Financial incentives and the role of personality A biopsychological view

(1)

Financial incentives and the role of personality

A biopsychological view

Master Thesis Nico Arjen Miedema

University of Groningen

Faculty of Economics and Business

MSc Business Administration

Specialization: Organizational and Management Control Juli 2013 Arjen Miedema Address: Birstum 13 Postal Code: 8494MZ City: Nes Mobile: 0651775858 MSc Business Administration

Specialization: Organizational & Management Control S1798375

Juli 2013

(2)

Financial incentives and the role of personality

A biopsychological view

Abstract

Financial incentives have been used for a very long time to motivate employees. Financial

incentives are traditionally used in results and action controls. The goal of this literature study is to get a better understanding of the motivating effect of incentives and how personality influences this, this will give a foundation for later laboratory and field studies.

Currently the workings of incentives are not fully understood. The framework presented by Ouchi (1979) is one of the most popular theories to determine if incentives should be employed. However research by Jansen et al. (2009) and Ariely (2007) show that the motivating effect of incentives is not as predictable as previously thought. Incentives do not always enhance motivation or increase performance, even in the situations described by Ouchi (1979) to require incentives.

In this study two hypothesis will be tested to determine the effect of an employees personality on the motivating effect of financial incentives. Personality is defined using the coping styles as discussed by Koolhaas et al. (1999) whilst incentives will be based on the management control definition. The coping styles defines two personality styles, the flexible reactive coping style and the routine seeking proactive coping style.

The hypothesis tested were Ha1: reactive individuals should respond differently to incentives than

proactive individuals, in the same circumstances. Ha2: reactive individuals will be more motivated by incentives than proactive individuals, in the same circumstances. Hypothesis one was found to

be true whilst hypothesis two was not supported, instead it was found, in a lab test, that pro-active individuals are more motivated by financial incentives than reactive individuals.

Keywords:

(3)

Introduction

Managers often use results controls to reward and punish their subordinates. It is generally assumed that employee compensation, using incentives, can be an effective way to motivate employees. If done correctly, research has shown that incentives increase pay satisfaction of employees and also their productivity (Guzzo et al. 1985). However recent research by Jansen et al. (2009) show that the pay satisfaction and performance related to incentive compensation differs globally in the same industries. It was assumed until recently that a global best practice regarding incentive

compensation existed (Chiang & Birtch, 2007) however Jansen et al. disproved this.

It can be concluded that a management control problem exists, what determines the effectiveness of incentives is unknown as clearly not every factor influencing this has been identified. According to Jelinek et al. (1981) the central objective of any control system should be to increase the

predictability of any organizational task as well as making organizational activities easier to coordinate. However, the motivating effect of incentives are not fully understood. When managers employ results controls using for example the framework suggested by Ouchi (1979), results may not be as the manager expected. San Miguel (1977) explains that the individual has a lot of influence on the motivating effect of management control practices, for example he states that money, as a reward, only motivates to a certain extend. Ariely et al. (2007) complements this, by showing that monetary incentives do not need to motivate people more in this case the engagement in pro-social behavior was measured and found to be depending on the effect of the incentive on the persons social image. It shows that people take a lot of factors into consideration be it conscious or subconscious. The management control problem that this gives is unpredictability, a management control system that uses incentives to motivate employees may have large variances in individual employee performance, due to individual differences. However, few studies have examined the relationship between personality and other individual differences and the effectiveness of incentives.

In this study personality will be researched as a factor that might influence the effectiveness of (financial) incentives. Das (1986) shows that the personality of a controller has influence on the control actions taken by the manager. This may not seem relevant, but it shows that an individuals personality might influence the decisions he-she takes. As motivation is expressed in conscious behavior, this can be seen as a type of decision, namely to work harder, to earn more money for example. The study by Das (1986) shows the influence that an individuals personality can have, therefore it is also likely that this will also hold for employees.

In the last few years neuroscience has advanced in the understanding of differences between individuals and the role of personality. Boersma et al. (2011) show that different types of

personalities respond differently to certain situations. They show that some types of diseases, in this case insulin resistance, is more prone to develop in individuals with a certain personality type, this may not seem relevant but it proves that individuals differ and their response to certain stimuli differs as well. The responses to external stimuli differs between personality types. It is therefore interesting to see if incentives and personality have a connection. This could help management accountants better understand how incentives affect individuals and also give room for a new body of follow-up research.

Research question

As mentioned earlier the central theme of this study is the relationship between personality and the effectiveness of incentives. This leads to the following research question.

(5)

To answer this question the following sub questions need to be answered.

What are incentives? When do we use incentives? When are incentives effective? What is personality?

How to measure personality?

How does personality relate to incentives?

Is there a relationship between the motivating effect of incentives and personality? Conclusions that can be drawn regarding how an individuals personality influences the effectiveness of financial incentives?

Most of these sub questions are straightforward literature explorations, with the goal to provide a strong literature base for the thesis. However, to answer the research question, incentives and personality need to be expressed in either management accounting terms of neuro-scientific terms. This is important as both definitions originated in different disciplines in science. In other words, these definitions can not be used together, to answer the research question, without proper analyses first. With the subquestion “How does personality relate to incentives” the prior described issue can be answered.

It is important to note that the use of incentives and how to determine their effectiveness is needed to relate the end results back to the management accounting issue. The relationship between personality and incentives is also important to explore not only because of the main question but also to determine if incentives influence personality over time, in other words how stable is a personality and will it be affected by the incentives?

Literature overview

Introduction literature overview

In this section the literature used will be explained briefly to present a brief overview which will be used to formulate the hypotheses of this study. First incentives will be discussed followed by the types of control systems they are used in. Then the effectiveness of incentives is formulated followed by a short definition personality. This section will then be followed up with the research question and hypotheses.

Incentives

Incentives are means used by managers to change the direction of effort, duration and intensity of labour in the direction the manager aims for are part of action and results controls (Bonner and Sprinkle, 2002). Incentives are used in two forms of management control, namely action and results control. These are two control groups used in management control literature. They are used to design organizational control systems and are used to control employees to perform actions in favor of the organizations goals. Both of these will now be explained shortly.

