NYU Tax policy colloquium, week 5: Vanessa Williamson's "Read My Lips: Why Americans Are Proud to Pay Taxes"

Yesterday at the colloquium, Vanessa Willamson of the Brookings Institution presented two chapters from her book Read My Lips: Why Americans Are Proud to Pay Taxes. Specifically, we read chapters 2 and 4, which could be collectively titled "How the Taxpaying Experience Obscures Low-Income Taxpayers and Shapes Attitudes Towards Progressivity."

The book is the product of in-depth survey research (via one-hour phone interviews with a broad range of respondents) regarding Americans' views of the tax system. There are quantitative aspects to the research, but the main contribution is qualitative - exploring people's beliefs and sentiments in a setting that encourages them to be reflective rather than, say, emphasizing partisan talking points. The interviews were conducted in 2013-2014, hence after the Obama-Romney campaign but before Clinton-Trump.

The title is a bit of a misnomer, picked (I would presume) by the publisher with an eye to sales appeal rather than descriptive precision. The book doesn't so much document that the respondents are made to feel proud by the act of paying taxes as such, as that they view "taxpayers" as morally superior to "non-taxpayers" - even when they describe themselves as non-taxpayers. The pride attaches to the status, not the act, although it's true that without the act one presumably can't get the status..

The research was performed before Trump in 2016 proclaimed that not paying taxes is "smart," rather than discreditable. So it would be interesting to see if Republican voters' attitudes have been changed by this. I would like to hope not. Trump presents himself to his adorers as unique and special, a kind of fantasy projection for them. So perhaps they are like nerds who can cheer the violence wreaked by the protagonists in superhero films without either thereby becoming violent themselves or starting to admire violence by peers or even other political leaders. Presumably Trump supporters would still mind if a Democrat (or a "moderate" Republican?) paid no taxes on a large income. But then again,  it's way too early to judge the long-term damage to cooperative social norms generally that the current era is wreaking.

Anyway, back to the underlying common public attitude about taxpayers versus non-taxpayers. I view it as a culturally particular instantiation of a fundamental human psychological frame.

Starting with the more universal part: human beings, like lots of other social species, regularly engage in cooperative undertakings for general benefit, in which it is hard to observe others' degree of cooperation, or else in which cooperation and thus reliance is sequential. This creates prisoner's dilemmas that are eased by natural or cultural selection for having some degree of emotional inclination to cooperate (which may help to overcome both short-sightedness and the difficulty of faking a good attitude), while also raising the probability that defectors will be punished when they are found out, and that people will play tit-for-tat when deciding whether to cooperate.

As it happens, vampire bats show such attributes as well. When some have succeeded and others have failed in their nightly blood quests, the former will sometimes regurgitate and share a part of their booty with the latter. They keep track of who shares with them and who doesn't, which apparently requires having relatively large brains despite the caloric cost. So vampire bats not only sometimes front-end the generosity (someone has to go first), but at the back end distinguish between cooperators and defectors, and play at least a degree of tit-for-tat.

Blood-sharer = taxpayer? Obviously, here we are in a culturally contingent application of the underlying concept. It's one thing to say that human beings are often emotionally inclined to adopt a tit-for-tat framework that valorizes contributors while disparaging shirkers, and another to generalize about their predilection to base this on one's status as a taxpayer or non-taxpayer. And note that the main distinction Williamson's respondents had in mind was NOT that between users and non-users of aggressive tax shelters, so much as that between higher and lower earners (where the latter can be assumed to pay little or no federal income tax).  Pending parallel qualitative surveys conducted in other countries (which would be a great and most-welcome contribution), this might be a distinctively 21st century American way of coding contributor vs. shirker.

I'm certainly as subject as anyone to feeling the emotional tug of the contributor vs. shirker / tit-for-tat psychological frame. But I regard it as preposterous to apply that frame to people who earn enough vs. don't earn enough to pay significant (income) taxes. For example, (1) it's a current-year rather than a long-term assessment, (2) it looks at the gross not the net (taxes alone, rather than taxes vs. benefits), and (3) it treats low-earners as shirkers rather than as (a) unfortunate and (b) mostly internalizing the downside of their bad outcomes (it's not as if we had a generous social safety net).

