# Incentives and monitoring: impact on the financial and non-financial orientation of capital budgeting

Michael J. Turner (UQ Business School, The University of Queensland, St Lucia, Queensland, Australia)
Leonard V. Coote (UQ Business School, The University of Queensland, St Lucia, Queensland, Australia)

ISSN: 2049-372X

Publication date: 9 April 2018

## Abstract

### Purpose

While investment decisions may be financial decisions, there is a growing recognition that they are also often non-financially based decisions. The purpose of this study is to report findings focused on the project selection stage of capital budgeting, which has the objectives of exploring for: the relative degree of emphasis decision makers attach to a financial and non-financial orientation in capital budgeting; and the role, if any, that two agency theory variables have on the relative degree of emphasis: a personal incentive for project go-ahead and monitoring of project outcomes through a post-audit.

### Design/methodology/approach

Discrete choice experiments (DCEs) are used and framed in a between-subjects 2 (personal incentive) × 2 (monitoring) design. DCEs are well-suited to research questions which examine some tension between competing alternatives. For example, trade-offs involving the relative degree of emphasis decision makers attach to a financial and non-financial orientation in capital budgeting.

### Findings

In the absence of a personal incentive and monitoring, decision makers attach a significant degree of emphasis to cash inflows and cash outflows, both financial factors, and one strategic non-financial factor being improvement in the position of the firm vis-à-vis competitors in capital budgeting. However, when decision makers receive a personal incentive from project go-ahead, they attach a lower degree of emphasis to cash outflows. Alternatively, when there is monitoring through a post-audit and a personal incentive, decision makers attach a higher degree of emphasis to cash outflows.

### Practical implications

Decision makers attach a significant degree of emphasis to only a relatively narrow band of attributes in making a capital budgeting decision, which is true in both the absence of and in the presence of the agency conditions. There is also little support for the view that there is any higher degree of emphasis attached to a financial orientation vis-à-vis a non-financial orientation. A particularly important finding relates to the overarching goal of monitoring through a post-audit. One view is that it should foster more accurate forecasting by making forecasters aware that their efforts will be reviewed. However, the findings of this study appear to be more supportive of a view that post-audits might lead agents to become more conservative or even shy away from projects.

### Originality/value

The study makes contributions to the growing field of research which has the objective of exploring for the relative degree of emphasis decision makers attach to a financial and non-financial orientation in capital budgeting. In particular, it extends the prior research through its investigation of the role that two agency theory variables play in the relative degree of emphasis decision makers attach to a financial and non-financial orientation: a personal incentive for project go-ahead and monitoring of project outcomes through a post-audit.

## Keywords

#### Citation

Turner, M. and Coote, L. (2018), "Incentives and monitoring: impact on the financial and non-financial orientation of capital budgeting", Meditari Accountancy Research, Vol. 26 No. 1, pp. 122-144. https://doi.org/10.1108/MEDAR-02-2017-0117

### Publisher

:

Emerald Publishing Limited

## 1. Introduction

The corporate finance literature sees capital budgeting as comprising of four stages: project identification, project development, project selection and project control (see reviews by Burns and Walker, 2009; Gordon and Pinches, 1984; Mukherjee, 1987). This study reports findings focused on the project selection stage, which has the objectives of exploring for:

• the relative degree of emphasis decision makers attach to a financial and non-financial orientation in capital budgeting; and

• the role, if any, that two agency theory variables might have on the relative degree of emphasis: a personal incentive for project go-ahead; and monitoring of project outcomes through a post-audit[1].

As Frezatti et al. (2013, p. 300) explain:

Long-term investment decisions affect and are affected by agency conflicts due to the agent’s and the principal’s distinct perceptions of risk, the agent’s relationship with the principal (particularly regarding the agent’s performance assessment by the principal), and the appraisal methods and mechanisms that support each party’s decision processes[2].

While the selection stage has been the most investigated by survey researchers (Burns and Walker, 2009), it has focused largely on the application of financially oriented selection techniques (Haka, 2007). However, while investment decisions may be financial decisions, there is a growing recognition that they are also often non-financially based decisions. Nevertheless, studies have mainly considered on only the impact of national context (Carr and Harris, 2004; Carr and Tomkins, 1996, 1998; Shields et al., 1991), with little evidence provided regarding which contextual variables, other than country context, might be associated with the relative degree of emphasis decision makers attach to a non-financial orientation in capital budgeting (Chen, 2008; Verbeeten, 2006). This current study aims to address these shortcomings especially because application of financially oriented capital budgeting appraisal techniques does not always translate into better firm performance (Carr et al., 1994).

The method used in this current study are discrete choice experiments (DCEs), which are framed in a between-subjects 2 (personal incentive) × 2 (monitoring) design. Findings indicate that in the absence of a personal incentive and monitoring decision makers place a significant degree of emphasis on cash inflows and cash outflows, both financial factors, and only one strategic non-financial factor being improvement in the position of the firm vis-à-vis competitors. Decision makers, however, place no statistically significant degree of emphasis on any of the other attributes. However, when decision makers receive a personal incentive from project go-ahead they attach a lower degree of emphasis to cash outflows. Alternatively, when there is a presence of monitoring through a post-audit and a personal incentive, decision makers place a higher degree of emphasis on cash outflows. Under both agency conditions there are no significant differences in the relative degree of emphasis decision makers attach to the other attributes.

We offer reasoning as to why decision makers exhibit changes in only the relative degree of emphasis they attach to cash outflows and not any other attribute, which may stem from the principle of controllability (Merchant and Otley, 2006). Our research also carries practical implications, especially in relation to the benefits or not of post-audits (Farragher et al., 1999; Koch et al., 2009; Soares et al., 2007). The remainder of this study is organised as follows. The next section provides a review of the literature and given the exploratory nature of this study, propositions are also developed. Subsequent sections address, in turn, the method used and the study’s findings. The final section contains the discussion and conclusion, which highlights the main contributions, implications, limitations and avenues for future research.

## 2. Literature review and proposition development

### 2.1 The relative degree of emphasis attached to a financial and non-financial orientation

While a large volume of survey-based research signifies that one can be confident that the application of discounted cash flow capital budgeting techniques relative to other financially oriented techniques such as payback and accounting rate of return has increased substantially (Haka, 2007), there is only a relatively limited understanding as to whether this has changed over the same time period vis-à-vis non-financially oriented capital budgeting appraisal methods. For the purposes of this current study, a non-financial orientation can include strategic factors (e.g. the degree of fit of the capital budgeting project with business strategy), political/risk factors (e.g. the investment track record of the sponsor) and intuitive/performance factors (e.g. effects on the morale of personnel within an organisation) (Butler et al., 1993; Guilding, 2003).

Some of the earliest survey-based works concerning a non-financial orientation were carried out by Kamath and Elmer (1989) as well as Kamath and Oberst (1992) who assessed the impact of these on hospital capital budgeting by asking general managers to rank several non-financial factors specific to hospital capital budgeting. In these studies, almost all respondents stated that non-financial factors entered into the capital budgeting decision-making process. Porwal and Singhvi (1978) examined employee relations, community relations, shareholder relations and competitive position as non-financial factors in a survey of large manufacturing firms in India, finding differences in the relative degree of emphasis attached to these factors based on organisational profitability and size.

Motivated by the paucity of research surrounding the circumstances in which the use of discounted cash flow (DCF) methods may or may not be beneficial, Chen (2008) conducted a cross-sectional survey-based study using the responses of 115 CFOs of publicly traded manufacturing firms in the USA Respondents were asked to rate the relative importance of a DCF analysis (i.e. financial orientation) vis-à-vis a strategic analysis (i.e. non-financial orientation) in capital budgeting decision-making. Chen (2008) also sought to examine the extent to which the use of DCF techniques and non-financial factors might be associated with both product standardisation and firm strategy. While both methods played a significant role in capital budgeting, on a five-point scale from 1 = not important to 5 = extremely important, DCF techniques were found to be more important (mean 4.017, std. dev. 1.067) than non-financial factors (mean 3.616, std. dev. 1.156). Non-financial factors, however, were considered to be only a partial substitute when a DCF analysis were considered to be less effective. In this case, DCF analyses were found to be less effective within those environments presenting less product standardisation owing to the greater uncertainty surrounding inputs to DCF techniques.

Another key study is the survey-based research of Turner and Guilding (2012), which among other objectives, gathered responses from 101 hotel general managers concerning the relative degree to which financial and non-financial factors influenced whether an investment proposal were given the go-ahead. Using the typology developed by Butler et al. (1993), their survey featured items for financial, strategic (non-financial), political (non-financial) and intuitive (non-financial) factors. On a seven-point Likert scale from 1 = not at all to 7 = to a large extent, financial factors (mean 6.04, std. dev. 0.78) were found to have more influence than non-financial factors (mean 3.85, std. dev. 1.23) in influencing whether an investment proposal were given the go-ahead.

