ABSTRACT. How did the availability of a vaccine change the lives of people who had been restricted by the COVID-19 pandemic for a year? This question may seem too simple to warrant analysis. Because vaccination is one of the few means of preventing viral infection, the availability of a vaccine should cause people to feel relief. The proven safe and effective vaccine against the COVID-19 virus is an important tool for establishing herd immunity against it. Yet, worldwide, people have been hesitant or have outright refused to be vaccinated against COVID-19.

A vaccination program has two routes through which it affects human life—through the individual and through the population: When an individual is vaccinated, it changes that person’s assessment of their own safety. On the population level, the percentage of a population that is vaccinated affects the safety and activity of the people, which in turn affects people’s judgment, emotions, and daily life. This study investigates the effects of a vaccination program on these two levels. We are not aware of previous studies that have taken this approach. Specifically, we estimate the effects of a person being vaccinated and the vaccination rate in Israel on the person’s assessment of their chances of being infected with COVID-19, how symptomatic they would expect to be, their emotions, their expectations of future spread of COVID-19, and their economic expectations—both for their personal income and for Israel’s gross domestic product (GDP). Understanding these effects is important when designing communication strategies to address vaccine hesitancy and to foster vaccine uptake among the population. In addition, the current study is unique in its use of panel data from seven surveys administered between March 2020 and March 2021. At the time of the last survey, more than half of Israel’s population had been vaccinated following a rapid vaccination program, the first in the world, while uncertainty about the vaccine and the program was high.

We used a fixed effect model to analyze the questions at hand and found a positive effect for the vaccination program. We found that a person’s own vaccination affected both their assessment of their own safety and their emotions but did not affect their expectations regarding the situation in Israel as a whole. In contrast, the vaccination rate in Israel affected respondents’ expectations about the spread of COVID-19, their own income, and Israel’s GDP but did not affect their assessment of their own safety or their emotions. These results suggest that most Israeli people evaluated vaccination rationally and that they were relieved by the availability of a vaccine for themselves and when they considered the level of vaccination of the population as a whole. However, we found that religious level matters: Those who were more religious were less relieved by the level of vaccination in Israel.

The findings can serve as an example of the importance of vaccination for emotional and cognitive relief on the individual level; they also show that a high level of vaccination in the population brightens people’s economic expectations, which are important to the recovery of economic activity.

ABSTRACT. Introduction From the start of the COVID-19 pandemic countries have taken severe measures, including economic lockdowns, to curb the disease and to limit the number of hospital cases. We aim to provide insight into these effects by using a unique Dutch panel data set including consumers’ economic perceptions and expectations, feelings, and social interactions.

Theory Van Raaij and Gianotten (1990) showed that in the Netherlands, apart from the income, it were mainly consumer opinions about their personal or household economic situation that explained expenditures, credit and saving, not their opinions about the general economic situation. To the extent that the savings increase in 2020 is due to precautionary saving, and the inability to spend money during the lockdown period, COVID-19 might have been a factor explaining this behavior. An economically relevant indicator of financial perceptions is income evaluation. Van Praag and Frijters (1999) report on a survey question on what income people find good, sufficient, or bad. Possibly, income evaluation may depend on whether one believes more or less income is needed in a lockdown situation. The uncertainty experienced during the COVID-19 crisis might be associated with lower levels of happiness (Helliwell et al., 2021). The lockdown measures largely disrupted people’s opportunities to maintain social relations and interactions outside their households. The ability to maintain social relations tended to buffer the negative effects of the lockdown restrictions on mental well-being in the U.S. (Laurence and Kim, 2021). People in Belgium indicated that they consumed slightly more alcohol and increased smoking during the lockdown (Vanderbruggen et al., 2020). The impact of COVID-19 on the behavior of people depends very much on their environment, for example, whether they live in more urban or rural areas, or in heavily infected areas.

Method Sample We use data from the DNB Household Survey (Teppa and Vis, 2012) including roughly 2500 households. For our analysis, we assign responses from weeks 10–22 in 2020 to the lockdown period, and any other responses to the period thereafter. The last survey period was week 42, before any new restrictive measures were applied. Measures We included measures of the following indicators. Finance: Household financial situation, combining income management, saving and debts, and expectations; Financial flexibility, i.e., the easiness of reducing household expenditures; Household saving opinions; Relative standing, comparisons with other households; Income evaluation, i.e., what household income was deemed sufficient; Expected housing price changes in the next 2 years; Percentage of expected housing price increase in 10 years; Percentage of consumer price increase in 1 year. Emotions: Average of reported feelings of calm and happy; Average of reported feelings of mood swings, stress, anxiety, feeling down, depressed. We created two variables to capture the constructs positive emotions and negative emotions. Social relations: Average of reported sympathizing with and taking time for others; Average of reported interactions and cooperation with others at work. Health: General health; Health change in the past year. Smoking cigarettes; Drinking alcohol. Residence: Whether participants lived in cities or densely populated areas; Living in Brabant, i.e., the province in which the COVID-19 outbreak started.

Estimation strategy We estimated the influence of the lockdown period on our measures by specifying a lockdown*year variable as the difference-in-difference estimator (DiD), comparing lockdown with non-lockdown periods and years 2019 and 2020. Similar as in Greyling (2021a, 2021b) we assume that the year 2019 provides a true counterfactual for the 2020 levels, such that the dependent variable follows the same trend in both years.

Results Both the perceived household financial situation and financial flexibility were positively related to the lockdown. In addition, household saving and relative standing during the lockdown were more common in larger cities than in rural areas. The expectations concerning an increase of housing prices in the next two years were negatively related to the lockdown. However, in ten years the percentage of expected housing price increase was unrelated to the lockdown. These results point to the general expectation of a temporary fall and subsequent recovery of housing prices thereafter. During the lockdown period higher consumer prices were expected in the next twelve months than in other periods.

The lockdown resulted in lower levels of positive emotions across the sample. Overall happiness and negative emotions did not change. Lower levels of happiness were observed, during the lockdown, of people living in urban areas. During the lockdown period people in Brabant felt less socially aware, possibly because in Brabant the lockdown was much stricter than in the rest of the Netherlands, and also mortality rates were very high. In the lockdown, people from urban regions reported a drop in social interactions with others at work. General health condition, health change, smoking, and drinking were all unrelated to the lockdown.

Discussion Our study has focused mainly on perceptions and experiences rather than on observed behavior. Since the lockdown was relatively short (about three months), only transitory thoughts and feelings seemed to have been affected, i.e., price perceptions, household financial management, emotions, and social relations. We did not find evidence for more long-lasting effects, for example, on savings, perceived health, and (un)healthy behaviors, although such effects might have occurred later during the pandemic. We have used the measures from 2019 as counterfactual of what happened in 2020, i.e., of what might have happened in 2020 without the COVID-19 pandemic. The comparison of background variable distributions in the two years seems to mainly support this assumption. However, a longer time frame, both with and without the pandemic, would give a firmer basis of analysis.

Notes The full paper has been published in Frontiers in Sustainable Cities, 2022(4), Article 860151. In this paper use is made of data of the DNB Household Survey administered by Centerdata (Tilburg University, The Netherlands).

