Prepayment in private use of lease cars
S.J. van Baardewijk-Berg July 7th, 2017 One of the ways in which behavioural ethics insights are used is to improve tax compliance. In the Netherlands, an extraordinary amount of fraud occurs in the reporting of private kilometres driven in a company car. Because of the way the regulation is organised, loss aversion might be a driver for this unethical behaviour. We attempted to test whether a situation where prepayment eliminated this loss frame would decrease fraud. In two experiments using online scenarios, we did not find any evidence of such an effect. However, we hope to have contributed to the construction of paradigms for testing loss aversion in real world situations that can eventually be used to conduct the large scale field experiments needed to improve policy and administration for the better. Introduction ... 2 Study 1 ... 6 Methods ... 6 Results ... 8 Discussion ... 10 Study 2 ... 11 Methods ... 11 Results ... 14 Discussion ... 15 Conclusion ... 15 References ... 17Introduction
When asked, almost everyone reports that they know right from wrong, and that they act in accordance to what they know is right (Pittarello et al., 2015). This is not surprising, as we tell countless stories to each other about how heroes hold on to the good to defeat what is bad. Of course we want to be like them, and not like their selfish enemies, who seek out every opportunity to gain power or money. We love thinking, writing and debating about what constitutes good moral values. But has it helped us achieve our goal of being as good as we can be? The behavioural ethics field studies how we behave when we are actually confronted with ethical decisions in real life. In fact, there are many situations were most people do not act in accordance to their own ethical standards at all. Even though a person may be committed to being a good person, they might still make unethical decisions, for example when they do not expect an ethical dimension to the task they are performing, when they are trying to reach a specific goal or prevent a loss, or when the situation makes it easy to construct a lie (Mazar Amir & Ariely, 2008; Ayal & Gino, 2011; Kern & Chugh, 2009; Grolleau, Kocher and Sutan, 2014, Ordóñez and Welsh, 2015; Shalvi et al. 2011; Beaussart, 2013). However, the interesting thing is that they do not maximise their selfish gain either. Instead, it has been shown that people often bend the rules just a bit, up to the point where they can still justify their actions as ethical. This way they can preserve their self-concept as an honest person, while still profiting a bit more from the given situation than they otherwise would. So apparently, the decision making context and the system of responsibility can make a big difference in people’s capability to act ethically (Thaler & Sunstein, 2009; Luyendijk, 2016). This finding has big impact on policy in many organisations. It might seem possible to find a few ‘rotten apples’, who use every opportunity they get to maximalise their selfish gain, and root them out from the compliant mass. If many people commit small lies, however, this means that detection of those lies is near impossible. It is therefore very important to uncover the mechanisms behind these irrational transgressions, in order to design choice environments and incentives in such a way that people are automatically encouraged to act in accordance with what they would report is right. Many organisations are working on just this, often in collaboration with scientists, policy makers, administrators and business, to create better policy and structures to increase ethical behaviour in our society. One of the first terrains to be tackled is tax compliance, where behavioural ethics initiatives have a lot of success in working with tax authorities to increase tax compliance and thus tax revenue, most notably among those Behavioural Insights Teams in the United Kingdom (BIT Staff, 2013). Private use of lease cars We find a clear case of an ethical decision in the tax regulation surrounding lease cars in the Netherlands. As per the current regulation, as established in the ‘wet op de loonbelasting’, the ‘wet inkomstenbelasting’, and related rules, employers are free to choose if they want to allow their employees to use company cars company cars not only for work trips, but also for private trips. If these private trips amount to more than 500 kilometres a year, a fixed amount of tax has to be paid as the use of the car is deemed income for the employee. Usually this amount is 25% of the worth of the car, but this percentage depends on various properties of the specific vehicle.In most cases, companies allow for private use of the car. Their employees can drive the car freely and pay the income tax for private use, but they also have the choice to drive their car for 500 kilometres a year or less for private purposes instead and relieve themselves of this income tax. If employees want to make use of this opportunity, they can submit a declaration (a so-called ‘Verklaring geen privégebruik auto’) to the Tax Authority, wherein they agree to not drive more than 500 kilometres in all their private trips of that calendar year combined. If they do so, their employer no longer has to withhold the income tax for the private use of the car from their salary. The employees are now obliged to keep track all kilometres driven with the car in their private journey administration, so that a tax inspector could review it in case of an audit1, as every company car driver must report in their tax declaration at the end of the year how much private kilometres they have driven (Belastingdienst.nl, 2017). Currently, about 250.000 hold such a declaration, a minority of the roughly 900.000 company car drivers (Berentsen, 2017, Belastingbelangen.nl, 2011, CBS, 2017). Despite big fines of up to 5278 euros, it is estimated in the past by the Dutch Tax Authority that every year at least 30% of those report fraudulently about how much they have used their lease car for private purposes, (Belastingbelangen.nl, 2011). This calculates to 78.000 taxpayers, which is an extraordinary amount according to the Tax Authority. In some cases, tax evaders expend a lot of effort to make it seem like they use their lease car less often for private purposes than they actually do. What’s even more, it’s not uncommon for these tax evaders to go through a great deal of trouble to construct false evidence for their fraudulent reports (Belastingbelangen.nl, 2010, Besteman, 2009). One unique aspect of this regulation is that in a majority of cases, employees can choose to submit a declaration at the start of the year, while the requirement for this rebate has not yet come about in practice. While in theory employees are free to choose at the start of the year if they want to submit a declaration or not, it could be that they have a hard time estimating their driving needs and the feasibility of hardly using the car. However, making the wrong choice is theoretically not even a problem, because the declaration can be nullified throughout the year. If employees submit a declaration, but chance their mind in the course of the year and decide they want to cross the 500 kilometres limit after all, they can ask for nullification of their declaration via an online form, allowing them to cross the 500 kilometre limit after doing so2. Of course, this means that they then need to start paying the income tax again for the private use of the car, and have to repay, all at once, the income tax for the months that have passed up to this point (Belastingdienst.nl, 2017). To give an impression of the amounts of money involved, €3000,- is an average amount a driver can expect to pay over the course of a year. However, this number can differ greatly depending on value of the car, it’s fuel-efficiency and the tax bracket of the employee, ranging from €500,- to €7500,- (Solnah, 2013)3. 1 This situation has come about as a result of the complex legal structure of responsibility surrounding private use of lease cars, as it protects the employer from 2 In theory, taxpayers can be fined for nullifying the declaration after the 500 kilometre limit has already been crossed. 3 Additionally, there is a theoretical upper limit to private use of the car. If the monetary value of the private use of the car for a given calendar year exceeds the value of the part of the value of the car that had to be counted as income, this is called ‘excessive private use’ and the value of the actual private use is counted as income instead. However, note
Behavioural ethics theory suggests that two mechanisms apply to this structure that make it a more tempting situation for taxpayers to commit fraud in their tax declaration. Loss aversion First, loss aversion could play a big role in this situation. When people experience a loss as compared to an equally large gain, the experience of loss is somehow more extreme than with the gain. The effect works to such an extent that people generally only take on a gamble if the amount they stand to gain is twice as big as the amount they stand to lose. Here we find one of those instances where irrational decision making really shows, as people are willing to expend more of their energy, time and other resources to prevent a certain loss, than to achieve an equally large gain, which is contrary to what would be predicted from classical economic theory (Tversky & Kahneman, 1992). The mechanism behind this behaviour is formalised in prospect theory. According to this theory, ownership plays a big role. People judge a new prospect not objectively, by a standard such as economical value, or utility of the final position, but in a non-linear way, depending on if the prospect of a certain choice is viewed as a loss or a gain relative to their current situation, as illustrated by the ownership curve in Figure 1. One’s position on this curve drastically changes one’s view on the losses and gains to be had. In many situations, this reference point naturally involves ownership of some things and not other things. Removing something from this initial endowment is treated as a loss, but adding the same good (to an endowment without it) is treated as a gain. This specific phenomenon of undervaluing opportunity costs (forgone gains) as compared to out-of-pocket costs (losses) is referred to as the endowment effect (Thaler, 1980, Camerer, 2000)4. It causes people to act irrationally in even simple decisions as they overvalue prepaid rewards and fail to maximise their financial gain (Kahneman et al., 1990, Hochman et al., 2014). that without a declaration, lease car drivers are not required to keep a journey registration and do not have to be able to prove how much kilometres they have driven as private trips. Instead, this burden of proof lies with the Dutch Tax Authority. 4 Closely related concepts are: the mere ownership effect, the status-quo bias and buying-selling price gaps. Figure 1 – Expected psychological value of a loss or gain in relation to the current reference point.
