Trade Secrets Law and Corporate Disclosure: Causal Evidence on the Proprietary Cost Hypothesis

• Author: LIANDONG ZHANG,YUPENG LIN,YINGHUA LI
• Published date: 01 March 2018
• Date: 01 March 2018
• DOI: http://doi.org/10.1111/1475-679X.12187

DOI: 10.1111/1475-679X.12187

Journal of Accounting Research

Vol. 56 No. 1 March 2018

Printed in U.S.A.

Trade Secrets Law and Corporate

Disclosure: Causal Evidence on the

Proprietary Cost Hypothesis

YINGHUA LI,

∗YUPENG LIN,

†AND LIANDONG ZHANG ‡

Received 19 September 2015; accepted 24 August 2017

ABSTRACT

This study exploits the staggered adoption of the inevitable disclosure doc-

trine (IDD) by U.S. state courts as an exogenous shock that generates varia-

tions in the proprietary costs of disclosure. We find that firms respond to IDD

adoption by reducing the level of disclosure regarding their customers’ iden-

tities, supporting the proprietary cost hypothesis. Our results are stronger for

firms in industries with a higher degree of entry threats, for firms in more

volatile industries, and for firms with a lower degree of external financing de-

pendence. Overall, this study represents one of the first efforts in identifying

the causal effect of proprietary costs of disclosure on the supply of disclosure.

JEL codes: D82; D83; K11; L1; M41

Keywords: proprietary costs; corporate disclosure; customer identity; in-

evitable disclosure doctrine; trade secrets law

∗School of Accountancy, Arizona State University; †NUS Business School, National Univer-

sity of Singapore; ‡School of Accountancy, Singapore Management University.

Accepted by Christian Leuz. We appreciate helpful comments from two anonymous re-

viewers, Derek Chan, Qi Chen, Qiang Cheng, Richard Crowley, Sterling Huang, Kai-Wai

Hui, Bin Ke, Clive Lennox, Jing Li, Oliver Li, Hai Lu, Shuqing Luo, Wei Luo, Xiumin Mar-

tin, Ke Na, Chul Park, Hojun Seo, Charles Shi, Douglas Skinner, Charles Wang, Xin Wang,

Hao Xue, Haiwen Zhang, participants in the 2016 Tsinghua International Corporate Gov-

ernance Conference, and the seminar participants and colleagues at the City University of

Hong Kong, National University of Singapore, Singapore Management University, and Uni-

versity of Hong Kong. We thank Bill McDonald for sharing his 10-K headers data. Wethank Di

Zhao for capable research assistance. An online appendix to this paper can be downloaded at

http://research.chicagobooth.edu/arc/journal-of-accounting-research/online-supplements.

265

Copyright C, University of Chicago on behalf of the Accounting Research Center,2017

266 Y.LI,Y.LIN,AND L.ZHANG

1. Introduction

Corporate disclosure reduces information asymmetry between managers

and shareholders, which may increase a firm’s stock liquidity and decrease

its cost of capital. Thus, absent any costs, value-maximizing managers have

incentives to fully disclose their private information. Full disclosure, how-

ever, is rarely observed in the capital market. One potential constraint to

full disclosure is that some disclosures may damage a firm’s competitive

position in its product market. Verrecchia [1983] shows that, in the pres-

ence of proprietary costs, partial disclosure can be optimal, with the level

of disclosure decreasing in the proprietary costs of disclosure.

In this study, we seek to provide causal evidence on the proprietary cost

hypothesis. Although the impact of proprietary costs of disclosure on the

decision to disclose proprietary information is theoretically unambiguous,

the empirical literature to date fails to provide conclusive evidence on the

proprietary cost hypothesis, possibly because of the following challenges

(Berger [2011]).1First, the proprietary costs of disclosure are generally

unobservable to researchers and thus researchers often have to use indus-

try structure to approximate variations in proprietary costs. The relation

between industry structure and proprietary costs, however, is theoretically

ambiguous. For example, product market concentration can be either pos-

itively or negatively correlated with proprietary costs of disclosure, depend-

ing on whether firms face existing competitors or the threat of entry, and

on whether firms compete primarily on the basis of price or long-run ca-

pacity decisions (e.g., Darrough and Stoughton [1990], Verrecchia [1990a,

b], Wagenhofer [1990], Darrough [1993], Gigler, Hughes, and Rayburn

[1994], Clinch and Verrecchia [1997], Penno [1997], Dye [1998]). This

theoretical issue is further complicated by measurement errors in indus-

try structures (e.g., Ali, Klasa, and Yeung [2009], Berger and Hann [2007],

Dedman and Lennox [2009]). For example, Ali, Klasa, and Yeung [2009]

