Skip to main content
Log in

Why do pro forma and Street earnings not reflect changes in GAAP? Evidence from SFAS 123R

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

Abstract

This study examines how key market participants—managers and analysts—responded to SFAS 123R’s controversial requirement that firms recognize stock-based compensation expense. Despite mandated recognition of the expense, some firms’ managers exclude it from pro forma earnings and some firms’ analysts exclude it from Street earnings. We find evidence consistent with managers opportunistically excluding the expense to increase earnings, smooth earnings, and meet earnings benchmarks but no evidence that these exclusions result in an earnings measure that better predicts future firm performance. In contrast, we find that analysts exclude the expense from earnings forecasts when exclusion increases earnings’ predictive ability for future performance and that opportunism generally does not explain exclusion by analysts incremental to exclusion by managers. Thus our findings indicate that opportunism is the primary explanation for exclusion of the expense from pro forma earnings and predictive ability is the primary explanation for exclusion from Street earnings. Our findings suggest the controversy surrounding the recognition of stock-based compensation expense may be attributable to cross-sectional variation in the relevance of the expense for equity valuation, as well as to differing incentives of market participants.

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.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. The SEC describes the concern as relating to “the improper use of non-GAAP financial measures during the past 30 years.” Former SEC Chief Accountant Turner stated “people use these ‘pro-forma’ press releases as a vehicle to spin the investors” (Turner, as quoted in Stamas, “SEC looks into firms’ earnings releases ‘Pro-forma’ figures may mislead investors,” The Seattle Times, June 19, 2001). More recently, the SEC has referred to pro forma earnings as a fraud risk factor (Leone, “What’s on the SEC’s radar?” CFO.com, September 29, 2010).

  2. Throughout, stock-based compensation expense refers to the portion of stock-based compensation expense that was not required to be recognized prior to the issuance of SFAS 123R. This expense relates primarily to at-the-money stock options granted to employees.

  3. The term “Street earnings” sometimes refers to a firm’s realized GAAP earnings adjusted to exclude earnings components that have been excluded from the consensus analyst forecast. We use the term to refer to the forecast of Street earnings, that is the consensus analyst forecast.

  4. “Petition for Review and Repeal of FAS 123R, ‘Share-Based Payment’” filed with the SEC by opponents of SFAS 123R on February 27, 2008.

  5. Consistent with similar motivations for voluntary expense recognition and inclusion of the expense in earnings forecasts, stock-based compensation expense is included in consensus forecasts of all firms McConnell et al. (2003) identify as voluntarily recognizing the expense. As Sect. 5 explains, we exclude from our sample all firms that voluntarily recognized the expense. We are unaware of any of firms that included unrecognized stock-based compensation expense in pro forma earnings prior to 2006.

  6. I/B/E/S and First Call are data services of Thomson Financial that similarly treat actual earnings and analyst earnings forecasts. Thus we refer to them collectively as First Call–I/B/E/S except when discussing our data sources; First Call and I/B/E/S maintain separate databases. All data used in this study are taken from the I/B/E/S historical research database, with exception of footnote data, which are taken from the First Call database.

  7. Consistent with this, Zacks, another supplier of analyst earnings forecasts, includes stock-based compensation expense in all forecasts in its database; if the forecast submitted by the analyst excludes the expense, Zacks adjusts the forecast to include it. See Wall Street Journal Online, “Options expensing jars consensus,” April 4, 2006. Because stock-based compensation expense is not a “special” or transitory item, the forecast data provider can more readily determine whether stock-based compensation expense is included in a particular analyst’s forecast made prior to observing actual earnings. This determination is more difficult for special or transitory items examined in prior research. This is because these items, almost by definition, should have an expectation of zero, which means there is no practical difference between forecasting the item to be zero and excluding the item from the forecast. Thus such items are only meaningfully excluded after actual earnings are observed.

  8. However, Cohen et al. (2007) find that the increased explanatory power of Street earnings is attributable to measurement error in earnings surprises.

