Skip to main content
Log in

Uncertainty about managerial horizon and voluntary disclosure

  • Published:
Review of Accounting Studies Aims and scope Submit manuscript

Abstract

I examine the relation between investors’ uncertainty about managerial horizon and corporate voluntary disclosure. I argue that investors’ uncertainty about the manager’s short-term stock price concern works as a disclosure friction that allows for a partial disclosure equilibrium. When investors are uncertain about managers’ horizons, short-horizon managers can withhold bad news by pooling with long-horizon managers who are not motivated to disclose, regardless of the content of the news, as they are largely indifferent to short-term stock prices. Based on this theoretical framework, I hypothesize that reducing investors’ uncertainty about managerial horizon reduces this disclosure friction, thereby pressuring short-horizon managers to provide more voluntary disclosure. I use the executive compensation disclosure mandate as an empirical setting that reduced investors’ uncertainty about managerial horizon. Employing a difference-in-differences research design, I find a significant increase in voluntary disclosure for treated firms relative to control firms, largely driven by firms with short-horizon managers.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Notes

  1. For instance, Murphy (2013) provides an overview of the literature that links reporting choices to CEO cash and equity compensation incentives. In a similar vein, Dichev, Graham, Harvey, and Rajgopal (2013) find that the vast majority of executives believe that executive pay affects firms’ reporting choices. More recently, Edmans, Fang, and Lewellen (2017) and Edmans, Goncalves-Pinto, Groen-Xu, and Wang (2018) identify executives’ short-term concerns using the amount of stocks and options scheduled to vest in a given period. Similarly, Gopalan, Milbourn, Song, and Thakor (2014) construct a measure of executive pay duration that reflects the vesting periods of different pay components to quantify the extent to which executive compensation is short-term.

  2. Throughout the analysis, I measure voluntary disclosure using indicator variables because this comports with the theoretical framework that models the likelihood of disclosure. My inferences remain unchanged when I use the frequency of earnings forecasts instead.

  3. For example, Anilowski, Feng, and Skinner (2007) find that the value-weighted proportion of forecasters reach approximately 50 percent in 2004.

  4. Following the CD&A adoption by the control firms in FY 2008, they are no longer a useful control group. For this reason, FY 2007 is the only fiscal year during which a clean variation exists in the firms’ treatment status. It is unclear ex-ante what the difference-in-differences effect of the treatment would look like for the treatment firms, relative to the control firms, in FY 2008 and afterward. The direction and magnitude of the effect will depend on whether the effect of the treatment stays constant, strengthens, or weakens over time.

  5. I define the control sample as firms with June to November fiscal year-ends to maximize the control sample size (a total of 171 firm-years). My inferences remain unchanged when I instead define the control sample as firms with September to November fiscal year-ends (a total of 100 firm-years).

  6. My inferences are robust to requiring a balanced sample.

  7. Because ISS Incentive Lab does not report the date an executive started working as a CEO and because the dataset starts in 1998, I can only start cumulating an executive’s tenure from 1998. This implies that my measure of tenure is potentially understated.

  8. While my approach almost precisely mimics that of Edmans et al. (2017), there are two major differences. First, I rely on ISS Incentive Lab database, due to data availability on WRDS, whereas Edmans et al. (2017) use Equilar database that they purchased. Equilar has a wider cross-sectional coverage inclusive of Russell 3000 firms, whereas ISS Incentive Lab covers around 2,000 US companies (around 1,200 active) including S&P 500 firms and a significant portion of the S&P 400. This biases my sample toward larger and more visible firms. Second, I calculate annualized Vesting measure, whereas Edmans et al. (2017) impose additional assumptions to calculate a decomposed Vesting measure at a quarterly-level.

  9. Vesting captures the ex-post sensitivity of equity that vests based on the number of options with a particular exercise price and expiration date that change from “unvested” to “vested” status, plus the number of shares that vest during that fiscal year. In that sense, it is similar to the measure proposed by Edmans et al. (2017) and does not require additional assumptions about vesting schedule. This differs from the ex-ante measure proposed by Edmans et al. (2018), which infers vesting sensitivity at a monthly frequency by tracking stock and option grants over the past multiple years and imposing assumptions about the detailed vesting schedules.

  10. In particular, I calculate Vesting for each firm-executive pair for FY 2007 for the treatment sample of firms and for FY 2008 for the control sample of firms, as these are the earliest possible years with sufficient data to calculate the variable. Then I merge the resulting Vesting values with my sample of firm-years, conditional on them having the same CIK identifier and the same CEO participant ID. By extrapolating post-period Vesting values to the pre-period observations, I am implicitly assuming that managerial horizon remains relatively stable across this short window of three years.

  11. Even though many CEOs hold already-vested equity, this could be because they face implicit or explicit constraints on selling it, and vesting could alleviate such restrictions. For instance, the CEO might be restricted by the corporate ownership guideline, for which only the vested equity counts. In this case, additional vesting can alleviate selling constraints and allow the CEO to sell some of the equity she owns. Furthermore, the CEO may voluntarily hold equity to maintain voting rights. In this case, additional vesting will allow the CEO to sell equity while preserving her voting rights. Therefore it seems reasonable to assume that equity vesting leads to equity sales and thus induces the CEOs to become more concerned about maximizing short-term stock prices.

  12. This approach is similar to that of Hutton, Lee, and Shu (2012), where a firm-year is defined as a loss year when the management’s EPS forecast is negative. Li (2008) takes a similar approach and defines a bad-earnings news using a net income loss indicator.

  13. An advantage of defining a bad-news forecast relative to a zero cutoff is that it is defined based on an absolute benchmark, rather than a relative one like investor consensus. It is critical that I define a bad-news forecast based on an absolute benchmark in this setting, because the reduction in investors’ uncertainty about managerial horizon affects disclosure threshold by influencing investor consensus in the absence of disclosure. In other words, defining a bad-news forecast based on a relative benchmark like investor consensus would be problematic, because the theory predicts that firms will always only disclose information that exceeds investor consensus.

