JOURNAL OF MANAGEMENT ACCOUNTING RESEARCH American Accounting Association Vol. 25 DOI: 10.2308/jmar-50492 2013 pp. 59–63

DISCUSSION OF

Knowing Versus Telling Private Information About a Rival Dhananjay (DJ) Nanda

University of Miami

ABSTRACT: Bagnoli and Watts (2013) show that a firm will always disclose its private

information when this information solely affects its rival’s product market decisions. This

result is robust to different competitive scenarios (Cournot or Bertrand competition),

features (product heterogeneity or private information quantity), and levels of commitment

(ex ante or ex post). I highlight how this result fits in the accounting disclosure literature,

describe the intuition behind the theory, and discuss its implications for future work.

Keywords: discretionary disclosure; product market competition; private information.

INTRODUCTION

O ver the last three decades, economics, finance, and accounting scholars have produced a large theoretical and empirical literature that examines firms’ discretionary disclosure policies. A central premise of the theory is ‘‘any entity making a disclosure will disclose information that

is favorable to the entity, and not disclose information unfavorable to the entity’’ (Dye 2001). Bagnoli and Watts (2013) add to this body of work by theoretically examining firms’ incentives to disclose information that solely pertains to its rivals’ product market prospects. They establish that a firm possessing such information will always disclose it, a result robust to several modeling assumptions.

In this discussion, I first provide a brief summary of the insights gained from the extant discretionary disclosure literature to highlight Bagnoli and Watts’ contribution. I then discuss their modeling assumptions to isolate the intuition behind their result. Finally, I conclude with remarks on the model’s implications and an area that perhaps provides a setting where one observes the modeled information environment.

DISCRETIONARY DISCLOSURE

Since the early 1980s, scholars have examined the incentives of informed agents revealing their

information to uninformed counterparties in an exchange economy (see Grossman 1981; Milgrom

1981).1 The most parsimonious representation of this exchange is described by a model of an asset seller

I thank Ramji Balakrishanan (editor) for helpful suggestions.

Published Online: April 2013

1 I will refrain from providing a survey of the entire discretionary disclosure literature as it is beyond the scope of this discussion. Verrecchia (2001) and Dye (2001) provide a comprehensive review of the theoretical accounting disclosure literature.

59

that possesses information about the asset’s value that potential buyers for that asset do not. Absent the

seller disclosing his/her private information, the buyers rationally suspect that the withheld information

adversely reflects on the asset’s value. Consequently, the price that sellers are willing to pay for the asset

incorporates a discount reflecting their conjecture regarding the amount by which the asset value is lower.

In equilibrium, the discount equals the amount that makes it in the seller’s best interest to disclose his/her

private information about the asset’s value. This tradeoff represents the ‘‘full disclosure’’ equilibrium and

is often referred to as the ‘‘unraveling’’ result.

What are the modeling assumptions that sustain the unraveling result? First, buyers know that

the seller has value-relevant information about the asset offered for sale. Second, buyers are

homogeneous in interpreting disclosure (and non-disclosure), i.e., they have identical beliefs about

the seller’s private information. Third, the seller’s disclosure is credible, i.e., it is truthful. Finally,

disclosure is costless to the seller and buyers alike. Subsequent work has attempted to overcome the

unraveling result of full disclosure in part to explain empirically observed partial disclosure;

economic agents do (sometimes) withhold information. Specifically, researchers have shown that

firms/managers do withhold information when the assumptions behind the ‘‘full disclosure’’ model

are violated. Broadly characterized, these modeling innovations are: (1) the seller may or may not

be informed (Dye 1985; Jung and Kwon 1988), (2) buyers have heterogeneous beliefs about the

seller’s information (Lintner 1969), (3) the seller’s disclosure may not be truthful (Milgrom and

Roberts 1986), and (4) the seller faces disclosure costs (Verrecchia 1983). In the presence of these

disclosure frictions, researchers have shown that the full disclosure equilibrium need not hold.

Bagnoli and Watts build on the disclosure-cost (innovation 4) narrative in the theoretical

accounting discretionary disclosure literature. Specifically, prior work has shown that firms facing

proprietary costs, i.e., costs only incurred if information is disclosed, will not always disclose

information as potential counterparties now interpret non-disclosure as avoiding costs rather than

suppressing unfavorable information. The simplest representation of proprietary costs is an

exogenous fixed cost only incurred if a firm chooses to disclose its private information (Verrecchia

1983). In this case the disclosure equilibrium is a threshold: only information that is sufficiently

favorable (above the threshold) is disclosed. Subsequent work endogenizes the proprietary cost of

disclosure by considering firm disclosure policy in an imperfectly competitive product market

where the cost of disclosing private information arises from competitors gaining a strategic

advantage in their pricing or output choices (Darrough and Stoughton 1990). In this literature, the

disclosure equilibrium is affected by modeling choices: cost- or demand-related information,

Cournot or Bertrand competition, and whether the firm chooses its disclosure policy before or after

observing the information (see Darrough 1993).

