外文文献Accounting Information, Disclosure, and the Cost of Capital

DOI: 10.1111/j.1475-679X.2007.00238.xJournal of Accounting ResearchVol. 45 No. 2 May 2007Printed in U.S.A.Accounting Information,Disclosure, and the Cost of CapitalRICHARD LAMBERT,CHRISTIAN LEUZ,AND ROBERT E. VERRECCHIAReceived 7 October 2005; accepted 18 December 2006ABSTRACTIn this paper we examine whether and how accounting information abouta firm manifests in its cost of capital, despite the forces of diversification. Webuild a model that is consistent with the Capital Asset Pricing Model andexplicitly allows for multiple securities whose cash flows are correlated. Wedemonstrate that the quality of accounting information can influence thecost of capital, both directly and indirectly. The direct effect occurs becausehigher quality disclosures affect the firms assessed covariances with otherfirms cash flows, which is nondiversifiable. The indirect effect occurs becausehigher quality disclosures affect a firms real decisions, which likely changesthe firms ratio of the expected future cash flows to the covariance of thesecash flows with the sum of all the cash flows in the market. We show that thiseffect can go in either direction, but also derive conditions under which anincrease in information quality leads to an unambiguous decline in the costof capital.1. IntroductionThe link between accounting information and the cost of capital offirms is one of the most fundamental issues in accounting. Standard settersThe Wharton School, University of Pennsylvania; Graduate School of Business, Universityof Chicago. We thank Stan Baiman, John Cochrane, Gene Fama, Wayne Guay, Raffi Indjejikian,Eugene Kandel, Christian Laux, D. J. Nanda, Haresh Sapra, Cathy Schrand, Phillip Stocken,seminar participants at the Journal of Accounting Research conference, Ohio State University, andthe University of Pennsylvania, and an anonymous referee for their helpful comments on thispaper and previous drafts of work on this topic.385Copyright C, University of Chicago on behalf of the Institute of Professional Accounting, 2007386 R. LAMBERT, C. LEUZ, AND R. E. VERRECCHIAfrequently refer to it. For example, Arthur Levitt, the former chairman ofthe Securities and Exchange Commission (SEC), suggests that “high qualityaccounting standards .reduce capital costs” (Levitt 1998, p. 81). Simi-larly, Neel Foster, a former member of the Financial Accounting StandardsBoard (FASB), claims that “More information always equates to less un-certainty, and .people pay more for certainty. In the context of financialinformation, the end result is that better disclosure results in a lower costof capital” (Foster 2003, p. 1). While these claims have intuitive appeal,there is surprisingly little theoretical work on the hypothesized link.In particular, it is unclear to what extent accounting information or firmdisclosures reduce nondiversifiable risks in economies with multiple securi-ties. Asset pricing models, such as the Capital Asset Pricing Model (CAPM),and portfolio theory emphasize the importance of distinguishing betweenrisks that are diversifiable and those that are not. Thus, the challenge foraccounting researchers is to demonstrate whether and how firms account-ing information manifests in their cost of capital, despite the forces ofdiversification.This paper examines both of these questions. We define the cost of cap-ital as the expected return on a firms stock. This definition is consistentwith standard asset pricing models in finance (e.g., Fama and Miller 1972,p. 303), as well as numerous studies in accounting that use discountedcash flow or abnormal earnings models to infer firms cost of capital (e.g.,Botosan 1997, Gebhardt, Lee, and Swaminathan 2001).1In our model,we explicitly allow for multiple firms whose cash flows are correlated. Incontrast, most analytical models in accounting examine the role of informa-tion in single-firm settings (see Verrecchia 2001 for a survey). While thisliterature yields many useful insights, its applicability to cost of capital issuesis limited. In single-firm settings, firm-specific variance is priced becausethere are no alternative securities that would allow investors to diversifyidiosyncratic risk.We begin with a model of a multisecurity economy that is consistent withthe CAPM. We then recast the CAPM, which is expressed in terms of returns,into a more easily interpreted formulation that is expressed in terms of theexpected values and covariances of future cash flows. We show that the ratioof the expected future cash flow to the covariance of the firms cash flowwith the sum of all cash flows in the market is a key determinant of the costof capital.Next, we add an information structure that allows us to study the effectsof accounting information. We characterize firms accounting reports asnoisy information about future cash flows, which comports well with actualreporting behavior. We demonstrate that accounting information influencesa firms cost of capital in two ways: (1) direct effectswhere higher quality1We also discuss the impact of information on price, as the latter is sometimes used as ameasure of cost of capital. See, for example, Easley and OHara 2004 and Hughes, Liu, andLiu 2005.INFORMATION AND THE COST OF CAPITAL 387accounting information does not affect cash flows per se, but affects themarket participants assessments of the distribution of future cash flows; and(2) indirect effectswhere higher quality accounting information affectsa firms real decisions, which, in turn, influence its expected value andcovariances of firm cash flows.In the first category, we show (not surprisingly) that higher quality infor-mation reduces the assessed variance of a firms cash flow. Analogous to thespirit of the CAPM, however, we show this effect is diversifiable in a “largeeconomy.” We discuss what the concept of “diversification” means, and showthat an economically sensible definition requires