The distorting effect of the reward system on target setting at corporate organizations

(1)

Amsterdam Business School

The distorting effect of the reward system on target setting at

corporate organizations

Master Thesis

Name: Christiaan Meijs Student number: 10833455

Thesis supervisor: dr. E.G. Van de Mortel Date: June 20, 2016

Word count: 13667

MSc Accountancy & Control, specialization Control

(2)

2

Statement of Originality

This document is written by student Christiaan Meijs 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.

(3)

3

Abstract

Purpose:

This study seeks to identify how corporate firms deal with the distorting effect of the reward system on target setting. This thesis describes the management control system design that causes this tension by using current literature and empirical research. In addition, the current literature does not describe how corporate firms deal with this distorting effect. With empirical research this study seeks to identify how corporate firms deal with that.

Methodology:

This research is based on a case study of a large international hotel chain. The results are based on semi-structured interviews with mid-level managers from individual hotels. Besides the interviews, budgets from 2010-2015 from one of the hotels are analyzed to support the findings of the interview.

Findings:

The main findings of this research show that head office is aware of the distorting effect of the reward system on target setting and imposes several controls. Mid-level managers that are included in the budgeting process try to decrease their future targets in the budget to increase their chance of obtaining their bonus. This is referred to as budgetary slack. The information gap between head office and individual hotels allows mid-level managers to do so. This decreases the accuracy of the targets. Head office imposes various controls to reduce slack and increase the accuracy of the targets and budgets. The effect on the accuracy of the controls is depending on the control itself. Some controls imposed by head office even increase the budgetary slack.

Limitations and future research:

This thesis relies on the honesty and openness of the interviewees. Seen the topic of the interviews, managers might be less honest to avoid sanctions from their employer. Secondly, only mid-level managers are interviewed, the result of this thesis builds on their perception. Recommendations for future research are to seek for the perception of head office and board members. Moreover, a quantitative study could identify the degree of negative and positive effects on the accuracy.

(4)

4

Forecasts are widely used by all sorts of companies around the world. They serve many roles such as planning, control, performance measurement, and incentivizing performance. Merchant and Van der Stede (2012) define forecasting as an important practice of establishing targets for future performance. Setting a target is a quantitative expression of the company’s strategy and objectives. This enables the company to communicate the objectives throughout the organization and to effectively allocate resources. Forecasts are also commonly used for performance measurement and incentives: the reward system. Forecasting may include a certain degree of lower-level participation. In extreme cases forecasts are prepared top-down or bottom-up, where in the last polar-case the firm relies most on lower-level specialized knowledge. High participative budgeting is commonly used in corporate firms; those firms often have a tall organizational structure where lower-level employees hold much information. This study examines if there is tension between target setting and the reward system. Not all entities deal with this field of tension, certain factors must be present. Firstly, the company must have higher-level and lower-level managers and a certain degree of information asymmetry. The higher-level managers rely on information from lower-level employees during the target setting process. Secondly, the entity must have a bonus system that relies on targets set under participative budgeting. An example of an industry entailing those factors is the hospitality industry, particularly hotels. Hotels are often part of a larger hotel chain. This creates a situation where corporate office relies on their lower-level managers, such as the financial controller, when setting next year’s targets. Moreover, it is common that those lower-level managers are rewarded for attaining their target.

1.2 Research focus

This thesis focuses on two related topics: the target setting and the reward system.

I. Target setting enables managers to anticipate the future, such as allocation of resources, informing potential investors, and for attracting new loans.

II. The reward system is occupied with setting a bonus target for employees and provides a form of reward if the goal is attained. The goal is to provide managers with an incentive for high performance. Research has shown that firms set their targets at such a point that they are attainable but also

(7)

7

challenging, to prevent strategic behavior by managers (Indjejikian et al., 2014).

Previous research has shown that these two elements can create a conflict of interest under certain circumstances. If managers meet or exceed prior-year targets, research has shown that is not likely that their next-year target will be higher (Indjejikian et al., 2014). Consequently, firms set a lower target than they actually expect to prevent strategic behavior by managers.

1.3 Research question

The question that then arises is how useful these forecasts are for planning purposes. Since the reward system may interfere with the planning purpose of target setting. The example below illustrates this tension.

Company Y sets a sales target for employee X at €100 for the year 2017. The actual performance of employee X is €120. Company Y considers two options for what the target should be for 2017.

I. Target setting for planning purposes requires the most accurate target, which implies that if employee X was able to reach €120 this year Employee X will be able to do so in 2017, under the ceteris paribus assumption. Consequently management sets the sales target at €120 or higher.

II. The design of the reward system may interfere with the target setting. A lower target may be necessary to prevent strategic behavior or budget games. A lower performance will lead to a lower target next year. Management of company Y urges to avoid that employee X ‘stops’ performing when the target is reached. Company Y sets the target at €110.

Conflict: setting the most likely target (€120) ‘punishes’ employee X for the high performance, consequently the reward system creates a conflict of purposes when forecasting. Setting the target to avoid budget games (€110) will lower the usability of the forecast and budget for management and planning purposes. So what target does company Y set and how do they deal with this tension? This study seeks to answer that question.

Current literature addresses the effect of the reward system on forecasting (Indjejikian et al., 2014) (Indjejikian, Metejka, & Schoeltzer, 2014) and also best practices for target setting

(8)

8

(Tanlu, 2008), however no current literature addresses the field of tension between those two purposes. Intuitively it makes sense that companies struggle with this field of tension, however it has never been addressed in a study how companies cope with this challenge. This leads to the following research question and sub-questions:

“How do corporate firms cope with the field of tension between the reward system and reliable target setting for planning purposes when forecasting?

Sub-questions:

- ‘’How does the higher-manager perceive the field of tension?’’ - “How does the lower-manager perceive the field of tension?” - ‘’What is the process of forecasting in the hospitality industry?” - “Which purpose is dominant when setting the target?”

