Over the past two years, heterodox economic theory has burst into the public eye more than ever as conventional macroeconomic models have failed to explain the economy we’ve been living in since 2020. In particular, theories around consolidation and corporate power as factors in macroeconomic trends–from neo-Brandeisian antitrust policy to theories of profit seeking as a driver of inflation–have exploded onto the scene. While “heterodox economics” isn’t really a singular thing–it’s more a banner term for anything that breaks from the well established schools of thought–the ideas it represents challenge decades of consensus within macro- and financial economics. This development, of course, has left the proponents of the traditional models rather perturbed.
One of the heterodox ideas that has seen the most media attention is the idea of sellers’ inflation: the theory that inflation can, at least partially, be a result of companies using economic shocks as smokescreens to exercise their market power and raise the prices they charge. The name most associated with this theory is Isabella Weber, a professor of economics at the University of Massachusetts, but there are certainly other economists who support this theory (and many more who support elements of it but are holding out for more empirical evidence before jumping into the rather fraught public debate.)
Conventional economists have been bristling about sellers’ inflation being presented as an alternative to the more staid explanation of a wage-price spiral (we’ll come back to that), but in recent months there have been extremely aggressive (and often condescending, self-important, and factually incorrect) attacks on the idea and its proponents. Despite this, sellers’ inflation really is not that far from a lot of long standing economic theory, and the idea is grounded in key assumptions about firm behavior that are deeply held across most economic models.
My goal here is threefold: first, to explain what the sellers’ inflation and conventional models actually are; second, to break down the most common lines of attack against sellers’ inflation; third, to demonstrate that, whatever its shortcomings, sellers’ inflation is better supported than the traditional wage-price spiral. Many even seem to recognize this, shifting to an explanation of corporations just reacting to increased demand. As we’ll see, that explanation is even weaker.
As briefly mentioned above, sellers’ inflation is the idea that, in significantly concentrated sectors of the economy, coordinated price hikes can be a significant driver of inflation. While the concept’s opponents generally prefer to call it “greedflation,” largely as a way of making it seem less intellectually serious, the experts actually advancing the theory never use that term for a very simple reason: it doesn’t really have anything to do with variance in how greedy corporations are. It does rely on corporations being “greedy,” but so do all mainstream economic theories of corporate behavior. Economic models around firm behavior practically always assume companies to be profit maximizing, conduct which can easily be described as greedy. As we’ll see, this is just one of many points in which sellers’ inflation is actually very much aligned with prevailing economic theory.
Under the sellers’ inflation model, inflation begins with a series of shocks to the macroeconomy: a global pandemic causes an economic crash. Governments respond with massive fiscal stimulus, but the economy experiences huge supply chain disruptions that are further worsened with the Russian invasion of Ukraine. All of these events caused inflation to increase either by decreasing supply or increasing demand. The stimulus checks increased demand by boosting consumers’ spending power–exactly what it was supposed to do. Both strained supply chains and the sanctions cutting Russia off from global trade restricted supply. Contrary to what some opponents of sellers’ inflation will say, the theory does not deny the stimulus being inflationary (though some individual proponents might). Rather, sellers’ inflation is an explanation for the sustained inflation we saw over the past two years. Those shocks led to a mismatch between demand and supply for consumer goods, but something kept inflation high even after the effects of those shocks should have waned.
The culprit is corporate power. With such a whirlwind of economic shocks, consumers are less able to tell when prices are rising to offset increases in the cost of production versus when prices are being raised purely to boost profit. This, too, is not at odds with conventional macro wisdom. Every basic model of supply and demand tells us that when supply dwindles and demand soars, the price level will rise. Sellers’ inflation is an explanation of how and why prices rise and why prices will increase more in an economy with fewer firms and less competition.
Sellers’ inflation is really just a specific application of the theory of rent-seeking, which has been largely accepted since it was introduced by David Ricardo, a contemporary of the father of modern economics, Adam Smith. (Indeed, this point, which I raised nearly a year and a half ago in Common Dreams, was recently explored in a new paper from scholars at the University of London.) As anyone who has ever watched a crime show could tell you, when you want to solve a whodunnit, you need to look at motive, means, and opportunity. The greed (which, again, is at the same level it always is) is the motive. Corporations will always seek to charge as high of a price as they can without being dangerously undercut by competitors. Sellers’ inflation doesn’t posit a massive increase in corporate greed, but a unique economic environment that allows firms to act upon the greed they have possessed.
