Who’s biased? A meta-analysis of buyer–seller differences in the pricing of lotteries.

Abstract

A large body of empirical research has examined the impact of trading perspective on pricing of consumer products, with the typical finding being that selling prices exceed buying prices (i.e., the endowment effect). Using a meta-analytic approach, we examine to what extent the endowment effect also emerges in the pricing of monetary lotteries. As monetary lotteries have a clearly defined normative value, we also assess whether one trading perspective is more biased than the other. We consider several indicators of bias: absolute deviation from expected values, rank correlation with expected values, overall variance, and per-unit variance. The meta-analysis, which includes 35 articles, indicates that selling prices considerably exceed buying prices (Cohen's d = 0.58). Importantly, we also find that selling prices deviate less from the lotteries' expected values than buying prices, both in absolute and in relative terms. Selling prices also exhibit lower variance per unit. Hierarchical Bayesian modeling with cumulative prospect theory indicates that buyers have lower probability sensitivity and a more pronounced response bias. The finding that selling prices are more in line with normative standards than buying prices challenges the prominent account whereby sellers' valuations are upward biased due to loss aversion, and supports alternative theoretical accounts. (PsycINFO Database Record

The Loss of Loss Aversion: Will It Loom Larger Than Its Gain?

“…Another explanation of the endowment effect, which similarly does not require loss aversion, comes from Weaver and Frederick () and Isoni () (see also Simonson & Drolet, ; Yechiam, Ashby, & Pachur, ). These authors provide a differential reference price account.…”

“…In addition, the status quo bias occurs even when the status quo and change options are otherwise equivalent and thus not coded in terms of losses and gains (Gal, 2006; replicated in this article) Endowment effect Kahneman et al (1990) Participants indicate a higher WTA than WTP for the same good Loss and gain are confounded with inaction and action as well as other factors. When losses and gains are decoupled from inaction and action in the retention paradigm, no evidence for loss aversion is present (Gal & Rucker, 2017a; see also Isoni, 2011;Yechiam et al, 2017;Weaver & Frederick, 2012) Risky bet premium Tversky & Kahneman (1992) Participants demand a substantial premium over an expected value of zero to accept a bet with even odds of gain and loss. It is thought that the possible loss of money looms larger than the possible gain of money Inaction and action are confounded with loss and gain; when losses and gains are decoupled from inaction and action, no evidence for loss aversion is present (Ert & Erev, 2013;Gal, 2006;Yechiam & Hochman, 2013) Hedonic impact ratings…”

Too good to be true? Psychological responses to uncommon options in risk–reward environments

“…The task was presented in the form of a game show called “Keep or sell?” (“Behalten oder Verkaufen?”). We used selling rather than buying prices because the former have been shown to deviate less from gambles' EVs than the latter (however, both paradigms elicit responses strongly linked to EVs; see Yechiam et al, ); in general, both formats are suitable for exposing participants to different risk–reward structures. To motivate participants to indicate their true valuations of the gambles, we implemented a Becker‐DeGroot‐Marschak auction (Becker et al, ) as follows: Participants entered a price at which they would be willing to sell each gamble by moving the mouse along a rating scale (0 E $ − 2,500 E $) and confirming the value with a click.…”

“…In pseudocode, We excluded trials in which participants indicated prices that exceeded the payoff offered in the gamble by more than 100E$, as this suggests lack of attention to the task (we did the same in Experiment 2). We used this rather liberal exclusion criterion for two reasons: First, as a recent meta-analysis shows, selling prices frequently exceed gambles' EVs ( Yechiam et al, 2017, Table 2). Second, the EV of a gamble becomes harder to compute given four-digit payoffs (0 − 2, 500E$), compared with smaller payoff ranges.…”

Too Good to Be True? Psychological Responses to Uncommon Options in Risk–Reward Environments

“…("Behalten oder Verkaufen?"). We used selling rather than buying prices because the former have been shown to deviate less from gambles' EVs than the latter (however, both paradigms elicit responses strongly linked to EVs; see Yechiam et al, 2017); in general, both formats are suitable for exposing participants to different risk-reward structures. Experiments were coded in PsychoPy (Peirce, 2007).…”