The degree in which non-financial performance measures
are used by powerful CEO to inflate their compensation.
Lorenzo Singodikromo 10247424
21 June 2015
Thesis: MSc Accountancy & Control, control track
Universiteit van Amsterdam, Faculty of Economics and Business Supervisor: dr. P. Kroos
Statement of Originality
This document is written by student Lorenzo Singodikromo who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Abstract
This thesis looks whether non-financial performance measures are used as a tool by powerful CEOs to diversify compensation risk in their bonus plans. Based on the findings there seems no indication that powerful CEO include (additional) non-financial performance measures in their bonus plan in order to diversify their risk. On the contrary, the results show that CEO tenure is negatively associated with the use of non-financial performance measures and the interaction with volatility in market performance. Based on this, CEOs who are in the same organization for longer periods tend to rely more on financial rather than non-financial performance measures.
Table of Content 1. Introduction ... 5 1.1 Background ... 5 1.2 Research question ... 6 1.3 Contribution ... 7 1.4 Structure ... 7 2. Literature review ... 8 2.1 Agency theory ... 8 2.2 Performance measures... 9
2.2.1 Characteristics of good performance measures ... 9
2.2.2 Financial performance measures ... 10
2.2.3 Non-financial performance measures ... 12
2.3 CEO power ... 13
2.4 CEO power and performance measures ... 14
3. Methodology ... 16 3.1 Sample ... 16 3.2 Empirical models... 17 3.3 Variable measurements ... 17 3.3.1 Dependent variable ... 17 3.3.2 Independent variables ... 18 3.3.3 Control variables... 19 4. Empirical results ... 20 4.1 Descriptive statistics ... 20 4.2 Main analyses ... 23 4.3 Additional analyses ... 27
4.3.1 Alternative specification of the dependent variable ... 27
4.3.2 Addressing the potential of outliers/influentials ... 30
5. Conclusion ... 30
References ... 32
1. Introduction
1.1 Background
Publicly held firms are characterized by separation of ownership and control. This means that the shareholders (owners) have little or no direct control over the decision making process of management (Marks, 1999). The main advantage of this separation is the possibility of risk diversification for shareholders. They should be able to diversify and freely change their investment strategy in response to the market (Marks, 1999).
Besides this advantage, separation of ownership and control also features some problems. Jensen and Meckling (1976) refer to these problems as agency problems. First, there is a misalignment of objectives between the agent and the principal. Second, there is information asymmetry in which the agent has more information than the shareholder. The above problems are agency costs for the owners. The agents may take actions which may only benefit themselves at the expense of the shareholders (Jensen & Meckling, 1976).
One way to address the agency problems is with the use of performance measures. Performance measures provide information about the agent his actions. Furthermore firms can attach incentives to performance measures outcomes so that the agent will act in the best interest of the firm.
The traditional method of measuring and rewarding the performance of executives is based on financial measures. These financial measures provide the different levels of
management accurate and objective information that will help them in their decision making (Hopwood, 1972). Financial measures such as earnings or return on investments are
frequently used because they are seen as sensitive indicators of executive performance (Indjejkian & Matejka, 2012).
Despite the importance of accounting-based numbers, there are some flaws when these measures are used as an evaluation tool for the manager. First, the information provided from the traditional measures is incomplete and is a biased indicator of executive
performance (Hopwood, 1972). The actions of the manager are not fully captured by the measures. The wrong conclusions might be taken when they are only evaluated on financial numbers. Furthermore the most important limitation of only using financial measures is the inability to connect the long term vision of the firm to the short term actions of managers (Kaplan & Norton, 1996). The measures itself only focuses on achieving short term targets
which does not benefit the long term objectives of the firm. Financial performance measures are thus imperfect.
Due to the above limitations of financial metrics, non-financial performance measures are frequently used in incentive plans. The use of non-financial performance measures will lead to a greater focus on the long term rather than the short term (Banker et al., 2000). The study of Ittner et al. (1997) affirms this finding. Firms with an innovation and differentiation strategy need to be long term orientated, thus more emphasis is placed on non-financial performance measures in their annual bonus contract. Sliwka (2002) concludes that the use of non-financial performance measures in manager’s compensation plan gives manager’s
incentive to exert higher long-term effort. According to his study financial performance measures often neglect the latter purpose and therefore additional non-financial performance measures become valuable.
Incentive plans have recently been criticized as they are perceived to be the subject of CEO power. It is argued that incentive plans are a means for powerful CEOs to inflate their compensation instead of serving its purpose of aligning the interests between CEOs and shareholders. Core et al. (1999) show how firms characterized by low corporate governance have higher CEO compensation. When corporate governance is weak, CEOs have more influence and can exert their power in this setting. Likewise, Morse et al. (2011) find how powerful CEO can influence the weighting of different performance measures towards the better performance measures. In this way powerful CEO can inflate their compensation.
Little research looked into the question whether powerful CEO may diversify their compensation risk by including multiple financial and nonfinancial performance measures. This will represent the focus of my study.
1.2 Research question
This study looks at the question whether CEO may diversify their compensation risk by being evaluated on multiple financial and nonfinancial performance measures. The above
discussion leads to the research question, which is formulated as follow:
Do powerful CEOs reduce their compensation risk through the incremental use of non-financial performance measures in their executive bonus plan?
1.3 Contribution
First, this study contributes to the literature on non-financial performance measures. Where prior studies predominantly looked into the contribution of non-financial measures relative to the financial measures, this thesis is focused on the question whether the nonfinancial
measures that are used in CEO compensation contract serve the interests of powerful CEOs. In this way, this study also contributes to the literature on CEO power. Both the studies of Core et al. (1999) and Morse et al. (2011) show that powerful CEO may influence their compensation scheme in a way that will benefit them more. This research will extend this literature by looking at the way whether CEO may diversify their compensation risk by being evaluated on multiple financial and nonfinancial performance measures.
This study will also contribute in a more practical way. Incentive plans composed of a combination of financial and nonfinancial measures, are typically used as a means to
overcome the conflict of interest that may exist between CEOs and shareholders. It is important for shareholders to become aware that incentive plans are not only purposefully designed to improve the operation of the firm. It is possible that the use of nonfinancial measures may also be motivated because it is in the best interest of the CEO, at the expense of the shareholders. Knowledge about the degree that the use of nonfinancial measures is explained by CEO power enables parties such as shareholders to anticipate this behaviour, and possibly correct for this.
1.4 Structure
The structure of the paper is as follow. The literature review and hypothesis development will be discussed in chapter two. The research methodology will be discussed in chapter three. An explanation of the empirical results will be given in chapter four. The conclusion will be given in chapter five.
2. Literature review
2.1 Agency theory
The owner-managed firm has two characteristics. First, the owner has a claim of the firm’s profit. Second, the owner can make management decisions (Marks, 1999). In publicly held firms the shareholders (owner) have a claim to the profits but do not have the right to actively participate in management decision making (Fama & Jensen, 1983). This is defined as
separation of ownership & control.
One reason why firms tend to choose for this separation is that (some) risk can be shared among shareholders. Shareholders that own shares are not required to have any other active role in the firm; they can freely sell or trade the shares to other people. This allows for unrestricted risk sharing among shareholders (Fama & Jensen, 1983). Moreover by making this risk sharing available, investors are able to diversify and change their investment strategy in response to the market (Marks, 1999).
