Employee Stock Option Incentives

(1)

Employee Stock Option Incentives*

Bernard Schipper University of Amsterdam

July 18, 2007

__________________________________________________________________________________________ * I would like to thank David Clifford, Dr. Zacharias Sautner and an anonymous referee for their helpful discussions and comments.

(2)

Employee Stock Option Incentives

Abstract

I investigate Employee Stock Option (abbreviated ESO) plan participation and Tax-Free Savings (abbreviated TFS) plan participation using a unique dataset consisting of 23,000 employees working for a Dutch Financial Institution. I find that participation in both ESO plans and TFS plans is coupled with incentives for the employees. Employees participating in either plan are better retained by the company. Employees in higher salary scales seem to be looking for compensation using ESO plans and TFS plans. And finally, performance of participating employees is significantly better than performance of non-participating colleagues. Inconsistent with traditional ESO theories, no link could be found between attraction and ESO or TFS participation. Taken together, my findings suggest that ESO and/or TFS participation may play a substantial role in motivating and retaining employees of all salary levels, while ESO participation motivates higher salaried workers to create value for the company.

(3)

List of Tables

1a Predicted Relationships between Participation Behaviour and Rational and Psychological

Factors 8

1b Predicted Relationships between Participation Behaviour and Subject Variables 8

2 Definition of Variables 15

3 Descriptive Statistics 16

4 Correlation 22

5 Differences between Groups - ESO Participation Behaviour of Employees 23 6 Differences between Groups - TFS Participation Behaviour of Employees 24 7 Retention and ESO Plan and/or Tax-Free Savings Plan Participation 25 8 Performance and ESO Plan and/or Tax-Free Savings Plan Participation 26 9 ESO Plan and/or Tax-Free Savings Plan Participation 27

(5)

1 Introduction

Employee stock option (hereafter referred to as “ESO”) plans have important financial consequences for companies issuing them and employees receiving them. From the issuer perspective, ESO plans are expensive to setup and require resources to maintain the plan for the duration of the program. In addition to this, the FASB has decided in December 2004 to revise SFAS 123 and require companies to expense granted compensation costs for employee stock options which significantly impacts the income statement according to Botosan and Plumlee (2001). From the employee perspective, it has been shown by Hall and Murphy (2003) that ESO may amount to a significant part of compensation, even as Sautner and Weber (2005) argue it is difficult for employees to assess the true value of the options awarded to them.

Since 1994, many employees in the Netherlands have been eligible to participate in a tax-free savings (hereafter referred to as “TFS”) plan. This is essentially risk-free and has proved to be quite popular among Dutch employees. In recent years, employees at some companies have had the choice to forego the risk-free tax-advantage and use the quota of the TFS plans to buy into an ESO program. The ESO plans are risky as their payout is dependent on the company’s future share price, which is uncertain. It is therefore relevant to see why employees actually participate in offered ESO plans and interesting to test for proof of existence of incentives effects. Designers of ESO plans should be well aware of the incentives for employers and employees. ESO ownership has been associated with many incentive effects, such as attracting personnel, retention of personnel, compensation of personnel and improvement of performance of personnel. A lot of research has been done to assess the impact of ESO plan ownership on these incentive effects.

An important argument is that the availability of ESOs can attract motivated and entrepreneurial employees. Companies effectively borrow from employees; they receive employment services today in return for uncertain and variable payouts in the future. The argument is made that short term cash flows are not hampered since there is no immediate cash requirement for the granted stock options. This would be especially beneficial for companies in a phase of high growth. According to Hall and Murphy (2003), this form of financing is very costly as the cost of options granted to employees exceeds the value employees attribute to them. Opponents have also pointed out that risk-averse undiversified employees are not likely to be an efficient provider of capital. I will test to see if new employees are inclined to buy into ESO plans, controlling for other variables such as age, gender,

(6)

possibilities to execute stock options is worth considerable less in this case. Moreover, retention incentives only last as long as the vesting period. After the vesting period ends employees are free to execute their options; they can do so at any time and start working for a new employer, evaporating retention benefits. Companies might then be obliged to install a new ESO plan to keep incentive benefits. Gonsalez and Gurtoviy (2006) have argued that deferred payments alone cannot keep qualified personnel at a firm, in some circumstances a cash component needs to be included as well to retain personnel. As the employees of the company on which we have a data sample will not forfeit their options in case they leave the company, retention incentives are expected to be lower than shown in publications. We will however be able to compare ESO participation and TFS participation of employees who leave the company and employees who do stay on, and see if a preference exists for either plan.

The third and perhaps most important argument in favour of ESO plans is that they give senior management a greater incentive to act in the interest of shareholders by providing a direct link between compensation and company stock-price performance. Agency problems between shareholders and managers are mitigated by linking their compensation directly to stock-price performance. Hall and Murphy (2003) find that as most options are granted to employees well below executive level, this incentive might be true for top-level employees but will not hold for shares granted to lower level employees. I will test to see whether senior management is indeed more inclined to use buy into stock option, also when controlling for other variables such as age, gender and years working in the firm.

The fourth incentive coupled with ESO plans is improvement of performance for employees who participate in ESO plans. Rationally, it will be hard to argue that lower level employees are motivated by the ESOs they received as they are unlikely to be able to influence the share price. They generally have little influence on the overall execution part of a business, and are unable to make risk-seeking choices which are required to improve the share price. Top level management however, might be encouraged to take on risks they would normally not take. Without ESO plans they would be comfortable with their compensation packages. If part of this pay package is deferred and largely dependent on extraordinary results to drive the share price, top level management with ESO grants will be more motivated to improve performance. I will see if evidence exists to back-up these assertions.

My research has not provided any evidence that employees are attracted by the ESO plan or by the TFS plan. It seems that participation for both is stronger for employees with more years at the firm. Also when we look specifically at employees that recently started, there seems to be a negative relation between employee ESO and TFS participation and employees new to the firm.

(7)

There seems to be evidence that employees are retained by participating in the ESO plan. Employees who left voluntarily did not participate quite as often as non-leavers. There is a negative relation between voluntary leavers and TFS participation. This negative relation is quite pronounced.

From a compensation perspective, when controlling for additional parameters we found salary scale does have a large impact on ESO participation. A higher salary scale does also increase the chances of TFS participation, but not to the extent as shown for ESO participation. Higher salaried employees might feel they are in the position to improve company performance and share price.

From a performance perspective, when controlling for additional parameters we found ESO participation has a large impact on assessment scores. TFS participation does also increase the chances of a high assessment score, very much equal to the result shown for ESO participation. It is apparently very motivating to take part in either and ESO or a TFS plan, this might be due to loyalty of the employee as a result of participating or perhaps due to the fact that employees truly believe they can improve the share price. There seems no doubt that ESO plan and TFS plan participation are very beneficiary for the performance of the company.

