In addition, we investigate whether there is a difference in behavior between professionals who are in direct contact with customers and those that are not. In particular, we run the regression of Table 1 (column (13) ALL) only with those professionals with direct customer contact (i.e., compliance, client advisor, customer support, fund placement, investment advisor, investment banking, private equity/banking, relationship manager, sales) in one specification and with those professionals without direct customer contact in another specification. As outlined in columns 3 and 4 of Table A.3 in the online appendix, we find that coefficients are very similar across both sub-pools of professionals with slightly lowered significance levels due to lowered sample sizes. Thus, we infer that the effects observed in our study are a general finding, applying to professionals with and without direct customer contact in their daily job. In addition, concisefinance we added fixed effects for each job function as outlined in column 2 of Table A.3 in the online appendix. We show that results do not change when adding job function fixed effects, indicating that job functions do not necessarily drive our main result.
As a robustness check for Experiment we recruited another 160 financial professionals and administered a second experiment, with a mobile laboratory in the field (see online appendix A.2 for details on the design and results). This experiment is a modified setting of the lab-in-the-field investment experiment in Kirchler et al. (2018) with the focus on professionals’ decision making for real customers. The results from the lab-in-the-field experiment corroborate our findings from the online experiment by showing that rank-driven behavior is activated once professionals’ incentives are no longer flat. Here, underperforming professionals increase their risk taking when investing customers’ funds compared to their outperforming peers. Again, we also find a null result in the treatment with professionals’ flat incentives, providing tentative evidence that rank-driven behavior could be moderated when incentives are flat.
Online survey evidence for preferences regarding relative performance, competitiveness, and risk
Our results in this paper so far and the findings in Kirchler et al. (2018) reveal that the rank-driven behavior of financial professionals is robust across various settings, including investment decisions for others. This raises the question to what extent financial professionals differ in their rank-driven behavior compared to other groups, such as their customers. If customers are equally rank-driven as professionals, it is possible that customers not only enjoy the monetary, but also the non-monetary benefits from their investment manager’s (or private banker’s), outperforming their peers. Hence, if customers benefit from the higher ranking, the rank-driven behavior of professionals could be in the interests of customers, as it directly translates into non-monetary customer benefits. If, however, preferences for relative performance are stronger for professionals than the average customer, then any increased risk taking due to social competition among peers is at least partly violating customer interests. Given the performance-oriented business culture in the financial industry and the possibility that very competitive individuals self-select into this sector and are shaped by it, financial professionals might differ from other groups in their preferences for relative performance, competitiveness, and risk.
To shed more light on the role that professionals’ individual characteristics might play, we administered an online survey to financial professionals, a representative general population sample, and individuals from other competitive professions like professional sports and academia (the survey questions are outlined in the online appendix A.5). In particular, we asked survey questions measuring risk attitudes according to the German SOEP (Dohmen et al., 2011) and attitudes toward social status, financial success, and relative performance like in Cohn, Fehr, Maréchal, 2014, Cohn, Fehr, Maréchal, 2017.22 In addition, we asked for preferences regarding social status and relative performance in specific domains (job, hobbies, family, friends) and how self-perceived attitudes towards social status and relative performance developed during childhood and adolescence.23