Does time inconsistency differ between gain and loss? An intra-personal comparison using a non-parametric elicitation method
Does time inconsistency differ between gain and loss? An intra-personal comparison using a non-parametric elicitation method
复制标题
收益和损失之间的时间不一致是否存在差异？
DOI：
10.1007/s11238-019-09728-1
复制
发表时间：
2020
影响因子：
0.8
通讯作者：
K
中科院分区：
文献类型：
--
作者：
Shiba;S.;Shimizu;K
一些关于时间偏好的研究发现，收益偏好和损失偏好的时间不一致。然而，同一个人内两者之间的关系仍不清楚；也就是说，一个人在获得收益时表现出时间不一致，在损失时也会表现出时间不一致吗？本文报告了实验室环境中个体收益和损失的时间不一致情况。为了精确比较个人的得失时间不一致程度，我们使用罗德的“DI（减少急躁）指数”（Manag Sci 65(4):1700–1716, 2018）来测量时间不一致的程度，而不仅仅是识别 TI 是否存在。该指数表示一个人表现出当前偏见的程度，并且很容易扩展到同一个人内的收益和损失偏好之间的比较。此外，它允许实验测试所谓的未来偏差，这一直是最近时间不一致文献的焦点领域。它是通过非参数方法得出的，避免了分析中的任何规范错误。我们的研究结果如下：首先，我们发现未来的偏好不仅对收益而且对损失都有偏好，并且我们证实这种趋势与之前对收益的偏好的发现是一致的。其次，在个人层面上发现收益和损失的时间不一致之间存在正相关关系。事实上，在大多数情况下我们无法发现两者之间存在显着差异。
Several studies on time preference have found time inconsistency in both gain and loss preferences. However, the relationship between the two within the same person remains unclear; that is, does an individual who demonstrates time inconsistency for gain outcomes do so for losses as well? This paper reports on individuals’ time inconsistency for gains and losses in a laboratory setting. To obtain a precise comparison of individuals’ time inconsistency for gains and losses, we used Rohde’s “DI (decreasing impatience)-index” (Manag Sci 65(4):1700–1716, 2018) and measured the level of time inconsistency, rather than merely identifying whether TI was present. This index represents how strongly a person exhibits present bias, and easily extends to the comparison between gain and loss preferences within the same person. Further, it allows the experiment to test for so-calledfuture bias, which has been a focus area in recent time inconsistency literature. It is elicited through a non-parametric method, which avoids any specification errors in the analysis. Our findings are as follows: first, we found future bias in preferences for not only gains but also losses, and we confirmed that this tendency is consistent with previous findings on preferences for gains. Second, a positive correlation between time inconsistency for gains and losses was found at the individual level. Indeed, we could not find a significant difference between the two in most cases.