1.4 Contribution

As mentioned before, the current literature fails to research the field of tension between the two purposes. By empirical research this thesis seeks to fill this gap in the literature. This thesis has two main contributions to the literature. Firstly, it identifies the management control design causing this field of tension. As a result, this study adds to the literature a deeper understanding of the conflicting purposes of forecasting through interviews with managers from the hospitality industry. Secondly, this thesis describes how corporate firms cope with this field of tension. Forecasts have been extensively discussed in the literature, but mainly the general weaknesses of forecasting have been outlined (Hansen et al., 2003). This study does not aim to identify more weaknesses of forecasting; rather it investigates how companies deal with some of those weaknesses and what the causes are of those weaknesses. More specifically, it investigates if the reward system has a distorting effect on the target setting and how companies deal with it.

1.5 Paper structure

The remainder of the paper is structured as follows: section 2 reviews the current literature on forecasting and budget games and presents the analytical framework, section 3 describes the research setting and data collection, section 4 presents the findings and conclusions to the propositions, and section 5 concludes this thesis.

(9)

9

2. Literature review

The thesis builds further on existing literature. A wide variety of studies present the benefits and flaws of forecasting and their solutions. The four main topics covered in the literature review are: the process of target setting, management control systems, the reward system, and information asymmetry.

2.1 Difficulties of target setting

Issues that arise with target setting have been discussed extensively in the literature, however they lack in-depth analysis. All organizations use forecasts to predict and manage their future (KPMG, 2007). A study of KPMG in 2007 found that only one out of five companies produces reliable forecasts, and that inaccurate forecasting results in decreasing share value. One of the current flaws presented by KPMG is that incentives are linked to budgeting; therefore the majority of the forecasts are lower than the actual performance. This subsection reviews the current literature to identify causes of poor forecasting. Firstly, unreliable input affects the reliability of forecasts (KPMG, 2007). Managers lack confidence in the financial data used for forecasting. According to the author of that study, this is due to a degree of uncertainty of the data, for example because of volatile markets. This occurs especially in the hotel industry where demand is more difficult to predict compared to production organizations because of the high impact of external factors (e.g. economic and political situation) (Winita, 2005). Another reason why managers perceive the data to be unreliable is because managers do not use all the available sources. Many economic and governmental reports are widely available but not used by companies. Not surprisingly, the source used the most is internal, historic financial data. Secondly, firms often ignore the opportunities arising from technology. Most hotels only make use of spreadsheets, while bespoke forecasting and planning tools are available. These impediments relate to the usefulness of forecast for planning purposes. Thirdly, most products and services in the hotel industry are customized, since they are based on the guests’ needs and wants. Customized products and services are more difficult to quantify in forecasts than standardized products from a manufacturing company.

The obstacles for an effective reward system have also been researched extensively, mostly referred to as budget games or strategic behavior by mid-level managers. Firstly, target ratcheting is the manipulation of future targets by reducing the current performance by accounting manipulation or even real economic decisions. Since most companies, also the hotel industry, rely on historic performance in the process of forecasting, lower prior

(10)

10

performance will lead to a more attainable target for the next period (Bouwens and Kroos, 2008; Indjejikian et. al, 2014-B). Secondly, there are two conflicting factors when deciding on the difficulty of a target. The target should be challenging in order to incentivize high performance. But on the other hand the target should be attainable, if the target is set too high it demotivates the managers because the target is ‘out-of-the-money’ (Bhimani, A., Horngren, C., Datar, S., and Rajan, M., 2012; Hope and Fraser, 2003; Tanlu, 2008).

The impediments described above are crucial to understand the process of target setting, however they describe the difficulties of the forecasting with regards to planning and the reward system. As mentioned before, the two purposes combined form an obstacle as well. As long as companies use the same targets for budgeting and rewarding there will remain a conflict of interest (KPMG, 2007; Chow et al., 1988). This thesis expands the current literature by examining the way the hotel industry deals with this conflict of interest.

2.2 Information asymmetry

Information asymmetry historically has been introduced as the gap of information between shareholders and the board (Bartov, and Bodnar, 1996). Information asymmetry exists because most of the decision-rights are allocated to the board of directors. This is known as decentralization. This type of information asymmetry embraces the question whether the board of directors acts in the same interest as the investors. Nowadays the notion of information asymmetry is also present within the organization itself; between lower and higher level employees.

The agency theory suggests that employers (principals) have different objectives than the employees (agents). Within the scope of this thesis the principal is the head office or director and the employees are the top-level managers of individual hotels and restaurants. Agents act in self-interest that not necessarily meets the organizational goals (Chiang, and Birtch, 2010; Eisenhardt, 1989). Information asymmetry enables opportunistic behavior by the agents, since the principal is unable to exactly monitor what the agents is doing or that monitoring is too expensive. The hospitality industry deals with a large degree of information asymmetry, since head office and individual hotels are physically separated from each other. Often even cross-border. Head office relies on the information provided by the individual hotel. Under this assumption my first proposition is:

Proposition 1: decentralization leads to more information asymmetry than in a centralized organization.

(11)

11

The distortion from the reward system can occur when mid-level financial managers are included in the budgeting process. Including lower-level employees in the budgeting process is referred to as participative budgeting. Heinle, Ross, and Saouma (2014) describe the process of participative budgeting and the key players involved. The current trend towards corporate decentralization requires a more bottom-up approach of forecasting. Due to information asymmetry corporate management can only rely on managerial reports. This on the one hand improves target setting since lower-level employees hold valuable information.

On the other hand this motivates mid-level management to increase their performance. Information asymmetry and thus the reliance on figures form mid-level management leads to struggles. Mid-level management may have an incentive to misreport budget figures to increase their chance of reaching their target. Top-down budgeting eliminates this risk because top-management decides the forecasting figures. However, this does leave crucial information out of sight, especially in highly decentralized firms, such as the hospitality industry, where lower-level employees hold much information.

2.3 Target setting

This section defines target setting, it presents the purpose of target setting, and it identifies the process of target setting. Targets, such as budgets and forecasts, define the level of performance an organization needs to achieve for each of its key performance indicators (Ferreira and Otley, 2009) and form the basis for rewarding performance. Organizations use target setting for a wide variety of purposes. Firstly, it enables the company to communicate and coordinate plans of business functions into a corporate plan that is aligned with the strategy. Secondly, it serves as a planning tool for allocating resources. Thirdly, it authorizes certain employees to make independent decisions for the budgeted amount. Fourthly, budgets serve as an early warning trigger if the company deviates from its objectives. Lastly, targets form the basis for another major function in the management control system: evaluating performance and incentives (Shastri, and Stout, 2008; Bhimani, Horngren, Datar, and Rajan, 2012). Identifying these purposes is crucial to obtain a better understanding of the budget process and the motives of budget preparers, since better budgets and forecasts lead to better performance (KPMG, 2007).

