self-fulfilling prophecy



Reflections on the economic theory of “inflation expectations”

Which comes first: inflation or inflation expectations? Seriously, this question doesn't disturb economists' sleep. For they seem to firmly believe that inflation expectations determine inflation. Central bank governors, on the other hand, do not sleep well until they are supposed to be able to stabilize inflation expectations.

Well, common sense dictates that there are only two reasonable answers to that question. The most immediate answer says that inflation precedes inflation expectations, as the second does not exist without the first, but the first can exist without the second. The most robust answer says that they – that is, the objective phenomenon and the subjective experience – influence each other. However, it may seem strange, but economists unapologetically favor the answer that appears to be false. For them, the stench produces the shit.

For this reason, a recently published article by a member of the North American central bank, Jeremy R. Rudd, caused a certain sensation in the college of economists in several countries, in particular in Brazil. Here is the title of the article: “Why do we think inflation expectations matter for inflation? We should?"[I] Here is how the author himself points to the object of his criticism: “Economists and economic policy makers believe that the expectations of companies and families about future inflation are key determinants of current inflation”.

In the article referred to here, this apostate in the world of the orthodox shows that several preachers supported theories that gave an explicit and relevant role for inflation expectations in inflation itself. Renowned authors who even won the Nobel Prize are cited in this way. Among them, he mentions the names of Milton Friedman, Edmund Phelps, Robert Lucas, from the universities of Chicago and Columbia in the United States.

Rudd reviews the texts of this current of thought, suggesting that “this belief” – which is, incidentally, very widespread among economists – “is supported by extremely fragile foundations”. Phelps' assumption - he says - is essentially and following the best practices. Friedman assumes that markets are always in equilibrium. These authors assume that there is no monetary illusion, reject that the path of the economic system is “path dependent”, implicitly admit that the equilibrium in the model of generalized markets is stable (which, as is well known, is false).

But why was this causation introduced into economic theory? – asks this author. His answer says, first, that this determination is present in common sense; there it has the status of a self-fulfilling prophecy. Afterwards, he claims that this prejudice gained dignity because it was mathematically modeled by “great expression” economists and came to exist in the macroeconomic models themselves. In this way, the expectation of inflation “was reified as a feature of the reality that everyone knows exists”.

Now suppose that common sense has a point, and that inflation and the expectation of inflation seem to feed off each other. In this case, it is possible to consider two possible situations: the case in which there is positive feedback between them and the dynamics are explosive, and the case that this feedback is negative and the dynamics are stabilizing. For ease of understanding, these two cases are shown here in graphs. It should also be noted that inflation is constantly receiving “shocks”, which are said to be “exogenous”.

In the first case, where the dynamics are explosive, more inflation implies more expected inflation, which, in turn, implies more inflation. In the second case in which the dynamic is stabilizing, the expectation of inflation is approximately equal to inflation, but the latter is optimistic, as it always assumes that inflation will fall.

The critical author's assumption is that common sense is also good sense. If inflation is growing above 4% per year, the case of explosive dynamics is valid. If inflation is below this value, that is, around 2% per year, the case of stabilizing dynamics is valid. There is, therefore, a bifurcation in the behavior of economic agents and it is found somewhere between these two empirically inferred limits. It is then a matter of knowing why this behavior is observed over time.

Based on the graph below, which roughly presents consumer price and wage inflation in the United States between 1960 and 2020, he concludes that “the trend of the inflation rate above 4% is associated with persistently high inflation dynamics – observed between 1965-1980 – while an inflation around 2% – observed after 1990 – this is not the case”. In fact, if up to 1980 wages accompanied prices, after that date, price growth always surpassed wage growth.

Rudd's answer to the previous question implicitly shifts the focus of the analysis from the dynamics driven by inflation expectations to the dynamics of class struggle, even though he does not use that term. The aforementioned bifurcation is explained from the graph above. “an important characteristic of the inflationary dynamics after 1990” – he says – “seems to be the non-existence of a powerful price/wage spiral” – contrary to what happened before that date. Hence, he concludes “it is unlikely that well-anchored inflationary expectations can explain the stability observed” in the most recent period.

