Notes Toward a Theory of Inflation

Remember the “Ha-ha money printer go brrrr”-meme¹ from back in the early days of the pandemic in 2020? The “Nooo!”-character on the left-hand side of the meme represents someone like mainstream economists and wall street commentators, the IMF, and the World Bank. The “ha-ha” guy operating the money-printing machine ostensibly represents the Federal Reserve, though you could make the case that it is actually the public themselves, who despite the objections of the first character enjoy knowing the empirical fact that printing money does not lead to inflation, even if they do not understand the mechanics of why this is so.

Twitter user @femalelandlords, March 9, 2020. URL: https://knowyourmeme.com/memes/money-printer-go-brrr#fn1 last accessed February 27, 2022 — Twitter user @femalelandlords, March 9, 2020.

Most people take it as given that printing more money causes inflation, and they can be forgiven for thinking so, as it has been hammered home to them by mainstream economics professors in every econ 101 course the world over for many decades. But the money printer meme wouldn’t have taken off like it did if there weren’t significant doubt that this formulation was true, at least within certain circles (primarily economists, financial professionals, and policymakers.) The CFA Institute, about as mainstream a source on financial planning as one could find, published a recent article myth-busting the idea that changes to aggregate measures of the money supply can predict where inflation may go:

Central bank money printing is largely irrelevant to money supply and inflation,…Given their typical mandate to create moderate inflation, the all-powerful central banks seem quite powerless.²

This suggests we need to throw out the old theories. Something about them is fundamentally wrong, yet they linger for lack of alternatives. We need to develop new ones.

But why were they wrong? It’s a question that may not be so important for meme-making, but is very important and perhaps central to economic planning. From 2020 to the present day, everyone from central bankers to supply chain managers to the general public has been confronted again and again with the question: what actually causes inflation? Inflation is typically thought of as an increase in the overall level of prices in the economy. Unfortunately, nobody seems to actually understand with certainty why such general and continuous price increases actually take place. If that sounds a bit scandalous, don’t take my word for it: former Federal Reserve governor Daniel Tarullo, a technical expert if there ever was one, made headlines in 2017 for saying that his central bank has no coherent theory of inflation:

The substantive point is that we do not, at present, have a theory of inflation dynamics that works sufficiently well to be of use for the business of real-time monetary policymaking.³

That’s somewhat distressing, given that the Fed’s reason for existing is “real-time monetary policymaking.” This is a little like finding out neither of the pilots on your flight know how to land the plane.

But the governor of the Fed’s language is imprecise. It’s not that economists have no theories of inflation; really, it’s that they have too many. What causes price rises? Maybe it’s the absolute number of dollars circulating. If not, maybe it’s“too many dollars chasing too few goods.” Maybe it’s got something to do with business costs, since when these go up a company will often (not always?) raise its prices. (But aren’t business costs prices? Whose prices, and why’d they go up?) A new theory recently published in the Financial Times, with which I align in a few important ways, claims to come from a Modern Monetary Theory (MMT) point of view and argues that the fundamental cause of inflation is “unfilled orders” (orders which came in for a particular product but couldn’t be produced or shipped due to some logistical or production issue) and its proper solution is “demand management” and replacing the “budget constraint” with an “inflation constraint” – though exactly which theories of inflation this vindicates and which it refutes are left rather ambiguous in the piece.⁴ This is in fact typical of how economists write about inflation across the board. They tend not to distinguish rigorously between the contending theories; rather, they will argue about inflation sporadically on a case-by-case basis – in a newspaper op-ed, say – and draw on elements in a grab-bag manner from various different theories of inflation at the same time, even if these seem to contradict each other. As a result, it’s not always clear what the main theories even are, much less how they relate to each other. No wonder the Fed throws its hands up.

Leftists need a theory of inflation at least as badly as economists at the Fed do. We need one in order to ameliorate any potential price increases which would stymy classic socialist programs, like socialized medicine, industrial planning, and speaking more to our present moment, a Green Transition.

Survey of Existing Theories of Inflation

Quantity Theory of Money

As we shall see, over the millennia the exact causes of inflation have dogged economists and philosophers, challenging them to develop logical arguments and empirical approximations explaining why a price may rise. Historically, there have been “price revolutions,” or secular, long-term changes to price-changes which result in an overall higher price level. Economists have often begun their theories by assessing these historical episodes. The 16th century price revolution, for example, is assumed, among many economic historians, to have given rise to the Quantity Theory of Money (QTM) – a theory which holds that the price level is ultimately a function of the amount of money in circulation. The QTM is often given by a famous equation MV = PQ, where M is the supply of all money, V is the “velocity” of money or the rate at which the money is being spent; P is the overall price level; and Q is the quantity of goods and services available to be purchased.⁵ There are many theoretical reactions to this equation, to the QTM, and its possible implications more broadly. These reactions can be broken down into three distinct categories.

First, some economists view QTM as essentially a hydraulic operation of the economy whereby, given an increase in the money supply (M), all else being equal, the price level (P) must rise in order for the equation to balance out. The remaining two terms V and Q are held constant as it is assumed they operate on a fundamentally longer timescale (for instance, that although the money supply and prices can change quickly and smoothly, the amount of output has hard limits or at least cannot change as rapidly). Others allow for movement in the other two terms of the equation (V and Q), to conclude that while M is certainly important for the price level, it is mediated by the velocity of money and the quantity of output (or goods and services) in the economy, in ways that the hydraulic interpretation does not accurately represent. Both of these views differ only in the degree of importance which they place on the money supply affecting the price level. In this sense both can be thought of as distinctively “monetarist” views of inflation.