Drawing on Chen (2008), an analysis based on a DCF analysis is considered more important than one based on non-financial factors. Furthermore, financial factors tend to have a greater influence than non-financial factors in whether an investment proposal is given the go-ahead (Turner and Guilding, 2012). A further factor worth noting is Porter’s (1996) observation that users generally prioritise quantitative information over qualitative information. From these perspectives, decision makers are therefore expected to attach a higher degree of emphasis to a financial orientation relative to a non-financial orientation in capital budgeting. P1 has been worded in a manner consistent with this expectation[3]:

P1.

Decision makers attach a higher degree of emphasis to a financial orientation than a non-financial orientation in capital budgeting.

### 2.2 The impact of a personal incentive for project go-ahead

Capital budgeting can be viewed as transpiring within a context where the agency model is central (Haka, 2007). This means that while firm maximisation plays an important role in a manager’s investment decisions (Fatemi et al., 1983), it is not the sole criterion. Self-interest, for example, can often be a major factor (Kida et al., 2001). From this perspective attribution theory (Kelley, 1973) and legitimacy theory (Cialdini and Goldstein, 2004) suggest that the appraiser of a capital budgeting project will make an underlying assessment about the extrinsic motivations or incentives which face the proposal preparer (Cheng and Mahama, 2011). Here it is important to recognise that there is an innate human propensity to assume that extrinsic rewards play a guiding role in shaping employee behaviour (Kadous et al., 2005). Accordingly, where the preparer of a capital budgeting proposal stands to receive a personal incentive for securing project go-ahead, their approver is expected to interpret this behaviour by establishing causality in light of the extrinsic motivation (Wagner and Gooding, 1997).

From these perspectives, it is expected that where the preparer of a capital budgeting proposal stands to receive a personal incentive for securing project go-ahead, they will perceive for there to be a lower overall likelihood of gaining project acceptance from their approver. Given the importance of financial factors in capital budgeting (Bennouna et al., 2010; Gitman and Vandenberg, 2000; Graham and Harvey, 2001; Truong et al., 2008) and decision makers preference for them (Chen, 2008; Turner and Guilding, 2012), it is therefore expected that decision makers will perceive for it to be in their best interests to attach a higher degree of emphasis to projected cash inflows and to attach a lower degree of emphasis to projected cash outflows so as to raise the likelihood of the project gaining acceptance (Cantarelli et al., 2008)[4]. P2a and P2b have been worded in a manner consistent with this expectation.

P2a.

Decision makers who receive a personal incentive for project go-ahead will attach a higher degree of emphasis to cash inflows in capital budgeting.

P2b.

Decision makers who receive a personal incentive for project go-ahead will attach a lower degree of emphasis to cash outflows in capital budgeting.

Although non-financial information is often considered less important in capital budgeting (Chen, 2008; Turner and Guilding, 2012), non-financial information is nevertheless also often used to some extent to assist in capital budgeting decision-making (Moyer et al., 2001). As a successful capital budgeting project does not necessarily have to be justifiable on purely financial grounds (Carr et al., 1994). It is therefore expected that decision makers will perceive for it to be in their best interests to attach a higher degree of emphasis to non-financial factors so as to raise the likelihood of the project gaining go-ahead. P2c has been worded in a manner consistent with this expectation.

P2c.

Decision makers who receive a personal incentive for project go-ahead will attach a higher degree of emphasis to non-financial factors in capital budgeting.

### 2.3 The impact of a personal incentive and monitoring

To improve the accuracy of capital budgeting forecasts firms often engage in monitoring control activities to assist in the promotion and acceptance of only those projects that will be viable and beneficial from the firms’ point-of-view. In capital budgeting the most common mechanism by which monitoring control can be enacted is through the use of a post-audit (Haka, 2007). While there have been few studies focused upon the control stage of the capital budgeting process (Burns and Walker, 2009), Pohlman et al. (1988) found that 75 per cent of their respondent firms had performed a post-audit. Furthermore, Gitman and Vandenberg (2000) reported that 47 per cent of the firms they surveyed had formal procedures for evaluating the operating performance of existing projects.

Although there are various definitions of what a post-audit is, it is generally agreed that they can be introduced depending on the timing of the process either: “early” at the start-up of an investment project and/or “intermediate” during the operational phase, which can include controls associated with verification of budgeted performance and/or “final” at the end of the life-cycle of an investment (Giovanni and Maccarrone, 2001). The primary objectives of a post-audit are to serve as a mechanism of financial control, to provide information pertaining to future capital expenditure decisions, to eliminate psychological and/or political impediments associated with asset control and abandonment, and they can have a psychological impact on the proposers of capital investments (Gordon and Myers, 1991).

Post-audits are essentially designed to evaluate the efficiency of managers and to assess the accuracy of the basic assumptions and estimates that they initially made about a project before it received the go-ahead (Chenhall and Morris, 1993). In this way the overarching goal of a post-audit is to foster more accurate forecasting by making forecasters aware that their efforts will be reviewed (Farragher, et al., 1999) and that the review outcomes will be used to administer rewards such as promotion and punishments such as denials of promotion (Koch et al., 2009). On the one hand, post-audits can lead to behavioural-related improvements whereby the actions of the agent are better aligned with those of their principal (Azzone and Maccarrone, 2001). However, on the other hand post-audits can have deleterious effects whereby an agent can become more conservative or even shy away from projects when proposing capital budgeting projects (Koch et al., 2009). A further issue is that post-audits traditionally emphasise numerical figures as opposed to non-financial information because financial information tends to be considered more objective than an analysis based on non-financial information (Farragher et al., 1999).

From each of these perspectives imposition of monitoring through a post-audit is expected to result in decision makers’ recommendations being scrutinised more closely by their principal. Drawing on psychology literature, when monitoring is combined together with a personal incentive it is therefore expected to make the features of the decision makers’ choice to recommend a project or not become more difficult or to require more difficult trade-offs, meaning that they are likely to display a higher propensity to avoid making the choice (i.e. to not recommend) (Luce, 1998). This is because more difficult choices typically involve some higher level of internal tension owing to the clashes between opposing or contradictory impulses (Weber et al., 2001). Tension can be associated with feelings of uncertainty, unpleasant feelings of distress (Janis and Mann, 1977) or of anxiety, worry and unease (Luce, 1998).

Decision makers are therefore expected to have a higher fear of failure where monitoring through a post-audit is in place, relative to when it is absent, because poor project performance is more readily capable of being detected. Under these circumstances, decision makers may deem it better to have a bad outcome as a result of inaction (i.e. by not recommending a project) rather than failure as a result of action (i.e. by recommending a project) (Zeelenberg et al., 2001). Choice avoidance, for example, can be a viable coping mechanism in the face of difficult choices (Lazarus, 1991). This is important because as individuals experience feelings such as fear or worry they tend to judge the riskiness of investments as higher and therefore choose safer investments with lower payoffs (Loewenstein et al., 2001). Decision makers can therefore be motivated to avoid decision alternatives that produce negative feelings (Zeelenberg and Beattie, 1997). As Sprinkle et al. (2008, p. 437) explain, “risk aversion leads individuals to […] select “safe” projects […] [that] reduce firm welfare”.

The combination of monitoring and a personal incentive for securing project go-ahead is therefore expected to lead decision makers to become more fearful and risk averse about recommending a project owing to the greater potential for bad project performance to be detected by way of the post-audit. Furthermore, evidence of bad project performance can also impose costly reputational losses to the decision maker involved (Ozbas, 2005). As post-audits typically follow a financially orientated imperative (Farragher et al., 1999), it is therefore expected that decision makers will become more conservative in terms of the proposals they put forward for recommendation, which will result in changes to the relative degree of emphasis they attach to cash inflows and cash outflows. Specifically, decision makers are expected to attach a lower degree of emphasis to cash inflows and a higher degree of emphasis to cash outflows so as to make their project recommendation carry a lower risk of failure. P3a and P3b have been worded in a manner consistent with these expectations:

P3a.

Decision makers who receive a personal incentive for project go-ahead and who are monitored through a post-audit will attach a lower degree of emphasis to cash inflows in capital budgeting.

P3b.

Decision makers who receive a personal incentive for project go-ahead and who are monitored through a post-audit will attach a higher degree of emphasis to cash outflows in capital budgeting.

