Journal of Economic Psychology 33 (2012) 10591069

Contents lists available at SciVerse ScienceDirect

Journal of Economic Psychology

journal homepage: www.elsevier.com/locate/joep

The effect of anger and anxiety traits on investment decisions

Elisa Gambetti , Fiorella Giusberti
Department of Psychology, University of Bologna, Italy

a r t i c l e

i n f o

a b s t r a c t
This study investigates the extent to which people make nancial decisions on the basis of their dispositional tendency to engage in a specic emotion, such as anger or anxiety. We predicted that trait anger is associated with the decision to invest, whereas trait anxiety motivates individuals to avoid investments. We employed a six question survey, considering real life investment decisions, stock trend predictability and preference toward risk investments, and three hypothetic scenarios to measure the participants risk attitudes in the area of nance. The results showed that trait anger predicted risky decisions: it was positively associated with the willingness to invest money in different kinds of stocks, preferring medium/long-term investments, and with high predictability assessment in the forecast of stock trends. Contrarily, trait anxiety predicted conservative nancial decisions: it was associated with the decision not to invest savings, to hold interest-bearing accounts, and with low predictability of stock trends. In hypothetic scenarios trait anger predicted a medium risk portfolio and the decision to wait before selling both loss and gain investments, while trait anxiety was associated with the preference for a low risk portfolio and with the decision to immediately sell a stock both if it increases or decreases in value. These data are consistent with cognitive models of emotions, highlighting their functional utility and extend the knowledge of the relationship between personality traits and real life investment decision-making. 2012 Elsevier B.V. All rights reserved.

Article history: Received 2 January 2012 Received in revised form 9 July 2012 Accepted 10 July 2012 Available online 20 July 2012 JEL classication code: 3120-personality traits & processes PsycINFO classication code: D14-personal nance Keywords: Anger Anxiety Personality traits Economic decision-making Personal nance

1. Introduction To successfully navigate the nancial world, an individual must be able to distinguish and respond appropriately to different situations which require more proactive, but potentially costly, strategies or more passive, conservative decisions. Emotions such as anxiety and anger serve these different functions: anger has been dened as an emotional state that consists of feeling that varies in intensity, from mild irritation or annoyance to fury and rage (Spielberger & Sydeman, 1994) and it tends to promote approach and proactive behavior in the form of attack (e.g., Lerner & Tiedens, 2006). On the other hand, anxiety has been dened as an emotional response involving unpleasant feelings of tension, apprehensive and worried thoughts and it prompts avoidant and conservative behavior (Raghunathan & Pham, 1999; Wilt, Oehlberg, & Revelle, 2011). Since emotions have developed through evolutionary processes, they are often functional and important in both the assessment of situations and the consequent decisions (e.g., Finucane, Alhakami, Slovic, & Johnson, 2000; Hogarth, Portell, Cuxart, & Kolev, 2011). Previous research has found that emotional states can alter peoples goals, attitudes and perception, (e.g., Forgas, 2000; Zajonc, 2000). It has been widely accepted that decision making can be inuenced by emotional states, involving differences in the way individuals appraise events (e.g., Slovic, Finucane, Peters, & MacGregor, 2004). According to cognitive models (Beck, 1999; Eysenck, 1997), emotions are supported and regulated by a variety of cognitive processes: no emotion can
Corresponding author. Address: Department of Psychology, University of Bologna, Viale Berti Pichat 5, PO Box 40127, Bologna, Italy. Tel.: +39 051 2091885; fax: +39 051 243086. E-mail address: e.gambetti@unibo.it (E. Gambetti).
0167-4870/$ - see front matter 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.joep.2012.07.001