Push & Pull Theories of Inflation

Other economists view the equation as merely a tautology showing that at any given time in an economy the amount of money actively being spent (M times V) is equal to the amount in money of the output being sold, and nothing more. A tautology is not an explanation, this argument goes, and there’s good reason to suppose inflation must rely on other factors. These economists almost all fall into the camp of non-monetarists, of which there are numerous schools of economic thought represented. Their explanations of inflation and its causes rely upon macroeconomic indicators of supply and demand interacting with one another in complex ways so as to produce the changes to the price levels we experience. Often, economists in this vein tend to talk about “cost-push” and “demand-pull” as possible explanations for inflation.

Demand-Pull Theories

Demand-pull inflation theories tend to argue that an increase in demand, especially relative to supply, will cause price increases across the economy for one reason or another. The classic QTM story is usually that any increase of the money supply will lead to a rise in prices, all other things being equal. A demand-pull story begins distinguishing itself from monetarism when it relates this story in some way to production. It’s not the increase in the money-supply as such, it’s that too much monetary demand relative to the supply of goods in the economy as a whole (“too much money chasing too few goods”) causes the price of goods to be “bid up.” What does bidding up mean in this context? Hard to say. It might be, for example, that more demand at the same level of supply raises prices due to the supposed price elasticity of demand. (In reality, most prices have no lawlike relationship to demand under normal circumstances.)

Other demand-pull stories tend to stick closer to reality and frame the difference in terms of too much spending relative to productive capacity. In other words, if demand increases to the point where too many orders are being placed for goods, but the economy’s ability to produce those goods remains the same, the result will be an “overheated” economy that cannot produce as much as is needed. This results in shortages that lead to price rises. Thus, demand must be “managed” in order to prevent inflation. This “aggregate demand management” was central to Keynes’s analysis of war planning in How to Pay For the War, and tends to be accepted by Post-Keynesian economists.

These phrases; “too much money chasing too few goods,” “bidding up prices,” and “an overheated economy” are ambiguous and can be read several different ways in terms of the actual cause of the price-increases. And to make matters worse, they also get mixed and matched despite sometimes being distinct stories.

Cost-Push Theories

Then there are the cost-push theories of inflation. Economist Marc Lavoie in his chapter on inflation in Post Keynesian Economics: New Foundations makes clear the Post Keynesians’ rejection of the quantity theory of money in the strongest of terms.⁶ He admirably summarizes the macro-level cost-push model they support, instead. Not only does the money supply not determine the level of prices, Lavoie writes, but neither does the growth in the money supply determine the rate of inflation or any such aggregate growth rate therein. Fair enough, as this much has been borne out empirically as discussed above.

Instead, the core theory of inflation presented by Lavoie is built upon economist Joan Robinson’s observations that ‘in our model, as in reality, the level of the money-wage rate obtain[ed] at any particular moment is an historical accident’,⁷ and J.M. Keynes’ prediction that ‘the long run stability or instability of prices will depend on the strength of the upward trend of the wage-unit (or, more precisely, of the cost-unit) compared with the rate of increase in the efficiency of the productive system’.⁸ In other words, the overall level of wages in the economy is the primary cost in the economy; the upward trend in wages ultimately determines costs of production for firms all across the economy; and if there is a strong upward trend in wages, those costs will push up prices, making wages the cost that pushes prices upwards.

Lavoie is quick to clarify that he does not believe inflation materializes solely or even mainly from wage increases, but rather that it is an important factor within a distributional conflict between workers and capitalists over wages and prices at the level of the whole economy. Workers who are relatively more organized, for example those in unions, will negotiate with capitalists for wages that match some indexed cost of living over a certain time period, while capitalists will respond either with a lower counteroffer, or by accepting the worker’s offer but gradually increasing their prices over time, ideally at a rate higher than the rate of wage increase they agreed to with the workers, so as to still achieve their desired markup over costs.

It is within this gradient between wage growth and price growth (which Lavoie calls the “aspiration gap”) that the dynamics of inflation play out. And the mechanism by which he says inflation could escape the confines of one firm, is by the diffusion of information concerning new deals struck between labor and capital in this manner into the wider economy. Using this information, other bargaining outfits representing workers in labor negotiations would be more likely to demand higher wages and potentially trigger capitalists to set higher prices on a sustained, economy-wide level. The potential for a wage-price spiral to develop through this diffusion of information about wage growth or price increases, at the economy-wide level, constitutes the basis for potential inflation, according to Lavoie. The rest of his chapter focuses on introducing the potential for commodity price changes into the model (something neither the capitalists nor the workers have control over in his model). Lavoie ends his chapter with an expansion of the model to include a sort of residual component containing all the supply-side factors outside of the central wage-price dynamic, such as higher commodity prices, higher prices of imported goods from abroad, a slowdown of productivity growth, increases in sales taxes, etc., and constructs it so that it is still conditional upon the original “aspiration gap” between desired wages and desired prices. His interest in these other supply-side factors, however, suggests that other forces besides the overall wage level might be at play in cost-push dynamics.

