Tax policy colloquium, week 11: Steve Bank's "Manufacturing Tax Populism: Revisiting the 1962 Campaign Against Dividend and Interest Withholding"

Yesterday at the colloquium, Steve Bank of UCLA Law School (a former Chicago student of mine, way back in the day) presented a tax history paper concerning an interesting episode in modern U.S. tax law: the failed effort by the Kennedy Administration, as part of what became the 1962 tax act, to enact withholding for people’s dividend and interest income, in response to substantial under-reporting. The effort failed due to unexpectedly intense opposition from members of the general public whom the banks successfully riled up and organized into what was apparently a genuine, not merely Astroturf, grass roots opposition movement. Faced with this level of unusual public upset, Congress settled for expanding information reporting, although it would be some time before the IRS would have the ability to make much use of this resource (which nonetheless over time led to increased compliance).

The article isn’t posted because it’s an early draft. It will be a chapter in a larger book concerning a U.S. change in social norms towards greater respectability and acceptance for aggressive tax avoidance behavior. One might view this as a likely consequence of the income tax’s becoming a “mass tax” rather than just a “class tax” on the rich, during World War II. But the change in norms was lagged and perhaps related to broader cultural trends, such as the Vietnam / Watergate / inflation decade-plus of gloom that looks sunny now by our standards, along with the Republican-led shift in political norms starting with Reagan.

Case studies like this don’t need policy payoffs to be interesting – I find them so for their own sake. But three topics in particular struck me as of particular interest here, from the standpoint of conducting a tax policy seminar.

1.         TAX POLITICS

The 1962 “rebellion” against dividend and interest withholding mainly took the form of a flood of letters, postcards, phone calls, et al to Congress, complaining about the dividend and interest withholding proposal. The banks (along with dividend-paying public companies) hated the proposal, at least in part because they would need to incur the costs of administering it, and perhaps also because it helped them attract customers if non-reporting meant that they were in effect offering people tax-free income. Their efforts to get customers, depositors, shareholders, etc. to protest bore fruit to an extent that possibly startled them, and that showed the proposal was touching a nerve (albeit augmented by deliberate misinformation that the banks were happy to spread).

The 1962 act was mainly tax cut legislation, with the investment tax credit as a centerpiece. Its main “populist” feature – a term the paper uses, and that I’ll further discuss below – was a proposed sharp cutback on deductibility for business meals and entertainment. That effort, by the way, in my view largely failed. Congress enacted a new Code provision, section 274, shot through with provisions limiting business meal and entertainment deductibility, but JFK had proposed outright repeal for a lot of the stuff, and instead Congress just put in a bunch of toothless rules about its not being “lavish and extravagant,” and being “directly related” or some such thing to the pursuit of a particular business deal. (Probably the most efficacious new provision was one requiring greater substantiation of these sorts of expenses, rather than allowing them to be estimated.)

Well yes, it's true that JFK said “the three martini lunch must pass from the scene,” or some such thing, and that indeed it has, but I think we can agree that the causation was otherwise.

Anyway, back to dividend and interest withholding. It was a partial pay-for, meant to reduce the overall cost of the act by increasing compliance (rather than enacting a new tax of any sort). It would have had a flat 20% withholding rate, meaning that rich people would still owe more and some at the low end would need to file income tax returns so they could claim refunds. But the resistance to it was “populist,” in the sense that it took the form of: Why is the Administration going after us, mere pikers when it comes to tax avoidance or evasion, when so many rich people and big corporations are getting away with murder? Go after them first, if you’re going after anyone!

Since Congress backed off – albeit, expanding information reporting rather than doing nothing – this is a rare case in which members of the general public affected political outcomes. But it involved the main such vein in which the public does at times influence legislative outcomes. (Let’s note, of course, the point, that the people who protested here weren’t “the public” in the sense of everyone – they were a particular group, albeit one outside the usual corridors of influence.)

To put this outcome in perspective, it’s useful to consider recent political science research about the question of whose preferences affect U.S. political outcomes. A 2014 paper by Martin Gilens and Benjamin Page has a neat chart showing the effects on the probability of a policy’s being adopted changes as support for it rises among (a) average citizens, (b) economic elites, and (c) organized interest groups.

For (a), the general public, the line is flat – rising public support for a proposal from 0 towards 100% has almost no effect whatsoever on the likelihood of adoption. Well-known recent examples would include, e.g., the preference even among Republican voters for increasing taxation of the rich, and the public’s support for addressing climate change.

