Abstract
We examined the impact of anticipating poor economic conditions on financial risk taking. In Experiment 1, the salience of poor future economic prospects was manipulated among young adults. Those who were reminded of their poor future economic prospects were more likely to take the opportunity to gamble with their money than those in the control condition. In Experiment 2, we once again manipulated the salience of poor economic prospects. Extending the results of Experiment 1, participants who were reminded of their poor economic prospects bet more money on a spin of a roulette wheel than those in a control condition. Importantly, we show that the relationship between poor economic prospects and gambling is mediated by belief in the necessity of taking financial risks to make money. Implications of economic downturns for gambling and other forms of risk taking are discussed.
In periods of economic downturn personal debt tends to increase, job prospects worsen, and earning losses sustained typically take 8–10 years postrecession to fade (Oreopoulos, von Wachter, & Heisz, 2006). Thus, when facing poor current and future economic conditions, a prudent course would be to conserve existing funds (i.e., save), pay down existing debt, and carefully select long-term, low-risk investments (see Flatters & Willmott, 2009; Friedman, 1957; Sass, Monk, & Haverstick, 2010) where there is a potential for positive expected value. A less prudent course would be to play games of chance—a form of financial risk taking where there is a negative expected value (see Labrie, Kaplan, LaPlante, Nelson, & Shaffer, 2008). Yet, human behavior does not always conform to such rational choice predictions (Akerlof, 2007; Akerlof & Shiller, 2009). In contrast to rational choice models, we test the idea that economic threat might, counterintuitively, increase people’s willingness to take financial risks in games of chance (i.e., gambling).
In line with a rational choice argument, reports of consumer behavior in times of economic difficulty suggest that people typically tighten their grip on personal savings and restrict spending on general consumptive expenditures such as food, clothing, and out-of-pocket expenses (McCully, 2011). Objectively, thrift is sensible in the face of decreasing disposable income (i.e., “don’t spend what you don’t have”). According to traditional models of economic decision making (see expected utility theory; Friedman & Savage, 1948), people make financial choices that provide them the best return on investment, that is, they maximize utility. Thus, in the presence of economic threat, people should engage in fiscally conservative decision making and refrain from activities that have a negative expected value—as is the case with gambling.
It is now understood, however, that decision-making models based on the notion that people maximize personal economic utility do not fully capture the financial decisions people make, especially when there is a risk of loss (see Bromiley, 2010; Kahneman & Tversky, 1979; Larrick, Heath, & Wu, 2009; Mishra & Fiddick, 2012; Rabin, 1998; Shafir & LaBoeuf, 2002; Tversky & Kahneman, 1992; Weber & Johnson, 2009). For example, prospect theory’s value function (Kahneman & Tversky, 1979) explains how behavior is altered by the presence of a reference point such as a financial expectation or goal. The value function incorporates three principles. The first principle, the reference point, partitions outcomes into gains and losses (or successes and failures) relative to a reference point (e.g., how much money people expect to have in their bank account). The second principle, loss aversion, is that outcomes perceived to be losses are more painful than similar sized gains are pleasurable. The third principle, diminishing sensitivity, is that outcomes have a smaller marginal impact when they are more distant from the reference point. As a result, the pain a gambler experiences following a loss of an additional $10 when $120 had already been lost is less than if only $20 had already been lost.
Based on prospect theory, we contend that the salience of an economic downturn should lead people to perceive that losses are on their financial horizon (relative to their financial reference point). The potential of financial loss is unpleasant, which should result in loss aversion (i.e., people feel uncomfortable about the prospect of financial loss). As perceptions of economic threat intensify by anticipating a downturn in the economy, and the possibility of being below their financial reference point increases, people should be more willing to take financial risks to offset the perceived threat to their financial future. In this way, engaging in games of chance might offer people the possibility of being at their reference point at the end of the day, thus leading some to believe that taking a gamble on a negative expected value makes some sense (see Ali, 1977; McGlothlin, 1956). In this light, the taking of financial risks might be perceived as prudent under certain economic conditions.