  14. Looking at mean values of Forecast, short-horizon firms are more likely to issue a forecast compared to long-horizon companies, but the differences are not statistically significant. Since my theoretical premise requires that short-horizon managers are more motivated to issue a forecast than long-horizon managers, I formally test this conjecture by regressing Forecast on the quintile ranks of Vesting and a set of control variables (untabulated). Consistent with the theoretical intuition, I find a positive and significant coefficient estimate on the ranked Vesting variable.

  15. The two coefficients on the interaction terms in columns (2) and (3) are statistically different at a 1% level.

  16. The two coefficients on the interaction terms in columns (2) and (3) are not statistically different (p-value 0.287).

  17. The two coefficients on the interaction terms in columns (2) and (3) are statistically different at a 10% level.

  18. The two coefficients on the interaction terms in columns (2) and (3) are statistically different at a 5% level.

  19. The two coefficients on the interaction terms in columns (2) and (3) are statistically different at a 1% level.

  20. The two coefficients on the interaction terms in columns (2) and (3) are statistically different at a 5% level.

  21. The two coefficients on the interaction terms in columns (2) and (3) are statistically different at a 5% level.

  22. In sum, the covariates include firm size (Size), market-to-book ratio (Mtb), return-on-assets (Roa), an indicator variable for loss (Loss), special items (Special), leverage (Leverage), earnings volatility (Earnvol), return volatility (Retvol), executive tenure (Tenure), institutional ownership (Instown), the interaction terms for these variables interacted with a short-horizon indicator variable (ShortHorizonDummy), and the short-horizon indicator variable itself (ShortHorizonDummy).

  23. The two coefficients on the interaction terms are statistically different at a 1% level.

  24. The two coefficients on the interaction terms are statistically different at a 5% level.

  25. The two coefficients on the interaction terms are statistically different at a 1% level.

  26. The two coefficients on the interaction terms are statistically different at a 5% level.

  27. The two coefficients on the interaction terms are statistically different at a 5% level.

  28. Another possible solution is \({t}_{U}^{*}=\frac{N\delta -M-\sqrt{{N}^{2}{\delta }^{2}+NM+{M}^{2}}}{N}\). However, since this solution is smaller than 0, it describes an unrealistic equilibrium where all realized outcomes are voluntarily disclosed. Therefore I focus on the equilibrium solution Eq. (8) in my analysis.

  29. To see this, consider a stock grant \(i{^{\prime}}\) that vests equally over \({t}_{i{^{\prime}}}\) years. Since a fraction \(1/{t}_{i{^{\prime}}}\) of the grant is vested each year, the term \(Restricted stoc{k}_{i}\times {t}_{i}\) in the Duration calculation equation above should be replaced by \(Restricted stoc{k}_{i{^{\prime}}}\times \left(\frac{1}{{t}_{i{^{\prime}}}}+\frac{2}{{t}_{i{^{\prime}}}}+\cdots +\frac{{t}_{i{^{\prime}}}}{{t}_{i{^{\prime}}}}\right)=\frac{Restricted stoc{k}_{i{^{\prime}}}}{{t}_{i{^{\prime}}}}\times \frac{{t}_{{i}^{^{\prime}}}({t}_{{i}^{^{\prime}}}+1)}{2}=Restricted stoc{k}_{i{^{\prime}}}\times \frac{{t}_{i{^{\prime}}}+1}{2}\); duration for option grants with ratable vesting schedules can be calculated in the same way. See also, FN 10 of Gopalan et al. (2014).

References

  • Alford, Andrew W., and Philip G. Berger. (1999). A simultaneous equations analysis of forecast accuracy, analyst following, and trading volume. Journal of Accounting, Auditing & Finance 14 (3): 219–240.

    Article  Google Scholar 

  • Anilowski, Carol, Mei Feng, and Douglas J. Skinner. (2007). Does earnings guidance affect market returns? The nature and information content of aggregate earnings guidance. Journal of Accounting and Economics 44 (1–2): 36–63.

    Article  Google Scholar 

  • Balakrishnan, Karthik, Mary B. Billings, Bryan Kelly, and A. Alexander Ljungqvist. (2014). Shaping liquidity: On the causal effects of voluntary disclosure. The Journal of Finance 69 (5): 2237–2278.

    Article  Google Scholar 

  • Bamber, Linda S., and Youngsoon S. Cheon. (1998). Discretionary management earnings forecast disclosures: Antecedents and outcomes associated with forecast venue and forecast specificity choices. Journal of Accounting Research 36 (2): 167–190.

    Article  Google Scholar 

  • Bertomeu, Jeremy, Paul Ma, and Iván. Marinovic. (2020). How often do managers withhold information? The Accounting Review 95 (4): 73–102.

    Article  Google Scholar 

  • Boone, Audra L., and Joshua T. White. (2015). The effect of institutional ownership on firm transparency and information production. Journal of Financial Economics 117 (3): 508–533.

    Article  Google Scholar 

  • Bushee, Brian J. (2001). Do institutional investors prefer near-term earnings over long-run value? Contemporary Accounting Research 18 (2): 207–246.

    Article  Google Scholar 

  • Dichev, Ilia D., John R. Graham, Campbell R. Harvey, and Shiva Rajgopal. (2013). Earnings quality: Evidence from the field. Journal of Accounting and Economics 56 (2–3): 1–33.

    Article  Google Scholar 

  • Dye, Ronald A. (1985). Disclosure of nonproprietary information. Journal of Accounting Research 23 (1): 123–145.

    Article  Google Scholar 

  • Edmans, Alex, Luis Goncalves-Pinto, Moqi Groen-Xu, and Yanbo Wang. (2018). Strategic news releases in equity vesting months. The Review of Financial Studies 31 (11): 4099–4141.

    Article  Google Scholar 

  • Edmans, Alex, Vivian W. Fang, and Katharina A. Lewellen. (2017). Equity vesting and investment. The Review of Financial Studies 30 (7): 2229–2271.

    Article  Google Scholar 

  • Ferri, Fabrizio, Ronghuo Zheng, and Yuan Zou. (2018). Uncertainty about managers’ reporting objectives and investors’ response to earnings reports: Evidence from the 2006 executive compensation disclosures. Journal of Accounting and Economics 66 (2–3): 339–365.

    Article  Google Scholar 

  • Fischer, Paul E., and Robert E. Verrecchia. (2000). Reporting bias. The Accounting Review 75 (2): 229–245.

