Another issue on which TS [Thaler and Sunstein] are silent--and should not be--is the issue of income tax withholding. The U.S. federal government moved to withholding during World War II because that was when it changed the income tax from a "class tax"--that is, a tax on high-income people--to the "mass tax"--that is, a tax on all but very low-income people. Afraid that people would be faced with a huge tax bill if they were required to pay their taxes annually or quarterly, the federal government implemented an "installment plan," requiring employers to withhold taxes from their employees' paychecks. I remember having a discussion about this in the late 1970s with Thaler, when he and I were both young assistant professors at the University of Rochester. We both noted that, because of withholding, many people do not seem aware of how much they really pay in income taxes and that, if they get a big refund, they think they are not paying much. This is the kind of bias that, in other contexts, TS decry. They seem to want people to be aware of the real cost of various things and not to under- or over-estimate. So a straightforward way to do that with taxes is to end withholding. If people were required to pay taxes on an annual or quarterly basis, they, especially high-income people, would become more aware of the cost of government and would likely estimate it to be higher than they now think it is. The result would likely be more resistance to government expansion. It's pretty clear that libertarians would like this outcome. And, given TS's basis for libertarian paternalism--people's systematic tendency to misestimate based on readily available data--and their alleged libertarianism, they should like both the process that leads to the result--namely, people having better information--and the result itself. Yet they say not a word about tax withholding.
This is from David R. Henderson, "Libertarian Paternalism: Leviathan in Sheep's Clothing?", Society, June 2014. Although I recommend my own article (duh), I don't recommend that you pay $39.95 to read it. If you are affiliated with a university, you can probably get it free on line. Notice, by the way, that they made a mistake in the abstract. The last word should be "more," not "true."
Another excerpt, under a subsection titled "The Invisible Gorilla:"
Libertarians are also particularly well situated to take the TS thinking in a new direction where TS seem to fear to tread: applying the insights of behavioral economics to a particularly powerful group of people, namely, government officials.
Let me explain. In a fascinating chapter, "Invisible Gorillas and Human Herds," Sunstein tells about an experiment in which people who were asked to watch a video and count basketball passes totally missed seeing a gorilla in the midst of the players. The lesson for businesses, individuals, and governments is, he writes, "that we are all at risk of missing a lot that is happening in the background (and possibly even the foreground) of our lives." Indeed.
That brings me to the 800-pound gorilla in the room--government--and a large irony in Simpler that Sunstein seems unaware of. In one passage, he notes that he delayed getting vaccinated for H1N1. Why do I highlight that fact? Because it illustrates a fundamental contradiction. Sunstein's delay shows that even he is subject to the thinking that he wants the government to "nudge," or outright coerce, us out of. He even admits, just four pages earlier, that for many people, "including those who work in government, what may matter most is today, tomorrow, and next week." Yet, he wants us to trust these selfsame government officials to make major decisions--about drugs, medical care, cars, and cigarettes, to name only a few--for us. If those government officials can't be trusted to take the long view when their own wellbeing is at stake, why would Sunstein think that we can trust them to do so for a nation of strangers?
The very insights that TS have about human behavior apply even more strongly to the humans in government because, besides being human and having the limits that regular humans have, they have even less incentive than the average human has to make good decisions about our lives. If something in our lives goes wrong, even badly wrong, due to a government official's decision, the government official typically bears very little cost for that bad decision. That explains why government so often makes bad decisions.