The effects of option incentives on backdating and earnings management.I. INTRODUCTION
CEO (1) (Chief Executive Officer) The highest individual in command of an organization. Typically the president of the company, the CEO reports to the Chairman of the Board. stock options are typically granted at-the-money, where the exercise price of the option is set at the market price on the grant date. Lower share price on the grant date and price appreciation afterwards af·ter·ward also af·ter·wards
At a later time; subsequently.
afterwards or afterward
later [Old English æfterweard]
Adv. 1. can help managers profit from their option compensation. Previous research suggests that managers can pursue a variety of opportunistic opportunistic /op·por·tu·nis·tic/ (op?er-tldbomacn-is´tik)
1. denoting a microorganism which does not ordinarily cause disease but becomes pathogenic under certain circumstances.
2. managerial behaviors to increase the value of their option grants, including backdating Predating a document or instrument prior to the date it was actually drawn. The negotiability of an instrument is not affected by the fact that it is backdated. option grants to periods of lower share prices (Yermack, 1997; Aboody and Kasznik, 2000; Chauvin Chauvin is a common French name.
Shenoy ( Devnagri: शणय ) is a common Surname amongst the Goud Saraswat Brahmins.
Saraswat Brahmins originate from Goa, India. The word 'Shenoy' itself means a temple accountant. Therefore almost every village in Goa had a Shenoy. , 2001; Lie, 2005; Heron and Lie, 2007; and Narayanan Narayanan may mean
To suppress evidence is to keep it from being admitted at trial by showing either that it was illegally obtained or that it is irrelevant. share price when options are granted (Baker, Collins, and Reitenga, 2003; Cheng and Warfield For other uses, see .
Warfield is a village and civil parish in the English county of Berkshire. It is part of the unitary authority of Bracknell Forest. Geography
Warfield is a mostly rural parish made up of a number of small settlements. , 2005; Bergstresser and Philippon, 2006; Cornett For the place in England, see .
The cornett, cornetto or zink is an early wind instrument, dating from the Medieval, Renaissance and Baroque periods. It was used in what are now called alta capellas or wind ensembles. , Marcus Marcus, in the Bible: see Mark, Saint. , and Tehranian, 2008; McAnally, Srivastava, and Weaver
The Weavers are small passerine birds related to the finches.
These are seed-eating birds with rounded conical bills, most of which breed in sub-Saharan Africa, with fewer species in tropical , 2008). In other words Adv. 1. in other words - otherwise stated; "in other words, we are broke"
put differently , newly granted options induce in·duce
1. To bring about or stimulate the occurrence of something, such as labor.
2. To initiate or increase the production of an enzyme or other protein at the level of genetic transcription.
3. CEOs to engage in either option backdating or downwards down·ward
adv. or down·wards
1. In, to, or toward a lower place, level, or position: floating downward.
2. earnings management. But CEOs' option portfolios do not include only newly granted options. In addition to new options granted in the current year, CEOs' option portfolios also have vested vested adj. referring to having an absolute right or title, when previously the holder of the right or title only had an expectation. Examples: after 20 years of employment Larry Loyal's pension rights are now vested. (See: vest, vested remainder) and unvested outstanding options that CEOs retain from their grants in previous years. These two elements can also motivate CEOs to pursue self-interest self-in·ter·est
1. Selfish or excessive regard for one's personal advantage or interest.
2. Personal advantage or interest.
self . Therefore, the purpose of this paper is to examine how various elements of CEOs' option portfolios create conflicting incentives for both earnings management and option backdating, and investigate the link between the two opportunistic behaviors in order to explain the effects.
We examine a sample of 9,568 CEO-year observations in Standard & Poor's Execucomp database from 1992 through 2005, among which 685 CEO years are labeled as the potential backdaters according to according to
1. As stated or indicated by; on the authority of: according to historians.
2. In keeping with: according to instructions.
3. the methodology in Collins, Gong and Li (2009), and 8,883 CEO years are labeled as the benchmark A performance test of hardware and/or software. There are various programs that very accurately test the raw power of a single machine, the interaction in a single client/server system (one server/multiple clients) and the transactions per second in a transaction processing system. firms. We first test the effects of various elements of CEOs' option portfolios on accrual-based earnings management and the occurrence of backdating separately. Given the conflicting effects of various option elements on both earnings management and option backdating, we then examine the link between these two opportunistic actions in order to explain the effects.
Our main findings are as the following. We find that higher incentives from newly granted options are associated with income-decreasing earnings management and higher incidence of backdating, whereas higher incentives from unvested options are related to income-increasing earnings management and lower incidence of backdating. However, incentives from vested options have an insignificant effect on both earnings management and backdating. We further show that negative total accruals Accruals
Accounts on a balance sheet that represent liabilities and non-cash-based assets used in accrual-based accounting. These accounts include, among many others, accounts payable, accounts receivable, goodwill, future tax liability and future interest expense. and discretionary accruals are associated with higher likelihood of backdating, which suggests that CEOs engaging in option backdating are more likely to use income-decreasing earnings management to get more favorable fa·vor·a·ble
1. Advantageous; helpful: favorable winds.
2. Encouraging; propitious: a favorable diagnosis.
3. price for their new option grants as well.
Due to the correlation between downward earnings management and option backdating, the need for managers to diversify diversify
To acquire a variety of assets that do not tend to change in value at the same time. To diversify a securities portfolio is to purchase different types of securities in different companies in unrelated industries. the increased-risk associated with stock-based compensation (Ofek and Yermack, 2000; Cheng and Warfield, 2005) can clearly explain the effects of CEOs' outstanding options on both earnings management and the occurrence of backdating. Those managers with high incentives from unvested options in a given year are more likely to take income-increasing accruals to beat-up short-term stock price when they sell stock to diversify risk (Cheng and Warfield, 2005), and then have disincentives for backdating. However, we further show that the negative relation between unvested options and the occurrence of backdating is largely driven by managers with less persistent incentives from unvested options. Those managers with highly persistent incentives from unvested options are more willing to bear the risk involved in their compensation. As a result, they have lower incentives for upward earnings management for selling purpose and then have a less motivation to reduce the incidence of backdating. In addition, by the time that options become exercisable, vested options likely have a constant hedge ratio Hedge Ratio
1. A ratio comparing the value of a position protected via a hedge with the size of the entire position itself.
2. A ratio comparing the value of futures contracts purchased or sold to the value of the cash commodity being hedged.
1. and generate no needs for further risk diversification Diversification
A risk management technique that mixes a wide variety of investments within a portfolio. It is designed to minimize the impact of any one security on overall portfolio performance.
Diversification is possibly the greatest way to reduce the risk. (Cheng and Warfield, 2005), and produce unclear preferences for the choice of accrual-based earnings management (1). Due to the insignificant effect on earnings management, vested options do not generate a significant effect on backdating either.
Our paper provides new evidence on the conflicting effects of various option elements on both option backdating and earnings management. Especially, we study the occurrence of backdating in previously unexplored situations where managers face conflicting incentives from newly granted options and outstanding options. We are not aware of any previous studies that investigate the occurrence of backdating relate to such conflicting incentives. For example, Collins, Gong and Li (2009) only consider the effect of newly granted options on backdating. Minnick and Zhao Zhao can mean:
Furthermore, our paper extends the literature on backdating and earnings management. The prior studies separately examine backdating and option grant related earnings management. We combine and extend the two strands of literature by providing new evidence that self-serving CEOs are likely to take both opportunistic actions at the same time when they are granted options.
In addition, our findings have implications for corporate executive compensation policies. The evidence that managers take into account the conflicting incentives from their whole option portfolios in determining their opportunistic actions should be of interest to board of directors who contemplate compensation contracts for managers. The findings that indicate multiple opportunistic actions related to option grants are also relevant to the current public policy debate regarding options and executive pay in general.
The paper proceeds as follows. In Section II we briefly review the literature and develop our hypotheses. Section III describes our sample and data. Section IV presents empirical em·pir·i·cal
1. Relying on or derived from observation or experiment.
2. Verifiable or provable by means of observation or experiment.
3. tests and results, and Section V concludes the paper.
II. LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
A. Literature Review
Option grants to CEOs are designed to align align (līn),
v to move the teeth into their proper positions to conform to the line of occlusion. the interests of managers with those of shareholders (Jensen Noun 1. Jensen - modernistic Danish writer (1873-1950)
Johannes Vilhelm Jensen and Meckling, 1976; Jensen and Murphy, 1990). However, the literature also documents that option awards can induce opportunistic behaviors by management. For example, one line of research suggests that managers manipulate manipulate
To cause a security to sell at an artificial price. Although investment bankers are permitted to manipulate temporarily the stock they underwrite, most other forms of manipulation are illegal. the timing of option award dates as a means of increasing the fair value of their awards (Yermack, 1997; Aboody and Kasznik, 2000; Chauvin and Shenoy, 2001; Lie, 2005; Heron and Lie, 2007; and Narayanan and Seyhun, 2006). Specifically, options are found to be awarded prior to increases in stock price, and after decreases in stock price. The other line of research suggests that option grants emerge as a particularly strong predictor of income-decreasing earnings management (Baker, Collins, and Reitenga, 2003; Cheng and Warfield, 2005; Bergstresser and Philippon, 2006; Cornett, Marcus, and Tehranian, 2008; McAnally, Srivastava, and Weaver, 2008), which can directly benefits managers via lower strike price of their option awards.