1060

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

be without a cognitive appraisal attributing a meaning to situations. Peters, Vstfll, Grling, and Slovic (2006) suggested that cognitive assessments of the environment, activated by specic emotions, have an important inuence on decision making: they act as informational, motivational and processing functions and so they have specic impacts on outcome effects. Previous studies have shown that anger increases the tendency to perceive situations as predictable, comprehensible and under individual control (Ellsworth, Scherer, Davidson, Scherer, & Goldsmith, 2003), and it increases optimism and feelings of invulnerability (Quigley & Tedeschi, 1996). As a consequence, anger is related to the perception of low risk across new situations (e.g., Lowenstein & Lerner, 2003). On the other hand, anxiety is linked to an attentional bias toward threat-related information and to evaluating ambiguous stimuli as negative ones (e.g., Bar-Haim, Lamy, & Glickman, 2005; Gu, Ge, Jiang, & Luo, 2010). Moreover, anxiety is associated with the perception of uncertainty, unpleasantness and low situational control (e.g. Smith & Ellsworth, 1985). In this sense, when an individual feels anxious the perception of risk across a situation increases, because of a disproportional dwelling on loss outcomes. In the assessment of emotions, it is important to distinguish between the intensity of the experience of the emotional states and the individual differences in the tendency to react with these specic emotions across time and situations, that is personality traits (e.g., Lazarus, 1994). Over the last 15 years there have been an increasing number of studies about the inuence that personality traits have on nancial perception and decisions (e.g., Gibson, 2006). Traditionally, economic theory implies that investment decisions should be based on expected utility. That is, the best decisions are those that maximize the expected utility of the money obtained. Economists dene risk as an objective index that has to be used when deciding on which solution to invest in. They highlight the importance of looking at the risk/return ratio of the selected investment. From this point of view, investment risk is the variance of stock utility (e.g., Rabin & Thaler, 2001). However, several studies showed that people do not assess investment risk objectively (e.g., Parker & Fischhoff, 2005). Research in judgment and decision making has shown that risk perception, that is the intuitive judgment about the occurrence of negative outcomes and the severity of the associated consequences, and risk acceptance, which involves the subjective balancing of benets with risks and their acceptability, are inuenced by many other factors aside from the utility of money, such as personality traits. A variety of studies have attempted to explore types and traits correlated with investment perception and behavior. For example, Carducci and Wong (1998) found that persons with a Type A personality are more willing to take higher levels of risk in all nancial matters than Type B individuals. There is also evidence of a desire for sensation seeking by some persons in terms of their nancial management (Wong & Carducci, 1991). These personality characteristics, in combination with specic socioeconomic background (i.e. being male, older, married, professionally employed with higher incomes, more education, more nancial knowledge, and increased economic expectations), predicted risk acceptance, tolerating declines in the investments prices while waiting for them to increase in value, in everyday money matters (Grable, 2000). Fenton OCreevy, Nicholson, Soane, and Willman (2004) found that successful professional traders for European investment banks, who have likely high levels of risk acceptance, tend to be emotionally stable and open to new experiences. Extraversion and conscientiousness were both found to be positively related to short-term investment intentions (Mayeld, Perdue, & Wooten, 2008). On the other hand, other authors have concluded, without studying large samples of traders, that personality traits are themselves not important for trading and investment decisions (e.g., Lo & Repin, 2005). Regarding the anger and anxiety traits, there is evidence suggesting a relationship between trait anger and risky decisions in hypothetic nancial, social and health scenarios (e.g., Gambetti & Giusberti, 2009), as well as a relationship between individual differences in trait anxiety, worry, and social anxiety and risk-avoidance in a behavioral risk-taking task (e.g., Maner et al., 2007). Recently, a study found a signicant negative correlation between trait anxiety and investment behavior, but only in the case of the immediate resolution of risk (van Winden, Krawczyk, & Hopfensitz, 2011). In general, although the literature has shown that anger and anxiety traits predict risk perception and risk acceptance, few studies have investigated the relationship between such personality traits and real life investment decisions. Research in this eld is in fact primarily experimental and there is a general disagreement about the level of external validity of the results because this kind of studies tested participants in articial conditions that bear little resemblance to the outside world (e.g., Gigerenzer, 2008). Thus, it is important to further investigate whether anger and anxiety traits play different roles in investment perception and decision in real domains of human life. 1.1. The current study Some studies suggest that personality traits resemble momentary emotions in important ways and should thus yield similar effects on judgments and decisions (e.g., Gross, Sutton, & Ketelaar, 1998). Since the evaluation of an expected outcome is an important step towards decision making (e.g., Paulus, 2005), individual differences in anger or anxiety predispose respectively to a positive or a negative outcome evaluation (e.g., Fischhoff, Gonzalez, Lerner, & Small, 2005; Gu et al., 2010). This kind of codication is a key to adjust the assessment of preferences among possible options of the subsequent decision making. As cited above, trait anger and trait anxiety appear to predispose, respectively, to risky and to avoidant decisions (e.g., Mitte, 2007). However, the impact of individual differences in anger or anxiety-proneness on real life nancial decisions has till now been noticeably understudied, even though they are particularly noteworthy: they may be unconscious and unrelated to the decision at hand, nonetheless they can have the potential to inuence the perception of nancial products and the consequent decisions in important ways. The main purpose of the current study was to evaluate the relationship between anger and anxiety traits and nancial behavior, considering real life investment decisions. We conducted an experiment in which participants were confronted

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

1061

with their actual investment choices and their preferences, in the recent past and at present, checking for demographic characteristics (such as age, education, gender, marital status, employment status and income), experience in economic/nancial topics and amount of savings. We supposed that trait anger would predict risky decisions, that is investments with high risk and therefore high expected returns. On the other hand, trait anxiety would predict conservative decisions, that is sound investments for savings. In addition, we sought to evaluate the stock trend predictability, that is, how much a person believes the uctuation in price of an investment to be predictable (e.g., to what extent is the value of a specic share or state bond in the market foreseeable?). In such terms, this factor measures the sense of control regarding the possibility to forecast the variation in trend of stocks. The appraisal-tendency theory (Lerner & Keltner, 2000) suggests that anger promotes the perception of situations as predictable and under individual control: on this basis, we expected that trait anger would activate a cognitive assessment of high stock trend predictability. On the contrary, given that anxiety is linked to a tendency to avoid potentially threatening decisions (i.e., risky ones), a perception of high uncertainty and low personal control over a situation (e.g. Frijda, 1986), we expected that trait anxiety would activate a perception of low stock trend predictability. Finally, we were interested in risk attitude. In line with the potential goal expected from highly anxious individuals to reduce uncertainty (e.g., Bensi & Giusberti, 2007) and to protect themselves from potential threats (Maner & Schmidt, 2006), we expected that, in hypothetic nancial scenarios, anxious individuals would be inclined to immediately sell a stock both if it increases or decreases in value, with the goal to, respectively, be sure to obtain a gain or avoid even more losses. Conversely, on the basis of studies highlighting that anger leads to increased risk taking and optimism (e.g., Lerner & Tiedens, 2006; Quigley & Tedeschi, 1996), we expected that angry individuals, in hypothetic situations, would be prone to wait before selling both if the value of an investment rises or decreases in an attempt to, respectively, increase even more their gains or recover from a temporary loss situation. 2. Material and methods 2.1. Participants Two hundred and fourteen individuals (114 women) were recruited for the study at work (i.e., ofces, shops, stores, university and banks). The average age of the sample was 38.8 years (SD = 11.7), ranging from 20 to 60 years. All participants provided informed consent prior to the study and their participation was voluntary. Demographic information for the sample is given in Table 1. Participants were asked to rate on a 4-point rating scale their experience in economic/nancial topics. 32.7% of the sample was inexperienced, 22.4% had no specic training but in the past had learned something from personal experience, 7.5% were interested in these topics and kept themselves abreast of investments and 55.1% studied nancial topics or worked in the eld of economics. Finally, participants were asked to indicate on a scale ranging from 0 to 100 how much money they saved in a year. This measure indicated the perception of money supply that could possibly be used for investments.