more than simply examin-ing what happens when the number of securities in the economy becomeslarge.Moreover, we demonstrate that an increase in the quality of a firms dis-closure about its own future cash flows has a direct effect on the assessedcovariances with other firms cash flows. This result builds on and extendsthe work on “estimation risk” in finance.2In this literature, information typ-ically arises from a historical time series of return observations. In particular,Barry and Brown 1985 and Coles, Loewenstein, and Suay 1995 comparetwo information environments: In one environment the same amount of in-formation (e.g., the same number of historical time-series observations) isavailable for all firms in the economy, whereas in the other information en-vironment there are more observations for one group of firms than another.They find that the betas of the “high information” securities are lower thanthey would be in the equal information case. They cannot unambiguouslysign, however, the difference in betas for the “low information” securities inthe unequal- versus equal-information environments. Moreover, these stud-ies do not address the question of how an individual firms disclosures caninfluence its cost of capital within an unequal information environment.Rather than restricting attention to information as historical observationsof returns, our paper uses a more conventional information-economics ap-proach in which information is modeled as a noisy signal of the realizationof cash flows in the future. With this approach, we allow for more generalchanges in the information environment, and we are able to prove muchstronger results. In particular, we show that higher quality accounting infor-mation and financial disclosures affect the assessed covariances with otherfirms, and this effect unambiguously moves a firms cost of capital closerto the risk-free rate. Moreover, this effect is not diversifiable because it ispresent for each of the firms covariance terms and hence does not disap-pear in “large economies.”Next, we discuss the effects of disclosure regulation on the cost of capi-tal of firms. Based on our framework, increasing the quality of mandateddisclosures should in general move the cost of capital closer to the risk-free2See Brown 1979, Barry and Brown 1984, 1985, Coles and Loewenstein 1988, andColes, Loewenstein, and Suay 1995.388 R. LAMBERT, C. LEUZ, AND R. E. VERRECCHIArate for all firms in the economy. In addition to the effect of an individualfirms disclosures, there is an externality from the disclosures of other firms,which may provide a rationale for disclosure regulation. We also argue thatthe magnitude of the cost-of-capital effect of mandated disclosure is unequalacross firms. In particular, the reduction in the assessed covariances betweenfirms and the market does not result in a decrease in the beta coefficientof each firm. After all, regardless of information quality in the economy, theaverage beta across firms has to be 1.0. Therefore, even though firms cost ofcapital (and the aggregate risk premium) declines with improved mandateddisclosure, their beta coefficients need not.In the “indirect effect” category, we show that the quality of accountinginformation influences a firms cost of capital through its effect on a firmsreal decisions. First, we demonstrate that if better information reduces theamount of firm cash flow that managers appropriate for themselves, theimprovements in disclosure not only increase firm price, but in generalalso reduce a firms cost of capital. Second, we allow information qualityto change a firms real decisions, for example, with respect to productionor investment. In this case, information quality changes decisions, whichchanges the ratio of expected cash flow to nondiversifiable covariance riskand hence influences a firms cost of capital. We derive conditions underwhich an increase in information quality results in an unambiguous decreasein a firms cost of capital.Our paper makes several contributions. First, we extend and generalizeprior work on estimation risk. We show that information quality directly in-fluences a firms cost of capital and that improvements in information qualityby individual firms unambiguously affect their nondiversifiable risks. Thisfinding is important as it suggests that a firms beta factor is a function of itsinformation quality and disclosures. In this sense, our study provides theoret-ical guidance to empirical studies that examine the link between firms dis-closures and/or information quality, and their cost of capital (e.g., Botosan1997, Botosan and Plumlee 2002, Francis et al. 2004, Ashbaugh-Skaifeet al. 2005, Berger, Chen, and Li 2005, Core, Guay, and Verdi 2006).We discuss this guidance in more detail in our conclusion. In addition,our study provides an explanation for international differences in the eq-uity risk premium or firms average cost of equity capital, stemming fromdifferences in disclosure regulation across countries (e.g., Hail and Leuz2006).It is important to recognize, however, that the effects of a firms disclo-sures on its cost of capital, as demonstrated by our model, are fully capturedby an appropriately specified, forward-looking beta and the expected returnon the market as a whole.3Thus, our model does not provide support for an3In this sense, our paper also relates to empirical studies on the conditional CAPM, exam-ining whether the CAPM holds conditionally on time t information and whether measurementerror in the factor moments may explain the role of the Fama-French factors (e.g., Jagannathanand Wang 1996, Lettau and Ludvigson 2001).INFORMATION AND THE COST OF CAPITAL 389additional risk factor capturing “information risk.”4One way to justify theinclusion of additional information variables in a cost of capital model is tonote that empirical proxies for beta, which for instance are based on histori-cal data alone, may not capture