  9. Brown and Sivakumar (2003) observe that comparing GAAP earnings and Street earnings biases in favor of Street earnings because GAAP earnings is not designed to measure recurring operating earnings, which is more the motivation for Street earnings. Because operating earnings is not defined by GAAP, Bhattacharya et al. (2003) and Brown and Sivakumar (2003) study operating earnings as defined by S&P.

  10. Landsman et al. (2007) obtain similar inferences using tests of predictive ability based on the Ohlson (1995, 1999) valuation model.

  11. A large portion of firms in our sample with stock-based compensation expense excluded from Street earnings are in the computer industry. Such firms might be considered glamour firms. The glamour firm findings in Baik et al. (2009) suggest that excluded stock-based compensation expense would have predictive ability for these firms. We find the opposite, which suggests that our findings are not explained by the glamour firm effect in Baik et al. (2009).

  12. Lougee and Marquardt (2004) use firms’ earnings response coefficients and corresponding R 2s to explain the likelihood that a firm subsequently reports pro forma earnings. However, as in other related research, Lougee and Marquardt (2004) compare predictive ability only for firms with pro forma exclusions and do not consider whether cross-sectional differences in an item’s predictive ability explain its exclusion. Curtis et al. (2011) test the opportunism and predictive ability explanations for net positive special items by confirming that the items have less predictive ability for future operating earnings than past operating earnings and documenting no association between the items and earnings announcement returns.

  13. For a few firms we confirmed that Compustat data #399, our measure of COMPX, is the difference, per share, between GAAP net income and what net income would have been had stock-based compensation expense been measured using the grant date value measurement approach disclosed under SFAS 123, which SFAS 123 requires firms to disclose. COMPX reflects the amount by which earnings that excludes the expense is higher than earnings that includes it.

  14. Untabulated tests reveal that our inferences are unaffected if we use the variability of net income minus stock-based compensation expense instead of σ(COMPXA).

  15. This is consistent with the statement by TIBCO Software Inc. in its December 21, 2006 8-K, explaining its exclusion of stock-based compensation expense from pro forma earnings: “the nature of the stock-based compensation expense also makes it very difficult to estimate prospectively, since the expense will vary with changes in the stock price and market conditions at the time of new grants, varying valuation methodologies, subjective assumptions and different award types.”

  16. These variables also could identify situations in which investors would like more information about stock-based compensation expense and managers oblige. However, managers could provide such information without using pro forma earnings, as regulated by Regulation G, that excludes stock-based compensation expense, which is how we define EX_PROFORMA. See footnote 25.

  17. These firms are Bank of America, Bear Stearns, Credit Suisse First Boston, Goldman Sachs, Merrill Lynch, Sanford C. Bernstein, and UBS. See, for example S. Taub, “Staggered start for options expensing” (CFO.com. June 1, 2005) and “Stock options: So who’s counting,” (New York Times, nytimes.com, August 6, 2005). Our inferences are unaffected if POLICY equals one if a majority of the brokers following the firm have a policy to include the expense, or if we eliminate firms for which a majority of brokers following the firm have such a policy.

  18. Our inferences are unaffected if we define SIZE as the natural logarithm of total assets.

  19. Using GAAP earnings in our predictive ability equations, rather than pro forma or Street earnings, helps avoid confounding effects related to other exclusion decisions that managers or analysts make. This is because GAAP earnings is not dependent on those decisions.

  20. For example, in Eq. (2) if earnings, EARN, were based on net income after subtracting stock-based compensation expense, we could find a significant negative relation between EARN t+1 and COMPXA t solely because the expense is recurring and positively autocorrelated. The mean (median) firm-specific AR(1) coefficient for COMPXA over our sample period is 0.39 (0.44). By basing EARN on net income without subtracting stock-based compensation expense, we rule out this possibility.