    Article  Google Scholar 

  • Gipper, Brandon. (2020). The economic effects of expanded compensation disclosures. Journal of Accounting and Economics, forthcoming.

  • Glaeser, Stephen, Jeremy Michels, and Robert E. Verrecchia. (2020). Discretionary disclosure and manager horizon: Evidence from patenting. Review of Accounting Studies 25 (2): 597–635.

    Article  Google Scholar 

  • Gopalan, Radhakrishnan, Todd Milbourn, Fenghua Song, and Anjan V. Thakor. (2014). Duration of executive compensation. The Journal of Finance 69 (6): 2777–2817.

    Article  Google Scholar 

  • Hirst, D. Eric., Lisa Koonce, and Shankar Venkataraman. (2008). Management earnings forecasts: A review and framework. Accounting Horizons 22 (3): 315–338.

    Article  Google Scholar 

  • Hutton, Amy P., Lian Fen Lee, and Susan Z. Shu. (2012). Do managers always know better? The relative accuracy of management and analyst forecasts. Journal of Accounting Research 50 (5): 1217–1244.

    Article  Google Scholar 

  • Jensen, Michael C., and William H. Meckling. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3 (4): 305–360.

    Article  Google Scholar 

  • Jung, Woon-Oh., and Young K. Kwon. (1988). Disclosure when the market is unsure of information endowment of managers. Journal of Accounting Research 26 (1): 146–153.

    Article  Google Scholar 

  • Lang, Mark, and Russell Lundholm. (1993). Cross-sectional determinants of analyst ratings of corporate disclosures. Journal of Accounting Research 31 (2): 246–271.

    Article  Google Scholar 

  • Li, Feng. (2008). Annual report readability, current earnings, and earnings persistence. Journal of Accounting and Economics 45 (2–3): 221–247.

    Article  Google Scholar 

  • Miller, Gregory S. (2002). Earnings performance and discretionary disclosure. Journal of Accounting Research 40 (1): 173–204.

    Article  Google Scholar 

  • Murphy, Kevin J. (2013). Executive compensation: Where we are, and how we got there. Handbook of the Economics of Finance 2: 211–356.

    Article  Google Scholar 

  • Nagar, Venky, Dhananjay Nanda, and Peter Wysocki. (2003). Discretionary disclosure and stock-based incentives. Journal of Accounting and Economics 34 (1–3): 283–309.

    Article  Google Scholar 

  • Pownall, Grace, Charles Wasley, and Gregory Waymire. (1993). The stock price effects of alternative types of management earnings forecasts. The Accounting Review 68 (4): 896–912.

    Google Scholar 

  • Rogers, Jonathan L., and Phillip C. Stocken. (2005). Credibility of management forecasts. The Accounting Review 80 (4): 1233–1260.

    Article  Google Scholar 

  • Sinha, Nishi, and Dov Fried. (2008). Clustered disclosures by competing firms: The choice of fiscal year-ends. Journal of Accounting, Auditing & Finance 23 (4): 493–516.

    Article  Google Scholar 

  • Smith, David B., and Susan Pourciau. (1988). A comparison of the financial characteristics of December and non-December year-end companies. Journal of Accounting and Economics 10 (4): 335–344.

    Article  Google Scholar 

  • Suijs, Jeroen. (2007). Voluntary disclosure of information when firms are uncertain of investor response. Journal of Accounting and Economics 43 (2–3): 391–410.

    Article  Google Scholar 

  • Verrecchia, Robert E. (1983). Discretionary disclosure. Journal of Accounting and Economics 5: 179–194.

    Article  Google Scholar 

  • Waymire, Gregory. (1985). Earnings volatility and voluntary management forecast disclosure. Journal of Accounting Research 23 (1): 268–295.

    Article  Google Scholar 

Download references

Acknowledgements

I thank seminar participants at The Wharton School, Russell Lundholm (editor), Jeremy Michels, Daniel Taylor, Robert Verrecchia, and an anonymous referee for helpful comments.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Jung Min Kim.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix A: A model of voluntary disclosure

This appendix illustrates a simple model of voluntary disclosure based on the Dye (1985) and Jung and Kwon (1988) disclosure model. I assume an economy of \(N\) firms with short-horizon managers and \(M\) firms with long-horizon managers. Each firm’s earnings (\(\tilde{x }\)) are uniformly distributed in between 0 and 1 (i.e., \(x\sim Unif(\mathrm{0,1})\)). The manager can privately observe the realization of \(x\) and choose to disclose it. Disclosing the realization of \(x\) will impose disclosure cost of \(\delta\) on the firm. This cost can be thought of as the cost of preparing and disseminating information or the cost associated with disclosing information which may be proprietary in nature.

I assume that short-horizon managers aim to maximize current stock price and thus are motivated to disclose as long as the realized outcome is favorable. On the contrary, long-horizon managers are indifferent to short-term stock price movements, and they are not motivated to disclose, regardless of the realized values. In equilibrium, only the short-horizon managers will disclose so long as the realized outcome net of disclosure costs is greater than the expected stock price in the event of nondisclosure. The expected stock price in the event of nondisclosure differs depending on the degree of investors’ uncertainty about managerial horizon. Below I illustrate how the disclosure threshold and the likelihood of disclosure changes as investors’ uncertainty about managerial horizon changes.