Other decision-making and risk-taking models lend support for our contention that increased financial risk taking may occur when the threat of economic downturn is salient. Individuals who feel their financial situations have worsened (experiencing of a downturn) tend to feel deprived (Liang & Fairchild, 1979)—a situation positively associated with willingness to gamble (Callan, Ellard, Shead, & Hodgins, 2008; Callan, Shead, & Olson, 2011). Specifically, Callan, Ellard, Shead, and Hodgins (2008) suggested that gambling allows people to alleviate feelings of financial deprivation due to the perceived opportunity to gain that which they believe is deserved, that is, financial security. In addition, peoples’ subjective wants and needs impact risk taking. When a perceived need (e.g., money) is deemed not to be satisfied (and outside immediate grasp), risk taking to obtain the needed object increases (see Mishra & Fiddick, 2012; Weber, Shafir, & Blais, 2004). Thus, gambling provides the potential to gain financial stability (e.g., the “Jackpot” on a slot machine is hit), and as a result help people meet their needs or offset feelings of deprivation through additional monies. Unfortunately, gambling is a risky strategy to achieve these goals, given its negative expected value, which is exacerbated by decision makers’ tendency to overvalue the probability of success in games of chance (see Prakash, Chang, Hamid, & Smyser, 1996; Wohl & Enzle, 2002). This is particularly troublesome in small stakes gambles (i.e., scratch off tickets, slot machines)—games with small probabilities of success that require only a small bankroll to play (Kahneman & Tversky, 1979; Prelec & Loewenstein, 1991; Weber & Chapman, 2005). Accordingly, we propose that under poor economic conditions, unmet financial needs might result in greater willingness to gamble.
Circumstantial evidence for our proposition that economic threat may facilitate gambling is found in the negative association between lottery ticket purchasing and financial stability (see Blalock, Just, & Simon, 2007). Specifically, lottery play is most prevalent among those with the least buying power (those who are poorer). The poor purchase lottery tickets, in part, because it provides the hope of extrication from the difficult financial situation they face. Unfortunately, due to the negative expected value inherent in games of chance, remaining disposable income will eventually be lost—a situation of greatest detriment to the poor who, by definition, have limited funds. When coupled with the fact that people overestimate the likelihood that gambling will yield financial success, an economic downturn is ripe for increased gambling. Support for this contention is provided by Canadian Gaming Association (CGA, 2008, 2011) data showing an overall 5% gambling revenue increase during the 2008 recession, with casino and lottery revenue growths of 11%. Therefore, from a macroeconomic standpoint, when economic threat is made salient, people are likely to respond with a greater willingness to gamble.
Overview of the Current Research
Using the current world economic crisis as a backdrop, the present research examined the effect of perceived financial insecurity on risk taking in a negative expected value domain. First, we assessed the yet unexplored possibility that geopolitical forces (the world economy), by threatening the individual’s future economic prospects, might influence gambling behavior. Second, we examine a possible reason why economic threat might result in increased risk taking in games of chance. Specifically, in Experiment 1, we investigated in whether participants were willing to gamble when poor economic prospects were made salient through a future economic threat manipulation. To this end, university students read that due to the downturn in the world economy, their cohort had poor economic prospects (e.g., their degree will likely not help them find a job) because of higher unemployment. They were then given the option to gamble with their participant remuneration ($10). It was hypothesized that when such economic threat was salient, willingness to gamble would increase. Although counter to our general hypothesis, it is also possible that poor economic prospects could lead to a general desire to spend money even in ways that lack explicit financial risk. If it were the case that economic threat elevates a general desire to consume, participants should be more likely to forfeit the opportunity to gamble and instead take the $10 remuneration in the economic threat condition relative to the control condition. Taking the money they had received from the experiment would enable participants to purchase anything they desire once they left the study.
In Experiment 2, we again manipulated economic threat to examine whether the salience of poor economic prospects heightens betting behavior (i.e., amount of money bet while gambling). Similar to Experiment 1, we hypothesized that the salience of poor economic prospects would result in greater amounts of money bet (compared to a control group). We also examined whether perceived need to take financial risks mediates the effect of economic threat on gambling. Negative mood was assessed as an alternative potential mediator of the economic threat-gambling link.
Experiment 1 Method
Participants
A total of 51 (27 male and 24 female) undergraduates at a large Canadian University participated in Study 1. They ranged in age from 17 to 42 years, with a mean age of 22.08 years (standard deviation [SD] = 4.48). Participants received $10 for their participation.
Procedure
Participants were randomly assigned to one of the two conditions. In the economic threat condition, participants read a news article, ostensibly from a well-respected national news magazine. The article warned the reader (in this case, current university students) that “the global financial crisis will soon be felt by students.” It was explained that the current financial crisis would result in “additional tuition fees,” “greater difficulty securing loans,” and that an investment in higher education “may not even be enough to secure the few positions available in this unstable economic climate.” Participants in the control condition read an article about money production at the Royal Canadian Mint. All the participants then completed a filler questionnaire about their perception of the economy. Upon completion, participants were given $10 as remuneration for their time. However, prior to their departure, they were offered the opportunity to participate in a separate study about gambling. Participants were told that they would be ushered to a casino laboratory where they could gamble with their $10 on any of a number of three-reel slot machines with a single pay line. Moreover, they were informed that the odds of success were set to mimic that of the local casino. Participants’ choice of whether or not to gamble was the dependent measure. Once the participant’s choice to gamble or not gamble was made, the experimental session came to a close. All participants were fully debriefed.