ABSTRACT. Research shows that the advocacy of materialism, understood as beliefs that link wealth and consumption with personal achievement and happiness (Richins & Dawson, 1992), is influenced by higher media consumption (e.g., Kasser, Ryan, Couchman, & Sheldon, 2004; Shrum, Burroughs, & Rindfleisch, 2005), personal and social insecurities (e.g., Pieters, 2013), and adverse emotions (e.g., Donnelly, Ksendzova, Howell, Vohs, & Baumeister, 2016). Moreover, it has been suggested that in consumer-oriented societies, mortality reminders intensify materialistic orientations and desires (Arndt, Solomon, Kasser, & Sheldon, 2004). Lockdown restrictions introduced in most countries to control the spread of the COVID-19 virus in early 2020 substantially altered these factors. For example, early figures suggest that people spent 20% more time watching broadcast TV and 27% more time on streaming platforms (GlobalWebIndex, 2020). Moreover, across different countries including the United Kingdom (UK), there has been a substantial increase in reported rates of anxiety and depression (Commonwealth Fund, 2020), a rise in loneliness (ONS, 2021), and a decrease in well-being, happiness, and a sense of self-worth (Fujiwara et al., 2020). In addition, the COVID-19 outbreak has consistently featured in news headlines globally, with narratives around issues of “illness”, “death”, and “survival” making health and health-related behaviours a frequent topic of conversations. Given this context, it is possible that the societal and behavioural changes that “stay-at-home” restrictions have brought about, and the recurrent mortality reminders may be boosting the endorsement of materialistic values. However, no prior studies have looked at possible changes in materialistic values during the COVID-19 pandemic or have holistically looked at how behavioural and emotional changes experienced due to lockdown restrictions may be contributing to this possible rise in materialism. Consequently, the present research aims to identify the role of different behavioural and affective changes brought by the COVID-19 pandemic in the endorsement of materialism, and to examine possible changes in attitudes towards money. Study 1 looked at the effects that an increase in media consumption, social isolation, negative affect, stress and anxiety, and the perceived self-threat of COVID-19 have on an individual’s current endorsement of materialism. 741 UK residents were recruited through an online subject pool. Participants’ age ranged between 18 and 79 (M = 34.90, SD = 11.63) and 71.8% were females. We measured materialism using the 9-items Materialistic Value Scale (Richins, 2004) and asked participants to indicate if their media consumption had increased or decreased within the past six months, their perception of social isolation, negative affect, stress and anxiety with two, perception of COVID-19 as a life threat, and the financial impact of the pandemic. A hierarchical linear regression was performed with the materialistic values scale as the dependent variable, subjective economic status, age, gender, and the economic impact of COVID-19 as control, and the changes in the factors identified by prior literature as predictors. First, we introduced the participant’s demographic characteristics. This model accounted for 10.32% of the variance in materialism and revealed that only age was significant at predicting materialism, β = -.26, p < .001, bootstrapped 95% CI [-.02, -.01]. In the second step, we introduced the economic contextual effects. This variable was significant at predicting materialism, β = .13, p < .001, bootstrapped 95% CI [.03, .09] and showed that the second model explained 2.14% more variance. Finally, the variables looking at increases in media consumption, social isolation, negative moods, stress and anxiety, and the perception of COVID-19 as a life threat were introduced as predictors. The results revealed that only increases in media consumption, β = .15, p < .001, bootstrapped 95% CI [.06, .18], and in stress due to COVID-19, β = .14, p = .045, bootstrapped 95% CI [.00, .13], were significant predictors of materialism. This third model explained 4.18% the variance in materialism. Study 2 assessed possible value changes due to the COVID-19 pandemic and lockdown restrictions in the UK. T1 sample (N = 200) was composed of 69% female, age ranging from 18 to 67 years old (M = 24.95, SD = 7.97). T2 sample (n = 87) was composed of 71.3% female, age ranging from 18 to 67 years old (M = 26.36, SD = 9.45). A MANOVA was used to compare the participants that completed the questionnaire during Time 1 and Time 2 with the participants that only completed Time 1. This revealed non-significant differences in any of the scales and subfactors collected at Time 1 (all ps > .05). Participants’ goals were assessed using 18 items taken from the Aspiration Index (Kasser & Ryan, 1993). A series of paired samples t-test was used to assess the participants’ changes in the endorsement of different life goals between T1 and T2. The results revealed non-significant changes in the importance given to self-development, affiliation, community, image, and popularity (all ps > .05). However, there was significant difference in the importance that participants gave to the goal of financial success between T1 (M = 5.41, SE = 1.34) and T2 (M = 4.99, SE = 1.27), t(86) = 4.21, p < .001, Hedges’ g = .32, bootstrapped 95% CI [.21, .61]. Our findings suggest that during the COVID-19 pandemic, individuals have lessened the importance given to money despite a rise in the factors that have been found to facilitate the endorsement of materialism. We observe that changes in media consumption and stress and anxiety due to the pandemic predicted people’s current levels of materialism. However, contextual alterations to these factors due to the COVID-19 outbreak show a limited influence in the advocacy of materialistic values. Our work suggest that the COVID-19 pandemic has introduced changes in people’s attitudes towards money that could lead to deeper societal changes in a post-pandemic world. For example, aligned with our findings, it has been observed an increase in job resignations in the US and the UK, potentially fuelled by people’s reflections on their life priorities during the COVID-19 outbreak (Christian, 2021). Moreover, this lessened focus on money could also shift consumer behaviours approaches towards conspicuous consumption and luxury goods.

ABSTRACT. For people across the globe, accessing resources such as vaccinations and furlough payments has been key for managing the health and financial risks following the COVID-19 pandemic. However, getting access to these resources has often proved challenging. For instance, in some countries, the demand for COVID-19 vaccinations was much higher than the availability. COVID-19 created a unique case to evaluate the possibility that the greater the (perceived) risk magnitude posed by the pandemic, the more likely it could be that individuals would engage in unethical behaviours to gain access to prized resources (e.g., vaccination). More specifically, we postulate that individuals would be more likely to cheat about their health or financial status in order to improve the probability of obtaining, respectively, vaccinations or furlough payments. To study this important topic, we have created a bespoke scenarios based on the COVID-19 pandemic and conducted three experimental studies to evaluate whether objective risk, perceived risk and other factors would impact participants’ intention to be dishonest in order to improve their chances of accessing pandemic risk management resources. Study 1 (N = 302) evaluated the extent to which the objective and perceived risk of the COVID-19 pandemic influenced individuals to provide false information about their health status to access a limited supply of vaccination. In Study 2 (N = 201) we assessed whether the extent to which individuals provided false information varied when we manipulated the information that would help them access a vaccination or access to furlough payments. Finally, Study 3 (N = 270) evaluated whether the willingness of individuals to cheat in order to access furlough payments was influenced by the risk (probability) that their dishonesty could be detected by the government. Across all three studies, our results show that approximately one third of all participants would cheat to access vaccinations or furlough payments. Study 1 found that the perceived risk, but not the objective risk, of the disease significantly predicted dishonesty when attempting to access vaccinations. While Study 2 found no difference in levels of dishonesty between those attempting to obtain a vaccination and those attempting to obtain furlough payments, the results showed that perceived risk significantly predicted dishonesty for vaccinations but not for furlough payments. Finally, Study 3 found a significant positive relationship between dishonesty and the risk of being detected. Taken together, our studies and results have important implications for ensuring that, during a pandemic, risk management resources can be allocated in a fair and judicious and that fraud can be reduced.

ABSTRACT. An upstart usually requires financing. The starting entrepreneur may, first, provide some own financing from personal savings. The entrepreneur may, second, also get some financing from the family. The entrepreneur may, third, also get some external financing, i.e. financing from outside the family, e.g. from banks.