demonstrate that using industry concentration measures from the Census

(including both private and public firms) overturns many results of previ-

ous literature using Compustat concentration measures.

Second, measures of proprietary costs of disclosure (such as industry

structure) may actually capture omitted capital market benefits and agency

costs of disclosure. As one example, industry structure measures are often

influenced by the number of firms in an industry. The greater the num-

ber of firms, the lower are firms’ capital market benefits associated with

their own disclosure because capital market participants learn from infor-

mation spillovers provided by related firms.2Thus, it is unclear whether

1See subsection 2.3 for a detailed review of the empirical literature on proprietary costs of

disclosure.

2See, for example, Admati and Pfleiderer [2000], Baginski and Hinson [2016], Shroff,

Verdi, and Yost[2016], Breuer, Hombach, and M¨

uller [2016], and Berger,Minnis, and Suther-

land [2017].

TRADE SECRETS LAW AND CORPORATE DISCLOSURE 267

the relation between high competition (measured by low concentration)

and low disclosure is due to high proprietary costs or low capital market

benefits. Segment disclosure is another context in which proprietary costs,

capital market benefits, and agency costs are complicatedly entwined. It is

often argued that managers do not disclose segment information to hide

highly profitable segments for proprietary cost motives (Berger and Hann

[2007]).3However,given that the average profitability is publically available

in the income statement, managers must also hide some unprofitable seg-

ments, or otherwise the reported average profitability will reveal the truth

(Leuz [2004]). Thus, it is possible that managers are in fact hiding unprof-

itable segments through nondisclosure to avoid shareholder scrutiny and

enjoy a quiet life or to maintain the level of stock prices.4From these exam-

ples, we can see that it is difficult to disentangle the effects of proprietary

costs, agency costs, and capital market benefits on disclosure choices. A re-

lated challenge plaguing this literature is that the information disclosed by

firms is often not merely or not even majorly relevant for competitors. For

instance, while management earnings forecasts are informative to investors,

it is unclear what specific competitive advantage a firm sacrifices in disclos-

ing earnings forecasts shortly before the actual earnings announcements

(Lang and Sul [2014]).

We strive to address these challenges using a research design with two

key features. First, we exploit a trade secrets law setting that provides plau-

sibly exogenous shifts in the proprietary costs of disclosure, holding other

determinants of disclosure approximately fixed. Unlike most studies in the

prior literature, this setting does not rely on the theory and measurement

of industry structure. Second, we investigate a disclosure item that is unam-

biguously known to the firm and is primarily more relevant to competitors

than to other stakeholders. This feature further helps mitigate concerns

about correlated omitted variables related to agency costs or capital market

benefits of disclosure.5

Specifically, our empirical strategy is based on the staggered adoption of

the inevitable disclosure doctrine (IDD) by U.S. state courts since the 1970s

(Klasa et al. [2017]). The IDD is a theory in trade secrets law that grants the

employer (i.e., the plaintiff) an injunction to prevent a current or former

employee (i.e., the defendant) from working for another company, if the

3Managers also claim that they “do not want to explicitly reveal sensitive proprietary in-

formation ‘on a platter’ to competitors, even if such information could be partially inferred

by competitors from other sources, such as trade journals or trade conferences” (Graham,

Harvey, and Rajgopal [2005, p. 62]).

4Cho [2015] finds that more transparent segment disclosure is associated with less severe

agency problems and higher investment efficiency.

5This exercise is in the spirit of Leuz [2004], which distinguishes between disclosures that

are more likely to be proprietary and those that are more relevant to capital markets. The

author looks at cash flow statements, which are likely not proprietary but relevant to capital

markets.