  21. Using the same dependent variable in Eqs. (2) and (3), and Eqs. (4) and (5), permits us to conduct meaningful tests of differences in explanatory power based on regression R 2s. Our tests of significance are based on a nonparametric goodness of fit test because Clarke (2003, 2007) shows the test is superior to the Vuong (1989) test when the sets of explanatory variables are highly correlated, as they are in our setting.

  22. When estimating all of our equations, we eliminate outliers. Following Belsley et al. (1980), we classify outliers as those observations with studentized residuals greater than three in absolute value. All inferences are unaffected by the inclusion of outliers.

  23. Some related research (for example, Gu and Chen 2004; Baik et al. 2009; Christensen et al. 2011) uses the First Call footnote file to determine whether analyst forecasts exclude an earnings component. We use the Bear Stearns list because of the additional analysis performed by Bear Stearns to verify whether the First Call footnotes were correct with respect to the exclusion of stock-based compensation expense. Also, we identified three types of errors in the footnote file. First, not all firms have footnotes indicating whether stock-based compensation expense is included in or excluded from the forecast. For example, there is no footnote related to stock-based compensation expense for Google’s 2006 earnings forecast, but we confirmed that stock-based compensation expense is excluded from the forecast as the Bear Stearns list indicates. Second, some firms have multiple footnotes that conflict as to whether the forecast includes or excludes the expense. Third, some firms have a footnote indicating exclusion, but we confirmed from press releases and analyst reports that the forecasts include the expense, consistent with the Bear Stearns analysis. See Baik et al. (2009) for other caveats regarding use of the First Call footnote file.

  24. These eliminations were of potential Street Includers; no Street Excluder voluntarily recognized the expense.

  25. Regulation G requires firms that disclose a non-GAAP earnings measure to label the measure as non-GAAP and reconcile the non-GAAP measure to the most comparable GAAP measure, namely earnings or earnings per share. Some firms disclose other earnings-based measures typically labeled “adjusted EBITDA” or something similar. We do not label these firms as Pro Forma Excluders unless the firm also excludes stock-based compensation expense from a non-GAAP measure of earnings or earnings per share that is reconciled to GAAP earnings or earnings per share.

  26. Street and Pro Forma Includers could include firms with exclusions of other, possibly recurring, earnings components. To the extent that the incentives for stock-based compensation expense exclusion are the same as for exclusion of other earnings components, our tests will be biased against finding support for the opportunism explanation. However, exclusions of other earnings components will not affect our predictive ability tests because those tests focus only on the predictive ability of stock-based compensation expense for future GAAP earnings.

  27. Bhattacharya et al. (2003) report that Street and pro forma earnings differ about one-third of the time, and thus our proportion of 14 % (256/1,845) might appear low. However, our sample includes firms that did not have exclusions from either Street or pro forma earnings, which appear as firms for which analysts and managers agree, whereas the Bhattacharya et al. (2003) sample includes only firms with exclusions of one or the other type.

  28. In Sect. 6.3 we estimate Eqs. (1), (4), and (5) separately for computer and non-computer firms.

  29. Our inferences are unaffected if we eliminate the five Street Excluders that are Pro Forma Includers.

  30. In both panels, the R 2s of both Model 1 and Model 2 are higher for Includers than for Excluders. The R 2s in panel A for Pro Forma Includers (Excluders) are 46.88 and 45.61 % (28.64 and 30.16 %) and in panel B for Street Includers (Excluders) are 42.77 and 43.63 % (36.23 and 35.25 %). The smaller R 2s for Excluders is consistent with Excluders having more transitory earnings components than Includers. However, differences in R 2s between Includers and Excluders do not affect our inferences, which are based on the difference in R 2s for Model 1 and Model 2 within each Includer and Excluder group, not between the groups. The only difference between Model 1 and Model 2 is whether stock-based compensation expense is included in the explanatory earnings variable.

