Home | Previous | Next | Huddart Site Map
Options for compensation
Kenneth J Klassen
Volume: 135 Issue: 6
Start Page: 41-44
Subject Terms: Valuation Stock options Compensation plans Incentives Accounting policies
9190: United States
4120: Accounting policies & procedures
6400: Employee benefits & compensation
Geographic Names: United States US Canada
Abstract: Stock options are a material component of executive compensation in both Canada and the US. The average value of stock option grants, valued using the Black-Scholes method (a theory based formula used to value publicly traded options), grew by a factor of six in the 1990s. Stock options are no longer reserved just for executives, as many companies have adopted broader employee option plans. The use of stock options in the US and Canada varies in predictable ways with measures of the firms agency costs. That is, in companies where the agency problem between the shareholders and managers is more severe, more stock options, relative to salary and bonus, are used. There are many who feel the Black-Scholes method of valuing stock options overstates the value of employee stock options.
Full Text: Copyright CANADIAN INSTITUTE OF CHARTERED ACCOUNTANTS Aug 2002
[Headnote] WHILE EMPLOYEE STOCK OPTIONS GROW IN POPULARITY, SOME WONDER IF THE VALUE OF SUCH PLANS IS IN FACT OVERSTATED
Enlarge 200% Enlarge 400%
Average CEO compensation for 1,000 major US corporations
Stock options are a material component of executive compensation in both Canada and the United States. In 1991, one-third of Canada's largest 100 public companies offered stock options to compensate employees; by 1995, that number had grown to two-thirds. In 2000, a review of proxy statements filed with the Toronto Stock Exchange did not reveal any companies in the top 100 not using them.
Average compensation levels - salary and bonus measured in nominal dollars - for US CEOs in the 1990s doubled, but the average value of stock option grants, valued using the Black-Scholes method (a theory based formula used to value publicly traded options), grew by a factor of six. While the average value of these numbers is heavily influenced by some companies that use very large stock option awards, median values, which are not offected by these extreme values, reveal a similar trend.
In 1991, the median salary and bonus was US$670,000 and the median value of stock option grants was US$230,000. By 1999, these values had risen to US$1.13 million and US$1.21 million, respectively.' Canadian data compiled in 1993 through 1995 for 184 firms that granted options reveals the average cash salary and bonus across the three years was $630,000 and average value of stock option grants was $1.67 million.2 More recent Canadian data is difficult to obtain in an electronic form.
Stock options are no longer reserved just for executives, as many companies have adopted broader employee option plans. A random sample of 10 of the 100 largest Canadian companies revealed the top five executives' option grants account for 44% of all company options granted in 2000. There is considerable dispersion, with two companies granting more than 80% of options to the top managers, and six companies granting less than 20% to the top managers.
Companies use stock options for a variety of reasons. In 1979, economist Bengt Holmstrom3 pioneered a model of compensation, which, with subsequent theoretical work,4 begins with two assumptions: the actions of the executive are not observable (and therefore cannot be contracted upon); and the employee is effort-- averse or will not voluntarily act in the best interest of the owners. This creates a setting with a so-called agency problem. When these two conditions exist, standard principal-agent models show it is not possible to motivate the employee with salary alone. Instead, the owner, or principal, must create an incentive contract that varies with observable outcomes. These models have been applied to the executive compensation setting.
Principal-agent models show that when employee compensation is linked to the overall performance of the firm, the employee can be motivated to work harder when the performance measures relate to the employee's actions. Unfortunately, the company performance measures vary with many factors beyond the employee's effort.
When the employee is risk-averse, introducing this incentive contract imposes risks because compensation may be low even when the employee works hard. As a consequence, the employee must be paid more on average to be willing to accept this risk.
This additional cost, or agency cost, can be quite large, particularly when the performance measure contracted upon is weakly related to the employee's efforts. Using multiple measures, each of which depends differently on the unobserved action of the employee, can improve the overall contract efficiency and reduce the agency cost home by the shareholders.5
Incentive contracts come in many forms: bonus arrangements, stock ownership requirements, stock option grants and such hybrids as stock appreciation rights.6 Many firms, consistent with the theoretical models, use a variety of techniques to capture the best of all worlds.
