Theories of inflation: part one – the mainstream

by Michael Roberts

Guglielmo Carchedi and I have been working on a theory of inflation for several years.  We completed a detailed paper some time ago which was accepted for publication by a Marxist journal.  However, publication takes forever, so I thought I would provide a shortened version of our work on my blog.

For the blog, I have divided the paper into various parts: 1) mainstream theories of inflation; 2) heterodox and other Marxist theories; 3) our theory, called a ‘value theory of inflation’; and finally 4) a discussion of the application of our theory to the recent spike in inflation since the paper was completed.

Let me start with part one: mainstream theories.  

What causes inflation has been a puzzle for mainstream economics. As Walter Munchau in the Financial Times put it in 2020: “central bankers do not really understand how inflation works. There are lots of theories and approaches, theoretical and statistical, but none that has been able to explain persistently what is going on in the real world.” Charles Goodhart of the London School of Economics was harsher in 2021: “The world at the moment is in a really a rather extraordinary state because we have no general theory of inflation.”  Why is this? It’s because conventional or mainstream theories of inflation have failed to provide any robust explanation of why there are changes in the prices of goods and services in modern economies. 

The mainstream view of inflation can be divided into three groups: 1) the monetarist theory; 2) the Keynesian ‘excess demand’/cost-push theories; and 3) central bank expectations theory.

The monetarist theory of inflation holds that inflation is purely a monetary phenomenon, i.e. that inflation is determined by changing money quantities relative to the changes in the quantity of output. Leading monetarist Milton Friedman argued that “inflation is always and everywhere a monetary phenomenon in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output.“

The monetarist perspective is based on a mathematical identity 𝑀𝑉 = 𝑃𝑌, where 𝑀 represents the money supply, 𝑉 is the velocity of money (which measures how frequently money is transacted within the economy), 𝑃 is the general price level, and 𝑌 is the real output of the economy. MV adds up to money in circulation and PY adds up to the nominal amount of income in an economy. Monetarism argues that the left hand side of the equation drives the right-hand side. So, assuming that the velocity of money (V) remains constant, then if M rises faster than Y, there will be an inflation of prices. Thus, changes in M drive changes in P. 

But this equation is an identity; the causal direction cannot be assumed. Marx’s position was the opposite of the monetarist direction. For Marx, money is not value, but the representation of value. So it is changes in the value/price of commodities that determines the volume of money in circulation. “If the velocity of circulation is given, then the quantity of the means of circulation is simply determined by the prices of commodities. Prices are thus high or low, not because more or less money is in circulation, but there is more or less money in circulation because prices are high or low.” 

Changes in P (prices of production) determine changes in MV (money in circulation).And prices change as a result of value changes (Y), as measured in the amount of labour time used to produce all the commodities. If the value of commodities rises/falls (ie more or less labour is necessary for their production), more/less money is needed for their circulation. This is because money is the representation of value, not value itself, which comes from the expenditure of human labour, not from money.

Who is right, Friedman or Marx?  Do changes in MV lead to changes in PY, or vice versa? Is there a close correlation between changes in money supply (M) and changes in prices (P)? Figure 1 shows annual percentage changes in US money supply (M2) and inflation as measured by the ‘implicit’ GDP price deflator). The correlation between the two seems relatively close up to the 1990s.  But from then on, US money supply growth accelerated as a trend while the price deflator decelerated. There was even a negative correlation. Over the whole period from 1960, the correlation between money supply and prices was low (0.21).

Figure 1. Source: FRED, authors’ calculations

A key factor in the weak correlation between changes in money supply and changes in prices is that money supply and money in circulation are not the same. Money hoarding and use of money to make purchases of financial assets can make a considerable difference to the relationship between money supply and prices. Indeed, as Figure 1 shows, the rising gap between money supply growth and GDP deflator from the mid-1990s onwards suggests a diversion of money out of productive assets into financial assets. 

Thus Friedman’s monetarist theory does not hold to explain US inflation in the post-war period.  In a later post, I shall show how it is changes in prices of production (or the value of commodities) that drives ‘money in circulation’ – the opposite of monetarism.

The second mainstream theory is Keynesian. This is dominant among mainstream explanations. A classic example of the ‘cost-push version of this theory was recently offered by Jason Fulman, former US White House economic advisor: “When wages go up that leads prices to go up. If airline fuel or food ingredients go up in price, then airlines or restaurants raise their prices. Similarly, if wages for flight attendants or servers go up then they also raise prices. This follows from basic micro and common sense.” (Fulman, 2022). The causes of inflation, apparently, are rising raw material costs and attempts by workers to get higher wages. This forces companies to raise prices to maintain profits.  This is the cost-push inflation theory.

Marx answered this theory way back in 1865. When debating with the trade unionist Weston who argued that wage rises would cause inflation, Marx argued that rising wages are the “reaction of labour against the previous action of capital” and that “wage rises generally happen in the track of previous price rises.” (Marx, 1865).  Moreover, it escapes Fulman that the effects on prices of wage increases or of materials could be countered by less profits. Given a certain quantity of value, if wages rise/fall, profits fall/rise, prices can remain unchanged.  Fulman’s argument assumes that profitability must be maintained ‘at all costs’.