When a personal incentive for project go-ahead and monitoring through a post-audit are present, because post-audits typically have a financial imperative (Farragher et al., 1999), decision makers’ are expected to be less risk averse in connection with placing a higher degree of emphasis on non-financial information so as to raise the likelihood of project acceptance. P3c has been worded in a manner consistent with this expectation:

P3c.

Decision makers who receive a personal incentive for project go-ahead and who are monitored through a post-audit will attach a higher degree of emphasis to non-financial factors in capital budgeting.

## 3. Method

### 3.1 Research design

DCEs have been used extensively in several fields such as applied economics and psychology, marketing and transport and logistics studies (Hensher et al., 2005; Louviere et al., 2000) and their use in accountancy research is called for (Turner and Coote, 2017) and increasing (Chung and Hensher, 2015a, 2015b; Jones et al., 2012). The main advantage of DCEs is that they are well-suited to research questions which examine some tension between competing alternatives. Tension here meaning two or more phenomena in a dynamic relationship that involve both competition and complementarity (English, 2001), such as the trade-offs in the relative degree of emphasis decision makers attach to financial and non-financial information in capital budgeting.

The specific design of the DCEs implemented in this study includes nine attributes. These attributes have been selected to be representative of the types of information that typically confront decision makers at the point of evaluating a capital budgeting project for recommendation. For this purpose Butler et al.’s (1993) categorisation of financial and non-financial information in capital budgeting has been adopted with minor adaptation to the specifics of the contextual case scenario. As shown in Table I, along with their identifying code, this has resulted in the following: two financial factors, cash inflows (F1) and cash outflows (F2); three strategic factors, fit of the capital budgeting project with business strategy (S1), improvement in the position of the hotel vis-à-vis competitors (S2) and improvement in the overall performance of the hotel (S3); two risk factors, the impact of economic changes on the project (i.e. internal or external changes) (R1) and the impact on the financial position of the hotel if the project failed (R2); and two performance factors, the effect of the project on product quality (P1) and the effect of the project on the morale of hotel staff (P2).

The questionnaire was developed in close consultation with fourteen academics and six practitioners who piloted it at various stages. Consistent with the recommendations of Brownell (1995) the instrument was also reviewed by experts in survey methods. Several iterations of the questionnaire were pilot tested. Piloting was achieved by first providing a copy of the questionnaire to the reviewers. Following this, telephone discussions and, where possible, face-to-face meetings were carried out with the individual concerned who was asked to comment on the questionnaire’s layout, the relevance and appropriateness of questions and any ambiguities. Through the piloting phase it emerged that the capital budgeting project would need to be of a long-term nature and of a large size to draw out the requisite agency challenges (Frezatti et al., 2013). This resulted in the case study scenario for this current study focusing on a potential $20m investment to expand a hotel via the construction of 80 new rooms. Throughout the piloting phase much input was sought concerning development of the levels for each of the study’s nine attributes. In terms of the financial attributes, it emerged that it would be most representative of realism that it be important for respondents to be able to receive either a positive or a negative net cash flow possibility across the various choice sets they would be presented with. For example, to capture the possibility of a project having a negative net cash flow effect but non-financial benefits. In terms of the non-financial attributes, labels for each of these except for the attribute S1 were developed as having from a very small improvement, impact or effect through to having a very large improvement, impact or effect meaning that they were non-directional with only a positive potential effect. Fit of the capital budgeting project with business strategy (S1), however, used a negative/positive scale which ranged from a very poor fit to a very good fit. Decision makers who piloted the survey saw this attribute as requiring a dual-directional scale to capture the potential of a manager promoting a project that would not fit well with the firm’s strategy so as to achieve, for example, short-term goals, as opposed to a firms’ longer-term strategy. Table I provides information about this current study’s: meta-attributes (column one); attributes and an identifying code (column two); and levels for each of the attributes (column three). As each of the nine attributes are defined by four levels each, there are many thousands of choice alternatives (49). For this reason, a fractional factorial design has been adopted which enables the entire design to be systematically divided into versions that each possess a unique combination of choice sets and which can be randomly allocated to groups of decision makers (Hensher et al., 2005). Following Louviere et al. (2000), a form of fractional factorial design known as an orthogonal main effects plan was used to design the choice sets and alternatives for both the “A” and “B” choices and the “not recommend” option. The not recommend option is necessary to act as a baseline alternative, corresponding to the status quo or do-nothing situation. This results in a closer representation of a real-world decision-making situation because one of the options must always be in the decision makers’ currently feasible choice set. An orthogonal main effects plan enables generation of a more parsimonious design (in 64 rows) than does the full factorial enumeration (in 128 rows). To reduce respondent fatigue, the 64 possible choice sets were then blocked into 8 blocks, with each block containing 8 choice sets. Hence, each decision maker was required to complete 8 choice sets from one of 8 randomly assigned choice set blocks. In doing this, an equal number of each of the questionnaire versions was administered and received from the respondents. Appendix 2 contains an example of one of the 64 choice sets presented to the respondents and the capital budgeting scenario posed is available in Appendix 1. To investigate the impact of a personal incentive and monitoring on the relative degree of emphasis decision makers attach to a financial and non-financial orientation, the DCEs were placed within a master experiment defined by the agency variables. The incentive structure was operationalised through the variable; personal incentive to agent for project go-ahead. The monitoring structure was operationalised through the variable; monitoring of agent. Hence, the agency conditions of a personal incentive to agent for project go-ahead and monitoring of agent were manipulated in a between-subjects fashion by varying the experimental case scenarios. These variables are described further below. The variables studied gave rise to four conditions (2 × 2). Of the four conditions, the singular condition monitoring of agent was excluded from the analysis. Based on the principles of agency theory, there must first be personal incentive until the need arises to monitor. As such, an examination of monitoring was considered to be not theoretically correct. ### 3.2 Research context and scenarios The research instrument contained a case scenario which adopted the approach taken in Carr and Tomkins (1998) by focusing on a single industry. The particular industry chosen is the hotel industry, which is characterised as having high capital intensity (ISHC, 2015). We incorporated Guilding’s (2003) view that capital budgeting is rendered more complex in hotels that operate with a management contract[5]. Framing the scenario in this way allowed for a clearly defined principal-agent relationship to be explained and an industry context where capital budgeting is important (Collier and Gregory, 1995). The capital budgeting scenario related to a$20m proposed investment to expand a hotel via the construction of 80 new rooms. Subjects were informed by their instructor at the beginning of the session that the proposed investment would have a 10-year life-span and that the information in each choice set represented annualised information, which would perpetuate for the 10-year period. This was considered pertinent because of the horizon problem encountered in an agency relationship whereby an agent can achieve short-term goals, while their principal often desires the achievement of longer-term goals (Lambert, 2001).

Manipulations of personal incentive and monitoring were embedded in the scenarios. In addition to the common part of the scenario (see Appendix 1), participants in the personal incentive condition read the following statement:

The hotel operating company you are employed by is enthusiastic for the proposed project to proceed as they will enjoy a significant increase in management fees. Your operating company has therefore agreed to pay you a considerable salary increase should the project go-ahead.

Participants in the monitoring condition read the following statement:

Should the project go-ahead, the project will be subject to an audit after it gets underway by the property’s owner. Post-audits are designed to monitor the progress and your performance in relation to the investment project (i.e., budgeted versus actual). The management contract in your hotel stipulates that your appointment as General Manager requires approval of the owner. Further, the owner has authority to remove you as General Manager for unacceptable performance.

### 3.3 Subjects

Subjects for the experiment consisted of student accountants enrolled in a third-year accounting course at a large public university in Australia. Our choice of student accountants as surrogates for professional accountants is justified in several ways. First, Waldron and Fisher (2017) highlight the suitability of accounting students as surrogates for professional accountants, especially where the task involves an investment decision. Second, Swain and Haka (2000) demonstrate there to be no clear differences in the capital budgeting decisions of experienced and inexperienced capital budgeters. Third, there are numerous examples of studies in the capital budgeting literature which have provided strong arguments in favour of the use of student accountants as surrogates for professional accountants (Brown and Solomon, 1993; Cheng and Mahama, 2011; Cheng et al., 2003; Kerler et al., 2014).

A pen-and-paper questionnaire was administered in two separate sessions late in the academic semester at the conclusion of the regularly scheduled lectures. In each session, the instructor clarified that participation was voluntary and would take no more than 20 min. Subjects were asked to not communicate with each other while they were filling in the questionnaire. Subjects were randomly assigned to treatment conditions as different versions of the experimental materials were randomly assigned. The study adhered to all relevant guidelines of the ethical review process at the University.