When investors are uncertain about managerial horizon

When investors are completely uncertain about managerial horizon, they will value a firm in the event of nondisclosure at the same value, regardless of whether the firm is actually a short- or a long-horizon firm. This valuation depends on their prior beliefs about the earnings distributions and the proportion of short- versus long-horizon firms in the economy. Given this valuation, a long-horizon manager does not disclose, regardless of the realization of \(x\). In contrast, a short-horizon manager discloses only if disclosing yields a higher current stock price than withholding. In other words, a short-horizon manager will disclose if and only if

$$x-\delta\geq \mathrm{Pr}{(i=SH\vert\left\{m=SH\right\}\cap\{d=ND\})}\times\frac{\mathit t}{2}+\mathrm{Pr}{(i=LH\vert\left\{m=SH\right\}\cap\{d=ND\})}\times\frac{1}{2}.$$
(4)

\(i\in \{SH,LH\}\) describes the investor’s assessment of the managerial horizon, where SH represents a short-horizon manager and LH represents a long-horizon manager. \(m\in \{SH,LH\}\) describes the manager’s true horizon. \(d\in \{D,ND\}\) describes the disclosure status, where D represents disclosure and ND represents nondisclosure. \(t\) describes the threshold level of disclosure, and \({t}_{U}^{*}\) describes the short-horizon manager’s equilibrium disclosure threshold when investors are uncertain about managerial horizon. \({t}_{U}^{*}\) will be set a point that satisfies the following equation.

$${t}_{U}-\delta =\mathrm{Pr}\left(i=SH|\left\{m=SH\right\}\cap \left\{d=ND\right\}\right)\times \frac{{t}_{U}}{2}+\mathrm{Pr}\left(i=LH|\left\{m=SH\right\}\cap \left\{d=ND\right\}\right)\times \frac{1}{2}.$$
(5)

Investors’ uncertainty about managerial horizon manifests through the two probabilities in equation (5), \(\mathrm{Pr}\left(i=SH|\left\{m=SH\right\}\cap \{d=ND\}\right)\) and \(\mathrm{Pr}\left(i=LH|\left\{m=SH\right\}\cap \{d=ND\}\right)\). When investors are completely uncertain about managerial horizon, in the event of nondisclosure, investors will assess a firm as a short-horizon firm with the same probability, regardless of whether the firm is actually a short-horizon firm or not. This can be expressed as follows.

$$\mathrm{Pr}\left(i=SH|\left\{m=SH\right\}\cap \{d=ND\}\right)=\mathrm{Pr}\left(i=SH|d=ND\right)=\frac{{t}_{U}N}{{t}_{U}N+M}.$$
(6)
$$\mathrm{Pr}{(i=LH\vert\left\{m=SH\right\}\cap\{d=ND\})}=\mathrm{Pr}{(i=LH\vert d=ND)}=\frac{\mathit M}{{\mathit t}_{\mathit U} N+ M}.$$
(7)

Note from equations (6) and (7) that conditioning on whether a firm is actually a short-horizon firm (\(m=SH\)) has no effect on investors’ probabilistic assessment, as investors are completely uncertain about managerial horizon. Substituting (6) and (7) into equation (5) and solving for \({t}_{U}^{*}\), I getFootnote 28

$$t_{\mathit U}^{\ast}=\frac{\mathit N\delta\mathit-\mathit M+\sqrt{ N^{\mathit2}\delta^{\mathit2}\mathit+\mathit N\mathit M\mathit+\mathit M^{\mathit2}}}{\mathit N}.$$
(8)

For \({t}_{U}^{*}\) to be an interior disclosure equilibrium, it must satisfy \(0<{t}_{U}^{*}<1\). This requires an additional assumption that \(\delta <\frac{1}{2}\). In other words, a sufficient condition for the existence of a unique, interior \({t}_{U}^{*}\) is that the disclosure costs be “relatively small”. To facilitate the discussion, henceforth I assume that \(\delta <\frac{1}{2}\). In equilibrium, \((1-{t}_{U}^{*})N\) short-horizon firms with earnings realization (\(x\)) above \({t}_{U}^{*}\) will disclose.

When investors’ uncertainty about managerial horizon decreases

Next I examine how the disclosure equilibrium changes when the investors’ uncertainty about managerial horizon goes down. Colloquially speaking, reduction in the investors’ uncertainty about managerial horizon means that, when a short-horizon firm withholds information, investors are more (less) likely to accurately (inaccurately) assess the firm as a short-horizon (long-horizon) company than if they were completely uncertain about managerial horizon. I denote the reduction in investors’ uncertainty about managerial horizon by \(\varepsilon\). Now, the investors’ probabilistic beliefs (\({\mathrm{Pr}}^{C}(\bullet )\)) can be expressed as follows.

$${\mathrm{Pr}}^{ C}{(i=SH\vert\left\{m=SH\right\}\cap\{d=ND\})}=\frac{ t_{ U}^{\ast} N}{ t_{ U}^{\ast} N+ M}+\varepsilon,$$
(9)
$${\mathrm{Pr}}^{C}\left(i=LH|\left\{m=SH\right\}\cap \{d=ND\}\right)=\frac{M}{{t}_{U}^{*}N+M}-\varepsilon,$$
(10)

where \(\varepsilon >0\), \(0\le \frac{{t}_{U}^{*}N}{{t}_{U}^{*}N+M}+\varepsilon \le 1\), and \(0\le \frac{M}{{t}_{U}^{*}N+M}-\varepsilon \le 1\).

It is straightforward to show that \({\mathrm{Pr}}^{C}\left(i=SH|\left\{m=SH\right\}\cap \{d=ND\}\right)>{\mathrm{Pr}}^{C}\left(i=SH|d=ND\right)\) and that \({\mathrm{Pr}}^{C}\left(i=LH|\left\{m=SH\right\}\cap \{d=ND\}\right)<{\mathrm{Pr}}^{C}\left(i=LH|d=ND\right)\), which reflects the decrease in investors’ uncertainty about managerial horizon.

Similar to equation (5), the new equilibrium threshold level of disclosure \({t}_{C}^{*}\) will be set a point that satisfies the following equation.

$${t}_{C}-\delta ={\mathrm{Pr}}^{C}\left(i=SH|\left\{m=SH\right\}\cap \{d=ND\}\right)\times \frac{{t}_{C}}{2}+{\mathrm{Pr}}^{C}\left(i=LH|\left\{m=SH\right\}\cap \{d=ND\}\right)\times \frac{1}{2}.$$
(11)

I substitute equations (8), (9) and (10) into equation (11) and solve for \({t}_{C}^{*}\), the short-horizon manager’s threshold level of disclosure when investors are less uncertain about managerial horizon.

$${t}_{C}^{*}=\frac{\left(-2\delta +\varepsilon \right)N\delta -M+(-2\delta +\varepsilon )\sqrt{{N}^{2}{\delta }^{2}+NM+{M}^{2}}}{\left(-1+\varepsilon \right)N\delta -M+(-1+\varepsilon )\sqrt{{N}^{2}{\delta }^{2}+NM+{M}^{2}}}.$$
(12)

It is straightforward to show that \({t}_{C}^{*}={t}_{U}^{*}\) when \(\varepsilon =0\). In addition, given the earlier assumption that \(\delta <\frac{1}{2}\), one can verify that \(\frac{d{t}_{C}^{*}}{d\varepsilon }<0\). Therefore \({t}_{C}^{*}<{t}_{U}^{*}\) when \(\varepsilon >0\), and the equilibrium threshold level of disclosure decreases, as investors’ uncertainty about managerial horizon decreases (\(\varepsilon\) increases). In equilibrium, \((1-{t}_{C}^{*})N\)(\(>(1-{t}_{U}^{*})N\)) short-horizon firms with earnings realization (\(x\)) above \({t}_{C}^{*}\) will disclose.