By contrast, rising support for a proposal by either economic elites or interest groups has a very sharp upward-sloping effect on the likelihood of policy adoption.

Of course, we knew this already. So why was something like interest and dividend withholding in the 1962 act different?

This is by no means a unique episode. Consider, for example, public opposition to Treasury proposals some decades later to (a) require contemporaneous “auto logs,” documenting people’s claims to have used their cars for business, and (b) crack down on people’s making personal use of frequent flyer miles that had been earned through deductible or reimbursed travel. There are plenty of other examples as well from recent decades.

In each of the above three cases, the outrage was generated by policymakers’ efforts to change the status quo on the ground by collecting taxes that were legally due but not being paid. And in each case the proposal would have imposed compliance burdens, in addition to requiring people to pay more than they had become used to paying. Each time this stimulated defensive pushback, focusing both on the compliance burden and on the “unfairness” of targeting decent and hardworking regular folks, etc.

The lesson that I’d derive from this for policymakers, if I were advising them on how to get what they want without arousing general public opposition, could be called “Don’t wake the sleeping baby.” You can do whatever you like in the next room while the baby is sleeping, as long as you don’t start banging gongs (or whatever) in a manner that will rouse it from its slumbers. Changing settled practice and imposing confusing and anxiety-creating compliance burdens is a good example of unduly banging gongs.

But most of the time the baby sleeps quite soundly. Consider the DC legislative process since Trump took office. The public has had very little influence on any of it, with one exception – the attempted repeal of the ACA. That indeed roused public pushback that, by the barest of margins, led to the effort’s failure. (Several Republican Senators were uneasy about voting to take away health insurance from millions of their own voters, but even so the thing nearly happened.)

The public had next to no discernible impact on the passage or content of the 2017 tax act. That was an inside-the-Beltway process on both sides. Probably the most widely noted provision was reducing state and local tax deductions, which may have helped lose a few blue state Republicans their seats the next year. But that was a narrow yet salient adverse change (its effect perhaps overestimated by voters who were previously subject to the alternative minimum tax), and of course recently we’ve seen news about how people can’t tell apart the 2017 act’s actual impact on their liabilities from its apparent impact, the latter being based on what refunds they get, net of changes to withholding.

2.         TAX POPULISM

The paper views the 1962 outcome as to dividend and interest withholding an example of “tax populism,” because discontent so focused on the issue: Why go after us, when the rich and big corporations are getting away with so much more than we are?

BTW, the rich certainly were “getting away” with a lot in those days, in the sense that there was a 91% statutory rate that almost no one was actually paying. There has been a shift since then towards lowering the statutory rate but also requiring that rate to be closer than it was in 1962 to the actual effective rate. In effect, the regime then was: We’ll have very high rates as a kind of public statement, but we’ll tolerate all sorts of devices, a great many of them unambiguously legal, to avoid actually imposing those rates.

But is this “tax populism”? Two particular thoughts on this topic:

(a) If this is populism, it’s very limited and pretextual – People aren’t saying: “Go after tax avoidance by the rich!” They’re saying: “Don’t go after us UNTIL you’ve done that first!” But this is unaccompanied by any strong appetite for actually doing the latter.

Strange as it may seem today, that era’s closest Democratic Party equivalent to, say, Elizabeth Warren or Bernie Sanders today was the Minnesota senator Hubert Humphrey. Among the era’s most progressive mainstream politicians on racial justice issues, Humphrey had long been waiving the era’s more truly “populist” banner for what was then called tax reform. He wanted to attack the era’s income tax preferences that, say, favored the oil industry, or kept effective rate for high-income individuals so far below the top statutory rates.

(As an aside, “tax reform” then meant base-broadening so that the rich and corporations would pay more. In the 1980s, it came to mean instead revenue-neutral and distribution-neutral base-broadening plus rate cuts. In the 2010s, it came to mean absolutely nothing at all.)

Anyway, my sense is that Humphrey never got much mileage out of this tax and other left-populism, either personally or politically. Tax reform in its first manifestation never got too far, in part because the sleeping baby really didn’t care much after all. The scaling back of the 1962 effort to address three martini lunches et al then taught the same lesson.

As for Humphrey himself, JFK famously crushed him in the 1960 West Virginia Democratic primary, being handsomer and far better funded. After that Humphrey, in what sadly became his slavering desperation for the presidency, first became Lyndon Johnson’s abused and widely mocked pro-Vietnam War toady, and then tried a nails-scratching-the-blackboard lurch to the right in 1972 that still got him nowhere. It was a sad ending to a career that had featured earlier moral high points.