Here we account for the financing of startups before and after the pandemic disruption in 2020, complementing a recent study of business angel funding (Tolba et al, 2021). Expectedly, the pandemic disruption entailed a decline in external financing for startups (Hypothesis 1). Therefore, expectedly, the pandemic disruption pressured the entrepreneur to increase proportion of own financing (Hypothesis 2). Furthermore, the pandemic disruption expectedly enhanced solidarity within families, and therefore we hypothesize that the pandemic disruption caused an increase in family financing relative to external financing (Hypothesis 3).

These hypotheses are here tested by analyzing startups in Spain before the pandemic (in 2019) and after the pandemic disruption (in mid-2020 and 2021), with a representative sample of 1,782 starting entrepreneurs, surveyed by the Global Entrepreneurship Monitor. The entrepreneurs were asked for the percentage of their financing that came from personal savings, the percentage from family, and the percentage from external sources.

Our results show an increase in own financing and decline in external financing, as hypothesized (H1 and H2).

The hypothesis that family financing increased relative to external financing, is tested by linear regression. Our results show an increase in entrepreneurs’ financing from family relative to external financing, as hypothesized.

In conclusion, the pandemic disruption changed financing of startups. Specifically, as hypothesized, the disruption caused a decrease in external financing, but an increase in proportion own financing, and also an increase in family financing relative to external financing.

Reference. Tolba, A., A. Ismail, and T. Schøtt. 2021. People financing entrepreneurs within and outside the family: Pandemic decline and resilience in cultures around the World. Journal of Risk and Financial Management, 14(12), 610.

ABSTRACT. Goldstein and Rothschild (2014, Judgement and Decision Making, hereafter GR) compared the accuracy of subjective probability distributions elicited via an interactive graphical interface with those elicited by stated descriptive statistics. They inferred from the greater accuracy produced by graphical elicitation that by eliciting descriptive statistics researchers underestimate the accuracy of laypeople’s mental representations, with potential implications for surveys and experimental measurement. If this interpretation is correct, then distributions derived by graphical elicitation should be better predictors of individuals’ decisions than stated descriptive statistics. We test this implication by eliciting distributions using the methods of GR and relating them to willingness to gamble based on the same distribution. We report two experiments, one using a student sample, the other a sample of consumers.

Experiment 1

A student sample (n=76) was shown a sequence of 100 random draws from a bag of balls (computerised) using the skewed distributions employed by GR. Distributions were elicited using a graphical interface and stated descriptive statistics, in counterbalanced order, with an incentive to be accurate. Participants were then invited to make four bets on a number to be drawn from the bag, two certainty equivalents and two probability equivalents. One of the four was chosen at random and executed for real.

In contrast to GR, among this incentivised student sample, there was no difference in accuracy between the stated statistics and the graphical interface. Moreover, mixed effects regression models using the repeated measures to control for individual level willingness to gamble revealed that the stated statistics had a stronger relationship with certainty and probability equivalents derived from real bets. The implication is that, at least for this relatively numerate sample, the stated statistics are more representative of the subjective probability distribution as it applies to decision making.

Experiment 2

The second experiment repeated the first with a sample of consumers (n=89), larger incentive for accuracy and a graphical interface that more closely matched that of GR. Two distributions of balls were employed, one skewed and one symmetric, presented in a pseudorandomised order. Four bets were undertaken based on each distribution and one was again executed at random.

The graphical interface was marginally more accurate than the stated statistics, although both were more accurate than in GR (who used a sample from Mechanical Turk). There was some evidence that this might have been caused by some respondents failing to understand the concept of an average. Despite marginally lower accuracy, the stated statistics again had a stronger relationship than the distribution on willingness to bet.

Conclusion

We found no evidence that individuals make decisions based on mental representations that are more accurate than implied by their stated descriptive statistics. Indeed, stated statistics were more strongly related to decision making than the equivalent statistics derived from a graphical elicitation of the subjective distribution. Taking our findings and those of GR together, one possibility is that the graphical interface generates superior accuracy, especially among those less numerate, because it aggregates multiple small decisions. If so, and in keeping with our findings on willingness to take bets, this does not imply that individuals have more accurate representations of subjective probability distributions than implied by their descriptive statistics. The graphical elicitation may give insight into the accuracy of people to reproduce rather than represent a probability distribution.

ABSTRACT. An important contribution of the bounded rationality approach to behavioural economics is recognizing that smart people do not have the capability to behave in a fashion recommended by conventional economics. This raises the question of what capabilities are required for individuals, households, firms, and governments to make the best possible decisions; what some might refer to as rational or boundedly rational decisions. I argue that one needs to model smart decisions and relatively smart decisions in terms of a decision-making capability frontier. This would allow us to discuss and measure the extent to which decisions are sub-optimal—effectively x-inefficiency in decision-making. This frontier is not the conventional economics’ one, well critiqued by scholars from a wide array of methodological perspectives. Rather, it is derived from an understanding of what human agents require to make smart (rational, optimal) decisions.

Unlike in the conventional economics wisdom, I do not assume that decision-makers are born, naturally endowed, or gifted with the capabilities to make optimal decisions. Therefore, this frontier is contextualized by context and by the objectives or end goals of the decision-making process. We built upon Sen and Nussbaum’s work on capabilities, x-efficiency theory (Leibenstein’s based firm-based model extended into the domain of choice behaviour), and choice x-inefficiency modelling that has its roots in conventional choice theory. We exemplify this approach discussing financial literacy, choice theory and firm decision-making. The capabilities approach is also contrasted to the nudging approach to decision-making wherein it is assumed that individuals tend to make relatively irrational decisions because the decision-making hardware with which we are endowed. This approach has psychological and neurological roots.

The capabilities approach developed here suggests poor decision-making and, consequently, poor or sub-optimal decisions have less to do with naivety, emotions, herding, and biases, and more to do with a capabilities gap that can be repaired through smart public policy. In this case, experts or choice architects, are not focused on nudging individuals to make decisions that they think are in the best interest of the individual. Rather, the point of focus is shifted to the realm of ‘institutional design’ wherein experts attempt to provide individuals with the capabilities to make smart decisions; decisions that they view to be in their best interest (that of the decision-makers) and that of their family and communities. This approach can also incorporate important cases of negative and positive externalities that are a consequence of individual decision-making which, as conventional economists would also maintain, soft or even hard nudging might be required to change the behaviour of the individuals, especially when such behaviour causes harm to others. This approach also speaks to the heuristics used by rational decision-makers, which can be either effective or ineffective given the capabilities of decision-makers and their decision-making environment.