Besides the above advantages, the separation of ownership and control also features some costs. Managers will most likely lack incentives to manage the firm the same way as an owner, leading to inefficient decision making. Jensen and Meckling (1976) characterize this as an agency problem. In the agency theory, the principal (shareholders) gives the agent (management) some authority to act on behalf of the principal (Jensen & Meckling, 1976). While the agent has to act in the best interest of the principal, it is often seen that this is not the case. People tend to think only of themselves. Thus people want to increase and maximise their own utility and wealth. Given this assumption, both players have different goals which will lead to a conflict of interest. In other words there is a misalignment of objectives between the agent and the principal (Jensen & Meckling, 1976). Furthermore another important problem is the inability to effectively control the actions of the agent (Fama & Jensen, 1983). There is information asymmetry in which the agent has more information about their actions than the principal. Monitoring agents with conflicting interest will therefore lead to (more) agency costs.
In order to address this problem it is possible to implement an optimal incentive contract (Holmstrom, 1979). In this contract a compensation plan is made of different performance measures in order to act as an incentive for the agent to do what is best for the firm (shareholder). With this optimal contract the shareholders do not have to actively
for monitoring actions is that not all the actions of the management can be observed by shareholders due to the information asymmetry. The outcomes of the actions which are observable and which the management can influence should be linked to the relevant
performance measure in the compensation scheme (Holmstrom, 1979). In the next paragraph the choice of performance measures in incentive contracts will be discussed in greater depth.
2.2 Performance measures
In this paragraph the use of performance measures will be explained. First, the characteristics of what defines the quality of a performance measure will be described. Second, advantages and disadvantages of financial performance measures will be given. At last, the use of non-financial performance measures will be explained.
2.2.1 Characteristics of good performance measures
A performance measure provides the relevant information about the actions of the agent. Furthermore firms can attach incentives to performance measures outcomes so that the interest of both the agent and the shareholders are aligned. Due to this, performance measures are used to influence the behaviour of the agent in a way that is beneficial for the
shareholders (Holmstrom, 1979). The ideal performance measure is an executive’s individual contribution to firm value. Given that this measure is not available, most incentive plans contract on available outcome measures. First, each performance measure that provides incremental information about the agent their actions should be included in an incentive contract (Holmstrom, 1979). Subsequently, the weight of these performance measures in an incentive plan depends on the characteristics of these measures.
First, the weight of a measure is increasing in the sensitivity of the measure to the manager’s actions. The sensitivity is determined to what extent an agent can affect the related measure. For example, divisional profit is a more sensitive measure for a divisional manager compared to firm-level profit. Second, the measure is not only influenced by the actions of the agent but also by external factors. This is called noise. For example, stock market performance is typically perceived to be noisier compared to accounting performance. A good performance measure is thus described as a measure with high sensitivity of the effort of the agent and with a relative low level of noise. This way compensation can be linked to
the performance measure and will only give incentive for the agent to act in accordance to the objectives of the firm (Holmstrom, 1979).
In the past the focus was more on ‘single-task agency relation’ where the goal of the principal was to motivate the agent to increase his effort along a single dimension (Kim & Suh, 1991). ‘Multi-action agency models’ allow consideration of the congruence of the performance measure (Banker & Thevaranjan, 2000). An agent is responsible for multiple tasks in which he has to exert certain effort. The agent will thus try to allocate his effort in an optimal way that will give him the most benefits with the least amount of effort, leading to an important incentive problem for shareholders (Banker & Thevaranjan, 2000). Performance measures should therefore not only be evaluated on the basis of the sensitivity and noise, but also on the congruence.
Congruence can be explained as to what extent the increase in measure leads to an increase in the firm value. Therefore, a good performance measure has a high congruence if an increase in the performance measure also implies an increase in firm value. For example, given that accounting earnings can be improved by making less investments in R&D (as most of it is immediately expensed), which goes at the expense of firm value, accounting earnings are typically perceived as a low congruence performance measure (Dechow & Sloan, 1991). However the study of Datar et al. (2001) concluded that in some cases where there is a lot of noise involved, the firm is willing to sacrifice part of the goal congruence in order to reduce the overall risk of the agent their compensation. This way the risk premium for risk-averse agents is also reduced. A performance measure is therefore valuable if it is precise in measuring effort (high sensitivity) and is congruent with the goals of the firm (Banker & Thevaranjan, 2000). Furthermore a performance measure should have little noise in order to reduce the risk of the agent (Datar et al., 2001).
2.2.2 Financial performance measures
The traditional method of measuring and rewarding the performance of executives is based on financial measures. These financial measures provide the different levels of management accurate and objective information that will help them in their decision making (Hopwood, 1972). Financial (accounting) measures, such as net income or return on assets represent the most complete performance measures because the results of the actions of the manager will eventually flow through the financial statements (Moers, 2006). All the actions of the manager will be summarized in a single comprehensive performance measure. This gives a
clear overview of the performance of the agent. Therefore when employees are evaluated and incentivised by a summary financial performance measure, this can increase delegation of decision rights. Managers can decide themself which actions to pursue to improve
performance on the financial measure (Moers, 2006).
Besides that, these financial measures (net income or return on assets) are frequently used because they are seen as sensitive indicators of executive performance (Indjejkian & Matejka, 2012). In general these measures have a high level of sensitivity for executives and can easily be improved by rewarding the right incentive to the relevant agent, such as
divisional profit to a divisional manager. Moreover accounting regulation and external auditors reduces the susceptibility to manipulation, meaning that the managers should not be able to manipulate the measures to their own benefit. Furthermore accounting measures are not so susceptible to noise. The conservative nature of the accounting system where
economic events have to be realized before they are reflected by the accounting system makes earnings less susceptible to noise, compared for example to stock market returns.
On the other hand, financial measures are often criticized for having weak congruence as these performance measures are backward looking measures. They look at the economic effect of the management decisions made in the past. Financial measures are unable to connect the long term vision of the firm to the short term actions of managers (Kaplan & Norton, 1996). Dechow and Sloan (1991) found in their research that CEOs in their final years will lower investment expenditures to improve short term earnings. Future earnings which are beneficial for the firm will be sacrificed in order to achieve short term profits. The reason this happens is because of accounting conservatism mentioned before. The effect of the investments leading to future benefits will only be reflected if it is impounded in sales. Given this conservative behaviour it is likely that financial performance measures will therefore give managers incentives to lower investments. Moreover financial measures are lagging indicators due to this backward oriented focus. They are claimed to provide little information about the managerial actions on those dimensions that are predictive of the future success of an organisation, such as efforts to increase customer satisfaction. The information provided from these measures is incomplete and is a biased indicator of executive
performance (Hopwood, 1972).
To sum up, financial performance measures are highly sensitive. Moreover the accounting regulations reduce the susceptibility to manipulation. Furthermore accounting measures are not so susceptible to noise. On the other hand, the measures are being criticized for having weak congruence with the objectives of the firm. Due to this limitation,
non-financial performance measures are often complemented with non-financial measures. These measures will be discussed in the next paragraph.
