The Information Content of Corporate Earnings: Evidence from the Securities Exchange Act of 1934

• Published date: 01 September 2022
• Author: OLIVER BINZ,JOHN R. GRAHAM
• Date: 01 September 2022
• DOI: http://doi.org/10.1111/1475-679X.12425

DOI: 10.1111/1475-679X.12425

Journal of Accounting Research

Vol. 60 No. 4 September 2022

Printed in U.S.A.

The Information Content of

Corporate Earnings: Evidence from

the Securities Exchange Act of 1934

OLIVER BINZ∗AND JOHN R. GRAHAM†

Received 7 January 2021; accepted 31 January 2022

ABSTRACT

We examine whether the Securities Exchange Act of 1934 increased the infor-

mation content of corporate earnings disclosures. Prior research questions

whether the Act improved disclosure quality but generally relies on long-

window tests and yields mixed results. We focus on whether the Act increased

earnings informativeness, improving upon prior designs by focusing on short

earnings announcement windows and employing a difference-in-differences

design to control for potential contemporaneous structural changes. We doc-

ument an increase in earnings informativeness following the Act, which is

larger for treatment firms (which withheld disclosure before the Act) than

∗INSEAD Asia Campus; †Fuqua School of Business, Duke University, and the NBER

February 2022

Accepted by Philip Berger. Weappreciate helpful comments and suggestions from Hami Ami-

raslani, Dan Bens, Thomas Bourveau, Greg Burke, Gavin Cassar, Scott Dyreng, Fabrizio Ferri,

Elia Ferracuti, Robert Holthausen, Xu Jiang, Peter Joos, Matt Kubic, Mark Lang, Bill Mayew,

Francine McKenna, Steve Monahan, Suresh Nallareddy, Katherine Schipper, Mani Sethura-

man, Kevin Standridge, Robert Stoumbos, Rahul Vashishtha, Mohan Venkatachalam, Lauren

Vollon, an anonymous reviewer, and the workshop participants at Duke University, INSEAD,

the European Accounting Symposium for Young Scholars (EASYS) webinar,the Early Insights

in Accounting webinar, the Virtual Corporate Finance Fridays webinar, and the University

of Miami. We thank Adarsh Abhinav, Sai Challa, Akshay Jituri, Maksym Kosachevskyy, Jairaj

Joshi, Tyler Johnson, Rishi Badola, VacheDarbinyan, Darshit Parekh, Srushti Deshpande, and

Sheilvy Soetanto for excellent research assistance. All errors remain our own. An Online Ap-

pendix to this paper can be downloaded at http://research.chicagobooth.edu/arc/journal-

of-accounting-research/online- supplements

1379

© 2022 The Chookaszian Accounting Research Center at the University of Chicago Booth School of

Business.

This is an open access article under the terms of the Creative Commons Attribution License, which

permits use, distribution and reproduction in any medium, provided the original work is properly cited.

1380 o. binz and j. graham

for control firms. The increase in informativeness is more pronounced for

firms that are subject to stronger enforcement.

JEL codes: G14, G18, M40, M41, M48

Keywords: mandatory disclosure; earnings announcements; information

content; valuation; enforcement

1. Introduction

The Securities Exchange Act of 1934 (hereafter “the Act”) is the most ex-

pansive secondary market regulation enacted in the history of the United

States.1The Act was the first federal law to mandate disclosure of audited

financial statements, established the Securities and Exchange Commission

(SEC) as an enforcement body, and is still the basis of most financial liti-

gation.2Given that the Act changed the reporting environment from no

federal mandatory reporting to mandatory reporting and increased en-

forcement, it provides an ideal setting to study whether its implementation

improved the informativeness of accounting numbers. A report from the

President and the Council of Economic Advisers [2003, pp. 96–97] indi-

cates that this important issue is still unresolved: “whether SEC enforced

disclosure rules actually improve the quality of information that investors

receive remains a subject of debate among researchers almost 70 years af-

ter the SEC’s creation.” Some even argue that the Act did not improve the

quality of information at all (Benston [1969, 1973]).