There were a total of 269 students who had enrolled in the course, 79 students were in attendance at the lectures when the questionnaire was administered and a response was collected from each of these 79 students. The response rate based on total enrolment in the course is therefore 29.37 per cent. A total of 5 responses were discarded, however, due to being incomplete (i.e. missing data) meaning that there were a total of 74 usable responses. Each of these responses were subjected to and passed recommended data screening procedures (Hair et al., 2010). As each respondent chose between the competing alternatives 8 different times, this represents a total of 592 observed choices to analyse.

According to Orme (2010), an acceptable sample size is: n ≥ 500c/ta, where n is the number of respondents, t is the number of tasks (i.e. choice sets completed), a is the number of alternatives per task (i.e. A, B and no-choice) and c is the largest number of levels for any one attribute. Hence, for this current study n = 74, t = 8, a = 3 and c = 4; meaning that the calculation is (500 × 4)/(8 × 3) = 24. Hence our total sample size of 74 (592 choices) responses is more than adequate and we use this sample for our tests involving P2a, P2b and P2c (see Table IV) and P3a, P3b and P3c (see Table V). Our tests involving P1, however, adopt only those responses which were obtained in the absence of an agency condition, which means that this model’s sample size is n = 18 (144 choices). Although this is slightly smaller than the recommendation (Orme, 2010), the results (see Table III) are similar to those in the control groups of the agency models (see Panel A of Table IV and Panel A of Table V). Hence, we do not believe that the sample size for the exploration of P1 poses any strong threat to the validity of the results of this current study. However, we do note this as a limitation and so the results in connection with P1 should be read with this borne in mind. The distribution of the 74 responses across each of the study’s experimental conditions, i.e. 2 (personal incentive) × 2 (monitoring) are reported in Table II.

## 4. Results

### 4.1 Econometric specification

The relative degree of emphasis decision makers attach to each of the nine attributes was explored with a conditional logistic regression (logit) model (McFadden, 1974). An implicit assumption going into the model is that all changes in one level of an attribute are expected to have a similar effect. For the purposes of simplifying the interpretations and specifying and estimating a more parsimonious model, each of the non-financial attributes (see Table I) were re-coded from four levels to two levels. This means that for each of the non-financial attributes, the two lower-levels (e.g. very poor fit and poor fit) have been combined into one lower level, and the two higher-levels (e.g. very good fit and good fit) have been combined into one higher level. As a check on our results, we explored other approaches to coding the non-financial attributes, such as the use of continuous coding for the non-financial attributes. These analyses showed our results to be robust to the different coding schemes. For this reason, we present results based on combining the lower- and upper-levels of the non-financial attributes without loss of generality.

### 4.2 Estimation of the base model

Table III provides estimation of the conditional logit base model, which produced a log likelihood value of −136.27 and a pseudo R2 of 13.86 per cent (n = 18; 144 choices). Table III provides the following information: meta-attributes (column one); attributes (column two); regression coefficients of the mean, otherwise known as betas (i.e. β) (column three); standard error (column four); z score (column five); and level of significance (column six). In Table III, the z-scores (column six) represent a standardisation estimate divided by the standard error and are used to assess the significance of the relative degree of emphasis decision makers attach to each of the attributes. Thus, the z-scores are applicable in determining the relative degree of emphasis decision makers attach to a financial and non-financial orientation because coefficients are on the same scale and they measure the deviation from zero of an estimated attribute. Accordingly, Table III indicates that in the absence of a personal incentive and in the absence of monitoring that decision makers attach a significant degree of emphasis to:

• cash inflows (F1) (coef. 0.229, z = 3.64, p < 0.01);

• improvement in the position of the hotel vis-à-vis competitors (S2) (coef. 0.446, z = 3.12, p < 0.01); and

• cash outflows (F2) (coef. 0.379, z = 3.08, p < 0.01).

From this perspective, it appears that while decision makers attach the most significant degree of emphasis to cash inflows (F1) (z = 3.64); a strategic factor (S2) (z = 3.12) has ranked ahead of cash outflows (F2) (z = 3.08), a financial factor. Furthermore, if cash inflows and cash outflows are considered as one financial factor (i.e. net cash flows), it is not clear as to whether decision makers attach any higher degree of emphasis to financial factors relative to non-financial factors. Hence, at the level of a single attribute analysis, it would appear that there is little support for P1. Table III indicates that decision makers did not attach any significant degree of emphasis to any of the other attributes.

### 4.3 Estimation model for a personal incentive

The model presented in Table IV produces a log likelihood value of −554.33 and a pseudo R2 of 14.77 per cent (n = 74; 592 choices). This model and the model in the next section are estimated using all 74 responses. Panel A provides the model’s main effects, which serve as a baseline control group from which the results in Panel B are additive. Panel A assumes that interactions between the attributes and the manipulated variables are not significant and as would be expected these results are similar to those of the base model (Table III). Decision makers, for example, continue to attach a significant degree of emphasis to:

• cash inflows (F1) (coef. 0.222, z = 7.40, p < 0.01);

• improvement in the position of the hotel vis-à-vis competitors (S2) (coef. 0.462, z = 6.81, p < 0.01); and

• cash outflows (F2) (coef. 0.277, z = 4.83, p < 0.01).

Panel B of Table IV presents the attribute interactions with a personal incentive and consistent with P2b the results show cash outflows to have a statistically significant negative coefficient (coef. −0.079, z = −1.69, p < 0.05). This indicates that decision makers who receive a personal incentive from project go-ahead attach a lower degree of emphasis to cash outflows. However, as decision makers did not exhibit any significant changes in the relative degree of emphasis they sought to attach to the other attributes, this means that there is no significant evidence to support either P2a or P2c.

### 4.4 Estimation model for the interaction between a personal incentive and monitoring[6]

The model presented in Table V produces a log likelihood value of −552.34 and a pseudo R2 of 15.07 per cent (n = 74; 592 choices). Panel A provides the model’s main effects and consistent with the base model (Table III) decision makers continue to attach a significant degree of emphasis to:

• cash inflows (F1) (coef. 0.218, z = 6.54, p < 0.01);

• improvement in the position of the hotel vis-à-vis competitors (S2) (coef. 0.452, z = 5.59, p < 0.01); and

• cash outflows (F2) (coef. 0.321, z = 4.87, p < 0.01).

Panel B of Table V presents the attribute interactions with a personal incentive and monitoring and consistent with P3b the results show cash outflows to have a statistically significant positive coefficient (coef. 0.093, z = 1.65, p < 0.05). This indicates that decision makers who receive a personal incentive and that are monitored attach a higher degree of emphasis to cash outflows. However, as decision makers did not exhibit any significant changes in the relative degree of emphasis they sought to attach to the other attributes, there is no significant evidence to support P3a or P3c.

## 5. Discussion and conclusion

This study advances the literature because instead of determining the use or non-use of capital budgeting techniques as in most surveys, this study has instead pursued the objectives of exploring for:

• the relative degree of emphasis decision makers attach to a financial and non-financial orientation in capital budgeting; and

• the role, if any, that two agency theory variables might have on the relative degree of emphasis: a personal incentive for project go-ahead; and monitoring of project outcomes through a post-audit.

With regard to the first of these objectives, decision makers have been found to attach a significant degree of emphasis to only a relatively narrow band of attributes in making capital budgeting decisions: cash inflows, cash outflows and one strategic factor being improvement in the position of the hotel vis-à-vis competitors. With regard to the significant strategic factor, this appears as consistent with the view that an enhancement of competitive advantage should be one of a managers’ main motivations (Burcher and Lee, 2000). Furthermore, that strategic factors are often considered to be the most important non-financial consideration in the evaluation of a capital budgeting proposal (Alkaraan and Northcott, 2006). For each of the other strategic factors, however, such as fit of the capital budgeting project with business strategy, decision makers did not attach any significant degree of emphasis to these, which appears as counterintuitive if the prior literature is recognised whereby strategic factors should be used to ensure a project’s link with a firm’s strategy (Alkaraan and Northcott, 2006; Carr and Tomkins, 1996). Our findings therefore build on the work of Chen (2008) and Turner and Guilding (2012) who while also demonstrating the usefulness of strategic factors in capital budgeting, found strategic factors to have a lower level of importance and influence than financial factors.

Further to research objective one, decision makers have been found to attach no significant degree of emphasis to either of the two risk attributes or the performance factors. In terms of risk factors, incorporation of risk is theorised to be most beneficial for the purpose of career concerns, reputation entrenchment or to obtain incentive components (Markowitz, 1959). Hence, it may be perhaps somewhat unsurprising that decision makers did not attach any significant degree of emphasis to risk factors in the absence of the agency conditions. Alternatively, it does appear as unusual, however, that no significant degree of emphasis was attached to risk factors especially in the presence of a personal incentive. Likewise, for performance factors we know that employers often take steps to grow their employee’s stock of value and investments into human capital and are made to create sustained shareholder value through the mechanism of profitable performance (Roslender, 2009). Moreover, improving the quality of a firm’s products is usually considered fundamental to a businesses’ success (Everett et al., 1997). Again, it is surprising that decision makers did not attach any significant degree of emphasis to these risk factors.