Results control

Results controls are controls that reward employees for achieving certain goals. The goals need to be well defined, employees need to be able to influence the desired areas being controlled, and the organization must be able to measure the result effectively. (Merchant and Van der Stede (2007), p25-35)

(6)

The use of Results controls can also give rise to behavioral displacement. This happens when the result measures are not in line with the company goals. Or because of over-quantification, this happens when the results measured are only the concrete and quantifiable results. This causes the hard to measure (in most cases the intangible) results to be overlooked. (Merchant en van der Stede, p 180-182)

Action control

Action controls are controls that directly influence the employees behavior. There are four basic forms namely, behavioral constraints, preaction reviews, action accountability and redundancy. In this proposal only the action accountability will be discussed as this is the only form that can be directly linked to a reward or punishment.

Action accountability holds employees accountable for the actions they take, they will be rewarded or punished accordingly to the desirability of the actions. Action controls solve problems related to a lack of direction, motivation and personal limitations.

The success of action controls depends on two variables, organizations must be able to determine what desirable actions are, and organizations are able to ensure that undesirable actions do not occur. (Merchant en van der Stede, 2007, p76-92)

Ouchi (1979) would place action controls in cell 3, in this cell there is perfect understanding of the transformation process, however the ability to measure outputs can be low.

Behavioural displacement is also associated with action controls. It is known as the mean-ends inversion, meaning that actions receive more attention and focus instead of the end goals. This causes employees to stop thinking about how to improve performance and their jobs but focus mainly on executing their tasks. As this bears the lowest risk. (Merchant en van der Stede, (2007) p182-183)

The effectiveness of incentives

In this study the effectiveness of incentives will be based upon the increase in business or individual performance associated with the use of the incentive. In next paragraphs proof will be presented that the effectiveness of incentives is dependent on many factors.

According to Banker et al (1996), the introduction of incentives in tasks that employees find

intrinsically interesting reduces performance and interest in the task, as employees feel that they are treated as incompetent and experience the incentive as controlling. Also according to Guzzo, Jette, and Katzell, (1985) performance is only improved by financial incentives if the task is boring and monotonous. Interesting to note is that Guzzo et al. (1985) find that financial compensation does not have a significant effect on withdrawal but does have it on output. (p.286)

Jansen et al. (2009), show that pay satisfaction differs from setting and thus also the attitude towards incentives is bound to more than just the incentive itself. They found that US car dealers almost solely rely on incentive compensation whilst in the Netherlands only a small part of the employee compensation is based on incentives. Moreover in the Netherlands the use of incentives seemed to lower performance. They suggest that this is strongly bound to culture, however this is merely based on correlations.

Interestingly, monetary incentives do not always enhance performance, according to Awasthi and Pratt (1990) employees, with low intuitive and perceptual abilities, do not perform better after the introduction of financial incentives. This proves that the individual abilities of an employee

influence the effectiveness of incentives. The traditional way of using incentives would not account for individual differences, however these might attribute greatly to the variances in performance related to incentives.

Personality

(7)

originally presented by Henry (1977), are stress response patterns. Two styles have currently been identified namely, proactive and reactive coping. This approach to personality has been chosen as it is better to measure than for example measuring the tolerance of ambiguity, as Das (1986) does. Therefore it can be used by managers in a practical setting. Also tolerance of ambiguity measures only one personality trait, the coping strategies are wider and larger in scope providing a clearer indication of a person's stress response style, the tolerance of ambiguity shows some overlap with the definitions of the coping strategies. People with low tolerances of ambiguity prefer more stable environments than people who don't (Das, 1986), this overlaps with the proactive coping style. Proactive coping is an active coping style, the animal is characterized by showing higher degrees of territorial control and aggression. This coping style is based upon the work of Cannon's (1915) fight-flight response. Proactive animals show a proactive behavioral response to environmental challenges (Koolhaas et al. 1999).

Reactive coping is a passive stress response. Engel and Schmale (1972) described it as a

''conservation-withdrawal'' response. In essence it means that an animal is more prone to show low levels of aggression and reacts more adaptive and flexible to environmental stimuli, only when absolutely needed (Koolhaas et al. 1999).

Ursin (1980), shows that the coping styles are also present in humans. Boersma et al. (2011) talks about coping strategies and how they relate to certain behaviors. Originally they researched the role of personality in the treatment of type 2 diabetes. Their conclusion is that reactive personalities are more prone to develop insulin resistance (a major risk factor in type 2 diabetes). However they also showed that personality type is important in the treatment of the patient. Not only are the hormonal differences between personality types which explains why some medication will only work on certain individuals also the patient's compliance to changes, in this case lifestyle, is directly influenced by personality types (Hassmen, 1993; Ryden et al., 2001)

These findings give reason to believe that personality may also influence the response on

incentives. Incentives are external stimuli and thus the response by reactive individuals should be different from proactive individuals. Likely the human reward system will play an important role, in the processing of incentives, by the brain. The reward system is a brain system that allows us to represent environmental stimuli on the basis of an abstract coding. This way a person can compare the value of diverse stimuli, from different sources, and use this as a basis to make decisions. (Peters and Büchel, 2010) The values attributed to certain stimuli is based upon previous

experiences (Kable and Gilmcher, 2009 ). Choices made by a person are based upon their judgment of stimuli and the emotional response they have (Loewenstein et al. 2001). Emotions are states that are entered by a person, in response to a reward or a punishment. Conditioning plays a large part in the development of emotional responses. First, stimuli will enforce the behavior that they provoke, secondly due to association, stimuli will be able to invoke emotional responses based on the anticipation of the results of the stimuli (Foxhall, 2010). Basically if behavior leads to a reward a person will respond more positive to it and learn that the stimuli is good for them (or so they perceive it, drugs etc. are also perceived as ''good'' by the brain). Linking this back to management control, if a reward is effective there is a basis to believe that a certain reward structure could reinforce itself, urging a person to participate in it more. This could mean that the coping strategy can change over time, as a response to an incentive and will lead to different conclusions of the research question.

This brings us to the hypothesis of this theses.

Ha1: reactive individuals should respond differently to incentives than proactive individuals, in the same circumstances.

(8)

The study will give insight into the possible effect of personality on the effectiveness of incentive on the individual employee. Currently, it is unknown to managers what the best course of action is to design their control systems. This study will provide insights that will improve the predictability of their (financial) incentive based control systems.