The remainder of the paper is structured as follows: section 2 will provide an overview of available literature on predicted relationships between participation behavior and rational and psychological factors, as well as predictions on behavioral and dataset variables. Section 3 looks at the research problem, research goals and testable hypotheses. Section 4 describes the dataset. Section 5 contains the empirical analysis and results on the determinants of ESO and TFS participation. The conclusion is in section 6, together with recommendations for further research.

(8)

2 ESO Participation Literature

A lot of papers have been written on ESO incentives and employee behaviour. The main argument for stock option plans is they give management incentives to act in the interest of shareholders by providing a direct link between stock performance and compensation. ESO plans are also set up to attract motivated employees without having to immediately expense cash. A third benefit of ESO plans is that they are usually structured in a way that employees can only benefit from them over a longer period of time. They therefore bring retention benefits. A fourth advantage is thought to be that ESO plans encourage managers to take on risk they would normally not take. Managers with ESO compensation are more dependent on either volatility to drive the option price up or dependent on extraordinary results to drive the share price. The share price generally should include all expected performance. Hall and Murphy (2003) have argued that stock options are granted to too many people, mostly below top-executive level. ESO plans are an inefficient way to attract, retain and motivate executives and (especially) lower-level employees. Lower level employees have no influence on firm performance and therefore do not have the performance incentives of top management. Due to accounting and cash-flow considerations they are falsely thought to be inexpensive by board and managers.

As I will be testing for ESO and TFS incentives, I’ve made an overview of predicted relationships: between Participation Behaviour and Rational and Psychological Factors. It is shown in table 1a: Predicted Relationships between Participation Behaviour and Rational and Psychological Factors. I would like to use this to further discuss available literature. I will use table 1b: Predicted Relationships between Participation Behaviour and Subject Variables to show what employee behaviour one can expect when looking at the company of which we have obtained a dataset and keeping the participation rules for these specific employees in mind.

Traditional theories have assumed that financial markets can be described using models in which agents act fully rational. When looking at rational explanations for participations in ESO plans TFS plans, it is clear that risk-averse employees will be more inclined to participate in the TFS plan than participate in the ESO plan. For this there are two main reasons; the first being that risk-averse employees do not like to have the uncertainty regarding the payout of the ESO plan. Furthermore, they do not like to participate in ESO plans as these are non-transferable. There is no market for them from which the value can be assumed and employees who hold ESOs are usually unable to fully diversify their holdings and human capital investments. Bettis et al (2004) have argued that it is difficulty to understand and value of ESO plans and incentives to both employers and employees as holdings in ESO plans cannot be transferred. Using a large dataset of 140,000 ESO exercises, they have found that

(9)

early exercise is common as exercise is usually executed just over two years after the vesting period and 4 years prior to the expiration date. There are differences between exercise behaviour depending on the volatility of the underlying stock, dividend yields and/or depending on the recent performance of the share. Lessons learned: a risk-averse employee is expected to exercise early to diversify her of his holdings, especially in case the volatility of the stock is greater. Exercised options forego a substantial part of the remaining option-life as a result.

The influence of stock holdings on ESO participation is also researched. People who hold a larger portion of their wealth in company stock will exercise early (Lambert et al, 2001), similarly, people with a large part in company stock are less likely to participate in ESO plans and will rationally go for the TFS plan. As shown by Lambert et al (2004), the value of a stock option can be significantly less when a large share of an employee’s wealth is in company stock. On the other hand, rational behaving poor employees cannot forego the certain tax-advantage and must choose the TFS plan when the vesting period is equal. With regards to firm specific human capital, employees who have a lot of firm-specific human capital would rationally like to diversify away from this and choose the TFS plan participation instead of the ESO plan.

Rationally, these findings are true and sometimes supported by empirical evidence. However, Barberis and Thaler (2002) argue that behavioural finance shows economic phenomena can often better be understood when one assumes that agents are not always acting fully rational. According to them there are two pillars of behavioural finance, one being ‘psychology’ and the other being ‘limits to arbitrage’. Psychology categorizes the various ways investors can behave irrationally. Limits to arbitrage discusses the inability of investors to arbitrage away the irrational behaviour of some investors, causing a price to include some irrational factor even after this has been recognized by the market as a whole. One of these psychological variables has been reported by Barber and Odean (2001). They argue that while confident investors trade excessively, this is proven to be supported by trading data split by gender. They find men trade 45% more than women, and this costs them annually 2.65% while women lose 1.72% from excessive trading. Lessons learned: men have more confidence in trading stock, and lose some return as a result. Women lose less as they are less confident therefore trade less. Rational investors only trade and purchase information when they expect this to improve their utility (Grossman and Stiglitz). Another psychological variable is optimism. Optimistic employees believe that future share prices are higher than they really are and as a result are more inclined to participate in ESO plans. A third psychological factor is mental accounting. People assign different values to shares they have received in a different manner; they use associations to form

(10)

choose the TFS plan tax advantage as they see this as a source of ‘cheap’ income, sort of a bonus rather than a source of future income.

I also have made some predictions on variables having to do with the employees participating in the specific ESO plan and TFS plan as offered by the company. As employees could always choose to use the remaining tax-free savings room after they had participated in the ESO plan, TFS plan participation is always expected from a rational point of view for this specific situation.

An employees’ age influences participation behaviour; older employees looking for retirement income will not have the time to wait until their options become valuable and might therefore choose to participate in the TFS plan. Younger employees however, will have more time and be more eager to participate in the ESO plan. I expect employees who participate in ESO plans to perform relatively well as they will be rewarded more in case they are able to drive the share price up. I further expect that employees working in HQ will participate more than Branch network employees as HQ employees feel they are more informed. Another reason why I expect HQ employees to participate more as they might be more loyal to the company as they are closer to management and feel more responsible, just as shown by Cohen (2006). I expect males to participate more in the ESO plan than in the TFS as they are usually overconfident and trade more as shown by Barber and Odean (1999) as well as by Agnew (2005). From a performance point of view I expect salary scale to have a positive effect on ESO participation. Higher salaried employees are wealthier and wealth is a positive influence as shown by Agnew (2005). She looked at 401(k) contributions on an individual level. She found that lower paid workers are less diversified as a result of concentration of fund, and found that high income employees were significantly more likely to invest in company stock than lower salaried employees. Mostly because they made better decisions on investment diversification, but also due to the fact that lower salaried workers had a harder time diversifying away from company related stock. Sautner (2005) looked at ESO plans from the firm perspective as well as the relationship between the design of stock option programs and corporate governance structures. He discusses why companies use stock option compensation as a component of employee compensation, how broad based stock option programs affect firm performance and how much stock option programs cost to the issuing firm. He states that ESO seem to be a reasonable mechanism to align the interests of top executives with shareholder, but do not have the same incentive effects for lower level employees as positive incentive effects are likely to be reduced by free-rider considerations and the cost resulting from the compensation risk imposed on lower level employees. Lower level employees would rather have cash than an uncertain future payout as they require cash flow now rather than later. This forces lower level employees to choose steady jobs over jobs with uncertain income. Sautner has shown that ESO participation is expected to increase as salary increases.