  31. Untabulated statistics from several alternative specifications of Eqs. (2) through (5) reveal the same inferences regarding predictive ability as we draw from the tabulated statistics. In particular, we estimate the Eq. (1) including an indicator variable that equals one if EARN t is negative and zero otherwise, and an interaction of the loss indicator and the equation’s earnings variable; (2) eliminating SIZE and BM from the set of explanatory variables; (3) including an interaction variable of EX and the equation’s earnings variable; (4) in Eqs. (3) and (5), permitting EARN t and COMPXA, or ΔEARN t and ΔCOMPXA t , to have different coefficients; and (5) using, sequentially, operating earnings from Compusat (Kolev et al. 2008) and cash from operations instead of GAAP earnings. We also estimate Eqs. (2) through (4) using (EARN t+1 − COMPXA t+1) as the dependent variable and estimate Eq. (5) using Δ(EARN t+1 − COMPXA t+1).

References

  • Aboody, D. (1996). Market valuation of employee stock options. Journal of Accounting and Economics, 22, 357–391.

    Article  Google Scholar 

  • Aboody, D., Barth, M. E., & Kasznik, R. (2004a). SFAS No. 123 stock-based compensation expense and equity market values. The Accounting Review, 79(2), 251–275.

    Article  Google Scholar 

  • Aboody, D., Barth, M. E., & Kasznik, R. (2004b). Firms’ voluntary recognition of stock-based compensation expense. Journal of Accounting Research, 42(2), 123–150.

    Article  Google Scholar 

  • Accounting Principles Board. (1973). Accounting for stock issued to employees. New York, NY: APB.

    Google Scholar 

  • Baik, B., Farber, D. B., & Petroni, K. (2009). Analyst incentives and street earnings. Journal of Accounting Research, 47(1), 1–25.

    Article  Google Scholar 

  • Barth, M. E., Beaver, W. H., & Landsman, W. R. (1998). Relative valuation roles of equity book value and net income as a function of financial health. Journal of Accounting and Economics, 25(1), 1–34.

    Article  Google Scholar 

  • Belsley, D. A., Kuh, E., & Welsch, R. E. (1980). Regression diagnostics. New York, NY: Wiley.

    Book  Google Scholar 

  • Blacconiere, W. G., Frederickson, J. R., Johnson, M. F., & Lewis, M. F. (2011). Are voluntary disclosures that disavow the reliability of mandated fair value information informative or opportunistic? Journal of Accounting and Economics, 52(2–3), 235–251.

    Article  Google Scholar 

  • Black, D., & Christensen, T. (2009). US managers’ use of ‘pro forma’ adjustments to meet strategic earnings benchmarks. Journal of Business, Finance and Accounting, 36, 297–326.

    Article  Google Scholar 

  • Black, F., & Scholes, M. (1973). The effects of dividend yield and dividend policy on common stock prices and returns. Journal of Financial Economics, 1, 1–22.

    Article  Google Scholar 

  • Bhattacharya, N., Black, E. L., Christensen, T. E., & Larson, C. R. (2003). Assessing the relative informativeness and permanence of pro forma earnings and GAAP operating earnings. Journal of Accounting and Economics, 36, 285–319.

    Article  Google Scholar 

  • Bradshaw, M. T., & Sloan, R. G. (2002). GAAP versus the Street: An empirical assessment of two alternative definitions of earnings. Journal of Accounting Research, 40(1), 41–66.

    Article  Google Scholar 

  • Brown, L. D., & Sivakumar, K. (2003). Comparing the value relevance of two operating income measures. Review of Accounting Studies, 8, 561–572.

    Article  Google Scholar 

  • Brown, L. D., & Y.-J. Lee. (2006). The impact of SFAS 123R on changes in option-based compensation. Working paper, Georgia State University.

  • Brown, N. C., Christensen, T. E., Elliott, W. B., & Mergenthaler, R. D. (2011). Investor sentiment and pro forma earnings disclosures. Journal of Accounting Research, 50, 1–40.

    Google Scholar 

  • Burgstahler, D., & Dichev, I. (1997). Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics, 24, 99–126.