Concentration is the means; when the market is in the hands of only one or a few firms, it becomes easier to raise prices for a couple of reasons. First, large firms have price-setting power, meaning they control enough of the sector that they are able to at least partially set the going rate for what they sell. Second, when there’s only a few firms in a sector, wink-wink-nudge-nudge pricing coordination is much easier. Just throw in some vague but loaded phrases in press releases or earnings calls that you know your competition will read and see if they take the same tack. For simplicity, imagine an industry dominated by two firms, A and B. At any given point, both are choosing between holding prices steady and raising them (assume lowering prices is off the table because it’s unprofitable, let’s keep it simple.) This sets up the classic game-theoretical model of the prisoner’s dilemma:
A Maintains Price | A Raises Price | |
B Maintains Price | →, → | ↓, ↑ |
B Raises Price | ↑, ↓ | ↑, ↑ |
In the chart above, the red arrows represent the change in A’s profit and the blue represent the change in B’s. If both hold the price steady, nothing changes, we’re at an equilibrium. If one and only one firm raises prices without the other, the price-hiker will lose money as price-conscious consumers switch to their competitor, who will now see higher profits. This makes the companies averse to raising prices on their own. But, if both raise their prices, both will be able to increase their profits. That’s why collusion happens. But, wait, isn’t that illegal? Yes, yes it is. But it is nigh on impossible to police implicit collusion, especially when there is a seemingly plausible alternative explanation for price hikes.
As James Galbraith wrote, in stable periods, firms prefer the safer equilibrium of holding prices relatively stable. As he explains:
In normal times, margins generally remain stable, because businesses value good customer relations and a predictable ratio of price to cost. But in disturbed and disrupted moments, increased margins are a hedge against cost uncertainties, and there develops a general climate of “get what you can, while you can.” The result is a dynamic of rising prices, rising costs, rising prices again — with wages always lagging behind.
And that gets us to opportunity, which is what the macroeconomic shocks provide. Firms probably did experience real increases in their production costs, which gives them good reason to raise their prices…to a point. But what has been documented by Groundwork Collaborative and separately by Isabella Weber and Evan Wasner is corporate executives openly discussing increasing returns using “pricing power,” which is code for charging more than is needed to offset their costs. This is them signaling that they see an opportunity to get to that second equilibrium in the chart above, where everyone makes more money. And since that same information and rationale is likely to be present at all of the firms in an industry, they all have the incentive (or greed if you prefer) to do the same. This is easiest to conceptualize in a sector with two firms, but it holds for one with more that is still concentrated. At some point, though, you reach a critical mass where suddenly there’s one or more firms who won’t go along with it. As the number of firms increases, it becomes more and more probable that one won’t just go along with it, which is why concentration facilitates coordination.
And that’s it. In an economy with significant levels of concentration — more than 75 percent industries in the American economy have become more concentrated since the 1990s — and the smokescreen of existing inflation, corporate pricing strategy can sustain rising prices due to the uncertainty. Now, if you ask twenty different supporters of sellers’ inflation, you’ll likely get twenty slightly different versions of the story. However, the main beats are mostly agreed upon: 1) firms are profit maximizing, 2) they always want to raise prices but usually won’t out of fear of either being undercut by the competition or being busted for illegal collusion, and 3) other inflationary pressures provide some level of plausible deniability which lowers the potential downside of price increases.
The evidence available to support theories of sellers’ inflation is one of the main points of contention between its proponents and detractors. Despite that, there is strong theoretical and empirical evidence that backs the theory up.
First is a basic issue of accounting that nobody in the traditional macro camp seems to have a good answer for. Profits are always equal to the difference between revenues (all the money a company brings in) and costs (all the money a company sends out).