Finally, I expect employees who have been employed for a longer time (yearswork) to be more loyal and therefore invest more in company stock as shown by Cohen (2006). He discusses the effect of loyalty on individual portfolio choices by using a retirement contribution dataset. He specifically

(11)

looked at the contributions of stand-alone firm employees and conglomerate employees and found stand-alone employees are more likely to invest in their firm’s shares than conglomerate employees. Cohen finds that the cost of loyalty to portfolio return is larger than the cost of home bias and can result in a 20% decrease in retirement income. Lessons learned: The loyalty explanation seems to affect retirement contributions and cannot be taken away by diversification from labour income, information based explanations or employee extrapolation of past results.

(12)

Table 1a: Predicted Relationships between Participation Behaviour and Rational and Psychological Factors

This table reports predicted relationships between various rational and psychological variables and ESO participation / TFS participation behavior. \+" means that a model or theory predicts that an increase in the respective variable will result in an increase in participation. Correspondingly, \-" means that a model or theory predicts that an increase in the variable will result in a decrease in participation, and hence in an earlier exercise decision. \?" means that

no prediction is possible.

Variable ESO Participation Behavior TFS Participation Behavior (Predicted Sign) (Predicted Sign)

Rational Variables

Risk Aversion - - +

Stockholdings - - +

Wealth + +

-Firm-specific Human Capital - +

Psychological Variables

Overconfidence +

-Optimism +

-Mental Accounting - +

Table 1b: Predicted Relationships between Participation Behaviour and Subject Variables

This table reports predicted relationships between subject variables and ESO participation / TFS participation behavior. These are predictions related to behavior of employees who work for the company. \+" means that a model or theory predicts that an increase in the respective variable will result in an increase in participation.

Correspondingly, \-" means that a model or theory predicts that an increase in the variable will result in a decrease in participation, and hence in an earlier exercise decision. \?" means that no prediction is possible.

Variable ESO Participation Behavior TFS Participation Behavior (Predicted Sign) (Predicted Sign)

Subject Variables age ? ? assessment + + hq + + male + + scale + + yearswork + +

(13)

3 Research Problem, Goal, Hypotheses and Methodology

3.1 Research Problem

ESO plans are expensive to setup and require resources to maintain the plan for the duration of the program. They are set up for the incentive effects associated with them, which exist in theory but are hard to prove empirically. Several studies have investigated the behaviour of employees investing in company stock through ESO plans exercise data or 401(K) contributions. In studies using ESO plan exercise data, usually a limited set of employees is compensated with stock options, which cannot be diversified away and which are designed to retain personnel. Not all employees can participate. For those who can, participation in the ESO is part of the compensation package for the employee and is therefore not free. A participation bias hampers insight into how ESOs can be used as an incentive to attract and retain employees, an incentive to compensate employees and an incentive to motivate employees. Additional empirical research on data without the participation bias would be useful.

3.2 Research Goal

Design of an ESO plan should be in line with its purpose. Information on ESO incentives to employees is of interest to firms issuing ESO plans. As we have been able to obtain an unbiased dataset, we can use it to see if incentives influence employee participation. As the data consists of individuals with very different characteristics, I will try to see if incentives are different as well among sub-sets of employees.

It would be relevant to see why employees participate in ESO plans when the choice to participate is free, and a risk-free alternative is available. It enables us to better understand the incentive effects associated with the ESO plans and TFS plans and testing for them in absence of the participation bias would yield useful results. I hope to answer the following question:

‘What motivates employees to participate in an Employee Stock Option program?’

Knowing which subsets of employees are interested in participation is relevant, but it is of much more interest to determine why employees are participating. I therefore look for evidence whether employee participation can be linked to specific incentive effects. Linking incentive effects to employee participation could help design future ESO plans.

(14)

3.3 Hypotheses

The research problem showed us that it would be useful to investigate the participants in an ESO plan. The research goal is to find out why employees participate. As several invective effects have been described in literature, we will attempt to find proof of existence of such incentive effects. We have therefore prepared four partial research questions, one partial research question for each aforementioned incentive effect. The partial research questions are:

I. ‘Are employees attracted by the ESO plan?’ II. ‘Are employees retained by the ESO plan?’

III. ‘Do employees look for compensation by participating in the ESO plan?’ IV. ‘Does employee performance improve by participating in the ESO plan?’

For as far as possible, hypotheses are formulated and tested for theories related to partial research questions.

Hypotheses for partial research question I:

H0: participation in the ESO plan is not different for changes in years with the company

H0: participation in the TFS plan is different for changes in years with the company

and

H0: participation in the ESO plan is not different for new employees

H0: participation in the TFS plan is different for new employees

We will test using the data we have on the amount of years employees are working for the company, in the second part we have created a new variable that enables us to look specifically at the subset of employees who started in 2003 and are therefore new to the company.

Hypotheses for partial research question II:

H0: retention is not different for participants of the ESO

H0: retention is different for participants of the TFS

We will test to see whether ESO participants have been employed by the company for a longer period than non-participants, and later do the same for TFS participants.

Hypotheses for partial research question III:

H0: participation in the ESO plan is not different for changes in salary scale

H0: participation in the TFS plan is different for changes in salary scale

We will test for ESO and TFS participation using the data we have on the salary scale of the employees.

Hypotheses for partial research question IV:

(15)

H0: performance is different for participants of the TFS plan

We will test to see whether ESO participants perform better than non-participants using the end of year score employees received from their managers. We later perform the same test for TFS participants. This is an important proxy of motivation inflicted by ESO and or TFS participation.

I will test these hypotheses in chapter 5: Empirical Analysis and Results.

3.4 Methodology of Research

To test whether proof can be found that a hypothesis is supported by facts, we use an econometrical test based on binary response models. We use binary response models when the range of values of the dependent variable is substantially restricted (Woolridge 2006, ch17). Tests using conventional linear probability tests can lead to fitted probabilities which are negative or in excess of one; in addition to this the partial effect of any explanatory variable (appearing in level form) is constant. The interest lies primary in the response probability. The test I applied is the PROBIT tests. It can be used to make the model linear. For the PROBIT, the test statistic used is the z-score. In case a coefficient is positive then the parameter has a positive relation to the dependent variable. When a coefficient is negative, the parameter has a negative relation to the dependent variable. In my case, I checked for robustness by using the specific PROBIT robust analysis. When checking for robustness very strong outliers are dropped, after which cases with large absolute residuals are down-weighted so these do not influence the coefficient too much. Including the check for robustness is a compromise between dropping moderate outliers and violating the assumption of the Ordinary Least Squares regression. The performed PROBIT analyses do exclude records with values missing for the included fields; this ensures solid results. Additional econometrical tests include the pairwise correlation (table 4) and a two-sample Wilcoxon rank-sum test (Mann-Whitney test) comparing the median values of a certain variable for the two groups (tables 5 and 6).