    Article  Google Scholar 

  • Carter, M. E., & Lynch, L. J. (2003). The consequences of the FASB’s 1998 proposal on accounting for stock option repricing. Journal of Accounting and Economics, 35, 51–72.

    Article  Google Scholar 

  • Choudhary, P., Rajgopal, S., & Venkatachalam, M. (2009). Accelerated vesting of employee stock options in anticipation of FAS 123-R. Journal of Accounting Research, 47(1), 1–42.

    Article  Google Scholar 

  • Christensen, T. E., Merkley, K. J., Tucker, J. W., & Venkataraman, S. (2011). Do managers use earnings guidance to influence street earnings exclusions? Review of Accounting Studies, 16, 501–527.

    Article  Google Scholar 

  • Christie, A. A. (1987). On cross-sectional analysis in accounting research. Journal of Accounting and Economics, 9(3), 231–258.

    Article  Google Scholar 

  • Clarke, K. (2003). Nonparametric model discrimination in international relations. Journal of Conflict Resolution, 47, 724–774.

    Article  Google Scholar 

  • Clarke, K. (2007). A simple distribution-free test for nonnested model selection. Political Analysis, 15(3), 347–363.

    Article  Google Scholar 

  • Cohen, D., Hann, R., & Ogneva, M. (2007). Another look at GAAP versus the Street: an empirical assessment of measurement error bias. Review of Accounting Studies, 12(2–3), 271–303.

    Article  Google Scholar 

  • Curtis, A., McMahon, S., McVay, S., & Whipple, B. (2011). Non-GAAP earnings: Informative or opportunistic? An analysis of transitory gains. Working paper. University of Utah.

  • Doyle, J. T., Lundholm, R. J., & Soliman, M. T. (2003). The predictive value of expenses excluded from pro forma earnings. Review of Accounting Studies, 8(2–3), 145–174.

    Article  Google Scholar 

  • Financial Accounting Standards Board. (1995). Statement of financial accounting standards no. 123: Accounting for stock-based compensation. Norwalk, CT: FASB.

  • Financial Accounting Standards Board. (2004). Statement of financial accounting standards no. 123 (revised 2004): Share-based payment. Norwalk, CT: FASB.

  • First Call (1999). Present methodology for US earnings estimates, company documentation. New York, NY: First Call.

  • Francis, J., & Philbrick, D. (1993). Analysts’ decisions as a product of a multi-task environment. Journal of Accounting Research, 31, 216–230.

    Article  Google Scholar 

  • Gow, I. D., Ormazabal, G., & Taylor, D. J. (2010). Correcting for cross-sectional and time-series dependence in accounting research. The Accounting Review, 85(2), 483–512.

    Article  Google Scholar 

  • Gu, Z., & Chen, T. (2004). Analysts’ treatment of nonrecurring items in street earnings. Journal of Accounting and Economics, 38, 129–170.

    Article  Google Scholar 

  • Hanlon, M., Rajgopal, S., & Shevlin, T. (2003). Are executive stock options associated with future earnings. Journal of Accounting and Economics, 36, 3–43.

    Article  Google Scholar 

  • Hagopian, K. (2006). Point of view: Expensing employee stock options is improper accounting. California Management Review, 48, 136–156.

    Google Scholar 

  • Johnston, D., & Rock, S. (2006). The effect of expense recognition on stock option grants and operating performance. Working paper.

  • Jones, S. D. (2006, April 4). Options expensing jars consensus. Wall Street Journal.

  • Kasznik, R., & McNichols, M. F. (2002). Does meeting earnings expectations matter? Evidence from analyst forecast revisions and share prices. Journal of Accounting Research, 40(3), 727–759.

    Article  Google Scholar 

  • Kolev, K., Marquardt, C. A., & McVay, S. E. (2008). SEC scrutiny and the evolution of non-GAAP reporting. The Accounting Review, 83(1), 157–184.

    Article  Google Scholar 

  • Kothari, S. P., & Zimmerman, J. L. (1995). Price and return models. Journal of Accounting and Economics, 20(2), 155–192.

