Survey_HKLRW_2020

Abstract

We randomly assign subjects to one of two main incentive conditions. In the flat condition, subjects receive a fixed fee as a reward for participating in the experiment. The fixed participation fee is 12 euro for subjects in the private and professional investor samples, and is 3 euro in the student sample. Payment of the fixed participation fee is independent of the choices made in the experiment. In addition to the participation fee, subjects in the incentives condition are paid the earnings resulting from their choice in one experimental task, which is randomly selected at the end of the experiment. In case the selected task involves a series of decisions, a second random draw determines the decision to be paid out. Subjects in both conditions are presented with the same experimental tasks and experimental instructions (except for minor differences necessary to explain the payment protocols). At the beginning of the experiment, we explicitly inform subjects in the flat group that the payoffs of their decisions are hypothetical and do not affect their final payoff. The experiment was administered online using Limesurvey. We recruited subjects for participation via e-mail. Through e-mail, we recruited student participants from the University of Innsbruck using Hroot (Bock et al., 2014). In addition, we invited private investors from a panel of 2,000 clients of a large German brick-and-mortar bank who regularly participate in short online surveys/experiments administered by Goethe University Frankfurt. Third, we recruited professional investors via two channels. Two-thirds of the professional investor sample were recruited from the proprietary subject pool of professional investors (www.before.world) at the University of Innsbruck, some of whom had participated in previous unrelated studies. The remaining professionals are fund managers from di?erent European countries whom we identified via their fund aliation using data from Morningstar. We collected data for all subsamples in April and May 2020. Overall, 1,727 subjects completed the experiment. Once the experiment started, we o?ered unlimited time to finish to avoid exerting time pressure on subjects who were potentially engaged with risk elicitation experiments for the first time. However, to screen out participants who plausibly did not take the experiment seriously and to avoid potential noise due to outliers, we drop subjects in the top (99%) and bottom percentiles (1%) of the response time (34 respondents). The final sample consists of 1,693 subjects, comprising 821 subjects from the private investor sample, 244 from the financial professional investor sample, and 628 from the student sample. The median response time in the final sample is 13.22 minutes with a standard deviation of 9.88 minutes.