Profits= Revenue – Costs
This is inviolable; that is simply the definition of profits. As I’ve written before, this means that the only two possible ways for a company to increase profits is by generating more revenue or cutting costs (or a combination of the two, but let’s keep it simple). Costs can’t be the primary driver in our case because we know they’re increasing, not decreasing. Inflationary pressures should still have increased production costs like labor and any kind of input that is imported. Companies also have been adamant about the fact that they are facing rising costs; that’s their whole justification for price hikes. And mainstream economists would agree. They blame lingering inflation on a wage price spiral, which says that workers demanding higher wages have driven cost increases that force companies to raise prices – resulting in higher inflation. As both sides agree that input costs are rising, the only possible explanation for increased profits is an increase in revenue. Revenue also has itself a handy little formula:
Revenue = Price * Units Sold
While the units sold may have increased, price was the bigger factor. We know this for at least two key reasons: because of evidence showing that output (the units sold) actually decreased and because of the evidence from earnings calls compiled by Groundwork. Executives said their strategy was to raise prices, not to sell more products. And there’s two very good reasons to believe the execs: (1) they know their firms better than anyone, and (2) they are legally required to tell the truth on those calls. (That second reason is also evidence of sellers’ inflation on its own; if the theory’s opponents don’t buy the explanation given by the executives to investors, they must think executives are committing securities fraud.)
In rebuttal to the accounting issue, Brian Albrech, chief economist at the International Center for Law and Economics, has argued that using accounting identities is wrongheaded:
Just as we never reason from a price change, we need to never reason from an accounting identity. My income equals my savings plus my consumption: I = S + C. But we would never say that if I spend more money, that will cause my income will rise.
This, on face, seems like a reasonable argument, except all it really shows is that Albrecht doesn’t understand basic math. Tracking just one part of the equation won’t automatically tell us what the others do…duh. But we can track what a variable is doing empirically and use that relationship to make sense of it. We would never say that someone spending more money on consumption causes their income to rise. But we certainly could say that if we observe an increase in personal consumption, then we can reason that either their income increased or their savings decreased. The mathematical definition holds, you just have to actually consider all of the variables. In fact, Albrecht agrees, but warns “Yes, the accounting identity must hold, and we need to keep track of that, but it tells us nothing about causation.” No, it tells us correlation. Which, by the way, is what econometrics and quantitative analyses tell us about as well.
The way you get to causation in economics is by tying theory and context to empirical correlations to explain those relationships. Albrecht’s case is just a very reductive view of the actual logic at play. He continues:
After all, any revenue PQ = Costs + Profits. So P = Costs/Q + Profits/Q. If inflation means that P goes up, it must be “caused” by costs or profits.
No, again. Stop it. This is like saying consumption causes income.
Once again, Albrecht is wrong here. This is like saying higher consumption will correspond to either higher income or lower savings. Additionally, there’s a key difference between the accounting identities for income and for profits: income is broken down into consumption and savings after you receive it, whereas costs and revenues must exist before profits. This makes causal inference in the latter much more reasonable; income is determined exogenously to that formula, but profits are endogenous to their accounting identity.
In addition to these observations, though, there is also various economic research that supports the idea of seller’s inflation. Some of the best empirical evidence comes from this report from the Federal Reserve Bank of Boston, this one from the Federal Reserve Bank of San Francisco, and this one from the International Monetary Fund.
Another key piece of evidence is a Bloomberg investigation that found that the biggest price increases came from the largest firms. If market power were not a factor, then prices should have been rising roughly proportionally across firms, regardless of their size. If anything, large firms’ economies of scale should have cut down on the need to hike prices. Especially because basic economic theory tells us that when demand increases, companies want to expand supply, which should have resulted in more products (especially from larger firms with more resources) and a corresponding drop in price increases. And yet, what we actually saw was a drop in production from major companies like Pepsi, who opted instead to increase profits by maintaining a shortfall in supply.
That said there’s plenty more, including this from the Kansas City Fed, this from Jacob Linger et al., this from French economists Malte Thie and Axelle Arquié, this from the European Central Bank, this one from the Roosevelt Institute, and more. The Bank of Canada has also endorsed the view. It seems unlikely that the Federal Reserve, European Central Bank and the Bank of Canada have all become bastions of activist economists unmoored from evidence. Perhaps it’s time those denying sellers’ inflation are labeled the ideologues.