In the relatively short amount of time that the behavioural ethics field has existed, there have been many examples of how loss and gain framing can impact behaviour drastically in many different situations with identical outcomes.When investigating how this mechanism applies to ethical decisions, it has been shown that people lie more to avoid a loss than to secure an equally large gain, and also report that lying to avoid losses is more acceptable. (Kern & Chugh, 2009; Grolleau, Kocher and Sutan, 2014; Schindler and Pfattheicher, 2017). When we apply this knowledge to the lease car situation, we find that declaration holders who do not keep their private use below the 500 kilometre limit will have to pay back money that was already transferred to them as salary. Declaration holders who want to drive more than 500 kilometres privately, or have already done so, therefore face a looming loss. According to the theory mentioned above, they would be willing to spend an irrational amount of their resources to avoid this loss, which might include expending the energy needed to deviate from honesty in one’s behaviour (Shalvi et al., 2015) Is is thus possible that this prospect of loss persuades them to commit fraud on their tax declaration by reporting that they have driven less than 500 kilometres, as such misreporting would prevent this loss. As a comparison, it is not unthinkable to redesign this situation to where misreporting on the tax declaration results in a gain of money. One way to do so would be to let all lease car drivers pay upfront, and allow them to report at the end of the year how much they have driven for private purposes. Now, they only have the option to receive their money back, which would make fraudulent reporting an act of cheating to gain. Approach This inherent loss frame for declaration holders could be an explanation for the prevalence of misreporting of private use of lease cars. Therefore, we attempt to answer the question: Does the moment of tax payment influence lease car drivers’ willingness to defraud their tax declaration? Based on the discussed theory, we expect a difference in reference point between the prepayment situation and the current situation. Declaration holders can be seen as owners of money, who can pay to receive the ability to drive, whereas lease car drivers without a declaration can be seen as owners of the ability to drive, which they can exchange for money. Therefore, by introducing prepayment of the tax for driving the lease car privately, loss aversion toward this tax money should decrease as less ownership over the money is felt. Instead, the new reference point becomes the position of being able to drive the car as much as one would like, while having the option of forgoing this ability for money. Thus, we hypothesise that: Pre-payment of taxes for private use of lease cars will decrease taxpayers’ willingness to defraud their tax declaration To test if the moment of pay influences this decision, we have created a paradigm where people are put into the situation of a lease car driver who gets presented with an ethical dilemma regarding crossing the 500 kilometre limit for private kilometres. The regulation presented will either follow a pay upfront or a pay in the end approach, so
that in one case, misreporting results in a possibility for monetary gain, but in the other, it results in the avoidance of a monetary loss. We predict that the self-reported likeliness of fraudulent reporting will be lower in the pay upfront condition.
Study 1
Methods Participants To perform the scenario, 120 participants were recruited via Amazon’s Mechanical Turk marketplace (MTurk, www.MTurk.com) using the TurkPrime platform (Litman et al., 2016), restricted to be accessing MTurk from the United States of America. An informed consent was given before the experiment started and the Ethics Committee of CREED, the Center for Research in Experimental Economics and Political Decision Making, approved the study. Participants received a financial compensation. To be included, participants needed to finish the survey within 4 hours, the shortest setting possible in Qualtrics. Data collection was continued until 120 participants had finished the study. Procedure Participants were directed towards a survey in Qualtrics (Qualtrics, Provo, UT), where the study was introduced and information was provided for reaching out to the researchers for any questions. Participants were randomly but evenly divided over the two scenarios. They were instructed to read the scenario and answer a verification question to make sure they understood the scenario. In a new window, they were presented with the problem. In yet a next window, they were asked how they would respond to the problem. The questionnaire then continued with various follow-up questions. Apart from the scenarios, all questions were exactly the same over conditions. Scenarios To make the regulations understandable for our participants, we made several changes from reality. First, we translated the 500 kilometres limit to a 300 miles limit, as our participants were all from the USA. Secondly, the declaration was implemented as an organisational policy, so that all participants in the pay in the end condition were in effect declaration holders as a default situation. They could however still opt out of this via an online form. Furthermore, we left some details out of the regulation, such as the different categories of fuel-efficiency, or the fact that the tax authority is involved in this process of submitting and nullifying the declaration, instead only focusing on the employee and the organisation they work at. In both conditions, participants were introduced to driving a company car and the related policy, which involved the duty to pay taxes over a part of the car if it was used for private purposes for more than 300 miles, in addition to work trips. In the pay upfront condition, participants were presented with the organisational policy titled “This is your car!”. This meant that the tax for private use was withheld from the driver’s salary over the course of the year. If they wanted, they could keep a journey registration and if at the end of the year, they drove 300 miles or less for private purposes, they could ask for a refund of the tax money.In the pay at the end condition, participants were presented with the organisational policy titled “This is a work car!”