![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
In this study, we investigate whether cross-country variation in investors’ mean openness to experience – one of the traits of the Five-Factor Model of personality – has an effect on the level of foreign bias present in international equity portfolios. Building on extant research showing a strong relationship between openness to experience and the attitude towards other cultures and risk, we argue that this personality trait is a construct that aggregates aspects of the psyche that influence the propensity to invest abroad. We find support for this hypothesis, showing that the more ‘open’ the country, the smaller the underrepresentation of foreign assets in its equity portfolio. Our analysis reveals that the impact of openness to experience is the most pronounced while making investments in countries with a weak information environment, or those that culturally or economically differ from the country the investor comes from, and suggests that international asset allocation is to some extent affected by psychological biases.
Acknowledgements
We are grateful to Andrew Vivian and two anonymous referees, whose insightful suggestions have substantially improved the paper. The authors also wish to thank Stuart Hyde and other participants of the third Emerging Scholars in Banking and Finance Conference that was held at Cass Business School in December 2012 for valuable comments.
Notes
1. These factors recurred in numerous personality ratings and thus led researchers in the 1980s to the conclusion that they are the fundamental dimensions of personality, that apply to people regardless of their age, sex, or origin. To give a general sense of what sort of behaviour can be linked to each of the dimensions that are not at the core of this study, it is advisable to present at least a few adjectives that describe individuals who score highly in each factor. For neuroticism, they include – anxious, unstable, and worrying; for extraversion – active, enthusiastic, and talkative; for agreeableness – forgiving, kind, and trusting; and for conscientiousness – organized, reliable, and thorough. We redirect the reader to McCrae and John (Citation1992) for a comprehensive introduction into the Five-Factor Model.
2. We thank an anonymous referee for suggesting the need to disentangle the two hypotheses and for valuable suggestions concerning how to robustly test them.
3. Interestingly, this preference is most evident in the case of companies from emerging economies, which might suggest a lesser degree of tolerance for the unfamiliar of investors in mature economies.
4. They refer to the Coordinated Portfolio Investment Survey (CPIS), which does not show the value of the holdings for a large portion of investor–recipient country pairs when the investment is made by a developing country.
5. McCrae and Terracciano (Citation2005) argue that the assessments made by members of cultures whose ‘Quality Index’ exceeds the threshold of 25 possess excellent psychometric properties. Non-Western countries exceed this threshold less often than Western countries.
6. This portfolio combines the total market capitalization of all public companies from the recipient (destination) countries used in this study.
7. Mean scores for each personality trait are assessed on the basis of the observer-rating Revised NEO Personality Inventory (NEO-PI-R), consisting of 240 items answered on a five point scale.
8. McCrae and Terracciano (Citation2005) show that three of the five personality traits are related to GDP per capita and the Human Development Index. We use the former variable (logged) in our study.
9. We use the turnover ratio as a proxy for market liquidity.
10. All of these measures are sourced from Thapa and Poshakwale (Citation2012).
11. This result is largely driven by the fact that developing countries forego investment opportunities in 53% of possible investor–recipient combinations, whereas developed countries only in 12% of such cases.
12. It is important to point out that although openness to experience is fairly strongly correlated with ‘GDP per capita in the investor country’ and ‘patriotism’, it is highly unlikely that the significance of our key variable is due to multicollinearity. This claim is supported by the fact that the variance inflation factors are below the thresholds recommended in the literature. Additionally, our findings are not adversely affected if we orthogonalize openness to experience to the potentially troublesome regressors, or simply drop them from the specifications (results available upon request).
13. The specification is the same as specification 4, bar the fact that it drops ‘uncertainty avoidance’ due to multicollinearity issues in some of the subsamples.
14. The standard errors are relatively large, possibly due to the large number of censored observations.
15. It is worth pointing out that the bilateral variables also measure the availability of information in the recipient country, but from the perspective of a specific investor country (i.e. in a relative, not absolute sense).
16. This procedure makes the coefficient for the openness to experience variable (the single term) interpretable.
17. We return to this point in the next section.
18. This does not, however, mean that in all instances when differences between the investor and recipient countries are very low, openness to experience has no effect on foreign bias. Additional regressions (unreported) suggest that the effect can be positive if the information environment of the recipient country is poor. Analogously, openness to experience can have a positive effect while investing in the least opaque countries if the economic or cultural differences between the investor and recipient countries become considerable.
19. We use specification 4 as the referential specification used in our tests. The results of the robustness analysis are reported in . For brevity, we only exhibit estimates of the effect of openness to experience on foreign bias.
20. The results are qualitatively similar when we use other variables that produce an interaction effect with openness to experience.
21. We control for corruption, expropriation risk, rule of law, and judicial efficiency. The data are sourced from La Porta et al. (Citation1998).
22. The specifications exclude the ‘GDP per capita in investor country’ and ‘patriotism’ variables due to multicollinearity issues.