2.2.3 Non-financial performance measures
Kaplan and Norton introduced the balanced scorecard (BSC) in 1992. This lead to new forms of performance measures which were used together with traditional financial performance measures in order to improve the overall value of the firm. The balanced scorecard was a revolutionary invention within the accounting world. Non-financial measures such as
customer or employee satisfaction, quality of products and market share became popular and were commonly used to evaluate managers (Banker et al., 2000). Short-term financial
information was complemented with qualitative and quantitative non-financial information. This meant that the decision making of the managers had a long-term oriented focus rather than a short-term focus (Kaplan & Norton, 1996). Moreover non-financial performance measures can add value in compensation schemes because it focuses on the long-term effort of managers.
A second benefit of the balanced scorecard related to non-financial performance measures is the linkage with strategy. Firms took into account the strategy of the firm and tried to link it with relevant performance measures indicators. One example is the study of Perera et al. (1997). In their study organisations with a customer-focused manufacturing strategy have more emphasis on non-financial performance measures. The performance of executives was more based on customer satisfaction rather than financial metrics such as profit. This was more in line with their main strategy, which was customer oriented, and leads to better executive performance.
Based on the above, it can be said that non-financial performance measures ensures that executives and employees focus on actions that are beneficial for future firm
performance. Non-financial measures provide information about actions which are leading indicators of future performance. In other words non-financial performance measures are perceived to have a high level of congruence with the firm’s objectives. Linking
non-financial performance measures to rewards will incentivise managers to allocate more effort and resources to those actions that will benefit the long term.
Furthermore the study of Campbell (2008) showed that non-financial performance measures are not only informative about the actions that managers take to, for example
increase quality but also on the acquired ability of managers to increase the drivers of future performance.
While some firms tend to rely solely on financials performance measures many firms complement the traditional measures with non-financial measures. Ittner et al. (1997) found that firms with a strategy based on innovation or quality tend to place a greater weight towards non-financial performance measures in the bonus contracts. The reason that these firms use more of these measures is that they need to be long-term oriented. Also firms will try to link the strategic objectives to a reward system in order to align the goals of the management and the organisation (Ittner et al., 1997). Combining the long term oriented objective to the relevant reward, will motivate and incentivise managers to take the desired actions.
Another reason to use non-financial performance measures is that they are better indicators of future financial performance than the traditional measures. The measures will motivate managers to take actions that will result in better outcomes in areas such as customer satisfaction, quality and innovations. These outcomes are the leading factors in future financial performance (Kaplan & Norton, 1992). Banker et al. (2000) researched this and came to the conclusion that customer satisfaction measures are indeed significantly associated with future financial performance measured in revenues and operating profit. They also found how adoption of an incentive plan based on non-financial performance measures is associated with improved non-financial and financial performance. This is consistent with the above literature stating that non-financial performance measures are able to motivate managers to act in the best interest of the firm. There is a high level of congruence with the firm’s objectives.
2.3 CEO power
In the literature there are multiple interpretations of CEO power. CEOs that can consistently make key decisions, regardless whether there are opposition or not from other executives, are deemed powerful (Adams et al., 2005). Bebchuck & Fried (2003) argue that the design of incentive plans is not only used as an instrument to align the interest of the managers and the shareholders, but may be part of the agency problem itself. They argue that incentive plans are not a solution to the agency problem but may actually encourage managers to misbehave and grant themselves more compensation. Powerful CEOs are able to inflate their own compensation and thus are able to extract rent in the form of compensation (Bebchuk &
Fried, 2003). Those CEOs have enough power and authority to gain influence over
shareholders and are thus able to extract more payments from the firm, which are often not related to their performance. Incentive plans are therefore susceptible to managerial power and as a result managers have significant influence over their own compensation schemes (Bebchuk & Fried, 2003).
The second point that they made is that the board of directors will also not always act in the best interest of the shareholder. The director’s behaviour is also part of the agency problem and their overall goal is to be re-appointed to the board (Bebchuk & Fried, 2003). In other words, the board of directors can also be affected by the power of the CEO. CEOs may have power that can influence the decision making of the board of directors (Grinstein & Hribar, 2003). CEOs who determine what is on the board agenda can control the amount of information that is given to the board. This is especially the case if the CEO is also the chairman of their own board (Jensen, 1993). Furthermore Jensen argues that the board of directors are often too large and the directors on the board own little amount of equity of the firm. This means that the directors have no or little incentive to monitor actions of the CEO and allows the CEO to influence the decision-making of the board (Grinstein & Hribar, 2003).
Another important relation is the relation between CEO power and the structure of corporate governance. The study of Core et al. (1999) looks at the compensation level of the CEO and the quality of the corporate governance. They show that firms characterized by low corporate governance have higher CEO compensation. When corporate governance is weak, CEOs have more influence and can exert their power in this setting. There are thus more agency problems in firms with a weak corporate governance structure (Core et al., 1999). Moreover Morse et al. (2011) find how powerful CEOs can influence the weighting of different performance measures towards the better performance measures. CEOs with more power were able to alter the focus of their incentive compensation to extract more payments from the firm. In this way powerful CEO can inflate their compensation. In addition the authors found this relation to be more pronounced in weak corporate governance settings (Morse et al., 2011).
2.4 CEO power and performance measures
The agency problems arising from separating ownership and control can lead to misalignment of goals between the CEO and the shareholders. Incentive contracts tied to performance
measures are intended to address these problems. However as was argued above it is possible that incentive contracts are not a solution but part of the problem itself (Bebchuk & Fried, 2003). They can be used as a tool for powerful CEO to grant themselves excessive
compensation. The CEO will likely have some power over the board because of his position (Adams et al., 2005). There are certain characteristics that make a CEO have more power and influence in the firm and the board of directors. These characteristics are often linked with weak corporate governance structure in organisations. Core et al. (1999) showed that CEO compensation is associated with weak corporate governance such as large board size, small fraction of outside directors and many busy directors.
Some empirical studies also showed how powerful CEOs can influence the design of the incentive contract. For example Morse et al. (2011) show how powerful CEO will shift the weight of different performance measures towards the better performing measures in order to inflate their compensation.
While financial performance measures are quite sensitive to the effort of the manager and not so susceptible to noise, they are criticized for backwards looking and having a weak congruence with the firm’s objectives. It is for this reason that some firms tend to
complement them with non-financial performance measures. These measures are linked with the firm’s strategy and are more forward looking (Kaplan & Norton, 1996). It can thus be said that these measures are better in evaluating the performance of the CEO.