(9)

Methodology and conceptual model

Introduction

In this section the research method, to answer the research question, will be discussed. It seems most appropriate to do an empirical study by doing a field study, however the choice was made to do an exploratory literature study instead, as the use of personality as a factor, has not been used prior in management accounting literature. To expand on the previous sentence, as the gap between theory and practice regarding personality and incentives is still large, this gap will have to be closed. The final step in this process will be a field study, however these need a scientific basis. Exploratory research leads to empirical research if the exploratory research had positive results (Scapens, 2004). Research regarding personality and reward-related behaviors has been done in neuroscience (Koolhaas et al., 1999) however, the definitions used in these studies do not exist in management control. In order to use the research and studies in neuroscience the definition of personality needs to be adapted to work with management control literature. This means that behaviors described in neuroscience need to be translated to a management control equivalent and financial incentives need to be translated to a behavioral equivalent in neuroscience. The goal of this study is to gain a wider understanding of the relation between personality and the workings of incentives. To do this empirical research is needed. While this study is not an empirical study it can be perceived as the first step to embed the role of employee personality in management control which could eventually lead to laboratory studies and field studies as suggested by Scapens (2004). As discussed before, the research question requires that financial incentives and personality are clearly defined. This to give a strong basis for the rest of the study and the analysis that follows. Incentives are a form of reward, money activates the reward system, in the human brain, somewhat similar to food and other ''natural incentives'' (Elliot et al. 2000), as other mammals also have the same personalities types it is worthwhile to look at animal studies (Ursin, 1980). It is necessary in this case to research how a species differs to a human. Animal studies usually use some form of incentive to motivate the animals to perform certain tasks. Research related to this might give insights to answer the research question.

Methodology, how to gauge personality.

It is also needed to understand how personality is measured in order to understand how to interpret the research. Boersma (2011) uses in her research two different scales. The first is the Bortner scale, this scale developed by Bortner and Roseman (1967) uses two different components, one is a questionnaire the subject has to answer. The other is a set of tasks that try to determine if someone has a proactive personality. These two were later found to have shown no relationship with one and another. It is therefore assumed that both measure different aspects of a proactive personality and can be used together to create a more objective and reliable scale.

The use of a questionnaire has often been used, to determine the personality type. Bortner and Rosenman (1967) used the Jenkins Activity Survey, they found that it correctly identified 72% of their population as proactive individuals. The advantage of using the questionnaire is the easy of using it and also the easy with which results can be replicated and are consistent through time (Jenkins et al., 1968). However it is prone to be influenced by cultural differences and practice effects (Bortner and Rosenman, 1967).

The tests have the advantage that they are much more accurate, moreover it is free from cultural differences as an influencing factor. However much more skill is required by the researcher in registering the results. (Bortner and Rosenman, 1967)

(10)

assigned to any of these five categories. The point of this section was to show that the proactive and reactive coping style are two extremes on the same continuum, and that most people are somewhere between these two extremes.

Conceptual model

For clarification of the goals and concepts, of this study, a conceptual model has been created (fig 1). In this model the relationship between certain concepts has been drawn to give an image of how the motivating effect of incentives and the employees personality relate according to the hypothesis. Three relationships that will be examined are identified with a letter, the latter relationships have been drawn to give a more complete and clarified image of the conceptual model. The three relationships are:

A: The relationship between an employees personality and his motivation. B: The relationship between incentives and employee motivation

C: The relationship between incentives and personality

(11)

(12)

Personality

Introduction to personality

To shortly recap, the purpose of this study is to determine if and how an individuals personality influences the effectiveness of financial incentives. In order to answer this question it is needed that personality is defined and understood. In this chapter the definition of personality, in this thesis, is discussed, followed by a literature overview of this definition. The definition used is that of the coping strategies. Subjects discussed will be the definition of personality (coping styles), the differences between personality types, how this is empirically tested, when people are able to show personality related behavior (when are we able to cope with our situation) and finally how we can examine personality in humans. This chapter will answer the subquestion “what is personality?”. Personality determines to a large extend the way in which an individual perceives and interacts with the environment. Thurstone (1987), defines personality as a collection of behavioral strategies that a person uses throughout life, to control his interactions with the environment. Interactions with the environment can affect an animals health. The physical characteristics of an aversive stimulus do not cause this but it is the cognitive appraisal of that stimulus, this determines the aversive character of the stimulus and can induce stress (Koolhaas et al., 1999). The ability to cope with these

situations determines the aversive impact of stimuli (Ursin, 1998).

Defining the coping styles

In the previous section personality was defined as a collection of behavioral strategies, while this definition is correct it is not practical to use this for the purposes of this study. The definition is vague and does not provide typologies of the strategies, making analyses of personality types impossible. In order to do this the coping styles are used. Koolhaas et al. (1999), defines coping styles as a ''coherent set of behavioral and physiological stress responses which is consistent over time and which is characteristic to a certain group of individuals''. According to Koolhaas et al.(1999), coping styles have been shaped by evolution. Ursin (1980), shows that coping styles are also prevalent in humans. At the end of this chapter human personality and coping will be discussed in more detail. These studies do need to be examined carefully, as the many different definitions of coping and personality can cause problems (Koolhaas et al., 1999).

The definition, of coping styles, given by Koolhaas et al. (1999) is very similar to Thurstone's (1987) definition of personality however, this theory provides better typologies of personality making analyses more practical. Currently two coping styles have been identified, the pro-active and reactive coping style (Henry and Stephens, 1977; Koolhaas et al., 1999). These will be defined in the next section.

Defining proactive coping

Proactive coping is an active coping style, the animal is characterized by showing higher degrees of territorial control and aggression. This coping style is based upon the work of Cannon's (1915) fight-flight response. Proactive animals show a proactive behavioral response to environmental challenges. Animals that are of this this type show more territorial control and aggression (Koolhaas et al, 1999).

Defining reactive coping

Reactive coping is a passive stress response. Engel and Schmale (1972) described it as a

(13)

Different medical, animal and psychological studies have shown that individuals differ in their coping ability. Factors that influence this are genotype, development and previous and early experience (Koolhaas et al. 1999). Because of this it can be hypothesized that a reward structure employed by a manager may be able to alter the personality of a person, this is also shown in the model (FIG 1) as relationship C. This will need further researching.

