Abstract
Prior literature documents that CEOs are rewarded for any positive component of income and are partially shielded from negative special items. However, the incidence of and rules pertaining to nonrecurring items significantly changed over the last two decades, calling for a reassessment of earlier research. This paper finds that executives now benefit less from positive nonrecurring items and are penalized more for negative special items, compared to earlier periods. The predictive value of the components of income helps explain this shift. Hand-collected data indicates that compensation committees are more likely to include a component of income that can predict future earnings in their CEO bonus performance measures. Changes in the predictive value of the nonrecurring components of income over time contribute to shifts in their treatment in calculating CEO pay.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Current cash compensation is tied to accounting performance in many bonus plans, while the value of long-term noncash compensation is more clearly associated with market performance (e.g., Duru and Iyengar 2001). Also, this paper’s focus on CEO pay is driven by the closer connection between overall firm performance and pay for CEOs, compared to other types of managers.
U.S. Generally Accepted Accounting Principles do not define the term “special items.” Rather, academics use this label for the set of events and transactions U.S. GAAP defines as “infrequent or unusual.” Compustat provides the categories of special items (see Appendix 1).
Appendix Table 9 details the four samples used in this paper.
The inclusion of special item refers to instances in which an executive’s cash compensation can increase when positive special items are reported and decrease when negative special items are reported. The exclusion of special items refers to instances in which special items do not affect an executive’s cash compensation, because the earnings number used to calculate the bonus is some measure taken before the effects of special items. For example, imagine a firm has pretax income of $100 and special items of –$5. If the committee excludes special items, the income number used will be $100 – (−$5) = $105. If the committee includes special items, the income number used will simply be $100. Thus, when special items are negative (positive), it benefits a CEO to have them excluded (included).
The cash component of CEO pay remains important, because it is substantial in magnitude, can be easily understood by CEOs, provides liquidity, and is common in compensation contracts. The average CEO in my current large sample receives about half of his or her annual compensation in cash. According to Murphy and Jensen (2011), incentive plans are effective only if participants understand how their actions affect the payoffs they will receive. Those authors as well as Sloan (1993) claim that the ability of CEOs to adjust earnings is much greater than their ability to adjust stock price. Cash compensation provides more liquidity than restricted stock grants or options. Finally, cash compensation always links CEO pay to accounting earnings, but significantly fewer equity award performance measures do so (Bettis et al. 2015).
That paper uses the term “persistence” for a concept Jones and Smith (2011) later define as predictive value.
Theoretically, compensation committees could adjust for only portions of special items. For example, they could include the impact of one type of special item, like IPRD, but exclude another type, like restructuring. However, I find evidence that this occurs only in a small fraction of a percent of the instances in my hand-collected sample.
The choice of inflation-adjusted cash compensation allows for the most accurate comparison to the findings of Gaver and Gaver. Their data source provided total cash compensation only for most years and not bonus. Running their analysis on bonus alone, during my sample period, indicates no significant and consistent differences (untabulated); thus this switch between bonus and total cash compensation should not be problematic. Using non-inflation-adjusted numbers leads to qualitatively similar results (untabulated).
Given the skew in the distribution of cash compensation, the analysis is repeated after taking the log transformation of the dependent variable. This leads to qualitatively similar results (untabulated).
Special items are included on the basis of their before-tax values. A smaller subset of after-tax special items is available in Compustat, but this variable is significantly less populated than the pretax amount.
Results are qualitatively similar if the cutoff is 1992, the beginning of my current large sample period.
A number of earlier papers use the Cochrane-Orcutt procedure as a substitute for the full transformation (e.g., Natarajan 1996). However, that method would needlessly reduce the size of the sample, given other papers find both methodologies lead to the same results.
There are three main reasons I drop these observations. First, the exclusion of only noncash special items is very rare. Second, untabulated analyses, using a multinomial logit, reveal these are simply driven by industry membership. The remaining results are qualitatively similar. Finally, the results are qualitatively similar when these firms are defined as excluding special items.
Gaver and Gaver’s (1998) data is taken from the Forbes magazine CEO compensation survey of the top 500 firms in terms of either net income, total assets, revenue, or market value. The results from the current period are qualitatively unchanged if only the top 500 firms, as determined by those metrics, are included in the sample.