Besides for optimal contract reasons, non-financial measures may also be included in executive compensations contracts due to the influence exerted by powerful CEOs. That is, CEOs are assumed to be risk-averse while a large part of their human and financial capital is tied to a specific firm. The addition of non-financial measures to their executive contract diversifies their compensation risk over a greater number of performance measures. Moreover, as non-financial performance measures are specific measures which focus on a subset of actions, they are typically only to a small amount correlated with the financial summary measures. Therefore it is expected that powerful CEOs may reduce their compensation risk by exerting pressure to include additional non-financial performance measures in their executive compensation contracts. The hypothesis is then formulated as follow:
Hypothesis 1: CEO power is positively associated with the use of non-financial performance measures in executive compensation contracts
3. Methodology
3.1 Sample
This archival research is based on an empirical analysis on the 500 largest firms in the United States. The initial sample covers all S&P 500 firms for the fiscal year 2013. Data regarding the CEO (and other executives) is collected from Execucomp and ISS (formerly
RiskMetrics). Financial data regarding the firm will also be collected from Compustat. The data of non-financial performance measures in executive compensation schemes is collected by hand from the proxy statements (DEF 14a) that can be found in the Edgar database
From the S&P 500 a random sample was taken. From this sample size some firms were excluded because of missing data on characteristics of the CEO (and other executives) such as tenure and compensation. This leads to the removal of 40 observations. Furthermore 26 firms dropped out due to missing financial data. Moreover utility firms with a SIC code between 4900 and 4999 and financial institutions with a SIC code between 6000 and 6999 were also removed because they are heavily regulated. This resulted in removing 118 observations. The final sample consisted of 200 firms. All the observations have been winsorized at 1% and 99% level. The table below summarizes the sample size. Appendix A gives additional information about the industry sector of the firms.
Table 1 Sample size
Initial sample from ISS 500
Removal due to random selection 116
Removal due to missing CEO data 40
Removal due to missing financial data 26
Removal of financial and utility firms 118
Final sample 200
Firms that solely use FPM 119
Firms that use NFPM 81
Firms that specify Numb_NF 56
3.2 Empirical models
To explore the research question and test the hypothesis, the following regression models will be used:
Model 1:
Nonfin= β0 + β1 CEO_Power + Controls +
ɛ
The first model is used to test whether the power of the CEO will affect the use of non-financial performance measures in their executive compensation. Based on the hypothesis, it is expected that powerful CEOs will use their influence to include non-financial performance measures in their executive compensation plan. β1 is therefore expected to be positive (β1 >
0).
Model 2:
Nonfin= β 0 + β1 CEO_Power + β2 Volat + β3 Volat*CEO_Power + Controls+ ɛ
The second model is used to further test the hypothesis whether the relation between volatility of firm performance and the inclusion of non-financial performance measures in CEO incentive plans is affected by the power of the CEO. Greater volatility in performance implies greater compensation risk and powerful CEOs could mitigate this by including nonfinancial measures in CEO bonus plans. So it is likely that CEOs with more power may diversify this risk by including (additional) non-financial performance measures in their compensation contract. β3 represents how CEO power affects the relationship between
volatility and the use of non-financial measures in CEO bonus plans. It is for this reason that β3 is expected to be positive (β3 > 0).
3.3 Variable measurements
3.3.1 Dependent variable
The dependent variable will be the non-financial performance measures (Nonfin). This variable is based on three different measures. First, a dummy variable will be made (NF) which is equal to 0 if non-financial measures are not used in the executive contracts and equal
to 1 if the compensation in the contract is (partly) based on non-financial measures. Second, if the above dummy variable is equal to 1, the total number of different non-financial performance measures (Numb_NF) will be counted if the information is provided in the proxy statement. At last, if the above dummy variable is equal to 1 the total weight (Weight_NF) of non-financial performance measures relative to the total performance
measures will be collected if the information is provided in the proxy statement. Examples of extracts from the proxy statement are provided in Appendix B
3.3.2 Independent variables
The main independent variable of interest consists of measures of CEO power. The first measure is tenure (CEO_Tenure). This is determined by how long the CEO has been in the position as CEO in the organisation. CEOs who serve for a long time in the same
organisation tend to be set in their ways and reflect more dominance to other executives (Hermalin & Weisbach, 1991). The length of tenure as CEO is also positively related to compensation (Morse et al., 2011), this indicates that CEO tenure is considered to be related to the amount of power the CEO can have.
The second measure of CEO power is the tenure of the CEO relative to the average tenure of all executives in the organisation (Rel_Tenure). One can assume that if the CEO stays longer in the same position compared to the other executives, they will have more influence in the organisation. Thus this measure gives a good indication of how powerful CEO can be.
The final measure of CEO power is the compensation of CEO relative to the average compensation of all executives in the organisation (Rel_comp). In the literature it has been argued that powerful CEOs are able to inflate their compensation (Core et al., 1999). Thus when the compensation of CEO is relative higher compared to the other executives in the organisation, it can indicate that those CEO have more influence and are able to earn greater compensation due to reasons other than performance. In other words, excess compensation is considered to be related to CEO power.
The second independent variable consists of measures of volatility in firm performance. The first measure of firm performance is the standard deviation of the stock returns
(STDev_stockret) of the firm over a five-year period (five years prior to fiscal year 2013). This market indicator is rather forward looking and the stock price should reflect future
benefits and risks related to performance. Therefore this measure gives a good indication of firm performance. Stock returns are calculated by taking the end of year stock price of year t minus the end of year stock price of t-1, divided by the end of year stock price of t-1. This method excludes dividends paid to shareholders.
However the problem with stock returns is that the measure is relative noisy and there are a lot of external factors that can influence this. This means that the relation between the CEO and stock returns might be low. It is for this reason that a second measure will be added. The second measure is the standard deviation of the return on assets (STDev_ROA) of the firm over a five-year period (five years prior to fiscal year 2013). This is a typical accounting measure used to estimate firm performance and firm risk. If the standard deviation is high, this means that there is a lot of volatility in the performance. Return on assets is calculated by dividing the net income by the total assets. Both measures are used in multiple studies such as Core et al. (1999), Morse et al. (2011) and Adams et al. (2005).
3.3.3 Control variables
The control variables are there to control for alternative explanations for the findings. The most important control variable in looking at the relation between CEO power and non-financial performance measures is the amount of long term equity of the CEO (LT_Equity) (Ittner et al., 1997). Based on the literature it is assumed that non-financial performance measures can be a great incentive for executives to focus on the long-term objectives of the firm. On the other hand, long term equity holdings of the CEO can also be a good incentive to focus on the long term. In order to control for the possibility that they will act as substitutes, this variable is included. Ittner et al. (1997) defines long-term equity as a ratio calculated by dividing the CEO’s equity holding by the CEO’s annual salary and bonus.
Two other variables are included to measure the use of non-financial performance measure in executive compensation contracts. The reason that firms are large with great investment opportunities may be because of their focus on the long-term perspective. Thus large firms with a lot of investment opportunities may have incentive contracts based on non-financial performance measures. Ittner et al. (1997) found that firms with a strategy based on innovation tend to place a greater weight towards non-financial performance measures in the bonus contracts. In order to take into account this relation firm size (Size) and investment opportunities (Growth) are included. Firm size is estimated by taking the natural logarithm of the total amount of assets. The investment opportunity is estimated with the use of the firm’s
average market to book ratio in fiscal year 2013. This variable is measured as the ratio of total market value of equity to total book value of equity (assets minus liabilities).
The summarized definitions of all the above variables can be found in Appendix C
4. Empirical results
4.1 Descriptive statistics
Table 2 reports some basic descriptive statistics. In fiscal year 2013 about 41% of the sample firms used at least one non-financial performance measure in their executive compensation contract. Furthermore, of all the firms that used these measures the average number of non-financial measures is three. In addition the average weight of non-non-financial performance measures relative to the total measures in the executive compensation contracts is around 25%.
