An influential set of laboratory studies (e.g. Kahneman et al. 1991) documents that ownership of an asset causes significant changes in subjects’ valuations of the asset. This ‘endowment effect’ runs counter to standard microeconomic theory, which assumes that preferences and beliefs are unaffected by ownership. These results are important to consider carefully, as they suggest there may be a wedge between buyers’ and sellers’ valuations of objects, with potential consequences for the design of market mechanisms.
These results have been called into question in markets outside of laboratory settings (List 2003, 2011), where evidence suggests that endowment effects attenuate sharply for more experienced traders. This is important, as it suggests there may be less need for concern about the prevalence of endowment effects in field settings. However, these studies, while cleanly identified and internally valid, have been limited in the types of markets they study, and by the number of market participants they track over time. As a result, it continues to be difficult to judge whether the endowment effect is materially important in a number of large-scale market settings.
In a new paper, we exploit a natural experiment in the Indian equity market to study the endowment effect (Anagol et al. 2016). During the period of study, India’s stock market regulator required the randomised allocation of shares to retail investors in situations in which initial public offerings (IPOs) were highly oversubscribed. In such situations, the lottery ensured that winners received a fixed number of shares (the minimum lot size) of the IPO stock, while losers in the lottery received zero shares. Using precise data on portfolio holdings, we track the behaviour of a total of 1.5 million winners and losers following the random endowment of the stock in lotteries in 54 IPOs occurring between 2007 and 2012.
We measure the endowment effect as the difference in the propensity of winners and losers to hold the IPO stock following the random allocation. The randomisation ensures that winners and losers are ex-ante identical in terms of their information sets, beliefs, and preferences. Moreover, they have equal opportunities to trade once the stock is listed in the market. If endowment effects do not exist in this setting, holdings of the randomly allocated stock amongst winners and losers should converge rapidly over time.
What we find
At the end of the first trading day following the random allocation of stock, we find that 70% of lottery winners hold the IPO stock, while only 0.7% of the losers hold the IPO stock. At the end of the listing month, lottery winners are 62% more likely to hold on to the IPO stock, and this effect size declines to 46% at the end of six months. Even 24 months following the IPO listing, winners are 36% more likely to hold the IPO stock than lottery losers. Throughout this period, lottery losers’ propensity to hold the stock stays relatively constant, between 1.5 and 1.7%.
We also find that the estimated endowment effect has little relationship with listing gains on the first day (Figure 1), suggesting among other things that these effects are not driven by a wealth effect accruing to lottery winners. However, they fall sharply with the experienced return on the stock in the month following listing (Figure 2), reminiscent of the well-known disposition effect (Shefrin and Statman 1985).
Figure 1 Listing returns and the endowment effect
Figure 2 First-month holding returns and the endowment effect
Similar to List (2003, 2011), we find that the estimated endowment effect is highly correlated with measures of market experience. Investors who were allotted shares in over eight past IPOs have endowment effects that are 17% smaller than investors with no past IPO experience. That said, such experienced investors are still 60% more likely than losers to hold the IPO stock at the end of the first day. Similarly, investors who trade frequently in stock markets (with more than 6 trades in the month prior to the random allotment of shares) are 14.3% less likely to hold the IPO stock compared to those with no past trades, but nonetheless exhibit a strong endowment effect. Overall, we find that while experience does attenuate the endowment effect, it doesn’t appear to eliminate it.
We have one interesting finding about the role of experience measured differently, i.e. personally experienced past returns. We find that if the IPO returns are substantially greater than a winner’s previously experienced returns in their portfolio, they are more likely to sell the IPO stock, and for losers, they become more likely to buy the stock. That is to say, the endowment effect reduces considerably when past personally experienced returns are taken into account, suggesting that experience-based learning (Malmendier and Nagel 2011, 2016) plays an important role in individual decision-making.
Explaining the endowment effect
We consider the extent to which standard explanations, such as wealth effects, monetary transaction costs, taxes, and disincentives for flipping (i.e. investors might believe that they will be penalised in allocations to future IPOs if they sell the stock quickly), explain the estimated endowment effect – and we find little evidence for such motivations.
Given that our results stem from inaction on the part of lottery winners (they don’t sell) and losers (they don’t buy), a natural characterisation of this behaviour is that investors are inertial. However, we observe that lottery winners trade the rest of the portfolio more than lottery losers following the random endowment (Anagol et al. 2015). Moreover, we find that the endowment effect is still evident for investors who made more than 20 trades in non-IPO stocks in the same month. These facts, as well as others that we document, suggest that while the endowment effect may be driven by inertial behaviour that is specific to the IPO stock, it is very unlikely that our results are driven by investors’ portfolio-wide inertial behaviour.
We go on to consider a number of theoretical models which have been used to explain laboratory findings of the endowment effect or empirical evidence of investor behaviour such as the disposition effect. We find that the leading theoretical model used to explain the endowment effect, namely the expectations-based reference dependent loss-averse preferences model of Kőszegi and Rabin (2006) appears unable to explain our results in a setup that realistically captures features of the natural experiment that we study. We also find that a number of other models, such as the Weaver and Frederick (2012) model of ‘bad deals’, the realisation utility model with loss-averse agents of Barberis and Xiong (2012), and the salience-based decision-maker model of Bordalo et al. (2012), are able to account for parts of our findings, but none on its own can explain the sum total of our results.
In particular, a major stumbling block for these models is their inability to explain why lottery losers rarely ever hold the stock. To explain this finding, we posit that a model in which losers are inattentive to the stock after they lose the lottery is necessary in addition to one of the other mechanisms. In sum, a combination of inattentive behaviour on the part of losers combined with non-standard preferences or non-standard beliefs of agents would appear to be consistent with our findings taken as a whole.
Anagol, S, V Balasubramaniam, and T Ramadorai (2016), "Endowment Effects in the Field: Evidence from India's IPO Lotteries", available at SSRN
Anagol, S, V Balasubramaniam, and T Ramadorai (2015), "The Effects of Experience on Investor Behaviour: Evidence from India's IPO Lotteries", available at SSRN 2568748
Barberis, N, and W Xiong (2012), "Realization utility", Journal of Financial Economics 104 (2), 251-271
Bordalo, P, N Gennaioli, and A Shleifer (2012), "Salience in experimental tests of the endowment effect", The American Economic Review 102 (3), 47-52
Kahneman, D, J L Knetsch, and R H Thaler (1991), "Anomalies: The endowment effect, loss aversion, and status quo bias", The Journal of Economic Perspectives 5 (1), 193-206
Kőszegi, B, and M Rabin (2006), "A model of reference-dependent preferences", The Quarterly Journal of Economics, 1133-1165
List, J A (2003), "Does market experience eliminate market anomalies?" Quarterly Journal of Economics 118 (1), 41-72
List, J A (2011), "Does market experience eliminate market anomalies? The case of exogenous market experience", The American Economic Review, 313-317.
Malmendier, U, and S Nagel (2011), "Depression Babies: Do Macroeconomic Experiences Affect Risk Taking?", The Quarterly Journal of Economics 126 (1), 373-416.
Malmendier, U, and S Nagel (2016), "Learning from inflation experiences", The Quarterly Journal of Economics 131 (1), 53-87
Marzilli Ericson, K M, and A Fuster (2014), "The Endowment Effect", Annual Revenue of Economics, 6 (1), 555-579
Shefrin, H, and M Statman (1985), "The disposition to sell winners too early and ride losers too long: Theory and evidence", The Journal of Finance 40 (3), 777-790