This requirement may introduce a survivorship bias. While this is not ideal, it is required by the methodology and has the benefit of exposing sustainable compensation decisions by firms.
The final sample also excludes outliers, in line with the methodology of Adut et al. (2003). On the basis of the results from equation 1, any firm with a coefficient on a nonrecurring component of income outside of three standard deviations of the median is eliminated. This method eliminates strong skews in the data. This removes six and 33 firms from the Gaver and Gaver large sample and current large sample, respectively. The percentage of data lost in eliminating the outliers similar to the loss of Adut et al. (2003).
Many firms provide reconciliations from GAAP to non-GAAP numbers used to calculate bonuses. For other firms, the paper uses the definition of income provided in the proxy statement and financial statement numbers to decide whether special items are included or excluded in determining executive compensation. This choice is not always clear and may introduce some noise; however, this is unlikely to bias the results. The dozen observations for which the inclusion or exclusion of special items is unclear are excluded from the sample.
This paper does not consider two types of nonrecurring items—extraordinary items and discontinued operations—given their relatively consistent exclusion from the definition of income used to calculate bonuses. Analyses using the current large sample show no significant link between these items and cash compensation.
The significant findings regarding negative special items only apply to instances when recurring income is positive (untabulated). Bonus plans typically reward CEOs only when firms are profitable. The finding that negative special items are only significant when recurring income is in a relevant range for bonus payouts is supported.
Gaver and Gaver (1998) indicate this finding is driven simply by negative recurring income, making a bonus unlikely. They find that a simple indicator for negative recurring income explains the significant coefficient. I find similar results (untabulated).
Adut et al. (2003) define their TREND variable in a number of ways. Their main analysis uses a growth index, calculated by dividing out-of-sample median compensation by median compensation in the first year. This paper uses their alternative measure, the percentage change in median compensation, relative to the prior year’s median compensation. This approach is necessary because this paper’s research design does not allow for out-of-sample data.
Almost all firms treat any special item reported during a given period in the same manner. The two observations with differential treatment of multiple special items do not allow for analysis and are excluded from the sample.
There is a noticeable increase in the predictive value of positive special items at the very end of the sample period. The question of whether this is sustainable trend or aberration to is left to future research.
Securities and Exchange Commission, Release Nos. 33–6962, 33–6966, and 34–32,723 and IRS Section 162(m).
References
Adut, D., Cready, W., & Lopez, T. (2003). Restructuring charges and CEO cash compensation: A reexamination. The Accounting Review, 78(1), 169–192.
Ali, A., & Zarowin, P. (1992). Permanent versus transitory components of annual earnings and estimation error in earnings response coefficients. Journal of Accounting and Economics, 15, 249–264.
Baber, W., Kang, K., & Kumar, K. (1998). Accounting earnings and executive compensation: The role of earnings persistence. Journal of Accounting and Economics, 25(2), 169–193.
Bartov, E., & Mohanram, P. (2014). Does income statement placement matter to investors? The case of gains/losses from early debt extinguishment. The Accounting Review, 89(6), 2021–2055.
Bettis, J., Bizjak, J., Coles, J., and Kalpathy, S. (2015). Performance vesting provisions in executive compensation. Working paper.
Curtis, A., Li, V., and Patrick, P. (2018). The use of adjusted earnings in performance evaluation. Working paper.
Darrough, M., Guler, L., & Wang, P. (2014). Goodwill impairment losses and CEO compensation. Journal of Accounting, Auditing and Finance, 29(4), 435–463.
Dechow, P., & Sloan, R. (1991). Executive incentives and the horizon problem: An empirical investigation. Journal of Accounting and Economics, 14(1), 51–89.
Dechow, P., Huson, M., & Sloan, R. (1994). The effects of restructuring charges on executives’ cash compensation. The Accounting Review, 69(1), 138–156.
Duru, A., & Iyengar, R. (2001). The relevance of firms’ accounting and market performance for CEO compensation. Journal of Applied Business Research, 17(4), 107–118.
Fairfield, P. M., Sweeney, R. J., & Yohn, T. L. (1996). Accounting classification and the predictive content of earnings. The Accounting Review, 71(3), 337–355.