For CEO tenure the average tenure is almost 10 years, with a median of 8 years. This is somewhat higher than the values reported in prior studies such as Morse et al. (2011) or Adams et al. (2005) where they both had an average value of 7 years. A possible explanation for this difference is that since they used data from late 1990 until early 2000, over time CEOs tend to stay longer in the organisation, which results in higher tenure. Moreover, the average CEO tenure relative to all executives is less than one (0.844) indicating that CEOs compared to other executives are in the organisation for a shorter duration. At last the average CEO total compensation compared to other executives is more than two (2.234). This means that the CEO earns on average more than twice as much as other executives.
For the volatility in firm performance the average standard deviation of the market performance (stock returns) is far higher than the average standard deviation of the
accounting measure (return on assets). The values are 0.409 and 0.044 respectively. This is in line with the intuition that a lot of external factors can affect the market performance, while accounting measures are less sensitive to outside factors.
The ratio of the CEO’s equity holding divided by the CEO’s annual salary and bonus (long term equity) is around eight. This indicates that the compensation in the form of equity incentives is eight times higher than cash compensations. At last, the market to book ratio (Growth) has an average value of 4.681. This means that on average the market value of
equity is much higher than the book value of equity, indicating that the average sample firm displays considerable growth options.
Table 2
Descriptive statistics
Non-financial performance measures (Nonfin)
Variable Obs Mean Std. Deviation Median P10 P25 P75 P90
NF 200 0.405 0.492 0 0 0 1 1
Numb_NF 56 3.230 1.897 3 1 2 4 6
Weight_NF 45 0.258 0.136 0.2 0.1 0.2 0.30 0.44
CEO power (CEO_Power)
Variable Obs Mean Std. Deviation Median P10 P25 P75 P90
CEO_Tenure 200 9.685 6.986 8 2 5 13 19
Rel_Tenure 200 0.844 0.609 0.698 0.174 0.436 1.133 1.657
Rel_Comp 200 2.234 1.639 1.943 0.879 1.393 2.598 3.768
Volatility in firm performance (Volat)
Variable Obs Mean Std. Deviation Median P10 P25 P75 P90
STDev_ROA 200 0.044 0.055 0.024 0.008 0.013 0.050 0.095
STDev_stockret 200 0.409 0.270 0.328 0.163 0.225 0.513 0.733
Control variables
Variable Obs Mean Std. Deviation Median P10 P25 P75 P90
LT_Equity 200 8.427 5.117 7.648 3.151 5.044 10.369 14.537
Size 200 9.657 1.092 9.541 8.320 8.905 10.456 11.149
Growth 200 4.160 4.423 3.059 1.216 2.009 4.990 9.484
Table 3 reports more descriptive statistics for the sample firms using financial (119 firms) and non-financial performance measures (81 firms) in the CEO compensation contracts, including t-tests for significant differences in means. The average tenure of the CEO is higher
in firms that did not use any non-financial performance measures. The same can be concluded for relative tenure. CEOs with longer (relative) tenure seem to have their incentives more strongly tied to financial measures. However, CEOs with higher
compensation relative to other executives (as another proxy for CEO power) seem to have more frequently incentive plans tied to non-financial performance measures. So, this
produces mixed findings with regard to the question whether CEOs with greater power have their incentive contracts more often tied to non-financial measures. In addition firm size (natural logarithm of assets) is significantly higher for firms that have non-financial measures in their executive contracts, while the market to book ratio (Growth) is higher for firms that only use financial performance measures. This indicates that bigger firms have tied their CEO incentive plan more often with one or more non-financial performance measures. On the other hand firms with greater growth opportunities rely more on financial performance measures. Volatility in performance (for both the accounting metric and the market metric) and the amount of long term equity of the CEO are not significant different for firms that use financial or non-financial performance measures.
Table 3
t-test of differences in means
Financial measures (n=119) Non-financial measures (n=81)
Variable Mean Std. Dev Mean Std. Dev t-test
CEO_Tenure 10.706 7.470 8.185 5.937 2.54 Rel_Tenure 0.933 0.651 0.714 0.518 2.54 Rel_Comp 2.074 1.217 2.470 2.099 -1.69 STDev_ROA 0.043 0.049 0.046 0.064 -0.43 STDev_stockret 0.417 0.241 0.396 0.241 0.55 LT_Equity 8.290 5.376 8.629 4.737 -0.46 Size 9.511 1.102 9.872 1.047 -2.32 Growth 4.892 4.290 3.085 4.422 2.89
Significant level of 5% or more is reported in bold.
Table 4 reports a Pearson correlation matrix between all the independent variables. First, CEO tenure and relative tenure are strongly correlated as they both take into account the tenure of the CEO at the respective sample firm. Furthermore, long-term equity holdings are
positively correlated with all three proxies of CEO power, while firm size is negatively correlated with the tenure-related proxies of CEO power and positively correlated with the compensation-related proxy for CEO power. This indicates that bigger firms have CEOs that earn higher relative compensation, which is consistent with the intuition that bigger firms demand higher quality CEOs who expect higher compensation. The negative correlation between firm size and the tenure-proxies suggest that CEOs tend to stay longer in smaller firms than in the bigger firms. One explanation for this is that the position of the CEO in bigger firms is highly competitive and CEOs who do not perform well are easily replaced, leading to a shorter tenure.
Table 4
Pearson correlation coefficients
Variable 1 2 3 4 5 6 7 8 1. CEO_Tenure 1.00 2. Rel_Tenure 1.00 1.00 3. Rel_Comp 0.08 0.08 -1.00 4. STDev_ROA -0.10 -0.10 0.10 1.00 5. STDev_stret -0.06 -0.06 0.16 0.38 1.00 6. LT_Equity 0.13 0.13 0.57 -0.03 0.13 1.00 7. Size -0.21 -0.21 0.32 0.02 -0.01 0.04 1.00 8. Growth 0.09 0.09 -0.03 0.01 -0.06 -0.01 -0.26 1.00 Significant level of 10% or more is reported in bold.
4.2 Main analyses
In order to test the hypothesis and to see whether there is a relation between CEO power and the use of non-financial performance measures a regression model is used. Because the dependent variable (NF) can only take two values (either 1 if non-financial performance measures are used in the CEO incentive contract or 0 otherwise), a logistic regression model is used to estimate the coefficients.
Nonfin= β0 + β1 CEO_Power + Controls +
ɛ
z-statistics in parentheses: *** p<0.01, ** p<0.05, * p<0.1
Table 5 reports the results of the first model. On the basis of the hypothesis, a positive relation is expected between the three proxies of CEO power and the use of non-financial measures in CEO incentive contracts. The three columns report the results for each of the three proxies of CEO power. In general, the findings lead to the rejection of the hypothesis. First, the coefficient on CEO tenure and relative tenure as two proxies for CEO power are both negative significant (-0.06 and -0.60 respectively). This indicates that when tenure increases (note that tenure is a widely used proxy for CEO power), firms are more likely to rely solely on financial information in their executive contracts. In other words, CEOs who are longer in the organisations actually have incentives that are solely tied to financial rather than non-financial measures. The coefficient on relative CEO compensation (as the third proxy for CEO power) is positive but not significant. Therefore, on the basis of these findings, the hypothesis is rejected.