The process of budgeting, forecasting, and planning varies across organizations. Targets can be set based on quantitative models, historic information, based on negotiation between lower- and higher-level managers, or a combination of the three (Merchant, Van der

(12)

12

Stede, 2012). Quantitative models are used when there is a causal relationship between input and output and this relationship is stable and predictable. This is for example often the case with variable costs. Forecasting based on historical information is based on prior performance. The prior performance is used to predict future outcomes. This can be done subjectively or making use of statistical models. Negotiating target setting is the process where lower- and higher-managers mutually set the target for next period, this also referred to as participative budgeting (Heinle, et. al, 2014). Negotiated targets are often present when there is a certain amount of information asymmetry between the different hierarchal layers. Consequently, using quantitative models and historical information do not require lower-management input since that is based on statistics. Therefore, these can be seen as top-down approaches. To conclude, model-based and historical-data forecasting are likely to be used under the assumption of top-down forecasting. Higher-management sets the target for the lower-level management. When the information asymmetry increases, participative or negotiating target setting is more appropriate.

The hotel industry focuses on customer satisfaction, which is largely dependent on the employees’ and managers’ performance. Past studies have shown a positive correlation between participative budgeting and the performance of a company (Winita, 2005). This is in line with the normative situation in the hotel industry, where the target is negotiated between the CFO at head office and the financial controllers at the individual hotels. Model-based and historical data are used to predetermine a certain upper and lower bound for the target. In the area between the upper and lower bound the negotiating between head office and the financial controller at hotel-level occurs. Under the assumption that lower-level employees hold valuable information the second proposition is:

Proposition 2: participative budgeting has a positive effect on the accuracy of target setting under the assumption that information asymmetry is present.

2.4 Reward system

Rewards are present in nearly every for-profit company (Murphy, 2000). Incentives are used to reward performance. This aligns the behavior of the subordinate with the expected behavior set by the superior. Rewards are known in a wide range of forms. These can range from recognition from management, financial bonuses and salary increase, to more implicit

(13)

13

rewards such as promotion (Ferreira and Otley, 2009). This section of the literature review describes the link between budgets and the incentive program.

Exceeding customers’ experience is key in the hospitality industry. Therefore, it is important that the employee’s behavior is aligned with the organization’s norms and values. Incentives and goal setting have a proven positive effect on aligning behavior to the best interest of the company (Bonner, and Sprinkle, 2002). However, recent media coverage also reveals the drawbacks of incentives. When there is too much pressure on making the numbers and a high incentive on performing, employees may start to manipulate the figures (Likierman, 2009). This can be done by ‘cooking the books’, but also by non-ethical behavior towards colleagues or clients. Managers are rewarded for their performance if they meet or exceed a target; this target is set in the budgets (Indjejikian et. al 2014-A). Bonus targets exist because of moral hazard, since it is difficult or sometimes even impossible to monitor the behavior of mid-level managers. According to KPMG (2007) the bonus target is equal to the target used for planning purposes. In addition, Shastri and Stout conducted a survey (2008) in which they report an increasing number of firms relying on participative budgeting. This leads to a problem since ‘easy’ goals lead to weaker performance compared to a situation with more challenging goals (Chow et. al., 1998). Under the assumption that the principal agent theory is applicable, my third proposition is:

Proposition 3: Information asymmetry has a negative effect on the accuracy of target setting via the reward system.

2.5 Management control system

Management control systems (MCS) have been frequently discussed in the literature. The general definition for MCS is the set of tools and measures than ensure that the organization achieves its strategic objectives (Merchant and Van der Stede,2012). The MCS serves as a planning and control tool. The MCS guides organizations and its employees towards the achievement of objectives by enabling agents to allocate resources aligned with the strategic direction. And as a control tool, by aligning the behavior of employees to the best interest of the company. Ferreira and Otley have compiled a framework of factors of a MCS. I use this framework in order to capture and identify all the aspects of a MCS. Ferreira and Otley (2009) introduced twelve questions of aspects and influencing factors of a management control system. It offers a comprehensive understanding of the MCS and provides the link

(14)

14

between components. Generally, the more clear and specific the previous aspect is set, the better the following step (question) can be identified. The Ferreira and Otley (FO) framework is used to describe the role and position of forecasting in the management control system. The model is purely explanatory; hence it does not propose how a management control system should be. This thesis only captures the factors that are occupied with forecasting. The other factors do play an important role in the MCS and although all factors are intertwined they fall beyond the scope of this thesis. More specifically, the relation between reward system and target setting is addressed in this thesis. Therefore, these components are discussed rigorously Concerning key performance indicators (KPI)s and performance evaluation, these components will be briefly discussed in this section since only the outcome of the design is relevant for this thesis.

Original Ferreira and Otley framework (2009) Explanation

Mission and vision (Q1) The way the values and purpose of the company are established and communicated throughout the organization.

Key Success Factors (Q2) Defines the indicators that are crucial for the company’s success.

Organizational Structure (Q3) The organizational structure and the impact on the design and use of the MCS.

Strategies and Plan (Q4) The strategy and the objectives that the company has decided on to be successful.

ü Key Performance Indicators (Q5) The KPIs are derived from the objectives. This also included the way they are communicated towards employees.

ü Target setting (Q6) The level of performance the company requires to be successful.

ü Performance Evaluation (Q7) The process of evaluating individuals, groups and the organization itself.

ü Reward Systems (Q8) The rewards employees receive for obtaining the target. Information flows (Q9) The information flow design

PMSs Use (Q10) The way the organization makes use of the MCS. PMSs Change (Q11) Which factors lead to changes in the MCS

Strength and Coherence (Q12) Describes the strengths of the links between the previous components.