Results and Discussion
As predicted, more participants in the economic threat condition chose to gamble (N = 13; 48%) than participants in the control condition (N = 6; 23%), χ2(1) = 4.56, p = .03. Thus, in line with our hypothesis, when poor economic prospects were made salient, people chose to gamble with their money when presented with the opportunity.
The observed result is counterintuitive from an objective standpoint. People should want to save when their economic future is framed as being bleak. Instead, participants who were manipulated to think their economic prospects were poor were more likely to gamble. The argument could be made that under economic threat, people might be more willing to spend money on anything and that gambling was the only “purchasing” option provided to the participant. However, if economic threat simply heightens peoples’ overall desire to spend their money, the most reasonable choice for participants would have been to take the money they had with them. With money in hand, participants could have purchased anything. Instead, they chose the risky option more often in the threat condition—an option that could result in monetary loss, but also provides the opportunity for greater financial gain.
The result obtained in Experiment 1 only provides evidence for a link between economic threat and willingness to gamble. What the results do not speak to is why people are willing to risk most often in this condition. We suggest that the effect of economic threat on intent to gamble is due to a heightened belief that people must take risks to make money. In times of economic crisis, people find it difficult to make money through traditional avenues (e.g., a well-paying job, the financial market), yet a desire to have financial security remains. Consequently, people may turn to gambling as a means of achieving that which they believe they deserve. Such feelings are likely aroused when poor employment opportunities and financial decline are made salient. Thus, when their poor economic prospects are salient, people will gamble more because they believe they need to take financial risks to make money. We test this possibility in Experiment 2.
Experiment 2
Replication of the counterintuitive effect of economic threat on gambling observed in Experiment 1 was warranted. We also wanted to extend the effects observed to gambling behavior—recall, participants in Experiment 1 simply expressed willingness to gamble, but were not permitted the opportunity to do so. In Experiment 2, following an economic threat manipulation, all participants were given the opportunity to gamble with $10 in seed funds on a single spin of a roulette wheel. In this way, we were able to assess the amount that participants were willing to wager under economic threat. We hypothesized that participants in the economic threat condition would bet more of their $10 than participants in the control condition.
Conducting a second experiment also permitted examination of the possible mechanisms by which the salience of an economic threat increases willingness to take financial risks. To this end, we assessed whether economic threat heightens the perceived need to take financial risks, which in turn leads to gambling. To the extent that people perceive a threat to their economic future, there should be an associated increase in the perceived need to take financial risks to achieve financial security. Importantly, unfulfilled needs, like financial security, tend to increase willingness to take risks to satisfy those needs (see Mishra & Fiddick, 2012; Weber et al., 2004). One avenue of financial risk taking is gambling.
An alternative mediating mechanism was also considered. Specifically, we assessed negative affect as a mediator of the economic threat effect on gambling. We chose negative affect because of its established association with both risk taking and gambling. Numerous studies have shown negative mood to predicts gambling propensity; because of the immersive nature of gambling, it may allow people to escape their aversive emotional state (Blaszczynski & Nower, 2002; Ledgerwood & Petry, 2006; Stewart, Zack, Collins, Klein, & Fragopoulos, 2008). Positive affect, however, does not have the same effect on the initiation of gambling (see Wohl & Enzle, 2003).
Method
Participants
Participants were 45 (23 male and 22 female) undergraduates at a large Canadian University. They ranged in age from 17 to 35 years (M = 21.29, SD = 3.92). As in Study 1, all participants received $10 in payment for their participation.
Procedures and Measures
The experimental manipulation of economic threat used in Experiment 2 was identical to that used in Experiment 1—we reminded some participants that their future financial position was somewhat precarious. They then completed a measure of perceived need to engage in financial risk taking (α = .90). The items used to assess this construct were “I think people need to take risks to make money,” “I think of gambling like a financial investment,” and “I think people should play it safe when it comes to their finances (reverse-scored).” Responses to these items were anchored at 1 (strongly disagree) and 7 (strongly agree). Participants then completed the Positive Affect Negative Affect scale (PANAS; Watson, Clark, & Tellegen, 1988). The PANAS consists of 20 emotions, with 10 emotions related to positive affect (α = .84) and 10 emotions related to negative affect (α = .90). Participants indicated to what extent they felt each emotion after reading the news article manipulating economic insecurity on a 5-point Likert-type scale anchored at 1 (strongly disagree) and 5 (strongly agree).