KNOWING VERSUS TELLING

Bagnoli and Watts study a firm’s disclosure policy in a product-market setting, with two

competing firms facing a linear demand curve and constant marginal production costs over a single

period. Both firms are solely concerned with maximizing firm profits at the end of the period.

Consequently, valuation, agency, and dynamic issues are absent. A firm privately observes

information that solely pertains to its rival’s operations (demand or cost related) and truthfully

discloses it, if at all, pre-production. The information stochastically observed by a firm is fully

revealing of either the demand or the cost parameters of its rival. The information structure modeled

in Bagnoli and Watts is its novel contribution to the accounting disclosure literature. Specifically,

the assumption that a firm’s private information is decision-irrelevant to itself but is relevant to its

rival’s decisions leads to a robust full disclosure equilibrium, enhancing our understanding of the

role played by proprietary costs in affecting firm disclosure policy.

60 Nanda

Journal of Management Accounting Research Volume 25, 2013

Why is it in the best interests of the firm to always disclose its private information about its

rival’s economic prospects? To describe the intuition behind the result, I begin with the ex ante disclosure commitment scenario where a firm chooses its policy before observing the information

so it does not know if the information is favorable or unfavorable to its own economic prospects. If

the firm chooses not to disclose, its rival is uninformed and acts based solely on the expected value

of the demand or cost parameters, which leads to either choosing to produce too much (or price too

low) or too little (price too high). Given the assumptions of linear demand and cost, the expected

profit of the informed firm is convex and increasing in the information’s favorability (to itself ).

Consequently, a firm always prefers disclosing its private information (a gamble) to withholding the

information and having its rival choose prices or quantities based on the expected values of the

parameters. Thus, the unraveling result of full disclosure occurs when a firm’s private information

does not have any direct effect on the informed firm’s demand or cost functions.2

If a firm chooses its disclosure policy after observing the information and knowing whether it is

favorable or unfavorable (ex post disclosure), a similar full disclosure equilibrium results—the

reason being that non-disclosure by the informed firm is rationally perceived by the rival as

information that is unfavorable to the informed firm. Consequently, the rival will increase price or

quantity, thereby reducing the informed firm’s expected profit. Consequently, absent any

exogenous disclosure costs, the informed firm always discloses its information. In both cases, ex ante and ex post disclosure, private information (solely) about a rival (solely) has proprietary benefits.

IMPLICATIONS AND EXTENSIONS

Bagnoli and Watts assume that the firm’s private information is solely decision-relevant to its

rival. This naturally leads to the inquiry: What information can a firm possess about a rival that is

decision-irrelevant for itself? The authors provide several examples in the manuscript of firms

disclosing information that potentially affects their rival’s product markets. For instance, Bagnoli

and Watts cite the example of AT&T’s disclosure of the effect of smartphones on wireless

networks, which is decision-relevant to Verizon which is introducing a new device that would use

such networks. However, it seems that this information is decision-relevant for AT&T as well

because the company also supports smartphones on its data networks. The other examples in the

paper also suggest that the publicly disclosed information is decision-relevant for both the

disclosing firm and its rival. So, why is it so difficult to cite examples that fit the modeled

information environment?

There are a few potential explanations for the difficulty in citing examples of firms disclosing

information that does not affect their own business. First, there are potential costs associated with

acquiring information about a rival and disclosing it publicly, which are unmodeled in Bagnoli and

Watts. For instance, industrial espionage is prohibited in the U.S. under the Economic Espionage

Act of 1996 (18 U.S.C. §§). Because acquiring and disclosing such information has costs, the

no-disclosure equilibrium likely prevails with regard to such information. Second, perhaps firms

that possess information about their rival choose to disclose it surreptitiously. For instance, blogs

and chat rooms provide some anonymity to a person disclosing information and are yet effective in

disseminating such information. Another likely reason is that the single-period model is perhaps too

restrictive in explaining empirically observed behaviors.

A single-period setting, as modeled in Bagnoli and Watts, is analytically appealing because of

its parsimony and ability to provide ‘‘clean’’ comparative static predictions. However, firm

2 If the information affects the informed firm’s demand parameter (m ¼ 1 case under Bertrand competition in Bagnoli and Watts), then it prefers to suppress disclosure.