. This meant that the readers had to keep a journey registration. If they saw they were going to cross 300 miles for private purposes, they would need to update the organisation, and pay the tax for private use of the car for that year at the end of the year. At the end of each scenario, a verification question was asked to make sure participants understood the rules presented in their scenario. Participants could only continue after giving the right answer. Task All participants were presented with the same problem, which stated that it is now October, and they have driven 280 miles for private purposes so far. They were told they had planned to go to a party, but going would mean crossing the 300 miles limit. They were informed they had the option of reporting this trip as a work trip without their employer knowing, but in doing so running a risk of being fined for fraud by the government. Measures After the problem, participants were asked: “How likely is it that you will report this trip as a work trip?”, to which they could answer on a 7 point scale, ranging from “Extremely likely” to “Extremely unlikely”. This way, one end of the scale represents unethical behaviour, namely to defraud one’s journey registration, and the other end represents other ways of solving the problem, which accounts for two very evident ways of dealing with the problem in a moral way, namely to report crossing the 500 kilometre limit, or to arrange alternative transport. Asking only this one scale could makes it more likely participants will not think too long about their answer and it does not offer them new solutions to the problem that they would not have thought of themselves. In a new question, different manipulation checks were done, to see if participants experienced the task as an ethical decision, felt like they were allowed to drive the car for private purposes, and if they felt like the car was theirs. This question also included a couple of statements regarding participants’ motivation in their main answer, to see if people responded as they thought a majority of people would, and to see if they were unwilling to pay taxes. Next, a version of the main question was asked that was rephrased to not be directly about the participants’ actions, as a way of making it easier for them to tell the truth. Then participants were asked how often they thought reporting as a work trip is justifiable, mirroring the way of questioning on the World Values Survey, a prominent instrument for reporting values, to compare their responses to the much larger sample that was used for this survey. Finally participants were asked to fill out the Moral Identity Scale, as we expect Moral Identity to confound with the responses to the main question (MIS, Aquino & Reed, 2002). See appendix 1 for the exact contents of the survey. Data analysis Analyses were run in JASP (JASP Team, 2016) and SPSS (IBM Corp, 2013). Likert-scales were treated as interval data as per Norman (2010). According to this same study, parametric tests are in fact very robust, even when data are non-normal or do not have homogeneity of variance.
All scores were recoded so that a higher score meant that more of the intended concept was measured. The MIS score was subsequently calculated by averaging over all items of the scale. Manipulation check To determine if the treatment had the intended result, t-tests were used to compare the various manipulation check questions over conditions. Compare group response to literature To see if the group reacted as expected, we looked at how justifiable the act of reporting as a work trip was deemed, and if certain outcomes were comparable over groups, such as the MIS score. Treatment effect To determine the effect of the treatment on the main outcome variable, likeliness of reporting as a work trip, an ANCOVA was used with the MIS score as a covariate. Exploratory tests Various t-tests and regressions were performed to get additional insight into what motivated the answers of the participants. This was done for the sample as a whole, but also for the two treatment groups separately. Results Of the 120 participants who fully completed the survey, 59 had been assigned the Pay upfront condition, 61 the Pay-in-the-end condition. Manipulation check On average, participants felt more like the car was theirs under the pay upfront scenario (M = 43.90, SE = 1.71) than under the pay at the end scenario (M = 2.18, SE = 1.30). This difference was significant t(118) = -6.215, p<.001. Also, participants felt more on average like they were allowed to drive the car for private purposes under the pay upfront scenario (M = 3.60, SE = 1.84) than under the pay at the end scenario (M = 2.64, SE = 1.58). This difference was significant t(118) = -2.940, p=0.004.
Figure 2 - Comparison of participants to US population on judging cheating on taxes or cheating on taxes by reporting as a work trip. Figure 3 - Comparison of responses to the main outcome measure in studies 1 and 2 0 10 20 30 40 50 60 70 80 % v al id r ep or ts
"Do you think * can always be justiLied,
never be justiLied, or something in
between?"
"cheating on taxes if you have a chance" (WVS - USA, n=2232, M=1,91, SD=1,8) "reporting this trip as a work trip" (Study 1, n=120, M=7,97, SD=2,32) "cheating on taxes if you have a chance (Study 2, n=120, M=7,49, SD=2,62) 0 5 10 15 20 25 30 35 40 45 50 % o f r esp on se s"How likely is it that you will report this
trip as a work trip?"
Study 1 Study 2Comparison of group response to literature The results from the question where it was asked to judge reporting as a work trip match the USA results from a similar question that was asked in the World Values Survey of 2010-2014 quite well, considering the experimental context (see Figure 2). 25% of participants reported it was slightly, moderately or extremely likely for them to report as a work trip (see Figure 3). Treatment effect On average, participants were not significantly more likely to report as a work trip under the pay at the end scenario (M = 2.56, SE = 1.96) than under the pay upfront scenario (M = 2.81, SE = 2.08). The covariate used in the ANCOVA, Moral Identity, was significantly related to the self-reported likeliness of reporting as a work trip, F = 3.97, p=0.049. There was not significant effect of the treatment on reporting as a work trip after controlling for the effect of Moral Identity, F = 0.02, p=0.90. No interaction effect between treatment and MIS score was found, F0.047, p=0.83. Third person perspective On average, participants reported that they could imagine someone else reporting as a work trip much more (M=5.63, SC=1.58) than themselves (M=2.68, SD=2.02). This difference was significant, t(119)=11.97, p=<.001. Also, there was no significant relationship between the likeliness of reporting as a work trip and being able to imagine someone would report as a work trip, r=-0.101, p=0.24. Discussion No main effect was found of the treatment on reporting as a work trip, even though it seemed to have manipulated sense of ownership succesfully. The nature of an online questionnaire is so that it is difficult to show a strong effect. Therefore, we attempted to improve on the effectivenes of the scenario in a second study by incorporating a concrete sum of money to increase salience and by simplifying the instructions on the presented regulations and options even further. Third person perspective The response on the question that was framed from a third person point of view gathered a very different response from the main outcome question. Without even a correlation between the two measures, it seems safe to say this result is not reflective of people’s reservation in revealing their true intention to us. Instead, judging from this result, we believe that quite a few participants misunderstood the question, and interpreted is as asking for their estimate of the likeliness the average person would report as a work trip. Therefore we dropped this question in the next iteration of the task. Additionally, we see the need to add another measure to increase our ability to see if loss aversion as a mechanism ties into the answers.