B. Hypotheses Development
The literature has shown that new option grants create incentives for CEOs to manipulate exercise prices, either through backdating or through downward earnings management. However, the option portfolios that CEOs hold not only include new options that are granted in the current year, but also retain unexercised vested and unvested options that are granted in previous years. These two elements can also motivate CEOs to pursue self-interest and affect the selection of opportunistic managerial actions such as earnings management and option backdating.
The incentives from outstanding options for earnings management can arise from either risk diversifications or option exercises. When a manager is granted options, the options usually are not exercisable until three or four years later. As unvested options cannot be exercised immediately, the effect of unvested options on earnings management through option exercises is unclear. However, managers with high incentives from unvested options have more needs to diversify increased risk associated with stock-based compensation and tend to sell shares. Because of the selling, managers have a motivation to increase the short-term stock price through income-increasing earnings management. Consistent with this argument, Cheng and Warfield (2005) find managers with high incentives from unvested options are more likely to report earnings that meet or just beat analysts' forecast.
Different from unvested options, vested options are exercisable. If managers choose to exercise their vested options, they have a motivation to take income-increasing accruals to beat up short-term stock price, as price appreciation can benefit their option exercises. However, managers can choose to hold their vested options instead and have incentives to manage earnings downward in the current period to reserve good earnings for the future when they decide to exercise. Therefore, the choice of earnings management associated with vested options through option exercises differs across firms and over time, depending on the relative value of the options to be exercised in the current period and the discounted value of options to be exercised in the future (Srivastava, 2005). As a result, the effect of vested options on earnings management through option exercises can be inconsistent Reciprocally contradictory or repugnant.
Things are said to be inconsistent when they are contrary to each other to the extent that one implies the negation of the other. . Similarly, the risk diversification argument also predicts an insignificant association between incentives from vested options and earnings management. It is because that by the time when options become exercisable, options likely have a constant hedge ratio and generate no needs for further risk diversification (Cheng and Warfield, 2005).
In sum, earnings management can not only benefit mangers' newly granted options, but also affect managers' profit from outstanding options. Therefore, we develop the following hypothesis:
Hypothesis 1: Accrual-based earnings management is negatively associated with incentives from newly granted options, and positively associated with incentives from unvested outstanding options. However, there is no significant association between accrual-based earnings management and incentives from vested options.
Next, we want to examine the effects of vested and unvested outstanding options on the occurrence of option backdating. The practice of using hindsight hind·sight
1. Perception of the significance and nature of events after they have occurred.
2. The rear sight of a firearm. to select a date in the past as the option grant date is supposed to affect the new option grants only. However, if option backdating correlates with earnings management, incentives from outstanding options can also affect the incidence of option backdating. On the one hand, if managers can easily backdate back·date
tr.v. back·dat·ed, back·dat·ing, back·dates
To mark or supply with a date that is earlier than the actual date: backdate a check. their options and already secure gains from their option compensation, it is possible that they are less interested in managing earnings. On the other hand, it is also reasonable for self-serving CEOs to take both opportunistic actions at the same time to reap even greater profits from their newly granted options. Especially, those CEOs who engage in backdating are considered to have more influences over board of directors (Collins, Gong, and Li, 2009). Consequently, they can also be more likely to circumvent cir·cum·vent
tr.v. cir·cum·vent·ed, cir·cum·vent·ing, cir·cum·vents
1. To surround (an enemy, for example); enclose or entrap.
2. To go around; bypass: circumvented the city. board monitoring to manipulate earnings for their own benefits. If managers engage in option backdating and downward earnings management at the same time, incentives driven by outstanding options for backdating will be different from those by newly granted options, since they have conflicting effects on earnings management. Specifically, as we test in Hypothesis 1, unvested outstanding options can create a preference for earnings management in a direction opposite to newly granted options, whereas vested outstanding options do not have a significant correlation with earnings management. Therefore, we test the following hypothesis:
Hypothesis 2: The incidence of option backdating is positively associated with incentives from newly granted options, and negatively associated with incentives from unvested outstanding options. However, there is no significant association between the incidence of option backdating and incentives from vested options.
III. SAMPLE AND DATA
To study the effects of CEO option incentives on earnings management and CEO option backdating, we start with a sample of CEOs from the Execucomp database by restricting our attention to nonfinancial Adj. 1. nonfinancial - not involving financial matters
financial, fiscal - involving financial matters; "fiscal responsibility" and non-utilities firms with available data for CEO compensation, tenure, age, and option portfolio data in the years between 1992 and 2005.
We then collect CEO stock option grant data from Thomson Financial Thomson Financial
A major provider of information, analytical tools, and consulting services to the financial community. The firm, a division of Thomson Corporation, is best known to investors for its First Call segment, which publishes consensus earnings Insider Trading database for our initial sample, after eliminating multiple grants that occur on the same date. The Insider Trading database provides data on insider trading activities reported on SEC forms 3, 4, 5, and 144. Following Collins, Gong, and Li (2006), we label a firm in a given year as a potential backdater if the stock price on at least one grant date in that year falls within the lowest decile decile
one of the groups when a series of ranked data is divided into ten equal parts, or dividing points between such groups. See also quartile. over 120 trading days before and 120 trading days after the grant date. A firm in a given year is labeled as a benchmark firm if the stock prices on all grant dates in that year do not fall in the bottom decile of the firm's 240-day stock price distribution.
In addition, we employ accounting data from Compustat '''Standard & Poor's Compustat® is a database of financial, statistical and market information on active and inactive companies throughout the world. Compustat® data has a reputation for extensive coverage, standardization, expertise and timeliness. , stock return data from the Center for Research in Security Prices This article or section needs sources or references that appear in reliable, third-party publications. Alone, primary sources and sources affiliated with the subject of this article are not sufficient for an accurate encyclopedia article. (CRSP CRSP Collaborative Research Support Program (USA)
CRSP Collaborative Research Support Program
CRSP Center for Research in Security Prices
CRSP Center for Research in Security Prices ), board data from the Corporate Library and proxy See proxy server.
(networking) proxy - A process that accepts requests for some service and passes them on to the real server. A proxy may run on dedicated hardware or may be purely software. statements, and institutional ownership data from CDA (1) (Compact Disc Audio) The compact disc file extension that is seen on the computer in Explorer or some other file manager. CDA files are actually pointers to the locations of the individual tracks on the CD medium. See CD-DA. Spectrum. To be included in the final sample, a firm must have data available from all the above sources for a given year. The requirement results in a sample of 9,568 CEO-year observations, among which 685 CEO years are for the potential backdaters and 8,883 CEO years are for the benchmark firms.
A. Measures of Earnings Management
Following the literature (Dechow, Sloan, and Sweeney Sweeney
in poems by T. S. Eliot, symbolizes the sensual, brutal, and materialistic 20th-century man. [Br. Poetry, Benét, 978]
See : Virility , 1995; Bartov, Gul gul
A stylized octagonal motif in Oriental rugs.
[Persian, rose; see julep.] , and Tsui, 2001; Bergstresser and Philippon, 2006; Cornett, Marcus, Tehranian, 2008), we construct total accruals and discretionary accruals as our measures of earnings management.
In order to construct the variable of total accruals, we first calculate earnings before extraordinary items and discontinued operations Discontinued operations
Divisions of a business that have been sold or written off and that no longer are maintained by the business. minus operating cash flows from continuing operations continuing operations
Parts of a business that are expected to be maintained as an ongoing segment of an overall business operation. Income and losses from continuing operations are reported separately if any segments have been discontinued during the (Cornett, Marcus, Tehranian, 2008). We then divide the number by the previous year's assets to obtain the measure of total accruals (Ratio_ta).
After the calculation of total accruals, we use the modified Jones (1991) model to construct the variable of discretionary accruals. Discretionary accruals equal the difference between total accruals and "normal" accruals. The modified Jones model estimates "normal" accruals as a fraction of lagged assets from the following model:
[TA.sub.jt] / [Assets.sub.jt-1] = [[alpha].sub.o] 1 / [Asssts.sub.jt-1] + [[beta].sub.1] [DELTA][Sales.sub.jt] / [Assets.sub.jt-1] + [[beta].sub.2] [PPE PPE (Brit) n abbr (Univ) (= philosophy, politics, and economics) → Studiengang bestehend aus Philosophie, Politologie und Volkswirtschaft
PPE n abbr (BRIT ) (SCOL .sub.jt] / [Assets.sub.jt-1] (1)
where [TA.sub.jt] denotes total accruals for firm j in year t, [Assets.sub.jt-1] denotes total assets for firm j in year t-1, [DELTA][Sales.sub.jt] denotes a change in sales for firm j in year t, and PP[E.sub.jt] denotes property, plant, equipment for firm j in year t. We estimate model (1) by using the firms in Compustat with the same two-digit SIC code as our sample firms in each year of the sample period.