Predictions

I make three predictions based on the equilibrium results.

First, as investors’ uncertainty about managerial horizon decreases, voluntary disclosure by the short-horizon firms increases.

Second, as investors’ uncertainty about managerial horizon decreases, more unfavorable news is disclosed by the short-horizon firms.

Third, as investors’ uncertainty about managerial horizon decreases, investors apply a greater discount on short-horizon firms and reduce the discount on long-horizon firms.

The first and second predictions can be expressed as \(\left(1-{t}_{C}^{*}\right)>\left(1-{t}_{U}^{*}\right)\) and \(E\left(x|x\ge {t}_{C}^{*}\right)<E\left(x|x\ge {t}_{U}^{*}\right)\), and they follow from the equilibrium result that \({t}_{C}^{*}<{t}_{U}^{*}\). These predictions suggest that the reduction in investors’ uncertainty about managerial horizon will pressure the short-horizon firms to provide more voluntary disclosure and reveal more of the bad news they previously had been withholding.

The third prediction has two parts. The first part states that investors discount firm value for the short-horizon firms following the decrease in the uncertainty about managerial horizon. This can be expressed as follows.

$${(\frac{1+t_U^\ast}2-\delta)}{(1-t_U^\ast)}+{(t_U^\ast-\delta)}t_{ U}^{\ast}>{(\frac{1+t_C^\ast}2-\delta)}{(1-t_C^\ast)}+{(t_C^\ast-\delta)}t_{ C}^{\ast}.$$
(13)

The left-hand-side of the inequality describes the investors’ average valuation of short-horizon firms when they are completely uncertain about managerial horizon. Similarly, the right-hand-side describes the investors’ average valuation of short-horizon firms when their uncertainty decreases. The inequality in (13) can be rearranged to yield \(\frac{1}{2}\left({t}_{U}^{*}+{t}_{C}^{*}\right)\left({t}_{U}^{*}-{t}_{C}^{*}\right)>0\), which holds because \({t}_{C}^{*}<{t}_{U}^{*}\). The second part of the third prediction states that investors reduce the discount for the long-horizon firms, as these firms are less likely to be pooled with nondisclosing short-horizon firms when investors are less uncertain about managerial horizon. Because the weighted average expected value of \(\tilde{x }\) across the two types of firms has to equal the unconditional expected value of \(\frac{1}{2}\), the second part of the third prediction can be expressed as follows.

$$\frac{\mathit M\mathit+\mathit N}{\mathit2\mathit M}-\frac{\mathit N}{\mathit M}{(\left(\frac{1+t_U^\ast}2\right)\left(1-t_U^\ast\right)+\left(t_U^\ast-\delta\right)t_U^\ast)}<\frac{\mathit M\mathit+\mathit N}{2 M}-\frac{\mathit N}{\mathit M}{(\left(\frac{1+t_C^\ast}2\right)\left(1-t_C^\ast\right)+\left(t_C^\ast-\delta\right)t_C^\ast)}.$$
(14)

The left-hand-side of the inequality describes the investors’ average valuation of long-horizon firms when they are completely uncertain about managerial horizon, and the right-hand-side describes the investors’ average valuation of long-horizon firms when their uncertainty decreases. Rearranging the inequality in (14) yields \(-\frac{N}{2M}\left({t}_{U}^{*}+{t}_{C}^{*}-2\delta \right)\left({t}_{U}^{*}-{t}_{C}^{*}\right)<0\), and this holds because \({t}_{U}^{*}-\delta >0\), \({t}_{C}^{*}-\delta >0\), and \({t}_{C}^{*}<{t}_{U}^{*}\).

Appendix B: Variable definitions

Variable

Definitions

Forecast

An indicator variable that equals one if the firm issues an earnings forecast (either a quarterly or an annual forecast) in a given fiscal year and zero otherwise

Treat

An indicator variable that equals one for firms with a December fiscal year-end and zero for firms with a June to November fiscal year-end

Post

An indicator variable that equals one in FY 2007 and zero in FY 2006

Size

Natural logarithm of total asset in millions

Mtb

Market value of equity divided by book value of equity

Roa

Income before extraordinary items divided by total asset

Loss

An indicator variable that equals one if the firm reports a negative Roa and zero otherwise

Special

Special items divided by total assets

Leverage

Sum of total long-term debt and total current liabilities divided by total asset

Earnvol

Standard deviation of the Roa over the past five fiscal years

Retvol

Standard deviation of monthly returns during a given fiscal year

Tenure

Number of years the CEO has worked for the company since 1998, according to ISS Incentive Lab

Instown

Percentage of shares owned by institutional investors at the end of the fiscal year as reported in Thomson Reuters 13F filings

Vesting

The change in the dollar value of the CEO’s vesting equity in a given fiscal year for a 100% change in the stock price, calculated as the number of vesting shares plus the aggregated delta of vesting options in a given fiscal year multiplied by the stock price at the end of the previous fiscal year, expressed in millions; sample observations with Vesting above (below) median are classified as short-horizon subsample (long-horizon subsample)

NegForecast

An indicator variable that equals one if the firm issues at least one negative (below zero) earnings forecast in a given fiscal year and zero if all the earnings forecasts issued by the firm in a given fiscal year are positive earnings forecasts