Table 1 Demographic information for the sample (N = 214). Variable Marital status Single Married Separated/divorced Employment status Student Employed Manager Self-employed Unemployed/homemaker Pensioner Did not respond Education Primary school Secondary/high school Bachelors degree Graduate degree Household income Less than 10,000 Euros 10,00020,000 Euros 20,00040,000 Euros More than 40,000 Euros Did not respond N (%) 104 (48.6%) 92 (43%) 18 (8.4%) 14 (6.9%) 108 (50%) 16 (7.5%) 44 (20.6%) 10 (4.7%) 10 (4.7%) 12 (5.6%) 8 (3.7%) 78 (36.4%) 60 (28%) 68 (31.8%) 34 (15.9%) 90 (42%) 60 (28.1%) 24 (11.2%) 6 (2.8%)

1062

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

2.2. Measurements of personality traits To evaluate anger, we used the Italian version of the State-Trait Anger Expression Inventory-2 (STAXI-2; Comunian, 2004) that used for trait anger a 10-item global scale (see the Appendix). The psychometric qualities of the original and translated scale have been examined in a large number of studies (e.g., Forgays, Forgays, & Spielberger, 1997), demonstrating strong internal consistency, and they have been validated against various indices of anger-related physiological arousal and other self-report measures of anger and hostility (e.g., Spielberger, 1999). Cronbachs alpha coefcient was 0.80 in the present sample. To assess trait anxiety we used the Italian version of the STAI-Y2 form (Pedrabissi & Santinello, 1989). Details are provided in the Appendix. The 20-item version of this scale have demonstrated good internal consistency as well as convergent and discriminant validity (e.g., Spielberger & Sydeman, 1994). Cronbachs alpha coefcient was 0.82 in the present sample. 2.3. Investment questionnaire The questionnaire included two parts to investigate the participants investment perceptions and decisions. Part one had six questions. Three questions (A, B, D) concerned real life investment decisions. Question C concerned stock trend predictability and questions E and F concerned the subjects preferences toward risk investment. Question A Have you ever invested your money? had a dichotomous answer, and question B What kind of investment have you chosen so far? targeted six kinds of investments: deposit books, shares, state bonds, industrials, property bonds and insurance products. Participants were asked, for each kind of investment, to rate again on a 10-point scale their stock trend predictability even if they had never invested (question C How much do you believe the trend of the investment to be predictable?). Question D How much is the amount of your investments in the last 24 months? involved a 4-point rating scale ranging from less than 5000 Euros to more than 20,000 Euros. Question E, What kind of investment do you prefer?, had a 5-point rating scale ranging from until 12 months to more than 10 years. Question F, What is the best investment for you?, involved the following answers (a) 12-month bond with interest rate of 2%, (b) 6-month bond with interest rate of 1.5%, (c) interest-bearing account with an interest rate of 1%. Part two consisted of three hypothetic scenarios to measure the participants risk attitudes in the nancial area. The rst situation was: Your bank offers you different security portfolios with the following characteristics: low gain/no loss, medium gain/medium loss, high gain/high loss. Which portfolio do you choose?. The second situation was: You have shares. What do you do if the stock declines? and had the following answers: (a) sell at a loss, (b) wait some days with the possibility to lose or to advance, (c) wait some weeks with the possibility to lose further or advance. The third situation was the same as the second scenario but in a gain context. 2.4. Procedure Participants completed the STAXI-2 Trait Anger scale and the STAI-Y2 in a counterbalanced manner, that is, either before or after completing the investment questionnaire. They were encouraged to answer as honestly as possible, told that there were no right or wrong answers, and assured that their answers would remain condential. The scenarios were presented randomly to eliminate the order effect. 3. Results For each item of the investment questionnaire we performed a descriptive analysis on the frequency of answers and a hierarchical multiple linear regression analysis, considering as independent variables in step one, the demographic variables (i.e., age, gender, education, marital status, employment status and household income), experience in economic/nancial topics, savings, and in step two the personality traits (i.e., trait anger and trait anxiety). 3.1. Investment decision 32% of the sample had never invested money, 68% had decided to invest. The results of the regression analysis showed that the investment decision was predicted by age and experience in step one (F(8, 192) = 9.84, p < 0.01, adjusted R2 = 0.09) and by both trait anxiety and trait anger in step two (F(2, 194) = 5.69, p < 0.01, R2 change = 0.06). Table 2 summarizes the regression coefcients from this analysis. 3.2. Savers and investors The answers to question B indicated that 37% of the sample, the savers, hold a current account and/or deposit books, whereas 63% of the sample, the investors, invested money in shares (32.7%), industrials (36.4%), state bonds (29.0%), insurance products (23.4%) and property bonds (5.6%). Regression analysis showed that the way to invest money was predicted by age, household income, savings and experience in step one (F(8, 195) = 12.54, p < 0.01, adjusted R2 = 0.10) and by both trait

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069 Table 2 Predictor variables, b coefcients and t values from hierarchical multiple linear regressions of A and B questions. Question A b Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety 0.28 0.05 0.06 0.05 0.04 0.16 0.29 0.13 0.28 0.21 t 2.50* 0.70 0.79 0.54 0.47 1.75 4.33** 1.81 4.10** 2.15* Question B b 0.38 0.03 0.06 0.11 0.09 0.20 0.32 0.16 0.30 0.21 t 3.61** 0.56 0.89 1.33 1.22 2.30* 4.80** 2.24* 4.46** 2.16*

1063

Notes: Regressions of question A Have you ever invested your money? and question B What kind of investment have you chosen so far? on demographic variables, experience in economic/nancial topics and savings (step 1) and personality traits (step 2). The dependent variables are dichotomous: question A (0 = no, 1 = yes), question B (1 = saver, 2 = investor). * 5% Signicance. ** 1% Signicance.