Before we get into the substance of critiques against sellers’ inflation as a theory, there are a few miscellaneous issues with the framing its opponents often use. There is a tendency for arguments against sellers’ inflation to use loaded words or skewed phrasing to implicitly undermine the legitimacy of people who are spearheading the push for greater scrutiny of corporations as a part of managing inflation.
For instance, Eric Levitz says the debate sees “many mainstream economists against
heterodox progressives.” This phrasing suggests that the debate is between economists on the one hand and proponents of sellers’ inflation on the other. But that’s not true! There are both economists and non-economists on both sides of the issue. Weber is an economist, as are the researchers at the Boston and San Francisco Feds. And others, including James Galbraith, Paul Donovan, Hal Singer, and Groundwork’s Chris Becker and Rakeen Mabud are on board. Notably, Lael Brainard, the head of President Biden’s Council on Economic Advisors (and former Federal Reserve Vice Chair) recently endorsed the view.
Or take how Kevin Bryan, a professor of management at the University of Toronto described Isabella Weber as a “young researcher” who “has literally 0 pubs on inflation.” Weber is old enough to have two PhDs and tenure at UMass and–will you look at that–has written about inflation before! Presenting her as young sets the stage for making her seem inexperienced, which saying she has no publications doubles down on. But his claims are false. Weber wrote a paper with Evan Wasner specifically about sellers’ inflation. But even if we take Bryan’s point as true and ignore the very real work Weber has done on inflation and pricing, Weber still has significant experience with political economy, which helps to explain how institutional power is able to influence markets—exactly the type of thinking sellers’ inflation is based upon.
(And this is nothing compared to the abuse that Weber endured after an op-ed in The Guardian provoked a frenzy of insulting, condescending attacks from many professional economists. For more on that, see Zach Carter’s New Yorker profile of Weber and/or this Twitter thread that documents Noah Smith’s outlash at Weber.)
But even the semantics that don’t get into ad hominem territory are confusing. Here is a list of the topline concerns that Kevin Bryan raised:
Let’s just run through that list of concerns real quick:
All of this is to set up the next point in that Twitter thread, which is that “being an Iconoclast is not the same thing as being rigorous, or being right.” True, but dodging the debate by attacking the credibility of an idea’s advocates and taking issue with the method of dissemination are also not the same as being rigorous. Or as being right.
These are just a couple of examples, but opponents of this theory really lean into making it sound like its champions are inexperienced and don’t know what they’re talking about. Aside from being in bad faith, this also indicates a lack of confidence in comparing the contemporary story to that of sellers’ inflation.
With the semantics out of the way, it’s time to get into the meat of the case(s) against sellers’ inflation. There is no singular, unified case here, more of a constellation of related ideas.
The first line of defense against theories of sellers’ inflation is asserting that traditional macroeconomics is good and has solved our inflation problem. For example, Chris Conlon of NYU has credited rate hikes with inflation cooling. Conlon says “I for one am glad Powell and Biden admin followed boring US textbook ideas.” But there’s a problem with that: the contemporary economic story does not actually explain how rate hikes can cool inflation without a corresponding rise in unemployment.
The traditional story starts in the same place as the sellers’ inflation story: macroeconomic shocks create inflation. (Although the traditionalists prefer to emphasize fiscal stimulus as the primary shock, rather than supply chains. The evidence largely indicates that stimulus did have some inflationary effect, but not much. The global nature of inflation also undercuts the idea that American domestic fiscal policy could be the main explanation.) The shock(s) create a supply and demand mismatch, with too much money chasing too few available goods. After that, however, the traditional mechanism for explaining inflation remaining high is supposed to be a wage-price spiral.