Discussion of Knowing Versus Telling Private Information About a Rival 61

Journal of Management Accounting Research Volume 25, 2013

disclosure policy is based on strategic considerations that are multi-period in nature. For instance, a

firm may suppress information that is potentially favorable to its rivals to discourage potential

entrants into its product market. Similarly, unfavorable information disclosures may create entry

barriers or potential exits by rivals. Consequently, multi-period considerations potentially lead to

other equilibriums than the full-disclosure equilibrium established in Bagnoli and Watts.

Multi-period models also provide the opportunity to explore other firm considerations that affect

disclosure policy such as agency issues. For instance, disclosing favorable information about a rival

may benefit the disclosing firm but may be personally costly to its manager if it adversely affects

his/her short-term performance appraisal. Overall, the Bagnoli and Watts framework provides a

starting point for future research to examine dynamic or multi-period considerations in firm

disclosures of information about rivals.

The labor market provides a potential setting where a party (firm or agent) often is privy to

information about its rival and this information is decision-irrelevant for itself. For instance, firms

possess information about an employee’s ability that is decision-relevant for a competitor

attempting to hire away the employee. Potential recruiters solicit recommendations from past

employers in an effort to gather this information. In academia, promotion and tenure decisions are

affected by evaluations of a candidate’s scholarship by employees of a rival institution. Disclosing

information in this setting is inherently strategic, with few proprietary costs, yet we often observe

credible full disclosure of private information. In fact, the private information in labor markets has

the Bagnoli and Watts feature of proprietary benefits that accrue to the disclosing entity. Even

though the labor market is not modeled in this paper, the insight that an entity with information

about its rival always discloses is applicable.

CONCLUSION

Bagnoli and Watts (2013) demonstrate that firms always disclose their private information

about their rival’s product market prospects, if this information does not pertain to their own

demand or cost characteristics and if disclosure is costless. Consequently, information about a rival

has proprietary benefits to a disclosing firm. The full-disclosure equilibrium derived in this paper is

robust to several modeling choices and assumptions. As such, this paper complements the existing

disclosure literature that has until now focused on proprietary costs of disclosure.

The Bagnoli and Watts paper sets the stage for further research on disclosure of information

that is not necessarily decision-relevant to the disclosing party. In my discussion, I identify two

potential areas of inquiry that may prove fruitful. First, to explore the strategic nature of disclosures,

multi-period models are required where dynamic incentive issues and firm entry-exit decisions can

be explored. Second, since we observe the Bagnoli and Watts modeled information structure in

labor markets, future extensions of this model should consider exploring disclosure behavior in

these settings.

REFERENCES

Bagnoli, M., and S. G. Watts. 2013. Knowing versus telling private information about a rival. Journal of

Management Accounting Research 25 (1).

Darrough, M. N. 1993. Disclosure policy and competition: Cournot vs. Bertrand. The Accounting Review

68: 534–561.

Darrough, M. N., and N.M. Stoughton. 1990. Financial disclosure policy in an entry game. Journal of

Accounting and Economics 12: 480–511.

Dye, R. A. 1985. Disclosure of nonproprietary information. Journal of Accounting Research 23: 123–145.

62 Nanda

Journal of Management Accounting Research Volume 25, 2013

Dye, R. A. 2001. An evaluation of ‘‘essays on disclosure’’ and the disclosing literature in accounting.

Journal of Accounting and Economics 32: 181–235.

Grossman, S. J. 1981. The role of warranties and private disclosure about product quality. Journal of Law and Economics 24: 461–483.

Jung, W. O., and Y. K. Kwon. 1988. Disclosure when the market is unsure of information endowment of

managers. Journal of Accounting Research 26: 146–153.

Lintner, J. 1969. The aggregation of investors’ diverse judgments and preferences in purely competitive

security markets. Journal of Financial and Quantitative Analysis 4: 347–400.

Milgrom, P. 1981. Good news and bad news: Representation theorems and applications. Bell Journal of Economics 12: 380–391.

Milgrom, P., and J. Roberts. 1986. Relying on the information of interested parties. Rand Journal of Economics 17: 18–32.

Verrecchia, R. E. 1983. Discretionary disclosure. Journal of Accounting and Economics 5: 365–380.

Verrecchia, R. E. 2001. Essays on disclosure. Journal of Accounting and Economics 32: 97–180.

Discussion of Knowing Versus Telling Private Information About a Rival 63

Journal of Management Accounting Research Volume 25, 2013

Copyright of Journal of Management Accounting Research is the property of American Accounting Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.