    Article  Google Scholar 

  • Landsman, W. R., Miller, B. L., & Yeh, S. (2007). Implications of components of income excluded from pro forma earnings for future profitability and equity valuation. Journal of Business Finance and Accounting, 34(3–4), 650–675.

    Article  Google Scholar 

  • Lambert, R. A. (2004). Discussion of ‘analysts’ treatment of non-recurring items in street earnings’ and ‘loss function assumptions in rational expectations tests on financial analysts’ earnings forecasts’. Journal of Accounting and Economics, 38(1–3), 205–222.

    Article  Google Scholar 

  • Larcker, D. F. (2003). Discussion of ‘are executive stock options associated with future earnings?’ Journal of Accounting and Economics, 36, 91–103.

    Article  Google Scholar 

  • Lin, H., & McNichols, M. F. (1998). Underwriting relationships, analysts’ earnings forecasts and investment recommendations. Journal of Accounting and Economics, 25, 101–127.

    Article  Google Scholar 

  • Lougee, B., & Marquardt, C. (2004). Earnings informativeness and strategic disclosure: An empirical examination of ‘pro forma’ net income’. The Accounting Review, 79, 769–795.

    Article  Google Scholar 

  • McConnell, P., Pegg, J., Senyek, C., & Mott, D. (2003, September 4). More companies voluntarily adopt fair value expensing of employee stock options. Bear Stearns Equity Research.

  • McVay, S. (2006). Earnings management using classification shifting: An examination of core earnings and special items. The Accounting Review, 81(3), 501–531.

    Article  Google Scholar 

  • Michaely, R., & Womack, K. (1999). Conflict of interest and the credibility of underwriter analyst recommendations. Review of Financial Studies, 12, 653–686.

    Article  Google Scholar 

  • Ohlson, J. A. (1995). Earnings, equity book values, and dividends in equity valuation. Contemporary Accounting Research, 31, 661–687.

    Google Scholar 

  • Ohlson, J. A. (1999). On transitory earnings. Review of Accounting Studies, 3, 145–162.

    Google Scholar 

  • Richardson, S., Teoh, S. H., & Wysocki, P. D. (2004). The walk-down to beatable analyst forecasts: The role of equity issuance and insider trading incentives. Contemporary Accounting Research, 21(4), 885–924.

    Article  Google Scholar 

  • Riedl, E. J. (2004). An examination of long-lived asset impairments. Accounting Review, 79(3), 823–852.

    Article  Google Scholar 

  • Riedl, E. J., & Srinivasan, S. (2010). Signaling firm performance through financial statement presentation: An analysis using special items. Contemporary Accounting Research, 27, 289–332.

    Article  Google Scholar 

  • Senyek, C., Pegg, J., Mott, D., & Calingasan, A. (2007, January 16). Consensus estimates exclude stock-based compensation expense for 32% of NASDAQ 100 companies. Bear Stearns Equity Research.

  • Vuong, Q. (1989). Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica, 57(2), 307–333.

    Article  Google Scholar 

Download references

Acknowledgments

We thank David Aboody; Tom Lys; Robert Magee; Sarah McVay; Richard Sloan (editor); workshop participants at the 2011 Review of Accounting Studies conference, especially discussant Theodore Christensen; American Accounting Association annual meeting, especially discussant Valentina Zamora; the Financial Accounting and Reporting Section midyear meeting, especially discussant Melissa Lewis; the Kellogg School of Management; Lancaster University; Santa Clara University; Stanford University; University of Southern California; Yale University; and two anonymous reviewers for helpful comments and suggestions. We thank our respective schools for financial support.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Mary E. Barth.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Barth, M.E., Gow, I.D. & Taylor, D.J. Why do pro forma and Street earnings not reflect changes in GAAP? Evidence from SFAS 123R. Rev Account Stud 17, 526–562 (2012). https://doi.org/10.1007/s11142-012-9192-9

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11142-012-9192-9

Keywords

JEL Classification

Navigation