The story goes something like this: the stimulus boosted consumer demand, which overheated the economy, and created more jobs than could be filled, meaning job seekers negotiated higher pay when they took positions. They then spent that extra money which increased demand further, leading to even higher prices as supply couldn’t keep up with demand. Workers saw that their cost of living went up, so they took the opportunity to demand better pay. Companies were forced to give in because they knew in a hot labor market, their workers could leave and earn more elsewhere if employers didn’t meet workers’ demands. Once their wages went up, those workers had more spending power, which they used to buy more things, further increasing demand. That elevated prices more, as the supply-demand mismatch increased. Now workers see their cost of living rising again, so they ask for another raise. If this pattern has held for a few rounds of pay negotiations, maybe workers ask for more than they otherwise would, trying to get out ahead of their spending power shrinking again. Rinse and repeat.
But we know that this story doesn’t describe the inflation that we saw over the last couple of years. Wage growth lagged behind inflation, which indicates that something else had to be driving price increases. Plus the Phillips curve, which is meant to illustrate this relationship between higher employment and higher inflation, has been broken in the US for years. It simply does not show a meaningful positive relationship any more.
It’s important that we understand this story as a whole. Levitz, in his piece, tries to separate the initial supply-demand mismatch from the wage-price spiral as a way of making the conventional model stack up better against sellers’ inflation. But that doesn’t actually hold because if you omit the wage-price spiral (which Levitz agrees seems dubious), the mainstream macro story has no mechanism for inflation staying high. If it were just a one-time stimulus, that would explain a one-time inflation spike, but once that money is all sent out (say by the end of 2021), there’s no source for further exacerbating the supply-demand mismatch (in say the end of 2022 or early 2023). (Remember, inflation is the rate of change of prices, so if prices spike and then stay the same afterwards, that plateau will reflect a higher price level but not sustained high inflation.)
Similarly, focusing on only the supply-side shocks provides no reason for why inflation remained elevated long after supply chain bottlenecks had cleared and shipping prices had fallen.
The incentive shift that occurs in concentrated markets is key to understanding this. In a competitive market, firms’ response to a surge in demand is to produce more. But, when the market is concentrated and some level of implicit coordination is possible, increased production is actually against a firm’s best interest, it will just put them back at that first equilibrium from earlier. They want to enjoy the high prices and hang out in the second equilibrium as long as they can
Sellers’ inflation, at least, has an internal mechanism that can explain how we got from one-off shocks to the economy to sustained inflation. Yet its opponents wrongly describe what that mechanism is. Remember the story from earlier: the motive of profit maximization, the means of market power in concentrated industries, and the opportunity of existing inflation. The most basic objection to this mechanism is to mischaracterize it as blaming sustained upward pressure on prices on an increase in the level of greed among corporations. That’s what economist Noah Smith did in a number of blogs that have aged quite poorly. But no one is seriously arguing companies are greedier, only that there is an innate level of greed, which conventional models also assume.
The strawmanning continues when we get to the means, which is what this Business Insider piece by Tevan Logan of Ohio State does by pointing out how Kingsford charcoal tried and failed to rent seek by raising prices, which just caused them to lose market share to retailers’ generic brands. Exactly! The competition in the charcoal market demonstrates why consolidation is a key ingredient in sellers’ inflation. If Kingsford had a product without so many generic substitutes, then consumers would not have had the chance to switch products. And that’s why a lot of the biggest price hikes occurred with goods like gas, meat, and eggs, all of which are controlled by cartel-esque oligopolies.
The opportunity component actually seems to be a point that there’s broad agreement on. For example, Conlon says that the “idea that firms might raise prices by more than their costs is neither surprising nor uncommon.” He goes on to suggest, however, that this is likely because firms expect costs to continue rising. There’s certainly an element of truth to that, but also consider the basic motivation of corporations: maximizing profits. As a result, if companies expect their costs to rise by, say, 5 percent over the next year and they’re going to adjust prices anyway, why not raise prices by 7 percent, more than enough to offset expected cost increases?
The theoretical case against sellers’ inflation is, as Eric Levitz noted, “deeply confused;” he was just wrong about which side was getting stumped.