References Altman M (2006) “What a difference an assumption makes: effort discretion, economic theory, and public polic.” In: Altman M (ed) Handbook of contemporary behavioral economics: foundations and developments. Armonk, New York, pp 125–164. Altman, M (2010) “A Behavioral and Institutional Foundation of Preference and Choice Behavior: Freedom to Choose and Choice X-inefficiencies,” Review of Social Economy 69: 395- 411. Altman M (2017) “A bounded rationality assessment of the new behavioral economics.” In: Frantz R, Chen S-H, Dopfer K, Heukelom F, Mousavi S (eds) Routledge handbook of behavioral economics. Routledge, London, pp 179–193. Altman, M (2019) “Implications of smart decision-making and heuristics for production theory and material welfare,” Mind and Society 18: 167-179. Altman M and Lamontagne L (2004) “Gender, human capabilities and culture within the household economy: Different path to socioeconomic well-being?” International Journal of Socio-Economics 31: 325–364. Berg N, Gigerenzer G (2010) “As-if behavioral economics: neoclassical economics in disguise?” History of Economic Ideas 18:133–166 Gigerenzer G (2007) Gut feelings: the intelligence of the unconscious. Viking, New York. Kahneman D (2003) “Maps of bounded rationality: psychology for behavioral economics,” American Economic Review 93: 1449–1475. Leibenstein H (1966) “Allocative efficiency vs. ‘x-efficiency’” American Economic Review 56: 392–415 Mousavi S (2018) “What do heuristics have to do with policymaking?” Journal Behavioral Economics for Policy 2:69–74. Nussbaum M (2003) “Capabilities as fundamental entitlements: Sen and social justice,” Feminist Economics 9: 33–59. Sen A (1987) Commodities and Capabilities. Oxford: Oxford University Press. Sen A (1999) Development as Freedom. New York: Knopf. Simon HA (1978) “Rationality as a process and as a product of thought,” American Economic Review 70:1–16. Simon HA (1986) “Rationality in psychology and economics,” Journal of Business 59:S209–S224. Smith VL (2005) “Behavioral economics research and the foundations of economics,” Journal of Socio-Economics 34:135–150. Thaler RH, Sunstein C (2008) Nudge: improving decisions about health, wealth, and happiness. Yale University Press, New Haven.

ABSTRACT. People like discounts and marketers know that. Temporarily lowered prices attract consumers to buy products or services they normally would not buy, or to buy more of them than they would normally buy. In this way, discounts are advantageous for both the consumer and for the businesses offering them. However, offering products or services on a discount may also have some serious drawbacks. For example, discounts may lead to a devaluation of the product or service (e.g., Arkes et al., 2002; Tybout, 1978). Moreover, missing the discount might lead to even more detrimental consequences in the long run, due to the phenomenon of inaction inertia: “bypassing an initial action opportunity decreases the likelihood that subsequent similar action opportunities will be taken” (Tykocinski et al., 1995; p. 793). For instance, imagine that last week you wanted to buy a ski pass at a reduced price, $40 instead of the regular $100. You somehow forgot about it and missed the opportunity. The ski pass is now offered for $90 instead of $100. Will you buy it? In Tykocinski et al.’s experiments, people were less likely to buy the $90 ski pass after missing the $40 one (i.e., with a large value difference between the two offers), than after missing a $80 discount (i.e., small value difference). Thus, consumers that did not buy the discounted product (inaction), were likely to refrain from buying the product later on (inertia). In marketing, inaction inertia might be one of the causes of the “post-promotion dip” in sales (e.g., Van Heerde et al., 2000) and can even make loyal customers switch to a different brand (Zeelenberg & Van Putten, 2005). Ample research has examined the inaction inertia effect and potential explanations for it (for reviews, see Van Putten et al., 2013a,b; for a large replication project on the original findings, see Chen et al., 2021). The effect has been replicated in many different contexts. Outside the consumer behavior context, inaction inertia has also been found to influence negotiations (Terris et al., 2020) and the acceptance of job offers (Foster & Diab, 2017). We examined whether all people equally likely to fall prey to the inaction inertia effect? Despite the prominence of inaction inertia both in applied contexts as well as in academic research, this question has received barely any attention. In the present research we focus on one of the most important personality characteristics in consumer research: dispositional greed. Greed is the insatiable desire for more. It has been shown to affect many consumer behaviors, such as saving and spending (Seuntjens et al., 2016), and choosing work over leisure (Zeelenberg et al., 2020). Individual differences in greediness can reliably be measured (Seuntjens et al., 2015). Examining the role of dispositional greed in inaction inertia helps to understand who is affected by it, and ultimately, what is causing the effect. We test two rivaling hypotheses that can be derived from the definition of greed. The “greedy people always want more” hypothesis assumes that greedy people are always attracted to discounts, as it provides them an opportunity to get something for a good price. Previously missed discounts are not relevant, as greedy people focus on acquisition (Seuntjens et al., 2015). Thus, for greedy people, we could expect an overall higher likelihood to act on the current discount, and a diminished effect of the missed discount (weaker inaction inertia). In contrast, the “greedy people hate losing” hypothesis assumes that greedy people are stingy, value money more, and show more loss aversion (Krekels, 2015). We could thus expect that missing discounts hurts more for greedy people. After all, it is not possible to acquire the best possible offer anymore. This may result in an overall lower likelihood to act on the current discount, and a reduced effect of the missed discount (weaker inaction inertia). This hypothesis is in line with Mussel and Hewig (2016), who found that greedy individuals show more negative affect in response to losses, and with Tykocinski and Pittman (1998) who argue that the inaction inertia effect occurs because consumers aim to distance themselves from missed discounts that are associated with negative affect. We conducted a high-powered study (N = 551) using six different inaction inertia scenarios (From Tykocinski et al. (1995) the ski pass, the car, and the fitness club scenario; From Zeelenberg et al. (2006) the couch scenario and a simpler version of the city trip scenario; From Krijnen et al. (2020) the retirement saving scenario). This study was preregistered (https://aspredicted.org/blind.php?x=GHM_N9Z), and the data, code and materials can be found on: https://researchbox.org/374&PEER_REVIEW_passcode=MIOHPU. This experiment had the single within-subjects factor value difference (small vs. large) and the between-subjects covariate greed. Participants were recruited on Prolific Academic in the summer of 2021 to participate in a study on financial decisions. As the inaction inertia scenarios all referred to US $ as currency, we recruited US American residents exclusively. After providing informed consent, half of the participants first answered six inaction inertia scenarios, followed by a questionnaire to assess dispositional greed. For the other half of the participants, the order was reversed. Three of the scenarios were presented in the small-difference format and three in the large-difference format. For each participant, it was randomly chosen which scenarios were in the small-difference or the large-difference format. Due to the nested structure of our data (six responses nested in 551 participants), we tested our hypotheses via multilevel moderated mediation analysis in the R package lavaan (Rosseel, 2012). The data provide support for the “greedy people always want more” hypothesis as we found that greedy people showed an overall higher likelihood to act on the current discount, and a diminished effect of the missed discount (weaker inaction inertia). The results will be discussed in relation to theories about the causes of inaction inertia, and in relation to the role of greed in consumer behavior.

ABSTRACT. This paper proposes a method for empirically mapping psychological personality traits to economic preferences. There is extensive evidence that economic preferences, cognitive ability, and personality predict a wide range of economic outcomes (see Heckman, Jagelka, and Kautz, 2021 for a recent summary of the literature). However, the question of whether they work through one another or side by side has not been conclusively answered. It is important to do so in order to determine the dimension of attributes which constitute human capital and explain differences in life outcomes. I demonstrate that careful modelling of measurement and decision errors allows one to establish the long supposed but empirically elusive link (see Almlund et al., 2011 and Becker et al., 2012) between economic and psychological frameworks for understanding differences in individuals’ behaviors.

I estimate a structural model of decision making under risk and delay using data from a unique field experiment in which each participant made over 100 choices on incentivized tasks designed to elicit risk and time preferences. I use extensive associated survey data to map both true economic preferences and the stochastic components of decision-making onto cognitive ability and factors related to three of the Big Five personality traits.