Discretionary accruals then are defined as a fraction of assets as
[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE re·pro·duce
v. re·pro·duced, re·pro·duc·ing, re·pro·duc·es
1. To produce a counterpart, image, or copy of.
2. Biology To generate (offspring) by sexual or asexual means. IN ASCII ASCII or American Standard Code for Information Interchange, a set of codes used to represent letters, numbers, a few symbols, and control characters. Originally designed for teletype operations, it has found wide application in computers. ] (2)
where hats denote de·note
tr.v. de·not·ed, de·not·ing, de·notes
1. To mark; indicate: a frown that denoted increasing impatience.
2. estimated values from model (1). The inclusion of [DELTA][Receivables Receivables
An asset designation applicable to all debts, unsettled transactions or other monetary obligations owed to a company by its debtors or customers. Receivables are recorded by a company's accountants and reported on the balance sheet, and they and include all debts owed .sub.jt] in equation (2) is the "modification" of the Jones (1991) model. This variable attempts to capture the extent to which a change in sales is due to aggressive recognition of questionable sales.
In addition to Ratio_ta and Ratio_da, we also use other variables derived from these two measures of earnings management, in order to capture changes in accruals between years, or to distinguish different directions of accruals. Please see the Appendix for the definition of these variables.
B. Measures of CEO Option Incentives
To examine the effects of the level of option incentives on earnings management and the likelihood of backdating, we use three measures of the pay-for-performance sensitivity: Pps_new, Pps_vest, and Pps_unvest. Pps_new is the pay performance sensitivity of newly granted options, which is the value change in CEO' s newly granted options in the current year for 1000-dollar change in the market value of equity. Pps_vest is the pay performance sensitivity of exercisable outstanding options that are granted from previous years and CEOs retain in their portfolios. Similarly, Pps_unvest is the pay performance sensitivity of unexercisable outstanding options.
To capture the value of option sensitivity, we begin by calculating the partial derivative derivative: see calculus.
In mathematics, a fundamental concept of differential calculus representing the instantaneous rate of change of a function. of individual stock option with respect to one-dollar change in share price (the Black and Scholes Scholes(/skowlz/ or /šowlz/) could refer to the following places:
The rate earned on a riskless asset. is the interest rate on seven-year constant-maturity Treasury bond, obtained from the website of the Federal Reserve Bank of St. Louis Louis, titular duke of Burgundy
Louis, 1682–1712, titular duke of Burgundy; grandson of King Louis XIV of France. He became heir to the throne on the death (1711) of his father, Louis the Great Dauphin. , and the volatility Volatility
1. A statistical measure of the tendency of a market or security to rise or fall sharply within a period of time.
2. A variable in option pricing formulas that denotes the extent to which the return of the underlying asset will fluctuate between now and the (i.e., Volat in the Appendix) is the standard deviation of stock price over the prior sixty months. For previously granted options, we follow Core and Guay's methodology (2002) to estimate the average exercise prices.
C. Measures of Other Governance Governance makes decisions that define expectations, grant power, or verify performance. It consists either of a separate process or of a specific part of management or leadership processes. Sometimes people set up a government to administer these processes and systems. Variables, Firm Characteristics, and CEO Characteristics
In order to examine the effects of CEO option incentives on the incidence of backdating and earnings management, we also control for various firm characteristics, CEO characteristics, and other governance characteristics such as board characteristics, institutional ownership, and CEO ownership, by following the backdating and earnings management literature. The Appendix defines the above variables in details.
In the following analysis, we winsorize all the variables except Ceoown at the top and bottom 1% of the observations, in order to mitigate mit·i·gate
To moderate in force or intensity.
miti·gation n. the inordinate influence of extreme values.
IV. EMPIRICAL ANALYSIS AND RESULTS
Table 1 presents the descriptive statistics descriptive statistics
see statistics. of the key variables for our sample. The five panels provide information on earnings management, CEO option incentives, other governance variables, firm characteristics, and CEO characteristics.
In addition, Table 2 reports correlations between our key variables. Both the correlation matrix Noun 1. correlation matrix - a matrix giving the correlations between all pairs of data sets
statistics - a branch of applied mathematics concerned with the collection and interpretation of quantitative data and the use of probability theory to estimate population and variance The discrepancy between what a party to a lawsuit alleges will be proved in pleadings and what the party actually proves at trial.
In Zoning law, an official permit to use property in a manner that departs from the way in which other property in the same locality influence factors (VIF VIF - VHDL Interface Format. Intermediate language used by the Vantage VHDL compiler. "A VHDL Compiler Based on Attribute Grammar Methodology", R. Farrow et al, SIGPLAN NOtices 24(7):120-130 (Jul 1989). ) do not indicate problematic multicollinearity for our regression regression, in psychology: see defense mechanism.
In statistics, a process for determining a line or curve that best represents the general trend of a data set. models in the following sections.
A. The Effects of Various Option Elements on Earnings Management
We run regressions of accrual-based earning management against variables for incentives from various option elements and the other variables defined in Section III. We estimate the following equation:
[Accruals.sub.it] = [[beta].sub.0] + [[beta].sub.1] [CEOOptionIncentives.sub.it] + [[beta].sub.2] [OtherGovernance.sub.it] + [[beta].sub.3] [FirmCharacteristics.sub.it] + [[beta].sub.4] [CEOCharacteristics.sub.it] + [[epsilon].sub.it] (3)
Variables for option incentives (i.e., Ppsnew, Ppsvest, and Ppsunvest) will enter one at a time before all of them are put together to test their conflicting effects.
Table 3 presents the results. (2) When only Pps_new is included, higher incentives from newly granted options are significantly related to decreasing total accruals and greater decreases in total accruals from the prior three-year average. When only Pps_vest is included, incentives from vested outstanding options are not significantly related to any accrual accrual,
n continually recurring short-term liabilities. Examples are accrued wages, taxes, and interest. measures. When only Pps_unvest is included, on the one hand, higher incentives from unvested outstanding options are significantly correlated cor·re·late
v. cor·re·lat·ed, cor·re·lat·ing, cor·re·lates
1. To put or bring into causal, complementary, parallel, or reciprocal relation.
2. with increasing total accruals, greater increases in total accruals from the prior three-year average, and lower likelihood of negative total accruals. On the other hand, higher incentives from unvested options are also related to lower discretionary accruals and greater decreases in discretionary accruals from the prior three-year average. However, the latter relation is only marginally significant. When all the option incentives are included as in Table 3, Pps_new is significantly related to decreasing total accruals and greater decreases in total accruals from the prior three-year average. Pps_unvest is significantly related to increasing total accruals, greater increase in total accruals from the prior three-year average, and lower likelihood of negative total accruals. At the same time, incentives from vested outstanding options continue to have no significant relationship with any accrual measures. Therefore, the results suggest that on average high incentives from newly granted options decrease earnings and high incentives from unvested options increase earnings via accruals.
The results support our Hypothesis 1. When CEOs are granted new options, downward earnings management can help managers profit from their option compensation. In addition, when CEOs have higher incentives from unvested outstanding options, upward earnings management can help them to sell shares to diversify risk associated with their option portfolios.
B. The Effects of Various Option Elements on Option Backdating
1. Univariate univariate adjective Determined, produced, or caused by only one variable Analysis between Potential Backdaters and the Benchmark Sample
Table 4 reports the univariate comparison for the key variables between potential backdaters (Backdate = 1) and the benchmark sample (Backdate = 0). As can be seen from Panel A of Table 4, potential backdaters generally have more negative values and greater decreases in accruals. For instance, potential backdaters have average (median) total accrual of -7.50% (-6.20%) of assets, which is significantly smaller than -6.60% (-5.80%) for the benchmark firms. Potential backdaters also have significantly higher value of the dummy variable This article is not about "dummy variables" as that term is usually understood in mathematics. See free variables and bound variables.
In regression analysis, a dummy variable for negative total accruals, which indicates more negative values. In addition, compared to the benchmark firms, potential backdaters have greater decreases in accruals from the previous three-year average. Specifically, in terms of discretionary accruals, the mean (median) deviation from the previous three-year average value is -8.9% (-0.8%) of assets, which is significantly smaller than 5.3% (-0.1%) of assets for the benchmark firms.
In addition, according to Panel B, on average, CEOs in potential backdaters have higher incentives from newly granted options (average value of $3.260 vs. $1.780 per $1,000 change in market value of equity, and median value Noun 1. median value - the value below which 50% of the cases fall
statistics - a branch of applied mathematics concerned with the collection and interpretation of quantitative data and the use of probability theory to estimate population of $1.740 vs. $0.780 per $1,000 change in market value of equity) and higher incentives from vested option holdings. Furthermore, according to Panel C and Panel D, potential backdaters have more institutional shareholders, higher institutional ownership, fewer directors on the boards, lower leverage, higher growth, and larger stock return volatility than the benchmark firms. Finally, Panel E shows that CEOs who are in potential backdaters are younger than those in the benchmark firms, but they are not significantly different from each other in terms of the length of time on tenure.