Differing from the procedure adopted in Experiment 1, participants were not given the choice of whether to participate in the ostensibly separate gambling study. Instead, participants were given $10 as gambling seed money. They were told any amount of money they had left at the end of the study they would be allowed to keep. Participants were then escorted to a roulette table and told they could refrain from betting (and leave the laboratory with $10) or place a wager (in $1 increments) of up to $10 on a single spin. The game was simplified, so that participants gambled on whether the roulette ball would land on a red or a black number (if the ball landed on “0” or “00,” there would be a re-spin). The expected payout from a win was described as 2:1. Specifically, participants were told, “if the ball lands on the color you wagered on, your bet will double,” such that a $2 bet would pay $4 following a win. If the ball landed on the color, the participants did not wager on, they were told they would lose their bet.
The amount of money gambled was assessed as a $1–$10 continuous dependent measure. After placing their wager, the session was ended and participants were fully debriefed.
Results
One-way analysis of variance revealed that participants in the economic threat condition (M = 4.00, SD = 1.19) wagered more than those in the control condition (M = 2.86, SD = 1.39), F(1, 43) = 5.09, p = .03, η2 = .11. In addition, participants believed that risk taking was needed to achieve financial gain more in the economic threat condition (M = 5.20, SD = .58) than in the control condition, (M = 4.32, SD = 1.61), F(1, 43) = 6.10, p = .02, η2 = .12. There was also a main effect of condition on negative affect, F(1, 43) = 19.16, p < .001, η2 = .31. Participants reported more negative affect in the economic threat condition (M = 3.22, SD = 1.18) than in the control condition (M = 1.87, SD = .85). There was no difference between the economic threat and control condition in reported positive affect (M = 3.90, SD = .83 and M = 3.72, SD = 1.21, respectively), F(1, 43) = .31, p = .58, η2 = .007. Finally, both perceived need to take financial risks and negative affect predicted amount wagered, B = .58, standard error (SE) = .19, t(43) = 2.99, p = .005 and B = .45, SE = .21, t(43) = 2.17, p = .04, respectively.
A multiple mediation model was then tested using Preacher and Hayes (2008) bootstrapping technique (with 1,000 iterations) to determine whether the indirect effect of economic threat (coded as 0 = control, 1 = economic threat) on amount wagered, via financial risk taking or negative affect, was significantly different than zero (see Figure 1). The indirect effect through financial risk taking was estimated to lie between .11 and 1.00 with 95% confidence. Because zero is not in the 95% confidence interval, the indirect effect is indeed significantly different from zero at p < .05 (two-tailed). However, the indirect effect through negative affect was estimated to lie between −.40 and 1.06 with 95% confidence. Thus, this indirect effect was not significant because the confidence interval included zero.
Mediation model with the economic threat manipulation as the independent variable (coded as 0 = control, 1 = economic threat), perceived need to engage financial risk taking and negative affect as mediators, and amount wagered as the dependent variable: Study 2. The unstandardized coefficients and standard error shown in parentheses reflect the inclusion of the mediators in the equation. Unstandardized coefficients and standard error with an asterisk indicate a significant path, p < .05.
Discussion
The results of Experiment 2 conceptually replicate and extend the findings obtained in Experiment 1. First, we showed that participants are not only willing to gamble more readily under economic threat, but they wager more when given the opportunity. That is, participants wagered more of their $10 participant remuneration in the economic threat condition than in the control condition. Second, a mechanism by which economic threat increases gambling behavior was revealed. Specifically, we found that economic threat elevated gambling, and it did so by encouraging the belief that risk taking is necessary to achieve financial gains. Thus, when poor economic prospects are salient, people are more likely to believe that making money is only possible when financial risks are taken (i.e., when traditional moneymaking routes are not available, alternate routes need to be considered). Put another way, the reason why participants were more willing to gamble when under threat was because that threat underscored the need to take financial risks. Importantly, Experiment 2 eliminated negative affect as an alternative path through which economic threat facilitates gambling.
General Discussion
We tested whether financial risk taking in the form of gambling occurs when economic threat is salient and found evidence for this contention. From a rational choice perspective, a poor economic outlook should make gambling less attractive (i.e., reduce willingness to engage in games of chance), but the findings from this research demonstrate otherwise. Reminding young adults that due to the economic downturn investing their money in education would yield lower returns than in the past resulted in a greater willingness to gamble (Experiment 1) and higher wagers when given a gambling opportunity (Experiment 2).
