Welcome back to Conglomerate Junior Scholars Workshop.  Today's paper is the second of two papers with a tax component:  A Republic of the Mind:  Cognitive Biases, Fiscal Federalism, and Section 164 of the Tax Code by Brian Galle.  Don't let the fancy title fool you -- this is a topic that affects any of us that pay any amount of taxes to a state or local government (i.e., all of us).  Today's expert reviewer is Kirk Stark, a law professor at UCLA, where he writes and teaches in the area of taxation, including state taxation.  Professor Stark's comments are above and below the fold:

Brian Galle’s paper, A Republic of the Mind: Cognitive Biases, Fiscal Federalism, and Section 164 of the Tax Code, examines several arguments that have been offered in support of and against the deduction under the current federal income tax for taxes paid to state and local governments. As background, readers should know that the deduction for state and local taxes (sometimes called SALT for short) is available under section 164 of the Internal Revenue Code. This is an itemized deduction claimed on Schedule A of the taxpayer’s Form 1040. Up until 1986, a deduction was allowed for income, sales and property taxes paid to state and local governments. As part of the Tax Reform Act of 1986, Congress repealed the deduction for sales taxes, following a vigorous effort on the part of the Reagan administration to repeal the deduction entirely. More recently, as part of the American Jobs Creation Act of 2004, Congress reinstated the deduction for sales taxes (kind of) by allowing taxpayers to choose between deducting income taxes or sales taxes.

In general, this election will only be of use to taxpayers in states without an income tax (most prominently Florida and Texas). Meanwhile, it is also worth noting that the deduction is not allowed at all in computing the alternative minimum tax, which has affected more and more taxpayers in recent years. In fact, the SALT deduction is the single largest AMT “preference item” and is the main cause for pushing millions of new taxpayers into the AMT in higher-tax states such as New York and California. As I have explained in some of my own work, the mounting AMT bite means that we are essentially in the process of repealing the deduction for state and local taxes. So the Galle paper on this subject is both timely and important.

Galle announces his main objective of Republic of the Mind in the opening page of the article. Noting that “the last comprehensive legal academic examination of the merits of the deduction” was published in 1996, he aims to offer an updated review of the key issues and examine the SALT deduction from several new angles. He begins with a general overview of the literature and then launches into substantive analysis beginning in Part II and continuing through Part V.

Throughout the article, Galle’s analysis of section 164 is wide-ranging and provocative. (Indeed, if asked to identify the chief weakness of the article, I would suggest that perhaps the article is too wide-ranging. More on that point in a bit…). Over the course of fifty or so pages, Galle reviews and critiques arguments relating to cognitive biases that taxpayers face in determining their preferences for state and local public services, costs that inhibit taxpayer mobility, the pros and cons of tax exporting, the effect of federal mandates on state/local fiscal behavior, the importance of federal grants to various state/local programs, the possibility of evaluating taxpayer “utility” on an objective (rather than subjective) basis, the costs of tax enforcement and how the SALT deduction might reduce those costs, the idea that the SALT deduction should be viewed as a federal “subsidy” for state/local spending, the role of redistribution in state/local spending and how that influences arguments for and against the SALT deduction, the presence of positive and negative externalities (so-called “spillovers”) in state/local government behavior, and various other topics. Each of these issues is treated with care, and Galle does a nice job of making sure the reader is not left with a one-sided perspective on any given topic.

As this long list of topics suggests, there are many interesting and promising ideas lurking throughout this paper, which leads me to a point about focus and framing. A senior colleague once offered some advice to me that might be worth considering here (and, in any event, is helpful for all of us to bear in mind): don’t be afraid to leave yourself something to write about in future articles! There may well be two or three potential articles within this paper, and my own (perhaps idiosyncratic) view is that Galle would do well to break them apart into separate projects. But enough with advice—let’s move on to the substance. Rather than reviewing and evaluating each one of these ideas, I would like to focus my remarks on what seems to be the central argument of the paper—i.e., the “cognitive bias” argument featured in the title of the piece and discussed primarily at pages 12-19.

In these pages, Galle reviews, describes and critiques what he calls the “equity view” of the SALT deduction. According to Galle, “most commentators initially seemed to view the deduction as simply an equitable measure intended to put on an equal footing those who earned similar incomes in jurisdictions with differing tax rates.” (page 5). Under this view, as Galle explains it, two taxpayers with, say, $100,000 of income each should not bear the same federal tax burden if one (Snoop) pays $10,000 in state/local taxes while the other (Dre) only pays $3,000. The “proper” and most theoretically justifiable treatment of these taxpayers, the argument would go, is that Snoop should pay tax on $90,000 of income, while Dre should pay tax on $97,000. To treat Snoop and Dre the same, the traditional equity argument would posit, would violate principles of “horizontal equity.” The idea here is that taxes represent true economic losses—much like theft or flushing money down the toilet, in the sense that they reduce well-being and thus should reduce income in a normative income tax.

The traditional response to that argument, again following Galle’s description, is that taxes are not like theft at all. As we all know, taxes aren’t just flushed down the toilet; they pay for (hopefully valuable) public goods and services. So upon closer inspection, it becomes apparent that state and local taxes are not all that different from any other type of consumption. To be sure, paying taxes isn’t exactly like spending the day at the Galleria or surfing amazon.com. There is an element of compulsion involved with taxes that one typically doesn’t see in private market transactions. But taxpayers aren’t powerless in this process—they participate in deciding who is going to put together the service bundle offered by the government and, increasingly, they get an up or down vote on whether or not they agree with the taxes used to pay for those services. Moreover, the fact is that taxpayers can choose among a multitude of state/local jurisdictions to get the package of taxes/services that most closely approximates their personal preferences. Indeed, in the perfect Tiebout world, where mobility is costless, information is perfect and the number of jurisdictions from which to choose is infinite, taxes will match the value of services exactly. Taking these ideas seriously, it would appear that, in some fundamental sense, state/local taxes are just like private market consumption and, therefore, should not be deductible for purposes of the federal income tax.