2. Logit The logit function is an important part of logistic regression: for more information, please see that article.
In mathematics, especially as applied in statistics, the logit Analysis of the Effects of Various Option Elements on Option Backdating
To further analyze an·a·lyze
1. To examine methodically by separating into parts and studying their interrelations.
2. To separate a chemical substance into its constituent elements to determine their nature or proportions.
3. the relation between incentives from various option elements and option backdating, we estimate a logit regression model, where the dependent variable (Backdater) equals one if the firm is a potential backdater, and zero if the firm is a non-backdater (i.e. the benchmark firm), and the independent variables are variables for option incentives and the other variables defined in Section III. We estimate the following equation:
P(Backdater) = [[beta].sub.0] + [[beta].suB.1][CEOOptionIncentives.sub.it] + [[beta].sub.2][OtherGovernance.sub.it] + [[beta].sub.3][FirmCharacteristics.sub.it] + [[beta].sub.4][CEOCharacteristics.sub.it] + [[epsilon].sub.it] (4)
Similar as Section IV.A, variables for option incentives (i.e. Ppsnew, Ppsvest, and Ppsunvest) will enter one at a time before all of them are put together to test their conflicting effects.
Table 5 presents the results. It shows that the sensitivity of newly granted option values to changes in stock price is significantly and positively related to the probability of firms' option backdating. The result suggests that as the option-based compensation becomes more important, and greater benefit can be generated from option backdating, CEOs have stronger incentives to engage in option backdating. This is consistent with Minnick and Zhao (2009) and Collins, Gong, and Li (2009). In addition, higher incentives from unvested options are related to lower probability of backdating, whereas incentives from vested options do not have a significant effect. The results are consistent with our Hypothesis 2.
According to Section IV.A, managers with high incentives from unvested options tend to manage earnings upward, which is in the direction opposite to the preference created by newly granted options. Therefore, it is possible that backdating is correlated with income-decreasing earnings management, which creates disincentives for backdating by those managers with high incentives from unvested outstanding options. But managers with high incentives from vested outstanding options do not have a similar disincentive dis·in·cen·tive
Something that prevents or discourages action; a deterrent.
something that discourages someone from behaving or acting in a particular way
Noun 1. for backdating, since vested options are shown to be insignificantly in·sig·nif·i·cant
1. Not significant, especially:
a. Lacking in importance; trivial.
b. Lacking power, position, or value; worthy of little regard.
c. Small in size or amount.
2. correlated with earnings management. We are going to test the implied correlation between backdating and downward earnings management in Section IV.C and IV.D.
Furthermore, among the governance variables, the number of institutional investors is positively related to the likelihood of backdating. It is possible that more institutional investors become interested in purchasing the stock in the year of backdating when stock price goes down due to income-decreasing earnings management. In addition, higher CEO stock ownership is related to a lower likelihood of backdating. This is consistent with the notion that higher ownership improves the incentive alignment Alignment is the adjustment of an object in relation with other objects, or a static orientation of some object or set of objects in relation to others.
C. The Effects of Earnings Management on Option Backdating
In order to confirm the correlation between backdating and income-decreasing earnings management, in this section, we run logit regressions of the likelihood of option backdating against both variables for earnings management and variables for option incentives, as well as other control variables. We estimate the following equation:
P(Backdater) = [[beta].sub.0] + [[beta].sub.1][EarningsManagement.sub.it] + [[beta].sub.2][CEOOptionIncentives.sub.it] + [[beta].sub.3][OtherGovernance.sub.it] + [[beta].sub.4][FirmCharacteristics.sub.it] + [[beta].sub.5][CEOCharacteristics.sub.it] + [[epsilon].sub.it] (5)
Table 6 reports the results. The incidence of option backdating is shown to be positively related to income-decreasing earnings management. As can be seen from regression (3) and (4), backdaters have greater decreases in both discretionary accruals and total accruals from the prior three-year average than the benchmark firms. Higher likelihood of backdating is also shown to be associated with more negative total accruals, according to regression (5). The results suggest that managers engaging in option backdating are also likely to use income-decreasing earnings management to get more favorable price for their new stock option grants as well.
In addition, the effects of various option elements continue to be consistent with our Hypothesis 2. That is, higher probability of backdating is related to greater incentives from newly granted options and smaller incentives from unvested options, whereas incentives from vested options do not have a significant effect on backdating.
D. The Effect of Persistent Unvested Option Incentives vs. Non-Persistent Unvested Option Incentives on Option Backdating
The evidence that managers with high incentives from unvested options are less likely to engage in backdating is consistent with their incentives to manage earnings upward (Stein Stein , William Howard 1911-1980.
American biochemist. He shared a 1972 Nobel Prize for pioneering studies of ribonuclease. , 1989; Cheng and Warfield, 2005). In particular, if CEOs have unusually high incentives from their unvested options in a given year, they are likely to sell stocks to diversify risk. As a result, they have incentives to manage earnings upward to beat up short-term stock price for selling purpose. However, if CEOs have highly persistent incentives from their unvested options, it indicates that they are willing to bear the risk involved in their compensation. As a result, they have lower incentives for upward earnings management because they have a less motivation to sell stocks. Consequently, if option backdating is associated with income-decreasing earnings management, the managers with highly persistent incentives from unvested options should be less likely to avoid backdating, compared to those with non-persistent incentives from unvested options. Therefore, in this section, we distinguish between persistent and non-persistent unvested option incentives and investigate their different effects on backdating.
To distinguish between persistent and non-persistent unvested option incentives, we first rank the pay-for-performance sensitivity of unvested options (Pps_unvest) in each year into percentiles. We then calculate the average ranking for each firm over the sample period. To increase the reliability of the classification, we only consider those firms that have at least five years of data for Pps_unvest over the sample period. If a firm's average ranking is above the top 25% or below the bottom 25% of the average ranking distribution, then that firm is regarded as having highly persistent incentives from unvested options (consistently high or consistently low Pps_unvest, respectively). Under this approach, about 60% of the firm-years exhibit highly persistent Pps_unvest. We use a dummy variable, Recurring re·cur
intr.v. re·curred, re·cur·ring, re·curs
1. To happen, come up, or show up again or repeatedly.
2. To return to one's attention or memory.
3. To return in thought or discourse. , to denote these firm-years. (3) We then add an interaction of this dummy Sham; make-believe; pretended; imitation. Person who serves in place of another, or who serves until the proper person is named or available to take his place (e.g., dummy corporate directors; dummy owners of real estate). with Pps_unvest to the logit regressions for the likelihood of backdating to capture the incremental effect for the firms with highly persistent incentives from unvested options.
Table 7 reports the results. As in previous sections, pay-for-performance sensitivity of unvested options continues to be related to a lower probability of backdating. However, the coefficient coefficient /co·ef·fi·cient/ (ko?ah-fish´int)
1. an expression of the change or effect produced by variation in certain factors, or of the ratio between two different quantities.
2. on the interaction term does not have any significant effects on the incidence of backdating. The results suggest that the negative relation between Pps_unvest and the likelihood of backdating is largely driven by the managers with less persistent incentives from their unvested options. These managers have incentives to manage earnings upward to beat up short-term stock price when they sell stocks to diversify risk. As backdating is shown to be associated with income-decreasing earnings management, they have a disincentive for backdating. Differently, those managers with highly persistent incentives from their unvested options have lower incentives for upward earnings management because they are less likely to sell stocks for risk diversification purpose. As a result, they have a less motivation to reduce the incidence of backdating as well.
In this paper, we study the conflicting incentives from the various elements of CEO option portfolios on both earning management and the occurrence of backdating, and investigate the link between these two opportunistic actions to explain the conflicting effects. We find that higher incentives from newly granted options are associated with income-decreasing earnings management and higher incidence of backdating, whereas higher incentives from unvested options are related to income-increasing earnings management and lower incidence of backdating. However, incentives from vested options have an insignificant effect on both earnings management and backdating. We further show that the correlation between downward earnings management and option backdating, and the need for managers to diversify the increased-risk associated with stock-based compensation can clearly explain the effects of CEOs' outstanding options on both earnings management and the occurrence of backdating.
Our study is the first to examine how conflicting incentives associated with newly granted options and outstanding options affect the occurrence of option backdating. Our paper also combines and extends the literature on backdating and option grant related earnings management, by providing new evidence that self-serving CEOs are likely to take both opportunistic actions at the same time when they are granted options. Our findings have implications for corporate executive compensation policies, and are relevant to the current public policy debate regarding options and executive pay in general.
Appendix Definition of the key variables Variable Definition Panel A: Earnings management Ratio_ta Total accruals/the previous year's assets Ratio_da Discretionary accruals/the previous year's assets Dev_rta Deviation of Ratiota from the prior 3-year average Dev_rda Deviation of Ratioda from the prior 3-year average Neg_ta Dummy equal to unity if total accruals are negative Neg_da Dummy equal to unity if discretionary accruals are negative Panel B: CEO compensation Pps_new The value change in CEO's newly granted options per 1000-dollar change in the market value of equity Pps_vest The value change in CEO's outstanding exercisable options per 1000-dollar change in the market value of equity Pps_unvest The value change in CEO's outstanding unexercisable options per 1000-dollar change in the market value of equity Recurring Dummy equal to unity if average ranking of Pps_unvest over the sample period is in top or bottom 25% of the average ranking distribution Panel C: Other governance variables Ceoown CEOs' holdings of common shares/total shares outstanding Ln_ninst log(number of institutional investors) Instown Total shares held by institutional investors/total shares outstanding Ln_bdsize Log(the number of directors on the board) Pctbdind The proportion of outsiders on the board Duality Dummy equal to unity if the CEO is also the chairman of the board Panel D: Firm characteristics Mve Market value of equity Lev Book value of debt/(book value of debt + market value of equity) Nisd The standard deviation of net income during the three-year period from two years before to the current year Q Market value of assets/book value of assets Ln_ta Log(total assets in million dollars) Volat Standard deviation of daily stock returns over the prior 5 years Panel E: CEO characteristic Age CEO's age Ceotenure The length of time in whole years since the CEO was on the position Panel F: Option backdating Backdate Dummy equal to unity if the firm is a potential backdater in that year
Aboody, D., and R. Kasznik, 2000, "CEO Stock Option Awards and the Timing of Corporate Voluntary Disclosures, "Journal of Accounting and Economics, 29, 73-100.