My structural model has two main parts: a factor model used to derive latent cognitive ability and personality traits from multiple noisy observed indicators; and a model of decision-making under risk and delay based on the assumption that decisions are driven by expected utility maximizing behavior which itself depends on an individual's risk and time preferences but is subject to random errors. I allow preferences to depend both on observed heterogeneity and on unobserved factors related to cognitive ability and personality. In addition, I allow the structural parameters of the model to depend on "true" unobserved heterogeneity (unrelated to any observed characteristics or measures) in the form of unobserved types.

My main contribution is to show that up to 60% of the heterogeneity in both the true (or average) risk and time preferences, in their individual-level stability, and in people's propensity to make mistakes can be explained by cognitive ability and factors related to three of the Big Five personality traits: extraversion, conscientiousness, and emotional stability. True differences in desired outcomes are related to differences in personality whereas actual mistakes in decisions are related to cognitive skill.

Heterogeneity in preferences explains a majority of the variation in observed choices between risky lotteries and between payments occurring at different points in time. Indeed, the five estimated structural parameters have explanatory power which is an order of magnitude larger than that of nearly two dozen demographic and socio-economic variables. While risk and time preferences account for a vast majority of the explained variation in average risky or intertemporal choices, parameters related to randomness in decision making predict inconsistencies in individual behavior. The results are robust to a range of alternative assumptions regarding functional form and placement of the error term.

My approach generalizes to settings in which one wishes to relate parameters of economic models to observables with multiple available noisy measures. It incorporates a flexible error structure which accounts for errors in both decision making and in measurement, and thus allows to separate signal from noise in observed choices.

ABSTRACT. We use data from the US Health and Retirement Study to investigate the link between portfolio decisions and discriminatory experiences. We focus on the perception individuals have after being exposed to several discriminatory practices, and whether that treatment is due to their race or other reasons. We show that experiencing racial discrimination has a bigger effect on shaping portfolio decisions compared to any other type of discrimination. Specifically, racial discrimination reduces the probability of holding stocks and bonds by 5.3 and 1.6 percentage points respectively. Experiences of racism explain not only the choice to opt-in risky financial assets, but also the amount of assets held. Respondents who report being racially discriminated against, are non-White, and thus such experiences add to the racial wealth inequality. In general, being a non-White American reduces the probability of holding stocks and housing property by about 12 percentage points compared to a White American.

ABSTRACT. Social norms have proven to be a powerful nudge to make people adopt prosocial behavior. Indeed, informing people that most of their peers are behaving virtuously encourages them to improve their own behavior. However, since the feedback is based on the average behavior of a given population, the desirable behavior targeted has to be already practiced by a majority of the population, in order to avoid a boomerang effect, meaning that people above the average tend to act less virtuously to get closer to the average behavior. In that study, our purpose is to create a norm that can be implemented as a nudge in behaviors where current social norm nudges are inefficient, thus increasing the range of prosocial behaviors that can be improved by social norms nudge. We build a new norm, used in a nudge framework, that provides information based on the most altruistic people of the population. We find that this new norm, i.e. ”Norm from the top”, increases the contribution to prosocial behavior in the context of filling out additional surveys at the end of an experiment, whereas the standard norm does not have a significant effect, due in part to the boomerang effect. These results show the potential relevance of using the Norm from the top to promote prosocial behaviors practiced by a minority of people.

ABSTRACT. The design of effective incentive schemes that are both successful in motivating employees and keeping down costs is of critical importance. Research has demonstrated that prosocial incentives, where individuals’ effort benefits a charitable organization, can sometimes be more effective than standard monetary incentives. However, most research has focused on the intensive margin, examining effort conditional on participation in the activity. We examine the effectiveness of standard and prosocial incentives on the extensive margin, corresponding to people’s decisions to opt-in to an incentivized activity. In addition, we test the effectiveness of optional prosocial incentives, where individuals can choose between keeping or donating all or part of their payment. Across four experiments that vary the type and size of incentives, we find that individuals are more likely to avoid activities that involve any prosocial incentive. Our results highlight the importance of considering the margin of decisions when designing incentive schemes.

ABSTRACT. Predicting human behavior is one of the challenges behavioral scientists had been struggling with for decades. How would one act in a given situation? How can we direct people to behave in a desired fashion? Here, we focus on predicting pro-social behavior in economic contexts, and propose a new look into the drivers of such behavior. While there are many drivers of prosocial behavior (Dovidio et al., 2006), recent work largely focused on two: cognitive resources (Rand et al., 2012) and personality traits (Camerer, 2003; Hengel, 2011). Here, we link these two lines of research and ask: Does personality need cognitive resources to shape prosocial behavior? Put differently, we explore whether people’s stable dispositional traits (their “true” self) manifest themselves when time allows for deliberation. Thus, our work fits (and contributes to) the recent call for a person-by-situation approach to further our understanding of human prosocial behavior (Mischkowski & Glöckner, 2016; Mischkowski et al., 2018) and can help explain some of the inconsistencies in the recent literature. Personality traits account for a significant amount of variability in prosocial behavior, both measured with economic games and in real life settings (Boone et al., 1999; Camerer, 2003; Hengel, 2011). Accordingly, broad personality models such as the HEXACO (Ashton & Lee, 2007) and the big-5 (Costa & McCrae, 2009), as well as specific traits such as social value orientation (Van Lange, 1999) have been linked to different forms of prosocial behavior. Importantly, recent work (Thielmann et al., 2020) suggests that certain traits are more (or less) likely to affect specific prosocial behaviors measured in economic games based on situational affordances. Whereas many situational factors can increase the effect of personality traits on behavior, here we build on trait activation theory and focus on time pressure as a unique situational factor which might reduce people’s ability to rely on personality traits (Edwards et al., 2014). According to trait activation theory (Tett, & Burnett, 2003), individual differences are unlikely to emerge in situations in which people’s options are constrained (Cooper & Withey, 2009; Mischel, 1977). Indeed, previous studies had shown that prosocial people are more likely to act in line with their preferences when they have ample cognitive resources to overcome the temptation of acting selfishly (Bieleke et al., 2017; Hilbig et al., 2018). Similarly, anti-social preferences such as Machiavellism, have also been shown to be resource-dependent (Jones & Paulhus, 2017). Here, we aim to tap into the nature of the cognitive resources X personality interaction in economic games, and study when people’s personality traits are most likely to shape their behavior. The current research In two experiments, participants played an economic game in which they had a chance to act pro-socially or selfishly. In both experiments, half of the participants were assigned to a time pressure condition, and the other half were not under time pressure. In each experiment, we measured a specific personality trait expected to affect behavior in the game played, and tested whether the time pressure condition moderated to effect of personality on behavior. In Experiment 1 (n=184) participants played 6 rounds in the role of player A (trustor) in the trust game (Berg et al., 1995). In each round, participants were endowed with a sum of money and were asked whether they wanted to transfer a predetermined amount of it to player B. They were informed that this amount would be tripled and transferred to player B, which would then have to opportunity to send some of his endowment back to player A. Half the participants were assigned to the time pressure condition and had to respond within 5 seconds, whereas the other half were assigned to the no time pressure condition in which they were forced to wait at least 10 seconds before they could make their decision. Before playing the game, participants’ risk preferences were measured using the DOSPERT scale (Blaise & Weber, 2006). Risk preferences had been shown to predict the behavior of the trustor in the trust game (Houser et al., 2010). Results of Experiment 1 revealed a condition X DOSPERT interaction, b = -.27, SE = .15, p = .06. A slope analysis showed no effect of DOSPERT in the time pressure condition (b = .17, SE = .10, p = .10) but a significant effect in the no time pressure condition, b = .44, SE = .10, p < .001. In Experiment 2 (n=352, preregistered) participants played 5 rounds of the dictator game (Forsythe et al., 1994). In each round, participants were endowed with a sum of money and were asked whether they wanted to transfer a predetermined amount of that sum to another player, who had no opportunity to respond in any way. As in Experiment 1, time pressure was manipulated between-participants. Before playing the game, participants completed the Honesty/Humility subscale of the HEXACO model (Ashton & Lee, 2009), which was shown to predict behavior in the dictator game (Hilbig et al., 2013). Results of Experiment 2 conceptually replicated those of Experiment 1 and revealed the predicted condition X Honesty/Humility interaction, b = -1.09, SE = .52, p = .04. A slope analysis showed no effect of Honesty/Humility in the time pressure condition (p = .58) but a significant effect in the no time pressure condition, b = .90, SE = .38, p = .02. Taken together, our results reveal that personality is able to predict pro-social behavior only when participants have ample time. Under time pressure, the effects of the personality traits on behavior disappeared. Our findings suggest that personality does not affect people’s automatic responses, but that is shapes their behavior when given time to think about their preferences.

