Research on Security Prices (CRSP) database.

Daily

Empirical results

We estimated Eqs. (1), (2) via ordinary least square regression. The upper and lower 1% of UEit and CARit,[0,2] were winsorized (i.e., extreme values were replaced with the 1% or 99% percentile values) to guard against outliers. The upper 1% of NewsFrequencyit was also winsorized. Winsorizing selected variables prevents the potential negative effects of extreme values when conducting regression analysis. Our main results would remain qualitatively the same if unwinsorized data were used. The

Discussion

Our empirical results have several important implications for managers and investors. First, if managers prefer higher CARs around earnings events, then only when the unexpected earnings are negative should managers release information regarding forthcoming earnings during the pre-announcement period. The reason can be explained by taking expectation to Eq. (2) and differentiating with respect to NewsFrequencyit:∂ECARit,[0,2]∂NewsFrequencyit=c1+clUEit≡NNitwhere NNit is the expected change of CAR

Conclusions and future research directions

This study investigates the influence of news coverage on the ERC, which measures the information content of earnings. We collected news articles in the Wall Street Journal from August 1999 through February 2007 to construct measures for news coverage on S&P 500 companies. Combined with data from classical financial databases such as IBES, Compustat and CRSP, we were able to study the effect of news coverage on earnings surprise.

Our empirical results indicate that news coverage has a

Acknowledgements

This work was supported in part by the US National Science Foundation under grant CNS-0709338 and the National Science Council of Taiwan (NSC97-2410-H002-125-MY3). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.

Kuo-Tay Chen is an associate professor in the Department of Accounting at National Taiwan University. He received a Ph.D. degree in Management Information Systems from the University of Texas at Austin. His current research interests focus on applications of data mining and text mining to accounting and finance, information systems auditing, and applications of social network analysis to supply chain management and corporate governance. He has published in International Journal of Accounting

References (33)

  • J.R. Graham et al.

    The economic implications of corporate financial reporting

    Journal of Accounting & Economics

    (2005)
  • C. Hayn

    The information content of losses

    Journal of Accounting & Economics

    (1995)
  • R. Jacobson et al.

    Myopic management behavior with efficient, but imperfect, financial markets — a comparison of information asymmetries in the United States and Japan

    Journal of Accounting & Economics

    (1993)
  • S.P. Kothari

    Capital markets research in accounting

    Journal of Accounting & Economics

    (2001)
  • S.P. Kothari et al.

    Information in prices about future earnings: implications for earnings response coefficients

    Journal of Accounting and Economics

    (1992)
  • Z. Ma et al.

    Discovering company revenue relations from news: a network approach

    Decision Support Systems

    (2009)

    • Cited by (0)

    Kuo-Tay Chen is an associate professor in the Department of Accounting at National Taiwan University. He received a Ph.D. degree in Management Information Systems from the University of Texas at Austin. His current research interests focus on applications of data mining and text mining to accounting and finance, information systems auditing, and applications of social network analysis to supply chain management and corporate governance. He has published in International Journal of Accounting Studies (Taiwan), Journal of Contemporary Accounting (Taiwan), Taiwan Accounting Review, Data & Knowledge Engineering.

    Hsin-Min Lu received the bachelor's degree in business administration and MA degree in economics from the National Taiwan University, and the PhD degree in information systems from the University of Arizona. He is an Assistant Professor in the Department of Information Management at the National Taiwan University. His research interests include data mining, text mining, and applied econometrics.

    Tsai-Jyh Chen is a professor of risk management and insurance at the National Chengchi University (NCCU). She received her Ph.D. from the Wharton School, University of Pennsylvania. She was the chairman of the Department of Risk Management and Insurance and the director of the Graduate Institute of Insurance at the NCCU. Currently she serves as the director of English Taught Program in the commerce college of NCCU and the chapter advisor of International Honor Society Beta Gamma Sigma — NCCU Chapter, She has also provided services for the governmental and professional organizations. Dr. Chen has authored and translated several books and published many articles in academic and professional journals. She is the associate editor of the Insurance Issues and Practices (Taiwan Insurance Institute).

    Shu-Hsing Li received the bachelor's degree in business administration from National Chengchi University in Taiwan, and the PhD degree in accounting from New York University. He is now Professor of Accounting at National Taiwan University, and Chair Professor of Accounting at Tunghai University. He has also taught at Rutgers University and the University of Hawaii at Manoa. His academic publications have appeared in the Accounting Review, Journal of Accounting, Auditing and Finance, European Journal of Operational Research, Review of Quantitative Finance and Accounting, IEEE Intelligent Systems, and other scholarly journals. He is currently the Editor of NTU Management Review, and the Director of Enterprise Risk Management and Business Intelligence Research Center at National Taiwan University. He is the leading scholar in Taiwan working on the transfer pricing for multinational companies.

    Jian-Shuen Lian received his bachelor's degree in Accounting from the National Taiwan University, Taipei, in 1995, and master degree in Information Science from the University of Pittsburgh, PA, in 2000. Now he is an Accounting PHD candidate at the National Taiwan University, Taipei.

    Hsinchun Chen received the BS degree from the National Chiao-Tung University in Taiwan, the MBA degree from the State University of New York at Buffalo, and the PhD degree in information systems from New York University. He is a McClelland professor of management information systems at the University of Arizona. He has served as a scientific counselor/advisor of the US National Library of Medicine, the Academia Sinica (Taiwan), and the National Library of China (China). He is a fellow of the IEEE and the AAAS. He received the IEEE Computer Society 2006 Technical Achievement Award. He was ranked #8 in publication productivity in information systems (CAIS 2005) and #1 in Digital Library research (IP&M 2005) in two bibliometric studies. His COPLINK system, which has been quoted as a national model for public safety information sharing and analysis, has been adopted in more than 550 law enforcement and intelligence agencies in 20 states.

    View full text
    Copyright © 2010 Elsevier B.V. All rights reserved.