One benefit of a bonus arrangement is that it can be based on several measures, quantitative and qualitative. There is much debate over the appropriate factors to include and general principles such as controllability, incentive alignment and signal-to-noise are frequently cited.7 Accounting numbers are used in the majority of bonus computation calculations, with some form of accounting profit being common.8
Interestingly, there is little evidence of stock price information being explicitly included in bonus contracts.
Using qualitative measures allow compensation committees to use discretion in determining bonus compensation. While introducing uncertainty in the process, discretion allows the committee to mitigate risk from uncontrollable factors that affect the quantitative elements of the bonus.
Given that stock-based compensation is infrequently included in bonus arrangements, stock ownership, stock options and other stock-based compensation are used to ensure that the employee internalizes the desires of the shareholders, that is, for a higher stock price.
Stock options and similar contracts maybe a better form of stock-based compensation because they generally have greater sensitivity to stock price than direct stock ownership, and they avoid the risk of reduced wealth that occurs when the stock price falls. However, one outcome of using stock options over a long period of time may be an increase in share ownership as the options are exercised. A recent study documented that, for 500 public firms in the US, the value of the top five executives' option holdings averaged 1% of the market value of the firm, while the value of their share ownership was 4% of firm value.9
Accounting and finance researchers have documented that the use of stock options in the US and Canada varies in predictable ways with measures of the firm's agency costs. That is, in companies where the agency problem between the shareholders and managers is more severe, more stock options, relative to salary and bonus, are used.10
A review of academic studies" shows there are many theories for stock option use. In addition to aspects of the agency problems described earlier, common assertions relate to the age of the CEO, the stock owned by the manager, industry regulation, liquidity, financial reporting incentives and tax incentives. It should be noted, due to data constraints, the evidence is from larger public companies, which is a severe limitation on our understanding of compensation in smaller and/or private companies.
A CEO's age may affect investment choices. It is hypothesized that as the CEO nears retirement, he or she will forgo valuable longer-term investment, like in R&D, because incentive plans based on accounting performance will penalize the employee in the short run and the successor will receive the benefits. To counteract this horizon problem, greater use of stock-based compensation is suggested for CEOs nearing retirement. This assertion does not seem to be supported by the data.
If the CEO already holds a large amount of his or her personal wealth in the shares of the employer, incentives are hypothesized to already be properly aligned between the executive and the shareholders. Therefore, options are used more frequently when the CEO holds less of the company's stock. Of course, the more options used, the more stock will be held by the CEO over time. The conflicting outcomes seem to balance one another as there is no documented relation between stock ownership by the CEO and stock option grants.
Regulators in industries such as financial services and utilities are believed to reduce managers' discretion, thereby requiring less stock-based compensation to be used. Finance researcher David Yermack found utilities and insurance firms in the US use significantly fewer options than other industries, while banks do not differ from other industries.
Stock options do not require the company to expend cash directly. In fact, if the options are exercised, cash is received. Therefore, cash constrained companies are expected to use stock options to a greater degree. This hypothesis is not supported by compensation studies of public companies.
Financial reporting incentives are considered to be a key reason, at least among accountants, for the rise in stock option use. Stock options are the only form of compensation that do not reduce reported income. Therefore, firms that face greater pressures to have a higher income are hypothesized to use stock options to a greater degree. Studies of stock options that focus on this issue find that this relation exists in the US and Canada. 12
Taxation is also uniquely applied to stock options. In the US, there are several options, some are tax deductible some not. A tax deductible stock option faces approximately the same treatment as other forms of long-term compensation. Consistent with this, there is no evidence taxes influence stock option use in the US.
On the other hand, stock options never give rise to a tax deduction for employers in Canada. Thus, firms that pay taxes (i.e., firms that neither are in a loss position currently nor have loss carry-forwards to eliminate their taxable income) face a trade-off between the benefits of using stock options and the tax deductions they would receive by not using stock options. Canadian data show that such tradeoffs are made, on average.
In summary, there are many assertions for why stock options are used to differing degrees across companies.
Evidence suggests companies use more stock options if they face more severe agency problems or have a greater need to report more income; firms use fewer options, on average, if they are in the regulated utilities or insurance industries or, in Canada, pay taxes currently.