Table 3 Personality traits and the stock trend predictability of each kind of investment (i.e., deposit books, shares, state bonds, industrials, property bonds and insurance products). b (a) Dependent variable: stock trend predictability of deposit books Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety (b) Dependent variable: stock trend predictability of shares Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety (c) Dependent variable: stock trend predictability of state bonds Step 1 Age Gender Education Marital status t F 7.78** 0.20 0.01 0.05 0.04 0.10 0.14 0.49 0.31 0.32 0.38 1.79 0.12 0.67 0.42 1.24 1.41 6.46** 4.28** 21.56** 4.42** 5.13** 0.34 Adjusted R2 0.24

1.23 0.37 0.01 0.01 0.08 0.11 0.17 0.04 0.03 0.19 0.14 2.93** 0.16 0.17 0.82 1.20 1.58 0.39 0.39 4.23* 2.45* 2.12*

0.03

0.04

2.95** 0.11 0.09 0.08 0.03 0.90 1.14 0.94 0.30

0.08

(continued on next page)

1064 Table 3 (continued)

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

b Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety (d) Dependent variable: stock trend predictability of industrials Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety (e) Dependent variable: stock trend predictability of property bonds Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety (f) Dependent variable: stock trend predictability of insurance products Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety 0.14 0.05 0.26 0.16 0.25 0.27

t 1.62 0.47 3.22** 2.03*

Adjusted R2

11.16** 3.29** 3.52**

0.14

4.69** 0.18 0.03 0.04 0.01 0.15 0.07 0.24 0.37 0.41 0.29 1.56 0.38 0.48 0.10 1.72 0.74 3.05** 4.90** 13.73** 5.55** 3.93**

0.14

0.29

2.95** 0.11 0.09 0.08 0.03 0.14 0.05 0.26 0.16 0.35 0.30 0.89 1.14 0.94 0.29 1.62 0.47 3.22** 2.03* 8.48** 4.57** 3.85**

0.08

0.14

2.39* 0.01 0.07 0.01 0.05 0.09 0.03 0.28 0.11 0.26 0.16 0.08 0.86 0.09 0.53 1.06 0.30 3.42** 1.39 9.90** 3.34** 2.05*

0.06

0.13

Notes: Regressions of question C How much do you believe the trend of the investment to be predictable? on demographic variables, experience in economic/nancial topics and savings (step 1) and personality traits (step 2). Dependent variables take values 0100. * 5% Signicance. ** 1% Signicance.

anxiety and trait anger in step two: F(2, 199) = 7.41, p < 0.01, R2 change = 0.03. In particular, as shown in Table 2, trait anxiety predicted holding current account and/or deposit books, whereas trait anger predicted the proneness to invest money in different kinds of stocks. 3.3. Stock trend predictability To evaluate the relation between personality traits and the stock trend predictability, we performed a series of regression analyses on the question C scores. The results are shown in Table 3 and reveal that the stock trend predictability of each kind of investment was positively associated with trait anger and negatively with trait anxiety.

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069 Table 4 Predictor variables, b coefcients and t values from hierarchical multiple linear regressions of D, E, F questions. Question D b Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety 0.42 0.07 0.03 0.09 0.01 0.11 0.18 0.23 0.29 0.19 t 3.87** 1.02 0.40 1.05 0.05 1.24 2.51* 3.23** 3.63** 2.26* Question E b 0.17 0.22 0.09 0.12 0.03 0.02 0.04 0.12 0.15 0.04 t 2.12* 2.79* 1.23 1.23 0.39 0.19 0.53 1.55 2.68* 0.41 Question F b 0.02 0.21 0.16 0.06 0.01 0.25 0.15 0.05 0.22 0.20 t

1065

0.14 2.67* 2.07* 0.63 0.17 2.51* 1.88 0.66 2.87** 2.59**

Notes: Regressions of question D How much is the amount of your investments in the last 24 months?, question E What kind of investment do you prefer? and question F What is the best investment for you? on demographic variables, experience in economic/nancial topics and savings (step 1) and personality traits (step 2). The dependent variables ranged from 1 (0 Euros) to 4 (more than 20,000 Euros) for question D, from 1 (until 12 months) to 5 (more than 10 years) for question E, from 1 (12-month bond with interest rate of 2%) to 3 (interest-bearing account with an interest rate of 1%) for question F. * 5% Signicance. ** 1% Signicance.