Proof for behavioral characteristics

In this section a short sidestep is made to explain the various behavioral characteristic of both coping styles, in more detail, giving a better and more complete understanding of both coping styles. Firstly the importance of aggression in coping is examined, followed by the differences in problem solving and how the environment and coping relate.

Most of the following findings are based on animal research. However, Ursin (1980) showed that coping strategies are also present in humans therefore, many if not all of the characteristics apply to humans, more on this in the final part of the chapter (coping in humans). It is important to

understand these behaviors as they are an integral part of an animals survival strategy, therefore they have a strong biological basis and cannot be changed easily (Koolhaas et al., 1999). These behaviors could have significant effects in a business setting.

The importance of differences in aggression

The first important behavioral characteristic that divers between both coping styles is their aggressiveness, proactive individuals are more aggressive as reactive individuals. Koolhaas et al. (1999) have based their coping styles around the difference in aggressiveness. In other words, to identify a proactive individual aggressiveness is the most important factor to study (Koolhaas et al., 1999). To measure aggressiveness usually the latency before attacking is measured. Oortmerssen and Bakker (1981), show that attack latency in mice, this is the time an animal waits before

attacking, can be divided in two groups: those with a short and long attack latency. Also they prove that male aggressiveness is heritable. This study provides evidence that animals differ in

aggressiveness and coping. Koolhaas et al. (1999) shows that animals with a short attack latency are proactive whilst those with a long latency are reactive. The different characteristics associated with high and low aggression are now discussed.

The behavioral differences in problem solving: fight-flight vs conservation-withdrawal

In this section the influence of coping style on problem solving is examined. This is important to solve the research question as there may be performance differences related to personality type instead of the motivating effect of rewards, this could be due to differences in behavior. This section is structured around differences in aversive shock avoidance and the differences in the way changes in the environment are handled.

The influence of coping style on shock avoidance

(14)

both strategies are equally effective in avoiding an electrical shock. To conclude it seems that proactive mice are more actively involved in the avoidance of negative stimuli whilst reactive mice are more passive in their response to negative stimuli. These results also hold for humans (Ursin, 1998). In a business environment this could mean that proactive individuals will be more actively engaged in problem solving while reactive individuals will avoid problems, depending on the setting this could be preferable or not. Whilst this does not seem to have relation with incentives it could have a modulating effect on the performance.

The influence of environmental changes and coping

This section will cover how environmental changes influence individuals differently, depending on their coping style. First the differences between proactive and reactive personality types relating to environmental changes are discussed followed by a brief overview and possible management implications. These results will give an insight into the differences in coping flexibility of both coping styles, in a dynamic business environment, individuals with higher flexibility will probably perform better, this needs to be taken into account to gauge the effect of incentives is due to motivation or perhaps to the selection of staff better fitting with the tasks given.

Benus et al. (1990) show in their experiments that, aggressive mice are more prone to develop routines, while the non-aggressive mice do not. The aggressive mice form routines quickly and do not respond to minor changes in the environment. When their routine is broken by a major

environmental change they have more trouble to adapt in comparison to the non-aggressive mice, which are much more flexible and aware of their environment. Also the reactive mice show conditioned immobility (Benus et al. 1987). However, the proactive mice can perform tasks faster after the formation of a routine, something that is easier for aggressive (proactive) mice. Managerial implications can easily be hypothesized with this characteristic, some employees can handle

variability with more ease than others. However, in a static environment they will probably be outperformed by employees with proactive coping style. It is interesting to see how rewards will relate to this.

Summarizing the coping styles and their possible modulating effects on performance

As mentioned before proactive individuals are more aggressive, create routines and defend their territories more fiercely. Also they are more prone to actively deal with threats and, in case violence will not work, flee. Reactive animals excel in variable environments, they are flexible, they can show conditioned immobility and are more aware of their surroundings. Also they show low levels of aggression.

These two coping styles are extremes, but they may give a lot of insights into the functioning of a staff member and how to make him more effective. It could be that employee A performs better than employee B, not due to cognitive ability or skill, but just because A functions better in a variable environment whilst B excels in routine tasks. Assigning employees to tasks that fit their personality, if factors like skill, education and experience are ceteris paribus, would probably increase

production without assigning additional rewards.

In a management control context this could explain the variance in the performance of employees receiving incentives, but the effects of stress and personality should also be taken into

consideration. Inability to cope due to excessive stress could also be caused by the control mechanism this will be explained in the following section. The current data presented does not allow to draw good conclusions about the effect of personality and incentives on performance or motivation. This concludes the sidestep to explain the behaviors related to coping styles in more detail.

What determines the effectiveness of coping

(15)

experience and predictability (Weinberg and Levine, 1980). Coping is a stress response and requires certain factors to be effective (Koolhaas et al., 1999). Without successful coping individuals will be exposed to high amounts of stress, these will be detrimental to their health in the long term

(Maslach and Golding, 1998). Also the lack of successful coping will not only hurt employees it will also make them harder to control as they will be less predictable when using the results of this study. This section is important as it provides information about how a management control system allows employees to cope more efficiently, it also provides this study with key factors that need to be met in order for the phenomena described in this study to hold. This section is structured as follows: first the factors that aid coping are introduced, followed by a more in depth explanation and concluding with managerial implications of these findings.

Weinberg and Levine (1980) mention in their study that psychological factors can increase and decrease arousal by altering the response to aversive stimuli. These factors are: predictability of stressors; control or the ability to make coping responses during stress; the amount of feedback information an animal receives after an aversive stimulus; and finally prior experience with these kind of stimuli. Koolhaas et al. (1999) mentions a few of these factors, control and predictability, as key determinants for successful coping. However, Weinberg and Levine show in their study that the effectiveness of predictability, in reducing stress, is dependent on control and feedback information, this will now be explained.

In their study, Weinberg and Levine (1980) review multiple studies, which test the results of predictability on stress, the conclusion of this review was that predictability is a complex and confusing factor. Predictability was thought to mainly decrease the stress related to an aversive stimulus. However, the research results were contradicting. Some studies found that a predictable aversive shock was more aversive for an animal than an unexpected shock, whilst other studies show exactly the opposite.

Weinberg and Levine (1980) then proceeded to find an explanation for this phenomena. They found in their review that many studies do not solely test for predictability but have hidden variables that influence the results. These variables were control and feedback. Both of these variables will be defined now and their relation to predictability will be explained.