Study 2
Methods Participants Identical to study 1, 120 participants were recruited via Amazon’s Mechanical Turk marketplace using the TurkPrime platform to perform the scenario, restricted to be accessing MTurk from the United States of America. An informed consent was given before the experiment started and the Ethics Committee of CREED approved the study. Participants received a financial compensation. To be included, participants needed to finish the survey within 4 hours, the shortest setting possible in Qualtrics. Data collection was continued until 120 participants had finished the study. Procedure Also identical to study one, participants were directed towards a survey in Qualtrics, where the study was introduced and information was provided for reaching out to the researchers for any questions. Participants were randomly but evenly divided over the two scenarios. Participants were instructed to read the scenario and answer a verification question to make sure they understood the scenario. In a new window, they were presented with the problem. In yet a next window, they were asked how they would respond to the problem. The questionnaire then continued with various follow-up questions. Apart from the scenarios, all questions were exactly the same over conditions. Scenarios Similar to study one, in both conditions, participants were introduced to driving a company car and the related policy, which involved the duty to pay taxes if it was used for private purposes for more than 300 miles, in addition to work trips. Compared to study 1, the regulation was explained more clearly, making sure to state both benefits and drawbacks of the situation. Also, it was specified that the amount of taxes concerned was 3000.00 US Dollar. In the pay upfront condition, participants were presented with the lease program that included use the car for all private purposes, for which the tax for private use was withheld from the readers’ salary over the course of the year. However, if they wanted they could keep a journey registration and if at the end of the year, they had 300 miles or less for private purposes, they could ask for a refund of the tax money they had already paid. In the pay at the end condition, participants were presented with the lease program that they were not allowed to use the car for private purposes. This meant that the readers had to keep a journey registration. However, no tax was withheld from their salary. They could change their program at any time, to include private use of the car. If they saw they were going to cross 300 miles for private purposes, they would need to change their program online, and pay the tax for private use of the car for that year at the end of the year. At the end of each scenario, a verification question was asked to make sure participants understood the rules presented in their scenario. Participants could only continue after giving the right answer.Task Again, all participants were presented with the same problem, which stated that it is now October, and they have driven 280 miles for private purposes so far. They were told they had planned to go to a party, but going would mean crossing the 300 miles limit. They were informed they had the option of reporting this trip as a work trip without their employer knowing. Measures After the problem, participants were asked: “How likely is it that you will report this trip as a work trip?”, to which they could answer on a 7 point scale, ranging from “Extremely likely” to “Extremely unlikely”. Again, different items were included in a new question to check the manipulation effect, and see if participants felt like they were allowed to drive the car for private purposes, if they felt like the car was theirs, if they were motivated by not losing money or by not losing the ability to drive. However, in this second study, this question also included a couple of statements regarding participants’ attitudes towards money and towards driving a car, to see if loss aversion influenced their decision. These items were intended to measure money and driving attitudes, as was sucessfully done in Carmon & Ariely (2000) to get an indication of the influence of loss aversion on the overall result. In this study, students who were big fans of basketball were randomly assigned tickets for an important basketball game. Ticket owners, when asked if they were willing to sell their ticket, would name very high prices at which those without a ticket could buy it from them, as the prospect of losing the ticket seemed very painful to them. However, students who had not been assigned a ticket, when asked if they were willing to buy, focused on the prospect of loss of money and therefore named much lower prices that were acceptable for them to buy a ticket from the sellers. In addition to this, the researchers measured the attitudes towards basketball and money in the two groups of students. They found that within the ticket owner group, the selling price they named correlated to their basketball attitude, but not their money attitude. Conversely, in the money owner group, the buying price they reported correlated with their money attitude, and not their basketball attitude. This indicates that when determining a price for tickets, ticket owners focused more on their love of basketball, while money owners focused on the importance of saving money and the many other things they would be able to use it for. Thus, the endowment effect in this situation, shown by the difference in reported acceptable selling and buying prices, was shown to be motivated by loss aversion. With our scenario’s, it is uncertain if we can ask directly how participants value the ability to drive a lease car for private purposes, as with the assesment of acceptable selling and buying prices for the tickets. However, we can still measure attitudes towards driving and money. If loss aversion plays a role in the likeliness of participants to misreport, we may be able to find a negative correlation between the likeliness of misreporting and driving attitude in the pay upfront group, whereas we would expect to find a positive correlation between the likeliness of misreporting and money attitude in the pay in the end group (see Table 1).
Table 1 – Expected correlation of attitudes with the likeliness of reporting as a work trip under the different conditions
Report
Pay upfront only Report Pay in the end only
Money attitude No correlation Positive correlation
Driving attitude Negative correlation No correlation Whereas in the original study, these attitudes were measured with only one or two questions, in our survey we chose to add several items per attitude measure, inspired by the Osgood semantic differential (Osgood, 1957, Kervyn et al., 2013). Lastly, participants were asked to fill out the MIS; to judge how often they thought cheating on taxes if you have a chance is justifiable, exactly like the question that was asked of a much larger sample of the US population in the World Values Survey, a prominent instrument for reporting values; and to report their age, sex and estimated gross yearly income. See appendix 1 for the exact contents of the survey. Data analysis As in study one, analyses were run in JASP and SPSS, and likert-scales were treated as interval data. According to this same study, parametric tests are in fact very robust, even when data are non-normal or do not have homogeneity of variance. All scores were recoded so that a higher score meant that more of the intended concept was measured. The MIS score was subsequently calculated by averaging over all items of the scale. Also, ‘money attitude’ and ‘driving attitude’ were created by averaging the four related items for each measure. Manipulation check To determine if the treatment had the intended result, t-tests were used to compare the various manipulation check questions over conditions. Comparison of group response to literature To see if the group reacted as expected, we looked at the amount of fraud, how justifiable the act of reporting as a work trip was deemed, and if certain outcomes were comparable over groups, such as the MIS score. Treatment effect To determine the effect of the treatment on the main outcome variable, likeliness of reporting as a work trip, an ANCOVA was used with the MIS score as a covariate. Loss aversion To get an indication of the influence of loss aversion on participants’ decisions, a few approaches are possible. As in Carmon & Ariely (2000), regressions can be performed on the attitudes to see if money attitude correlated with higher likeliness of reporting as a work trip in the pay in the end condition, but not in de pay upfront condition and if, conversely, driving attitude correlated with lower likeliness of reporting as a work trip
in the pay upfront condition, but not in the pay in the end condition, indicating the manipulation changed people’s attitudes, theoretically because it determined people’s viewpoints. Additionally, we used t-tests to see if participants in the pay in the end condition report a higher motivation to not lose their money, whereas participants in the pay upfront condition report a higher motivation to not lose their ability to drive. Exploratory tests Various t-tests and regressions were performed to get additional insight into what motivated the answers of the participants. This was done for the sample as a whole, but also for the two treatment groups separately. Results In total, 121 participants fully completed the survey. Of those, one participant was excluded for reporting the middlemost response on most questions while finishing in 0 seconds. Of the remaining 120 participants (66 men, mean age: 35.66, SD = 10.89), 59 had been assigned the Pay upfront condition, 61 the Pay-in-the-end condition. Manipulation check On average, participants did not feel more like the car was theirs under the pay upfront scenario (M = 2.80, SE = 1.65) than under the pay at the end scenario (M = 2.84, SE = 1.92). This difference was not significant t(118) = 0.121, p=0.904. Also, on average, participants did not feel more like they were allowed to drive the car for private purposes under the pay upfront scenario (M = 2.97, SE = 1.884) than under the pay at the end scenario (M = 2.59, SE = 1.792). This difference was not significant t(118) = -1.120, p=0.265. Compare group response to literature This sample judged cheating on taxes differently from the USA results from the identical question that was asked in the World Values Survey of 2010-2014 (see Fig. 2). 32% of participants reported it was slightly, moderately or extremely likely for them to report as a work trip (see Fig. 3). Treatment effect On average, participants were not more likely to report as a work trip under the pay at the end scenario (M = 3.18, SE = 2.15) than under the pay upfront scenario (M = 3.44, SE = 1.93). The covariate in the ANCOVA, Moral Identity, was significantly related to the self-reported likeliness of reporting as a work trip, F = 27.23, p<.001. There was no significant effect of the treatment on reporting as a work trip after controlling for the effect of Moral Identity, F = 2.318, p=.131. No interaction effect between treatment and MIS score was found, F=2.583, p=0.111. On average, participants reported a higher likeliness of reporting as a work trip in study 2 (M=3.31, SD=2.04), than in study 1 (M=2.68, SD=2.02). This difference was significant, t(118)=-2.386, p=0.02.