3.4. Amount of investments The amount of investments in the last 24 months was zero for 34.6% of the sample, less than 5.000 Euros for 27.1%, from 5000 to 10,000 Euros for 15.9%, from 10,000 to 20,000 for 7.5%, and more than 20,000 Euros for 11.7%. 3.2% did not answer question D. Regression analysis showed that the scores in question D were signicantly predicted by age, experience and savings in step one: F(8, 206) = 5.25, p < 0.01, adjusted R2 = 0.21. The amount of investments was positively associated with trait anger and negatively to trait anxiety in step two: F(2, 179) = 4.46, p < 0.01, R2 change = 0.03. Table 4 summarizes the regression coefcients from this analysis. 3.5. Duration of investments 1.1% of the sample did not answer question E, 21.1% preferred investments until 12 months, 39.4% 13 years, 24% 3 5 years, 7.7% 510 years and 6.7% more than 10 years. We performed a regression analysis on the investment duration. As shown in Table 4, the scores in question E were signicantly predicted by age and gender (females were more prone to choosing long-term investments) in step one: F(8, 206) = 2.17, p < 0.05, adjusted R2 = 0.21. Only trait anger predicted the duration of investments in step two: F(2, 205) = 4.12, p < 0.01, R2 change = 0.22. Trait anxiety was excluded by the regression model. 3.6. Investment preferences 54.2% of the sample preferred 12-month bonds with an interest rate of 2%, 19.6% 6-month investments with an interest rate of 1.5%, and 21.5% an interest-bearing account with an interest rate of 1%. 4.7% of the sample did not answer question F. Regression analysis (see Table 4) showed that gender (female preferred an interest-bearing account), education and household income predicted scores in question F in step one: F(8, 206) = 3.01, p < 0.05, adjusted R2 = 0.08. Trait anger predicted the preference for 12-month bonds with an interest rate of 2%, while trait anxiety predicted the preference for an interest-bearing account with an interest rate of 1% in step two: F(2, 200) = 5.21, p < 0.01, R2 change = 0.41. 3.7. Low risk vs. medium risk As regards scenario 1, 45.7% of the sample chose a portfolio with low gain and no loss, 52.4% with medium gain and medium loss, and 1.9% with high gain and high loss (ceiling effect). Regression analysis showed that the scores in scenario 1 were signicantly predicted by marital status (singles chose a portfolio with low risk whereas married subjects preferred a portfolio with medium risk), experience and savings in step one: F(8, 206) = 8.80, p < 0.01, adjusted R2 = 0.10. Trait anxiety predicted the preference for a portfolio with low risk, whereas trait anger predicted the preference for a portfolio with medium risk in step two: F(2, 200) = 9.65, p < 0.01, R2 change = 0.01. Table 5 summarizes the regression coefcients from this analysis.

1066

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

Table 5 Predictor variables, b coefcients and t values from hierarchical multiple linear regressions of the three scenarios. Scenario 1 b Step 1 Age Gender Education Marital status Employment status Household income Experience Savings Step 2 Trait anger Trait anxiety 0.07 0.02 0.07 0.36 0.17 0.02 0.33 0.21 0.15 0.13 t 0.62 0.21 0.87 3.69** 1.94 0.26 4.87** 2.77* 2.15* 2.03* Scenario 2 b 0.05 0.01 0.03 0.30 0.22 0.40 0.27 0.05 0.15 0.29 t 0.46 0.18 0.42 3.30** 2.73 4.38** 3.98** 0.70 2.25* 4.16** Scenario 3 b 0.16 0.01 0.11 0.31 0.12 0.01 0.14 0.13 0.21 0.16 t 2.09* 0.19 1.36 3.18* 1.42 0.13 1.95* 1.88 3.75** 2.15*

Notes: Regressions of the three scenarios on demographic variables, experience in economic/nancial topics and savings (step 1) and personality traits (step 2). The dependent variable is dichotomous in scenario 1 (1 = low gain/no loss, 2 = medium gain/medium loss), ranged from 1 (sell shares at a loss) to 3 (wait some weeks) in scenario 2, and from 1 (sell shares on the rise) to 3 (wait some weeks) in scenario 3. * 5% Signicance. ** 1% Signicance.

3.8. Loss investments As regards scenario 2, 25% of the sample decided to immediately sell the shares at a loss, 24% chose to wait some days with the possibility to lose or advance, 51% preferred to wait some weeks with the possibility to further lose or advance. We performed a regression analysis (see Table 5) showing that the quickness of selling loss investments was signicantly predicted by marital status, household income and experience in step one: F(8, 206) = 10.54, p < 0.01, adjusted R2 = 0.13. Trait anger predicted the decision to wait, while trait anxiety predicted the decision to immediately sell loss investments in step two: F(2, 210) = 9.02, p < 0.01, R2 change = 0.08. 3.9. Gain investments As regards scenario 3, 68.9% of the sample preferred to immediately sell shares on the rise, 23.3% chose to wait some days, and 7.8% chose to wait some weeks. As shown in Table 5, the scores in scenario 3 were signicantly predicted by age, marital status and experience in step one: F(8, 206) = 3.45, p < 0.05, adjusted R2 = 0.08. Trait anger predicted the decision to wait in shares on the rise and trait anxiety predicted the decision to sell gain investments in step two: F(2, 212) = 5.25, p < 0.01, R2 change = 0.01. 4. Discussion The current study has investigated the role of anger and anxiety traits in nancial decision-making. In agreement with our hypotheses, the results show that individual differences in the proneness to feel such emotions are strongly and specically linked with real life investment decisions, sense of control over stock trend and risk preferences and attitudes. As far as real life investments are concerned, trait anger predicts risky nancial decisions: it is positively associated with the tendency to invest money in different kinds of stocks (i.e., shares, industrials, state bonds, insurance products and property bonds) and with the amount of investments. On the other hand, trait anxiety predicts conservative nancial decisions. It is related to the decision not to invest savings and to hold a current account and/or deposit books; trait anxiety is also negatively associated with the amount of investment. These results are in accordance with studies showing that trait anger has a role in risk-taking decisions, perceiving high familiarity and low salience over hypothetic situations (Gambetti & Giusberti, 2009). It is clear that also in a real economic context trait anger promotes decision-making biases that increase ones perception of having control over situations and ones proneness for risky choices. On the contrary, trait anxiety has been reliably associated with a proneness to avoid risk-taking, giving a selective attentional bias towards threatening stimuli, and a low sense of personal control over a situation (Butler & Mathews, 1987; Frijda, 1986; Maner et al., 2007). In evolutionary terms, the main function of anxiety is to keep us away from cues that indicate the presence of a danger, possibly including potential threat (Nesse, 2005). As a consequence, anxious people should perceive higher uncertainty and greater probability of loss across situations (especially nancial investments) and therefore they appear inclined to make risk-avoidant decisions for their savings: in fact, they opt for interest-bearing accounts and not for investments. As expected, stock trend predictability is positively associated with trait anger and negatively associated with trait anxiety. This result is in line with studies suggesting that positive emotional states experienced while awaiting outcomes, such as hope and excitement, induce people to take risks and to be condent in their ability to evaluate investment options (e.g.,