Control in these experiments means that animals have the ability to alter their environment (bury a

prod for example) or flee from an aversive stimulus or avoid this. In summary an animal needs to have the ability to actively respond to a stimuli. If this is the case a predictable shock will be less aversive to the animal as he can do something to alleviate the shock. If an animal has no control over the aversive stimuli then a predictable shock (or different aversive stimulus) is more aversive in terms of physiological response (Weinberg and Levine, 1980).

Feedback in these experiments means that animals get information, from stimuli, about the

effectiveness of their response and/or that the aversive stimuli is over and that they are safe (for a certain period of time). Lack of feedback can increase the effects of stress responses (Weinberg and Levine, 1980). Feedback relates to predictability, if a shock is inescapable then predictability can only reduce stress if the animal receives feedback. This can be explained with the ''safety signal'' hypothesis. According to this hypothesis if a stimulus is a reliable stimuli that predicts a shock, the absence of this stimulus will indicate safety. In this scenario the animal knows that it is safe as long as the stimulus does not appear. If this is not the case the animal will stay in a state of chronic fear for the remainder of the experiment (Seligman and Meyer, 1970).

In summary, control, feedback and prior experience will all enhance the coping ability of an animal, both control and feedback modulate with predictability of stimuli. The next section will use the information presented here and give an idea how management can benefit from these results.

Managerial implications related to effective coping

(16)

sense.

Important managerial implications can be drawn from the previous section. Management control literature suggest that in order for incentives to be effective, in motivating employees and possibly aligning the goals of the organization with those of the employee, the measures of performance must be in the control of the employee (Merchant and van der Stede, 2007). For an animal to cope with a situation it helps if it has control over the situation. Having control over the performance measure not only enhances the effectiveness of the incentive, in controlling the employee, it also provides an employee with a way to reduce stress. This is important as prolonged stress is

unhealthy, for example it can induce burnouts in employees (Maslach and Golding, 1998). During a burnout an employee will be less effective until he or she mentally breaks down and will be unable to work. Recovery is almost always lengthy and in most cases paid for by the company itself (this due to contract nature and laws in most countries), and therefore it is economically better to avoid this (Sheline 2000). While this may not seem relevant to the relationship between personality and incentives, it does demonstrate the additional purposes of having a control system that allows employees to cope. Coping is the expression of personality in times of stress. However if a management control system is designed without taking predictability, control and feedback into account a person may not be able to cope. Therefore rendering their coping strategy, to reduce stress, useless and making the motivating effect of incentives even more unpredictable as they currently are. In summary, a person must be able to cope with the situation to make sure that his coping strategy is in effect. If this is not the case the person is unable to cope and, with the literature presented here, unpredictable. In other words, the persons personality will have little effect on the effectiveness of incentives. Moreover, if coping is inhibited for prolonged periods of time, health issues may arise which will induce great costs upon the organization.

Coping in humans

Up until now the coping styles have been discussed, how they could relate to business and also what factors are needed in order for an animal to cope. However, most of the results presented were executed in animal research, therefore this section will prove that coping does occur in humans. Also the reasoning to use coping styles instead of different personality scale has not been properly introduced yet, this will also be done in this section.

As discussed before humans also employ coping strategies (Ursin, 1980), how they cope is dependent on their personality. Traditionally personality is defined using the the big five model. This model defines personality based on five traits, these are neuroticism, extraversion,

conscientiousness, agreeableness and openness to experience, people express these traits in a different degree (Digman, 1989). These traits are easy to measure and also very consistent through the aging process of a person (McCrae and Costa, 1987). However, there are issues with using this method to determining personality. Most of these personality tests are based on self evaluations done by the subjects themselves and can be biased because of this (Boersma, 2011). Because of this, the choice has been made to use a more objective characterization of personality.

The theory used, in this thesis, to describe human personality, is based upon physiological responses to alterations in the environment (Boersma, 2011). Boersma, (2011) uses a different characterization of personality styles, she refers to the type A and type B personality styles. These have both been proven to be present in humans, type A and type B are both extremes of one another (Boersma, 2011). The theory describes two personality types namely, type A and type B. Type A individuals are more aggressive, impatient and competitive. Also they have a high catecholamine responses to environmental stressors, catecholamines are hormones that prepare an animal physically for a fight-flight response (Vickers et al., 1981). Type B personalities display much more passive behavioral strategies when exposed to stressors. They have a higher release of cortisol ( Lundberg and

(17)

The advantage that these findings give is access to a larger body of scientific research, and give additional support for the coping styles. Papers that discuss type A and B personalities can now be examined to answer the research question. Moreover these theories can be used in practice to give an unbiased reading of personality that will make management control easier. This, because more unbiased result are more reliable, in the case that management control is adapted to personality type a manager would need accurate measures of personality to design a proper control system that would fit his employees.

Conclusion personality

In order to answer the research question how does an individuals personality influence the

effectiveness of financial incentives? It was needed to build an understanding of personality. In this study personality is defined as the coping style a person uses. A coping style is ''a coherent set of behavioral and physiological stress responses which is consistent over time and which is

characteristic to a certain group of individuals''.

The coping styles can be divided in to two extremes on the one side are the proactive individuals and on the other side are the reactive individuals. Proactive individuals are more aggressive and develop routines more easily compared to reactive individuals, they also are less flexible and have more trouble adapting to a changing environment. Proactive individuals show a flight-fight response.

Reactive individuals are less aggressive and more flexible than proactive individuals. They excel in changing environments compared to proactive individuals. Reactive individuals show a

conservation-withdrawal response.

It was shown that the task and type of personality can have a modulating effect on performance without the influence of incentives. Tasks that require a lot of employee flexibility are easier to perform for reactive individuals, for example. This would enhance their performance but this would not be due to incentives. So in order to determine if the motivating effect of incentives differs between proactive and reactive individuals this needs to be taken into consideration.

A person needs to be able to cope, it was shown how not being able to cope is detrimental to

(18)

Incentives

Introduction

In order to determine if and how incentives influence people with proactive personalities differently from people with reactive personalities it is needed to understand what incentives are, how they are assumed to function, the critiques that are present relating to the current theories, when incentives are traditionally used, and finally if this use of incentives is appropriate based on current

knowledge.