Table 5
Logit Regression: Test of model 1
relation between CEO power and non-financial performance measures
Variables Predicted sign CEO_Tenure Rel_Tenure Rel_Comp
CEO_Power + -0.05** (-2.12) -0.60** (-2.12) 0.14 (1.03) LT_Equity 0.02 (0.81) 0.02 (0.81) -0.01 (-0.28) Size 0.17 (1.18) 0.17 (1.18) 0.16 (1.02) Growth -0.10** (-2.25) -0.10** (-2.25) -0.10** (-2.35) Constant -1.40 (-0.93) -1.40 (-0.93) -1.75 (-1.15) Number of obs 200 200 200 LR chi2(4) 16.66 16.66 12.98 Prob > chi2 0.002 0.002 0.011 Pseudo R2 0.062 0.062 0.048
With regard to the control variables, the variable growth (market to book ratio) has a value of -0.10, which is negative and significant (p<0.05). This shows us that firms with greater growth opportunities tend to rely on financial rather than non-financial performance measures. All models are highly significant as the first two models are significant at the 1% level while the final model has a p-value of 0.011.
In addition a second model was used to further test the hypothesis whether the relation between volatility of firm performance and the inclusion of non-financial performance measures in CEO incentive plans is affected by the power of the CEO. The interaction of volatility in firm performance and CEO power was added in this model in order to test whether powerful CEOs diversify risk (when the volatility in performance increases) by including non-financial performance measures in their contracts. Table 6 reports the results of the second model.
Again, the three columns provide the regression results for each of the three proxies of CEO power. In general, these results also lead to the rejection of the hypothesis. There are no findings that indicate that CEO power positively influences the relationship between volatility and the use of non-financial measures in CEO contracts. On the contrary, the interaction between volatility in market performance and (relative) tenure is negative and significant. This goes against the expectations of the hypothesis and states that CEOs with a longer tenure (and thus more power) when facing volatile market performance tend to rely more on financial rather than non-financial performance measures. The interaction between the volatility in market performance and relative CEO compensation is positive but not significant. Furthermore, all three interactions between the proxies of CEO power and the volatility in accounting performance are not significant.
With regard to the control variables, just like the first model, the variable growth (market to book ratio) has a value of -0.10, which is negative significant.
The Likelihood Ratio Chi-square test gives a value between 15.10 and 22.70. This model is highly significant for all levels (except for the proxy CEO relative compensation, which is not significant below a significance level of 0.05).
Based on the above results, it can be concluded that there is no indication that powerful CEO include non-financial performance measures in their compensation contract in order to diversify their risk. There is no significant relation between CEO power and the relationship between volatility and the use of nonfinancial measures in CEO bonus plans. Based on the
above results, the hypothesis has to be rejected. One interesting (surprising) finding is however that CEO (relative) tenure is negatively associated with the use of non-financial performance measures and the interaction with market volatility. While the literature agrees that non-financial performance measures can be used as a tool to manipulate compensation payments, these results suggest the contrary.
Nonfin= β 0 + β1 CEO_Power + β2 Volat + β3 Volat*CEO_Power + Controls+ ɛ
z-statistics in parentheses: *** p<0.01, ** p<0.05, * p<0.1
Table 6
Logit Regression: Test of model 2
relation between CEO power, volatility and non-financial performance measures
Variables Predicted sign CEO_Tenure Rel_Tenure Rel_Comp
CEO_Power 0.06 (1.10) 0.74 (1.10) -0.04 (-0.16) STDev_ROA -0.74 (-0.15) -0.74 (-0.15) 1.9 (0.37) STDev_stockret 2.30 (1.60) 2.30 (1.60) -1.58 (-1.12) STDev_ROA* CEO_Power + 0.12 (0.20) 1.38 (0.20) 0.05 (0.03) STDev_stockret* CEO_Power + -0.34** (-2.15) -3.95** (-2.15) 0.32 (0.69) LT_Equity 0.04 (1.25) 0.04 (1.25) 0.00 (-0.04) Size 0.18 (1.21) 0.18 (1.21) 0.18 (1.10) Growth -0.10** (-2.29) -0.10** (-2.29) -0.10** (-2.38) Constant -2.29 (-1.42) -2.29 (-1.42) -1.33 (-0.85) Number of obs 200 200 200 LR chi2(8) 22.70 22.70 15.10 Prob > chi2 0.004 0.004 0.057 Pseudo R2 0.0841 0.0841 0.0559
4.3 Additional analyses
4.3.1 Alternative specification of the dependent variable
The relation between CEO power and the use of non-financial performance measures used in executive compensation contracts was tested in the previous section. The same analyses are repeated, but now the number of non-financial performance measures and the total weight assigned to non-financial performance measures are used as an alternative dependent
variable. Because a linear relation will be expected the ordinary least squares (OLS) method is used to estimate the coefficients.
Table 7 and 8 reports the results. Based on these results none of the variables are significant. There is no significant relation between either of the variables and the number of non-financial measures used in the executive compensation contracts. Moreover there is also no significant relation between the weight of non-financial performance measures and the other variables. This confirms the earlier results from the previous tests where there is no positive significant relation between CEO power and the relationship between volatility and the use of nonfinancial measures in CEO bonus plans. Moreover the first model for both the number and weight of non-financial performance measures is not significant. The second model (with the inclusion of volatility in performance) is significant below 5%. Therefore looking at the number of non-financial measures or the weight assigned to non-financial measures in CEO bonus plans does not change the previous conclusion. CEO power has no significant effect on the use of non-financial performance measures in CEO incentive contracts.