Figure 1: original FO framework and relevant factors (P) for this thesis

These four aspects of the model are linked with forecasting: Key Performance Measures (KPIs) (Q5), Target setting (Q6), Performance Evaluation (Q7), and Reward Systems (Q8). An application of the FO framework of the hospitality setting is used to identify factors within the MCS that impact forecasting. Besides, the interrelationship between the different factors is addressed.

(15)

15 Figure 2: Relevant factors

Source: adapted from Ferreira and Otley, 2009

Key Performance Indicators (KPIs) are the measures used at the organization to evaluate achievement of objectives. KPIs make the company’s strategies and plans measurable. In the hospitality industry KPIs can be both financial and non-financial. Within individual hotels that operate under a hotel chain, financial controllers are responsible for the gross operating income (GOI). The other expenses (e.g. operating expenses, depreciation and interest) are clustered and fall under the responsibility of head office. The GOI can be split in smaller KPIs. On the revenue side, revenue per available room (RevPar) is the combination between occupancy and average room rate. RevPar is calculated by multiplying the occupancy and average room rate. Food and beverage revenue is the second KPI on the revenue side. On the cost side, the KPI cost of sales (COS) is used to measure the percentage of food and beverage costs compared to revenue. A non-financial KPIs is employee turnover percentage, which is the percentage of employees that leave the hotel. The hotel industry is notorious because of its high turnover. The second non-financial KPI is customer satisfaction. These KPIs are communicated top-down; head office emphasizes the hotel management to focus on these indicators. Consequently, the targets are built around these KPIs and determine the bonus target.

Target setting sets the level of performance the organization needs to achieve for each KPI. Appropriate and straightforward KPIs enhance target setting. The process of target setting in the hospitality industry is relatively clear, since it involves little employees. Usually

(16)

16

only the top-management team of a hotel receives a bonus; the mid-level managers and supervisors are often not eligible for a bonus. In the situation that non-managers are eligible to receive a bonus they are not included in the budgeting process. This means that the bonus target is communicated top-down to lower-level employees. As mentioned in the previous section, head office sets an upper and lower bound for the target. The financial controllers from individual hotels set the target for the above-mentioned KPIs. After which head office and the financial controller come to a mutually accepted target by negotiating.

Performance evaluations are used to assess whether the organization met the set target for each KPI. Performance measurement not only includes performance on individual level, but also of groups (e.g. departments) or the organization as a whole. Moreover, distinction is made between subjective and objective performance evaluation. Performance evaluation in the hospitality is mainly objective. Since the chosen KPIs are by their nature quantitative, the performance can be objectively measured. Performance evaluation is mostly performed by head office assessing whether the managers obtained their bonus target or not. Only in rare cases all employees, lower and higher level, are rewarded for their performance. Subjective performance evaluation could be used to filter out flaws in the performance system or to filter out uncontrollable external factors (Gibbs et. al, 2004). Whether subjective performance is used in the hospitality industry has yet to be researched. Also budget games (Bouwens and Kroos, 2008) are discussed in this aspect of the MCS. These topics are addressed in the interviews.

The financial and non-financial rewards that come forth of the performance evaluation belong to the fourth aspect of the PMS. The reward system within the field of hospitality focuses on financial bonuses. This is strengthened by the fact that only the highest managers in a hotel are eligible for a bonus. The control activities that are present to stimulate wanted behavior and that are present to avoid unwanted behavior in the hospitality have not been addressed in the literature. This topic is addressed in the interviews.

Contextual factors and culture explain why certain choices are made in the management control system. Since these contingencies are already included in the above description they do not require additional elaboration (Ferreira and Otley, 2009).

2.6 Theoretical model

The theoretical model below is an adapted version of the Ferreira and Otley (2009) framework. Proposition 1 states that decentralization leads to more information asymmetry compared to centralized organizations. Proposition 2 states that decentralization has a positive

(17)

17

effect on the accuracy of target setting under the assumption that information asymmetry is present. On the other hand, proposition 3 states that decentralization has a negative influence on the target setting via the reward system due to strategic behavior of lower-level employees. Proposition 4 is about the MCS design choices that affect the accuracy of the target. This thesis focuses on how corporate firms deal with this conflict of interest when forecasting. Forecasts should be usable for planning purposes and also incentivize employees for performance. The interviews focus on the direct relationship between target setting and the reward system. It aims to reveal what the effects of the design decisions of the reward system are on target setting.

Figure 3: theoretical model

Adapted from Ferreira and Otley (2009)

PROPOSITIONS

P1: + Decentralization leads to more information asymmetry than in a centralized organization.

P2: + Participative budgeting has a positive effect on the accuracy of target setting under the assumption that information asymmetry is present.

P3: - Information asymmetry has a negative effect on the accuracy of target setting via the reward system.

P4: ? The effect of certain design choices of the reward system on the accuracy of the target is unclear.

(18)

18

3. Methodology

The objective is to explore and create an initial understanding on how firms deal with the distorting effect of the reward system on target setting. A qualitative method is the most logical choice for this research since it allows studying the process in the real world. Therefore this thesis mainly includes qualitative research in the form of interviews, which is supplemented with archival data such as forecasts from the hospitality industry to which I have been granted access.

3.1 Research setting

This study researches the subject in a large hotel chain: AccorHotels. Forecasts are widely used in the hospitality industry; moreover they include a high degree of decentralization and participative budgeting. Besides that I have contacts within this hotel chain, which makes the data accessible. The hotel chain is a multinational company operating is nearly all continents and roughly 100 countries. AccorHotels does not operate under the name Accor, it has multiple brands such as: Sofitel, Novotel and Ibis. All brands have their own target markets and specifications. Accor’s head office is located in Paris. At head office in Paris, the executive board of Accor is positioned. The next organization level is the support office. These support offices are located in different regions. At support office, the regional directors are located. The Netherlands is part of the regional support office Benelux, located at Schiphol. General managers of the individual hotels are directly reporting to support office. Below the general manger is the management team of the individual hotels, they report directly to the general manager. Except the financial specialist, they report to the general manager and the support office directly.