Hybrid Cost-Push / Demand-Pull Theories

And then there are those who mix and match both cost-push and demand-pull stories of inflation, without much of an attempt to stitch them together into a single theory with a unifying, driving mechanism. A good example of this position are the proponents of MMT. In the chapter on inflation in her book The Deficit Myth, economist Stephanie Kelton gives an overview of both the demand-pull and the cost-push theories, but then jumps to a list of heuristic tools for what to do to ameliorate inflation should it happen, including increased taxes to offset excessive spending, a federal job guarantee to set a wage floor and hopefully promote productivity, having the Congressional Budget Office evaluate any spending proposal on the basis of an inflation constraint rather than a budget constraint. She does not present an MMT theory of what causes inflation other than to say it is some combination of the two existing types of theories of cost-push and demand-pull. The upshot for MMT economists like Kelton is intended as a practical one: there are real limits to spending, as opposed to the fake limit of the budget constraint (biophysical ones, like the number of workers, the materials, the physical processes and technology available) that dictate whether or not a particular spending proposal could cause inflation to build up. Insofar as the budget deficits and national debt itself plays no operational role on the government’s ability to spend and hence doesn’t affect companies in the US’s ability or willingness to set prices higher, I agree with Kelton and the MMTers here. But when it comes time to shift the conversation from a set of heuristics to use for spotting inflation when it happens (or could happen) to the root causes of inflation, I find their explanations to be incomplete.

However, exactly how inflation happens gets muddled in these treatments of the concept, and is subordinated to the benefits of particular fiscal policies. Sometimes it is due to “bottlenecks” in the real economy. Other times, it is due to “too much money chasing too few goods”. Often in Kelton’s book the two concepts get mixed together in curious ways, for example, citing bottlenecks in the context of an ostensibly demand-pull oriented inflation:

Rising demand can begin to put pressure on prices, and bottlenecks can develop in industries that are experiencing the greatest strain on capacity. Inflation can heat up. Once the economy hits this full employment wall, any additional spending (not just government spending) will be inflationary. That’s overspending, and it can even happen if the government’s budget is balanced or in surplus.⁹

No reference is made to prices being set by anyone, or how price-increases propagate through the economy, only to strained capacity somehow leading to inflation “heating up”everywhere.She concludes by arguing that if this process of capacity-hitting-a-wall, so to speak, continued up to full employment, then any spending after that point would be overspending and therefore inflationary. But is the spending putting pressure on prices? Or, is it the increased costs of production, from the bottlenecks? While some of this sounds quite intuitive, it lacks specific theoretical underpinnings that would ultimately form the basis for the inflation-fighting toolbox the MMTers admirably seek out.

Elsewhere in MMT literature, the same slippage between demand-pull and cost-push verbiage can be found. Take for example the textbook Macroeconomics, written by economists Bill Mitchell, L. Randall Wray, and Martin Watts. In a ten page sequence, they go one-by-one through the theories: demand-pull inflation of the sort described by Keynes in How to Pay for the War, and cost-push inflation of the sort described by Lavoie (both of the Robinson-style “wage barrier” and the Kalecki-style “distributional conflict” varieties). They even mention certain heterodox microeconomists about markup pricing, which as we’ll see are centrally important to my own inflation theory.¹⁰ Despite suggesting that the authors have created a unified MMT theory of inflation, it is unclear precisely how these separate theories fit together at the level of concrete mechanisms in the economy, and the prose in places where such a connection is suggested is opaque.

For example, they claim cost-push inflation is actually conditional on there first being a certain degree of demand-pull inflation, while at the same time admitting that the two are difficult to tell apart in the first place:

Cost push inflation requires certain aggregate demand conditions for it to be sustained. In this regard, it is hard to differentiate between an inflationary process which was initiated from supply side [cost-push] pressures from one that was initiated by demand side [demand-pull] pressures.

This is immediately followed up by a concrete example of a situation potentially leading to inflation; a crucial raw material is suddenly scarce or becomes more expensive. The example is supposed to show how the theories work together to produce an explanation, but in fact the result is pretty confusing:

For example, an imported raw material shock means that a nation’s real income that is available for distribution to domestic claimants is lower. This will not be inflationary unless it triggers an ongoing distributional conflict as domestic claimants (workers and capital) try to pass on the real loss to each other. However, that conflict needs ‘oxygen’ in the form of ongoing economic activity in sectors where the [wage-price] spiral is robust. In that sense, the conditions that will lead to an accelerating inflation – high levels of economic activity – will also sustain an inflationary spiral emanating from the demand side. ¹¹

There are three separate stories here about inflation that have been mashed together, two cost-push and one demand-pull. The initial raw materials shock is a kind of cost-push story, though the reason why the cost of this particular material shooting up should impact the rest of the economy is left vague – the authors cite the blow it strikes to “the nation’s real income,” which presumably means the amount of goods and services you can buy with the same amount of money decreased as a result of this “shock.” This decrease in “real income” apparently touches off a distributional conflict between workers and capitalists over the shrunken pool of the nation’s output, with the workers demanding higher wages and and the capitalists demanding higher mark-ups. That distributional conflict is a second cost-push story, since for economists like Lavoie it is a source of inflation via the wage-price or price-wage spiral. Nevermind that such distributional conflicts necessarily take place within a single firm, and yet to be inflationary they must spread across the whole economy; or, that there must be something left unspecified tying the specific raw material that got “shocked” to the specific workplaces engaging in distributional conflict. This is because, it turns out, this distributional conflict isn’t inflationary in itself; it’s only inflationary when the conflict gets “oxygen” from “high levels of economic activity,” i.e. when it’s prolonged or deepened by demand-pull. The authors, therefore, explain this scenario by subordinating cost-push causes to demand-pull ones while asserting that they can all happen together, yet it’s only when demand-pull exists, it seems, that you get actual inflation.