t-statistics in parentheses: *** p<0.01, ** p<0.05, * p<0.1
Table 7
OLS Regression: Test of model 1
Relation between CEO power and (alternative) non-financial performance measures
Number_NF Pred. sign CEO_Tenure Rel_Tenure Rel_Comp Weight_NF Pred. sign CEO_Tenure Rel_Tenure Rel_Comp
CEO_Power + -0.02 (-0.46) -0.26 (-0.46) -0.23 (-0.78) CEO_Power + -0.00 (-0.63) -0.03 (-0.63) -0.00 (-0.00) LT_Equity 0.00 (0.03) 0.00 (0.03) 0.05 (0.58) LT_Equity 0.00 (0.54) 0.00 (0.54) 0.00 (0.31) Size -0.20 (-0.86) -0.20 (-0.86) -0.08 (-0.28) Size -0.02 (-0.81) -0.02 (-0.81) -0.02 (-0.67) Growth 0.04 (0.65) 0.04 (0.65) 0.04 (0.63) Growth 0.00 (0.48) 0.00 (0.48) 0.00 (0.50) Constant 5.23** (2.16) 5.23** (2.16) 3.95* (1.46) Constant 0.42* (1.92) 0.42* (1.92) 0.39 (1.62)
Number of obs 56 56 56 Number of obs 45 45 45
F statistic 0.37 0.37 0.48 F statistic 0.40 0.40 0.29
Prob > F 0.826 0.826 0.752 Prob > F 0.810 0.810 0.880
t-statistics in parentheses: *** p<0.01, ** p<0.05, * p<0.1
Table 8
OLS Regression: Test of model 2
relation between CEO power, volatility and (alternative) non-financial performance measures
Number_NF Pred. sign CEO_Tenure Rel_Tenure Rel_Comp Weight_NF Pred. sign CEO_Tenure Rel_Tenure Rel_Comp
CEO_Power 0.08 (0.65) -0.92 (0.65) -0.29 (-0.54) CEO_Power -0.00 (-0.47) -0.05 (-0.47) 0.02 (0.36) STDev_ROA 4.99 (0.45) 4.99 (0.45) -1.04 (-0.10) STDev_ROA -0.01 (-0.01) -0.01 (-0.01) 0.36 (0.34) STDev_stockret 1.60 (0.63) 1.60 (0.63) 0.02 (0.01) STDev_stockret -0.05 (-0.22) -0.05 (-0.22) 0.06 (0.33) STDev_ROA*CEO_Power + -0.98 (-0.70) -11.25 (-0.70) -0.29 (-0.54) STDev_ROA* CEO_Power + 0.04 (0.36) 0.51 (0.36) 0.036 (0.12) STDev_stockret*CEO_Power + -0.27 (-0.70) -3.08 (-0.70) -0.00 (-0.01) STDev_stockret*CEO_Power + 0.00 (0.16) 0.05 (0.16) -0.03 (-0.46) LT_Equity 0.03 (0.38) 0.03 (0.38) 0.06 (0.60) LT_Equity 0.00 (0.28) 0.00 (0.28) -0.00 (-0.09) Size -0.14 (-0.54) -0.14 (-0.54) -0.06 (-0.18) Size -0.02 (-0.80) -0.02 (-0.80) -0.02 (-0.65) Growth 0.04 (0.62) 0.04 (0.62) 0.04 (0.60) Growth 0.00 (0.27) 0.00 (0.27) 0.00 (0.31) Constant 3.84 (1.31) 3.84 (1.31) 3.85 (1.32) Constant 0.47 (1.67) 0.47 (1.67) 0.39 (1.44)
Number of obs 56 56 56 Number of obs 45 45 45
F statistic 0.33 0.33 0.22 F statistic 0.28 0.28 0.28
Prob > F 0.952 0.952 0.985 Prob > F 0.969 0.969 0.968
4.3.2 Addressing the potential of outliers/influentials
The results of the regression models may be strongly influenced by a limited number of observations. These outliers or influentials may shift the results towards a certain direction. In order to test for this possibility an additional robust regression is used. The findings are not tabulated and the results suggest that there is no significant change in the variables compared to the previous tests. This means that after using the robust regression which gives results which are less sensitive to outliers, the main conclusion from the earlier tests will not change. The coefficients of the tenure related proxies for CEO power (and the interaction with stock market volatility) are still negative significant, while CEO relative compensation (and interaction with volatility in firm performance) is not significant. Therefore there is no indication that powerful CEOs include non-financial performance measures in their bonus contract in order to diversify their compensation risk. Appendix D reports the results.
5. Conclusion
It is argued that incentive plans are a means for powerful CEOs to inflate their compensation instead of serving its original purpose of aligning the interests between CEOs and
shareholders. Prior research focused mostly on whether powerful CEO can influence their compensation in weak corporate governance settings. Furthermore prior studies found that powerful CEO can increase their compensation by shifting the weighting of different performance measures towards the better performance measures. This thesis adds to the literature by empirically testing whether powerful CEO may diversify their compensation risk, by including multiple financial and non-financial performance measures in their compensation plan. However based on the results in this thesis there is no indication that powerful CEO use non-financial performance measures as a tool to diversify compensation risk in their bonus plan.
First, based on the results of the first model there is no significant relation between CEO power and the use of non-financial performance measures in CEO bonus plan. One surprising result is however that CEO tenure is negatively correlated with the use of non-financial performance measures. While it is argued that non-non-financial performance measures can be used by powerful CEO to influence and manipulate compensation from the bonus
plan, these results show the opposite. If CEOs are longer in the organisation they actually use less non-financial performance measures and thus rely more on financial information.
Second, another surprising result is that the interaction between volatility in market performance (based on stock returns) and CEO (relative) tenure is negatively associated with the use of non-financial performance measures. Again the literature suggests that volatility in (market) performance will lead to greater compensation risk for CEOs. If this is the case, then powerful CEO will try to limit this risk by including multiple non-financial measures.
However based on the results in this thesis powerful CEO will rely more on financial measures when there is market volatility. One possible explanation for both these
contradicting results is that CEOs who are in the same organisation for a longer period of time are set in the traditional way of measuring performance. Financial (accounting) measures are still widely used by all organisations even though those measures are lacking, the main reason for this is that financial measures are easy to understand. It may be that CEOs who are in the organisation for a longer period of time find it better and easier to rely on financial rather than non-financial performance measures.
This thesis however has some limitations. First, the small sample size will have a big influence on the found results. Due to the lack of observations the models used in the
alternative analyses to measure the number and weight of non-financial performance
measures do not give any significant results. The number of observations is simply too small to get good reliable results. In addition to this, the results found in the main analyses should be taken with caution. Again the small sample size may not be representative. It might be interesting to see whether additional research may find similar research.
Furthermore because data is collected manually from the proxy statements, some errors can be made in interpreting the information in the proxy statements. While in most cases it is very clear if non-financial performance measures are used. Some firms use vague explanations in their proxy statement, multiple interpretations are therefore possible. Related to this is the fact that the information in proxy statements is self-reported by the firms. This will also affect the reliability and the validity of the information. Given the limitations
mentioned earlier and the fact that the findings are contradicting to the predictions, additional research on the subject whether powerful CEO diversify their compensation risk by including non-financial performance measures in their bonus plan is recommended.
References
Adams, R. B., Almeida, H., & Ferreira, D. (2005). Powerful CEOs and their impact on corporate performance. The Review of Financial Studies, 18(4), 1403-1432.
Banker, R. D., Potter, G., & Srinivasan, D. (2000). An empirical investigation of an incentive plan that includes nonfinancial performance measures. Accounting Review;
Account.Rev., 75(1), 65-92.
Banker, R., & Thevaranjan, A. (2000) "Goal Congruence and Evaluation of Performance Measures," working paper, University of Texas at Dallas.
Bebchuk, L., & Fried, J. (2003). Executive compensation as an agency problem. Journal of Economic Perspectives; J.Econ.Perspect., 17(3), 71-92.
Bushman, R., R. Indjejikian, and A. Smith. 1996. CEO compensation: The role of individual performance evaluation. Journal of Accounting and Economics 21 (April): 161-193.
Campbell, D. (2008). Nonfinancial performance measures and promotion- based incentives. Journal of Accounting Research, 46(2), 297-332.
Core, J., Holthausen, R. W., & Larcker, D. (1999). Corporate governance, chief executive officer compensation, and firm performance. Journal of Financial Economics; J.Financ.Econ., 51(3), 371-406.
Datar, S., Kulp, S. C., & Lambert, R. (2001). Balancing performance measures. Journal of Accounting Research; J.Account.Res., 39(1), 75-92.