The viewpoint brings to mind Romney's infamous "47%" comment in 2012. So why did he get flogged so badly for it in the public sphere? I think his main problem was overdoing it, by smearing 47% of the population as wholly unwilling to take responsibility for their own lives. Just as, back in the day, blatant racists risked offending casual and unconscious racists when they too-clearly stated beliefs that the latter shared, so Romney made problems for himself here. Indeed, one could almost call his statement a Kinsley gaffe - defined as occuring "when a politician tells the truth - some obvious truth he isn't supposed to say" - except that here it wasn't an actual truth, but rather something that many people believed to be true.

Viewing "taxpayers" as better than "non-taxpayers," where the difference lies in having enough earnings rather than in the degree of one's tax avoidance  behavior, is a normative choice, albeit one I find unappealing. So it's not factually "untrue." But the Williamson book suggests that people do indeed commit factual error by applying the frame solely with regard to income taxes - as distinct from, say, sales and payroll taxes.

The book further argues, I think persuasively, that this narrow-sightedness reflects differences in the degrees of "effort" associated with paying different taxes. Payroll taxes are collected so seamlessly (other than from independent contractors), and indeed so invisibly so far as the employer half is concerned, that people scarcely notice they are paying them. By contrast, income taxes, despite withholding's role in easing or even reversing the April 15 out-of-pocket bite, are associated with such stressful acts as getting one's financial info together, getting the return done, feeling anxiety about whether (amid all the bewildering complications) one has found all the right tricks for reducing one's liability, etc.

Sales taxes are a funny intermediate case. Williamson finds that poor people, whose budgets even a modest sales tax can stretch, are keenly aware of them. But higher-earners treat the sales tax as invisible on small purchases, and worth noticing only on big purchases where the bite is large enough to notice. Hence, astonishingly, they seem to view flat-rate sales taxes as progressive! In effect, the heuristic of ignoring trivially small amounts leads them to view the sales tax almost as if it were a luxury tax - without, it seems, any reliance on the fact that, say, groceries but not restaurant meals are typically exempted.

Decades of Republican rhetoric has surely done a lot to spread the view that "taxes" just means income taxes, which leads to distorted views of who is a taxpayer. But the book suggests that it is also about the income tax's being (despite withholding) a relatively high-effort and high-visibility, rather than low-effort, tax instrument.

Here's a further subtlety from the book regarding what the notion of "effort," for this purpose, seems to mean. Gasoline taxes are pretty much hidden - you just pay the after-tax price at the pump, and the tax isn't even separately stated. Yet apparently people are keenly aware that, if you do a whole lot of driving, you will pay a whole lot of gasoline tax. So they notice and complain about it, even though it's low-visibility apart from the general understanding that it's there (and rises with mileage).

One obvious payoff to these findings is that, if you are a policymaker who likes Tax A better than Tax B, you should try to structure things so that Tax A is low-effort or relatively invisible, whole Tax B is more high-effort and visible. So perhaps the Democrats ought to go all-in on seeking to enact automatic filing a la ReadyReturn in California. The aim would be, not to win votes ex ante by promising it, or gratitude ex post by delivering it, but rather to reshape the cognitive fiscal landscape in their favor once automatic presumptive filing was in place. But this would require the Democrats to be strategic and forward-looking, which is asking quite a lot of them.

One further set of findings from the book is as follows: While people tend to have mixed feelings about progressive redistribution of market income - seeming to combine an implicit "equal sacrifice" theory, plus aversion to high-end inequality even though they immensely under-estimate its scope, with concern about work effort and high-earners' distributive desert - they do on balance seem to favor rising marginal and average tax rates. (Which is not to say that they have a firm grasp of the distinction between these two measures.)

Yet  people erroneously think that a "flat tax" would be more progressive than the then-existing (pre-2017) graduated income tax, because they exaggerate the empirical significance of "loopholes." What they think the term "flat tax" means is unclear. No "deductions"? But presumably they don't think it would be a tax on the gross, rather than the net, from a given business. And of course, the question of what constitutes a "loophole," or for that matter a deduction, may not be well-understood.

Not clear how much any of this matters to actual political outcomes, given the falsity (as per Achen and Bartels) of the "folk theory of democracy," under which voters' policy preferences guide outcomes. But still all worth knowing, as information about our world.