Does this actually explain anything? It’s clear enough that demand-pull prevails in the authors’ explanation; and a supply shock eventually leading to inflation sounds like something that can happen in the real world, such as in the 70s oil shock or even the chips shortage of 2020-present. Why is “oxygen” from demand-pull necessary for it to be “really” inflation? And what precisely is “oxygen” a metaphor for? If it’s a story about demand for specific output exceeding productive capacity, what effect exactly do the material shock and the distributional conflict have on those two factors? Above all, why and in response to what are people raising their prices, and how precisely does this spread to the whole economy?

To make matters worse, the way they define inflation in the first place seems arbitrary:

Inflation is the continuous rise in the price level, so the price level has to be rising for a number of time periods. A one-off price rise is not an inflationary episode.

A “continuous rise” can be interpreted as a string of consecutive “one-off” rises. In that case, it would be good to know what a “one-off” rise is, anyway? Is it one month’s worth of rising prices? Two months? More? And by how much do they need to rise during such rises? Wouldn’t it be more appropriate for them to have defined inflation in terms of the underlying dynamics of the conflict theory of inflation they cite from Kalecki? The authors seem to anticipate this problem by invoking the concept of a “normal price level”, which is defined as:

…the prices that firms are willing to charge when they are operating at normal capacity and earning a profit rate that satisfies their strategic aspirations.

We now need to know what “normal capacity”and “strategic aspirations” are. Unfortunately, neither term is defined anywhere in the entire textbook. But looking elsewhere, one finds normal capacity to be a cost accounting term defined as,

…the amount of production volume that can be reasonably expected over the long term. Normal capacity takes into account the downtime associated with periodic maintenance activities, crewing problems, and so forth. When budgeting for the amount of production that can be attained, normal capacity should be used, rather than the theoretical capacity level, since the probability of attaining normal capacity is quite high. The normal capacity level can decline over time as production equipment ages, since the equipment requires more maintenance effort.¹²

If we use this definition of normal capacity and plug it into the authors’ definition of inflation, then it appears the demarcation line between “one-off” and “continuous” becomes slightly clearer, though not by much. By “continuous”, perhaps they mean a period of time where in no single accounting period was there a normal price level, meaning, at no time in an inflationary episode would you expect to see firms budgeting out normal capacity in light of their strategic aspirations. In other words, a one-off inflation might be better thought of as one accounting period of elevated (non-normal) pricing, separated on either side by accounting periods where normal capacity was both planned for and experienced.

But we do not get this level of granularity or precision for the definition of an inflationary episode in the textbook. Instead, it is a quite ambiguous and arbitrary one that strikes me as reminiscent of the inflationary episode of 2020-present, whereby MMT economists have been able to discount the price rises that have lasted for over a year as “temporary” or “transitory” without stating at what point they would become “continuous”. Furthermore and interestingly, all this discussion of capacity and what a normal amount of it is, seems to imply the primacy of rising costs for inflation. However we are robbed of any real conclusion to this effect, due to the aforementioned language mixing between the cost accounting of normal price levels, and demand-pull being the “oxygen” that causes the economy to “…sustain an inflationary spiral emanating from the demand side.” If demand-pull only matters because it creates cost increases, but cost-push only matters when there’s demand-pull, then what actually causes prices to go up, and when? And what precisely do markups, raw material costs, and distributional conflicts have to do with all that?

The Supply Chain Theory of Inflation

These are the theories that we’ve inherited from the past. However, in light of recent advances in heterodox microeconomics, new theories are possible. I think such a theory could be derived from those economists who have developed observation-based accounts of firm behavior, especially their pricing methods. Economists like Frederic S. Lee fall into this camp. Lee emphasized the cost structure of individual firms and of supply chains in his explanations of price changes, and the interdependence of those firms with supply chains. Although Lee never completed a theory of inflation, one flows directly from his work: a cost increase that is felt along sufficiently widespread supply chains can contain or unleash price increases in the wider economy. This view is fundamentally a microeconomic one and in that sense is completely distinct from the other two which depend upon macroeconomic factors to explain inflation, though in other respects there is common ground. It would allow us to eschew once and for all macroeconomic explanations of inflation (including those predicated on acceptance of the equation of exchange and QTM) altogether, in favor of a microeconomic one.¹³

And there’s good reason to believe this is the way to go. Contemporary theoretical discussions of inflation are taking off in interesting directions, perhaps due to the interesting times in which we live: a once-in-a-century pandemic may well be the cause of yet another of these price revolutions, in which case maybe a new theory of inflation is in order. However, to get there, it seems logical that one must first formulate a theory of price, since in the end this is what we are ultimately discussing when we talk about inflation. As economist J.W. Mason recently remarked on his website:

Inflation is just an increase in prices, so for every theory of price setting there’s a corresponding theory of inflation.