To date, researchers have examined firms’ unconditional stock return

performance around the Act and documented no change in returns but a

decrease in return volatility. However, the scarcity of evidence and the sharp

disagreement about interpretation led Easterbrook and Fischel [1984, p.

714] to conclude that “there is no good evidence that the disclosure rules

are beneficial [but] there is [also] no good evidence that the rules are

harmful, or very costly.” These prior studies generally rely on long-run re-

turns tests, which measure the net benefit of the Act and cannot speak

to specific costs and benefits (e.g., whether firms’ financial disclosures be-

came more informative as a result of the Act). We hand-collect financial

statement data and earnings announcement dates for the period around

the Act and directly analyze short windows of market activity to determine

whether the Act provided a specific benefit: more informative financial re-

porting.

1Rajan and Zingales [2003] argue that the Securities Act of 1933 and the Securities Ex-

change Act of 1934 laid “the accounting, regulatory, and legal foundation […] for today’s

vibrant financial system in the United States,” which they identify as one of the main drivers

of the unprecedentedly rapid economic growth observed over the past century.

2Of the core private class action securities fraud lawsuits filed in 2019, 87% were based on

the Act’s Section 10b-5, which regulates security purchases and sales (Cornerstone Research

[2020], figure 9).

the information content of corporate earnings 1381

We examine two market outcome variables to measure the informative-

ness of earnings announcements: perfect-foresight hedge portfolio returns

and earnings response coefficients (ERCs).3Returns on perfect-foresight

hedge portfolios (that are long in stocks of firms that report an earnings

increase and short in stocks of firms that report an earnings decrease) mea-

sure the value of accounting to investors as the trading return investors

could earn if they knew the accounting information before it was released

to the public (Ball and Brown [1968]). ERCs, the coefficients obtained

from regressing returns on earnings news, measure how much investors

bid up (down) stock prices upon the announcement of favorable (unfavor-

able) news. Thus, if the Act increased financial reporting informativeness,

we would expect perfect-foresight hedge portfolio returns and ERCs to in-

crease.

We begin our analysis of a large, representative sample of New YorkStock

Exchange (NYSE) firms by studying the cumulative returns of perfect-

foresight hedge portfolios. Consistent with a large increase in the value

of accounting information, hedge portfolio returns during earnings an-

nouncement windows are significantly higher after than before the Act.4

By calculating market responses conditional upon the underlying earnings

news and over a very short window, our test offers the statistical advantage

of clearly attributing the documented effect to earnings disclosure rather

than to dissemination through alternative information channels, and it mit-

igates the effects of other confounding events. Still, it is possible that con-

temporaneous structural developments unrelated to the Act changed the

informativeness of financial reports.

3We examine the robustness of our results to alternative approaches, such as the one pro-

posed by Ball and Shivakumar [2008], and generally find evidence confirming our inferences.

However, we focus on perfect-foresight hedge portfolio returns and ERCs because they are

based on earnings information, while these alternative approaches are not. Given that our re-

search question is whether the Act increased the information content of corporate earnings,

conditioning our inferences on earnings information is crucial because alternative approaches

are generally reflective not only of the earnings information, but also of all other information

released in the earnings announcement (e.g., Beaver, McNichols, and Wang[2020]).

4These findings raise the question of whether the net benefit of the Act was positive or

negative. We are not able to answer this question because of a lack of data on the costs of

the Act. If one were to accept the prior literature’s conclusion that the Act did not on net

affect welfare, this suggests that the costs of the Act are approximately as large as the benefits

we estimate. Phillips and Zecher [1981] estimate that the aggregate direct costs of complying

with SEC regulation were $1 billion in 1980. However, they also estimate that in the same year

firms incurred costs of approximately $2 billion just by distributing voluntary disclosures to

investors, which also suggests that firms would have incurred much of the compliance costs

voluntarily even in the absence of regulation. Thus, direct compliance costs appear to be

small. However, indirect compliance costs—such as proprietary costs arising from product

market competition; increased noise arising from mandatory, uninformative disclosures; and

the substitution of mandatorily disclosed, less useful information for voluntarily disclosed,

more useful information—may be substantial but are hard to measure (Verrecchia [1983],

Easterbrook and Fischel [1984]).