The need to know how incentives function is needed to determine if the results from this study are different or suggest other mechanics at work, also without this knowledge it is impossible to see which parts of the current theory are supported by this study and which are not supported and which are incorrect, also these results can be used to support or disprove the critiques on the different incentive theories. It is needed to know when incentives are used, and if this is correct, to determine if the research used to analyze the research question is applicable to a traditional business

environment or that incentives could be employed differently for greater effects. This chapter will answer the subquestions: What are incentives? and When do we use incentives?.

The chapter will start with defining incentives and explaining in what type of control situations incentives are usable. Secondly, the functioning of incentives and the critiques related to this will be examined. And finally the use of incentives will be examined.

Defining incentives

In this section incentives will be defined, also their current role in management control will be explained. While the latter does not directly contribute to answering the research question it does build a better theoretical foundation and understanding of the current use of incentives, providing information that can be used to make suggestions for more effective use of incentives based upon the results of this thesis.

Incentives are means used by managers to change the direction of effort, duration and intensity of labour in the direction the manager aims for. The direction of effort, by the employee, relates to the difference in goals of the employee and manager. Incentives can change the goals of the employee to better match those of the manager (Bonner and Sprinkle, 2002).

Financial incentives are typically employed in results and action controls. They are used to reward employees. Incentives are almost always prevalent in results control. They are less prevalent in action controls. In the next section both forms of control will be discussed and explained, this will provide a better understanding of management control and the reason incentives are used in these control mechanisms (Merchant and van der Stede, 2007).

Results control

Results controls are controls that reward employees for achieving certain goals. The goals need to be well defined, employees need to be able to influence the desired areas being controlled, and the organization must be able to measure the result effectively (Merchant and Van der Stede, 2007). In the model provided by Ouchi (1969) results controls are used if the ability to measure outputs is high and the knowledge about the transformation process is limited or present.

Using results controls can also give rise to behavioral displacement. This happens when the result measures are not in line with the company goals. Or because of over-quantification, this happens when the results measured are only the concrete and quantifiable results. This causes the hard to measure (in most cases the intangible) results to be overlooked (Merchant en van der Stede, 2007).

(19)

Action controls are controls that directly influence the employees behavior. There are four basic forms namely, behavioral constraints, preaction reviews, action accountability and redundancy. In this proposal only the action accountability will be discussed as this is the only form that can be directly linked to a reward or punishment (Merchant en van der Stede, 2007) .

Action accountability holds employees accountable for the actions they take, they will be rewarded or punished accordingly to the desirability of the actions. Incentives can be used in this form of control to motivate employees to perform certain actions, this relates to the action accountability. Action controls solve problems related to a lack of direction, motivation and personal limitations. The success of action controls depends on two variables, organizations must be able to determine what desirable actions are, and organizations are able to ensure that undesirable actions do not occur. (Merchant en van der Stede, 2007).

Behavioural displacement is also associated with action controls. It is known as the mean-ends inversion, meaning that actions receive more attention and focus instead of the end goals. This causes employees to stop thinking about how to improve performance and their jobs but focus mainly on executing their tasks. As this bears the lowest risk (Merchant en van der Stede, 2007).

Functioning of incentives

In this section the different theories that describe the functioning of incentives will be explained and commented upon. The need to know how incentives function is needed to determine if the results from this study are different or suggest other mechanics that influence the functioning of incentives, also without this knowledge it is impossible to reason which parts of the current theory are

supported by this study and which are not supported and which are simply not tested, also these results can be used to support or disprove the critiques on the different incentive theories. There are several theories that explain the functioning of incentives. These are, expectancy theory, goal-setting theory, agency theory and lastly social-cognitive theory (Bonner and Sprinkle, 2002). These theories will now be explained followed by critiques.

Expectancy theory is based around the assumption that employees act to maximize their expected

satisfaction of the received outcomes. According to expectancy theory the motivation of an employee is depend on two factors, namely (1) the relationship between a particular outcome and the effort it takes to reach this outcome and (2) the attractiveness towards the outcome (Vroom, 1964).

Money is very attractive to most employees, as one can buy goods with it and it is perceived as status. (Bonner and Sprinkle, 2002) According to this theory, because payment is linked to effort, this should stimulate employees to perform better (Bonner and Sprinkle, 2002).

Agency theory builds on the assumption that employees are fully rational and maximize expected

utility. Moreover employees have well defined preferences, are strictly risk-averse and are motivated solely by self-interest. Self-interest revolves around the maximization of leisure and wealth of the individual (Eisenhardt, 1989).

Expectancy theory and agency theory are very similar in their assumption that incentives motivate as they increase the wealth of the employee, which is a driving source of motivation. Agency theory is different in that it accounts for the risk averse nature of economic agents. All shared risk, that is not under direct influence of the employee, should be compensated in the form of a risk premium, otherwise the employee carries the risk of the company, this would be unfair as employees do not have the ability to influence the results, but do get punished for poor results, which are beyond their control, while they do not get compensated for this additional risk. This is unfair as they are hired to do a certain task and should only be rewarded based on this task (Holmstrom, 1989).

(20)

the employee, are the primary sources of motivation to perform better under incentive induced effort. The more concrete and challenging the employees goals, the harder he will work to achieve them. In this theory concrete and assigned goals are much more effective than setting expectancies of the employee (Locke and Latham, 1990).

It is not fully understood how, in the parameters of this theory, incentives influence effort (Bonner and Sprinkle, 2002). Locke et al. (1981), hypothesize that monetary rewards may stimulate employees to set goals. Employees may also set more difficult goals for themselves and the goal commitment may be higher due to pay for effort.

Social-cognitive theory is a continuation of goal-setting theory and expectancy theory. It introduces

self regulating cognitive processes that relate to effort. The individuals belief that he can achieve goals by executing certain actions is essential according to this theory to the motivating effect of incentives, this is called self-efficacy. Self-efficacy regulates employees efforts and can stimulate the development and growth of effort direction, effort duration, effort intensity and strategy

development, by giving the employees the confidence that their effort will pay off (Bandura, 1991). Incentives attribute to better performance in a complex manner, as self-efficacy influences so many factors. An increase in incentives results in higher task interest and this results in higher effort. An increase in effort should increase the performance of the employee. An increased performance usually leads to the development of task expertise and increased self-efficacy (Bandura, 1997). In essence these four theories revolve around increasing the effort an employee engages in and this will be complemented by an increase in performance. The problem with these theories is that they are hard to check in practice and the results of incentives are found to vary substantially. (Bonner et al. 2000).