If inflation theory is downstream of price theory, then no account of inflation can begin with the macro-economy at all since prices are set at the micro-level. Rather, you need to look at particular industrial sectors, their supply chains, and ultimately the pricing decisions of their firms. Only then are the true causes of inflation (both the internal failures of the industrial system and external shocks to it which can cause price rises) revealed.¹⁴ What follows will be my attempt to build up such a theory from Leesian principles.

A note before we begin…

I understand the impulse to construct a purely logical argument for inflation can lead to all manner of oversimplifications based on bad assumptions. This is why in the ensuing paragraphs, rather than doing that, I will adopt an approach called “grounded theory” wherein I shall endeavor to build up each step of the process of inflation using assumptions grounded in historical evidence collected by mostly heterodox economists who have seriously investigated this question of what causes inflation. I do this with a sense of humility – if I’m wrong, I’m wrong. However, what I hope to show here is that, properly grounded in historical evidence, one can build up a powerful understanding of what a price is, why they increase when they do, and hence what inflation is.

The Argument

Just what is a price? Suppose something is priced at $12. What does that mean? By itself, a number is just a number. The $ symbol gives that number a specific meaning – it’s how much of that kind of money is required. What is money? Fundamentally, it’s an accounting system used in economic planning – or, to use a technical term for this, a unit of account. Most units of account today correspond to the currency of a nation (such as dollars, euros, or yen). A price is thus a number in a unit of account, and a unit of account is the demarcation of numbers of money. Although a particular money itself gets issued only by designated issuers (the US is the sole issuer of the US dollar, for example), anyone can in practice denominate their price using US dollars (i.e., use the dollar as their unit of account) if they wish to. Which leads me to my next point.

What does a unit of account do, and what does it mean to have a certain number of monetary units? In one sense, you could take a metrological approach.¹⁵ Inches can be used to measure distance – and so can centimeters. Different measuring systems can be used to give measurements to the same phenomenon. This analogy tells you something very important about units of account. They’re socially constructed, and the units themselves are totally arbitrary. What’s important about them is that they form a system that can relate one price – one number, one mark along the measuring stick – to another.

But – and this is a crucial point – there is an important difference between scientific measurement systems and accounting systems. The measuring stick is measuring something – distance exists outside of you, and you use the stick to take a measure of it. Prices have no meaning outside the unit of account which makes them possible. That’s especially clear from the point of view of a price-setter. It’s better to think of prices not as a system of measure but as a system of counting – an accounting system! – where you can count up as much as you want. You’re not taking the measure of any one thing outside yourself; you’re deciding how much you need to count up in order to accomplish your strategic objectives. That’s what it means to set a price – to create a number in the unit of account. Usually, as we’ll see, this is something you’ll do while keeping in mind other prices set by other people in the same system – how much they’ve counted up, and the various consequences that might have for you depending on how you relate to them. For practical purposes, this number corresponds to how much money one needs to make a payment – how much something is worth. But what it’s worth is not an inherent quality we discovered within the thing itself; rather, it’s something we imposed upon it by setting a particular price.

So who sets prices, and why?

People set prices – not “the market”. Businesses employ people whose main responsibility is pricing, when the business is large enough for such division of labor. Even in small businesses the owner or a general manager still set aside time specifically towards evaluating their current prices and determining if any changes should be made. The people whose job it is to set prices do so by looking at the prices their own business has to pay to its suppliers in order to have all the inputs it needs to provide their goods or services to the customers. From the point of view of a business, these input prices are referred to as its costs. The business’s price is typically defined by these pricing managers as some additional amount (called a “markup”) over the sum of the costs that the business incurs in the course of production. In the context of an industrial firm, these costs are generally attributable to labor and to the physical inputs that the business purchases from its suppliers. Within this context of industrial production, there exist numerous flavors of pricing procedures, but many of them can be summarized by this general cost-plus-markup algorithm.

Economists Adolph Berle and Gardiner Means first theorized this phenomenon in their classic 1932 work The Modern Corporation and Private Property.¹⁶ In it, they use the empirical evidence of the cost-plus-markup procedure as evidence for what they call “administered prices”, or prices that are set not by the interactions of larger macroeconomic forces of supply and demand, but rather consciously through the decisions of pricing managers at the firm level. Berle and Means’s theory of administered prices has been confirmed empirically many times through surveys of the pricing managers themselves.¹⁷

In fact, Lee rather convincingly argues that this is not the superpower of large-scale corporations or oligopolistic firms, but that potentially all prices are administered and remain stable with respect to changes in demand. Thus, in his memorable phrasing, “There is no functional lawlike relationship” between supply-and-demand, and prices.