(19)

19 Figure 4: organizational chart AccorHotels The Netherlands

3.2 Interviews

The research is conducted within AccorHotels as a case study. Due to my connections within this hotel chain I strive to have in-depth interviews. Interviews are the most suitable research method because this research mainly focuses on behavioral aspects. The interviews are semi-structured and include open and closed questions. Different management layers in the hotel chain are interviewed, varying from managers at support office, to general managers of the individual hotels and members of the management team at individual hotels. Two of the eight interviews are not held within the hotel chain to increase the transferability of this study. The literature review and adaption of the Ferreira and Otley model are the basis for the interview questions.

3.3 Data collection

The data for this research is derived from conducting 8 interviews within hospitality chains. The 8 interviewees are mid-level managers, the financial controllers or general managers of individual hotels and restaurants chains in the Netherlands. They are responsible for preparing the budgets for their locations. After completion, the budget is sent to the Dutch head office

(20)

20

for approval. The CFO at the Dutch head office makes amendments and approves the budgets. The interviews with mid-level finance managers should reveal what impact the distorting effect of the reward system has on target setting. All the interviewees are mid-level manager, this means that lower-level managers report to them. Mid-level managers, in turn, report to their superior. Figure 5 introduces the interviewees. The interviews are conducted anonymously. Therefore, the interviewees are labeled with a letter instead of their names. The letters can be found in figure 4 as well to see the position of the interviewee within the organization.

# Interviewee Position Segment

Duration (minutes) 1. A Revenue manager Hotel head office 60 2. B Rooms division manager Hotel-level 45

3. C Revenue manager Hotel-level 50

4. D Financial controller Hotel-level 70+30*

5. E General manager Hotel-level 45

6. F HR manager Hotel-level 60

7. G Store-manager Restaurant Mid-level 60 8. H Store-manager Restaurant Mid-level 40

* 70 minutes interview and 30 minutes discussion of documents

Figure 5: interviewees

3.4 Interview questions

The interview questions are based on the theoretical model in figure 3. An exception is the beginning of the interview, which starts with general questions about the function and background of the interviewee. Following the model the next topic is decentralization, where the interviewee is asked about his or her responsibilities in the organization and his or her role with regards to target setting. Then the interview questions focus on the four relevant steps from the Ferreira and Otley model: KPIs, target setting, performance evaluation, and reward system. After which the questions focus on the effect of information asymmetry on the target setting process. To conclude, the interviewee is asked about the tension between setting ambitious goals and the desire to achieve the bonus targets. Some of the questions are

(21)

21

recommended by the Ferreira and Otley model itself and are copied. The list of interview questions can be reviewed in appendix A.

3.5 Documents

Besides the interviews, archival data was analyzed to back-up or confirm findings from the interviews. The profit and loss statements from 2010 until 2015 are used. Focus is on the actual total revenue, total cost and gross operating income compared to previous year and budget. Noteworthy findings or reoccurring findings are discussed with the responsible financial controller of the relevant period.

3.6 Data analysis

The data analysis of the interviews is divided into three processes: data reduction, data display and data interpretation (O’Dwyer, 2004). Before processing the data, the interviews were taped and notes were made during the interviews. The interviews and notes were combined into a transcript in Microsoft Word. The interviews were conducted in Dutch; consequently the transcripts are also in Dutch. After transcription the next step was data reduction. This implies reducing the extensive amount of data in relevant information. The data was then labeled into themes. These themes are: decentralization and information asymmetry, KPIs and target setting, performance evaluation and reward system, and MCS’s design. The second step was data display (Huberman and Miles, 1994). This process incorporates the reduced transcripts into matrices. The third step was data interpretation (O’Dwyer, 2004). This step implies the detailed reviewing of the reduced data and the literature. This last step eventually led to the construction of a narrative.

(22)

22

4. Results

This chapter presents the results of the interviews and the supporting document. The subsections are conforming the propositions and the theoretical model. The first part is about the effect of decentralization on information asymmetry. The second part is about the effect of participative budgeting on the accuracy of the target. The third section is about the reward system. The final part is about the MCS design choices and the effect on the accuracy of the target. The matrices display the findings in an orderly manner. The subject description is the description of the interview question. The questions are numbered and added to the table under ‘question’. For the full questions, please refer to appendix A. The given answers are grouped and presented under features. ‘Amount’ represents how many times that specific answer is given. Please note that sometimes an interviewee provided multiple answers. Therefore, the amount does not necessarily add up to exactly 8. The text under the matrices explains the findings in more detail. Each section concludes with the discussion of the related proposition.

4.1 Decentralization and information asymmetry

This section is about the allocation of decision-right to lower-level employees and its effect with regards to information asymmetry.

4.1.1 Decentralization

Subject description Question Features Amount

Professional background 1 Accountancy/finance Nonfinancial 4 4 Responsibility 2 Cost center

Revenue center Profit center

2 2 4 Role in budgeting 2 Cost side

Revenue side Entire budget

2 2 4 Decision rights 3 Independent decision making

Consulted decision making

6 2 Decentralization 3 Perceives highly decentralized

Perceives moderate decentralized

6 2

Figure 6: matrix decentralization

Professional background

From the interviewees, four have a professional background in finance, bookkeeping or accounting. The other interviewees have a broader management background, which always include a financial perspective but also e.g. human resources and hotel operational experience.

(23)

23 Responsibilities

The interviewees were asked about their responsibilities within the organization. All respondents perceive that their employer expects them to achieve a certain result and not solely performing a task. “I am ultimately responsible for my division, from a financial perspective up to human resources. At the end of the year I am judged on my performance, it is up to me how I will achieve the desired performance level within the bounds of the company rules” (G). This is aligned with the theory, which states that in decentralized organizations managers are evaluated on their results.

The responsibilities of the interviewees are drivers. Two interviewees indicate that they are solely responsible for costs. Their department may generate revenue but they are not involved in the process, this responsibility is referred to as a cost center. One of the interviewees, the rooms division's manager, stated: “I monitor and take actions on the cost side, for example the amount of FTEs, I obtain the revenue figures from the revenue manager, I do not influence that” (B). Two interviewees stated that they are solely responsible for creating and monitoring revenue; they cannot influence costs. The restaurant managers, general managers, and financial controller are responsible for both costs and revenues, the so-called profit centers. “I need to increase my gross operating income, by increasing my revenue and decreasing costs” (D).