The other side of the opposition to sellers’ inflation focuses on the empirics. To be fair, there’s certainly more work that needs to be done. But that’s about as far as the critique goes. The response is just “the data isn’t there.” I’ll refer you to Groundwork’s excellent work on executives saying that they are raising prices beyond costs, Weber’s paper, the Boston and San Francisco Fed papers, Bloomberg’s findings about larger firms charging higher prices, Linger et al.’s case study of concentration and price in rent increases, and the IMF working paper.
Setting aside the very real empirical evidence in support of seller’s inflation, the argument about a lack of empirics still gives no reason to default to the traditional model of inflation. Even if we accept a lack of data for sellers’ inflation, we have quite a lot of data that directly contradicts the mainstream story. Surely, something unproven is still preferable to something disproven.
Some economists like Olivier Blanchard have raised questions about methodology and the need for more work. Great! That’s what good discourse is all about; being skeptical of ideas is fine, as long as you don’t throw them out on gut instinct. Unfortunately, critics often simply reject the theory, rather than express skepticism. When they do, however, they often fall into the same methodological gaps in which they accuse “greedflation” proponents. For example, Chris Conlon egregiously conflating correlation and causation of the Fed’s monetary policy. Or Brian Albrecht taking issue with inductive logic while siding with a traditional story that makes up ever more convoluted, illusory concepts.
The traditional model of inflation is broken. The Phillips curve is no longer a useful tool for understanding inflation, a wage-price spiral flies in the face of reality, and there’s no viable alternative mechanism for sustained inflation within the demand-side model. Enter sellers’ inflation.
From the same starting point, and drawing on several cornerstone pieces of economic theory, sellers’ inflation is able to provide a consistent vehicle for one-off shocks to create prolonged upward pressure on price levels as firms exercise their market power. The bedrock ideas of the theory are consistent with seminal economic thought from the likes of David Ricardo and even Adam Smith himself and has the support of a number of subject matter experts. Is it a perfect theory? No, but to paraphrase President Biden, don’t compare it to the ideal, compare it to the alternative. More empirics would be preferable, but the case for sellers’ inflation remains much stronger than the case for a fiscal stimulus igniting a wage-price spiral, which is entirely anathema to most of the evidence we do have.
One way or another, inflation is trending down and, by some measures, is closing in on the target rate again. Many have rushed to credit the Federal Reserve for following the textbook course, but they don’t have any internal story about how the Fed could have done that without increasing employment. As Nobel laureate Paul Krugman (who supported rate hikes and once bashed the theory of sellers’ inflation) asked, “Where’s the rise in economic slack?” The conventional story is missing its second chapter and yet its advocates are eager to point to an ending they can’t explain as all the justification they need to avoid reconsidering their priors. One possibility Krugman notes, which Matthew Klein explicates here, is that inflation really was transitory the whole time. The sharp upward pressures were, indeed, caused by one-off shocks from the pandemic, supply chains, and Russian aggression, but the effects had unusually long tails. This theory aligns very well with sellers’ inflation; corporate price hikes could simply be the explanation for such long lasting effects.
Additionally, as Hal Singer pointed out, the recent drop in inflation corresponds to a downturn in corporate profits. Some, including Noah Smith (in that tweet’s comments), disagree and argue that both lower profits and less inflation are caused by new slack in demand. But that doesn’t really match what we’re seeing across macroeconomic data. True, employment growth has slowed, as has the growth of personal consumption, but that still doesn’t match up with the type of deflationary pressure that we were supposed to need; Larry Summers was citing figures as high as 6 percent unemployment. Plus, the metrics that do show demand softening largely only show that employment and consumption are steadying, not decreasing. On top of that, the contraction in output that The Wall Street Journal identified makes the case for simple shifts in demand driving price levels dubious. Additionally, if a wage-price spiral were at fault, leveling off employment growth would not be enough, the labor market would still be too tight (aka inflationary), hence why we’d need to increase unemployment.
Good economic theories always need more work to apply them to new situations and produce quality empirics. But pretending that sellers’ inflation is a wacky idea while the conventional macro story maps perfectly onto the economy of the past three years is thumbing your nose at the most complete story available, significant empirical evidence, and centuries of economic theory.
Dylan Gyauch-Lewis is Senior Researcher at the Revolving Door Project.