(16)

4 Dataset and Descriptive Statistics

4.1 Company Dataset

Since 1994, many employees in the Netherlands have been eligible to participate in TFS plans. These are essentially risk-free and have proved to be quite popular among Dutch employees. In recent years, employees at some companies have had the choice to forego the risk-free tax-advantage and use the quotas of the TFS plan to buy into an ESO program. The ESO plans are risky as their performance is dependent on the company’s future share price, which is unknown.

Through the human resources department of large Dutch financial institution we obtained employee specific data. All employees eligible for participation in the TFS plan, the ESO plan or both have been included. In addition to employee specifics such as gender and age are the individual participation choices made regarding participation in the TFS plan, participation in the ESO plan, participation in both the TFS plan as well as the ESO program or no participation in either plans. For a list of included fields please see table 2: Definition of Variables.

4.2 Description of the Data

The dataset holds data for 2004. We have used this year as it was the only available year that participation in the ESO plan was possible. The data consists of 23,327 employees eligible for participation in the TFS plan and / or ESO plan. The data was obtained from one database which holds data on several years; therefore there were no mismatches for fields which are unlikely to change such as gender, year of birth, nationality and starting date. Even though the dataset was no always complete, there was no evidence for issues associated with the data and therefore no data cleaning took place.

There were 11,676 males included in the data and 11,651 females, making the data even balanced with respect to gender. Regarding age, 25% was born in 1956 or before, while 25% was born in or after 1971. The mean year of birth was 1964 with a standard deviation of 9.3 years. Regarding years at the firm, 25% was employed for 25 or more years, while 25% was employed for 7 years or shorter. The mean year of employment was 1988, which means the average years with the firm was 16. The standard deviation for the years employed was 11.0 years. For a more in-depth description of the data please see table 3: Descriptive Statistics.

4.3

Description of the TFS- and the ESO plans.

In 2004 there were four available choices open to all employees employed on May 24, 2004 of the company’s Dutch business. The four options were participation in the TFS plan, participation in the ESO plan, participation in both plans or no participation. The vesting period for participation in both the salary-savings scheme and the ESO plan are identical, which encourages testing for variances in

(17)

employee behaviour between the two plans. I will continue to discuss the available options to the employees. For a full text provided by the company please refer to Appendix A.

Available choices for the employees:

I Participate in the TFS.

The salary savings scheme is a tax-efficient way of saving, because one does not have to pay tax or social security contributions on the gross earnings. The plan runs for 4 years, from May 2004 until December 31, 2008. A maximum of EUR 613 can be deducted from gross salary, invested against 3% in a savings account. Interest is paid yearly to the participants salary account. After the contract period, the EUR 613 in the savings account can be withdrawn without having to pay taxes. Money is vested until December 31, 2008. People who would like to withdraw early are liable to pay either 42% of 52% dependent of the tax bracket at the time of the start of the project, resulting in lower pay-out of EUR 355.54 or EUR 294.24 respectively. Full withdrawal without tax-penalty is possible under the following conditions:

• Purchase of a home for own use

• Payment of premiums for a life insurance or annuity policy • Voluntary pension contributions

• Payment of study expenses • Contribution to unpaid leave

• Contribution to start up of own business

II Participate in the ESO plan.

Participation in the Equity Option Scheme gives one the opportunity to acquire options to buy ordinary shares in The Company. Upon exercise there is a chance of achieving tax-free capital gains. The ESO plan runs for 7 years, from May 2004 until May 2011. Options are vested until December 31, 2008.

The maximum tax value of the options was 2 * EUR 613 = EUR 1,226 for 2004. As the price of the underlying share (ordinary Company share) was EUR 17.12 and the tenor was 7 years (the plans runs from 2004 - 2011, the tax value of one option was set by the government to be 26% of the exercise price; the value of one option therefore was 0.26% * EUR 17.12 = EUR 4.4512. For each employee,

(18)

down to EUR 613 – EUR 235.91/2 = EUR 495.04. Employees under the 42% tax bracket therefore incurred costs of 42% * EUR 235.91/2 = EUR 49.54. Employees under the 52% tax bracket therefore incurred costs of 52% * EUR 235.91/2 = EUR 61.34. Early withdrawal is not possible.

III Participate in the ESO plan and the TFS.

In this case, the tax room for the participation in the Salary Savings Scheme is lowered by the participation of the Equity Option Scheme. As 53 options were made available with a taxable value of 53 * 26% * EUR 17.12 (strike price) the tax value of the options was EUR 235.91. Participation in the Salary Savings Scheme was therefore lowered to EUR 613 – EUR 235.91 / 2 = EUR 495.04 for those employees who opted to participate in the TFS plan in case the ESO plan would not reach the tax-free roof.

(19)

Table 2: Definition of Variables

This table summarizes and defines variables used in the empirical analysis. All variables were taken either directly from or derived from the company data sample.

Variable Description

age Employee age in 2004.

aor 1 for participation in ESO plan, 0 for not participating.

assessment Assessment score of the employee, ranging 0-5. Higher is better.

bonus 1 for received bonus, 0 for not received bonus.

br 1 for employed at Branch Network, 0 for not employed at branch network.

highscale 1 if employee was employed in scale 15 or 16.

hq 1 for employed ate HQ, 0 for not employed at HQ.

id Unique employee identification number.

job_code Job description code.

job_description Employee Job description.

left_vol 1 in case employee left voluntary, 0 in case employee did not leave.

male 1 in case the employee is a male, 0 for female.

manager_id Unique identification number of the manager of the employee.

nationality Employee nationality.

new 1 if employee was first employed in 2003.

reason_for_leave In case the employee left the company: the reason why the employee left the company.

scale Salary scale raging from 1 to 16. Higher is better.

spaar 1 for participation in Tax-Free Savings plan, 0 for not participating.

sub 1 for employed at subsidiary, 0 for not employed at subsidiary.

(20)

Table 3: Descriptive Statistics

This table provides basic descriptive statistics on the variables in the company data sample. All data is concerning 2004.

assessment Assessment score of the employee, ranging 0-5. Higher is better. br 1 for employed at Branch Network, 0 for not employed at branch network. bonus 1 for received bonus, 0 for not received bonus.

hq 1 for employed ate HQ, 0 for not employed at HQ. male 1 in case the employee is a male, 0 for female. scale Salary scale raging from 1 to 16. Higher is better.

spaar 1 for participation in Tax-Free Savings plan, 0 for not participating. sub 1 for employed at subsidiary, 0 for not employed at subsidiary. yearswork Number of years the employee has been employed by the company.