Table 2 - Correlation of attitudes with the likeliness of reporting as a work trip under the different conditions
Report
Pay upfront only Report Pay in the end only Money attitude r=0.015, p=0.912 r=-0.433, p<0.001*** Driving attitude r=0.368, p=0.004* r=0.095, p=0.475 Loss aversion As shown in Table 2, the measured attitudes correlated with the likeliness of reporting as a work trip as expected, however money attitude was inversely correlated to likeliness of reporting as a work trip in the pay in the end condition, instead of a positive correlation as hypothesised. On the other hand, driving attitude was positively correlated to likeliness of reporting as a work trip, again the opposite of the inverse correlation that was expected. On average, participants were not more motivated to not lose money in the pay in the end condition as compared to the pay upfront condition, t(118) = 1.627, p=.106. Also, participants were not more motivated to not lose the ability to drive in the pay upfront condition as compared to the pay upfront condition, t(118) = 1.781, p=.077. Discussion Again, no main effect was found of the manipulation on the likeliness of reporting as a work trip. Contrary to the first study, the sense of ownership over the lease car was unaffected by the treatment, and participants reported to be significantly more likely to misreport.
Conclusion
Overall, manipulating the moment of tax payment was not found to influence taxpayers’ willingness to defraud their tax declarations. This is an unexpected find, as loss and gain framing is very well replicated. We did however find a consistent effect of Moral Identity on the self-reported likeliness to misreport, which at least indicates the outcome measure is indeed indicative of unethical behaviour. Also, our sample of participants seemed representative of the USA population, and we find a usable distribution of responses to the main outcome measure. The correlations on the loss aversion measures were in the expected place, but the opposite direction. However, this measure was intended to serve an explanatory value in relation to the main outcome, the absence of which renders them irrelevant. Methodological limitations The current research deviates from the literature in two important ways: Firstly, the actual experience of loss might be necessary for this mechanism to work. Even though we tried to make the scenario as salient as possible, it can still be that without actual consequences to the decision and no real experience of driving the car, it was not possible to establish a representative reference point with the participants. Secondly, the example of lease car driving is not as clear cut as the settings used inliterature. The regulation is very complicated, and although we skipped over quite some details in the scenarios used for this study, we cannot be sure all participants took away the same ideas of their possible options. This already carries important implications to take on board for future work towards implementation though, namely that when people try to understand the regulations and administrative requirements and possibilities, the large amount of information presented provides the opportunity for different people to frame the situation completely differently. On a related note, real-life tax payers may not feel as much opportunity to choose and may instead just do as they are told by employers, tax advisors or friends. Furthermore, participants who were in the pay upfront condition, and thus free to decide if they wanted to keep a journey registration, were presented with the information that they had driven 280 miles so far. As they would not know this apart from keeping a journey registration, and it is highly unlikely to achieve this low amount of use when driving the car freely for private purposes, this could be interpreted as evidence of a prior commitment to try keep below 300 miles of private use in order to apply for a tax return at the end of the year. If some participants indeed perceived the information in this way, this might put these participants, who were in the pay upfront condition, in a different mindset from what was intended. In particular, as it is already October in the problem, they might feel like they are very close to achieving the goal of reporting for a tax return. Goal setting, especially timed performance goals such as the current example, have been shown to increase fraud as people try to meet goals in any way possible, even if this includes bending the rules (Ordóñez & Welsh, 2015). However, such behaviour would still classify as ‘cheating to gain’, so if participants would indeed interpret the problem differently in this way, this would merely make it harder to show a main effect. Another factor that might have influenced the results is our way of measuring fraudulent behaviour, as the given scenario could have left participants with many reactions. It was suggested in the scenario to report the private trip as a work trip, thereby defrauding one’s journey registration. However, other ways of solving the problem in an unethical way are possible. Most prominently among those is to try and roll back the odometer in the car, so that the kilometres do not need to be accounted for in the journey administration, or to bribe a technician to do so. Again, the difference between real world situations and an online scenario might be large here, and in real life, we expect more diversity in how confident people feel about their fraud and how easily this is accomplished. Going forward, it is worthwhile to further explore the effect of loss and gain framing in the specific situation of lease car driving, as there is a large potential impact on policy, if only because any simplification of tax regulation is of such importance for compliance. Even though it may be difficult to research this specific group of regulations in online populations, this research can be a valuable contribution in fine-tuning a paradigm that will allow us to work towards a field experiment with actual large groups of taxpayers, it is important to know exactly which questions to ask. Implementing a pay upfront regulation could possibly save Tax Authorities, and thus society, a lot of money and administrative costs, not only because it could increase compliance, but also because it would be easier for taxpayers to understand and for administrators to explain it to them.