With regard to the second research objective of this study, when decision makers received a personal incentive for project go-ahead they sought to attach a lower degree of emphasis to cash outflows. Alternatively, when monitoring through a post-audit was introduced together with the presence of a personal incentive, decision makers then sought to attach a higher degree of emphasis to cash outflows. Under both of these agency conditions decision makers demonstrated no other significant changes in the relative degree of emphasis they sought to attach to the other financial factor, cash inflows or any of the non-financial attributes. A possible reason as to why decision makers exhibited significant changes in only the relative degree of emphasis they sought to attach to cash outflows may relate to the controllability principle (Merchant and Otley, 2006).

According to the controllability principle, if individuals have no impact on an outcome then it should serve no meaningful purpose to hold them accountable for that outcome. As Antle and Demski (1988, p. 701) explain, for a manager to “control” an evaluation statistic (e.g. cash outflows), the question which needs to asked is “whether his or her supply of inputs is able to affect the probability distribution of the output statistic”. If this is the case, decision makers may perceive a relative lack of control in connection with cash inflows along with the other non-financial attributes but to have significant control over cash outflows. This may be true if one considers the often-large sums of cash payments which are required at the start-up of a project, which are often based on a vendor’s quote and hence would carry a reasonable degree of certainty. Contrast this to cash inflows, which are arguably subject to greater uncertainty, especially from the external environment. Uncontrollable factors, for example, can fall into three broad categories: horizontal interdependencies (i.e. decisions made by other managers in a company); vertical or hierarchical interdependencies (i.e. decisions made by superiors); and external factors (i.e. unforeseen changes in the economic and competitive environment, natural catastrophes, etc.) (Giraud et al., 2008). Another reason might be that measuring non-financial outcomes may be somewhat difficult and subjective leading decision makers to perceive there to be little control over such.

The findings of this study have potentially important practical implications, especially in terms of the design of capital budgeting monitoring control systems. As was highlighted earlier, it has been claimed that an important goal of a post-audit is to foster more accurate forecasting by making forecasters aware that their efforts will be reviewed (Farragher et al., 1999). Alternatively, findings of this research appear to be more supportive of the view that post-audits can lead agents to become more conservative or even shy away from projects (Koch et al., 2009), as demonstrated through decision makers attaching a significantly higher degree of emphasis to cash outflows and to hence make a capital budgeting project appear as less favourable.

As Eccles (1991) explains, where managers perceive an appraisal system to be unfair due to the inclusion of factors beyond their control, they are likely to engage in behaviour to protect themselves but which might be harmful to their company. A conjecture then is that this protection mechanism manifests in decision makers compensating for their perceived relative lack of control over cash inflows and non-financial factors, by attaching a higher degree of emphasis to cash outflows. That is, to reduce their feelings of a fear of failure and negative consequences. However, this type of action might as well lead to a lower number of capital budgeting projects receiving a decision maker’s recommendation relative to when monitoring was absent (noting here that the opposite is likely to be true in the presence of only a personal incentive). If this is true, a firm’s growth might well be hindered to some extent through the application of monitoring through traditional post-audit control methods that attach emphasis to both cash inflows and cash outflows. Recall that post-audits traditionally emphasise numerical figures as opposed to non-financial information because financial information tends to be considered more objective than an analysis based on non-financial information (Farragher et al., 1999). Firms may therefore find it beneficial to design post-audits in a way that attaches a higher level of emphasis towards cash outflows as these likely carry a perceived higher degree of controllability.

Such a recommendation, however, should be tempered in light of the literature which has analytically demonstrated that under certain conditions, for example, information asymmetry, it can often be in a principal’s best interests to use any and all information at their disposal to assess their agent’s efforts, including any indicators beyond their control (Antle and Demski, 1988; Holmstrom, 1979). Some authors have highlighted, for example, that it is not always in the best interests of a firm to apply the controllability principle (Manzoni, 2002). That is, holding managers accountable for performance factors that they cannot fully control but to which they can have at least some partial influence, can induce them to maintain attention to those factors capable of their influence (Dearden, 1987; Merchant, 1989).

### 5.1 Limitations and avenues for future research

Firms often have more acceptable projects than funds available to invest, which can lead to capital rationing, which occurs about 40 per cent of the time (Gitman and Vandenberg, 2000). As this current study did not examine the impact of capital rationing, future research could usefully explore for this impact on the financial and non-financial orientation of capital budgeting. Another fruitful extension of this research would be to recognise that while cash inflows and outflows are often the key inputs to financially oriented capital budgeting appraisal techniques such as payback and accounting rate of return, more sophisticated DCF appraisal techniques, such as net present value and internal rate of return must also typically assign risk premiums subjectively or based upon sensitivity analysis (Shao, 1994). Future research could hence examine for the role of other financial factors in capital budgeting such as the assignment of discount rates or required rates of return.

As with any experimental research, when interpreting the results of this study it should be borne in mind that the findings relate only to the capital budgeting scenario posed. The extent to which the results are generalisable to other projects and contexts is therefore something for which future research is called. Another limitation is that the labels assigned to each of the non-financial attributes in this current study, except for fit of the capital budgeting project with business strategy were non-directional. While these labels were rationalised based on theory and on the extensive piloting phase, it could be that having both directional and non-directional scales may have impacted upon the results of this current study. Future research may therefore consider research designs that use a uniform approach to label development.

## Table I.

Meta-Attributes, attributes and levels

Meta-attributes Attributes (identifying code) Levels appearing on questionnaire
Financial Cash inflows (F1) $8m per year$6m per year
$4m per year$2m per year
Cash outflows (F2) $4m per year$3m per year
$2m per year$1m per year
Strategic (non-financial) Fit of the capital budgeting project with business strategy (S1) Very good fit
Good fit
Poor fit
Very poor fit
Improvement in the position of the hotel vis-à-vis competitors (S2) Very large improvement
Large improvement
Small improvement
Very small improvement
Improvement in the overall performance of the hotel (S3) Very large improvement
Large improvement
Small improvement
Very small improvement
Risk (non-financial) The impact of economic changes on the project (i.e. internal or external changes) (R1) Very large impact
Large impact
Small impact
Very small impact
The impact on the financial position of the hotel if the project failed (R2) Very large impact
Large impact
Small impact
Very small impact
Performance (non-financial) The effect of the project on product quality (P1) Very large effect
Large effect
Small effect
Very small effect
The effect of the project on the morale of hotel staff (P2) Very large effect
Large effect
Small effect
Very small effect

## Table II.

Distribution of responses across the experimental conditions

Personal incentive Monitoring Respondents Choices
Absent Absent 18 144
Absent Present 16 128
Present Absent 22 176
Present Present 18 144
74 592

## Table III.

Emphasis attached to a financial and non-financial orientation

Meta-attributes Attributes Coef. of the mean (β) Std. err. z P > |z|A
Financial Cash inflows (F1) 0.229 0.063 3.64 0.000,*
Cash outflows (F2) 0.379 0.123 3.08 0.000,*
Strategic (non-financial) Fit of the capital budgeting project with business strategy (S1) 0.063 0.132 0.48 0.633
Improvement in the position of the hotel vis-à-vis competitors (S2) 0.446 0.143 3.12 0.000,*
Improvement in the overall performance of the hotel (S3) 0.175 0.141 1.24 0.216
Risk (non-financial) The impact of economic changes on the project (i.e. internal or external changes) (R1) −0.013 0.134 −0.09 0.924
The impact on the financial position of the hotel if the project failed (R2) 0.155 0.139 1.12 0.264
Performance (non-financial) The effect of the project on product quality (P1) −0.266 0.138 −1.92 0.054
The effect of the project on the morale of hotel staff (P2) −0.211 0.143 −1.47 0.141
Not recommend 0.024 0.473 0.05 0.960
Log likelihood = −136.27
Pseudo R2 = 0.1386
n = 18; 144 choices
Notes:

A One-tailed test;

Two-tailed test;