If a business becomes powerful enough within its industry, it may well begin coordinating the prices of its competitors through a process known as “price leadership”. This occurs when smaller companies simply look to see what the pricing decisions of the power company is, and then either copy it entirely, or otherwise closely follow them. It may sound like the smaller companies are completely at the mercy of the powerful price leader, but this is not always the case. Smaller businesses could follow the price leader, or they could follow different competitive strategies (Porter, “Competitive Strategy: Techniques for…”) Cost leadership, differentiation , and market segmentation. Pricing strategies are typically downstream from the more grand strategy of the business — prices aren’t everything, but they are clearly quite important.¹⁸ Frederic S. Lee, also situates what he calls “strategic prices” within a broader business strategy. In addition to price leadership, there are other more direct forms of coordination in price-setting, such as through collaboratively setting a price through trade associations, business conferences, and cartels. Or, businesses may have prices set for them by government regulation. Finally, prices may be set by a combination of these factors. For example, a group of businesses could collaboratively set a price at a trade association meeting, and then other smaller businesses in the same industry who are not members of the association (or who at least were not invited to the meeting) end up using the price that was set by this collective price leader.¹⁹

Remember the surveys of pricing managers mentioned earlier? One answer not found anywhere by the economists conducting the surveys on the question of what factors go into a decision to raise a price was what any of the monetary aggregates happen to be at the time, or where the pricing managers thought they may end up at in the future. The money supply quite literally does not factor into pricing procedures themselves. This is a major problem for any of the theories of inflation that rest upon the idea of the quantity theory of money being operationally part of pricing.

This is a really important point. If surveys of real-world price setters indicate that they simply don’t take the money supply into consideration, or even know what it is, then by definition that means the money supply has at best a tenuous connection to prices in the real economy. Whether you have an economy where the total money supply is $100 billion or $100 trillion, an economy of one dollar or infinite dollars, these surveys suggest that the money supply will only affect the behavior of price-setters indirectly at most, through its effects on something they actually care about – their costs, their revenues, etc. And for that to be the case, you’d need to have a map of the causal mechanisms that takes you from a change in the money supply to a change in prices. Most theories inspired by the Quantity Theory simply do not.

What’s most important here is not simply to present a complete logical model, as mainstream economists sometimes think they have done, but rather to take an arguably more scientific approach. First, I want to present empirical evidence of pricing procedures. I want my model to be built upon a bare minimum of assumptions, and for these to be grounded in empirical observations of actually-existing price behavior. In other words, to the greatest extent possible, I want to use real-world mechanisms to explain real-world phenomena.

Now that we have a definition for what a price actually is, we can move onto the next question: what is inflation?

Most economics textbooks define inflation as “a general increase in the price level” where “price level” is, in principle, all the prices across the entire economy. In practice however, price level is a stand-in term to describe some critical portion of all prices currently on offer, usually as an index of a set of prices deemed representative of the larger economy. If enough of the prices in this socially constructed index go up and stay up for long enough of a time, then definitionally according to economists, you’ve got inflation. Such macroeconomic inflation indexes are numerous and increasingly more bespoke in their construction the further you go down that rabbit hole, but the general idea is you track the relative price movements of a basket of goods and services deemed representative in some sense of economic life for the average participant.²⁰ These indexes are what you see mainstream economists refer to when they discuss the topic on television.

It is a very macroeconomic presentation of the concept, though. Who is setting these prices? We have to remember that it is pricing managers and business owners reviewing their costs (the prices they’ve paid to their suppliers), and assigning a markup over and above those costs for each unit of goods or service on offer.

Businesses exist within what are called supply chains: systems of businesses, organizations, and other institutions producing inputs which are used by other members of the system to produce outputs. Together, the supply chain aims to, at the very least, socially reproduce itself as it was at the beginning of the production period – ideally generating a surplus that could be used to further expand production, or be retained in case of a rainy day. Supply chains have varying degrees of coordination — some are consciously planned by firms down to each individual piece of product; others are not planned at all in the aggregate, with planning only really taking place on the level of the individual firm. But no matter what, production across the economy as a whole is coordinated by supply chains, insofar as flows of orders up supply chains determine what and how much everybody produces.

The prices at which each business in the supply chain sells its goods must generally cover all their own costs plus the markup. Businesses often enter into long term supply agreements with other firms along their supply chain, so as to smooth out their costs and make them more predictable. This is done to reduce uncertainty of supply – if you become unsure of whether you’ll have enough inputs into your own production process to make good on customer orders for your product, this may lead to increases in costs.

Business managers have to plan for contingencies in case of shortages of inputs. You can’t plan for everything – such as so-called “act of god” events like a major hurricane, or an important supplier unexpectedly going out of business – but you can still plan for more reasonable risks, like if one of your suppliers faces a shortage of their own, which means you get your inputs more slowly, or (drumroll please) you are forced to accept a higher price for those inputs.

Why would your supplier do this to you? Aren’t you guys friends? Well you are, but in this case they had their own unforeseen supply issues, and much like you, they need to at least meet their own costs in order to continue as a going concern. But you do, too, so therefore it would be prudent as part of your overall business strategy to account for this possibility by either stockpiling enough of the inventory you may need should such a shortage occur, or by saving more money to ride it out without the benefit of revenue from those missed orders you would likely realize in this scenario.

There is an alternative option: you could raise your own prices in response to your supplier raising theirs. This could work, if the orders from your customers continue to come in high enough quantities to generate sufficient revenue to fund your operations. But there’s always the chance they might not. So there is actually a quite delicate balance to be struck between absorbing the cost increases of your suppliers in these rare situations, and passing them onto your customers – in the business literature, this quality of retaining customers by not increasing prices even in the face of rising costs is referred to simply as “goodwill”. In fact, it has been confirmed empirically a number of times that businesses are rather reluctant to raise their prices out of these and other related concerns.²¹ There is quite a good deal of price stability observed in supply chains, all else being equal – it is just predicated on the basic survival constraint of supply chains continuing to be met.²²

The supply chain theory of inflation is essentially that widespread price increases result from a sudden, unexpected cost increase across a supply chain that none of its participant firms can adjust to without raising prices.