Critiques on the incentive theories

In this section the afore mentioned theories will be subjected to critiques. This can show why these prior theories do not sufficiently explain the workings of incentives. Agency theory is commonly used to explain the functioning of incentives. It probably is the best known theory of the four described. Moreover, it is based om empirical evidence while some of the other theories are hypothetical. Agency theories has had its share of critique, most of them relate to the (bounded) rationality and self interest assumption.

Eisenhardt (1989), assumes that economic agents have bounded rationality, however in other articles it is assumed that agents are perfectly rational (Baiman, 1990). Bounded rationality means that agents choose for suboptimal decision options even though they fully understand the situation, and how they could achieve more profit (Webster, 2012). Bonner (2001), also shows that in many cases this is true, information is not used to its fullest potential.

Webster (2012), shows in his research how individuals will not choose the most optimal solution in a game. The game was an ultimatum game, players have to give a part of their wealth to the other player, the other player can accept or decline, in case of the latter both players do not get anything. Because of risk uncertainty but also social-norms and their believes in what is fair players choose suboptimal options. Zak et al. (2007), shows that administering oxytocin increases a persons willingness to be generous. This willingness to be fair inhibited the persons rationality making him choose a suboptimal option. The former results suggest that a persons focus on self interest can be inhibited or lowered by this. Research has shown that oxytocin level differentiation greatly correlates with three character traits, these are aggressiveness, detachment and anxiety (Uvnäs-Moberg et al., 1993). It shows that aggression correlates with oxytocin levels and thus that personality may give an understanding into the decision making process of an individual.

(21)

theories are also based on the self interest assumption, as showed before people do not seem to act solely out of self interest. Therefore the explanatory power of the other theories is also

compromised as one of their core assumptions is not true in all cases.

Agency theory and the other theories give an idea of how incentives might work, but they ignore many of the individual differences that could attribute to the effectiveness of incentives. In relation to the management problem this means that the current theories do not posses the explanatory power to correctly predict the effectiveness of incentives on individuals.

When to use Incentives

In this section the situations in which incentives should be implemented according to literature is discussed along with critiques on this literature. This is important for the research question as it is needed to know when incentives are used, and if this is correct, to determine if the research used to analyze the research question is applicable to a traditional business environment or that incentives could be employed differently for greater effects. The paper by Ouchi (1969) is probably the most popular article regarding the design/choice of a control system. Ouchi (1969), shows in his paper a model that guides managers in the choice of control system for a business. The model is based on two variables, (1) the knowledge of the transformation process and (2) the ability to measure outputs. Depending on these, the control system is chosen. As discussed before the use of incentives is only appropriate for results and sometimes action controls. Results controls can be found in output measurement cells while action controls can be found in behavior measurement cells (See table 1).

Table 1

Conditions determining the measurement of behavior and of output

Knowledge of the transformation process Perfect Imperfect Ability to

Measure Outputs High Behavior or Output Measurement (Action and results control)

Output Measurement (Results control) Low Behavior Measurement

(Action control) Rituals and Ceremony Control Looking at the previous sections of this chapter and the introduction, Ouchi's model seems overly simplistic. Going with the findings of Jansen et al. (2009) it seems that even in similar businesses, depending on currently unknown variables (culture is suggested but not proven to be the cause) the effectiveness of incentives can vary. Ouchi's simplistic model does not work, it does not provide the manager with adequate framework to find the most effective form of control for a business.

Bonner et al. (2000) executed a literature study on the effectiveness of financial incentives. As mentioned in the introduction, of this study, the effectiveness of incentives depends on many factors. Bonner et al. (2000), showed with their extensive literature study that not all incentive-based designs influence performance in a positive way. They found that in only half of the experiments, that they reviewed, that the participants improved their performance. Moreover, the cognitive complexity of a task and the expertise of the participant seemed to determine in a large sense if incentives would increase performance. If tasks are cognitively complex and the expertise of the participant is low, then incentives will most likely not increase the performance of the participant.

(22)

subjected to incentives. (Bonner et al. 2000) This is just as goal-setting theory suggest.(Locke and Latham, 1990)

It can be concluded that the effective use of incentives is depended on many factors. Not only is the type of task important also the cognitive abilities and expertise of the employee is relevant as is the ability for the manager to measure performance.

Conclusions regarding incentives

To conclude there are four theories to explain the working of incentives in the management control literature. These theories revolve around increasing the effort an employee is willing to engage in and this will be complemented by an increase in performance. Some of these theories are difficult to test or observe. Moreover the core assumptions of agency theory do not seem to hold all the time in practice. People are not fully rational and not completely driven by self interest. The research by Zak et al. (2007) shows that oxcytocin is a hormone that has influence on the willingness, of people, to be generous. People's behavior is controlled in a certain extend by their hormonal levels these hormonal levels are also found to correlate with character traits. As written in the introduction people differ hormonally (Hassmen, 1993; Ryden et al., 2001). All in all this means that people differ in their behavior due to their personality traits, this is in part the result of differences in hormonal levels, but also other processes like the reward system, the reward system is a brain region that helps a person make decisions (Simon et al., 2010).

Getting back to management control, it is clear now that the current theories about incentives and their functioning are not complete and correct, and thus incentives are not fully understood. This has management control implications, as it is unlikely to design the most effective control

mechanism if the mechanisms are not fully understood. It can be concluded that the effective use of incentives is depended on many factors, personality is one of these factors. It is still unknown how proactive and reactive individuals behave differently as a response to incentives, however

(23)

The reward system

Introduction

In order to determine how incentives change behavior and how this relates to personality, the way in which people, with different personalities, respond to incentives needs to be examined. Elliot et al. (2000) shows that humans treat money as a reward, as the incentives discussed in this study are financial it makes sense to focus on this literature to bridge the gap between personality and

incentives. Elliot et al. (2000) showed in a test that money activates many of the same brain regions as primary reward sources like food, sex and water. These brain regions are part of the reward system, this is a system that guides an animal in making decisions and motivates it to take action or not. As these are the only brain regions to specifically respond to incentives, undiscriminated for personality, studying these brain regions can provide insights into the differences, or the absence of differences, between proactive and reactive individuals, as shown by Simon et al. (2010) who shows that certain brain regions respond more intensely to rewards if the person posses certain personality traits.