Btm

The ratio of average book value of equity to its fiscal year-end market value

Beta

Market beta calculated based on past 60 months returns from WRDS Beta Suite

Pastret

Cumulative buy-and-hold monthly return over the past 12 months

Mktret

Cumulative value-weighted monthly market return over the concurrent 12 months

Vesting_totcomp

Vesting scaled by the total compensation as reported in Incentive Lab; sample observations with Vesting_totcomp above (below) median are classified as short-horizon subsample (long-horizon) subsample

Vesting_own

Vesting scaled by the value of beneficially owned shares as reported in Incentive Lab; sample observations with Vesting_own above (below) median are classified as short-horizon (long-horizon) subsample

Duration

The weighted average duration of the four components of pay (i.e., salary, bonus, restricted stock grant, and stock option grant) that accounts for a cliff versus a ratable vesting schedule of the equity grants; sample observations with Duration below (above) median are classified as short-horizon subsample (long-horizon) subsample

TotalComp

Natural logarithm of one plus the CEO’s total compensation in dollars

CashCompRatio

Sum of the CEO’s salary and bonus divided by the CEO’s total compensation

EquityCompRatio

Sum of the CEO’s stock awards and option awards divided by the CEO’s total compensation

Appendix C: An example of a CD&A disclosure

This appendix illustrates how the CD&A disclosure mandate enhanced disclosure about executive compensation, facilitating investors’ understanding of executives’ horizon. As an example, I choose Briggs & Stratton Corporation’s proxy statements. The company produces engines and power equipment. The excerpts focus on the company’s discussion of its executives’ outstanding equity awards and their exercises before and after the CD&A mandate. Note that after the CD&A mandate, the company presents much more granular information about each executive’s outstanding equity awards at a tranche level, with detailed information on exercise price, expiration date, and vesting schedule.

Before CD&A mandate

Aggregated option/SAR exercises in last fiscal year and FY-end option/SAR values

   

Number of securities underlying unexercised options/SARs at fiscal year end (#)

Value of unexercised in-the-money options/SARs at fiscal year end ($)

Name

Shares acquired on exercise (#)

Value realized ($)

Exercisable

Unexercisable

Exercisable

Unexercisable

J.S. Shiely

50,000

599,839

232,460

857,881

1,703,477

363,194

T.J. Teske

8,000

103,120

50,380

158,214

377,390

89,427

T.R. Savage

11,000

144,810

85,460

194,312

620,235

78,399

J.E. Brenn

76,326

911,253

61,600

192,800

434,560

78,003

P.M. Neylon

49,420

537,164

22,120

201,754

171,762

87,708

No SARs are outstanding. Options at fiscal year end include options granted on August 15, 2006 for fiscal year 2006.

After CD&A mandate

Outstanding equity awards at FY 2007 fiscal year-end

Name (a)

Option awards

Stock awards

No. of securities underlying unexercised options exercisable (#) (b)

No. of securities underlying unexercised options unexercisable (#) (c)

Option exercise price ($/share) (d)

Option expiration date (e)

No. of shares or units of stock that have not vested (#) (f)

Market value of shares or units of stock that have not vested ($) (g)

J.S. Shiely

22,000

 

$23.110

8/3/07

15,582

$491,768

 

92,540

 

24.595

8/7/08

  
 

79,920

 

23.345

8/13/09

  
 

242,240

 

30.440

8/15/13

  
  

348,560

36.680

8/13/14

  
  

105,721

38.830

8/16/10

  
  

161,360

29.865

8/15/11

  

J.E. Brenn

41,680

 

24.595

8/7/08

6,000

189,360

 

19,920

 

23.345

8/13/09

3,827

120,780

 

45,940

 

30.440

8/15/13

  
  

83,020

36.680

8/13/14

  
  

25,910

38.830

8/16/10

  
  

37,930

29.865

8/15/11

  

T.J. Teske

5,474

 

23.110

8/3/07

6,000

189,360

 

18,180

 

24.595

8/7/08

4,155

131,132

 

9,740

 

23.345

8/13/09

  
 

22,780

 

30.440

8/15/13

  
  

55,600

36.680

8/13/14

  
  

20,264

38.830

8/16/10

  
  

59,570

29.865

8/15/11

  

T.R. Savage

5,474

 

23.110

8/3/07

6,000

189,360

 

45,720

 

24.595

8/7/08

3,662

115,573

 

19,980

 

23.345

8/13/09

  
 

46,420

 

30.440

8/15/13

  
  

83,700

36.680

8/13/14

  
  

26,202

38.830

8/16/10

  
  

37,990

29.865

8/15/11

  

W.H. Reitman

3,060

 

24.595

8/7/08

2,046

64,572

 

15,160

 

23.345

8/13/09

  
 

31,960

 

30.440

8/15/13

  
  

38,040

36.680

8/13/14

  
  

12,166

38.830

8/16/10

  
  

22,570

29.865

8/15/11

  

Column (b): options that expire in 2007 vested on August 3, 2003; options that expire in 2008 vested on August 7, 2004; options that expire in 2009 vested on August 13, 2005; and options that expire in 2013 vested on August 15, 2006.

Column (c): options that expire in 2014 vested on August 13, 2007; options that expire in 2010 will vest on August 16, 2008; and options that expire in 2011 will vest on August 15, 2009.

Column (f): restricted stock awarded in 2003 will vest on August 15, 2008; and restricted and deferred stock awarded in 2005 will vest on August 16, 2010.

Column (g): based on the $31.56 per share closing price of a share of the company’s common stock as of the last business day of fiscal year 2007.