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

1067

Kuhnen & Knutson, 2011), while negative emotions, such as worry and anxiety, have the opposite effects (van Winden et al., 2011). On this basis, anger can be considered a positive emotion, especially in the sense of motivating future behavior (Lerner & Tiedens, 2006). This notion echoes other research demonstrating the functional utility of emotions: whether the experience of an emotion is liked or disliked does not depend entirely on whether it is positive or negative but also on an appraisal of the situation in which it is experienced (e.g., Vstfjll & Grling, 2006). With respect to risk preferences, trait anger is positively associated with the investment duration, whereas trait anxiety is not signicantly associated with this item. Moreover, trait anger predicts medium-term investments (e.g. 12-month bonds), while trait anxiety predicts a preference for interest-bearing accounts with low interest rate but with the possibility to withdraw money at any time. Long term investments are investments which are made for more than 5 years and are the ones which bear low risks and offer high prot on the outcome. In the case of long term investments people usually do not have the opportunity to change the conditions of the investments as often as the market changes, so this kind of investment does not respond to market uctuations as quickly as short-term investments do, and the investor is kept safe. In general, the results showed that people are not aware of this fact and in fact only 14.4% of the sample chose long-term investments. This said, it could seem paradoxical that trait anger is quite frequently associated with a preference for both risky (medium-term) and safe (long-term) investment decisions, whereas trait anxiety is not associated with investment duration (that is, with longer and safer investments). Nevertheless, angry individuals could be willing to make denitive decisions for a solid period of time because they are not worried about the impossibility to withdraw their savings, thus showing a high perception of safety and control over their nancial choice. At the same time, anxious individuals perceive future situations as uncertain and unpredictable, therefore they could be afraid of not being able to withdraw their savings as much as they would like in case of need: in such terms, each investment decision, independently of its duration, is considered a risky decision. With regards risk attitudes, trait anger predicts a portfolio with medium risk and it is positively related with the decision to wait before selling both loss and gain investments. On the other hand, in line with our hypotheses, trait anxiety is associated with the preference for a portfolio with low gain and no loss, as well as with the decision to immediately sell both loss and gain investments. A possible explanation for these results is that trait anger activates a defensive optimism where the relevance and the impact of negative events are de-emphasized (Hemenover & Zhang, 2004). In particular, in both gain and loss contexts, trait anger predicts an attitude to invest even if the decision is risky, and to wait before selling, running the risk of losing money but also of gaining. Trait anger leads individuals, independently of the context, to underestimate the impact of the rst possibility (to lose money) and to overvalue the second hypothesis (to gain money). On the other hand, trait anxiety leads individuals to make decisions to save their money even if such action may bring about a certain loss. According to Bensi and Giusberti (2007), the goal of trait anxiety is to lower as much as possible any feelings of anxiety in an attempt to reduce uncertainty. Both in gain and loss situations, trait anxious people appear to pursue the goal to save their current capital by avoiding the uncertainty of the market, thus resolving their uneasiness even at the cost of a loss, and gaining a greater sense of safety and control over potential threats. Finally, regarding the demographic variables, experience was the factor that regularly predicts real life investment decisions (with the exception of risk preferences and investment duration), such as stock trend predictability and risk attitude. Such datum is in line with the studies suggesting that respondents with relatively more investment experience had more risk-tolerant responses and higher-risk portfolios than less experienced investors (e.g., Chen & Corter, 2005). This result is quite signicant if we consider that we found trait anxiety to be negatively associated with experience. In such terms, the more a person is anxious, the less probability there is to gain experience in economic decisions. On this basis, anxiety seems to predispose to risk-avoidant behavior in the nancial eld which, in turn, reduces the possibility to gain knowledge and consequently contributes to the choice not to invest, causing a vicious circle. 4.1. Limitations and future research Some limitations can be identied in this study that remain to be considered in future work. First, the correlation between individual differences in the propensity to feel angry and anxiety feelings and investment decisions could be spurious due to a linkage with another factors, for example, a person who is low on the ability to inhibit or self-relate would also be likely to make impulsive decisions and take risks; thus more research is needed. Second, as state and trait emotions tend to be highly correlated, our results are consistent with the possibility that emotional states also contribute to investment decisions. Nevertheless, we did not manipulate emotions, so we cannot evaluate if and how this element facilitates nancial decision-making. Third, it should be considered that Italy was hit hard by the economic crisis of 20072011. The loss of economic resources, in fact, had a strong and close relationship to anxiety and anger (nal-Karagven, 2009) and consequently may inuence investment decisions leading people to be over prudent. Finally, the major limitation of this study is using hypothetical rather than actual scenarios in determining participants risk attitudes. Future studies could verify whether, also in actual investment situations, trait anger is predictive of a portfolio with medium risk whereas trait anxiety of a low risk portfolio. 5. Conclusions We claim that the results have potential implications for both theories of decision making and for investment companies or banks. Economists have suggested that the best decisions in the nancial eld are not linked to saving money, but to