Furthermore, encouraging moral behaviour in a small part of the tax declaration process can contribute to a habit of moral behaviour for a large group of people. One interesting suggestion would be to test a scenario like this with an actual sample of lease car drivers, among which declaration holders, as a lab-in-the-field equivalent of the current research. They are certain to have the relevant reference point. Another possibility may be to develop a simulation in the form of a game, where the experience of ownership of the car and freedom to drive is made salient as people can freely drive around to explore a game world. If this driving is intrinsically rewarding, which could be influenced by giving people small quests to for example drive to places where they can personalise their vehicle, it may be better possible to put the players in a position where they want to keep driving, even though they might cross a limit of their set amount of distance. Finally, we recommend considering again the measure of implied loss aversion, used in study 2, as it could not be put to use in the current research. It will be interesting to see if the results from Carmon and Ariely could be replicated in any way, and if loss aversion can be shown to influence tax evasion in this situation. It will be interesting to see what it will take for people to bend the rules just a bit, and, maybe even more so, what will take away the opportunity for them to do so with a clean conscience. This knowledge could be of great help in creating a society with better incentives and regulations so that all of us can get a little closer to being the “good guys” we want to be.
References
• Ayal, S., & Gino, F. (2011). Honest rationales for dishonest behavior. The social psychology of morality: Exploring the causes of good and evil. Washington, DC: American Psychological Association, 149-66. • Beaussart, M. L., Andrews, C. J., & Kaufman, J. C. (2013). Creative liars: The relationship between creativity and integrity. Thinking Skills and Creativity, 9, 129-134. • Belastingbelangen.nl (2010). Veel fraude met bijtelling privé-gebruik auto. Retrieved from: http://www.belastingbelangen.nl/artikel/Veel_fraude_met_bijtelling_prive-gebruik_auto on July 6th, 2017 • Belastingbelangen.nl (2011). Privégebruik auto: € 15 miljoen boetes en naheffingen. Retrieved from http://www.belastingbelangen.nl/artikel/Privegebruik_auto_Euro_15_miljoen_boete s_en_naheffingen/?sqr=fraude%20privegebruik& on July 5th, 2017 • Belastingdienst.nl. U gaat met een auto van uw werkgever naar uw werk. Retrieved from https://belastingdienst.nl/wps/wcm/connect/bldcontentnl/belastingdienst/prive/ auto_en_vervoer/u_reist_naar_uw_werk/auto_van_uw_werkgever/, on July 5th, 2017. See also: Wet op de loonbelasting, 1964, Hoofdstuk 2, Artikel 13bishttp://wetten.overheid.nl/BWBR0002471/2017-04-01#HoofdstukII_Artikel13bis Consulted on July 5th, 2017; Uitvoeringsbesluit loonbelasting, 1965, Hoofdstuk 3, Artikel 8, 9 http://wetten.overheid.nl/BWBR0002489/2017-07-01#Hoofdstuk3_Artikel8, Consulted on July 5th, 2017; Uitvoeringsregeling loonbelasting, 2011, Hoofdstuk 3, Artikel 3.13 http://wetten.overheid.nl/BWBR0028236/2017-04-01#Hoofdstuk3_Artikel3.13, Consulted on July 5th, 2017; Wet inkomstenbelasting, 2001, § 3.2.2, Artikel 3.20 http://wetten.overheid.nl/BWBR0011353/2017-07-01#Hoofdstuk3_Afdeling3.2_Paragraaf3.2.2_Artikel3.20, Consulted on July 5th, 2017. • Berentsen, L. (2017). Leaserijder controleren met camerabeelden mag niet. Het Financiele Dagblad, February 25th, p. 7. • Besteman, T. (2009). Fraude met leaseauto. De Telegraaf , December 11th; retrieved from http://www.telegraaf.nl/dft/geld/belasting/20524772/__Fraude_met_leaseauto__.ht ml on February 21st, 2017 • BIT Staff, 2013. Behavioural Insights Tax Trials Win Civil Service Award. Retrieved from http://www.behaviouralinsights.co.uk/tax/behavioural-insights-tax-trials-win-civil-service-award/ on July 5th, 2017 • Camerer, C. F. (2000). Prospect Theory in the Wild: Evidence from the Field. Advances in Behavioral Economics, 148. • Carmon, Z., & Ariely, D. (2000). Focusing on the forgone: How value can appear so different to buyers and sellers. Journal of consumer research, 27(3), 360-370. • CBS (2017). Jaarmonitor Wegvoertuigen: Aantallen. Retrieved from https://www.cbs.nl/nl-nl/achtergrond/2017/21/jaarmonitor-wegvoertuigen-aantallen on July 7th, 2017. See also: CBS.nl, Personenauto’s. Retrieved fom https://www.cbs.nl/nl- nl/dossier/transport-en-mobiliteit/hoofdcategorieen/vervoermiddelen/categorie-vervoermiddelen/personenauto-s on July 7th, 2017 • Grolleau, G., Kocher, M. G., & Sutan, A. (2014). Cheating and loss aversion: do people lie more to avoid a loss?. • Hochman, G., Ayal, S., & Ariely, D. (2014). Keeping your gains close but your money closer: The prepayment effect in riskless choices. Journal of Economic Behavior & Organization, 107, 582-594. • IBM Corp. Released 2013. IBM SPSS Statistics for Windows, Version 22.0. Armonk, NY: IBM Corp. • JASP Team (2016). JASP (Version 0.8.0.0)[Computer software]
• Kahneman, D., Knetsch, J. L., & Thaler, R. H. (1990). Experimental tests of the endowment effect and the Coase theorem. Journal of political Economy, 98(6), 1325-1348. • Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica: Journal of the econometric society, 263-291. • Kern, M. C., & Chugh, D. (2009). Bounded ethicality: The perils of loss framing. Psychological Science, 20(3), 378-384. • Kervyn, N., Fiske, S. T., & Yzerbyt, V. Y. (2013). Integrating the stereotype content model (warmth and competence) and the Osgood semantic differential (evaluation, potency, and activity). European journal of social psychology, 43(7), 673-681. • Litman, L., Robinson, J., & Abberbock, T. (2016). TurkPrime.com: A versatile crowdsourcing data acquisition platform for the behavioral sciences. Behavior Research Methods, 1 - 10. • Luyendijk, J. (2016). Dit kan niet waar zijn: onder bankiers. Atlas Contact, Uitgeverij. • Mazar, N., Amir, O., & Ariely, D. (2008). The dishonesty of honest people: A theory of self-concept maintenance. Journal of marketing research, 45(6), 633-644. • Norman, G. (2010). Likert scales, levels of measurement and the “laws” of statistics. Advances in health sciences education, 15(5), 625-632. • Ordóñez, L. D., & Welsh, D. T. (2015). Immoral goals: how goal setting may lead to unethical behavior. Current Opinion in Psychology, 6, 93-96. • Osgood, C.E, Suci, G., & Tannenbaum, P. (1957). The Measurement of Meaning. University of Illinois Press; via McLeod, S. A. (2009) Attitude Measurement.