Option exercise and stock vested during fiscal year 2007

Name

Option awards

Stock awards

No. of shares acquired on exercise (#)

Value realized on exercise ($)

No. of shares acquired on vesting (#)

Value realized on vesting ($)

J.S. Shiely

38,000

$241,060

0

0

J.E. Brenn

0

0

0

0

T.J. Teske

16,986

141,439

0

0

T.R. Savage

14,286

131,831

0

0

W.H. Reitman

3,074

17,504

0

0

Appendix D: The effect of CD&A disclosure mandate on ERC

This appendix studies the effect of CD&A disclosure mandate on the earnings response coefficient. Fischer and Verrecchia (2000) theoretically show that a decrease in investors’ uncertainty about the managerial preferences increases the earnings response coefficient (ERC). Ferri et al. (2018) empirically test the intuition from Fischer and Verrecchia (2000) and show that the CD&A mandate increases the ERC of the affected firms, relative to the control firms, consistent with the enhanced compensation disclosure reducing investors’ uncertainty about the manager’s disclosure preferences. This appendix replicates their findings by estimating the following regression equation.

$$\begin{array}{c}{Price\left(BHAR\right)}_{i,t}=\beta_0+{\boldsymbol\beta}_1Earning{s\left(Surprise\right)}_{i,t}\times{Treat}_i\times{Post}_t+\beta_2Earnings{\left(Surprise\right)}_{i,t}\\+\beta_3Earning{s\left(Surprise\right)}_{i,t}\times{Treat}_i+\beta_4Earning{s\left(Surprise\right)}_{i,t}\times Post_t+\\\beta_5Treat_i\times{Post}_t+\beta_6{Treat}_i+\beta_7{Post}_t+\Gamma Controls{}_{i,t}+\varepsilon_{i,t}.\end{array}$$

Price is stock price at the end of the fiscal year. BHAR is the market-adjusted buy-and-hold cumulative monthly return during the fiscal year. Earnings is the earnings per share, and Surprise is the current year’s EPS minus the past year’s EPS, deflated by stock price at the beginning of the period. I include market beta (Beta), firm size (Size), market-to-book ratio (Mtb), and leverage (Leverage) as the controls. Continuous variables are winsorized at the 1st and 99th percentiles. Robust t-statistics clustered by firm are in parentheses. ***, **, and * indicate significance at the 1%, 5%, and 10% two-tailed levels, respectively.

 

Price

Price

BHAR

BHAR

 

(1)

(2)

(3)

(4)

Earnings × Treat × Post

3.767***

3.451***

  
 

(2.76)

(2.60)

  

Surprise × Treat × Post

  

1.739**

1.383*

   

(1.97)

(1.69)

Earnings

9.804***

8.769***

  
 

(9.53)

(7.47)

  

Earnings × Treat

 − 4.431***

 − 4.028***

  
 

(− 3.87)

(− 3.24)

  

Earnings × Post

 − 5.328***

 − 4.733***

  
 

(− 5.14)

(− 4.42)

  

Surprise

  

1.959***

1.804***

   

(2.76)

(2.75)

Surprise × Treat

  

 − 1.623**

 − 1.284*

   

(− 2.21)

(− 1.88)

Surprise × Post

  

 − 0.063

0.016

   

(− 0.08)

(0.02)

Treat × Post

 − 16.816***

 − 15.604***

 − 0.061

 − 0.053

 

(− 4.29)

(− 4.07)

(− 1.43)

(− 1.21)

Treat

15.631***

12.730***

0.031

0.041

 

(6.06)

(4.73)

(1.14)

(1.43)

Post

21.848***

19.930***

0.069*

0.056

 

(6.97)

(6.22)

(1.87)

(1.47)

Controls

No

Yes

No

Yes

Observations

1,193

1,193

1,193

1,193

R2

0.34

0.39

0.11

0.15

Appendix E: A numerical example of Vesting calculation

This appendix illustrates the calculation steps to derive the price sensitivity of the CEO’s vesting stocks and options (Vesting). I calculate Vesting for a sample CEO and present the company’s disclosure tables retrieved from Incentive Lab for FY 2007, following Appendix B of Edmans et al. (2017). As an example, I use Harold M. Korell, CEO of Southwestern Energy Company, and calculate the price sensitivity of his vesting stocks and options (Vesting) for FY 2007.

First, I obtain data from Incentive Lab for Harold M. Korell:

Outstanding options as reported in Incentive Lab

 

Equity type

Number of securities

Strike price

Expiry date (YYYYMMDD)

As of Dec 31, 2007

(1)

Unexercisable options

20,320

$35.49

20121208

(2)

Unexercisable options

40,667

$40.67

20131211

(3)

Unexercisable options

39,490

$54.36

20141213

(4)

Exercisable options

292,334

$1.50

20091216

(5)

Exercisable options

746,220

$1.86

20101220

(6)

Exercisable options

93,507

$2.41

20111220

(7)

Exercisable options

345,096

$2.87

20121211

(8)

Exercisable options

225,799

$5.29

20131210

(9)

Exercisable options

146,920

$12.45

20111209

(10)

Exercisable options

40,640

$35.49

20121208

(11)

Exercisable options

20,333

$40.67

20131211

As of Dec 31, 2006

(12)

Unexercisable options

48,972

$12.45

20111209

(13)

Unexercisable options

40,640

$35.49

20121208

(14)

Unexercisable options

61,000

$40.67

20131211

(15)

Exercisable options

392,336

$1.50

20091216

(16)

Exercisable options

746,220

$1.86

20101214

(17)

Exercisable options

93,508

$2.41

20111220

(18)

Exercisable options

345,096

$2.87

20121211

(19)

Exercisable options

244,720

$5.29

20131210

(20)

Exercisable options

97,948

$12.45

20111209

(21)

Exercisable options

20,320

$35.49

20121208

12.1 Newly granted options as reported in Incentive Lab

 

Grant date (YYYYMMDD)

Number of securities

Strike price

Expiry date (YYYYMMDD)

(22)

20071213

39,490

$54.36

20141213

To calculate the number of vesting options for FY 2007, I match and group the outstanding options by strike price and expiry date. I then infer the number of vesting options from the following relationship, for each strike price-expiry date pair.

$$VestingOptNu{m}_{t}=UnvestedOptNu{m}_{t-1}+NewOptNu{m}_{t}-UnvestedOptNu{m}_{t}$$

After identifying vesting options in FY 2007 at the grant level (shown in the table below), I then use the Black–Scholes formula to calculate each option’s delta.

Calculated number and delta of vesting options as of Dec 31, 2007

Calculated number of vesting options

Number of Securities

Strike price

Expiry date (YYYYMMDD)

Z

Delta

(12)

48,972

$12.45

20111209

2.0552

16,822.63

(13) − (1)

20,320

$35.49

20121208

0.4872

4,892.43

(14) − (2)

20,333

$40.67

20131211

0.3793

4,616.44

(22) − (3)

0

$54.36

20141213

  
    

SumDelta = 26,331.5

Next, I obtain share data from Incentive Lab for Harold M. Korell.