We’ve established from the economic survey literature presented above that firms are strongly biased towards price stability – they don’t want to raise them if they don’t have to, and would much prefer to increase output rather than price in order to take advantage of revenue opportunities. And if they do raise prices, it is typically only gradually over the course of many years. An important question for inflation theory, then, is what causes whoever is the first one to raise prices to do so?

The original price-hike, or progenitor price increase, if you will, is the logical starting point for any potential inflation to develop. We know that firms use rules of thumb for their pricing in the face of uncertainty – if events outside of their control upend production such that there are notable shortages of necessary inputs for a firm, they will first coordinate with their suppliers, utilizing every option other than a price increase to try and make up for lost revenue. Firms will choose to absorb losses from missed or canceled orders without raising their prices for a time, but they won’t put up with this forever. The longer the disrupted supply conditions last; the more likely it is that the firm will choose to eventually raise its prices, thus becoming the first firm to do so in a supply chain. If that price increase continues unabated, then this means a prolonged increase in costs for their customers. And as a result, those customer firms may well join in on the price-hike, triggering cost and therefore price increases down the supply chain, which constitutes inflation.

If inflation does indeed materialize after the initial price-hike, in general, it takes on one of two forms. First, the most common cause of inflation is a series of price increases by individual firms made in response to biophysical shortages leading to cost increases of some key input into their production processes – for example, shortages or input price increases due to a drought or a natural disaster. Recently in 2021, the semiconductor chip shortage resulting from Covid-19 related disruptions of the global supply chain network have led to price increases for myriad products who share the chips as a common input into their production processes, in addition to labor disruptions from commercial lockdowns, quarantines, and social distancing distancing protocols put in place to control the spread of the Covid-19 virus during the global pandemic.

Second, countries could experience balance of payments crises affecting firms at the micro level, wherein suppliers getting their inputs from abroad suddenly find those inputs to be more expensive, or their forex revenues to be diminished due to unfavorable exchange rate fluctuations that increase their total costs.²³

These two forces – biophysical shortages and balance of payments fluctuations – and the cost increases they cause to ripple across supply chains are the primary cause of the widespread price increases we call inflation. The observant reader may have noticed that these phase changes I have described have certain echoes with the theories of inflation we examined before. Shortages and balance of payments crises, after all, sound like cost-push scenarios. But there’s an important difference: where macro cost-push stories say that some single cost factor, such as wages, goes up and raises prices across the economy all at once, my theory sees cost increases rippling down specific supply chains. It only creates a general price rise when affected supply chains are widespread.

An Aside About “Demand Pull”

This raises the question – if the supply chain theory has a micro version of “cost-push,” is there a micro version of “demand-pull?” What are some specific situations where increased orders can exacerbate pre-existing cost increases?

A particularly fascinating and complex situation where order flow can exacerbate costs is referred to in the supply chain management literature as the “bullwhip effect.”²⁴ If there is a sudden and unexpected surge in customer orders, dramatic, and expected to continue indefinitely (or even if there’s just uncertainty as to whether they will), firms may well choose to put in large orders that reverberate like a bullwhip up the supply chain, getting bigger with each step When it happens it is a major cause for concern, as the aberrant increases to order volume heading upstream on the supply chain ultimately has little if anything to do with the end-customer demand, and the ensuing cost increases to the businesses will likely outstrip the revenue they as a group take in. It can lead to a situation where one or more businesses caught up in this phenomenon can vastly overestimate the orders they will receive and wind up holding a ton of unsold inventory. Fortunately, the bullwhip effect is well enough understood by supply chain professionals today that they are often quick to identify scenarios where it could play out, and coordinate among themselves to avoid it. Still, it does happen.

The bullwhip effect usually doesn’t lead to price increases. But can it? If the cost increases it causes persist, there may be significant pressure on the businesses to increase prices.²⁵ If this is the case, then perhaps the bullwhip effect accounts for what the economists call demand-pull inflation. But even if so, this means it has very different underlying causes. It is not the result of printing too much money so that people have too much money in their pockets and are driving up prices with their excess consumption, but rather a sudden shock of increased demand that exceeds the productive capacity of firms to meet it over a long enough period of time to whittle down firm’s choices of how to respond to the increased orders other than to raise prices. Furthermore, the pressure that firms are under in a bullwhip effect scenario comes from their increased costs and lowered revenue.

Thus you could perhaps say demand does affect prices, sometimes, in some situations, but only insofar as it affects costs, since inflation is cost increases leading to price increases across supply chains – but even this is wrong and bad language. Really from the perspective of a supply chain manager, “demand” is just order flows across specific supply chains, not a macroeconomic phenomenon. What those increased order flows boil down to for the firm and its workers is a situation where people in specific workplaces are being asked to do what can’t be done. If quantity cannot be increased, then eventually price will be increased in order to generate an amount of revenue that socially reproduces the firm based upon what orders they can fill. But these hypothetical “demand-pull” effects, to the extent they exist at all, are weaker and secondary; if they exist, they’re really just cost-push with extra steps; and they operate within specific supply chains, only affecting the economy as a whole to the extent that supply chain does. And both cost-push and demand-pull models are macro abstractions that miss what’s really important, which is microeconomic pricing decisions within firms embedded in supply chains.²⁶ People are setting prices, and any theory of inflation must explain why those agents set the prices they do. Inflation is a phenomenon not of money, but of prices; and prices are always administered by someone.

