سمت تاریک مشوق های تیم : شواهد تجربی بر کیفیت مشاوره از متخصصان خدمات مالی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|4627||2013||7 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of Economic Behavior & Organization, Available online 27 March 2013
In an experiment with professionals from the financial services sector, we investigate the impact of a team incentive scheme on the recommendation quality of investment products when advisors benefit from advising lower quality products. Experimental results reveal that, when group affiliation is strong, inferior products are recommended significantly more often under team incentives than under individual incentives.
Consumers often have to rely on experts’ advice when making investment decisions, especially in the presence of strong informational asymmetries, a lack of expertise, and uncertainty about future profits. However, in the wake of the recent financial crisis, claims have been made that financial advisors mislead private investors about the riskiness of products, inducing investments in inferior financial assets. Indeed, in their professional routine, financial service experts constantly face the dilemma of how to balance their own interests with those of their customers. Several experimental studies have shown that broadly suggested disclosure of conflict of interest does not necessarily help to reduce the misadvising problem (Cain et al., 2005, Cain et al., 2011 and Ismayilov and Potters, 2013). Therefore, further search for effective mechanisms for solving this problem is still required. The impact of monetary profit on misadvising, as well as misreporting, has recently gained increasing interest and is often seen as one of the important drivers of unethical behavior (Gneezy, 2005, Inderst and Ottaviani, 2009, Sutter, 2009, Popova, 2010, Gibson et al., 2013, Angelova and Regner, 2013 and Gneezy et al., 2013). It has been shown that not only the size of a monetary incentive itself is relevant for unethical behavior but also the type of the incentive scheme (Schweitzer et al., 2004, Denis et al., 2006, Ordonez et al., 2009, Cadsby et al., 2010 and Conrads et al., 2013). For example, Cadsby et al. (2010) find experimental evidence that performance is over reported more often under target-based incentive schemes than under piece rate or tournament schemes. Denis et al. (2006) find a positive correlation between the likelihood of stock manipulation fraud and intensity of option-based compensation of CEOs. In a recent experiment, where Conrads et al. (2013) employed the dice-rolling game of Fischbacher and Heusi (2013), students lied significantly more often when team incentives were offered. However, in this experiment, lies had no negative impact on any other subject but raised the costs for the experimenter. Even if the advisors may not be focused purely on their narrow personal interest, they may still be tempted to adjust their recommendations to benefit their team or the financial institution they work for. They may do this, for instance, because of strong team identification or loyalty to the employer at the expense of their customers’ interests. Thus, advice from potentially biased experts might lead to sub-optimal investment decisions, particularly when their monetary incentives are tied to the short-term goals of the financial institution rather than their customers’ interests. In general, the effect of team incentives on potential misadvising may be twofold: On the one hand, a simple economic model would predict that misadvising is less prominent under a team bonus, as the individual marginal monetary benefit from misadvising is smaller, and a free-rider problem occurs as demonstrated by Holmstrom (1982). However, recent research indicates that the underlying mechanism of cheating under team incentives cannot be explained by purely economic considerations. For example, several experimental studies have pointed out that splitting the benefits with another person increases the likelihood of cheating (Gino and Pierce, 2010, Wiltermuth, 2011, Conrads et al., 2013 and Gino et al., 2013). In this way, Conrads et al. (2013) observed that lying in teams is partially driven by the opportunity to hide individual misdeeds. Gino et al. (2009) showed that corrupt social norms may serve as a trigger of unethical behavior in teams. The key research question of this paper is how a team or an individual incentive scheme affects advice quality, and whether the effects are moderated by the strength of the relationship between team associates. Mazar et al. (2008) demonstrated that a desire to keep a positive self-image facilitates deception. Considering our research question, team incentives may provide a justification for dishonest acts by means of, e.g., “helping the team members” which reduces the perceived immorality of the self. In this case, team incentives may help to reduce the perceived moral costs of dishonesty toward customers and increase the delivery of bad advice. The justification of dishonesty for team benefit may be easier especially when the relationship between the team members is strong. Therefore, we hypothesize that the detrimental effect of team incentives is moderated by the strength of the relationship among team members. To the best of our knowledge, very little research has been conducted on individual closeness and misbehavior. Gino and Pierce (2010) observed a positive effect of feeling empathy with others in helping them out by lying, whereas Wiltermuth (2011) found no significant effect of experimental matching with either a friend or a stranger on misreporting. However, investigating how closeness among individuals alters the likelihood of engaging in dishonest behavior remains of great importance, as human decisions are often made in social environments where actors have strong social ties. Therefore, in our experiment we focus on the disadvantages of team incentives among individuals with different degrees of group affiliation. We address this question by implementing a simple sender-receiver game, in which advisors, represented by financial sector professionals who took part in training, recommended an investment product to customers. Each customer, represented by a student participant, decided whether or not to buy the recommended product. Customers were not informed about the product's features and as participants interacted only once, the products were pure credence goods (Darby and Karni, 1973 and Dulleck and Kerschbamer, 2006). The advisors, however, were informed about the revenue distribution and the size of the commission rate attributed to each product. The product quality in our setting was operationalized by high expected return and low risk, and was inversely related to the size of the commission rate paid to the advisor. In particular, we compared the quality of the recommended products under an individual commission rate and a team bonus payment. For the latter, the commissions of three advisors were paid into a team account, which then was evenly distributed among its members. Additionally, we used the difference in the amount of time spent in the joint training seminars as a natural variation of subjects’ closeness. Both product design and incentive schemes are derived from common situations in the financial advisory business. Being financial professionals, advisors are confronted with very similar situations in their day-to-day business where they can typically choose from a set of products with different commission rates, returns on investment, and risk. Furthermore, retail bankers’ compensation schemes often combine components of individual and group commissions. We thus believe that conducting this experiment with financial service professionals employs a useful social framing that adds to the external validity of the experiment (e.g., Carpenter et al., 2005). Our main finding is that, compared to individual incentive schemes, advisors who were strongly affiliated with fellow team members recommended lower quality products when facing team incentives. However, we did not observe any difference in recommendation quality between treatments when relationship strength was weak. The remainder of this paper is organized as follows. In Section 2 we describe the experimental design and procedure. Section 3 presents the results and the last section concludes.
نتیجه گیری انگلیسی
A potential negative impact of individual incentive schemes, such as relative performance pay or target-based incentives, on ethical behavior has been broadly discussed in economics and management research (i.e., Schweitzer et al., 2004, Denis et al., 2006, Ordonez et al., 2009 and Cadsby et al., 2010). Moreover, some recent studies (e.g., Conrads et al., 2013) have indicated that team incentives may induce lying. Our results show that team incentives can induce a lower recommendation quality for credence goods by professional financial service advisors when they are from a sample with closer social ties. In the experiment, retail banking professionals were asked to recommend one out of six investment products to customers. Here, advisors had the opportunity to reap material benefits by recommending financially unattractive products to customers. We find that group incentives lead to lower recommendation quality when the group affiliation is strong. As close affiliation between team members is what we encounter in real-world organizations, where group bonuses are paid to team members who typically work together over long time horizons, this problem is particularly relevant. Our results indicate that under close group affiliation, individual incentives may be beneficial to induce more honest behavior, potentially because the selfish nature of distorted advice is more obvious and may raise the individual costs of lying.