Variable Mean Medean 25% perc. 75% perc. Std. Dev Male Female Male+Female Obs.

age 40.4 40 48 33 9.34 23,327 aor 4,708 3,292 8,000 23,327 assessment 3.1 3 3 4 0.95 21,352 bonus 4,873 5,541 10,414 23,327 br 7,822 8,987 16,809 hq 3,367 2,283 5,650 male 11,676 11,651 23,327 23,327 scale 7.0 7 4 8 2.98 21,213 spaar 6,241 5,959 12,200 23,327 started 1988.3 1990 1979 1998 10.96 23,327 sub 486 382 868 year_of_birth 1963.6 1964 1956 1971 9.34 23,327 yearswork 15.7 14 25 6 10.96 23,327

(21)

5 Empirical Analysis and Results

The purpose of testing a hypothesis is to determine the probability that it is supported by facts. The hypothesis formed is an informed /educated guess. For each partial research question, several theories are available. I will formulate hypotheses and test these theories.

The first partial question was:

I. ‘Are employees attracted by the ESO plan?’

We can see whether new employees are just as attracted to participate in ESO and TFS plans as employees who have been with the company for a longer time.

Years at the Company and Participation in 2004.

We would expect employees who have been working at the company for a longer time to show more loyalty towards the company and hence participate more in the ESO plan. We can see if our sample proves that participation does go up as the amount of years with the company increases. I formulated the following hypotheses:

H0: participation in the ESO plan is not different for changes in years with the company

H1: participation in the ESO plan is different for changes in years with the company

In the correlation matrix (table 4), it is shown that there is a positive correlation between the amount of years worked in the company and the participation in ESO plans. In table 9 it is shown that yearswork has a positive influence on participation, even when including additional parameters such as age, assessment score, whether the employee worked at Head-Office or not, gender and salary scale. H0 is

rejected. Employee wealth might influence participation as well, but as I have controlled for salary scale this might serve as a proxy for this and the result remains robust.

We can also have a look specifically look at whether yearswork has the same effect on participation in the TFS plan as previously shown on participation in the ESO plan. TFS participation is expected to be higher for new employees than ESO participation. I formulated the following hypotheses:

(22)

parameters such as age, assessment score, whether the employee worked at Head-Office or not, gender and salary scale. It seems that TFS participation is more attractive for employees with more years at the company. H0 is rejected. One could conclude that years at the company has a more pronounced

influence on TFS participation than on ESO plan participation. This is additional evidence that long term employees participate more in ESO plans and TFS plans, but more often choose the safe option of the TFS plan.

We would expect new employees to show less loyalty towards the company and hence participate less in the ESO plan. We can see if our sample proves that participation does go down if we look specifically at employees started in 2003. I formulated the following hypotheses:

H0: participation in the ESO plan is not different for new employees

H1: participation in the ESO plan is different for new employees

In table 9 it is shown that new employees have a negative influence on participation, even when including additional parameters such as age, assessment score, whether the employee worked at Head-Office or not, gender and salary scale. H0 is rejected. Employee wealth might influence participation

for the ESO plan, but as I have controlled for salary scale this might serve as a proxy for this and the result remains robust.

We can also have a look specifically look at whether the variable new has the same effect on participation in the TFS plan as previously shown on participation in the ESO plan. TFS participation is expected to be higher for new employees than ESO participation. I formulated the following hypotheses:

H0: participation in the TFS plan is not different for new employees

H1: participation in the TFS plan is different for new employees

When looking at table 9, one can see that TFS are not very attractive for new employees, but participation is higher than for ESO plans, even when including additional variables. It seems that TFS participation is more attractive for employees with more years at the company. H0 is rejected. One

could conclude that new employees are very not attracted to the company by ESO plans but have more interest in TFS plans.

II. ‘Are employees retained by the ESO plan?’

From a retention incentive perspective, one would think that ESO participation has positive retention effects. Retention is expected to be higher for employees who participate than employees who do not participate, as one would think that employees who participate are more committed to their company. I formulated the following hypothesis:

(23)

H0: retention is not different for participants of the ESO

H1: retention is different for participants of the ESO

I tested for the influence of ESO participation on retention including additional parameters such as age, assessment grade, gender, years at the company, salary scale and whether the employee worked at Head-Office or not. As shown in Table 7: Retention and ESO Plan and/or TFS Plan Participation, there is a strong negative correlation between participation in either ESO plans or TFS plans and whether employees left voluntary or did not leave. H0 is rejected.

From a retention incentive perspective, it is also interesting whether participation in the TFS plan has the same effect as previously shown by participation in the ESO plan. Retention is expected to be higher for employees who participate than employees who do not participate, as one would think that employees who participate are more committed to their company. I formulated the following hypotheses:

H0: retention is not different for participants of the TFS plan

H1: retention is different for participants of the TFS plan

I tested for the TFS plan participation influence on retention including additional parameters such as assessment grade, gender, years at the company, salary scale and whether the employee worked at Head-Office or not. Also shown in Table 7: Retention and ESO Plan and/or TFS Plan Participation; there is a strong negative relation between participation in the TFS plans and whether employees left voluntary. This strong negative correlation exist with additional variables are added to the analysis. H0

is rejected. The influence on participating in the TFS plan seems to be slightly higher than previously shown by the participation in the ESO plan.

III. ‘Do employees look for compensation by participating in the ESO plan?’

The perhaps most important argument in favour of ESO plans is that they give senior management a greater incentive to act in the interest of shareholders by providing a direct link between compensation and company stock-price performance. Agency problems between shareholders and managers are mitigated by linking their compensation directly to stock-price performance. Hall and

(24)

From a compensation incentive perspective, one would think that ESO participation has positive compensation effects. We test to see whether high salaried employees are participating more than lower salaried workers, since the high salaried workers may feel they have a better. Salary scale is expected to be higher for employees who participate than employees who do not participate, as one would think that employees with high salary scales might feel have influence on their compensation through the ESO plan and therefore participate more. I formulated the following hypothesis:

H0: ESO participation is not different for changes in salary scale

H1: ESO participation is different for changes in salary scale

I tested for the influence of salary scale on ESO participation including additional parameters such as age, assessment grade, whether the employee worked at Head-Office or not, gender and years at the company. As shown in Table 9: ESO Plan and/or TFS Plan Participation, there is a strong positive relation between salary scale and ESO participation, with both a strong coefficient as well as a strong t-statistic. H0 is rejected.

It is also interesting whether salary scale has the same impact on TFS plan participation as it has on ESO participation. We expect TFS participation to be positively related to salary scale. I formulated the following hypotheses:

H0: TFS participation is not different for changes in salary scale

H1: TFS participation is different for changes in salary scale

I tested for the influence of salary scale on TFS participation including additional parameters such as age, assessment grade, whether the employee worked at Head-Office or not, gender and years at the company. As shown in Table 9: ESO Plan and/or TFS Plan Participation, there is a strong positive relation between salary scale and ESO participation, but not as strong as shown previously when we looked at the influence of salary scale on ESO participation. This might be a result of higher salaried workers being able to better diversify their holdings and making better investment decisions as shown by Agnew (2005). H0 is rejected.

IV. ‘Does employee performance improve by participating in the ESO plan?’

Assessment Scores and Participation in 2004.