*

p < 0.01; **p < 0.05

## Table IV.

Effect of a personal incentive

Meta-attributes Attributes Coef. of the mean (β) Std. err. z P > |z|A
Panel A: main effects
Financial Cash inflows (F1) 0.222 0.030 7.40 0.000,*
Cash outflows (F2) 0.277 0.057 4.83 0.000,*
Strategic (non-financial) Fit of the capital budgeting project with business strategy (S1) 0.031 0.063 0.48 0.628
Improvement in the position of the hotel vis-à-vis competitors (S2) 0.462 0.068 6.81 0.000,*
Improvement in the overall performance of the hotel (S3) 0.102 0.067 1.53 0.126
Risk (non-financial) The impact of economic changes on the project (i.e. internal or external changes) (R1) 0.057 0.065 0.87 0.385
The impact on the financial position of the hotel if the project failed (R2) 0.078 0.065 1.19 0.235
Performance (non-financial) The effect of the project on product quality (P1) −0.072 0.067 −1.09 0.277
The effect of the project on the morale of hotel staff (P2) −0.114 0.068 −1.68 0.092
Not recommend −0.200 0.230 −0.87 0.386
Panel B: attribute interaction with a personal incentive
Financial Incentive × cash inflows (F1) 0.140 0.023 0.01 0.195
Incentive × cash outflows (F2) −0.079 0.047 −1.69 0.046,**
Strategic (non-financial) Incentive × fit of the capital budgeting project with business strategy (S1) −0.056 0.063 −0.89 0.375
Incentive × improvement in the position of the hotel vis-à-vis competitors (S2) −0.010 0.068 −0.14 0.888
Incentive × improvement in the overall performance of the hotel (S3) 0.014 0.067 0.21 0.417
Risk (non-financial) Incentive × the impact of economic changes on the project (i.e. internal or external changes) (R1) 0.065 0.065 0.99 0.161
Incentive × the impact on the financial position of the hotel if the project failed (R2) 0.011 0.065 0.17 0.432
Performance (non-financial) Incentive × the effect of the project on product quality (P1) 0.038 0.067 0.57 0.285
Incentive × the effect of the project on the morale of hotel staff (P2) −0.003 0.068 −0.05 0.961
Log likelihood −554.33
Pseudo R2 = 0.1477
n = 74; 592 choices
Notes:

A

One-tailed test;

Two-tailed test;

*

p < 0.01;

**

p < 0.05

## Table V.

Effect of a personal incentive and monitoring

Meta-attributes Attributes of the mean (β) Std. err. z P > |z|A
Panel A: main effects
Financial Cash inflows (F1) 0.218 0.033 6.54 0.000,*
Cash outflows (F2) 0.321 0.066 4.87 0.000,*
Strategic (non-financial) Fit of the capital budgeting project with business strategy (S1) 0.035 0.075 0.47 0.640
Improvement in the position of the hotel vis-à-vis competitors (S2) 0.452 0.081 5.59 0.000,*
Improvement in the overall performance of the hotel (S3) 0.124 0.080 1.55 0.122
Risk (non-financial) The impact of economic changes on the project (i.e. internal or external changes) (R1) 0.038 0.077 0.49 0.622
The impact on the financial position of the hotel if the project failed (R2) 0.107 0.079 1.36 0.172
Performance(non-financial) The effect of the project on product quality (P1) −0.138 0.079 −1.75 0.081
The effect of the project on the morale of hotel staff (P2) −0.154 0.081 −1.91 0.056
Not recommend −0.207 0.231 −0.90 0.369
Panel B: attribute interaction with a personal benefit and monitoring
Financial Incentive × Monitoring × cash inflows (F1) −0.010 0.027 −0.36 0.359
Incentive × Monitoring × cash outflows (F2) 0.093 0.057 1.65 0.049,**
Strategic (non-financial) Incentive × Monitoring × fit of the capital budgeting project with business strategy (S1) 0.027 0.075 0.36 0.359
Incentive × Monitoring × improvement in the position of the hotel vis-à-vis competitors (S2) −0.010 0.081 −0.13 0.900
Incentive × Monitoring × improvement in the overall performance of the hotel (S3) 0.050 0.080 0.62 0.268
Risk (non-financial) Incentive × Monitoring × the impact of economic changes on the project (i.e., internal or external changes) (R1) −0.051 0.077 −0.66 0.508
Incentive × Monitoring × the impact on the financial position of the hotel if the project failed (R2) 0.047 0.079 0.60 0.274
Performance (non-financial) Incentive × Monitoring × the effect of the project on product quality (P1) −0.132 0.079 −1.68 0.093
Incentive × Monitoring × the effect of the project on the morale of hotel staff (P2) −0.057 0.081 −0.70 0.481
Log likelihood = −552.340
Pseudo R2 = 0.1507
n = 74; 592 choices
Notes:

A

One-tailed test;

Two-tailed test;

*

p < 0.01;

**

p < 0.05

## Table AI.

Example choice set

A B
Cash inflows $6m per year$2m per year
Cash outflows $4m per year$1m per year
Fit with business strategy Very poor fit Very good fit
Improvement vis-à-vis competitors Very small improvement Very small improvement
Improvement in the overall performance Large improvement Small improvement
The impact of possible economic changes Very small impact Very large impact
The impact of failure on financial position Small impact Large impact
The effect on product quality Small effect Very small effect
The effect on the morale of hotel staff Very small effect Very small effect

## Notes

1.

For the purposes of this study the relative degree of emphasis decision makers attach to a financial and non-financial orientation in capital budgeting can be interpreted as either purposeful emphasis or decision makers may unconsciously attach emphasis.

2.

A substantial literature exists on the agency problem and the associated goal incongruence between principals and agents (Baiman, 1990; Berle and Means, 1932; Fama, 1980; Jensen and Meckling, 1976). The pioneers of agency theory were Jensen and Meckling (1976) who describe an agency relationship as arising when there is a contract whereby one party (the principal) appoints another party (the agent) to perform some service on behalf of the principal.

3.

Proposition 1 relates to the first objective of this current study: to explore for the relative degree of emphasis decision makers attach to a financial and non-financial orientation in capital budgeting. While our pursuit of this objective builds on Chen (2008), it is different because with regard to the financial orientation of this current study it includes cash inflows and cash outflows, rather than Chen’s focus on a DCF analysis. Furthermore, Chen examined only strategic factors, whereas we include strategic factors, risk factors and performance factors as pertaining to a non-financial orientation. We also build on Turner and Guilding (2012) who while gathering information from the same financial and non-financial perspectives as we do in this current study, our research is different because it gathers data through the use of DCEs, rather than Turner and Guilding’s use of traditional Likert type scale measurement. While a decision maker might, for example, consider both a financial and a non-financial orientation to be important or influential; DCEs instead force decision makers to make trade-offs among all of the available attributes in making a clear choice of which they prefer. This is argued to represent a more realistic and real world decision-making context vis-à-vis the use of Likert type scale measurement (Turner and Coote, 2017). Finally, it is worth noting that the work of Chen (2008) also gathered data through the use of Likert type scales.

4.

The way that a decision maker “perceives” the relative degree of emphasis that they attach to a financial and non-financial orientation in capital budgeting is expected to extenuate from the way in which they regard, understand or interpret the situation. For this reason, perceptions tend to arise through both conscious and unconscious cognitive thought processes (Marcel, 1983).

5.

Under a hotel management contract, one party (the owner) owns the hotel and its associated assets and a second party (the operator) manages the hotel’s day-to-day operational activities through their general manager (Rushmore, 2002).

6.

Interaction terms were also added to a model (but not reported here) to capture the effects of the singular agency theory variable monitoring. As expected, decision makers did not attach any significant degree of emphasis to a financial or a non-financial orientation. This means that decision makers reported no change in the relative degree of emphasis they attach to a financial and non-financial orientation where only the monitoring condition was present.

## Appendix 1. Questionnaire scenario

You are the General Manager of a prestigious 5-star CBD located hotel operating with a management contract. A hotel management contract is a written agreement between a hotel owner and operator that involves the hotel owner retaining legal ownership of the hotel site, building, plant and equipment, furnishings and inventories, while the operator assumes responsibility for managing the day-to-day operations of the hotel. You are formally employed by the hotel operator.

(This paragraph was present in all versions of the questionnaire)

The owner of your hotel is considering a \$20m investment to expand the hotel via construction of 80 new rooms. You are responsible for preparing the capital expenditure proposal for this project, which will be used by the owner to help them decide whether to go-ahead with the project or not. Note that the project is 100 per cent funded by the owner. The operating company does not contribute any capital towards the proposed project.

(This paragraph was present in all versions of the questionnaire)

The hotel operating company you are employed by is enthusiastic for the proposed project to proceed, as they will enjoy a significant increase in management fees. Your operating company has therefore agreed to pay you a considerable salary increase should the project go-ahead.

(This paragraph was only present in those scenarios where personal incentive to agent for project go-ahead was manipulated)

Should the project go-ahead, the project will be subject to an audit after it gets underway by the property’s owner. Post-audits are designed to monitor the progress and your performance in relation to the investment project (i.e. budgeted versus actual). The management contract in your hotel stipulates that your appointment as General Manager requires approval of the owner. Further, the owner has authority to remove you as General Manager for unacceptable performance.