Adapting is important for animals, to adapt means to survive. In order to adapt an animal needs to change its behavior, basic survival needs such as food, water and sex drive this adaptation process. Adaption is guided by a very specific system in the brain, the reward system. The physical

characteristics of the rewards system will be explained later on in this chapter, this in order to give a better understanding without over complicating the material (Elliot et al. 2000). This chapter will start with explaining how the reward system functions and why it exists, followed by a more in depth look into the most important brain regions, related to financial incentives, their location and role in the processing of rewards, followed by an overview of how incentives are processed. This will give an understanding of how people respond to financial rewards in a biological way and gives a link between incentives and personality, as personality influences behavior just as the reward system it can be hypothesized that people with different personalities have different functioning rewards systems, this will be examined in the analysis.

The function and functioning of the reward system

In this section the function of the reward system will be explained. The reward system is a system that has a very specific role in the decision process, it allows animals to compare stimuli from different environmental sources by giving them a value and basing subsequent behavior on this. In other words, the reward system compares alternatives and generates a motivating signal. Choices get more complex when there is comparing between different sources or if there is a delay in the timing of the received reward. Complexity also arises due to uncertainty of the reward (J. Peters, C. Büchel, 2010) The reward system also reinforces previous executed behavior, if the results were positive. This is done through dopaminergic systems, which have been shown to respond to reinforcing stimuli like food and drugs.

The current theory about reward systems is that they attribute different values to environmental stimuli. These values are generated within the brain regions to help compare alternative choices and help decide in the behavior the animal will engage in. There are different types of value, each type is a step in the decision process (Peters and Büchel, 2010). The different types of value are outcome value, goal value, decision value and action value. These will now be explained.

Outcome value refers to the value of a reinforcer during the consumption of this reinforcer, for

example eating food. The costs of attaining the reward are not taken into consideration (Peters and Büchel, 2010)

Goal value is, like outcome value, value excluding the costs of attaining the reward. However, goal

(24)

comparisons between fundamentally different stimuli, for example comparing apples to pears (Peters and Büchel, 2010).

Decision value, is the net worth of a specific decision, it takes the costs of attaining the reward into

consideration. This causes some implications, for example the value of an immediate reward is higher than that of a delayed reward and the value of a probabilistic reward is lower than that of a certain reward, granted that the rewards are all the same in this example (Peters and Büchel, 2010). Before making a choice people attribute values to a range of options, this is called action value. Based on the complexity of the decision, action value can be based on outcome, goal or decision value. The option with the highest action value is the one that is executed at that time (Peters and Büchel, 2010).

The reward system compares the values of different rewards. This is important to understand in a management accounting context. An employee has many choices confronting him during the day, he could work hard and for example become tired or make his money with the least amount of effort possible. For an incentive to be effective it needs to motivate the employee to put in the work to achieve this reward. Therefore it is important to understand how reward is processed to determine differences in the response of different personality types.

To better understand the functioning of rewards the individual brain regions that relate to the processing of rewards need to be identified. The ventral striatum is reliably linked to reward-related processes (Elliot et al, 2000) but also midbrain regions of the ventral tegmental area and substantia nigra (Elliot et al, 2000), the amygdala (Elliot et al, 2000) and regions of the forebrain (Elliot et al, 2000). All these structures have also been proven to play a role in the case of humans (Elliot et al., 2000). Elliot et al. (2000) have shown that humans process abstract rewards like money by

activating many of these brain systems. According to Kable and Gilmicher (2009) the ventromedial sectors and the prefrontal cortex are involved in the determination of value. Choice has been attributed to the lateral prefrontal and parietal areas in the sensory-motor hierarchy. Many brain regions have been identified now, however the regions that are particularly useful for answering the research question are those related to the motivating effects of incentives and the processing of monetary rewards, these regions are the ventral striatum and the orbitofrontal cortex (Elliot et al., 2000). These will both be discussed now, starting with a overview of their location and then discussing their role in more detail.

The position of the ventral striatum

In this section the position of the ventral striatum in the brain is discussed (see fig 2). This includes it´s position but also all the brain regions it is connected to, while this does not directly contribute to the research question it gives a better understanding of the ventral striatum. The function of the ventral striatum is discussed in a later section. This choice was made to simplify the information and make it more accessible to those without a neurological background whilst maintaining information that could be needed for more complex references.

Elliot, et al. (2000) show that the ventral striatum plays a role in the processing of monetary rewards, which are mostly used for incentives. Therefore to answer the research question it is needed to get an understanding of the ventral striatum.

The ventral striatum is a part of the Basal ganglia, these are found to be associated with directing intended movement. The Basal ganglia is ussually divided in an anterior limbic section, an associative loop and the posterior motorloop, all of these have different topografical connectivity patterns with the neurocortex (Peters and Büchel, 2010).

(25)

originating from the thalamus. The ventral striatum is also able to send information to the midbrain and directly influence the processing of information in the limbic areas. The ventral striatum is connected to many brain regions and can alter the processing of information in these areas (Peters and Büchel, 2010).

Fig 2

The position of the orbitofrontal cortex

As in the previous section the position of the orbitofrontal cortex (see fig 3) is discussed before explaining its function and role in the reward system. Elliot et al, (2000) show that during tasks that involve monetary rewards there is activity in the orbitofrontal cortex. This region has been

associated with mediating emotional influences on decision making (Damasio, 1994), therefore it is important for the research question, as it will influence the action values of options considered. The position of the orbitofrontal cortex will now be discussed.

The orbitofrontal cortex, which is involved in the decision making process, is located in the ventral part of the frontal lobes. It can be separated in a sensory and limbic network. The medial

orbitofrontal cortex is connected with the limbic areas like the hippocampus. The lateral orbitofrontal cortex receives mostly sensory input. Different regions of the lateral orbitofrontal cortex receive visual, auditive, olfactory, taste and somatosensory input and input from different organs. The dorsolateral and ventrolateral prefrontal cortex areas are associated with the processing of information in the orbitofrontal cortex (Peters and Büchel, 2010).

Financial incentives and the role of personality A biopsychological view

Financial incentives and the role of personality

A biopsychological view

Financial incentives and the role of personality

A biopsychological view

Contents

Introduction

Methodology and conceptual model

Personality

Incentives

The reward system