While use of stock options has exploded in the past few decades, another coincident event has been a dramatic rise in stock repurchases by companies.13 In 1998, for the first time total corporate distributions for repurchases in the US exceeded those for dividends.
Several theories have emerged to explain the rise in both repurchases and stock options. The first suggests using stock options, particularly at the levels in the late 1990s, creates significant dilution for many companies. To offset the dilution, companies repurchase their shares. An alternative explanation suggests that managers who have stock options repurchase shares for a self-interested reason: to improve the value of their options. Repurchasing stock has two possible effects on option value. First, by distributing profit with repurchases rather than dividends, the value of the options will rise since it is generally not protected from dividends. Second, a repurchase may be seen as a positive sign by the market participants, and so the value of the company is improved as market participants are more optimistic about the firm's future.
In a subtle, but important, variation on the dilution hypothesis, Scott Weisbenner, in his MIT dissertation, suggests dilution in reported earnings per share is important not dilution of the actual stock ownership.14 Since diluted earnings per share attempt to capture the potential dilution from stock options before they are exercised, firms may try to maintain their earnings per share with repurchases, which would occur earlier than the actual dilution that occurs on exercise.
To assess which explanation is consistent with repurchase behaviour, studies have included a variety of measures.15 Recent ones include variables to capture both total stock options granted to all employees and stock options granted to executives only. Dilution is assumed to be related to all stock options outstanding whereas the executives' incentives will relate to only their own stock option holdings.
While there is evidence of both measures being related to repurchases, the evidence is generally stronger for the relation between executive stock options and repurchases. Thus, a by-product of using stock options to motivate executives appears to be they will also undertake more significant stock repurchases.
In January, public enterprises, cooperatives, deposit-taking institutions and life insurance companies were required to adopt the new Handbook Section 3870 on stock-based payments, including employee stock options. Other enterprises will be required to adopt the standard by next January. The Canadian standard is similar to the US standard issued in 1995. The standards require companies to value options granted using an option valuation model such as Black-Scholes. While the precise method is not prescribed, the method chosen must include the exercise price; the expected life of the option; the current price of the underlying stock and its expected volatility and dividends; and, in Canada, the Canadian risk free rate of equal duration.
There are many who feel the BlackScholes method of valuing stock options overstates the value of employee stock options.16 The ability to trade standard options in a secondary market and the risk neutrality of the option holder are key assumptions in the Black-Scholes method.
When applying the standard, there is considerable judgment in determining many of the formulas' values. For example, the expected life of the option can be as long as the legal life of the option, typically 10 years, or as short as the vesting period for the option, perhaps one year. In such models as the Black-Scholes, computed option values decline as the expected option life becomes smaller. However, the ability to exercise sooner actually makes the option more valuable to the employee.
Obviously, it is too early to find evidence of the new standard's implications for Canadian stock option use. However, trends from the US can inform us about the future in Canada. Early analysis shows few companies following the FASB standard report the expense on their income statements. Most choose the pro forma disclosure approach permitted under both standards. A survey's of 172 annual reports from 1997 showed 30% of the companies indicated no material impact resulted from the application of fair value, 57% reported a pro forma adjustment that was less than 5% decline in net income. Only 13% disclosed a pro forma income number at least 5% less than that otherwise reported. None reported the expense on the income statement.
The academic research discussed above showed companies facing greater need for income tended to use more stock options. It is interesting to speculate if the disclosure of the pro forma income adjustment from stock options will reduce the use of stock options by these companies. If compensation committees believe market participants will view a large pro forma income difference negatively, there will be a reduced reliance on stock options in some firms. However, since there are many important reasons to use stock options to compensate employees, the effect of the new standard on stock options is likely to be small. However, the improvement in the information supplied to users of the financial statement will significantly benefit those trying to understand the impact of stock options on the enterprise.
[Sidebar] Stock options and similar contracts may be a better form of stock-based compensation because they have a greater sensitivity to stock price than direct stock ownership
[Footnote] 1 Data retrieved from the Execucomp database compiled by Standard & Poor's. 2 K. Klassen and A. Mawani, "The impact of financial and tax reporting incentives on option grants to Canadian CEOs," Contemporary Accounting Research, v 17 (2000), pp. 227-262.