1068

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

investing in a diversied portfolio for a medium-long term (e.g., Barber & Odean, 2000). In our study, trait anxious individuals did not risk investing money in low predictable stocks nor in those usually perceived as sound, such as property bonds, industrials or insurance products: trait anxious individuals seem to be over prudent whereas trait anger individuals make better decisions as regards economic gains. As our ndings suggest, personality traits may shape investment choices and decisions. These data t with a growing body of evidence suggesting that individual differences in affective experiences inuence risk decision-making (e.g., Maner et al., 2007; Slovic et al., 2004). Our study suggests the utility of a motivationbased approach to decision-making and is consistent with other theories positing emotion-specic inuences on judgments and decisions (e.g., Lerner & Keltner, 2000). These ndings also have practical implications for advisors or banks interested in identifying people most at risk for taking potentially risky decisions and, in turn, people not disposed to invest their money or to make investment decisions. Acknowledgement This paper is the result of research nancially supported by the Fondazione Cassa di Risparmio di Cesena. Appendix A A.1. Instructions for personality traits tests A number of statements which people have used to describe themselves are given below. Read each statement and then choose the appropriate number to the right of the statement to indicate how you generally feel. There are no right or wrong answers. Do not spend too much time on any one statement but give the answer which seems to describe how you generally feel. [answer on scale from 1 = almost never to 4 = almost always]. A.2. Trait anxiety (STAI-Y2; Pedrabissi & Santinello, 1989) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. I feel pleasant. I tire quickly. I feel like crying. I wish I could be as happy as others seem to be. I am losing out on things because I cant make up my mind soon enough. I feel rested. I am calm, cool and collected. I feel that difculties are piling up so that I cannot overcome them. I worry too much over something that really doesnt matter. I am happy. I am inclined to take things hard. I lack self-condence. I feel secure. I try to avoid facing a crisis or difculty. I feel blue. I am content. Some unimportant thought runs through my mind and bothers me. I take disappointments so keenly that I cant put them out of my mind. I am a steady person. I get in a state of tension or turmoil as I think over my recent concerns and interests.

A.3. Trait anger (STAXI-2; Comunian, 2004) 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. I am a hotheaded person. I am quick tempered. I am an impulsive person. I get angry when I have to wait because of others mistakes. I feel annoyed when I am not given recognition for job well done. I y off the handle. When I get mad, I say nasty things. I get angry when Im told Im wrong in front of others. When I am frustrated, I feel like hitting someone. I feel infuriated when I do a good job and get a poor evaluation.