all information effects. In this case, however,it is incumbent on researchers to specify a “measurement error” model or,at least, provide a careful justification for the inclusion of information vari-ables, and their functional form, in the empirical specification. Based onour results, however, the most natural way to empirically analyze the linkbetween information quality and the cost of capital is via the beta factor.5A second contribution of our paper is that it provides a direct link be-tween information quality and the cost of capital, without reference tomarket liquidity. Prior work suggests an indirect link between disclosureand firms cost of capital based on market liquidity and adverse selectionin secondary markets (e.g., Diamond and Verrecchia 1991, Baiman andVerrecchia 1996, Easley and OHara 2004). These studies, however, an-alyze settings with a single firm (or settings where cash flows across firmsare uncorrelated). Thus, it is unclear whether the effects demonstrated inthese studies survive the forces of diversification and extend to more gen-eral multisecurity settings. We emphasize, however, that we do not disputethe possible role of market liquidity for firms cost of capital, as several em-pirical studies suggest (e.g., Amihud and Mendelson 1986, Chordia, Roll,and Subrahmanyam 2001, Easley, Hvidkjaer, and OHara 2002, Pastorand Stambaugh 2003). Our paper focuses on an alternative, and possiblymore direct, explanation as to how information quality influences nondi-versifiable risks.Finally, our paper contributes to the literature by showing that informa-tion quality has indirect effects on real decisions, which in turn manifest infirms cost of capital. In this sense, our study relates to work on real effectsof accounting information (e.g., Kanodia et al. 2000, Kanodia, Sapra, andVenugopalan 2004). These studies, however, do not analyze the effects onfirms cost of capital or nondiversifiable risks.The remainder of this paper is organized as follows. Section 2 sets up thebasic model in a world of homogeneous beliefs, defines terms, and derivesthe determinants of the cost of capital. Sections 3 and 4 analyze the directand indirect effects of accounting information on firms cost of capital,respectively. Section 5 summarizes our findings and concludes the paper.2. Model and Cost of Capital DerivationWe define cost of capital to be the expected return on the firms stock.Consistent with standard models of asset pricing, the expected rate of re-turn on a firm js stock is the rate, Rj, that equates the stock price at the4Note that our model does not preclude the existence of such an additional risk factor inan extended or different model. This issue is left for future research.5See, for example, Beaver, Kettler, and Scholes 1970 and Core, Guay, and Verdi 2006for an empirical analysis that relates accounting information to a firms beta.390 R. LAMBERT, C. LEUZ, AND R. E. VERRECCHIAbeginning of the period, Pj, to the cash flow at the end of the period,Vj: Pj(1 +Rj) =Vj,orRj=VjPjPj. Our analysis focuses on the expectedrate of return, which is E (Rj| Phi1) =E (Vj| Phi1)PjPj, where is the informationavailable to market participants to make their assessments regarding thedistribution of future cash flows.We assume there are J securities in the economy whose returns are corre-lated. The best known model of asset pricing in such a setting is the CAPM(Sharpe 1964, Lintner 1965). Therefore, we begin our analysis by pre-senting the conventional formulation of the CAPM, and then transformthis formulation and add an information structure to show how informa-tion quality affects expected returns. Assuming that returns are normallydistributed or, alternatively, that investors have quadratic utility functions,the CAPM expresses the expected return on a firms stock as a function ofthe risk-free rate, Rf, the expected return on the market, E (Rm), and thefirms beta coefficient, j:E (Rj| Phi1) = Rf+ E (RM| Phi1) Rfj= Rf+E (RM| Phi1) RfVar(RM| Phi1)Cov(Rj,Rm| Phi1). (1)Equation (1) shows that the only firm-specific parameter that affects thefirms cost of capital is its beta coefficient, or, more specifically, the covari-ance of its future return with that of the market portfolio. This covarianceis a forward-looking parameter, and it is based on the information avail-able to market participants. Consistent with the conventional formulationof the CAPM, we assume market participants possess homogeneous beliefsregarding the expected end-of-period cash flows and covariances.Because the CAPM is expressed solely in terms of covariances, this for-mulation might be interpreted as implying that other factors, for example,the expected cash flows, do not affect the firms cost of capital. It is im-portant to keep in mind, however, that the covariance term in the CAPMis expressed in terms of returns, not in terms of cash flows. The two are re-lated via the equation Cov(Rj,RM) = Cov(VjPj,VMPM) =1PjPMCov(Vj,VM).This expression implies that information can affect the expected return on afirms stock through its effect on inferences about the covariances of futurecash flows, or through the current period stock price, or both. Clearly thecurrent stock price is a function of the expected-end-of-period cash flow.In particular, the CAPM can be re-expressed in terms of prices instead ofreturns as follows (see Fama 1976, equation (83):Pj=E (Vj| Phi1) E (VM| Phi1) (1 + Rf)PMVar(VM| Phi1)bracketleftBiggCovparenleftBiggVj,Jsummationdisplayk=1VkvextendsinglevextendsinglevextendsinglevextendsinglevextendsinglePhi1parenrightBiggbracketrightBigg(1 + Rf),j = 1,., J . (2)INFORMATION AND THE COST OF CAPITAL 391Equation (2) indicates that the current price of a firm can be expressedas the expected end-of-period cash flow minus a reduction for risk. Thisrisk-adjusted expected value is then discounted to the beginning of theperiod at the risk-free rate. The risk reduction factor in the numerator ofequa