ABSTRACT. A fundamental characteristic of modern societies is economic inequality, where deprived individuals experience chronic economic scarcity. While such experiences have been shown to produce detrimental outcomes in regards to human judgment and decision-making, the consequences of such scarcity for our morality remain debated. We conduct one of the most comprehensive tests of the relationship between experiences of relative chronic economic scarcity and various measures linked to morality. In a pre-registered study, we analyse data from a large, cross-national survey (N = 46,450 across 67 countries) allowing us to address important limitations related to measurement validity and external validity in past research. Our findings demonstrate that experiences of relative chronic economic scarcity, as indexed by (1) low subjective socioeconomic status at the individual level, and (2) income inequality at the national macro level, predict higher levels of moral identity, higher morality-as-cooperation, a larger moral circle, and importantly; more prosocial behaviour. The results appear robust to several advanced control analyses. Finally, exploratory analyses indicate that observed income inequality at the national level does not significantly moderate the predicted effect of subjective socioeconomic status. Our findings have vital implications for understanding human morality under chronic resource scarcity.

ABSTRACT. Recent studies have shown that inducing people not to believe in free will creates a feeling of helplessness and greatly diminishes prosocial behavior. In fact, people are more likely to cheat in an academic test or to be aggressive towards other individuals. Since altruism is a strong manifestation of prosocial behavior, we test if belief in free will can have an influence on giving behavior. For that purpose, we designed three distinct treatments which were implemented through a survey experiment, namely one where the notion of free will was reinforced, another where the notion of free will was refuted, and another under a neutral context. The results were compared with giving behaviour in a hypothetical dictator game. We found that on average respondents were less generous under the treatment refuting the existence of free will than in the pro-free will treatment or in the Neutral treatment. Furthermore, we found that when a relationship between belief in free will and amounts given in the dictator game existed, it was just for those individuals who reported higher beliefs in free will (using the FAD-Plus scale). We also found a positive and significant relationship between belief in scientific determinism (a subscale of the complete free will scale) and generosity.

ABSTRACT. Empirical findings showed that both trust in and power of governmental authorities increased tax compliance and these effects were relatively stable across a wide range of nations (e.g., Batrancea et al., 2019; Kogler et al., 2013). We also know that perceived fairness of taxation entails a higher level of tax compliance (e.g., Hofmann et al., 2008). Based on the important roles both trust in and power of governmental actions play in the interaction between the citizens and governmental authorities, we propose that the perceptions of trust and power increase tax compliance through the perceived fairness of governmental actions. We also propose that perceived fairness has an overarching effect and citizens tend to reciprocate fair governmental operations not only by elevating their tax intentions, but also by increasing their work motivation – another important factor in the efficient provision of public goods.

Hence, by relying on and extending the slippery slope framework (Kirchler, 2007), in a scenario study (N = 365 Hungarian adults, the vast majority with work experience) it was investigated how trust in and power of governmental functions influence both tax compliance and work motivation. The potential mediational role of perceived fairness regarding the actions of governmental institutions was also studied. Based on previous studies, in a 2 × 2 between-subjects design a hypothetical country, Varosia was presented in four different scenarios, aimed at manipulating both trust in governmental institutions (i.e., high or low level of trust by presenting regular or seldom referenda, transparent or non-transparent legislation etc., respectively) and power of governmental authorities (i.e., high or low level of power by presenting efficient or inefficient prosecution of tax evaders etc., respectively). After being asked to imagine living and paying their taxes in Varosia, participants responded to manipulation check questions and reported their (1) intended tax compliance (measured with three items, e.g., How likely would you pay your tax completely honestly?), (2) voluntary tax compliance (five items, e.g., When I pay my taxes in Varosia as required by the regulations, I do so to support the state and other citizens.), (3) enforced tax compliance (five items, e.g., When I pay my taxes in Varosia as required by the regulations, I do so because the tax office often carries out audits.), (4) tax evasion (five items, e.g., A customer paid in cash and did not require an invoice. How likely is it that you would omit this income on your tax return?), (5) work motivation (nine items, e.g., The above-mentioned conditions and circumstances of Varosia would stimulate me to show a good level of work performance.) as well as (6) the perceived fairness of governmental institutions (three items, e.g., The governmental institutions and authorities act fairly towards citizens.). All measures showed good internal consistency indices.

Based on moderated mediation analyses the results showed that the trust–fairness relationship was moderated by power (i.e., higher trust was associated with a higher level of perceived fairness, especially when power was high), and a higher level of perceived fairness was related to higher levels of both intended tax compliance and work motivation. A similar result was found regarding voluntary tax compliance, but not regarding enforced tax compliance, suggesting that perceived fairness increased the intrinsic rather than the extrinsic motivation behind higher levels of both tax compliance and work effort. The moderated mediation model for tax evasion further supported the previous results: besides the above-mentioned positive and significant trust × power interaction on perceived fairness, this fair perception was negatively related to tax evasion. A follow-up study with a similar research design is planned to examine whether these results can be explained by reciprocity (i.e., the perceived quality of social exchange between the government and citizens) and/or by other positive elements of the perceived interaction between the citizens and the government (e.g., positive affect generated by experiencing fair actions in the interaction). The results highlight the overarching importance of institutional fairness in determining not only tax compliance, but also work motivation as two central factors in the creation and sharing of public goods. The application possibility of the results in policy making and the necessity of both transparency and consistency in democratic governance will be discussed.