Retrieved from www.simplypsychology.org/attitude-measurement.html on July 5th, 2017. • Qualtrics software, Versions January-July 2017 of Qualtrics. Copyright © 2017 Qualtrics. Qualtrics and all other Qualtrics product or service names are registered trademarks or trademarks of Qualtrics, Provo, UT, USA. http://www.qualtrics.com • Schindler, S., & Pfattheicher, S. (2017). The frame of the game: Loss-framing increases dishonest behavior. Journal of Experimental Social Psychology, 69, 172-177. • Shalvi, S., Dana, J., Handgraaf, M. J., & De Dreu, C. K. (2011). Justified ethicality: Observing desired counterfactuals modifies ethical perceptions and behavior. Organizational Behavior and Human Decision Processes, 115(2), 181-190. • Shalvi, S., Gino, F., Barkan, R., & Ayal, S. (2015). Self-serving justifications doing wrong and feeling moral. Current Directions in Psychological Science, 24(2), 125-130.
• Solnah (2013). Hoeveel netto bijtelling betaal ik? Retrieved from http://auto-en-vervoer.infonu.nl/auto/119241-hoeveel-netto-bijtelling-betaal-ik.html, on July 5th, 2017. • Thaler, R. (1980). Toward a positive theory of consumer choice. Journal of Economic Behavior & Organization, 1(1), 39-60. • Thaler, R.H., & Sunstein, C.R. (2009). Nudge: Improving Decisions about Health, Wealth, and Happiness. (2nd edition). Penguin Books. • Tversky, A., & Kahneman, D. (1992). Advances in prospect theory: Cumulative representation of uncertainty. Journal of Risk and uncertainty, 5(4), 297-323.
Welcome
Dear participant,
Thank you for participating in our study. We expect that it will take about
10 minutes. Remember that there are no right or wrong answers. For your
participation you will receive $0.25 after you have completed the survey.
Any information that is obtained during this research study will be
processed anonymously and will be treated confidentially. Moreover,
participation in this research study is strictly voluntary. You may withdraw
your participation at any time you choose without cost.
If you have any comments or questions related to the study, feel free to
contact us.
Please close the web page if you do not want to participate. If you
complete the survey, we understand that you agree to participate.
Yours sincerely,
S.J. van Baardewijk-Berg, University of Amsterdam
(Suzanna.Berg@student.uva.nl)
Instructions
We would like you to imagine yourself as an employee at a company.
Please read the following information carefully.
Scenario 1
Qualtrics Survey Software https://eu.qualtrics.com/ControlPanel/Ajax.php?action=GetSurvey...
You work at a reasonably large company and you have no plans of leaving.
Every day you drive to and from your work in a company car that has been
provided to you.
The organization’s policy is “This is a work car!” which means:
For work related trips:
You are allowed to drive this company car for all work related trips.
For private trips:
When it comes to private trips, the rules are as follows:
If you drive 300 miles or less per calendar year for private trips, you
don’t have to pay a thing
If you drive more than 300 miles per calendar year for private trips,
you have to update the organization and count 25% of the company
car’s worth towards your income for that year. This means that you
will have to pay taxes on the use of the car.
The rules are simple, you must keep track of how much you use the car
for private trips. If you see that you are going to cross the 300 mile limit,
you must send an online request to update the organization and count
25% of the company car’s worth towards your income for that year.
Sending the online requests means you will have to pay taxes on the car
at the end of the year.
To verify you understood please answer the following question.
If I cross 300 miles for private use:
Scenario 2
Nothing happens
I have to count 25% of the company car’s worth towards my income for that year
You work at a reasonably large company and you have no plans of leaving.
Every day you drive to and from work in a company car that has been
provided to you.
The organization’s policy is “This is your car!” which means:
For work related trips:
You are allowed to drive this company car for all work related trips.
For private trips:
You are also allowed to drive this company car for all private trips, and
25% of the company car’s worth counts towards your income. This means
that you pay taxes on the use of the car.
However, if you drive 300 miles or less per calendar year for private trips,
you can update the organization and then you do not have to pay a thing.
The rules are simple, you can keep track of how much you use the car for
private trips. At the end of the year, if you see that you have not crossed
the 300 mile limit, you can send an online request to change the program
you are enrolled into which allows you not to count 25% of the company
car’s worth towards your income for that year. Sending the online requests
means you will receive the taxes back that you have payed on the car over
the course of the year.
To verify you understood please answer the following question.
If I don’t cross 300 miles for private use:
Nothing happens, I have to count 25% of the company car’s worth towards my income for that year
If I sent the online request, I do not have to pay taxes on the car and will receive the tax money I have already payed back
Problem
It is October, and you have driven 280 miles for private use so far this year.
This means that if you want to stay within the 300 mile limit for this year,
you will have to drive 20 miles or less till December 31st. Coming
weekend, you had planned to visit your brother's party. However, he lives
20 miles away from your place. So using the car for this private purpose
will lead you to cross the 300 mile limit.
You can also report this trip as a work trip, without your employer
knowing. Sporadic tax audits by the government could get you fined for
fraud though.
Outcome
How likely is it that you will report this trip as a work trip?
Secondary Question 1
Extremely likely Moderately likely Slightly likely Neither likely nor unlikely Slightly unlikely Moderately unlikely Extremely unlikelyTo what extent do you agree with the following statements?
Secondary Question 2
Strongly agree Agree Somewhat agree Neither agree nor disagree Somewhat disagree Disagree I experienced the previous question as an ethical decision I felt like I was allowed to drive the car as much as I likedI felt like the car was mine I did not want to pay taxes for the private use of the car I wanted to do the same as the majority of people would do
Can you imagine someone would report this trip as a work trip?
Secondary Questions 3
Do you think reporting this trip as a work trip can always be justified, never
be justified, or something in between?
Moral Identity Scale: Introduction
Here are some characteristics that might describe a person:
Caring, Compassionate, Fair, Friendly, Generous, Helpful, Hardworking,
Honest, Kind
The person with these characteristics could be you or it could be someone
else. For a moment, visualize in your mind the kind of person who has
these characteristics. Imagine how that person would think, feel, and act.
When you have a clear image of what this person would be like, answer
the following questions.
Moral Identity Scale
Extremely well Not at all Always justifiable Never justifiable
Here are some characteristics that might describe a person:
Caring, Compassionate, Fair, Friendly, Generous, Helpful, Hardworking,
Honest, Kind
The person with these characteristics could be you or it could be someone
else. For a moment, visualize in your mind the kind of person who has
these characteristics. Imagine how that person would think, feel, and act.
