NYU Tax Policy Colloquium on Mason & Knoll, Unbundling Undue Burdens

 Yesterday at the colloquium, we discussed the above article, coauthored by Ruth Mason and Michael Knoll. In a welcome departure from COVID-era norms, they were actually both physically present for the event.

The article is the fruit of a LOT of research, including careful reading of a large number of Supreme Court cases on the dormant commerce clause (DCC), in both the tax and regulatory realms. This is one of those times when one says: I’m glad they did it, so I don’t have to. These cases are notoriously turgid, all over the place, and difficult to fit into a coherent framework. But they are also, I think, important.

1) SOME SHORTHAND BEFORE WE GET STARTED

Let me start with a bit of background shorthand from the paper, so that the terms noted here can simply be plugged in below.

Internal consistency – In the 2015 Supreme Court case of Maryland v. Wynne, the Supreme Court struck down a Maryland tax provision on DCC grounds as violating “internal consistency.” (Knoll and Mason wrote an amicus brief, urging analysis that the Court then adopted.) This term is useful but not immediately intuitive, and hence worth explaining briefly here.

Maryland’s income tax taxed residents at 3.2% percent on both Maryland and non-Maryland income. It provided no credit for non-Maryland state income taxes paid on non-Maryland income. It also taxed nonresidents on Maryland source income at 1.25%.

Maryland argued that this tax oughtn’t to present a DCC problem. After all, it taxed Maryland residents’ outside income at the same rate (not a higher rate) than their local income, and it taxed non-residents at a lower rate than residents. But the tax was struck down as discriminatory against interstate commerce because it violated internal consistency.

The question asked by internal consistency is whether, if each of say 2 states adopted the challenged rule, cross-border transactions between them would be disfavored relative to same-state. The answer here was YES. Suppose we have Maryland 1 and Maryland 2, each with the same rules. A resident of either state would pay tax at a 3.25% rate on in-state income. But income earned in the other state would face a 4.45% tax rate (3.2% under one’s own state’s residence-based tax, and 1.25% in the other state from the source-based tax, with no abatement given the lack of home-state credit).

Knoll and Mason, along with Alan Viard, have discussed elsewhere why internal consistency is a sound economic method for identifying provisions that are tariff-like in the sense of burdening cross-border activity relative to one-jurisdiction activity. Obviously, an overt tariff would violate internal consistency, since only cross-border trade would lead to its being incurred, but the aim here is to generalize the concern and the inquiry.

Note that the inquiry does not involve actually looking at what other jurisdictions do. Although I agree with Mason and Knoll that this is an appropriate mode of inquiry into both the DCC and international equivalents (e.g., the ECJ’s review of member tax provisions that are charged with violating the EU’s commitment to a single market), the point for now is just that it offers a powerful and convenient analytic.

Pike balancing – In Pike v. Bruce Church (1970), the Supreme Court struck down an Arizona rule requiring cantaloupes that were grown in-state to be boxed in-state. The claimed rationale for the rule related to letting consumers know that particular cantaloupes were from Arizona (thus, e.g., spreading awareness of their quality). But it violated the DCC by burdening companies from out-of-state that might have boxing facilities (and cantaloupes) from elsewhere. “Pike balancing” is the process of thus weighing benefits vs. the burden on interstate commerce where it is a single-state issue. That is, one is looking at the Arizona rule independently of the question of how it fits with other states’ rules.

Bibb balancing – In Bibb v. Navajo Freight Lines (1959), Illinois had required truckers that were using its roads to use curved mudflaps to protect the cars behind them, whereas other states were requiring straight mudflaps. The Supreme Court struck down the rule under the DCC. Here the issue involved “interaction burdens” rather than a single-state analysis. That is, there was nothing wrong with curved mudflaps as such. But given other states’ widespread adherence to a straight mudflaps requirement, trucks engaged in interstate commerce would have to stick on new mudflaps – kindly made available, for a fee, by local Illinois businesses – if they wished to use Illinois roads. Just as in Pike, the court balanced legitimate in-state benefits (e.g., suppose the curved mudflaps worked better, at least under Illinois’ road conditions) against the burden to interstate commerce. But Bibb balancing differs from Pike balancing because it not only requires examining a broader set of questions regarding how different states’ regulations interact with each other, but also (if a DCC violation is found) may in effect require the courts to pick between alternative rules and cement one of the choices in place simply because it has become the prevalent one. This has various downsides, such as its potentially impeding the emergence of new and improved approaches. But if one doesn’t do it, one is inviting states to play the kind of destructive game that Illinois may conceivably have been engaged in here.

Green margarine rules – This term isn’t in the paper, but I find it expositionally useful. Wisconsin is a big dairy state, and hence a butter producer. Margarine predominantly comes from outside the state. For many years, Wisconsin forbade the in-state sale of yellow margarine. (I remember hearing decades ago from my parents that Wisconsin required margarine to be green, which apparently is not true – it could be white under the rule. But let’s say green because it is more vivid.)

A green-margarine rule in a dairy state looks on its face like it is simply there to help consumers tell apart margarine and butter. But in context it may be sneaky protectionism, aimed at reducing consumer demand for the imported relative to the local product because yellow is the color that looks right to them in this context.

It seems to me that the green-margarine rule ought to have been challenged on DCC grounds, but if it was the challenge apparently failed, and there is no well-known precedent. So let’s just call this type of effort at disguised protectionism a “green margarine rule.” Whether we group a challenge to it under Pike or Bibb (an issue that I discuss below), it seems clear that there is a DCC issue that may need, under existing precedent (but for good reason) require the court to balance the burden to interstate commerce against the legitimate (i.e., non-protectionist) benefits.

2) UNDERLYING THEORETICAL BASIS FOR THE DORMANT COMMERCE CLAUSE

The DCC can reasonably be viewed as reflecting an underlying set of views about both economics and politics. These are worth describing briefly because they affect not only its merits (some Supreme Court members, and of greater intellectual interest also some commentators, want to get rid of it), but also its interpretation.

In brief, the DCC empowers the federal courts to strike down state rules (whether taxes or other regulation) that are protectionist. The classic protectionist rule is an overt tariff, but a DCC inquiry also looks at provisions that are challenged as being tariff-like.

The DCC thus applies a bit selectively with regard to the totality of distortions that might arise from a given state’s having distinctive taxes or other rules. Other problems may be addressed by Congress, exercising its legislative powers under the active commerce clause (ACC), but are thought to lie outside the proper scope of DCC review.

For example, whenever two states have different tax rules in any way, this can violate locational neutrality. For example, if State A has a lower income tax rate than State B, this creates an incentive to earn income in A rather than B. But while Congress may address locational neutrality problems under the ACC, the courts don’t do so under the DCC.

Likewise, there may be spillovers such as that resulting from a race to the bottom. An example is tax competition – if one views it as normatively creating a race to the bottom, rather than (as I discuss below) to the top. Again, Congress can address this under the ACC, but courts don’t under the DCC.

Another issue is tax exportation (actual or attempted), but to things from getting too longwinded I will leave it out here.

Economics – There is a long tradition, prominent (for example) in the work of Adam Smith, to the effects that tariffs are generally inefficient. Indeed, unless a given jurisdiction has the requisite market power, it is likely to make itself worse-off on balance, not just the people in other jurisdictions, by reason of imposing them. Moreover, while it’s true that “optimal tariffs” (where one has the requisite market power) can benefit the jurisdiction that imposes them, there is still likely to be deadweight loss from the standpoint of all jurisdictions – suggesting that they should all agree to forgo even optimal tariffs if they are making this choice behind the veil.

If we accept the generalization that tariffs are bad, and what’s more view the federal courts under the DCC as well-situated to police the field, then it may make sense to hand them this power. But locational neutrality has far less definite policy implications, once one accepts that it nonetheless makes sense to allow inter-jurisdictional rule variation. So relegating this to the ACC and legislative choice makes sense.

Likewise, the only difference between a “bad” race to the bottom and a “good” race to the top is how one views the suggested equilibrium normatively. Consider the classic case of Delaware’s preeminence with respect to corporate law. This was once viewed as a race to the bottom, allowing managers to exploit the shareholders by reason of incorporating in a place where the governance rules are lax. Then it came to be viewed as a race to the top, on the ground that capital markets work well enough to incentivize good place of incorporation choices by founders and managers. Thus, the question of whether the race to Delaware should be addressed at the federal level depends on one’s view of how well capital markets are functioning – an issue completely distinct from that of the spillover process as between the states, when Delaware succeeds in capturing this business. Once again, the implication is that this is a legislative policy choice, and hence reasonably placed within the reach of the national legislature under the ACC, but not handed to federal courts under the DCC.

Politics – Why is there even a problem with tariffs (other than optimal tariffs), if sub-federal jurisdictions don’t generally benefit from imposing them? So why even bother with higher-level oversight. The standard answer is that jurisdictions are prone to enacting tariffs, and other tariff-like protectionist provisions, due to interest group politics. For example, concentrated producer groups and their lobbyists prevail legislatively over diffuse consumer interests.

So there is also a political theory under which the states might actually want federal oversight, including via the federal courts and the DCC, in order to help protect them against themselves. But this theory is about protectionism in particular. In other scenarios, even if the states are willing to concede (and indeed have conceded) ACC powers, they may not want to make it federal oversight as automatic and widespread as the DCC may make it with respect to protectionism. Plus, again, the optimal remedy isn’t so clear with respect to locational neutrality and race to the bottom vs. the top scenarios.

3) SINGLE-STATE ANALYSIS

Again, both internal consistency as in Wynne, and Pike balancing, involve a single-state analysis. One is looking at a given state’s rules in isolation, not at how different states’ rules interact as in Bibb.

Under existing doctrine, endorsed by Knoll and Mason, internal consistency is distinctively relevant to tax cases, and results in per se finding of commerce clause violation – no Pike balancing needed. This potentially raises issues of tax exceptionalism that I discuss below. But herewith are a few efforts to poke at the distinctiveness, notwithstanding that (for reasons I’ll omit here) I am generally comfortable with this state of the doctrine.

Internal consistency in form, but arguably not in substance – Suppose a state with a retail sales tax imposes a use tax rate (for purchases of out-of-state) that exceeds the sales tax rate on sales made in-state. On its face, this is a clear violation of internal consistency, and indeed it could scarcely look any more tariff-like. But suppose that widespread avoidance of the use tax meant that this was, in the aggregate, moving towards the overall equalization of effective tax rates on in-state vs. in-bound sales. Would this potentially offer a defense, at least requiring the court to evaluate the underlying merits instead of striking down the higher nominal levy per se? If so, do we enter the realm of Pike balancing?

No internal consistency violation in form, but arguably yes in substance – Recall the Wisconsin “green margarine rule.” On its face, this has nothing to do with internal consistency. But suppose we consider this requirement equivalent to an explicit tax – and Wisconsin indeed had a margarine tax, on top of the green margarine rule – and generalized it as “Impose a special tax on mainly non-local businesses that compete with mainly local businesses” – then one could view it as an internal consistency violation. Is this a helpful perspective, at least in the quest for intellectually consistent rules across the board?

Can internal consistency apply beyond tax? – In the green margarine example right above, I considered both calling the rule in effect a tax, and stating it in a more general form (which the state imposing it would of course dispute). But what if we look at single-state analyses of other standard examples from outside the tax realm. Thus, recall Pike, in which, as discussed above, Arizona tried to require in-state boxing of Arizona cantaloupes. 

Suppose that all states required in-state boxing of in-state cantaloupes. Then you have 2 cantaloupe-growing states, Arizona and California. Any one-state cantaloupe grower can enjoy the efficiencies (if there on balance) of boxing everything at the same plant. But any two-state cantaloupe grower needs to have two boxing plants, one in each state. This might conceivably create competitive disadvantage for multi-state cantaloupe growers.

I run through this simply as a way of getting a better handle on how internal consistency fits into the broader doctrine. It may not matter substantively whether or not we call this an internal consistency violation – although it might conceivably aid us in identifying suspicious cases – unless the identification meant that there would be a per se violation in lieu of Pike balancing. So one has to consider the case for tax exceptionalism, which again I will get to shortly.

4) INTERACTION PROBLEMS

Again, Bibb balancing pertains to interaction problems, which involve a disparity between states’ rules, rather than how any one of them operates in isolation. And again, the problem with Illinois’ rule requiring curved mudflaps was that it imposed a cost on out-of-state or multi-state trucks given that straight mudflaps were generally required elsewhere. The paper argues that, while Bibb balancing may be necessary given the ingenuity of legislators (or lobbyists) in particular states – along with the barriers to Congressional action and the difficulty of proving malign intent even if one deems that a suitable focus in principle – it has also proven especially tricky in practice, including because it may effectively place an existing prevalent rule beyond the reach of evolutionary change that might conceivably have been for the better.

The paper usefully distinguishes 3 scenarios regarding how such evolution, if permitted to go forward, might systematically unfold. One is the Delaware scenario of race to the bottom (or to the top, depending on one’s normative view regarding the end state).

A second is the California scenario, so named because California is a big enough player, given the size of its economy, to prescribe rules that companies will then decide to follow nationwide. (E.g., if it has tougher emission rules for cars than the federal government, companies may decide that it’s cheapest simply to follow those rules nationwide.)

A third is the Balkans scenario, a term that the paper uses in the spirit of “balkanization”), in which markets get segmented by reason of incompatible regulatory rules. A classic instance of this problem was the instance in which the ECJ had to deal with Germany’s requiring that liqueurs have at least 25% alcohol, while France required that they not have more than 20% alcohol. The ECJ solved the problem here, arguably Solomonically, by invoking a rule called mutual recognition, under which each state had to allow the sale of liqueurs that met either rule.

DCC issues arise under the Balkans scenario, leaving the Delaware and California scenarios purely for the federal legislative power under the ACC. But one complication here is that it won’t always be so clear which scenario presents itself most forcefully.

To illustrate, consider Moorman Mfg. Co. v. Bair, the 1978 Supreme Court case that, by a 5-4 vote, upheld Iowa’s rule (which was novel at the time) replacing the standard 3-factor formula (sales, property, and payroll) for determining in-state income with a purely sales-based formula. Consider the following 3 scenarios that, based on knowledge at the time, might conceivably have transpired:

Scenario 1 – The prior near-consensus in favor of the 3-factor formula is required by each state’s choosing whatever formula worked for it best. E.g., Iowa reportedly chose sales because it was a “market state” in which there wasn’t much corporate property or payroll. Suppose that similarly self-interested choices were made by “property states” and “payroll states.” National businesses might then have found themselves including more than 100% of their national income in one state or another, disfavoring them relative to one-state businesses. This would systematically disfavor interstate commerce, relative to one-state activity, and arguably would have supported intervention under the DCC even if it doesn’t quite achieve Balkans levels of disjuncture. But this is not in fact what happened.

Scenario 2 – The diversity between states’ formulas ends up imposing tariff-like extra administrative and compliance burdens on national businesses. Again, this is not what happened, and it was perhaps never a realistic threat, but again we’d be in DCC territory. This resembles the 1967 National Bellas Hess case, overturned by Quill and then Wayfair, which held at the time that use tax compliance was simply too administratively burdensome to be demanded of national businesses. I recall the case saying that there were 50,000 separate sales tax districts, including mosquito abatement districts. And in a pre-computer age this simply imposed too much burden. It is plausible that National Bellas Hess was a reasonable decision when issued, and only became unreasonable later as the rise of computer technology and the Internet both (a) triggered the growth of interstate commerce, and (b) made compliance far cheaper than it had previously been.’

Again, this scenario might in principle have supported a DCC challenge to what Iowa was doing in Moorman, but even at the time probably was not factually plausible.

Scenario 3 – What actually happened is that states learned of the advantages that they would enjoy, via tax competition, from using sales-only allocation. States didn’t want to discourage the local placement of property and payroll, and sales were far less tax-elastic insofar as this depended on where the consumers lived. So there was a race to the bottom towards sales-only apportionment – AKA a race to the top, if you happen to view this as a better approach than use of the 3-factor formula. Good or bad, it does seem to lie outside the DCC’s concerns.

5) TAX EXCEPTIONALISM

Lately and with good reason, tax folk have tended to be more self-conscious than previously about tax exceptionalism. This is the view that tax cases are unique and distinctive, and hence should be treated differently than non-tax cases. (By using the word “cases,” I don’t mean to restrict myself to DCC cases or other litigation. It arises as an issue in a variety of realms.)

The paper early on criticizes doctrinal instances of what it calls “unjustified tax exceptionalism,” such as the 4-factor DCC test for tax cases under Complete Auto Transit v. Brady (1977). But it later explains why tax exceptionalism may be justified with respect to treating internal consistency violations as per se adversely dispositive in tax cases, but not non-tax cases. This is distinct from requiring Pike balancing.

It makes clear that it is being substantive rather than formalistic in how it defines tax cases for this purpose. For example, a Pigovian tax would be treated like regulation in terms of DCC review. Likewise, issues of administrative burden from complying with a tax (as in my hypothetical Scenario 2 for Moorman) would face balancing. But when it’s just a matter of getting revenue, the paper argues that money is money. So a state’s interest in getting tax revenues in a particular manner that violates internal consistency is trivial, since they can get the same $$ some other way.

In response, one could argue that not all tax revenue is the same, in that states may have particular reasons for wanting to raise $$ in one way rather than another. But agreed, in a case such as Maryland v. Wynne, once one has shown through internal consistency that the provision does indeed have the tariff-like effect (whether intended or not and actualized or not) of burdening cross-border activity relative to one-state activity), the state’s legitimate interest in getting the $$ exactly this way, rather than in some other way, might reasonably be viewed as fairly trivial (making the formal occurrence or not of balancing rather a non-event). Whereas, under, say, the Pike or green margarine scenarios, it may be more necessary to look in detail at the state interests, even if they end up being deemed trivial.