References Batrancea, L., Nichita, A., Olsen, J., Kogler, C., Kirchler, E., Hoelzl, E., Weiss, A., Torgler, B., Fooken, J., Fuller, J., Schaffner, M., Banuri, S., Hassanein, M., Alarcón-García, G., Aldemir, C., Apostol, O., Bank Weinberg, D., Batrancea, I., Belianin, A., ... Zukauskas, S. (2019). Trust and power as determinants of tax compliance across 44 nations. Journal of Economic Psychology, 74, 102191. https://doi.org/10.1016/j.joep.2019.102191

Hofmann, E., Hoelzl, E., & Kirchler, E. (2008). Preconditions of voluntary tax compliance: Knowledge and evaluation of taxation, norms, fairness, and motivation to cooperate. Zeitschrift für Psychologie/Journal of Psychology, 216(4), 209–217. https://doi.org/10.1027/0044-3409.216.4.209

Kirchler, E (2007). The economic psychology of tax behaviour. Cambridge University Press. https://doi.org/10.1017/CBO9780511628238

Kogler, C., Batrancea, L., Nichita, A., Pantya, J., Belianin, A., & Kirchler, E. (2013). Trust and power as determinants of tax compliance: Testing the assumptions of the slippery slope framework in Austria, Hungary, Romania and Russia. Journal of Economic Psychology, 34, 169–180. https://doi.org/10.1016/j.joep.2012.09.010

ABSTRACT. Objectives and Motivation: Several scholars brought evidence to the existence of a general aversion toward taxes. McCaffery and Baron (2006) found that the word “tax” generated a more negative reaction than the label “fee”, although labeling a monetary charge as a tax or a fee does not make any difference in financial terms. Eckel et al. (2005) showed how explicit forced contributions generated a more robust crowding out effect than unlabeled tuitions. Such results are in line with other studies that revealed the presence of negative explicit attitudes towards tax institutions (Kirchler, 1999; Berti & Kirchler, 2001), uncovering the perception of taxes as an emotionally charged sacrifice (Ferrari & Randisi, 2013), possibly decreasing tax morale (Schmölders, 2006) and increasing the acceptance of antitax behaviors, such as tax evasion (Kirchler, 1999; Kirchler & Braithwaite, 2007). Sussman and Olivola (2011) redefined the concept of tax aversion as “a preference to avoid tax-related costs than to avoid equal-sized (or larger) monetary costs unrelated to taxes” (Sussman & Olivola, 2011; p. S91). In other words, the phenomenon of tax aversion corresponds to an individual propensity in escaping taxes more pronounced than the equal tendency in avoiding any other sort of payments. However, following replications of the experiments conducted by Sussman and Olivola provided inconsistent results (Olsen et al., 2019). The present study suggests a complementary perspective on tax aversion, responding to the question if tax aversion might operate as a latent and relatively unconscious phenomenon, besides being an explicit propensity. The question is elicited and justified by reviewing the work of the German economist Gunter Schmölders, who suggested that the roots of the opposition towards taxes may be traced back to historical and cultural issues (Schmölders, 2006). Leaning on Schmölders’ thought, it is therefore plausible that tax aversion may stem from a more genuinely inner predisposition rather than an explicit and rational attitude. Such intuition drove the present research, aimed at measuring tax aversion at both explicit and implicit level and verifying the connection between implicit and explicit measures. Methods: We present the results of three independent studies conducted in the Italian context. The Implicit Association Test (IAT) was employed to assess implicit tax aversion, whereas other measures were used to explicitly evaluate the phenomenon. Specifically, to assess the behavioral outcomes of tax aversion, a replication of experiments 1 and 2 of Sussman and Olivola (2011) was performed; the Perceived Fiscal Sacrifice Scale (Ferrari & Randisi, 2013) was adopted to evaluate tax aversion from an explicit emotional perspective; last, three dimensions of Tax Compliance Inventory TAX-I (Kirchler & Wahl, 2010) were included to investigate voluntary and enforced compliance, and tax evasion. Results: Analyzing the scores of the IAT, results support the idea that tax aversion may operate as a latent phenomenon, confirming our leading hypothesis. On the one hand, by replicating experiments 1 and 2 (Sussman & Olivola, 2011) on the behavioral aversion towards taxes beyond monetary costs, we found no effect, thus partially confirming the results of Olsen et al. (2019). On the other hand, implicit (IAT) tax aversion scores predicted the behavioral tax aversion in both experiments. Indeed, a series of logistic regressions shows that the higher the implicit tax-aversion, the higher the willingness to accept time investment to receive a tax-related (VAT) discount (but not an equivalent tax-unrelated discount). However, implicit tax aversion measure (IAT) appears to be independent (no correlations) from the explicit, self-reported measures which were included in the study (the Perceived Fiscal Sacrifice Scale and the three dimensions of TAX-I: voluntary and enforced compliance, and tax evasion). Conclusion: These preliminary results show that IAT can be an effective tool to detect (implicit) tax aversion, although, in terms of efficiency, the usefulness of this new measure can be debatable.

ABSTRACT. The most influential theory in tax literature is the seminal work by Allingham and Sandmo (1972) and Srinivasan (1973), who applied the economics-of-crime approach by Becker (1968) to tax reporting behavior. Considerable extensions to the standard model have been added over the years (for an overview see Alm, 2019; Andreoni et al., 1998), however, the basic premise of the deterrence approach stayed the same. Taxpayers choose between compliance and evasion based on a simple cost-benefit analysis. If deterrence is low (i.e., low audit probability in combination with low fines), evasion is likely; if deterrence is high, taxpayers choose to pay their due taxes.

Despite the widespread application of the deterrence approach, there is mounting evidence pointing at its limitations (e.g., Alm & Malézieux, 2020). Most importantly, the deterrence approach strongly overpredicts evasion rates in the real world. In most countries, the probability of being audited and the fines for detected tax evasion are relatively low. As a consequence, high rates of tax evasion would be expected. Most taxpayers, however, comply with the tax laws (see Alm, 2019; Alm et al., 1992; Andreoni et al., 1998). Even in the artificial and simplified environment of the research lab, compliance is usually higher than the deterrence approach predicts (e.g., Alm & Malézieux, 2020; Alm et al., 2010; Engel et al., 2020; Slemrod, 1992). Additionally, some studies even do not observe positive deterring effects of audits and fines (Iyers et al., 2010; Kirchler et al., 2010, Slemrod et al., 2001). Despite these limitations and efforts to incorporate psychological considerations, the deterrence approach has remained at the core of most strategies to combat income tax evasion.

In a recent experiment (Kogler et al., 2022) we applied MouselabWEB, a tool to record the acquisition of information via a computer mouse, in a lab experiment to investigate whether individuals actually behave as assumed in the deterrence approach. That is, whether individuals integrate outcomes and probabilities in what can be called a compensatory weighting and adding process (Su et al., 2013). To this end, we tested the implicit assumptions about the underlying decision process that can be theoretically derived from the deterrence approach (see Orquin & Mueller Loose, 2013). Based on both information processing and behavioral data, we found contradictions to the deterrence approaches’ main assumptions. First, model-conform decisions were not associated with information acquisition patterns that indicate more expected value (EV)-like calculations: neither with more transitions between audit probability and income, nor with more transitions between audit probability and fine, and also not with more box openings in general. Second, compliance was considerably higher in general than the model would predict, and although audit probability and fine level influenced decisions, they were not always integrated adequately. Finally, explaining and providing explicit information about EV did not enhance model-conform decisions.

The aim of the present study was to investigate whether the result that explicit EV information does not lead to more model-conform decisions, is stable and generalizable beyond a mere numerical presentation of expected values. First, we implemented an extensive introduction where all relevant parameters, as well the concept of expected value, were explained in great detail. We included several checks to ensure that participants understood all this information. Second, we introduced three different variants of communicating and presenting EV information. We compare tax decisions where (1) neither verbal explanation of the concept of EV nor explicit EV information is provided (control condition) with decisions where we (2) only verbally explain the concept of EV before the actual compliance decisions, (3) verbally explain the EV and present the concrete numerical values for each compliance decision, and (4) verbally explain the EV and present explicit EV information in the form of a visual cue.