(This paragraph was only present in those scenarios where monitoring of agent was manipulated)

For each of the following choice sets, which alternative would you most likely take forward to the owner.

## Appendix 2

Alternative A □ Alternative B □Neither □

## References

Alkaraan, F. and Northcott, D. (2006), “Strategic capital investment decision-making: a role for emergent analysis tools? A study of practice in large UK manufacturing companies”, British Accounting Review, Vol. 38 No. 2, pp. 149-173.

Antle, R. and Demski, J.S. (1988), “The controllability principle in responsibility accounting”, The Accounting Review, Vol. 63 No. 4, pp. 700-718.

Azzone, G. and Maccarrone, P. (2001), “The design of the post-audit process in large organizations: evidence from a survey”, European Journal of Innovation Management, Vol. 4 No. 2, pp. 73-87.

Baiman, S. (1990), “Agency theory in managerial accounting: a second look”, Accounting Organizations and Society, Vol. 15 No. 4, pp. 341-371.

Bennouna, K., Meredith, G.G. and Marchant, T. (2010), “Improved capital budgeting decision making: evidence from Canada”, Management Decision, Vol. 48 No. 2, pp. 225-247.

Berle, A. and Means, G. (1932), The Modern Corporation and Private Property, Macmillan, New York, NY.

Brown, C.E. and Solomon, I. (1993), “An experimental investigation of explanations for outcome effects on appraisals of capital budgeting decisions”, Contemporary Accounting Research, Vol. 10 No. 1, pp. 83-111.

Brownell, P. (1995), Research Methods in Management Accounting, Coopers and Lybrand, Melbourne.

Burcher, P.G. and Lee, G.L. (2000), “Competitiveness strategies and AMT investment decisions”, Integrated Manufacturing Systems, Vol. 11 No. 5, pp. 340-347.

Burns, R.M. and Walker, J. (2009), “Capital budgeting surveys: the future is now”, Journal of Applied Finance, Vol. 19 Nos 1/2, pp. 78-90.

Butler, R., Davis, L., Pike, R. and Sharp, J. (1993), Strategic Investment Decisions: Theory, Practice, and Process, Routledge, London.

Cantarelli, C.C., Flyvbjerg, B., Molin, E.J.E. and van Wee, B. (2008), “Cost overruns in large-scale transportation infrastructure projects: explanations and their theoretical embeddedness”, European Journal of Transport Infrastructure Research, Vol. 10 No. 1, pp. 5-18.

Carr, C. and Harris, S. (2004), “The impact of diverse national values on strategic investment decisions in the context of globalization”, International Journal of Cross Cultural Management, Vol. 4 No. 1, pp. 77-99.

Carr, C. and Tomkins, C. (1996), “Strategic investment decisions: the importance of SCM: a comparative analysis of 51 case studies in UK, US and German companies”, Management Accounting Research, Vol. 7 No. 2, pp. 199-217.

Carr, C. and Tomkins, C. (1998), “Context, culture and the role of the finance function in strategic decisions: a comparative analysis of Britain, Germany, the USA and Japan”, Management Accounting Research, Vol. 9 No. 1, pp. 213-239.

Carr, C., Tomkins, C. and Bayliss, B. (1994), Strategic Investment Decisions: A Comparison of UK and W. German Practices in the Motor Components Industry, Aldershot: ICAEW/Avebury.

Chen, S. (2008), “DCF techniques and nonfinancial measures in capital budgeting: a contingency approach analysis”, Behavioral Research in Accounting, Vol. 20 No. 1, pp. 13-29.

Cheng, M.M. and Mahama, H. (2011), “The impact of capital proposal guidelines and perceived preparer biases on reviewers’ investment evaluation decisions”, Australian Journal of Management, Vol. 36 No. 3, pp. 349-370.

Cheng, M.M., Schulz, A.K.D., Luckett, P.F. and Booth, P. (2003), “The effects of hurdle rates on the level of escalation of commitment in capital budgeting”, Behavioral Research in Accounting, Vol. 15 No. 1, pp. 63-86.

Chenhall, R.H. and Morris, D. (1993), “The role of post completion audits, managerial learning, environmental uncertainty and performance”, Behavioral Research in Accounting, Vol. 5 No. 1, pp. 170-186.

Chung, D. and Hensher, D.A. (2015a), “Modelling risk perceptions of stakeholders in public–private partnership toll road contracts”, Abacus, Vol. 51 No. 3, pp. 437-483.

Chung, D. and Hensher, D.A. (2015b), “Risk management in public–private partnerships”, Australian Accounting Review, Vol. 25 No. 1, pp. 13-27.

Cialdini, R.B. and Goldstein, N.J. (2004), “Social influence: compliance and conformity”, Annual Review of Psychology, Vol. 55, pp. 591-621.

Collier, P. and Gregory, A. (1995), “Investment appraisal in service industries: a field study analysis of the UK hotels sector”, Management Accounting Research, Vol. 6 No. 1, pp. 33-57.

Dearden, J. (1987), “Measuring profit center managers”, Harvard Business Review, September-October, pp. 84-88.

Eccles, R.G. (1991), “The performance measurement manifesto”, Harvard Business Review, January-February, pp. 131-137.

English, T. (2001), “Tension analysis in international organizations: a tool for breaking down communication barriers”, The International Journal of Organizational Analysis, Vol. 9 No. 1, pp. 58-83.

Everett, E.A.J., Corbett, L.M., Flores, B.E., Harrison, N.J., Lee, T.S., Rho, B.-H., Ribera, J., Samson, D. and Westbrook, R. (1997), “An international study of quality improvement approach and firm performance”, International Journal of Operations & Production Management, Vol. 17 No. 9, pp. 842-873.

Fama, E. (1980), “Agency problems and the theory of the firm”, Journal of Political Economy, Vol. 88 No. 2, pp. 288-307.

Farragher, E.J., Kleiman, R.T. and Sahu, A.P. (1999), “Current capital investment practices”, Engineering Economics, Vol. 44 No. 2, pp. 137-149.

Fatemi, A.M., Ang, J.S. and Chua, J.H. (1983), “Evidence supporting shareholder wealth maximization in management controlled firms”, Applied Economics, Vol. 15 No. 1, pp. 49-60.

Frezatti, F., de Souza Bido, D., Da Cruz, A.P.C., Barroso, M.F.G. and de Camargo Machado, M.J. (2013), “Investment decisions on long-term assets: integrating strategic and financial perspectives”, European Accounting Review, Vol. 22 No. 2, pp. 297-336.

Giovanni, A. and Maccarrone, P. (2001), “The design of the investment post-audit process in large organisations: evidence from a survey”, European Journal of Innovation Management, Vol. 4 No. 2, pp. 73-87.

Giraud, F., Langevin, P. and Mendoza, C. (2008), “Justice as a rationale for the controllability principle: a study of managers’ opinions”, Management Accounting Research, Vol. 19 No. 1, pp. 32-44.

Gitman, L. and Vandenberg, P.A. (2000), “Cost of Capital techniques used by major US firms: 1997 vs. 1980”, Financial Practice and Education, Vol. 10 No. 2, pp. 53-68.

Gordon, L.A. and Myers, M.D. (1991), “Post-auditing capital projects: are you in step with the competition?”, Management Accounting, Vol. 72 No. 7, pp. 39-42.

Gordon, L.A. and Pinches, G. (1984), Improving Capital Budgeting: A Decision Support System Approach, Addison-Wesley, Reading, MA.

Graham, J. and Harvey, C. (2001), “The theory and practice of corporate finance: evidence from the field”, Journal of Financial Economics, Vol. 60 No. 2/3, pp. 187-243.

Guilding, C. (2003), “Hotel owner/operator structures: implications for capital budgeting process”, Management Accounting Research, Vol. 14 No. 3, pp. 179-199.

Hair, J.F., Black, W.C., Babin, B.J. and Anderson, R.E. (2010), Multivariate Data Analysis (7 ed.), Pearson, Upper Saddle River, NJ.

Haka, S.F. (2007), “A review of the literature on capital budgeting and investment appraisal: past, present, and future musings”, in Chapman, C., Hopwood, A.G. and Shields, M.D. (Eds), Handbook of Management Accounting Research, Vol. 2, Elsevier, New York, NY, pp. 697-728.

Hensher, D.A., Rose, J.M. and Greene, W.H. (2005), Applied Choice Analysis: A Primer, Cambridge University Press, Cambridge.

