Innovation and Financial Disclosure

• Published date: 01 June 2024
• Author: HUI CHEN,PIERRE JINGHONG LIANG,EVGENY PETROV
• Date: 01 June 2024
• DOI: http://doi.org/10.1111/1475-679X.12546

DOI: 10.1111/1475-679X.12546

Journal of Accounting Research

Vol. 62 No. 3 June 2024

Printed in U.S.A.

Innovation and Financial Disclosure

HUI CHEN,∗PIERRE JINGHONG LIANG,†

AND EVGENY PETROV ∗

ABSTRACT

We examine how financial disclosure policy affects a firm manager’s strat-

egy to innovate within a two-period bandit problem featuring two produc-

tion methods: an old method with a known probability of success, and a new

method with an unknown probability. Exploring the new method in the first

period provides the manager with decision-useful information for the second

period, thus creating a real option that is unavailable under exploiting the old

known production method. Voluntary disclosure of the firm’s financial per-

formance provides the manager with another option to potentially conceal

initial failure from the market. The interaction of these two options deter-

mines the manager’s incentive to explore. In equilibrium, a myopic manager

who cares about the interim market price may over- or under-explore com-

pared to the optimal exploration strategy that maximizes firm value. Our anal-

ysis shows that firms operating in an environment with voluntary disclosure

early in the trial stage and mandated requirement later are most motivated

to explore, while firms subject to early mandated disclosure and late volun-

tary disclosure are least likely to do so. We also provide empirical predictions

∗University of Zürich; †Carnegie Mellon University

Accepted by Haresh Sapra. We thank the anonymous reviewer, Jeremy Bertomeu, Robert

Göx, Ilan Guttman, Xiaojing Meng, Felix Niggemann, Ulrich Schäfer,Phillip Stocken, Elyashiv

Wiedman (discussant) and the participants at the 2023 Journal of Accounting Research Con-

ference, the Cambridge Accounting Research Camp, 14th Workshop on Accounting & Eco-

nomics in Rotterdam, the seminar at the University of Zürich, the Accounting and Finance

Research Seminar at the Hong Kong University of Science and Technology, and the Account-

ing and Economics Society Asia-Pacific webinar for helpful comments.

935

© 2024 The Authors. Journal of Accounting Research published by Wiley Periodicals LLC on behalf of 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.

936 h.chen,p.j.liang,ande.petrov

about the link between the disclosure environment and the intensity and ef-

ficiency of corporate innovation.

JEL codes: D82, M41, O30, O31, O32, O33

Keywords: innovation; bandit problem; exploration and exploitation; vol-

untary disclosure; real option; disclosure option

1. Introduction

In modern economies, firms constantly face the strategic choice between

exploitation of existing technologies and exploration of new methods.

While exploitation offers familiarity and certainty of the status quo, ex-

ploration could lead to innovation, discovery, and improved efficiency if

successful. Firms must make optimal decisions on how much to explore, as

both under-exploration and over-exploration result in a loss of efficiency.

Many factors influence firms’ incentive to innovate and explore; firms fac-

ing financial markets can be especially affected by the associated disclosure

requirements. In this paper, we study the effects of disclosure policy on

how firms innovate. Specifically, we examine a myopic manager’s strategy

in choosing exploration under different disclosure requirements and de-

scribe the scenarios that foster value-maximizing innovation decisions.

We use a two-period and two-armed bandit problem as in Dye [2004] and

Manso [2011] to study a firm’s strategic decision of innovation.1The first

period represents a trial on a small scale, while the second period repre-

sents a full-scale commercial launch. There are two production methods

available for the firm to use: an old method and a new method. The old

method has been tried and tested in the past, with a known probability

of success. The probability of success of the new method, however, is un-

known. The new method is ex ante believed to be less profitable than the

old method, but the exact profitability can only be learnt after trying, and

may turn out to be higher than the profitability of the old method. The

manager of the firm can choose to explore the new method or exploit the

old method. In equilibrium, the firm decides whether to explore in the first

period—if the new method is successful, the firm will find it optimal to use

it in the second period; if the new method fails, the firm can still switch

to the old conventional method. Exploration thus provides the firm with a

real option in the second period. If the firm decides to use the old method

in the first period, it will not switch to the new method later as starting to

explore in the second period is not optimal. To capture the firm’s innova-

tion activity, we define exploration as the manager’s choice to try out the

new method in the first period.

1The bandit problem is often used to analyze problems involving the tension between ex-

ploration and exploitation, that is, trying out new arms to find the one with the highest payoff

versus using the arm with the known payoff. For a discussion of the bandit problems in more

general forms, see Bergemann and Valimaki [2006].

innovation and financial disclosure 937

The cash flows associated with the second period are a multiple of those

in the first period, capturing the potential growth from the trial stage to the

full launch of the business. While these cash flows are not directly observ-

able, the firm’s accounting system enables the manager to learn the finan-

cial outcome of the operations, whether success or failure, with a certain

probability at the end of each period.2The manager could subsequently

disclose the outcome to the investors. There are two types of disclosure

scenarios: mandated full disclosure or voluntary disclosure, with the firm

having the option to either conceal information or disclose truthfully (Dye

[1985], Jung and Kwon [1988]). These two disclosure scenarios can be in-

terpreted in terms of a firm’s status as either public or private. While a

public firm faces mandated disclosure requirements, a privately held firm

has the discretion to disclose voluntarily. In either case, after observing the

disclosure (or nondisclosure), investors form expectations about the firm’s

future cash flows. These expectations are reflected in what we refer to as the

market price. For a public firm, this price is listed on a stock exchange; for a

private firm, it would be represented by the firm’s valuation as determined

by investors. There is a short-term price formed after the first period, as well

as a long-term price after the second period. In the absence of managerial

myopic preference for price maximization, the first-best exploration strat-

egy that maximizes the long-term firm value depends on the growth of cash

flows from the first to the second period and has a simple single-threshold

feature. Intuitively, exploring the new method is only worthwhile if the po-

tential second-period profit growth is sufficiently high; otherwise, the firm

should simply choose to exploit the old method.

As a baseline, we first examine a benchmark where there is only one dis-

closure and the long-term firm’s price is perfect, that is, the price at the

end of the second period is formed after the final cash flow is realized, and

thus equals the liquidation firm value. This setup allows us to focus on the

disclosure tension in the first period to develop the basic intuition. In equi-

librium, the firm’s exploration strategy is characterized with two thresholds

(and three regions), depending on the potential growth in cash flows at

the full-scale commercial stage in the second period. At the two ends, the

manager either always explores or never explores. In the middle region,

she adopts a mixed strategy of “exploring with a probability”. In this bench-

mark setting, the manager explores strictly less than under the first-best

scenario, but is more likely to explore the new method under voluntary

disclosure than under mandated disclosure. Lacking the disclosure option

under the mandatory regime, the myopic manager favors the known old

2The financial success or failure of a new project or product is not always observed im-

mediately, especially at trial stage. Even a project that has attained technical success is not

necessarily immediately profitable. As the cost incurred and benefits received are not always

easily identified and traced to each single project in a large company with multiple business

lines, a better accounting system would allow the manager to observe the future profitability

of the individual projects more clearly.