Fan, Y., Barua, A., Cready, W., & Thomas, W. (2010). Managing earnings using classification shifting: Evidence from quarterly special items. The Accounting Review, 85(4), 1303-1323.
Gaver, J., & Gaver, K. (1998). The relation between nonrecurring accounting transactions and CEO cash compensation. The Accounting Review, 73(2), 235–253.
Healy, P., Kang, S., & Palepu, K. (1987). The effect of accounting procedure on CEOs’ cash salary and bonus compensation. Journal of Accounting and Economics, 9(1), 7–34.
Henry, T., & Holtzman, M. (2007). Extraordinary items share exclusive company. Journal of Accountancy, 203(5), 80–83.
Ittner, C., Larcker, D., & Rajan, M. (1997). The choice of performance measure in annual Bonus contracts. The Accounting Review, 72(2), 231–255.
Johnson, P., Lopez, T., & Sanchez, J. (2011). Special items: A descriptive analysis. Accounting Horizons, 25(3), 511–536.
Jones, D., & Smith, K. (2011). Comparing the value relevance, predictive value, and persistence of other comprehensive income and special items. The Accounting Review, 86(6), 2047–2073.
Kormendi, R., & Lipe, R. (1986). Earnings innovations, earnings persistence, and stock returns. Journal of Business, 60(3), 323–345.
Lambert, R., & Larcker, D. (1987). An analysis of the use of accounting and market measures of performance in executive compensation contracts. Journal of Accounting Research, 25, 85–125.
Lee, Y. (2014). An examination of restructuring charges surrounding the implementation of SFAS 146. Review of Accounting Studies, 19, 539–572.
Murphy, K. (1985). Corporate performance and managerial remuneration. Journal of Accounting and Economics, 7(1–3), 11–42.
Murphy, K., and Jensen, M. (2011). CEO Bonus plans: And how to fix them. Working paper.
Natarajan, R. (1996). Stewardship value of earnings components: Additional evidence on the determinants of executive compensation. The Accounting Review, 71(1), 1–22.
Perry, T., & Zenner, M. (2001). Pay for performance? Government regulation and the structure of compensation contracts. Journal of Financial Economics, 62(2001), 453–488.
Potepa, J. (2018). The drivers of integration charges, the opportunistic use of their reversal, and the impact of SFAS 141(R). Working paper.
Sloan, R. (1993). Accounting earnings and top executive compensation. Journal of Accounting and Economics, 16, 55–100.
Acknowledgements
I would like to acknowledge the thoughts and suggestions of the editor (Patricia Dechow) as well as the anonymous referees. This paper was written as part of my dissertation at Yale University. I am grateful to my committee, Jake Thomas (chair), Kalin Kolev, Alina Lerman, and Frank Zhang, for their recommendations as well as other members of the Yale faculty including Rick Antle, Alan Schwartz, Shyam Sunder, and Tsahi Versano. This paper has benefited from the comments of seminar participants at the AAA Annual Meeting, Baruch, the FDIC, The George Washington University, and Yale University. Finally, I would like to thank Sok-Hyon Kang, Carol Marquardt, Oded Rozenbaum, Yanfeng Xue, and Yanan Zhang for their advice.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix 1
Compustat’s Definition and Categorization of Special Items.
Compustat’s variable special items (spi) corresponds to transactions that GAAP defines as unusual or infrequent but not both. Compustat provides a breakdown of special items into the following nine categories.
- 1.
Acquisition and mergers
- 2.
Gain/loss on the sale of assets
- 3.
Impairments of goodwill
- 4.
Settlements (litigation and insurance)
- 5.
Restructuring costs
- 6.
Write-downs
- 7.
Gain/loss on the early extinguishment of debt
- 8.
In-process research and development
- 9.
Other
Appendix 2
Appendix 3
Appendix 4
Examples from Compensation Contracts
Earnings Number Used in Bonus Calculation Includes the Effect of Special Items
Aptara Proxy Statement – Fiscal Year 2011
AptarGroup’s earnings per share (“EPS”): If diluted EPS equals the average of the EPS of the past three years (“Baseline EPS”), a baseline annual performance incentive of 30% of salary is determined. This baseline annual performance incentive percentage is then increased or decreased by a factor for each 1% increase/decrease above or below the Baseline EPS. For example, if EPS were at or below the Baseline EPS, this element percentage would be between 0% and 30% of salary. If EPS were at or moderately above the Baseline EPS, this element percentage would be expected to be between 30% and 60% of salary. If EPS were significantly above the Baseline EPS, this element percentage would be expected to be between 60% and 90% of salary. Due to AptarGroup achieving all-time high annual diluted earnings per share in 2011 of $2.65, this annual performance incentive element percentage for 2011 was 59% of salary.