Arthur Arthur, king of Britain: see Arthurian legend.
king and hero of Scotland, Wales, and England. [Arthurian Legend: Parrinder, 28]
See : Heroism , N., 2001, "Board Composition as the Outcome of an Internal Bargaining Process: Empirical Evidence," Journal of Corporate Finance, 7, 307-340.
Baker, T., D. Collins, and A. Reitenga, 2003, "Stock Option Compensation and Earnings Management Incentives," Journal of Accounting, Auditing and Finance, 18 (Fall), 557-582.
Bartov, E., F. Gul, and J. Tsui, 2001, "Discretionary-Accrual Models and Audit Qualifications," Journal of Accounting and Economics, 30, 421-452.
Bergstresser, D., and T. Philippon, 2006, "CEO Incentives and Earnings Management," Journal of Financial Economics, 80, 511-529.
Black, F., and M. Scholes, 1973, "The Pricing of Options and Corporate Liabilities," Journal of Political Economy, 81, 637-654.
1 City (1990 pop. 25,186), seat of Boone co., central Iowa, on the Des Moines River; inc. 1865. It is a railroad and industrial center with plants making machinery, steel fabrications, and plastic signs. , A., L. Field, J. Karpoff, and C. Raheja, 2007, "The Determinants of Board Size and Composition: an Empirical Analysis," Journal of Financial Economics, 85(1), 66-101.
Chauvin, K., and C. Shenoy, 2001, "Stock Price Decreases Prior to Executive Stock Option Grants," Journal of Corporate Finance, 7, 53-76.
Cheng, Q., and T. Warfield, 2005, "Equity Incentives and Earnings Management," The Accounting Review, 80 (2), 441-476.
Collins, D., G. Gong, and H. Li, 2006, "The Effect of the Sarbanes-Oxley Act See SOX. on the Timing Manipulation Manipulation
Dealing in a security to create a false appearance of active trading, in order to bring in more traders. Illegal. of CEO Option Awards," Working Paper, University of Iowa Not to be confused with Iowa State University.
The first faculty offered instruction at the University in March 1855 to students in the Old Mechanics Building, situated where Seashore Hall is now. In September 1855, the student body numbered 124, of which, 41 were women. .
Collins, D., G. Gong, and H. Li, 2009, "Corporate Governance Corporate Governance
The relationship between all the stakeholders in a company. This includes the shareholders, directors, and management of a company, as defined by the corporate charter, bylaws, formal policy, and rule of law. and Backdating of Executive Stock Options," Contemporary Accounting Research, 26(2), 403-445.
Core, J. E., and W. Guay, 2002, "Estimating the Value of Employee Stock Option Portfolios and Their Sensitivities To Price and Volatility," Journal of Accounting Research, 40, 613-630.
Cornett, M., A. Marcus, and H. Tehranian, 2008, "Corporate Governance and Pay-For-Performance: The Impact of Earnings Management," Journal of Financial Economics, 87, 357-373.
Dechow, P., R. Sloan, and A. Sweeney, 1995, "Detecting Earnings Management," Accounting Review, 70, 193-226.
Del Guercio, D., and J. Hawkins, 1999, "The Motivation and Impact of Pension Fund Activism," Journal of Financial Economics, 52, 293-340.
Finkelstein Finkelstein (פֿינק(ע)לשׁטײַן, פינקלשׁט(י)ין, , S., and D. C. Hambrick, 1988, "Chief Executive Compensation: A Synthesis and Reconsideration re·con·sid·er
v. re·con·sid·ered, re·con·sid·er·ing, re·con·sid·ers
1. To consider again, especially with intent to alter or modify a previous decision.
2. ," Strategic Management Journal, 9, 543-558.
Gibbons Famous people named Gibbons include:
Grinstein Y., and P. Hribar, 2004, "CEO Compensation and Incentives: Evidence from M&A Bonuses," Journal of Financial Economics, 73, 119-143.
Heron, R., and E. Lie, 2007, "Does Backdating Explain the Stock Price Pattern Around Executive Option Grants?" Journal of Financial Economics, 83, 271-295.
Jensen, M. C., 1993, "The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems," Journal of Finance, 48, 831-880.
Jensen, M., and W. Meckling, 1976, "Theory of the Firm: Managerial Behavior, Agency Costs Agency Costs
The costs resulting from an agent performing services for a principal.
Agency costs are generally the commissions earned by agents.
See also: Agency Problem, Agent, Principal
Agency costs and Ownership Structure," Journal of Financial Economics, 3, 305-360.
Jensen, M., and K. Murphy, 1990, "Performance Pay and Top-Management Incentives," Journal of Political Economy, 98, 225-264.
Jones, J. 1991, "Earnings Management during Import Relief Investigations," Journal of Accounting Research, 29, 193-228.
Lie, E., 2005, "On the Timing of CEO Stock Option Awards," Management Science, 51, 802-812.
McConnell McConnell may refer to:
McAnally, M. L., A. Srivastava, and C. Weaver, 2008, "Executive Stock Options, Missed Earnings Targets and Earnings Management," The Accounting Review, 83 (1), 185-216.
Minnick K., and M. Zhao, 2009, "Backdating and Director Incentives: Money or Reputation," Journal of Financial Research, 32(4), 449-477.
Morck, R., A. Shleifer, and R.W. Vishny, 1988, "Management Ownership and Market Valuation," Journal of Financial Economics, 20, 293-315.
Narayanan, M. P., and H.N. Seyhun, 2006, "Effect of Sarbanes-Oxley Act on the Influence of Executive Compensation," Working paper, University of Michigan (body, education) University of Michigan - A large cosmopolitan university in the Midwest USA. Over 50000 students are enrolled at the University of Michigan's three campuses. The students come from 50 states and over 100 foreign countries. .
Nesbitt Nesbitt is a family and place name.
Ofek, E., and D. Yermack, 2000, "Taking Stock: Equity-Based Compensation and the Evolution of Managerial Ownership," Journal of Finance, 55(3), 1367-1384.
Smith, M., 1996, "Shareholder Activism by Institutional Investors: Evidence from CalPERS CalPERS California Public Employees' Retirement System ," Journal of Finance, 51, 227-252.
Srivastava, A., 2005, "Earning Management Choices when Stock Option Incentives Conflicts," Working paper, Texas A&M University
Stein, J. C., 1989, "Efficient Capital Markets, Inefficient Firms: A Model of Myopic my·o·pi·a
1. A visual defect in which distant objects appear blurred because their images are focused in front of the retina rather than on it; nearsightedness. Also called short sight.
2. Corporation Behavior," Quarterly Journal of Economics The Quarterly Journal of Economics, or QJE, is an economics journal published by the Massachusetts Institute of Technology and edited at Harvard University's Department of Economics. Its current editors are Robert J. Barro, Edward L. Glaeser and Lawrence F. Katz. , 104 (4), 655-669.
Sweeney, A. P., 1994, "Debt-Covenant Violations and Managers' Accounting Response," Journal of Accounting and Economics, 17(3), 281-308.
Warfield, T. D., J. J. Wild, and K.L. Wild, 1995, "Managerial Ownership, Accounting Choices, and Informativeness of Earnings," Journal of Accounting and Economics, 20(1), 61-91.
Watts Watts, residential section of south central Los Angeles. Named after C. H. Watts, a Pasadena realtor, the section became part of Los Angeles in 1926. Artist Simon Rodia's celebrated Watts Towers are there. , R. L. and J. L. Zimmerman, 1986, Positive Accounting Theory, Englewood Englewood (ĕng`gəlwd).
1 City (1990 pop. 29,387), Arapahoe co., N central Colo., on the South Platte River, a residential and industrial suburb of Denver; inc. 1903. Cliffs, NJ: Prentice Hall Prentice Hall is a leading educational publisher. It is an imprint of Pearson Education, Inc., based in Upper Saddle River, New Jersey, USA. Prentice Hall publishes print and digital content for the 6-12 and higher education market. History
In 1913, law professor Dr. .
Yermack, D., 1995, "Do Corporations Award CEO Stock Options Effectively?" Journal of Financial Economics, 39, 237-269.
Yermack, D., 1997, "Good Timing: CEO Stock Option Awards and Company News Announcements," Journal of Finance, 52, 449-476.
(1.) The effect of vested options on earnings management can also occur through option exercises. However, the choice of earnings management associated with vested options through option exercises differ across firms and over time and can be inconsistent. Please refer to Section II.B. Hypothesis Development for details.
(2.) Table 3 only shows the results when all the option incentive variables are included. The regression results for Pps_new, Pps_vest, and Pps_unvest separately are available upon request.
(3.) Our empirical results do not essentially change if we only include those firms with consistently high Ppsunvest for Recurring = 1.