The total number of vesting shares in FY 2007 as reported in Incentive Lab

 

Number of shares acquired on vesting for the year ending on Dec 31, 2007

(23)

42,870

As a last step, to calculate the price-sensitivity measures of vesting options for FY 2007, I add the deltas calculated above with the total number of vesting shares and multiply the sum by the closing stock price of 35.05 at the end of FY 2006. In the end, Harold M. Korell’s Vesting for FY 2007 is calculated as (26,331.5 + 42,870) × 35.05 = 2,425,512.6.

Appendix F: Validation of Vesting as a measure of managerial horizon

This appendix validates my main measure of managerial horizon, Vesting. In particular, I study whether those managers classified as short-horizon based on Vesting exhibit greater equity sales, relative to those classified as long-horizon. To do so, I estimate the following regression equation.

$$Equitysal{eDummy}_{i,t}\left(Equitysal{eValue}_{i,t}\right)=\beta_0+{\boldsymbol\beta}_1ShortHorizonDummy_i+\Gamma Controls_{i,t}+\varepsilon_{i,t}$$

Following Edmans et al. (2018), I use the term “equity sales” to refer to standard stock sales, sales of shares obtained upon option exercise, or the CEO cancelling some shares to pay for taxes or the strike price upon option exercise. EquitysaleDummy is an indicator variable that equals one if the CEO makes an equity sale in a given fiscal year and zero otherwise. EquitysaleValue is the total value of the equity sold expressed in millions, calculated as the number of equity sold multiplied by the transaction price. I construct the equity sales variables from Thomson Reuters Insiders Data. ShortHorizonDummy is an indicator variable that equals one if the observation belongs to the short-horizon subsample. As controls, I include firm size (Size), market-to-book ratio (Mtb), past returns (Pastret), return volatility (Retvol), the CEO’s beneficial ownership (Holdings), and tenure (Tenure). Continuous variables are winsorized at the 1st and 99th percentiles. Robust t-statistics clustered by firm are in parentheses. ***, **, and * indicate significance at the 1%, 5%, and 10% two-tailed levels, respectively.

 

EquitysaleDummy

EquitysaleValue

 

(1)

(2)

ShortHorizonDummy

0.105***

4.580***

 

(3.15)

(6.45)

Size

0.013

1.242***

 

(1.07)

(3.54)

Mtb

0.003

0.141**

 

(0.74)

(2.11)

Pastret

0.062

2.497**

 

(1.54)

(2.28)

Retvol

 − 0.170

13.525

 

(− 0.37)

(1.20)

Holdings

 − 0.007*

0.295**

 

(− 1.95)

(2.09)

Tenure

0.006

0.666***

 

(1.12)

(5.06)

Observations

1,193

1,193

R2

0.03

0.14

Appendix G: A numerical example of Duration calculation

This appendix illustrates the calculation steps to derive the pay duration measure (Duration), following Section 1 from Gopalan et al. (2014). They follow the fixed income literature and calculate pay duration as the weighted average duration of the four components of pay (i.e., salary, bonus, restricted stock grants, and stock option grants). In cases in which the stock and option awards have a cliff vesting schedule, I estimate pay duration as follows.

$$Duration=\frac{\left(Salary+Bonus\right)\times 0+\sum_{i=1}^{{n}_{s}}Restricted stoc{k}_{i}\times {t}_{i}+\sum_{j=1}^{{n}_{o}}Optio{n}_{j}\times {t}_{j}}{Salary+Bonus+\sum_{i=1}^{{n}_{s}}Restricted stoc{k}_{i}+\sum_{j=1}^{{n}_{o}}Optio{n}_{j}}.$$

i denotes a restricted stock grant. j denotes an option grant. Salary and Bonus are the dollar values of annual salary and bonus, respectively. As I calculate duration relative to the year-end, Salary and Bonus have a vesting period of zero. Next, Restricted stocki is the grant date fair value of restricted stock grant i with corresponding vesting period ti (in years) as reported in Incentive Lab. During the year, the firm may have other stock grants with different vesting periods (different ti), and ns is the total number of such stock grants. Similarly, Optionj is the grant date fair value of option grant j with corresponding vesting period tj (in years) as reported in Incentive Lab. no is the total number of such option grants.

In cases where the restricted stock grant (option grant) has a ratable vesting schedule, I modify the above formula by replacing ti(j) with (ti(j) + 1)/2.Footnote 29

I next provide a numerical example for Duration calculation using Antonio M. Perez, CEO of Eastman Kodak Co., as an example. In particular, I calculate his pay duration (Duration) for FY 2006. First, I obtain stock and option grants data as well as salary and bonus compensation data from Incentive Lab for Antonio M. Perez.

Salary and bonus for FY 2006 as reported in Incentive Lab

 

Salary

Bonus

(1)

$1,096,168

$690,525

17.1 Restricted stock and option grants for FY 2006 as reported in Incentive Lab

 

Award type

Grant date (YYYYMMDD)

Grant date fair value

Vesting schedule

Vesting period (in years)

(2)

Restricted Stock

20060101

1,518,525

Cliff

2

(3)

Restricted Stock

20060512

776,794

Cliff

1.67

(4)

Options

20061212

2,956,582

Ratable

3

After identifying salary, bonus, and restricted stock and option grants for FY 2006, I use the Duration formula above to calculate Antonio M. Perez’s Duration for FY 2006 as follows.

$$Duration=\frac{\left(\mathrm{1,096,168}+\mathrm{690,525}\right)\times 0+\left(\mathrm{1,518,525}\times 2+\mathrm{776,794}\times 1.67\right)+\left(\mathrm{2,956,582}\times \frac{3+1}{2}\right)}{\mathrm{1,096,168}+\mathrm{690,525}+\mathrm{1,518,525}+\mathrm{776,794}+\mathrm{2,956,582}}=1.456.$$

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kim, J.M. Uncertainty about managerial horizon and voluntary disclosure. Rev Account Stud 28, 615–657 (2023). https://doi.org/10.1007/s11142-021-09652-2

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11142-021-09652-2

Keywords

Navigation