When you have a clear image of what this person would be like, answer
the following questions.
Strongly agree Agree Somewhat agree Neither agree nor disagree Somewhat disagree Disagree It would make me feel good to be a person who has these characteristics Being someone who has these characteristics is an important part of who I am. I often wear clothes that identify me as having these characteristics. I would be ashamed to be a person who had these characteristics. The types of things I do in
Powered by Qualtrics
Code & thanks
Thank you for participating!
The survey confirmation code is: SB955
In order to be paid, please enter this code into MTurk.
Please press the ">>" button in order to finish the survey.
Welcome
Dear participant,
Thank you for participating in our study. We expect that it will take about 7
minutes. Remember that there are no right or wrong answers. For your
participation you will receive $0.25 after you have completed the survey.
Any information that is obtained during this research study will be
processed anonymously and will be treated confidentially. Moreover,
participation in this research study is strictly voluntary. You may withdraw
your participation at any time you choose without cost.
If you have any comments or questions related to the study, feel free to
contact us.
Please close the web page if you do not want to participate. If you
complete the survey, we understand that you agree to participate.
Yours sincerely,
S.J. van Baardewijk, University of Amsterdam
(Suzanna.Berg@student.uva.nl)
Instructions
We would like you to imagine yourself as an employee at a company.
Please read the following information carefully.
Instructions 2
Qualtrics Survey Software https://eu.qualtrics.com/ControlPanel/Ajax.php?action=GetSurvey...
You work at a reasonably large company and you have no plans of leaving.
Every workday you drive to and from your work in a company car that has
been provided to you. You have no other car.
The company car can be used freely for all work related trips.
For private trips, you have to pay a fixed amount of tax, depending on if
you cross a limit of 300 miles:
More than 300 miles
If you drive more than 300 miles a year for private purposes, you will have
to pay $3000 in taxes for the private use of the car.
300 miles or less
If you drive 300 miles or less a year for private purposes, no taxes have to
be paid on the private use of the car.
Scenario 1
With your current lease program, you are not allowed to use the car for
private purposes.
What this means for you:
- You may not drive the car for more than 300 miles per year for private
trips.
- To prove this you must keep track of all your work and private trips of
the year.
- You don’t have to pay the $3000 taxes for the private use of the car.
It is however possible for you to change your program, by filling out an
online form.
If you change your program to include private use of the car, you will have
to pay the $3000 taxes for the private use of the car at the end of the year.
Of course, you have to change your program before you drive more than
300 miles for private purposes.
To verify you understood please answer the following question.
If I cross 300 miles for private use:
Scenario 2
With your lease program, you are allowed to use the car for private
purposes.
What this means for you:
- You are allowed to drive the company car for all private trips.
- You pay $3000 taxes for the benefit of driving the car privately. Your
employer pays these taxes for you, witholding them from your salary.
However, if you instead only drive the car for private purposes for 300
miles or less, you can fill out an online form at the end of the year. You will
then receive the $3000 tax money that was withheld on your salary back.
Of course, this is only possible when you have not yet driven more than
300 miles for private purposes. To prove this you must keep track of all
work and private trips.
To verify you understood please answer the following question.
Nothing happensI have to update my program and pay $3000 taxes over the private use of the car at the end of the year
If I don’t cross 300 miles for private use:
Problem
It is October, and you have driven 285 miles for private use so far this year.
This means that if you want to stay within the 300 mile limit for this year,
you will have to drive 15 miles or less till December 31st.
Coming weekend, you had planned to visit your close friend’s party.
However, he lives 20 miles away from your place. So using the car for this
private purpose will lead you to cross the 300 mile limit.
You can also report this trip as a work trip, without your employer
knowing.
Outcome
How likely is it that you will report this trip as a work trip?
I don’t have to keep track of work and private trips.If I sent the online request at the end of the year, I will receive the $3000 tax money I have already payed back.
Extremely likely Moderately likely Slightly likely Neither likely nor unlikely Slightly unlikely Moderately unlikely Extremely unlikely
Secondary Question 1
To what extent do you agree with the following statements?
Strongly agree Agree Somewhat agree Neither agree nor disagree Somewhat disagree Disagree Strongly disagree I do not often do things in order to save money I often drive a car Driving a car is a negative thing to do I feel strongly about saving money I am not concerned with money compared to others I am not an avid car driver compared to others Saving money is a positive thing to doTo what extent do you agree with the following statements, concerning the
task?
Strongly agree Agree Somewhat agree Neither agree nor disagree Somewhat disagree Disagree Strongly disagree I felt like I was allowed to drive the car as much as I liked I felt like the car was mine I felt like I did not want to loose my money I felt like I did not want to loose the ability to drive my carMoral Identity Scale: Introduction
Here are some characteristics that might describe a person:
Caring, Compassionate, Fair, Friendly, Generous, Helpful, Hardworking,
Honest, Kind
The person with these characteristics could be you or it could be someone
else. For a moment, visualize in your mind the kind of person who has
these characteristics. Imagine how that person would think, feel, and act.
When you have a clear image of what this person would be like, answer
the following questions.
Moral Identity Scale
Here are some characteristics that might describe a person:
Caring, Compassionate, Fair, Friendly, Generous, Helpful, Hardworking,
Honest, Kind
The person with these characteristics could be you or it could be someone
else. For a moment, visualize in your mind the kind of person who has
these characteristics. Imagine how that person would think, feel, and act.
When you have a clear image of what this person would be like, answer
the following questions.
Strongly agree Agree Somewhat agree Neither agree nor disagree Somewhat disagree Disagree It would make me feel good to be a person who has these characteristics Being someone who has these characteristics is an important part of who I am. I often wear clothes that identify me as having these characteristics. I would be ashamed to be a person who had these characteristics. The types of things I do in
WVS comparison
Do you think cheating on taxes if you have a chance can always be
justified, never be justified, or something in between?
Demographics
What is your year of birth?
What is your sex, as listed on your passport?
Always justifiable Never justifiable Male Female
Powered by Qualtrics
Information about income is very important to understand. Would you
please give your best guess?
Please indicate the answer that includes your entire household income in
2016 before taxes.
Code & thanks
Thank you for participating!
The survey confirmation code is: TU273
In order to be paid, please enter this code into MTurk.
Please press the ">>" button in order to finish the survey.
Less than $10,000 $10,000 to $19,999 $20,000 to $29,999 $30,000 to $39,999 $40,000 to $49,999 $50,000 to $59,999 $60,000 to $69,999 $70,000 to $79,999 $80,000 to $89,999 $90,000 to $99,999 $100,000 to $149,999 $150,000 or more