An IMF study in 2022 addressed this question by applying a cross-economy database of past episodes among advanced economies going back to the 1960s. It found that “wage-price spirals, at least defined as a sustained acceleration of prices and wages, are hard to find in the recent historical record. Of the 79 episodes identified with accelerating prices and wages going back to the 1960s, only a minority of them saw further acceleration after eight quarters. Moreover, sustained wage-price acceleration is even harder to find when looking at episodes similar to today, where real wages have significantly fallen. In those cases, nominal wages tended to catch-up to inflation to partially recover real wage losses, and growth rates tended to stabilize at a higher level than before the initial acceleration happened. Wage growth rates were eventually consistent with inflation and labor market tightness observed. This mechanism did not appear to lead to persistent acceleration dynamics that can be characterized as a wage-price spiral.” In inflationary episodes, wages just try to catch up with prices. But even then, wage increases do not cause ‘wage-price spirals’ – this echoes Marx’s view in 1865.

A variation of the Keynesian view is the ‘demand-pull’ theory of inflation, namely that inflationary pressure results from a positive ‘output gap’ – where actual GDP exceeds ‘potential GDP’, defined as output at the limits of productive capacity including full employment. So inflation results from ‘excessive demand’ in an economy (ie above the supply capacity limit). This ‘excessive demand’ could be due to an expansionary government fiscal policy or due to fast rising wages given full employment, compounded by a fall in ‘potential GDP’, possibly from supply chain constraints and energy shocks. 

What is the evidence for this ‘excessive demand’ theory? Keynesians refer to the ‘trade-off’ between changes in wages and prices and/or changes in unemployment and prices.The former would have a positive correlation and the latter an inverse correlation. In graphic form, this trade-off would take the form of curve, as was first argued by AW Phillips with his so-called Phillips curve in 1958. The evidence for the post-war US economy is that, far from being a curve (ie a trade-off), the Phillips relation is broadly flat.  Here is our own calculation for the US.

Figure 2. Source: FRED, author’s calculations

Other empirical studies also show the Phillips curve to be broadly flat – in other words, there is no inverse correlation between wages or prices and unemployment. President Daly of the Federal Reserve Board of San Francisco, concluded that “the relationship between unemployment and inflation has become very difficult to spot.”. BIS economist Borio agrees: “the response of inflation to a measure of labour market slack has tended to decline and become statistically indistinguishable from zero.” And more recently, he concluded that “inflation has proved unexpectedly unresponsive to economic slack – the Phillips curve is very flat.” (Borio 2021). The former Fed Chairman Jay Powell also acknowledged the problem: “There was a time where there was a tight connection between unemployment and inflation. That time is long gone.” (Powell 2021). This uncomfortable conclusion was also reached earlier by two prominent mainstream economists. Solow (2018) remarked that “the slope of the Phillips curve itself has been getting flatter, ever since the 1980s, and is now quite small.” And Gordon (2018) echoed this: “The slope of the short-run inflation—unemployment relationship has flattened.”

So why do economists and central bankers continue to peddle a theory that has little empirical support? Gavyn Davies, a Keynesian and former chief economist at Goldman Sachs, explained: “without the Phillips Curve, the whole complicated paraphernalia that underpins central bank policy suddenly looks very shaky. For this reason, the Phillips Curve will not be abandoned lightly by policy makers” (Davies 2017). ). But the theory is not made less shaky by denying the empirical evidence against the theory.

The third mainstream theory of inflation is even weaker. It is promoted by central banks and international agencies. It is the expectations theory of inflation. Here inflation is caused by the psychology of the economic agents. If prices rise, then consumers expectations for price rises also increase, leading to accelerating inflation. As the IMF puts it: “It is possible that changes in current and expected inflation are both driven by changes in expectations about the future state of the economy. For example, if firms and households expect that the economy will be in a recession in the near future and inflation will be lower than today, they will start cutting their consumption and investment expenditures now, putting downward pressure on inflation today.”

But the consumer or household expectations of a rise in inflation does not provide a theory for why prices are rising in the first place. Federal Reserve economist Rudd points out that “unsurprisingly, what little evidence we have suggests that firms pay little attention to forecasts of aggregate economic conditions, including inflation.”  Yes, if inflation over the long term falls, then ‘unsurprisingly’, households and firms will expect inflation to fall. So “there is a suggestive low-frequency correlation between an estimate of inflation’s long-run stochastic trend and survey measures of long-run expected inflation.” In the graph below, as inflation slows, so expectations or forecasts of slowing inflation also follow. But expectations follow inflation rates, not vice versa.

Rudd concludes: “economists and economic policymakers believe that households’ and firms’ expectations of future inflation are a key determinant of actual inflation. A review of the relevant theoretical and empirical literature suggests that this belief rests on extremely shaky foundations, and a case can be made that adhering to it uncritically could easily lead to serious policy errors.” Central banks should note.

So we have three mainstream theories of inflation: monetarism; Keynesian cost-push and ‘excess demand’; and expectations, none of which is convincing or borne out by the empirical evidence.  No wonder the mainstream has ‘no general theory of inflation’.

In the next part, I shall discuss heterodox and other Marxist theories of the cause of inflation.