Performance is expected to be higher for employees who participate than employees who do not participate, as one would think that employees who participate are more committed to their company. I formulated the following hypothesis:

H0: performance is not different for participants of the ESO plan

(25)

I tested for the influence of ESO participation on assessment score including additional parameters such as age, gender, whether employees worked at HQ or not, years at the company, salary scale whether employees were new or not. As shown in Table 8: Performance and ESO Plan and/or TFS Plan Participation, there is a strong positive correlation between participation in either ESO plans or TFS plans and employees Assessment Scores. H0 is rejected.

From a performance incentive perspective, it is also interesting whether participation in the TFS plan has the same effect as previously shown by participation in the ESO plan. Performance is expected to be higher for employees who participate than employees who do not participate, as one would think that employees who participate are more committed to their company. I formulated the following hypotheses:

H0: performance is not different for participants of the TFS plan

H1: performance is different for participants of the TFS plan

I tested for influence of TFS plan participation on performance including additional parameters such as age, gender, whether employees worked at HQ or not, years at the company, salary scale whether employees were new or not. Also shown in Table 8, there is a strong positive correlation between participation in the TFS plans and employees’ assessment scores in 2004. H0 is rejected. The positive

influence on performance of participation in the TFS plan is very much equal to the influence on performance of the ESO plan participation.

(26)

Table 4: Correlation

This table reports the correlation coefficients between the individual-level variables, as well as the significance level of the respective correlation coefficients. It further includes the number of observations of the respective variables (Obs.). * indicates significance at 10%, ** indicates significance at 5%, *** indicates significance at 1%.

age Employee age.

assessment Assessment score of the employee, ranging 0-5. Higher is better. hq 1 for employed ate HQ, 0 for not employed at HQ

male 1 in case the employee is a male, 0 for female scale Salary scale raging from 1 to 16. Higher is better.

spaar 1 for participation in Tax-Free Savings plan, 0 for not participating yearswork Number of years the employee has been employed by the company.

(1) (2) (3) (4) (5) (6) (7) (8) (9) aor (1) 1.000 spaar (2) 0.203 *** 1.000 age (3) 0.025 *** 0.156 *** 1.000 assessment (4) 0.100 *** 0.072 *** -0.037 *** 1.000 male (5) 0.127 *** 0.023 *** 0.248 *** 0.045 *** 1.000 yearswork (6) 0.053 *** 0.203 *** 0.796 *** -0.017 ** 0.185 *** 1.000 scale (7) 0.211 *** 0.092 *** 0.144 *** 0.222 *** 0.431 *** 0.021 *** 1.000 hq (8) 0.010 -0.009 -0.029 *** 0.103 *** 0.108 *** -0.100 *** 0.329 *** 1.000 br (9) -0.005 0.009 0.017 *** -0.107 *** -0.113 *** 0.096 *** -0.312 *** -0.908 *** 1.000 Obs. 23327 23327 23327 21352 23327 23327 23213 23327 23327

(27)

Table 5: Differences between Groups - ESO Participation Behavior of Employees

This table compares descriptive statistics for the group of employees that participated ijn the ESO plan with the group of employees which did not participate. The table contains means and medians of a large set of variables for the two groups. It further includes the number of observations of the respective variables (Obs.). The last column contains p-values of a two-sample Wilcoxon rank-sum test (Mann-Whitney test) comparing the median values of a certain variable for the two groups. The null hypothesis is that the two groups are from populations with the same medians. * indicates significance at 10%; ** indicates significance at 5%; *** indicates significance at 1%.

age Employee age.

aor 1 for participation in ESO plan, 0 for not participating

assessment Assessment score of the employee in 2004, ranging 0-5. Higher is better. hq 1 for employed ate HQ, 0 for not employed at HQ

yearswork Number of years the employee has been employed by the company.

ESO plan participating employees

Non ESO plan participating employees p-value age Mean 40.77 40.27 0.0001 *** Median 40 40 Std. Dev. 8.55 9.72 Obs. 8000 15327 assessment Mean 3.23 3.03 0.0000 *** Median 3 3 Std. Dev. 0.87 0.99 Obs. 7843 13509 male Mean 0.59 0.45 0.0000 *** Median 1 0 Std. Dev. 0.49 0.50 Obs. 8000 15327 yearswork Mean 16.47 15.23 0.0000 *** Median 15 14 Std. Dev. 10.19 11.32 Obs. 8000 15327 scale Mean 7.90 6.58 0.0000 *** Median 8 6 Std. Dev. 2.87 2.93 Obs. 8000 15213 hq Mean 0.25 0.24 0.1276 Median 1 0 Std. Dev. 0.43 0.43 Obs. 8000 15327

(28)

Table 6: Differences between Groups - TFS Participation Behavior of Employees

This table compares descriptive statistics for the group of employees that participated ijn the TFS plan with the group of employees which did not participate. The table contains means and medians of a large set of variables for the two groups. It further includes the number of observations of the respective variables (Obs.). The last column contains p-values of a two-sample Wilcoxon rank-sum test (Mann-Whitney test) comparing the median values of a certain variable for the two groups. The null hypothesis is that the two groups are from populations with the same medians. * indicates significance at 10%; ** indicates significance at 5%; *** indicates significance at 1%.

age Employee age

assessment Assessment score of the employee, ranging 0-5. Higher is better. hq 1 for employed ate HQ, 0 for not employed at HQ

spaar 1 for participation in Tax-Free Savings plan, 0 for not participating yearswork Number of years the employee has been employed by the company.

TFS plan participating employees

Non TFS plan participating employees p-value age Mean 41.32 38.90 0.0000 *** Median 42 38 Std. Dev. 9.40 9.68 Obs. 16244 11127 assessment Mean 3.17 3.03 0.0000 *** Median 3 3 Std. Dev. 0.89 1.02 Obs. 11939 9413 male Mean 0.51 0.49 0.0004 *** Median 1 0 Std. Dev. 0.50 0.50 Obs. 12200 11127 yearswork Mean 17.79 13.32 0.0000 *** Median 17 9 Std. Dev. 10.55 10.93 Obs. 12200 11127 scale Mean 7.30 6.75 0.0000 *** Median 7 6 Std. Dev. 2.87 3.07 Obs. 12200 11013 hq Mean 0.24 0.25 0.1509 Median 1 0 Std. Dev. 0.43 0.43 Obs. 12200 11127

(29)

Table 7: Retention and ESO Plan and/or Tax-Free Savings Plan Participation

This table shows the coefficients of 4 PROBIT analyses on the effect of several variables on the dependent left_vol. It includes the number of observations of the respective variables (Obs.). Also included are Prob > Chi2 and Pseudo R2 to validate the analyses. * indicates significance at 10%, ** indicates significance at 5%, *** indicates significance at 1%.

assessment Assessment score of the employee, ranging 0-5. Higher is better. left_vol 1 in case employee left voluntary, 0 in case employee did not leave. male 1 in case the employee is a male, 0 for female.

spaar 1 for participation in Tax-Free Savings plan, 0 for not participating. yearswork Number of years the employee has been employed by the company.

new 1 if employee was employed in 2003.

