The Psychology of Taxing Capital Income: Evidence from a Survey Experiment on the Realization Rule
95 Pages Posted: 18 May 2021 Last revised: 20 Jan 2022
Date Written: May 17, 2021
How to tax capital income is a critical issue today. The realization rule—requiring that property usually must be sold before gains are taxed—is central to taxing capital income, but often decreases the efficiency, equity, and simplicity of the tax system. Estimates suggest that the realization rule costs the government over $2 trillion over 10 years. Given these problems, it is unclear why the rule exists for assets that are easy to value and sell. Scholars have long speculated about the role of the public’s views here, but little is known empirically about them. We conduct the first survey experiment to understand the psychology of the realization rule, which has broad implications for the taxation of capital income.
We have three main findings. First, respondents strongly prefer to wait to tax gains on stocks until sale: 75% to 25%. This pattern persists across a variety of other assets and policy framings: indeed, nearly half of those without stock prefer raising everyone’s taxes (including their own) to taxing unsold stock gains. But the flip side is that there is surprisingly strong support for taxing gains on assets at sale or transfer, including at death, in areas where current law never taxes those gains. Second, these views change only modestly after randomized participants observe a policy debate composed of videos explaining both the pros and cons of taxing before sale, though the pro and con treatments have large effects individually. And, third, among many possible explanations of these attitudes, we find particular evidence for four: using a different mental account for unsold gains than other ways of getting richer; a tendency to support the status quo; concerns about complexity; and a desire to tax consumption, not income, in the context of capital gains.
Keywords: capital income taxation, realization rule, survey experiment, social economics
JEL Classification: D72, D91, G41, H20, H31, K34
Suggested Citation: Suggested Citation