Dirk Black and Dr. Ted Christensen, School of Accountancy
Pro forma reporting has received significant attention from managers, regulators, and researchers in recent years. Pro forma reporting occurs when managers adjust earnings for certain “one-time” items allegedly to present earnings that are more representative of the “core” operations of their firms. In addition, analysts make adjustments to reported earnings to arrive at earnings per share estimates. Prior research has examined various aspects of pro forma reporting. Bhattacharya, et al., 2006 and Frederickson and Miller, 2004 find that investment decisions of ordinary, retail investors are affected by pro forma earnings while those of professional investors are not. Different proxies exist for pro forma, or non-GAAP, earnings. Bradshaw and Sloan, 2002 find that “street earnings” (earnings from analysts) have greater information content than GAAP earnings. Bhattacharya, et al, 2003 find that manager-adjusted earnings (pro forma) have more information content than GAAP operating earnings. Another important part of the pro forma earnings research involves the effects of government regulation. Entwistle, et al., 2006, find that the frequency of pro forma earnings disclosures decreased from 2001 to 2003 in their sample of S&P 500 firms.
This study examines the differences in pro forma reporting between firms followed by analysts and those that are not. The research questions are as follows: (1) Are there differences in market reactions to pro forma press releases released by analyst-followed and non-analyst-followed firms? (2) Do firms followed by analysts make different adjustments than firms that are not followed by analysts? (3) How have managers and analysts changed their behavior given recent regulatory action (The Sarbanes-Oxley Act of 2002, hereafter “SOX”)? Using a sample of hand-collected data from 9,196 pro forma earnings press releases from 1998 to 2003, I compare the pro forma earnings announcements of analyst-followed firms (5,037) to the earnings announcements of non-analyst-followed firms (4,159). For the statistical tests summarized in the table below (after removing missing values), the total sample is 8,891, with 4,971 analyst-followed firm announcements and 3,921 non-analyst-followed firm announcements. The abnormal returns test described below uses 9,195 observations, after removing an observation with a missing value. I hypothesize that three-day cumulative abnormal returns surrounding earnings announcement dates for non-analyst-followed firms will generally be larger than abnormal returns for analyst-followed firms (scaled by stock price). I further postulate that analyst-followed firms will be less aggressive in making pro forma adjustments than non-analyst-followed firms, as analyst following provides somewhat of a check on behavior. Finally, I hypothesize that analyst-followed firms will exhibit a greater decrease in pro forma reporting than non-analyst-followed firms. In addition, I hypothesize that non-analyst-followed firms will be less sensitive to SOX than analyst-followed firms. The variables of interest in my tests are as follows:
ScaledReturn = 3DayCumAbReturn (surrounding announcement) / Price (5 days prior)
ManagerTotalAdjustments = ProFormaEPS – Compustat Data Item 181 (GAAP operating eps)
ScaledReturn is used to measure market reaction, while ManagerTotalAdjustments is used to measure “aggressiveness” by management. I find mixed results in testing the market reaction hypothesis. The t-statistic and the Wilcoxon z-score are in opposite directions. More research is needed in regard to market reaction. Future research with this project will involve regressing cumulative abnormal returns on the scaled change in GAAP operating earnings and a Non-IBES (Non-Analyst) indicator variable to obtain a more rigorous test of market reaction, as well as other regression analyses. The results below indicate that non-analyst-followed firms are significantly more aggressive in making pro forma adjustments than analyst-followed firms. This confirms my second hypothesis. I also find that firms have become less aggressive with the passage of the Sarbanes-Oxley Act of 2002. Finally, I find that analyst-followed firms exhibit a greater decrease in pro forma reporting than non-analyst-followed firms, consistent with my third hypothesis.
In conclusion, these results are highly significant and will lead to further study and possible academic publication. Further regression analysis is needed to obtain more insight into market behavior. In addition, a regression analysis in the future will test whether or not managers and analysts make the same adjustments to GAAP operating earnings.
References
- Bhattacharya, N., E.L. Black, T.E. Christensen, and R.D. Mergenthaler. 2006. Who
Trades on Pro Forma Earnings Information? The Accounting Review, Forthcoming. - Frederickson, J.R. and J.S. Miller. 2004. The Effects of Pro Forma Earnings Disclosures on Analysts’ and Nonprofessional Investors’ Equity Valuation Judgments. The Accounting Review, 79: 667-686.
- Bradshaw, M.T and R.G. Sloan. 2002. GAAP versus The Street: An Empirical Assessment of Two Alternative Definitions of Earnings. Journal of Accounting Research, 40: 41-66.
- Bhattacharya, N., E.L. Black, T.E. Christensen, and C.R. Larson. 2003. Assessing the relative informativeness and permanence of pro forma earnings and GAAP operating earnings. Journal of Accounting and Economics, 36: 285-319.