Galle's contribution to this debate—at least in that portion of the paper dealing with “cognitive bias”—is to complicate this Tiebout-inspired market picture by introducing the possibility that taxpayers’ decisions about state/local fiscal matters may be systematically skewed by certain cognitive tendencies. I won’t review all of the cognitive biases that Galle identifies, but a couple of examples will help to illustrate what he is getting at. For example, Galle suggests that the so-called “endowment effect” may lead taxpayers to overvalue the tax/service package they presently enjoy as compared to the one they might opt into by moving. Here Galle is relying on studies showing that individuals often place a higher value on assets they own (a mug, a car, whatever) than they do if they must purchase the same asset from someone else.

As another example, Galle suggests that taxpayers may underestimate the true cost of taxes if state/local policymakers utilize multiple small, low-visibility taxes rather than large, highly visible taxes. Here one might imagine policymakers opting for various commodity taxes, which are generally paid in dribs and drabs as the commodities are purchased, rather than all at once in a single tax filing. On the opposite extreme, we might see property taxes, which in many instances are paid in large chunks maybe once or twice a year.

In short, Galle emphasizes that there may be reason to doubt the rationality of taxpayer preferences regarding the tax/service packages they choose. As these examples suggest, taxpayer choices may be ill-informed or skewed by systematic psychological tendencies. Can we identify any generalizations regarding how or, more precisely, in what direction taxpayer preferences are skewed? As Galle himself explains, “the general trend of the biases I’ve described seems to be toward overpayment…[A] taxpayer might remain in a state where she pays considerably more than she is actually getting. It seems rather harder to construct a story in which biases might produce a consumer surplus.” This passage suggests that, if anything, Galle sees the cognitive spin on state/local taxes as putting a thumb on the scale in favor of deductibility. After all, if taxpayers are “overpaying” then we move a little closer to the idea of taxes as pure economic losses.

For what it’s worth, my sense is that the facts on the ground are exactly the opposite of what Galle suggests in this brief passage. That is, if anything, taxpayers in general tend to have an exaggerated view of the cost of state and local taxes they pay and underestimate the value of state/local public services they receive. But to be fair, this is an empirical point that I will set aside. For now, let us simply accept Galle’s claim that “cognitive bias” might skew the market for state/local public goods in the direction he suggests. What then should be done?

Rather than addressing that question as it might apply to the deduction for state and local taxes, I want to identify a point of potentially broader significance for the income tax that Galle’s analysis usefully highlights. If Galle is right that cognitive biases can skew market decisions, and if we take seriously the idea that tax rules should be structured so as to take account of these biases, then why stop at the deduction for state and local taxes? To focus the issue, consider the following example. Let’s assume for the moment that some sort of identifiable “cognitive bias” leads people to go to the movies much more frequently than they would if they were perfectly rational economic agents. That is, the rational taxpayer would pay $200 per year for movie tickets, while the cognitively biased taxpayer spends $500. Should a tax system sensitive to cognitive bias give those who spend $500 on movie tickets a deduction of $300?

If this question sounds fanciful, consider the case of David Zarin, who borrowed millions from a casino, lost it all in a compulsive gambling spree, and then settled his debt with the casino for pennies on the dollar. While Zarin’s case was decided on some technical aspects of section 108 of the tax code (dealing with income from discharge of indebtedness) it has provoked an ongoing debate among tax scholars about whether Zarin really experienced millions of dollars of consumption value in his gambling spree. In the end, Zarin won his case and was spared from having to pay tax on the “free consumption” the casino made available to him. In some ways, this is exactly like the cognitively biased state/local taxpayer in Galle’s discussion. One view of the Zarin outcome (though not necessarily the one I endorse) is that Zarin was appropriately excused from paying tax on the grounds of what Galle might call “objective” or “hypothetical” utility (see pages 38-41). Once we start down this road, however, it is difficult to find a principled place to stop.

As Galle rightly suggests, there are all sorts of reasons why people make market decisions—including decisions regarding what kind of state/local taxes and public services they want—that call into question the rationality of their decision-making processes. Aside from cognitive bias, sometimes people simply make mistakes. Sometimes people make impulse purchases. Sometimes they don’t shop around as much as perhaps they should. On the other end of the spectrum, sometimes people stumble on really good deals. The cumulative effect of all of these difficulties is to cast a shadow of doubt on private market transactions as “perfect” measures of taxpayer utility. And in the state/local context, that may make us somewhat suspicious of the Tiebout model and what Galles calls the traditional critique of the equity view.

Yet despite all of these imperfections in private market behavior, it is hard to imagine a more reliable system for determining how much benefit each individual happens to derive from the transactions into which she enters. At bottom, Galle’s suggestion that some sort of alternative approach is preferable (he alludes to his preference for a “Rawlsian” approach, but does not elaborate) is a challenge to the very idea of “consumer sovereignty.” Perhaps consumer sovereignty deserves to be challenged in the context of the market for state/local public goods and services. Galles has taken the important first step of offering some suggestions for why that might be the case.

As other readers will surely note, I have only scratched the surface of Galle’s ambitious paper. There is much here to comment on, and I hope that other readers will take the opportunity offered by the good folks at Conglomerate to post their views on the Galle paper. I enjoyed having the opportunity to read and comment on this paper, and I look forward to comments from others.

Junior Scholars | Bookmark

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8345157d569e200d834568ba053ef

Links to weblogs that reference Conglomerate Junior Scholars Workshop: Galle on Deducting State/Local Taxes from Federal Taxable Income: