...Archive for June 2022

When to tax excess margins

Comments at interfluidity are always great, and the previous post was no exception. I want to quickly address some critiques by commenters, and make clear something that was perhaps not so clear in the prior post:

As commenter Effem points out, taxing margins in the way I’ve proposed is similar to imposing a price control, except that it won’t leave goods priced into shortage (as long as the tax above the price threshold is not 100%). The simplest way to understand it is as an exercise of government-coordinated monopsony power to counter the monopoly power that enables firms to profit from imposing scarcity and raising prices. It’s only an appropriate remedy in this case.

Commenter Somebody offers the very straightforward argument that high prices contribute to supply by bringing higher-priced production on line, which taxing price rises should discourage. If a commodity is being competitively produced such that lower cost sources are fully exploited, preventing price rises would indeed impair supply. But if lower cost sources are not being exploited because an oligopoly industry prefers to profit from scarcity instead, it’s more efficient to eliminate the oligopoly rents and get the cheap production online then to hope price rises bring new entrants into more resource-intensive forms of production. To use the commenter’s example, we don’t want to tap Canadian tar oil sands while cheaper sources of oil remain available but unexploited by producer choice. If industry incumbents are intentionally limiting supply, letting prices rise would only bring out this supply if it would provoke competition by new entrants somehow immune to the barriers preventing exploitation of the more accessible sources.

What if, however, a commodity may occasionally be able to command very high margins, but over a full cycle, its price is very volatile and often below average cost? Then clipping high margins may render the commodity entirely uneconomical to exploit. This is Effem’s concern. One way to deal with this is to let the industry consolidate during downturns, and then let it extract oligopoly rents, so that good times cover the bad. But that is not a great approach. It distributes costs unpredictably, and an industry monopolistic enough to make this work may not confine itself to pricing in a normal rate of profit over the full cycle. Another way to manage this is to incentivize storage when the commodity is cheap that can be mobilized when the commodity is dear to blunt the cycle. Futures markets are supposed to do this to a degree: “spiky” commodities tend to go into backwardization, representing a negative cost of storage net of a convenience yield. But obviously futures markets and the storage they incentivize have not been sufficient to prevent destabilizing swings in the price of oil. For really important commodities, state interventions like the Strategic Petroleum Reserve should probably be expanded. But storage can only be a buffer, sufficiently sharp volatility will break it. And there are services that cannot be stored also subject to high fixed costs and feast/famine cycles, like airlines. These industries are always a regulatory challenge: If you manage to create textbook levels of competition, price falls to marginal cost, which in ordinary times with flush capacity means price does not cover average cost, and everybody goes bankrupt. If you let the industry consolidate into pricing power, it grows predatory, and users prior cycles’ bankruptcy launder progressively higher scarcity rents. Laissez-faire simply does not work for industries like this. The solution, broadly, is to subsidize these industries through famine, while clipping the potential scarcity rents in good times by, e.g. taxing unusual margins. We just subsidized the airlines in famine. Now, they seem to be enjoying scarcity rents, and are not overly urgent about expanding capacity. We need to routinize a full-cycle theory that serves the broad public, rather than our current practice under which industries panic us into bailouts during rough patches, then enjoy scarcity rents during recoveries.

On Twitter, @InquisitiveUrsa suggests using a price collar around forward commitments to encourage petroleum supply. The state could promise to purchase very large quantities of crude to refill and expand the strategic petroleum, promising to pay at least, say, $80 per barrel but requiring a commitment to sell for no more than say $100 per barrel from domestic suppliers. This is a great example of a full-cycle stabilization approach, rather than a half-cycle panicked subsidy or ad hoc price intervention. (I would prefer pairing the committed price floor with an excess margins tax, to eliminate a screw-the-government-we’ll-take-our-scarcity-rents revolt by industry.)

Commenter Brett worries that taxing excess margins would be bad for innovation. There are a bunch of cases to think about there. Innovation can yield high margins in three ways: (1) A market is broadly competitive, but a first-mover enjoys temporary pricing power during the lag time between the introduction of the innovation and when competitors are able to efficiently reproduce it; (2) A market is intentionally uncompetitive because we have used patent, copyright, or other so-called “intellectual property rights” to grant an innovator the capacity to price at high margins; and (3) no formal intellectual property right protects an innovator but network effects or some other barrier to competition vouchsafes persistent pricing power to one or a few key players. Case (1) is the least problematic. High margins won on new products for which there are no obvious barriers to competition should be celebrated, and certainly not taxed. As time passes, competition should discernibly come to reign in prices and margins. If that does not happen, we may eventually consider it an example of case (3). But we should give the innovator the benefit of the doubt for some time. With respect to case (2), as Dean Baker has pointed out, we are already talking about taxes. Intellectual property rights amount to a grant to private parties to use the coercive power of the state in order to receive what are effectively excise taxes on particular goods and services that otherwise would be provided more cheaply. This is intended as a reward to innovation, though I think that intellectual property rentiers have expanded it into an overly costly and poorly targeted incentive to innovate. Rather layering a new tax on persistent excess margins derived from a thicket of IP rights and an army of lawyers, perhaps it would be better just to pare back the IP rights and let competition compress the margins. But in our fallen world, perhaps restructuring and optimizing legal arrangements rabidly defended by powerful interests just won’t happen. In which case, taxing persistent excess margins might be a good backstop, letting innovators fully enjoy their intended monopoly for a some period, but creating hazard for those who don’t let up over the long terms intellectual property lobbyists and lawyers are now able to extend exclusivity. Under case (3), where network effects or some other intractable barrier prevents competition over a long term regardless of our collective choice to reward or not to reward the innovator, one way or another of course we should reign in the pricing power and excess margin. That may involve taxing it, or regulating more directly to ensure “natural monopolies” are managed in the public interest. The permanent winner-take-all network effects that have let Facebook, Google, Amazon, and Apple so thoroughly dominate have been terrible, not good, for innovation.

Longtime commenter benign brodwicz argues excess profits should be transferred to firm workers, rather than taxed. The Whitehouse / Warren proposal would have the tax dividended to individuals, which I think is a fine idea (though a hard sell in an inflationary moment, and unfortunately with a phase-out by income). I don’t think transferring scarcity rents to firm workers is a great idea. The transfers would partially substitute for salary, and the result would be worker interests more strongly aligned with shareholders’ to pursue and exploit monopoly.

Taxing persistent excess margins is in the same family as antitrust regulation. It’s a regulatory idea that in application has to be tailored to the particulars of an industry. In some industries, competition works fine, high margins are transient rewards to innovation until competitors catch up, and nothing should be done. But much of our economy is now concentrated or captured by conglomerations of various sorts who squeeze. It is time to squeeze back. Fundamentally, what we want is a regulatory regime that discourages firms and investors from aspiring to monopoly or tacit cartel. We want firms to understand that even if they achieve persistent monopoly rents, they won’t be able to keep them. Just as it is in the interest of the New York Yankees that their league be regulated so not even they can permanently dominate, we want firms in industries to work proactively with regulators to ensure continuing competition. They should understand the alternative will not be a lucrative stagnation but real risk of expropriation.

Tax excess margins

With prices of commodities like oil soaring, there have been calls for a so-called “windfall-profits tax” from people like Sheldon Whitehouse and Elizabeth Warren. Matt Yglesias brings up the usual objection: If prices are soaring, it’s because producers aren’t producing enough to satisfy demand. Further taxing profits, the story goes, would only blunt incentives to produce, and so would be counterproductive. Yglesias suggests imposing a tax designed to be “inframarginal”, that would tax profits at levels beneath those producers are pretty sure to generate, but leave higher levels of profit untouched to preserve incentives to produce. I agree with Josh Barro that for a variety of reasons, this proposal is “too clever by half”. It would be hard to get right, especially considering how the precedent it sets would be perceived over a longer timeframe.

Nevertheless, it is good that Yglesias moves beyond the neoliberal reflex to assume taxes must always reduce incentives to produce. In theory, a reliably inframarginal tax wouldn’t affect those incentives at all. But we can do even better than that! We can design a profit tax that actually improves incentives to produce!

Holding the cost of goods sold constant, firms have two broad strategies to increase profits: They can increase the quantity of goods that they sell, or they can increase the price. Shareholders may be pleased either way, but the rest of us have a preference for the first strategy. We would like firms to seek profits by increasing production at moderate prices, rather than imposing scarcity and charging high prices. If markets were subject to textbook competition, firms could never choose the second strategy. One firm’s scarce production at high prices would become other firms’ opportunity to expand quantity and market share. But in concentrated industries, or say, an industry dominated by an international cartel and a few large producers, firms may tacitly coordinate to choose the second strategy. Call it “capital discipline”.

When competition does not prevent firms from resorting to scarcity and pricing power, a tax can. The key is not to tax profits per se, but profit margins. Then firms are free to be as profitable as they want to be, if they increase the quantity they produce and sell at customary margins. But if they try to shirk producing and just raise prices, the gravy gets taken away.

You wouldn’t want to tax accounting-derived profit margins. Firms would just raise prices anyway, and find ways to pad costs to keep margins low. But for commodities like oil, we know how expensive oil has to be for even high-cost producers, like US-based frackers, to turn a decent profit on each barrel of oil sold. So all we have to do to penalize the scarcity strategy is tax revenue collected at a fair margin above that price. Suppose the all-in cost to frackers per barrel produced is $80/barrel. We simply impose a tax on revenue above $100/barrel. Once the price of oil rises to this level, it does no good for producers if the price jumps even higher, to $120 or $140. The state takes the extra $20 or $40 away. The only way, then, to increase profits is to sell more barrels, or produce them more efficiently. And that is the incentive structure we want. Rather than blunting incentives to produce, taxing excess margin restores the incentives that, in a textbook economy, competition would provide. The same approach could be applied to refiners’ “crack spreads”. (See David Dayen.)

Not all of our politicians are idiots. If you read the details of the Whitehouse/Warren proposal, it is structured largely this way. It would tax 50% of the difference between current prices and prices known to be profitable for producers during an earlier period. It’s designed to hold producers harmless, if they increase quantities, but reduce the benefit they receive from a higher price. It even exempts smaller producers, to encourage competition to do the work and obviate the tax.

There’s no reason to get too clever by half. Sure, as you’d expect from Elizabeth Warren, a so-called “windfall-profit tax” (that is really an excess margins tax) has a populist stick-it-to-the-price-gougers vibe. But it is also a well designed levy from a technocratic perspective that would enhance incentives to produce when competition is insufficient to discourage industries from seeking scarcity rents. If such taxes are imposed regularly, the threat of them would reduce the incentives for industries to consolidate in the first place. Taxing persistent excess margins is in the sweet spot where good politics and good policy intersect. We should do more of it.


Postscript regarding climate change: I am terrified of climate change. It feels perverse to be writing about fossil fuels as an ordinary commodity whose quantity of production at lowest possible cost we should seek to maximize. Over as short a term as possible, we would like fossil fuels not to be produced at all, to be left in the ground. But I broadly agree with Matt Yglesias that, as a practical, political matter, kneecapping supply is a bad approach to this end. The material pain that unusually high energy prices produce, unless mitigated by some sweetener (like a carbon dividend), provokes backlash that undermines political coalitions serious about climate. High fossil fuel prices not due to an overt tax make the people who work to sabotage climate solutions richer and more powerful. These political circumstances may change, perhaps very soon unfortunately, as each summer is deadlier and more frightening than the last. But for the moment, high fossil fuel prices provoke political reaction more effectively than they promote desirable efficiencies. As long as this remains true, the key to weaning ourselves must be to render fossil fuels expensive relative to carbon-neutral alternatives, rather than in absolute terms. That is, we need to drive down the perceived cost of substituting efficiencies or alternative energy sources so that reduced use of fossil fuels is not painful. This is a political rather than ethical claim. From an ethical perspective, we ought to be willing to tolerate large standard-of-living sacrifices (and the redistributions that would be necessary not to starve people) in order to preserve a habitable planet. Politically, however, we are not there yet, so we must invent spoonfuls of sugar to help the medicine go down.