Anyway, the point so far as 1962 dividend and interest withholding is concerned is that the public, or such of it as could be roused to harry Congress, might have in principle favored taxing the rich more, but was more than willing to settle for just being left alone itself. Or at least, left alone so far as ways that it would notice were concerned. The expansion of interest and dividend withholding arguably means that, at least over time, riled opponents of the withholding proposal fell well short of getting to preserve the status quo (for many) of significantly under-reporting interest and dividend income.

(b) What is “populism”? – This is a larger and richer subject than I can hope to address here. And there is no single answer, across times and places or even perhaps in a single given time and place. Trump and Elizabeth Warren, for example, have both been called populists, and they do not have a whole lot in common.

But at a very general level, I’d call populism a tool used by one elite against another elite in the effort to call in outside support as they struggle against each other for desired outcomes. Note that elites need not be economic – they can also be social, political, administrative, technological, etc. If you aren’t a member of some elite or other, then you aren’t in the political game to begin with.

Recall the 2012 presidential election. Romney was attacked by the Democrats as epitomizing the out-of-touch business elite (car elevators, firing acquired businesses’ employees when at Bain, etc). Obama was attacked by the Republicans as epitomizing some sort of global / intellectual / professional elite (didn't bowl, liked Grey Poupon mustard, etc.). The business elite on the one hand, and the professional / intellectual / academic elite on the other, have frequently been at odds for more than a century, at a minimum in both the US and the UK, as I show in my forthcoming book DANGEROUS GRANDIOSITY: LITERARY PERSPECTIVES ON HIGH-END INEQUALITY THROUGH THE FIRST GILDED AGE. [Still awaiting a contract to publish, though I’ve had nibbles.] Now, I would say that the super-rich / business elite is the true elite, the one with real power that might reasonably lead to resentment, but then again I’m biased by my own affiliations on the other side of this divide.

3.         THE 1962 CONTROVERSY OVER DIVIDEND AND INTEREST WITHHOLDING

Okay, but what about the 1962 political fight that Bank’s chapter documents and discusses? It was reprised, by the way, in 1982, when Congress actually enacted dividend and interest withholding (as part of that year’s revenue-raising TEFRA legislation), only to repeal it, in the face of similar public outcry, six months later and before it took effect.

Here I’d say, despite any seeming disparagement above of the outcry, that the public may have had a very good point. Dividend and interest withholding truly did impose burdens, and not just on the banks but on people who would have had to file for refunds. And, for behavioral economics reasons, it would likely have reduced saving by people who target-save specified amounts in distinct vehicles, and simply spend the rest of their after-tax income, for the same reason that people apparently save more through Roth IRAs than traditional IRAs, without realizing that they’re doing so, when they fail to equate taxes within the savings vehicle to those imposed outside it.

Yes, compliance was a problem, and it was perfectly reasonable for the technocrats to aim at increasing it, but they may have under-appreciated the burdens being imposed relative to working harder to make information reporting work.

Do we need interest and dividend withholding today? Probably not – compliance with respect to interest and dividends is pretty high due to information reporting.

But what about wage withholding? Well, the thing is that everyone wants that – apart from Grover Norquist types today, and Milton Friedman when he lamented having worked on its WW II era implementation.

The objections of a Friedman or a Norquist rest on the fact that people perceive their income tax burdens as smaller when they pay in small bits, and perhaps even get a refund on April 15, than if the whole thing came due then. But the taxpayers themselves want it to feel smaller. And they don’t want to have to plan for and then implement a large-scale cash withdrawal from their savings on Tax Day.

Against this background – scarcely at all applicable to dividend and interest withholding, especially given the rules for filing estimated taxes – someone who took to the campaign trail pushing the Friedman-Norquist viewpoint would likely see it fall quite flat.

Thus, consider Barry Goldwater’s 1964 presidential campaign. Should he have taken the hint from what happened in 1962 that withholding was generally unpopular, and therefore concluded that he should campaign for the repeal of wage withholding? I doubt this would have worked well for him, and his advisors surely knew that rather than overlooking an opportunity.

Bottom line (or one among others): perhaps the story told in Bank’s Manufacturing Tax Populism should be viewed as having a happy ending, in the sense that it yielded a very reasonable outcome: more information reporting, but no dividend and interest withholding. And the fact that its "success" of populism was accompanied by a more significant failure with regard to the treatment of high-end business meal and entertainment expenses, is instructive as well. (When that stuff was more efficaciously scaled back later, that was a revenue-driven, inside-the-Beltway story.)