Role in budgeting

There is a distinction in the role of the interviewees with regard to target setting. “The revenue manager, financial controller, and general manager lead the target setting and are coordinating the process” (F). There are three different roles. Firstly, two interviewees stated that they deliver revenue targets to the key responsible employees for budgeting. Secondly, two employees stated that they base cost targets on the revenue they received. Thirdly, the key responsible employees for budgeting combine the revenue and costs and prepare the full budget and targets. “Eventually the financial controller is ultimately responsible for the creation of the budget”(C).

Decision rights

Since the interviewees are all higher-level managers they all perceive that they have the right to make decisions. Six interviewees stated that they have almost full authority. I have a lot of room for independent decision-making. I could say that for 90-95% of my decisions I do not need authorization, only for really big investments” (A). The decision rights end at gross

(24)

24

operating income level. This means that the hotel managers can make decisions on pending costs and revenues, but need permission for changes in assets. Two interviewees indicate that they can make some independent decisions, but that due to the short lines of communication between them and their superior they are merely consulted for their advice, rather than making an actual decision.

Decentralization

The interviewees perceive AccorHotels as a highly decentralized organization with many organizational levels. The worldwide head-office is located in Paris, each region has its own support office. Each hotel has its own general manager, and all the departments have their own manager. Each layer reports to their superior management layer. The first reason for decentralization is the size of the company: the board in Paris cannot monitor all individual regions, let alone the individual hotels or divisions. Secondly, the managers at regional and hotel level hold much more information about the environment in which they operate. “It is my specialization, I know the market and specific knowledge about the city in which I operate” (A). That is the main reason for the interviewees why they have decision rights. However, not everything is decentralized. Large investments such as rebranding and the decision to buy or lease the building in which the hotel operates are taken at head office.

4.1.2 Information asymmetry

Effect of decentralization

21 Information asymmetry 8

Reliance 22 Fully rely on my targets Rely but I have to defend

2 6 Tension 23 I do not perceive tension

I perceive tension

I could understand that others perceive tension, but I do not

1 5 2

Figure 7: matrix information asymmetry

Effect of decentralization

All the interviewees perceive that decentralization leads to having more information about their specific business than their superior. “It is impossible for head office to find out what happens on micro level and internally in the organization” (D). Consequently, for all interviewees some degree of information asymmetry is present. The extent of the information asymmetry however differs. Highly specialized managers, such as revenue managers, believe

(25)

25

that there is a big information gap between them en their superior: “head office only is aware of what happens in an entire region or even country, if I sign a new contract with a large corporate client, head office would not have that information” (C). On the other hand less specialized function, especially cost centers, have less information asymmetry as became evident from the interview. For example, the golden standard for food costs in the hospitality sector is 25% of the sales price of food items, that knowledge is also held by head office. The extent of information asymmetry is influenced by another factor: the higher in the organizational chart the greater the information asymmetry between layers becomes. The short communication lines between the general manager and the rooms division manager leads to little information asymmetry: “the general manager of course has an idea what happens in the hotel and the area” (F). While the general manager has indicated that his superior does not have information about internal processes within his hotel, rather head office relies solely on regional figures.

Reliance

The interviewees are like-mined with regards to the reliance of their superior on their knowledge: “very much reliant, only the general manager and controller of the specific hotel have somewhat the same knowledge. Head office has no other option than to rely on my figures” (A). So head office does rely on the information provided by individual hotels. However, they tend to rely on the consolidated targets from all hotels combined. “Head office builds on the information I put into our ERP-system; that system consolidates the information from all hotels” (C). There is a difference between daily information provided to head office about the operating business and periodic information, such as budgets. Daily information is simply taken over by head office, however periodic information about target setting must be defended. The reason why this information must be defended is because the superior manager must be sure that the target is stretched enough. “I always need to defend my targets, to make sure that they are honest and realistic”(B).

Tension

The fact that head office relies on information from lower-level manager sometimes leads to tension. As described in the literature the agency theory states that head office (the principal) and the lower-level managers (agents) have different objectives. Most interviewees perceive tension because they are evaluated on their self-set targets. “It is a political game, horse-trading, between management and head office” (D). This political game is about improving

(26)

26

the performance of the business and setting realistic target on the one hand, and making sure that the targets, and thus the bonus are achievable, on the other hand. Besides tension, setting own targets also stimulates intrinsic motivation. “Setting my own target requires trust from my manager, that is much more valuable than getting targets top-down”(F). Moreover, the interviewees are aware of the importance of accurate target setting. If targets are lowered, for example reporting less expected revenue to head office, that also impacts the costs. If the revenue target is lowered, the variable costs such as labor cost should also be lowered. If the revenue than indeed is much higher than budgeted, the hotel is not authorized to hire more employees and their only option is to outsource or hire temporary employees. This is far more expensive and leads to a lower performance of the hotel.

4.1.3 Conclusion proposition 1

Proposition 1:

Decentralization leads to more information asymmetry than in a centralized organization.

Decentralization is the notion of allocating decision rights to lower-level employees. From the interviews it became evident that all of the interviewees perceive that the decision making process is decentralized from head office to them. The extent of decentralization differs among the interviewees since they have different responsibilities. Three types of responsibilities are found among the interviewees: cost centers, revenue centers and profit centers. Only large investments are made at head office. These responsibilities are also reflected in the target setting process.

The effect of decentralization is that lower-level managers hold much more information than head office about the individual hotels. Consequently, information asymmetry simultaneously occurs with decentralization. It became evident that the agents have more information about the operational business than their superiors. Head office only has an idea of the trends in the region. Therefore head office is forced to rely on the (budget) information from lower-levels.

There are two factors found during the interview that determine the extent of information asymmetry. Firstly, the information gap between agents and principals is higher for specialized functions, such as revenue management. Secondly, information asymmetry increases when the agent is higher in the organizational chart. The short communication lines

(27)

27

between e.g. the general manager and his rooms division’s manager are shorter than the communication line between the general manager and his regional manager.

To conclude, decentralization leads to more information asymmetry. The extent of the information gap in this study is determined by the level of specialism of the agent and the degree of responsibility of the agent. In contrast, less specialist jobs and shorter communication lines lead to less information asymmetry.

4.3 KPIs and target setting

This section describes the key performance indicators that apply to the managers and the level of performance required.

4.3.1 KPIs

KPIs 4, 5 Gross operating income or related measure

Combination of financial and non-financial measures 8 8 Role of KPIs 6 KPI are basis for bonus 8

Figure 8: matrix KPIs

KPIs

A wide range of KPIs is present in the organization. It is a combination of both financial and non-financial KPIs. The interviewees state that they have own specific KPIs and KPIs that reflect group performance. One KPI that applies for all interviewees is gross operating income (GOI). Specialist jobs such as revenue managers are rewarded on occupancy and average room rate compared to benchmark figures. Human resources managers are rewarded on non-financial KPIs such as employee turnover and employee absenteeism. Overall 60% are hotel KPIs such as GOI and guest satisfaction, and 40% are individual targets that are specified for each manager.

Role of KPIs

As shown in figure 8, all interviewees state that the KPIs form the basis for their bonus. They KPIs are not necessarily used for performance evaluation: “there is a difference between performance evaluation and bonus. Performance evaluation is based on competencies that are required for this job. Bonus targets are doing something extra for the company” (C). Every job-function has its own competencies that are set-up by the company. The personal performance evaluation is not linked to the bonus, unless the evaluation is really below standards. “If my personal evaluation is insufficient, than I will not receive a bonus” (G). Vice versa, an excellent performance evaluation does not lead to a higher bonus.

(28)

28 4.3.2 Target setting

Role of target setting 8 Monitoring tool

Planning and resource allocation Informing investors Authorization expenditure 4 3 2 2 Process 9, 10, 11 Bottom-up, top-down

Top-down, bottom-up, top-down

6 2 Final decision 11, 12 Hotel director

Head office

3 5

Figure 9: matrix target setting

Role of target setting

Questions about the function of target setting the interviewees did not provide a unanimous response. This does not mean that the interviewees disagree on the purpose of target setting, but more likely what they find is the most important purpose. “Budgeting is for planning on the long-term, forecasting is for day-to-day use” (E). Another interviewee states: “managers need to account for each large expenditure if it fits in the budget” (F). The financial controller states that the budget serves the purpose of sharing the outlook of next year to investors.

Process

The process of target setting depends on the role of the manager in the process. “We start the budgeting process with all the managers together, even non-finance managers since we see it as a team-effort” (E). The revenue manager provides a revenue prognosis for next year. “My prognosis is based on what I find realistic with regards to trends, market developments, and if we expect changes in customers” (A). After the preliminary revenue target the departmental managers, such as the restaurant manager and rooms division’s manager prepare the cost side of the budget. This cost side includes variable costs and the amount of FTEs required for that level of operation. The HR manager is involved by linking the FTEs to salaries and making sure that the appropriate amount of FTEs is budgeted for. Then the preliminary revenue, variable costs, and labor costs are forwarded to the financial controller. The financial controller checks the input and continues to budget for other variable costs and fixed charges. Eventually this leads to the preliminary budgeted gross operating income.

After the preliminary budget, guidelines are provided by head office in Paris. “Paris hires an external company for an external market scan” (D). Figures for the Benelux are sent to support office at Schiphol. The macro-figures are compared with the figures presented by the individual hotels. The preliminary budgets do not precisely meet the requirements from

(29)

29

head office. Since head office relies on macro-figures they do not have that information about every specific hotel. As described earlier in section 4.1, the managers from the individual hotels are specialists on micro-level developments. As a result, the budget expectations from head office do not necessarily meet the budget expectations from the individual hotels.

Negotiation starts after including the guidelines from head office in the budget: “I need to make sure that my targets are ambitious, but I also need to protect my managers from nonsense from head office” (E). The general manager is referring to nonsense as out-of-the-money targets from head office. Due to the information asymmetry head office is not aware of hotel specific situations that might affect growth. Therefore, the eventual budget is negotiated between the general manager and the regional director. After the negotiating process the regional manager and the general manager sign the budget.

Two interviewees indicate that they slightly deviate from the process stated above. “The guidelines come from head office, with those figures I start budgeting” (G). After that the process of negotiating is equal.

Final decision

Including lower-level managers in the process, participative budgeting, enables head office to use the valuable information held by managers. But does this information lead to more accurate targets? According to the interviewees the key reason for including lower-level managers in the process is improved target setting. “I used to get my targets from head office, that was really guesswork from head office. I used to never reach my targets or perform much and much better than my target”(G). Another interviewee explains why this occurs: “How could head office know how this hotel will perform next year? They only use macro-figures that they apply pro-rata for each hotel. I examine the local market and internal factors and use that for my targets”(C).

Although agents are included in the target setting process, a superior always takes the final decision. During the negotiating process between the general manager and the management team, the general manager makes the final decision. The same is for the negotiating process between general manager and head office. “If the CFO really believes that the target is not stretched enough, he will add something” (E). By adding something the general manager meant increasing the GOI. It is then up to the hotels how to process this amendment in the budget, by increasing revenue and/or decreasing costs.

(30)

30 4.3.3 Conclusion proposition 2

Proposition 2:

Participative budgeting has a positive effect on the accuracy of target setting under the assumption that information asymmetry is present.

All interviewees have indicated that they have mix of different KPIs. Both individual, job-related KPIs, as KPIs for the entire hotel. The role of the KPIs is stimulating high performance by making it the basis for their bonus.

The role of target setting is for planning and control purposes. Targets allow managers and head office to allocate resources and inform investors. They provide managers with an authorization to spend resources that fit in the budget. Most interviewees indicate that the process of target setting starts at hotel level. That is where the information is about the local market and internal opportunities and threats. The budgeting process at individual hotels is a team-effort. All managers are asked to deliver input for their specialization. After the preliminary budget is set, head office presents their thoughts about next year. This is based on regional figures produced by an external agency. These are macro-figures that are allocated pro-rata to the hotels. This provides head office a rough estimate of what next year’s performance will be. Next step is the negotiating process between individual hotels and head office.

As most interviewees indicate that they are specialists for their hotel and, due to information asymmetry, head office does not have that same information. To confirm proposition 2, it is crucial to determine if this specialized information is actually used to set targets accurately. Most interviewees confirm that this information is used and that if they were not included in the process, targets would be less accurate. To conclude, targets set by the lower-level managers include information about the local market and micro-level, which leads to a more precise forecast of future performance. Head office only relies on macro-level information, which is a very rough estimate of reality. Some evidence was found to confirm proposition 2, however this is based on the perception of the lower-level managers.

4.4 Performance evaluation and reward system

Following the sequence of the theoretical model, the next step is identifying the performance evaluation and reward system in the case company.

(31)

31 4.4.1 Performance evaluation

Which KPIs 13, 15 Group-level: GOI and customer satisfaction Individual: financial and non-financial measures

Individual: non-financial measures

8 6 2 Other experiences 14 No

Yes, without bonus system

Yes, bonus system for high executives only

6 2 1 Process 16, 17 Objective evaluation, no exception

Objective evaluation, exceptions possible Subjective evaluation

2 4 2

Figure 10: matrix performance evaluation

KPIs that form the basis

The term performance evaluation used by Ferreira and Otley has a different meaning in this context. As mentioned earlier, the performance evaluation and bonus are distinctive. Employees are rewarded in terms of salary for their performance. If they score well on their performance evaluation, there might be room for promotion or a salary increase. Vice versa, performing not up to standards lead, to stricter supervision or even ending the contract. “The bonus reward is for doing something extra for the company” (C). There is a strict distinction between the bonus and the performance evaluation. However, there is one situation possible where they are intertwined. If the employee scores insufficient, the lowest score possible, then the employee is not eligible for a bonus.

The bonus target is based on individual and group reward. 60% of the targets are company- or hotel-wide targets, such as gross operating income and customer satisfaction. The other 40% are individual targets. Those targets differ per manager. Non-financial measures are negotiated between the manager and the superior. Financial targets are usually set top-down.

Other experiences

Most interviewees note that they did not have experienced a different reward system and target setting at a previous job or employer. However, two interviewees did. “At my previous job I was not eligible for a bonus, having no bonus does not motivate for superior performance. It is about recognition” (A). The other interviewee stated that managers could not influence the overall performance that much: ”my previous employer believed that only the sales director and GM directly influenced the performance” (H).

(32)

32 Process

The targets that are set are based on previous year’s performance. The superior and manager decide on a target for the current year. After approval from head office, the managers receive a bonus-letter in which their targets are recorded. “If I reach my target I get a bonus, it is always measured objectively” (H). Another interviewee perceives it slightly more subtle: “I am evaluated objectively, unless I have a really good explanation for deviating from my target”(D). More human oriented jobs; such as HR managers are solely evaluated subjectively. This because the HR manager does not have quantitative targets. The choice between objective and subjective performance evaluation is based on the characteristic of the KPI.

4.4.2 Reward system

Reward-system 18 Each bonus component is assessed individually Failing to meet one target leads to no bonus reward

6 2 Difficulties 19, 20, 27 No, it is easier to make my targets and motivating

Yes, trade-off between realistic and reachable target Yes, it could lead to earnings management

Long process of target setting

2 3 3 3

Figure 11: matrix reward system

Reward-system

AccorHotels has set-up a reward system for its managers. 60% of the bonus targets are related to hotel performance, and 40% are individual bonus targets. Lower-level manager can earn up to 10% of their annual salary as a bonus, specialized mid-level managers up to 20%, and the general manager and head office managers can earn up to 45% of their annual gross salary as a bonus.

As stated in the table, six interviewees are rewarded individually on each target. “Each target in my bonus letter has a weight, if I reach my target I obtain a bonus for that component” (B). In addition, the interviewees stated that if they reach their target they get the full bonus for that component, if they manage to reach it with more than 95% they get part of the bonus. This applies for both the individual KPIs and for the group-level performance. Two interviewees indicate that they have a stricter reward-system. “If I obtain my individual bonus target but the group doesn’t, I do not get a bonus at all. And vice versa” (H).

(33)

33 Difficulties

Two out of eight interviewees do not perceive any difficulties with the current reward system. In their opinion the current system is motivating, because they can set their own targets which provides a sense of responsibility. “Setting my own targets is more valuable to me than obtaining a bonus” (F). Moreover, the two interviewees indicated that the targets are more realistic under the current system compared to when head office used to set the targets for next year. The two interviewees indicate that they are not that occupied with their bonus. ”I am not a finance guy, linking my day-to-day performance is really too far away from my annual bonus”(F). The other interviewee indicates that he does not find it ethically correct to steer his performance towards the bonus target, however he could understand that others do.

Three interviewees indicate that the current system leads to a difficult trade-off between an ambitious target and a reachable target for bonus purposes. “I need to make sure that our targets are ambitious, but also achievable for my employees” (E). This seems to play an important role during target setting as one of the interviewees stated: “It plays a major role, people are occupied with their bonus during target setting”(D). The three interviewees describe it as playing a double role, which creates tension. Head office tends to increase the budget proposal. “We shouldn’t budget too ambitious, head office always increases the target” (E). Another remarkable comment was: “it is like a major purchase, e.g. buying a kitchen, then you also set your first bid very low” (B).

Three interviewees even believe that it could lead to (real) earnings management. All the interviewees indicate that the target for next year always should be higher than current year’s performance. This leads to earnings-management and budget games: “I could imagine that some managers slow down their growth once they have reached their target, to make sure they can grow next year too” (A). This comment is strengthened by an example given by another interviewee. Last year Luxembourg had an excellent year due to a large political event, next year they had to grow even further, without this large event and with the closing of one of the hotels.

Some interviewees find that the process of target setting takes too long. The budget process starts in October and ends in March. Moreover, the bonus targets are not validated until June. This has two effects. Firstly, the managers do not know where to focus on. At the time of validation up to five months of the year have already past. Secondly, it works demotivating. “The budgets are already outdated when they are approved, which decreases my motivation to really put effort in it” (D).

The distorting effect of the reward system on target setting at corporate organizations

The distorting effect of the reward system on target setting at

corporate organizations

Table of Contents