Dependent: left_vol Independent: aor -0.4929 *** spaar -0.5530 *** age -0.0247 *** 0.0232 *** assessment -0.1393 *** -0.1389 *** hq 0.0817 0.0909 * male 0.1072 ** 0.0421 scale 0.0513 *** 0.0484 *** yearswork -0.0436 *** -0.0415 *** new 0.0971 0.11441 Obs. 21351 21351 Prob > Chi2 0.0000 0.0000 Pseudo R2 0.1636 0.1743 PROBIT, robust PROBIT, robust

(30)

Table 8: Performance and ESO Plan and/or Tax-Free Savings Plan Participation

This table shows the coefficients of 2 PROBIT analyses on the effect of several variables on the dependent assessment. It includes the number of observations of the respective variables (Obs.). Also included are Prob > Chi2 and Pseudo R2 to validate the analyses. * indicates significance at 10%, ** indicates significance at 5%, *** indicates significance at 1%.

assessment Assessment score of the employee, ranging 0-5. Higher is better. left_vol 1 in case employee left voluntary, 0 in case employee did not leave. male 1 in case the employee is a male, 0 for female.

spaar 1 for participation in Tax-Free Savings plan, 0 for not participating. yearswork Number of years the employee has been employed by the company.

Dependent: assessment Independent: aor 0.2462 *** spaar 0.2749 *** age -0.0124 *** -0.0138 *** hq 0.1997 *** 0.1886 *** male 0.0975 *** 0.1222 *** new -0.9254 *** -0.9329 *** scale 0.0200 *** 0.0231 *** yearswork -0.0015 -0.0031 Obs. 21351 21351 Prob > Chi2 0.0000 0.0000 Pseudo R2 0.0349 0.0382

(31)

Table 9: ESO Plan and/or Tax-Free Savings Plan Participation

This table shows the coefficients of 2 PROBIT robust analyses on the effect of several variables on the dependents aor and spaar. It includes the number of observations of the respective variables (Obs.). Also included are Prob > Chi2 and Pseudo R2 to validate the analyses. * indicates significance at 10%, ** indicates significance at 5%, *** indicates significance at 1%.

aor 1 for participation in ESO plan, 0 for not participating. assessment Assessment score of the employee, ranging 0-5. Higher is better. left_vol 1 in case employee left voluntary, 0 in case employee did not leave. male 1 in case the employee is a male, 0 for female.

spaar 1 for participation in Tax-Free Savings plan, 0 for not participating. new 1 if employee was employed in 2003.

Dependent: ESO or TFS (spaar)

ESO plan t-statistic TFS plan t-statistic

Independent: age -0.0154 -8.77 *** 0.0027 1.64 ** assessment 0.0785 8.01 *** 0.0815 8.54 *** hq -0.1597 -7.15 *** -0.0497 -2.29 ** male 0.1445 7.10 *** -0.1506 -7.53 *** scale 0.0878 24.05 *** 0.0349 9.61 *** yearswork 0.0127 8.65 *** 0.0236 17.08 *** new -0.4649 -5.62 *** -0.2884 -3.84 *** Obs. 21351 21351 Prob > Chi2 0.0000 0.0000 Pseudo R2 0.0432 0.0394

(32)

6 Conclusion and Recommendations

6.1 Conclusion

There is an ongoing debate on the usability of ESO plans to attract, retain, compensate and motivate employees. Especially broad based ESO plans are scrutinized and thought to be inefficient. Notwithstanding available literature, I have found that participation in both ESO plans and TFS plans is coupled with incentives for the employees. Employees participating in either plan are better retained by the company. Employees in higher salary scales seem to be looking for compensation using ESO plans and TFS plans. And finally, performance of participating employees is significantly better than performance of non-participating colleagues. Inconsistent with traditional ESO theories, no link could be made between attraction and ESO or TFS participation. Taken together, my findings suggest that ESO and/or TFS participation may play a substantial role in motivating and retaining employees of all salary levels, while ESO participation motivates higher salaried workers to create value for the company.

6.2 Recommendations

It would be very interesting to see if employees in certain jobs within the company would be more wiling to participate in ESO or TFS plans. Data quality issues do not permit me to analyse using the obtained data. With additional data, it might be possible to assess whether employees who have more insight into option valuation are more inclined to participate.

A second welcome addition to the data would be the reason why employees choose not to participate in the TFS plan. As participating in the TFS plan results in an essentially risk-free income, employees who do not participate forego this particular income without us really understanding why.

(33)

Literature

• Agnew, Julie R. Do Behavioural Biases Vary Across Individuals? Evidence from Individual Level 401(K) Data. Forthcoming Journal of Financial and Quantitative Analysis.

• Barber, B. M., and Odean, T. Boys will be Boys: Gender, Overconfidence, and Common Stock Investment. Quarterly Journal of Economics, 116 (2001), 261- 292.

• Barberis, Nicolas and Thaler, Richard H. A Survey of Behavioural Finance. National Bureau of Economic Research, September 2002.

• Bettis, J. Carr and Bizjak, John M. and Lemmon, Michael J. Exercise behaviour, valuation, and the incentive effects of employee stock options. Journal of Financial Economics, February 2004.

• Botosan, C., and Plumlee, M. Stock option expense: The sword of Damocles Revealed. Accounting Horizons, 15, 311–327, 2001.

• Cohen, Lauren. Loyalty Based Portfolio Choice. Yale School of Management, September 2006.

• Core, John and Guay, Wayne. Stock Option Plans for Non-Executive Employees. Journal of Financial Economics. 61(2), 253-287, 2001.

• Engelhardt, Gary V. and Madrian, Brigitte C. Employee Stock Purchase Plans. NBER Working Paper No. 10421, April 2004.

• Gonsalez, Luis and Gurtoviy, Ruslan. How Much to Pay in Cash? Employee Retention via Stock Option. February 2006.

• Hall, Brian J. and Murphy, Kevin J. Optimal Exercise Prices for Executive Stock Options, 2000. The American Economic Review, May 2000.

• Keller, Gerald and Warrack, Brian. Statistics for Management and Economics. 6th

edition 2003.

• Lambert, Richard A. and Larcker, David F. Stock options, Restricted Stock and Incentives, April 2004.

• Meulbroek, Lisa, 2001. The Efficiency of Equity-Linked Compensation: Understanding the Full Cost of Awarding Executive Stock Options, Financial Management 30, 5-30.

(34)

• Sautner, Zacharias and Weber, Martin. Subjective Stock Option values and Exercise Decisions: Determinants and Consistency. Universität Mannheim, October 2005.

• Sautner, Zacharias and Weber, Martin. Corporate Governance and the Design of Stock Option Programs. Universität Mannheim, October 2005.

• Sautner, Zacharias and Weber, Martin. How Do Managers Behave in Stock Option Plans? Evidence from Exercise and Survey Data. Universität Mannheim, May 2006.