Hatice Uzun (a), Yudan Zheng zheng (zhēng),
n a Chinese term for an acupuncture diagnosis achieved by thoroughly examining and interviewing a patient. (b) *
(a) School of Business, Public Administration and Information Sciences Long Island University, H700, 1 University Plaza For the hotel in New York City, see .
Plaza (IPA /'plaθa/ or /'plasa/ , Brooklyn Brooklyn (brk`lĭn), borough of New York City (1990 pop. 2,300,664), 71 sq mi (184 sq km), coextensive with Kings co., SE N.Y. , NY 11201 firstname.lastname@example.org See .edu.
(networking) edu - ("education") The top-level domain for educational establishments in the USA (and some other countries). E.g. "mit.edu". The UK equivalent is "ac.uk".
(b) School of Business, Public Administration and Information Sciences Long Island University, H700, 1 University Plaza, Brooklyn, NY 11201 email@example.com
* The authors thank K.C. Chen (the editor), an anonymous referee A judicial officer who presides over civil hearings but usually does not have the authority or power to render judgment.
Referees are usually appointed by a judge in the district in which the judge presides. , and the participants at the FMA FMA Full Metal Alchemist (gaming)
FMA Federal Marriage Amendment
FMA Financial Market Authority (Austrian: Österreichische Finanzmarktaufsicht)
FMA Financial Management Association 2010 annual meeting in New York New York, state, United States
New York, Middle Atlantic state of the United States. It is bordered by Vermont, Massachusetts, Connecticut, and the Atlantic Ocean (E), New Jersey and Pennsylvania (S), Lakes Erie and Ontario and the Canadian province of , Northeast Business & Economics Association 37th annual meeting in Morristown Morristown.
1 Town (1990 pop. 16,189), seat of Morris co., N N.J., on the Whippany River; settled c.1710, inc. 1865. Although chiefly residential, it has diverse manufactures, including electronic products, health and beauty aids, auto parts, and , NJ, and the Wharton School 2010 Global Conference on Corporate Governance and the Global Financial Crisis, Pennsylvania Pennsylvania (pĕnsəlvā`nyə), one of the Middle Atlantic states of the United States. It is bordered by New Jersey, across the Delaware River (E), Delaware (SE), Maryland (S), West Virginia (SW), Ohio (W), and Lake Erie and New York .
Table 1 Summary statistics The table reports the descriptive statistics for the key variables for the sample over 1992 through 2005. The variables are defined in the Appendix. The major variables are winsorized at the top and bottom 1% of the observations. Mve and Nisd are in million dollars. Variable Obs. Mean Median Std. Dev. Panel A: Earnings management Ratio_ta 9568 -0.067 -0.058 0.085 Ratio_da 9568 0.149 0.011 1.777 Dev_rta 9381 -0.005 0.000 0.089 Dev_rda 9288 0.043 -0.001 2.148 Neg_ta 9568 0.865 1.000 0.342 Neg_da 9568 0.448 0.000 0.497 Panel B: CEO compensation Pps_new 9568 1.8800 0.8400 3.216 Pps_vest 9568 6.2500 3.5500 8.014 Pps unvest 9568 2.2100 0.9500 3.416 Panel C: Other governance variables Ceoown 9568 2.344% 0.000% 6.172% Ln_ninst 9568 4.978 4.949 0.786 Instown 9568 63.844% 65.853% 18.081% Ln_bdsize 9568 2.185 2.197 0.279 Pctbdind 9568 63.767% 66.667% 17.458% Duality 9568 0.657 1.000 0.475 Panel D: Firm characteristics Mve 9568 6370.577 1328.494 16,156,293,600 Lev 9568 0.200 0.148 0.197 Nisd 9568 135.874 30.832 319,823,421 Q 9568 2.126 1.649 2.126 Ln_ta 9568 7.140 6.978 1.503 Volat 9568 0.428 0.382 0.183 Panel E: CEO characteristics Age 9568 55.613 56.000 7.292 Ceotenure 9568 7.411 5.000 7.402 Variable 10% 90% Panel A: Earnings management Ratio_ta -0.154 0.012 Ratio_da -0.189 0.508 Dev_rta -0.095 0.078 Dev_rda -0.691 0.527 Neg_ta 0.000 1.000 Neg_da 0.000 1.000 Panel B: CEO compensation Pps_new 0.000 4.711 Pps_vest 0.000 15.811 Pps unvest 0.000 6.019 Panel C: Other governance variables Ceoown 0.000% 6.800% Ln_ninst 4.025 5.994 Instown 38.331% 85.965% Ln_bdsize 1.792 2.565 Pctbdind 40.000% 85.714% Duality 0.000 1.000 Panel D: Firm characteristics Mve 204,947,000 13,837,100,000 Lev 0.000 0.487 Nisd 4,899,060 326,743,000 Q 1.035 3.661 Ln_ta 5.340 9.260 Volat 0.200 0.700 Panel E: CEO characteristics Age 46.000 64.000 Ceotenure 1.000 18.000 Table 2 Correlation matrix The table reports the correlation matrix for the key variables for the sample over 1992 through 2005. The variables are defined in the Appendix. Ratio_ta Ratio_da Pps_new Pps_vest Ratio ta 1.00 Ratio da 0.04 1.00 Pps new -0.08 0.01 1.00 Pps vest -0.01 0.00 -0.01 1.00 Pps unvest -0.01 0.00 -0.01 0.99 Ceoown 0.04 -0.01 -0.15 0.00 Ln ninst 0.03 0.00 0.15 0.01 Instown 0.05 0.01 0.10 0.00 Ln bdsize 0.07 -0.01 -0.05 0.00 Pctbdind -0.02 0.00 0.08 0.00 Duality 0.04 0.00 -0.04 0.00 Mve -0.01 0.00 0.08 0.00 Lev -0.03 0.00 -0.12 -0.01 Nisd -0.09 0.01 0.06 0.00 Q -0.11 -0.01 0.18 0.00 Ln_ta 0.03 0.00 0.04 0.01 Volat -0.18 0.06 0.21 0.02 Age 0.09 0.00 -0.17 -0.01 Ceotenure 0.06 -0.01 -0.11 0.00 Pps_unvest Ceoown Ln_ninst Instown Ratio ta Ratio da Pps new Pps vest Pps unvest 1.00 Ceoown 0.00 1.00 Ln ninst 0.01 -0.21 1.00 Instown 0.00 -0.26 0.43 1.00 Ln bdsize 0.00 -0.18 0.40 -0.05 Pctbdind 0.00 -0.24 0.20 0.24 Duality 0.00 0.13 0.15 0.04 Mve 0.00 -0.05 0.56 -0.02 Lev 0.00 -0.05 -0.15 -0.08 Nisd 0.00 -0.08 0.43 -0.02 Q 0.00 0.05 0.20 0.01 Ln_ta 0.01 -0.16 0.79 0.21 Volat 0.02 0.03 -0.29 -0.05 Age -0.01 0.13 0.06 -0.01 Ceotenure 0.00 0.39 -0.04 -0.05 Ln_bdsize Pctbdind Duality Mve Ratio ta Ratio da Pps new Pps vest Pps unvest Ceoown Ln ninst Instown Ln bdsize 1.00 Pctbdind 0.09 1.00 Duality 0.11 0.11 1.00 Mve 0.29 0.07 0.09 1.00 Lev 0.19 0.03 0.07 -0.10 Nisd 0.25 0.10 0.07 0.56 Q -0.12 -0.03 -0.03 0.20 Ln_ta 0.55 0.17 0.20 0.55 Volat -0.43 -0.03 -0.17 -0.19 Age 0.13 -0.01 0.28 0.05 Ceotenure -0.08 -0.17 0.25 -0.02 Lev Nisd Q Ln_ta Volat Age Ratio ta Ratio da Pps new Pps vest Pps unvest Ceoown Ln ninst Instown Ln bdsize Pctbdind Duality Mve Lev 1.00 Nisd 0.12 1.00 Q -0.40 -0.03 1.00 Ln_ta 0.28 0.55 -0.19 1.00 Volat -0.01 -0.02 0.10 -0.38 1.00 Age 0.07 0.02 -0.12 0.15 -0.23 1.00 Ceotenure -0.07 -0.08 0.02 -0.06 -0.04 0.40 Ceotenure Ratio ta Ratio da Pps new Pps vest Pps unvest Ceoown Ln ninst Instown Ln bdsize Pctbdind Duality Mve Lev Nisd Q Ln_ta Volat Age Ceotenure 1.00 Table 3 OLS regression of the effects of various option elements on earnings management The table shows the coefficients from OLS regressions of incentives from various option elements on earnings management. The dependent variables and the independent variables are defined in the Appendix. The major variables are winsorized at the top and bottom 1% of the observations. The variance inflation factors (VIF) do not indicate problematic multicollinearity. For each coefficient, p-values for t-tests are provided in parentheses. *, **, *** denote significance at the 0.10, 