E. Gambetti, F. Giusberti / Journal of Economic Psychology 33 (2012) 10591069

1069

References
Barber, B. M., & Odean, T. (2000). Trading is hazardous to your wealth: The common stock investment performance of individual investors. Journal of Finance, 55, 773806. Bar-Haim, Y., Lamy, D., & Glickman, S. (2005). Attentional bias in anxiety: A behavioral and ERP study. Brain and Cognition, 59, 1122. Beck, A. T. (1999). Prisoners of hate: The cognitive basis of anger, hostility, and violence. New York: Harper Collins. Bensi, L., & Giusberti, F. (2007). Trait anxiety and reasoning under uncertainty. Personality and Individual Differences, 43, 827838. Butler, G., & Mathews, A. (1987). Anticipatory anxiety and risk perceptions. Cognitive Therapy and Research, 11, 551565. Carducci, B. J., & Wong, A. S. (1998). Type A and risk taking in everyday money matters. Journal of Business and Psychology, 12, 355359. Chen, Y. J., & Corter, J. E. (2005). When mixed options are preferred in multiple-trial decision making. Journal of Behavioral Decision Making, 18, 126. Comunian, L. (2004). Manuale STAXI-2 State-Trait Anger Expression Inventory-2: Adattamento italiano. Firenze: Organizzazioni Speciali. Ellsworth, P. C., Scherer, K. R., Davidson, R. J., Scherer, K. R., & Goldsmith, H. H. (2003). Appraisal processes in emotion: Handbook of affective sciences. New York: Oxford University Press (pp. 572595). Eysenck, M. W. (1997). Anxiety and cognition: A unied theory. London: Erlbaum. Fenton OCreevy, M. N., Nicholson, E., Soane, P., & Willman, P. (2004). Traders: Risks, decisions, and management in nancial markets. Oxford: Oxford University Press. Finucane, M., Alhakami, A., Slovic, P., & Johnson, S. (2000). The affect heuristic in judgments of risks and benets. Journal of Behavioral Decision Making, 13, 117. Fischhoff, B., Gonzalez, R. M., Lerner, J. S., & Small, D. A. (2005). Evolving judgments of terror risks: Foresight, hindsight, and emotion. Journal of Experimental Psychology: Applied, 11, 124139. Forgas, J. P. (2000). Feeling and thinking: The role of affect in social cognition. Cambridge: Cambridge University Press. Forgays, D. G., Forgays, D. K., & Spielberger, C. D. (1997). Factor structure of the State-Trait Anger Expression Inventory. Journal of Personality Assessment, 69, 497507. Frijda, N. H. (1986). The emotions. Cambridge: Cambridge University Press. Gambetti, E., & Giusberti, F. (2009). Dispositional anger and risk decision-making. Mind and Society, 8, 720. Gibson, D. (2006). Emotions in nance. Distrust and uncertainty in global markets. Contemporary Sociology: A Journal of Reviews, 35, 483485. Gigerenzer, G. (2008). Rationality for mortals: How people cope with uncertainty. New York: Oxford University Press. Grable, J. E. (2000). Financial risk tolerance and additional factors that affect risk taking in everyday money matters. Journal of Business and Psychology, 14, 625630. Gross, J. J., Sutton, S. K., & Ketelaar, T. (1998). Relations between affect and personality: Support for the affect-level and affective-reactivity views. Personality and Social Psychology Bulletin, 24, 279288. Gu, R., Ge, Y., Jiang, Y., & Luo, Y. (2010). Anxiety and outcome evaluation: The good, the bad and the ambiguous. Biological Psychology, 85, 200206. Hemenover, S. H., & Zhang, S. (2004). Anger, personality and optimistic stress appraisals. Cognition and Emotions, 18, 363382. Hogarth, R. M., Portell, M., Cuxart, A., & Kolev, G. I. (2011). Emotion and reason in everyday risk perception. Journal of Behavioral Decision-Making, 24, 202222. Kuhnen, C. M., & Knutson, B. (2011). The inuence of affect on beliefs, preferences, and nancial decisions. Journal of Financial and Quantitative Analysis, 46, 605626. Lazarus, R. S. (1994). Universal antecedent of emotions. In P. Ekman & R. D. Davidson (Eds.), The nature of emotions: Fundamental questions (pp. 163171). New York: Oxford University Press. Lerner, J., & Keltner, D. (2000). Beyond valence: Toward a model of emotion-specic inuences on judgment and choice. Cognition and Emotion, 14, 473493. Lerner, J. S., & Tiedens, L. Z. (2006). Portrait of the angry decision maker: How appraisal tendencies shape angers inuence on cognition. Journal of Behavioural Decision Making, 19, 115137. Lo, A., & Repin, D. (2005). Fear and greed in nancial markets: A clinical study of day-traders. In Presentation to the Annual conference of the American Economics Association. Lowenstein, G., & Lerner, J. (2003). The role of affect in decision-making. In R. Davidson, K. Scherer, & H. Goldsmith (Eds.), Handbook of affective science. New York: Oxford University Press. Maner, J. K., Richey, J. A., Cromer, K., Mallott, M., Lejuez, C. W., Joiner, E., et al (2007). Dispositional anxiety and risk-avoidant decision-making. Personality and Individual Differences, 42, 665675. Maner, J. K., & Schmidt, N. B. (2006). The role of risk avoidance in anxiety. Behavior Therapy, 37, 181189. Mayeld, C., Perdue, G., & Wooten, K. (2008). Investment management and personality type. Financial Services Review, 17, 219236. Mitte, K. (2007). Anxiety and risk decision-making: The role of subjective probability and subjective cost of negative events. Personality and Individual Differences, 43, 243253. Nesse, R. M. (2005). Natural selection and the regulation of defences: A signal detection analysis of the smoke detector principle. Evolution and Human Behavior, 26, 88105. Parker, A., & Fischhoff, B. (2005). Decision-making competence. External validation through an individual-differences approach. Journal of Behavioral Decision Making, 18, 127. Paulus, M. P. (2005). Neurobiology of decision-making: Quo vadis? Cognitive Brain Research, 23(1), 210. Pedrabissi, L., & Santinello, M. (1989). STAI, State-Trait Anxiety Inventory, Forma Y: Manuale. Firenze: Organizzazioni Speciali. Peters, E., Vstfll, D., Grling, T., & Slovic, P. (2006). Affect and decision-making: a hot topic. Journal of Behavioral Decision Making, 19, 7985. Quigley, B. M., & Tedeschi, J. T. (1996). Mediating effects of blame attributions on feelings of anger. Personality and Social Psychology Bulletin, 22, 12801288. Rabin, M., & Thaler, R. H. (2001). Risk aversion. Journal of Economic Perspectives, 15, 219232. Raghunathan, R., & Pham, M. T. (1999). All negative moods are not equal: Motivational inuences of anxiety and sadness on decision making. Organizational Behavior and Human Decision Processes, 79, 5677. Slovic, P., Finucane, M. L., Peters, E., & MacGregor, D. G. (2004). Risk as analysis and risk as feelings: Some thoughts about affect, reason, risk and rationality. Risk Analysis, 24, 311322. Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotions. Journal of Personality and Social Psychology, 48, 813838. Spielberger, C. D. (1999). State-Trait Anger Expression Inventory-2: Professional manual. Odessa: Psychological Assessment Resources. Spielberger, C. D., & Sydeman, S. J. (1994). State-trait anxiety inventory and State-Trait Anger Expression Inventory. In M. E. Maruish (Ed.), The use of psychological testing for treatment planning and outcome assessment (pp. 292321). Hillsdale: Lawrence Erlbaum Associates. nal-Karagven, M. H. (2009). Psychological impact of an economic crisis: A conservation of resources approach. International Journal of Stress Management, 16, 177194. van Winden, F., Krawczyk, M., & Hopfensitz, A. (2011). Investment, resolution of risk, and the role of affect. Journal of Economic Psychology, 32, 918939. Vstfjll, D., & Grling, T. (2006). Preference for negative emotions. Emotion, 6, 326329. Wilt, J., Oehlberg, K., & Revelle, W. (2011). Anxiety in personality. Personality and Individual Differences, 50, 987993. Wong, A., & Carducci, B. J. (1991). Sensation seeking and nancial risk taking in everyday money matters. Journal of Business and Psychology, 5, 525530. Zajonc, R. B. (2000). Feeling and thinking: Closing the debate over the independence of affect. In Joseph P. Forgas (Ed.), Feeling and thinking: The role of affect in social cognition. Cambridge: Cambridge University Press.