• Hallock, Kevin and Olson, Craig. The Value of Stock Options to Non-Executive Employees. National Bureau of Economic Research, October 2004.

(35)

Appendix A:

The Company Employee Savings Schemes 2004

1 Introduction

You can once again take part in The Company’s employee savings schemes this

year. This guide to Employee Savings Schemes explains why it may be

worthwhile for you to take part, how you can do this and how to make your

choice(s).

There are two schemes in which you can participate:



The Equity Option Scheme



The Salary Savings Scheme

Participation in the Equity Option Scheme gives you the opportunity to acquire

options to buy ordinary shares in The Company. If you exercise the options, you

have a chance of achieving tax-free price earnings.

The Salary Savings Scheme is a tax-efficient way of saving, because you do not

have to pay tax or social security contributions on part of your gross earnings.

The rules of the Equity Option Scheme and Salary Savings Scheme are set out

in Annexes 7.11.1 and 7.11.2 of the CAO booklet. You can also view the text of

the CAO online at the

HR-Web.

In order to take part in one or both of the schemes, you must hold a personal

salary account and a salary savings account.

In order to exercise stock options, you need a TIN Code for the Investment Line.

See Section 4 (Participation) for information on participation in the schemes.

3 Equity Option Scheme

3.1 Who can participate?

With the exception of holiday workers and interns, anyone who is employed in

the bank’s Dutch business on 24 May 2004 may participate in the Equity Option

Scheme. Participation is possible only in the month of May. An employee who

leaves the bank’s service before 24 May (salary payment date) is not eligible to

participate.

Furthermore, participation is not open to employees who have been off work for

more than two years due to a full occupational disability. (In this context, ‘full

disability’ is a disability of between 80 and 100%, within the meaning of the

Invalidity Insurance Act (WAO)).

3.2 What does the scheme involve?

The Equity Option Scheme offers employees the opportunity to obtain options on

ordinary stocks in The Company. An option is the right to buy a number of stocks

at a predetermined price (the exercise price or strike price) within a given period

(the option term). Buying the stocks that you are entitled to is called exercising

Formatted: Bullets and Numbering

(37)

your option. Once you have exercised your option, you can keep the stocks or

sell them.

3.3 How much can I invest?

The Managing Board has made 500,000 options available for the Equity Option

Scheme. The options will be allocated to participants in the Equity Option

Scheme. The number of options allocated to you will then depend on the number

of people who choose to take part in the scheme. Each option has a tax value.

Given a term of seven years, the tax value is 26% of the value of the underlying

stock.

Example:

Suppose 500,000 options are available for the Equity Option Scheme, 12,500

employees choose to participate and the exercise price is EUR 15. Under such

circumstances, you will be allocated 40 options (since 500,000 divided by 12,500

is 40).

The tax value of each option will be EUR 3.90 (26% of EUR 15). Thus, the

options allocated to you are worth EUR 156.00 (40 times EUR 3.90). The

maximum that the law allows you to invest on a tax-efficient basis through an

option scheme and/or a salary savings scheme is EUR 1,226 (twice EUR 613).

So, to calculate the amount you could still pay into the Salary Savings Scheme if

you take part in the Equity Option Scheme, you need to divide the value of your

options by two, then deduct the resulting figure from EUR 613. In our example,

therefore, your salary savings ceiling would be EUR 613 – (EUR 156/2) =

EUR 535.

You can use the remaining savings allowance for the Salary Savings Scheme. If

you wish to do so, you can give notice of this with your choice for the Equity

Option Scheme.

3.4 The costs

In principle, participation in the Equity Option Scheme is free – you do not have

to pay the bank anything for the options allocated to you. Nothing is deducted

from a participant’s salary or profit-sharing bonus. However, if you participate in

the scheme, you lose all or some of your entitlement to save money on a

tax-efficient basis through the Salary Savings Scheme.

(38)

Employee Stock Option Incentives

Employee Stock Option Incentives*

Employee Stock Option Incentives

Contents

List of Tables

1

Introduction

2

ESO Participation Literature

3 Research Problem, Goal, Hypotheses and Methodology

4

Dataset and Descriptive Statistics

Description of the TFS- and the ESO plans.

5

Empirical Analysis and Results

6

Conclusion and Recommendations

Literature

Appendix A:

The Company Employee Savings Schemes 2004

Contents

1

Introduction

You can once again take part in The Company’s employee savings schemes this

year. This guide to Employee Savings Schemes explains why it may be

worthwhile for you to take part, how you can do this and how to make your

choice(s).

There are two schemes in which you can participate:

The Equity Option Scheme

The Salary Savings Scheme

Participation in the Equity Option Scheme gives you the opportunity to acquire

options to buy ordinary shares in The Company. If you exercise the options, you

have a chance of achieving tax-free price earnings.

The Salary Savings Scheme is a tax-efficient way of saving, because you do not

have to pay tax or social security contributions on part of your gross earnings.

The rules of the Equity Option Scheme and Salary Savings Scheme are set out

in Annexes 7.11.1 and 7.11.2 of the CAO booklet. You can also view the text of

the CAO online at the

HR-Web.

In order to take part in one or both of the schemes, you must hold a personal

salary account and a salary savings account.

In order to exercise stock options, you need a TIN Code for the Investment Line.

See Section 4 (Participation) for information on participation in the schemes.

3

Equity Option Scheme

3.1

Who can participate?

With the exception of holiday workers and interns, anyone who is employed in

the bank’s Dutch business on 24 May 2004 may participate in the Equity Option

Scheme. Participation is possible only in the month of May. An employee who

leaves the bank’s service before 24 May (salary payment date) is not eligible to

participate.

Furthermore, participation is not open to employees who have been off work for

more than two years due to a full occupational disability. (In this context, ‘full

disability’ is a disability of between 80 and 100%, within the meaning of the

Invalidity Insurance Act (WAO)).

3.2

What does the scheme involve?

The Equity Option Scheme offers employees the opportunity to obtain options on

ordinary stocks in The Company. An option is the right to buy a number of stocks

at a predetermined price (the exercise price or strike price) within a given period

(the option term). Buying the stocks that you are entitled to is called exercising

your option. Once you have exercised your option, you can keep the stocks or

sell them.

3.3

How much can I invest?

The Managing Board has made 500,000 options available for the Equity Option

Scheme. The options will be allocated to participants in the Equity Option

Scheme. The number of options allocated to you will then depend on the number

of people who choose to take part in the scheme. Each option has a tax value.

Given a term of seven years, the tax value is 26% of the value of the underlying

stock.

Example:

Suppose 500,000 options are available for the Equity Option Scheme, 12,500

employees choose to participate and the exercise price is EUR 15. Under such

circumstances, you will be allocated 40 options (since 500,000 divided by 12,500

is 40).

The tax value of each option will be EUR 3.90 (26% of EUR 15). Thus, the

options allocated to you are worth EUR 156.00 (40 times EUR 3.90). The

maximum that the law allows you to invest on a tax-efficient basis through an