Participants were recruited online via Prolific. The final sample size was N = 345 (165 women and 180 men) with a mean age of 35.3 years (SD = 13.0). As outlined, the design included four experimental conditions based on EV-relevant information, and the dependent variable tax compliance (dichotomous choice; full evasion vs. full compliance) was measured over 24 rounds. There were four within-subject factors that varied the tax-related decision parameters resulting in 24 unique combinations: Income (1200 vs. 1400 Experimental Currency Units), Tax Rate (25% vs. 40%), Audit Probability (10% vs. 20% vs. 30%), and Fine Level (paying back the evaded amount plus a fine of 50% vs. paying back the evaded amount plus a fine of 100%). The order of rounds was random for each participant. The information in each round was presented using MouselabWEB (Willemsen & Johnson, 2019), where the information of each of the four tax parameters was hidden behind labeled boxes. To see the respective information, participants had to move the mouse cursor over a box. Once the mouse cursor was moved outside the respective box the information was hidden again. Accordingly, frequency, duration, and sequence of information acquisition were recorded.

The results indicate that participants made more model-conform decisions when we first verbally explained the concept of EV and additionally explicitly provided visual EV information for each tax payment decision. However, the other two conditions (explanation of EV only; explanation of EV with additional numerical EV information) did not influence compliance behavior compared to the control condition, which is in line with our previous study (Kogler et al., 2022). Although the visual EV cue adjusted participants’ decisions in line with the assumptions of the deterrence approach, compliance did not drop below 40%. Considering that evasion was the dominant choice in all rounds, these results suggest that even in a highly artificial setting, deterrence alone can hardly explain tax compliance decisions. Consequently, we argue that deviations from the predictions of the deterrence model are not merely due to the complexity of EV-calculations, but rather that such considerations are not of utmost importance in the first place.

ABSTRACT. What is the compliance effect of experiencing a tax audit? Empirical studies typically report a positive effect, while tax experiments frequently report a negative effect. We show that whether a tax audit increases or decreases subsequent compliance hinges on the balance of learning opportunities, mis-perception of audit risk, and the confounding effect of censoring. For an endogenous audit rule under which taxpayers can learn to reduce their audit risk by reporting higher income, learning effects outweigh misperception. But, the reverse holds when audit selection is exogenous or there is no scope for learning.

ABSTRACT. Collaborative evasion of taxes and social security fees is prevalent in household services, when a household hires a service provider and no third party is involved. However, evidence on the determinants of collaborative tax evasion in general and the household context in particular is lacking. This paper examines two coordination mechanisms of collaborative tax evasion: A partner’s signaled intention and information about majority's evasion behavior (empirical evasion expectation). We implement an interactive tax evasion game in an online labor market (MTurk) with 560 participants. Our findings show that priming with an empirical evasion expectation increases the fraction of evaded transactions by 20 percentage points. Our treatment manipulation of intention signals does not render a significant effect on evasion. However, when willingness to evade is signaled first in the chat, the probability of evasion increases by 45 percentage points.

ABSTRACT. Taxation is crucial for any functioning society, as it contributes towards public goods and finances governmental services. Not surprisingly, taxation is a recurrent topic in people’s conversations (e.g., Onu & Oats, 2016). Hence, information (including sentiments and behaviors) spreads across individuals’ social networks – a phenomenon referred to as “network effects”. For example, neighborhood communication effectively diffuses information on enforcement, ultimately resulting in increased compliance (Drago et al., 2020; Rincke & Traxler, 2009). While the network effects literature focused on information diffusion across individuals in close contact or geographical proximity, little is known about how the information about taxes is shared online. This knowledge is important since much of the communication nowadays occurs on social media platforms. With over 4.2 billion users in 2021 (DataReportal, 2021), social media, such as Twitter, are becoming an increasingly important platform for sharing news and information, including taxes (e.g., www.reddit.com/r/tax, www.facebook.com/groups/usexpattax). This makes social media platforms a great tool to study network effects for taxation related topics. We examined what makes a message about taxes more or less likely to spread across the users’ networks on one of the most popular social media platforms: Twitter. We focused on the linguistic features of the message. Prior research on information spread on social media showed that moral-emotional language was associated with wider diffusion of messages in online discussions of polarizing political issues, such as gun control, same-sex marriage, or climate change (Brady et al., 2017). In contrast, for less polarizing topics (e.g., mathematics), the use of moral and emotional language might appear “unnecessary” and jeopardize rational debate, ultimately leading to less diffusion (Burton et al., 2019). Given these different findings in prior research, we explored the role of moral and moral-emotional language in information diffusion about taxes. In light of the literature on negativity bias (Rozin & Royzman, 2001), we additionally examined the role of valence by considering the effect of positive vs. negative emotional and moral-emotional language separately. Lastly, drawing from the prior research on the importance of specific emotions – anger and fear – for tax compliance decisions (Erard et al., 2018), we tested whether specific emotions identified as important in existing taxation research (e.g., anger and anxiety) are related to information diffusion. We gathered over 4 million English language tweets with the hashtags #tax, #taxes, and #taxation, ranging from years 2010 to 2020. For classifying language content, we used a dictionary-based approach. That is, we compared the words in our text corpus to a dictionary and then return the prevalence of the dictionary word occurring within each individual tweet. For assessing moral language, we used the Moral Foundations Dictionary 2.0 (Firmer et al., 2019), and for emotional language we used the Linguistic Inquiry and Word Count (LIWC, Pennebaker et al., 2015). Overall, we computed the frequencies of five word categories: moral, emotional, moral-emotional (i.e., words that were part of both emotional and moral dictionaries) words, and for two specific emotions, namely anger and anxiety. To measure information diffusion, we use the retweet count of those tweets. To test our assumption, we ran negative binomial regression models while controlling for the follower count of the respective user. Our results suggest that the use of emotional, moral, and moral-emotional language in a tweet predicted greater diffusion (i.e., more retweets). Furthermore, positive emotional words were a better predictor of information diffusion than negative emotional words. Among the specific emotions only the use of anger (but not anxiety) words predicted more retweets. In a next step, applying an unsupervised machine learning approach (i.e., topic modelling) to the tax related tweets, we identified 30 topics with different content. Using a theoretical framework differentiating between synergistic and antagonistic tax climates, we were able to further categorize these topics into four broader groups: 1. Opinions about Tax Politics, 2. Enforcement (antagonistic climate), 3. Information & Service (synergistic climate), and 4. Emotions. The most frequently observed group was Information & Service (synergistic climate), which also steadily gained prominence during the past decade. We proceeded by analyzing the information diffusion properties and sentiment of the tweets associated with the four groups. Information & Service tweets had the most positive sentiment but were shared the least, while tweets regarding Opinions about Tax Politics were shared most often. In sum, the results suggest that lay people’s conceptions about taxation – as discerned from conversations on social media (Twitter) – largely reflect a synergistic (vs. an antagonistic) climate. We believe that the current findings could have important policy implications. Knowing what affects the spread of information could allow tax offices optimizing their communication efforts. Tracking changes in negative and positive emotional language in tax-related tweets may allow for an unobtrusive assessment of tax morale or tax climate with naturally occurring communications. Furthermore, identifying the most frequently discussed tax related topics could help policy makers determine where action or reform is needed most urgently. These findings are a novel contribution to the literature on tax psychology, as they illustrate the usefulness of a previously unexplored, but widely used, medium for communicating information about taxes.