Past research has shown that during the early 1990s, consumers began to curtail the exorbitant spending habits that characterized the 1980s. Johnson (1994) argued the major cause of this shift was the recession of the mid-1990s and the corresponding increase in unemployment. During the recent 2008 recession, a similar pattern of reduced consumer spending was observed (Hampson & McGoldrick, 2013). We hypothesized that while consumer spending might generally decrease in times of economic woe, willingness to gamble might increase as a means to achieve financial gains. Specifically, economic threat should lead people to perceive they must take financial risks to make money, which results in increased willingness to gamble. In the current research, we tested this avenue to monetary gain by telling participants their economic outlook was poor. As predicted, our participants were more likely to take the opportunity to gamble with their money when they felt their economic future was under threat. Importantly, we showed that poor economic conditions lead to increased gambling due to the perception that risk taking is necessary to make money.
Our results suggest that a poor economic forecast can be a catalyst for gambling, and provides insight into why the gambling industry witnessed an increase in revenue during the 2008 recession (see CGA, 2008, 2011). We argue, like Blalock, Just, and Simon (2007), that gambling provides hope of financial security in economically insecure times—a need that is apt to facilitate risk taking (see Mishra & Fiddick, 2012; Weber et al., 2004). Because people tend to feel financially deprived when their buying power shrinks (see Liang & Fairchild, 1979), gambling may be one avenue that people perceive as capable of alleviating their feelings of financial deprivation (Callan et al., 2008, 2011)—if they win, then their financial picture will improve. Indeed, we showed the effect of economic threat on gambling was mediated by perceptions that risk taking is required to make money. We believe our research is the first to show not only that the salience of economic threat can lead to gambling but also the mechanism by which poor economic conditions elevate willingness to gamble.
Caveats and Directions for Future Research
It is, of course, possible that the results of our studies are restricted to a student population and economic threats that stem from increasing difficulty to secure student loans and expecting a poor job market. Given that constriction of credit and unemployment are issues potentially faced by every segment of the population, this possibility seems unlikely. It is also possible that participants were more likely to gamble because the experimenter provided the money they used to gamble. While people may well be more inclined to gamble with “house money,” this does not explain the between-condition differences in risk taking with these funds that we observed.
It should also be noted that gambling under economic threat might be influenced by a number of additional contextual factors (see Bromiley, 2009, 2010). First, gambling might be reduced if the gamble can either exceed or fall short of a person’s financial goal or reference point, and a safe option will be certain to reach the goal (Camerer & Weber, 1992). In such a case, falling short of the reference point or goal is unattractive due to loss aversion, and people will subsequently make more risk-averse decisions. Conversely, high-risk gambles that provide the opportunity to reach one’s reference point will be preferred over safer alternatives that fall short of the reference point. In brief, risk-seeking behavior in games of chance is dependent upon both the alternatives provided in a gamble and the ability of those alternatives to help individuals reach their goal. Second, people may differ in other salient reference points (e.g., degree of media exposure to wealthy people or interpersonal comparisons with one’s more affluent neighbors and coworkers; Loewenstein, Thompson, & Bazerman, 1989) that may create feelings of deprivation and an increased preference for gambling. Finally, the current research focused on only one possible outcome of economic threat, that is, increased financial risk taking through gambling. Future research should vary the expected value of the risk to assess possible variation in risk taking under economic threat. It would also be prudent to provide participants with options other than taking the monetary remuneration or gambling. Although we would anticipate that people would be more likely to take the monetary remuneration if economic threat increased a general desire to spend, we did not directly provide participants with the option to purchase any goods. To directly test this possibility, a method could be employed that includes a “purchasing” option (e.g., using the $10 remuneration to buy a book or a video game, or provide an option to win an iTunes gift card).
Conclusion
These studies provide evidence that at least one form of financial risk taking—playing games of chance—increases when economic threat is salient. Nevertheless, poor economic prospects may not elevate the desire to engage in other forms of risk taking. Gambling is a unique risk-taking behavior as it provides the prospect of a financial windfall. Unfortunately, it provides a poor return on investment. Thus, periods of macroeconomic hardship not only decrease available monies and job prospects for many people but can also lead to risky and detrimental financial decisions with the money currently in their possession.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by an Ontario Problem Gambling Research Centre (OPGRC) Investigator Support Grant (#3409) to Wohl.