0.05, and 0.01, levels respectively. Dep. Variables Indep. Ratio_ta Ratio_da Variables Coeff. P-value Coeff. P-value CEO option incentives Pps_new -0.002 *** (0.000) 0.002 (0.692) Pps_vest 0.000 (0.282) -0.002 (0.451) Pps_unvest 0.001** (0.022) -0.007 (0.132) Other governance variables Ceoown 0.041 ** (0.016) 0.190 (0.534) Ln_ninst 0.004 * (0.057) -0.041 (0.269) Instown 0.030 *** (0.000) 0.338 *** (0.007) Ln_bdsize 0.021 *** (0.000) -0.140 * (0.064) Pctbdind -0.018 *** (0.001) 0.288 *** (0.004) Duality 0.002 (0.384) -0.047 (0.239) Firm characteristics Mve 0.443 *** (0.000) 0.075 (0.966) Lev -0.001 (0.846) -0.034 (0.745) Nisd -39.500 *** (0.000) 63.400 (0.462) Q -0.005 *** (0.000) -0.013 (0.277) CEO characteristics Age 0.001 *** (0.000) -0.004 (0.151) Ceotenure 0.000 (0.489) 0.002 (0.444) Intercept -0.162 (0.000) 0.531 (0.020) # of obs. 9904 9568 F statistics 13.07 2.06 Prob. 0.000 0.009 Dep. Variables Indep. Dev_rta Dev_rda Variables Coeff. P-value Coeff. P-value CEO option incentives Pps_new -0.001 *** (0.012) 0.004 (0.584) Pps_vest 0.000 (0.189) 0.000 (0.968) Pps_unvest 0.001 ** (0.046) -0.009 (0.107) Other governance variables Ceoown 0.031 * (0.076) 0.424 (0.340) Ln_ninst 0.006 *** (0.002) -0.057 (0.201) Instown 0.018 *** (0.014) 0.254 (0.105) Ln_bdsize 0.013 *** (0.006) -0.043 (0.643) Pctbdind 0.003 (0.649) 0.238 * (0.059) Duality -0.001 (0.562) -0.07 (0.172) Firm characteristics Mve -0.113 (0.213) 1.170 (0.539) Lev -0.010* (0.084) 0.129 (0.301) Nisd 2.89 (0.635) 14.700 (0.876) Q 0.001 (0.204) -0.009 (0.571) CEO characteristics Age 0.000 (0.657) -0.002 (0.488 Ceotenure 0.000 * (0.055) 0.003 (0.408 Intercept -0.078 (0.000) 0.242 (0.389 # of obs. 9713 9288 F statistics 6.060 1.04 Prob. 0.000 0.409 Dep. Variables Indep. Neg_ta Neg_da Variables Coeff. P-value Coeff. P-value CEO option incentives Pps_new 0.001 (0.543) 0.002 (0.196) Pps_vest 0.000 (0.737) -0.001 (0.367) Pps_unvest -0.003 *** (0.004) 0.001 (0.545) Other governance variables Ceoown -0.033 (0.635) 0.150 (0.120) Ln_ninst 0.053 *** (0.000) 0.006 (0.579) Instown -0.056 ** (0.016) -0.040 (0.255) Ln_bdsize 0.010 (0.514) -0.031 (0.169) Pctbdind 0.091 *** (0.000) 0.022 (0.476) Duality -0.012 (0.140) -0.013 (0.258) Firm characteristics Mve -0.115 (0.675) 0.238 (0.611) Lev 0.054 *** (0.008) -0.015 (0.625) Nisd -28.200 ** (0.023) 21.300 (0.307) Q -0.012 *** (0.000) -0.010 *** (0.010) CEO characteristics Age 0.000 (0.818) -0.002 ** (0.047) Ceotenure -0.001 (0.214) 0.000 (0.939) Intercept 0.583 (0.000) 0.608 (0.000) # of obs. 9904 9568 F statistics 11.96 1.69 Prob. 0.000 0.046 Table 4 Univariate comparison of backdaters vs. non-backdaters The table compares the key variables between the potential backdaters and the benchmark firms over 1992 through 2005. The variables are defined in the Appendix. The major variables are winsorized at the top and bottom 1% of the observations. Mve and Nisd are in million dollars. We use t-test and Wilcoxon rank-sum test to determine mean and median difference between the two sub-samples respectively. *, **, *** denote significance at the 0.10, 0.05, and 0.01 levels. Variable Potential Backdaters Benchmark Obs. Mean Median Obs. Mean Median Panel A: Earnings management Ratio_ta 685 -0.075 -0.062 8883 -0.066 -0.058 Ratio_da 685 0.095 0.011 8883 0.153 0.011 Dev_rta 677 -0.014 -0.003 8704 -0.004 0 Dev_rda 665 -0.089 -0.008 8623 0.053 -0.001 Neg_ta 685 0.904 1 8883 0.862 1 Neg_da 685 0.439 0 8883 0.449 0 Panel B: CEO option incentives Pps_new 685 3.26 1.74 8883 1.78 0.78 Pps_vest 685 6.72 4.13 8883 6.21 3.53 Pps_unvest 685 2.12 0.99 8883 2.22 0.95 Panel C: Other governance variables Ceoown 685 1.88% 0.02% 8883 2.38% 0.00% Ln_ninst 685 5.029 4.977 8883 4.975 4.949 Instown 685 65.98% 68.53% 8883 63.68% 65.61% Ln_bdsize 685 2.142 2.197 8883 2.188 2.197 Pctbdind 685 63.91% 66.67% 8883 63.76% 66.67% Duality 685 0.653 1 8883 0.657 1 Panel D: Firm characteristics Mve 685 6046.795 1397.663 8883 6395.545 1324.14 Lev 685 0.18 0.132 8883 0.202 0.15 Nisd 685 136.955 33.215 8883 135.79 30.607 Q 685 2.279 1.735 8883 2.114 1.643 Ln_ta 685 7.114 6.975 8883 7.142 6.98 Volat 685 0.483 0.445 8883 0.424 0.379 Panel E: CEO characteristics Age 685 54.809 55 8883 55.675 56 Ceotenure 685 7.045 5 8883 7.439 5 Variable p-values for difference tests Mean Median Panel A: Earnings management Ratio_ta 0.008 *** 0.010 *** Ratio_da 0.411 0.635 Dev_rta 0.006 *** 0.033 ** Dev_rda 0.101 * 0.048 ** Neg_ta 0.000 *** 0.002 *** Neg_da 0.641 0.641 Panel B: CEO option incentives Pps_new 0.000 *** <.0001 *** Pps_vest 0.108 0.035 ** Pps_unvest 0.47 0.776 Panel C: Other governance variables Ceoown 0.015 ** 0.814 Ln_ninst 0.064 * 0.055 * Instown 0.001 *** 0.001 *** Ln_bdsize 0.000 *** <.0001 *** Pctbdind 0.82 0.773 Duality 0.814 0.814 Panel D: Firm characteristics Mve 0.586 0.203 Lev 0.005 *** 0.003 *** Nisd 0.927 0.356 Q 0.008 *** 0.005 *** Ln_ta 0.639 0.721 Volat 0.000 *** <.0001 *** Panel E: CEO characteristics Age 0.003 *** 0.015 ** Ceotenure 0.141 0.612 Table 5 Logit analysis of the effect of various option elements on the likelihood of backdating The table shows the coefficients from a logit regression of incentives from various option elements on the likelihood of backdating. The dependent variable equals one if the firm is a potential backdater, and zero if the firm is a non-backdater (i.e. the benchmark firm). The independent variables are defined in the Appendix. The major variables are winsorized at the top and bottom 1% of the observations. The models are estimated by maximum likelihood method. The variance inflation factors (VIF) do not indicate problematic multicollinearity. For each coefficient, p-values for t-tests are provided in parentheses. *, **, *** denote significance at the 0.10, 0.05, and 0.01 levels, respectively. Indep 1 2 Variables Coeff. P-value Coeff. P-value CEO option incentives Pps_new 0.083 *** (0.000) Pps_vest 0.000 (0.981) Pps_unvest Other governance variables Ceoown -0.017 ** (0.051) -0.019 ** (0.034) Ln_ninst 0.285 *** (0.002) 0.218 ** (0.019) Instown 0.300 (0.223) 0.326 (0.186) Ln_bdsize -0.280 (0.126) -0.363 ** (0.045) Pctbdind -0.183 (0.444) -0.263 (0.267) Duality 0.125 (0.165) 0.149 * (0.096) Firm characteristics Ln_ta 0.000 (0.998) -0.012 (0.816) Volat 1.241 *** (0.000) 1.531 *** (0.000) CEO characteristics Age -0.004 (0.530) -0.007 (0.277) Ceotenure 0.004 (0.527) 0.001 (0.897) Intercept -4.070 (0.000) -3.206 (0.000) # of obs 9944 9943 Chi-square 169.920 94.480 Prob 0.000 0.000 Indep 3 4 Variables Coeff. P-value Coeff. P-value CEO option incentives Pps_new 0.083 *** (0.000) Pps_vest -0.002 (0.712) Pps_unvest -0.033 *** (0.011) -0.028 ** (0.029) Other governance variables Ceoown -0.020 ** (0.025) -0.018 ** (0.043) Ln_ninst 0.215 ** (0.020) 0.282 *** (0.003) Instown 0.362 (0.140) 0.347 (0.163) Ln_bdsize -0.391 ** (0.031) -0.307 * (0.094) Pctbdind -0.252 (0.287) -0.172 (0.472) Duality 0.155 * (0.082) 0.134 (0.139) Firm characteristics Ln_ta -0.024 (0.632) -0.012 (0.810) Volat 1.575 *** (0.000) 1.281 *** (0.000) CEO characteristics Age -0.008 (0.198) -0.005 (0.422) Ceotenure 0.001 (0.859) 0.005 (0.460) Intercept -2.955 (0.000) -3.839 (0.000) # of obs 9942 9942 Chi-square 101.620 175.710 Prob 0.000 0.000 Table 6 Logit analysis of the effects of earnings management on the likelihood of backdating (*, **, *** significance at the 0.10, 0.05, and 0.01 level) 1 2 Indep. Coeff. P-value Coeff. P-value Variables Earnings management Ratio_ta 0.049 (0.910) Ratio_da -0.030 (0.181) Dev_rta Dev_rda Neg_ta Neg da CEO compensation Pps_new 0.083 *** (0.000) 0.084 *** (0.000) Pps_vest -0.002 (0.711) -0.002 (0.655) Pps_unvest -0.028 ** (0.029) -0.026 ** (0.046) Other governance variables Ceoown -0.018 ** (0.043) -0.013 (0.134) Ln_ninst 0.281 *** (0.003) 0.271 *** (0.004) Instown 0.349 (0.163) 0.349 (0.169) Ln_bdsize -0.307 * (0.094) -0.313 * (0.094) Pctbdind -0.173 (0.471) -0.147 (0.548) Duality 0.134 (0.137) 0.117 (0.203) Firm characteristics Ln_ta -0.011 (0.824) -0.003 (0.960) Volat 1.285 *** (0.000) 1.310 *** (0.000) CEO characteristics Age -0.005 (0.417) -0.005 (0.428) Ceotenure 0.005 (0.464) 0.001 (0.862) Intercept -3.834 (0.000) -3.848 (0.000) # of obs. 9933 9596 Chi-square 175.570 172.830 Prob. 0.000 0.000 3 4 Indep. Coeff. P-value Coeff. P-value Variables Earnings management Ratio_ta Ratio_da Dev_rta -0.811 ** (0.044) Dev_rda -0.033 * (0.090) Neg_ta Neg da CEO compensation Pps_new 0.083 *** (0.000) 0.083 *** (0.000) Pps_vest -0.002 (0.696) -0.003 (0.526) Pps_unvest -0.027 ** (0.042) -0.028 ** (0.042) Other governance variables Ceoown -0.021 ** (-0.021) -0.017 * (0.070) Ln_ninst 0.295 (0.295) 0.294 *** (0.003) Instown 0.296 (0.296) 0.228 (0.376) Ln_bdsize -0.395 (-0.395) -0.427 ** (0.025) Pctbdind -0.209 (-0.209) -0.149 (0.549) Duality 0.138 (0.138) 0.113 (0.225) Firm characteristics Ln_ta -0.012 (0.815) -0.011 (0.834) Volat 1.207 *** (0.000) 1.292 *** (0.000) CEO characteristics Age -0.004 (0.518) -0.003 (0.611) Ceotenure 0.003 (0.683) -0.003 (0.719) Intercept -3.654 (0.000) -3.613 (0.000) # of obs. 9741 9315 Chi-square 174.420 166.630 Prob. 0.000 0.000 5 6 Indep. Coeff. P-value Coeff. P-value Variables Earnings management Ratio_ta Ratio_da Dev_rta Dev_rda Neg_ta 0.311 ** (0.015) Neg da -0.051 (0.531) CEO compensation Pps_new 0.083 *** (0.000) 0.085 *** (0.000) Pps_vest -0.002 (0.721) -0.002 (0.651) Pps_unvest -0.027 ** (0.037) -0.026 ** (0.048) Other governance variables Ceoown -0.018 ** (0.046) -0.013 (0.135) Ln_ninst 0.283 *** (0.002) 0.272 *** (0.004) Instown 0.356 (0.153) 0.339 (0.182) Ln_bdsize -0.311 * (0.090) -0.315 * (0.092) Pctbdind -0.200 (0.404) -0.153 (0.533) Duality 0.137 (0.130) 0.118 (0.199) Firm characteristics Ln_ta -0.018 (0.714) -0.003 (0.960) Volat 1.255 *** (0.000) 1.289 *** (0.000) CEO characteristics Age -0.005 (0.413) -0.005 (0.421) Ceotenure 0.005 (0.454) 0.001 (0.865) Intercept -4.035 (0.000) -3.807 (0.000) # of obs. 9933 9596 Chi-square 181.870 171.380 Prob. 0.000 0.000 Table 7 Logit analysis of the effect of persistent unvested option incentives vs. non-persistent unvested option incentives on the likelihood of backdating The table shows the coefficients from a logit regression of the effect of incentives from persistent vs. non-persistent unvested options on the likelihood of backdating. The dependent variable equals one if the firm is a potential backdater, and zero if the firm is a non-backdater (i.e. the benchmark firm). The independent variables are defined in the Appendix. The major variables are winsorized at the top and bottom 1% of the observations. The models are estimated by maximum likelihood method. The variance inflation factors (VIF) do not indicate problematic multicollinearity. For each coefficient, p-values for t-tests are provided in parentheses. *, **, *** denote significance at the 0.10, 0.05, and 0.01 levels respectively. 1 2 Indep. Coeff. P-value Coeff. P-value Variables Earnings management Ratio_da 0.065 (0.884) -0.027 (0.229) Dev_rta Dev_rda Neg_ta Neg_da CEO compensation Pps_new 0.085 *** (0.000) 0.086 *** (0.000) Pps_vest -0.001 (0.791) -0.002 (0.712) Pps_unvest -0.032 ** (0.044) -0.029 * (0.077) Recurring * Pps_unvest 0.001 (0.913) -0.001 (0.952) Other governance variables Ceoown -0.021 ** (0.025) -0.016 * (0.085) Ln_ninst 0.268 *** (0.005) 0.267 *** (0.006) Instown 0.342 (0.182) 0.324 (0.212) Ln_bdsize -0.309* (0.101) -0.313 * (0.102) Pctbdind -0.171 (0.486) -0.122 (0.627) Duality 0.150 (0.106) 0.134 (0.155) Firm characteristics Ln_ta -0.011 (0.838) 0.007 (0.902) Volat 1.295 *** (0.000) 1.324 *** (0.000) CEO characteristics Age -0.006 (0.324) -0.006 (0.335) Ceotenure 0.005 (0.495) 0.001 (0.860) Intercept -3.698 (0.000) -3.738 (0.000) # of obs. 9479.000 0.335 Chi-square 171.890 0.860 Prob. 0.000 0.000 3 4 Indep. Coeff. P-value Coeff. P-value Variables Earnings management Ratio_da Dev_rta -0.693 * (0.094) Dev_rda -0.033 * (0.097) Neg_ta Neg_da CEO compensation Pps_new 0.085 *** (0.000) 0.085 *** (0.000) Pps_vest -0.002 (0.762) -0.003 (0.548) Pps_unvest -0.030 * (0.060) -0.030 * (0.073) Recurring * Pps_unvest 0.001 (0.940) -0.001 (0.951) Other governance variables Ceoown -0.021 ** (0.026) -0.017 * (0.077) Ln_ninst 0.285 *** (0.003) 0.290 *** (0.003) Instown 0.296 (0.249) 0.225 (0.394) Ln_bdsize -0.389 ** (0.040) -0.416 ** (0.033) Pctbdind -0.198 (0.422) -0.113 (0.659) Duality 0.150 (0.108) 0.124 (0.192) Firm characteristics Ln_ta -0.013 (0.801) -0.016 (0.770) Volat 1.223 *** (0.000) 1.307 *** (0.000) CEO characteristics Age -0.005 (0.413) -0.004 (0.493) Ceotenure 0.002 (0.725) -0.003 (0.723) Intercept -3.558 (0.000) -3.547 (0.000) # of obs. 9325.000 8926.000 Chi-square 171.820 164.790 Prob. 0.000 0.000 5 6 Indep. Coeff. P-value Coeff. P-value Variables Earnings management Ratio_da Dev_rta Dev_rda Neg_ta 0.327 *** (0.013) Neg_da -0.062 (0.455) CEO compensation Pps_new 0.084 *** (0.000) 0.086 *** (0.000) Pps_vest -0.001 (0.794) -0.002 (0.715) Pps_unvest -0.030 ** (0.054) -0.028 * (0.078) Recurring * Pps_unvest 0.001 (0.910) -0.001 (0.953) Other governance variables Ceoown -0.021 ** (0.026) -0.016 * (0.085) Ln_ninst 0.272 *** (0.004) 0.267 *** (0.006) Instown 0.349 (0.171) 0.313 (0.229) Ln_bdsize -0.313 * (0.096) -0.316 * (0.100) Pctbdind -0.201 (0.414) -0.128 (0.611) Duality 0.153 * (0.099) 0.135 (0.151) Firm characteristics Ln_ta -0.019 (0.714) -0.006 (0.906) Volat 1.264 *** (0.000) 1.304 *** (0.000) CEO characteristics Age -0.006 (0.325) -0.006 (0.327) Ceotenure 0.005 (0.481) 0.001 (0.862) Intercept -3.918 (0.000) -3.690 (0.000) # of obs. 9479.000 9168.000 Chi-square 178.460 168.190 Prob. 0.000 0.000