Holmstrom, B. (1979), “Moral hazard and observability”, The Bell Journal of Economics, Vol. 10 No. 1, pp. 74-91.

ISHC (2015), CapEx2014: A Study of Capital Expenditures in the Hotel Industry, American Hotel & Lodging Educational Institute, Lansing, MI, pp. 1-243.

Janis, I.L. and Mann, L. (1977), Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment, Free press, New York, NY.

Jensen, M. and Meckling, W. (1976), “Theory of the firm: managerial behaviour, agency costs and ownership structure”, Journal of Financial Economics, Vol. 3 No. 4, pp. 305-360.

Jones, S., Hensher, D.A., Rose, J. and Walker, R.G. (2012), “Infrastructure asset reporting options: a stated preference experiment”, Accounting Horizons, Vol. 26 No. 3, pp. 465-491.

Kadous, K., Koonce, L. and Towry, K.L. (2005), “Qualification and persuasion in managerial judgement”, Contemporary Accounting Research, Vol. 22 No. 3, pp. 643-686.

Kamath, R. and Elmer, J. (1989), “Capital investment decisions in hospitals: survey results”, Health Care Management Review, Vol. 14 No. 2, pp. 45-56.

Kamath, R. and Oberst, E. (1992), “Capital budgeting practices of large hospitals”, The Engineering Economist, Vol. 37 No. 3, pp. 203-232.

Kelley, H. (1973), “The processes of causal attribution”, American Psychologist, Vol. 28 No. 2, pp. 107-128.

Kerler, W.A. III, Fleming, A.S. and Allport, C.D. (2014), “How framed information and justification impact capital budgeting decisions”, in Epstein, M.J. and Lee, J.Y. (Eds), Advances in Management Accounting, Vol. 23, Emerald Group Publishing, Bingley, pp. 181-210.

Kida, T., Moreno, K. and Smith, J. (2001), “The influence of affect on managers’ capital budgeting decisions”, Contemporary Accounting Research, Vol. 18 No. 3, pp. 477-494.

Koch, B., Mayper, A.G. and Wilner, N. (2009), “The interaction of accountability and post-completion audits on capital budgeting decisions”, Academy of Accounting and Financial Studies Journal, Vol. 13, pp. 1-27.

Lambert, R.A. (2001), “Contracting theory and accounting”, Journal of Accounting and Economics, Vol. 32 No 1/3, pp. 3-87.

Lazarus, R.S. (1991), “Progress on a cognitive-motivational-relational theory of emotion”, American Psychologist, Vol. 46 No. 8, pp. 819-834.

Loewenstein, G.F., Weber, E.U., Hsee, C.K. and Welch, N. (2001), “Risk as feelings”, Psychological Bulletin, Vol. 127 No. 2, pp. 267-286.

Louviere, J.J., Hensher, D.A. and Swait, J.J. (2000), Stated Choice Methods: Analysis and Application, Cambridge University Press, New York, NY.

Luce, M.F. (1998), “Choosing to avoid: coping with negatively emotion-laden consumer decisions”, Journal of Consumer Research, Vol. 24 No. 4, pp. 409-433.

McFadden, D. (1974), “Conditional logit analysis of qualitative choice behavior”, in Zarembka, P. (Ed.), Frontiers in Econometrics, Academic Press, New York, NY, pp. 105-142.

Manzoni, J.-F. (2002), “Management control: toward a new paradigm?”, in Epstein, M.J. and Manzoni, J.-F. (Eds), Performance Measurement and Management Control, Elsevier Science, Kidlington, Oxford, pp. 15-46.

Marcel, A.J. (1983), “Conscious and unconscious perception: experiments on visual masking and word recognition”, Cognitive Psychology, Vol. 15 No. 2, pp. 197-237.

Markowitz, H. (1959), Portfolio Selection, Wiley, New York, NY.

Merchant, K. (1989), Rewarding Results: Motivating Profit Center Managers, Harvard Business School Press, Boston, MA.

Merchant, K. and Otley, D.T. (2006), “A review of the literature on control and accountability”, Handbooks of Management Accounting Research, Vol. 2, pp. 785-802.

Moyer, R.C., McGuingan, J.R. and Kretlow, W.J. (2001), Contemporary Financial Management, South-Western, Cincinnati, OH.

Mukherjee, T.K. (1987), “Capital-budgeting surveys: the past and the future”, Review of Business and Economic Research, Vol. 22 No. 2, pp. 37-56.

Orme, B. (2010), “Sample size issues for conjoint analysis”, in Johnson, R. (Ed.), Getting Started with Conjoint Analysis, Strategies for Product Design and Pricing Research, 2nd ed., Research Publishers LLC, Madison, WI, pp. 57-66.

Ozbas, O. (2005), “Integration, organizational processes, and allocation of resources”, Journal of Financial Economics, Vol. 75 No. 1, pp. 201-242.

Pohlman, R.A., Santiago, E.S. and Markel, F.L. (1988), “Cash flow estimation practices of large firms”, Financial Management, Vol. 17 No. 2, pp. 71-79.

Porter, T.M. (1996), Trust in Numbers: The Pursuit of Objectivity in Science and Public Life, Princeton University Press, Princeton, NJ.

Porwal, L.S. and Singhvi, S.S. (1978), “A comparative study of capital expenditure evaluation techniques”, Long Range Planning, Vol. 11 No. 5, pp. 25-31.

Roslender, R. (2009), “The prospects for satisfactorily measuring and reporting intangibles: time to embrace a new model of (ac)counting?”, Journal of Human Resource Costing & Accounting, Vol. 13 No. 4, pp. 338-359.

Rushmore, S. (2002), “Hotel investments handbook”,

Shao, L.P. (1994), “Capital investment evaluation procedures utilized by North and Latin American affiliates”, Multinational Business Review, Vol. 2 No. 1, pp. 15-21.

Shields, M.D., Chow, C.W., Kato, Y. and Nakagawa, Y. (1991), “Management accounting practices in the US and Japan: comparative survey findings and research implications”, Journal of International Financial Management and Accounting, Vol. 3 No. 1, pp. 61-77.

Soares, J.O., Coutinho, M.C. and Martins, C.V. (2007), “Forecasting errors in capital budgeting: a multi-firm post-audit study”, The Engineering Economist, Vol. 52 No. 1, pp. 21-39.

Sprinkle, G.B., Williamson, M.G. and Upton, D.R. (2008), “The effort and risk-taking effects of budget-based contracts”, Accounting, Organizations and Society, Vol. 33 No. 4, pp. 436-452.

Swain, M.R. and Haka, S.F. (2000), “Effects of information load on capital budgeting decisions”, Behavioral Research in Accounting, Vol. 12, p. 171.

Truong, G., Partington, G. and Peat, M. (2008), “Cost-of-capital estimation and capital budgeting practice in Australia”, Australian Journal of Management, Vol. 33 No. 1, pp. 95-121.

Turner, M.J. and Coote, L.V. (2017), “Discrete choice experiments: a research agenda for experimental accounting”, Meditari Accountancy Research, Vol. 25 No. 1, pp. 158-182.

Turner, M.J. and Guilding, C. (2012), “Factors affecting biasing of capital budgeting cash flow forecasts: evidence from the hotel industry”, Accounting and Business Research, Vol. 42 No. 5, pp. 519-545.

Verbeeten, F.H.M. (2006), “Do organizations adopt sophisticated capital budgeting practices to deal with uncertainty in the investment decision? A research note”, Management Accounting Research, Vol. 17 No. 1, pp. 106-120.

Wagner, J.A. III. and Gooding, R.Z. (1997), “Equivocal information and attribution: an investigation of patterns of managerial sensemaking”, Strategic Management Journal, Vol. 18 No. 4, pp. 275-286.

Waldron, M. and Fisher, R. (2017), “Values and ethical judgments: the adequacy of students as surrogates for professional accountants”, Meditari Accountancy Research, Vol. 25 No. 1, pp. 37-64.

Weber, E.U., Baron, J. and Loomes, G. (2001), Conflict and Tradeoffs in Decision Making, Cambridge University Press, Cambridge.

Zeelenberg, M. and Beattie, J. (1997), “Consequences of regret aversion 2: additional evidence for effects of feedback on decision making”, Organizational Behavior and Human Decision Processes, Vol. 72 No. 1, pp. 63-78.

Zeelenberg, M., Inman, J.J. and Pieters, R. (2001), “What we do when decisions go awry: behavioral consequences of experienced regret”, in Weber, E.U., Baron, J. and Loomes, G. (Eds), Conflict and Tradeoffs in Decision Making, Cambridge University Press, Cambridge, pp. 136-155.

## Corresponding author

Michael J. Turner can be contacted at: m.turner@business.uq.edu.au