Conclusion

There is currently no accepted coherent theory of inflation. The people whose job it is to prevent inflation – central bankers and government economists – draw from a number of competing theories in an ad hoc manner, even when those theories contradict each other.

There are different theories of inflation which, as we have seen, miss important details in different ways, and are at best only partially realistic. Many of these theories rely upon the Quantity Theory of Money (QTM), which end up mostly being wrong. And even those that don’t rely on QTM tend to focus solely on inflation through the macroeconomic lens, which is too distant a vantage point because price-setting is a microeconomic phenomenon. You have to look at specific pricing procedures – who sets what prices and why – in order to explain price rises across the economy. And those pricing procedures take place along particular supply chains – meaning inflation is primarily not an economy-wide phenomenon, but more likely takes place in specific sectors and industries within the economy.

Frederic S. Lee’s effort to establish micro-level cost-based pricing procedures occurring along supply chains as the basis of industrial price phenomena generally, has proven critical to my own understanding of the topic of inflation, and here I have attempted to build upon Lee’s price theory. Lee’s theory of prices could have been adapted into a theory of inflation, however he sadly passed away in 2014 before any such plan could materialize. Still, there are hints that this indeed was the plan in the work of his students.²⁷ I have attempted to produce a more general theory of inflation built upon Leesian work on pricing procedures and price dynamics and have compared what the world looks like in terms of my theory to what the world looks like in other leading theories of inflation to see which appears to be least wrong. For example, the so-called “demand-pull” focused theories of inflation emanating from post-Keynesian authors may well be better understood as cost-push theories in disguise (“cost-push with extra steps”). But perhaps more importantly, I hope this essay can at least serve as a framework for further debates on inflation grounded first in historical evidence. If I have made any fundamentally flawed assumptions of my own, then let them be pointed out only with reference to the historical record on firm pricing behavior.

As powerful as this augmented Leesian framework could potentially be, it still has some significant gaps to be addressed in future essays. You may be asking yourself, what about asset prices like stocks, bonds, derivatives, or real estate (or cryptocurrencies for that matter)? This is an important question, and one that clearly must be answered for completeness. It is important for any proposed theory of inflation to stipulate how asset prices are set, and integrate it into the larger comprehensive account of inflation. Unfortunately, Lee and the post-Keynesians at best only hint at an answer.²⁸ However, what if it were the case that, say, those who trade stocks for a living were also following pricing procedures?²⁹ It’s not so far-fetched, as any undergraduate in finance or an MBA student will tell you their textbooks do run through seminal pricing formulae for evaluating a stock based on a set of assumptions regarding many different criteria. In fact, there is an entire industry of analysts devoted to using quantitative methods to obtain what they view to be the fair prices for individual components of all different asset classes. Think Morningstar, the investment research company famous for its pricing model that incorporates the concept of “economic moat” or, the amount of existing and potential revenue opportunities that a particular company enjoys that keeps it safe so to speak from competitors and offers relatively easy growth.³⁰^,³¹ Or Standard & Poors ratings system for corporate and sovereign bonds.³² Could these institutions and their pricing methods be simply another version of the same pricing behavior of the industrial firms discussed earlier? Or, could they be distinct from industrial pricing procedures in ways that have little if anything to do with the supply and demand framework of neoclassical economics?³³ More research is needed to be certain.

People set prices, and by extension cause inflation through their price-setting behavior at the microeconomic level. Prices are very often determined as markups over the cost of production for firms, and there is an observed bias to stability along supply chains, with a degree of uncertainty for input availability that necessitates such rule of thumb-style planning. As noted, supply chains can periodically swing into inflationary spirals of their own making, as with the bullwhip effect. Furthermore, when there is an exogenous shock to the system that leads to price increases early in key supply chains, such as a natural disaster or a balance of payments crisis, firms downstream in those supply chains are more likely to raise prices in turn the longer the shock lasts. Therefore, it seems no market price exists outside of this context of human price-setting at the firm and industry levels, and we have seen evidence of this fact now in numerous firm pricing behavioral surveys. What we call inflation is nothing but a continual increase in prices, for however long and for whatever reason, in enough sectors of the economy for it to significantly inhibit the ability of households and firms to get what they need, given their current wages or revenues.

Whether it is an individual manager at a small business, a pricing strategy team at a large corporation, or a team of programmers building an algorithmic pricing procedure, in all cases people are applying their knowledge of specific industries, concerning both their business partners and their competitors, to decide on rules of thumb for what their prices ought to be set to, and doing their best to stick to those procedures in the presence of radical uncertainty. It follows from this that, in my view, any coherent theorizing on inflation must be proposed and debated within this realm of grounded theoretical work. When properly historicized, the study of inflation through the study of pricing procedures gives leftists a clarity they are sorely lacking as to the real causes of continuous price increases that are sufficiently long and severe as to throw the economic system into crisis. It gives us a better sense of the true limits to our desired policies and our true options under the limitations we face; under the pressure of the real world, as we try to change it.~

Cryptid

Fixed Capital

Socks

Insomnia, Four Voices

Can Fascism Think?

Our Neighbor From Hell