Dechow, P. M., & Sloan, R. G. (1991). Executive incentives and the horizon problem: An empirical investigation. Journal of Accounting and Economics, 14(1), 51-89.
Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. The Journal of Law and Economics, 26(2), 301.
Grinstein, Y., & Hribar, P. (2004). CEO compensation and incentives: Evidence from M&A bonuses. Journal of Financial Economics, 73(1), 119-143.
Hermalin, B., & Weisbach, M. S. (1991). The effects of board composition and direct incentives on firm performance. Financial Management; Financ.Manage., 20(4), 101-112.
Holmström, B. (1979). Moral hazard and observability. Bell Journal of Economics, 10(1), 74.
Hopwood, A. G. (1972). An empirical study of the role of accounting data in performance evaluation. Journal of Accounting Research, 10, 156-182.
Indjejikian, R., & Matejka, M. (2012). Accounting decentralization and performance evaluation of business unit managers. Accounting Review, 87(1), 261-290.
Ittner, C. D., Larcker, D., & Meyer, M. (2003). Subjectivity and the weighting of performance measures: Evidence from a balanced scorecard. Accounting Review; Account.Rev., 78(3), 725-758.
Ittner, C. D., Larcker, D., & Rajan, M. (1997). The choice of performance measures in annual bonus contracts. Accounting Review; Account.Rev., 72(2), 231-255.
Jensen, M. C. (1993). The modern industrial- revolution, exit, and the failure of internal control- systems. Journal of Finance; J.Financ., 48(3), 831-880.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305-360.
Kaplan, R., & Norton, D. (1992). The balanced scorecard - measures that drive performance. Harvard Business Review; Harv.Bus.Rev., 70(1), 71-79.
Kaplan, R., & Norton, D. (1996). Using the balanced scorecard as a strategic management system. Harvard Business Review; Harv.Bus.Rev., 74(1), 75-&.
Kim, S.K. and Suh, Y.S. Ranking of Accounting Information Systems for Management Control. Journal of Accounting Research (Autumn 1991): 386-396.
Marks S.G., 2000, ‘The separation of ownership and control’, Encyclopedia of Law and Economics, Volume III. The Regulation of Contracts, pp. 692–724
Moers, F. (2006). Performance measure properties and delegation. Accounting Review; Account.Rev., 81(4), 897-924.
Morse, A., Nanda, V., & Seru, A. (2011). Are incentive contracts rigged by powerful CEOs? Journal of Finance, 66(5), 1779-1821.
Perera, S., Harrison, G., & Poole, M. (1997). Customer- focused manufacturing strategy and the use of operations- based non- financial performance measures: A research
note. Accounting Organizations and Society; Account.Organ.Soc., 22(6), 557-572.
Sliwka, D. (2002). On the use of nonfinancial performance measures in management compensation. Journal of Economics & Management Strategy;
Appendix
Appendix A Industry sector
Total sample Firms that use NFPM
SIC Codes n % n %
Mining 1000 – 1499 22 11,00% 10 12,35%
Construction 1500 – 1999 3 1,50% 2 2,47%
Manufacturing 2000 – 3999 114 57,00% 52 64,20%
Transport & Communication (excluding utilities)
4000 – 4999 21 10,50% 6 7,41%
Wholesale and retail estate 5000 – 5999 9 4,50% 3 3,70%
Services 7000 - 8999 31 15,50% 8 9,88%
Total 200 100,00% 81 100,00%
Appendix B
Examples of proxy statement
Contracts without non-financial performance measures:
CITRIX SYSTEMS INC (proxy filed April 11, 2014, p. 23)
Furthermore, while the establishment of variable compensation targets for our executives necessarily involves judgment, the actual payouts against those targets are based on
predetermined, objective financial criteria reflective of our annual operating plan or total
FASTENAL CO (proxy filed February 26, 2014, p. 22)
The cash bonuses for all of our named executive officers other than Mr. Florness (CFO) and Mr. Rucinski (VP SALES) are based on growth in pre-tax earnings of the officer’s area of responsibility. The compensation committee selected pre-tax earnings as the appropriate metric for calculating cash bonuses for those officers because of the committee’s belief that the focus of the named executive officers should be on profitability.
Contracts with non-financial performance measures:
3M CO (proxy filed March 26, 2014, p. 49)
These performance evaluations are done according to 3M's overall performance assessment and management processes, which involve setting annual financial and non-financial (e.g., commitment to diversity) goals and objectives for each individual and then assessing the individual's overall performance against these goals and objectives at the end of the year.
BRISTOL-MYERS SQUIBB CO (proxy filed March 24, 2014, p. 34)
We structure our compensation program to align the interests of our executives with the interests of our stockholders. We believe that an executive's compensation should be directly tied to helping us achieve our mission and deliver value to our stockholders. Therefore, a substantial portion of an executive's compensation is variable and at risk in the form of annual bonus and equity awards that vary in value based on company financial results and our TSR over one or more years. In addition, a significant portion of each executive's pay depends on his or her individual performance against pre-determined strategic, financial
and operational objectives as well as key behaviors.
Contracts with multiple non-financial performance metrics:
CATERPILLAR INC (proxy filed April 21, 2014, no page numbers given)
The Committee approved the other NEOs’ annual incentive compensation and proposed adjustments based on 2013 performance and the benchmarking information discussed above. In making these determinations, the Committee considered the most critical results for each of the NEOs in 2013 with respect to their business units, which included many of the factors
described in the CEOs evaluation above, as well as financial performance measures such as accountable profit, OPACC, operating cash flow, return on equity and return on sales; and
non-financial measures including PINS, POPS-C, as delivered quality and reliability and
safety results; in addition to the successful launch of NPI programs, cost management, inventory reduction and diversity and inclusion initiatives.
CENTURYLINK INC (proxy filed April 16, 2014, pp. 36-37)
Bonuses payable with respect to attainment of “strategic initiatives” were based on the Compensation Committee’s assessment of each senior officer’s specific contributions
regarding some or all of the following five strategic initiatives selected by the Committee in consultation with management:
- attainment of growth goals for our facilities-based internet protocol television service (IPTV), marketed as PrismTM TV
- increases in the number of customers in our markets purchasing our data services, which we refer to as our data penetration rate
- the number of completed installations of fiber optic cables linking our network to nearby wireless towers
- simplification of our network designed to reduce costs and improve our operations - attainment of hosting growth goals
Contracts with specific weight on non-financial performance measures:
CVS HEALTH CORP (proxy filed March 28, 2014, p. 28)
Customer service and client satisfaction account for the remaining 20% of award funding; 10% is based on the Retail Customer Service score, which measures customer
service in the retail segment and 10% is based on an aggregation of client satisfaction metrics from the PBM segment, covering mail order, specialty pharmacy and account/client services.
CAMERON INTERNATIONAL CORP (proxy filed April 1, 2014, p. 43)
For 2013, the Committee weighted the EPS, cash flow from operations, Total Recordable
Incident (TRIR) and voluntary attrition rate objectives for corporate executives at 60%,
20%, 10% and 10%, respectively, to ensure management was focused on earnings, cash generation, safety and employee engagement.