(Using the annual report for this firm, it is possible to confirm that they report a diluted EPS of $2.65 and that this figure includes the effect of restructuring charges.)
Earnings Number Used in Bonus Calculation Excludes the Effect of Noncash Special Items
Aegion Corp. Proxy Statement– Fiscal Year 2010
For purposes of the Management Annual Incentive Plan, consolidated net income is determined from our audited financial statements for the year and is adjusted to exclude the following:
• losses associated with the write-down of assets of a discontinued business operation or a business operation to be liquidated,
• gains or losses on the sale of any subsidiary, business unit or division or their assets or business,
• gains or losses on the disposition of material capital assets or the refinancing of indebtedness,
• losses associated with the write-down of goodwill or other intangible assets due to impairment, gains or losses from material property casualty events or condemnation awards,
• other material income or loss, the realization of which is not directly attributable to current senior management,
• any effect from a change in generally accepted accounting principles from those previously used,
• income taxes or benefits of any of the above, and
• any other factors deemed relevant by the Compensation Committee.
…
Additionally, actual award payouts are at the discretion of the Committee and the Plan may be modified, suspended or terminated at any time. Our Committee also has final authority regarding any adjustments to consolidated net income for the purposes of the Plan.
In determining the net income target for the 2010 Management Annual Incentive Plan, the Committee considered the recommendations of executive management regarding current industry and market conditions and projections based on management’s internal market analysis and various market surveys, our 2010 business plan as approved by our Board of Directors in December 2009 and prior year operating results. The net income target for the 2010 Management Annual Incentive Plan was $66,511,000, subject to adjustment in accordance with the Plan. Although the target was considered to be aggressive, it was believed to be set at a level that promoted our compensation objectives.
Earnings Number Used in Bonus Calculation Excludes the Effect of Special Items
Invacare – Fiscal Year 2011
From the proxy statement:
The table below shows, for each of the Company’s three performance goals, (1) the applicable threshold, target and maximum performance level, (2) the percentage of the weighted portion of the executive’s target bonus amount that would be earned at the corresponding performance level and (3) the actual performance level achieved for 2011.
Adjusted Earnings Per Share(1) Weight: 40% of Bonus | Organic Net Sales Weight: 30% of Bonus | % SKU Reduction Weight: 30% of Bonus | ||||
Performance Level | Performance Goal ($) | % of Target Bonus Earned | Performance Goal ($ in millions) | % of Target Bonus Earned | Performance Goal | % of Target Bonus Earned |
<Threshold | <1.85 | 0 | <1722.1 | 0 | <20 | 0 |
Threshold | 1.85 | 2.5 | 1722.1 | 50 | <20 | 0 |
Target | 2.05 | 100 | 1791.0 | 100 | 20 | 100 |
Maximum | 2.20 | 150 | 1825.4 | 150 | 30 | 150 |
2011 Result | 2.05 | 100 | 1751.7 | 71.25 | 20 | 100 |
Overall Funding Level (% of Target) 91.38% | ||||||
-
(1) Adjusted earnings per share is net earnings, excluding the impact of restructuring charges, amortization of the convertible debt discount recorded in interest, asset write-downs related to goodwill and intangibles, loss on debt extinguishment including debt finance charges and fees.
(Using the annual report and earnings release for this firm, it is clear that the earnings per share number provided matches the earnings per share number in financial statements after adjusting for the changes described.)
Appendix 5
Rights and permissions
About this article
Cite this article
Potepa, J. The treatment of special items in determining CEO cash compensation. Rev Account Stud 25, 558–596 (2020). https://doi.org/10.1007/s11142-019-09523-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11142-019-09523-x
Keywords
- Executive compensation
- Executive bonuses
- Income-decreasing special items
- Income-increasing special items
- Extraordinary items
- Discontinued operations