[Footnote] 3 B. Holmstrom, "Moral hazard and observability," Bell Journal of Economics, v 10 (1979), pp. 74-91.
[Footnote] 4 Examples of subsequent theoretical work include S. Baiman and J. Demski, "Economically optimal performance evaluation and control systems," Journal of Accounting Research, v 18, supplement (1980), pp. 184220; R. Banker and S. Datar, "Sensitivity, precision and linear aggregation of signals for performance evaluation," Journal of Accounting Research, v 27 (1989), pp. 21-39. 5 A recent review article is a good source for the interested reader: R. Lambert, "Contracting theory and accounting," Journal of Accounting and Economics, v 32 (2001), pp. 3-87. 6 Column on the level of compensation by T. Scott and P. Tiessen, "Paying the boss - and how," CAmagazine, November 1995, pp. 35-38. 7 R. Antle and J. Demski, "The controllability principle in responsibility accounting," The Accounting Review, v 63 (1988), 700-718; Banker and Datar, 1989 (cited in note 4). 8 C. Inner, D. Larcker and M. Rajan, "The choice of performance measures in annual bonus contracts," The Accounting Review, v 72 (1997), pp. 231-255; K. Murphy, "Executive compensation," A. Orley and C. David (eds.) Handbook of Labor Economics, v 3 (1999), Amsterdam: North-Holland. 9 K. Klassen and R. Sivakumar, "Stock repurchases, managerial incentives and stock options," University of Waterloo working paper, 2002.
[Footnote] 10 W. Lewellen, C. Loderer and K. Martin, "Executive compensation and executive in
[Footnote] "Incentive problems," Journal of Accounting and Economics, v 9 (1987), pp. 287-310; R. Sloan, "Accounting earnings and top executive compensation," Journal of Accounting and Economics, v 16 (1993), pp. 55-100; D. Yermack, "Do corporations award CEO stock options effectively?" Journal of Financial Economics, v 39 (1995), pp. 237-269; K. Klassen and A. Mawani, 2000 (cited in note 2). 13 Yermack, 1995 (cited in note 10). 12 For U.S. evidence see S. Matsunaga, "The effect of financial reporting costs on the use of employee stock options," The Accounting Review, v 70 (1995), pp. 1-26; for Canadian evidence see Klassen and Mawani, 2000 (cited in note 2).
[Footnote] 13 J. Liang and S. Sharpe, "Share repurchases and employee stock options and their implications for S&P 500 share retirements and expected returns," Federal Reserve Board Working Paper, 1999; D. Ikenberry, J. Lakonishok and I. Vermaelen, "Stock repurchases in Canada: Performance and strategic trading," journal of Finance, v 55 (2000), pp. 2373-2397. 14 S. Weisbenner, "Corporate share repurchases in the 1990s: What role do stock options play?" University of Illinois Working Paper, 2002; D. Bens, V. Nagar, D. Skinner and E Wong, "Employee stock options, EPS dilution, and stock repurchases," University of Michigan Working Paper, 2002. 15 In addition to those cited in note 9 see C. Jolls, "Stock repurchases and incentive compensation," NBER Working Paper #6467 (1998).
[Footnote] 16 S. Huddart, Employee stock options, Journal of Accounting and Economics, v 18 (1994), pp 207-231. 17 The Pearl Meyer & Partners survey is summarized in "FAS 123 impact only minor, survey reports," Employee Benefit Plan Review, 53 (1998), p. 52.
[Author note] Kenneth J. Klassen, MAcc, PhD, CA, is an associate professor at the School of Accountancy, University of Waterloo, and holds the Deloitte & Touche professorship in taxation
[Author note] Technical Editor: John Friedlan, PhD, CA, is associate professor at the Schulich School of Business at York University
Smeal College of Business, Penn State University, University Park, PA 16802-3603 USA
(814) 863-8393 fax
Home | Previous | Next | Huddart Site Map
was last updated on
Mon, Nov 9, 2015.
Fri, Aug 26, 2016.
Unless otherwise noted, all material is:
Copyright ©1995-2015 Steven Huddart. All rights reserved.