He completes this explanation by indicating that “the wage bargaining process changed radically between these two great periods, marked by a whole transition that took place during the 1980s roughly. Here is a striking difference between the Keynesian period (1945-1982) and the post-World War II neoliberal period (1982-…) with regard to union activism: “Out of the few industries in which unions still formally participate of wage bargaining (currently, this amounts to only 6% of total employment) (…), this no longer exists in the United States”.

The competition of companies for more qualified workers makes them somewhat better off, but the bulk of this large social class ends up competing for low wages. As a result, real wages in the United States have failed to keep pace with labor productivity growth since the 1980s.

The similarity between the two explanations for the inflationary process in the United States is that both focus on an apparent dynamic, in the first case, between inflation and inflation expectations and, in the second, between wages and prices. Both are, therefore, in the field of vulgar economics – because their scientific nature does not go beyond that inherent in common sense, which is also always present in the appearance of what happens and can happen. There is, however, an important difference between them: the second correctly points to the class struggle, even if it does not delve into this issue.

After all, why do capitalists and workers have antagonistic interests in capitalism? Even without going into the issue of exploitation, it is clear, first, that more wages roughly means less profits and vice versa. And this is a crucial problem in capitalism because this system is primarily oriented towards profit – and not towards the production of goods and services –, towards the insatiable accumulation of capital through the appropriation of profit in the sphere of the production of real commodities. Now, the advent of neoliberalism ended up representing a historic defeat for workers in developed countries who were – and continue to be – led by distributive currents, which predominated and predominate in the trade union movement.

In any case, in order to understand the wage/price dynamic, it is first necessary to note that there is an asymmetry of power between capitalists and workers, something that had already been noticed by Adam Smith in the XNUMXth century, but which is usually ignored by a significant part of contemporary economists. . If workers have to organize and fight for nominal wage increases, generally not having family reserves to sustain a long fight, this does not happen to capitalists. Furthermore, the latter have the power to raise prices at the most convenient time, as they are limited only by competition between the capitalist enterprises themselves.

It so happens that in the history of capitalism two forms of competition prevailed: that based on the flexibility of prices, downwards and upwards, existed in the XNUMXth century and the beginning of the XNUMXth century, but was replaced, from then on, by the relative power to generate idle capacity in competing companies. In this case, prices do not fall, on the contrary, they tend to rise continuously.

Well, this “creeping inflation”, around 2% a year, is now observed in all price statistics produced by countries. And it is allowed by an accommodative monetary policy, which ceased to be constrained by the gold standard as early as the 1930s. Monetary policy now regulates inflation only indirectly, by managing the interest rate, and not directly, by controlling monetary aggregates. And this policy, it is evident, has an anti-work bias because it puts workers in a situation of always “chasing after losses”. And as noted in the previous graph, after 1990, they – but mainly the less qualified categories – were falling behind in this race.

Now, it is necessary to ask at what point in the accumulation process do capitalists start to raise prices beyond the creeping trend and, thus, in a much more perceptible way macroeconomically? It can be seen that this is possible because the most important sectors of the economic system are oligopolistic. One popular answer would be that capitalists are there to make money, not to satisfy people's needs and desires. And, as the most cynical economist of all time, Milton Friedman, would say, "there is no such thing as a free lunch." Now, the vast majority of lunches are not “given”, but are provided by the profitability of capital. If it declines, people may starve.

There is a formal answer to the question posed in the previous paragraph; behold, it employs the mathematical language currently dominant in macroeconomics. It was provided by Anwar Shaikh in his great book on how capitalism works.[ii] But here only an outline of this theory will be presented since it is conceptually more rigorous than those produced by the so-called “mainstream".

Note at the outset that inflation under fiat money – and this is basic – is determined by the interaction between supply and aggregate demand. Note, now, that the response of the supply of goods to aggregate demand can be an increase in production or an increase in prices, or even a combination of the two previous ones. What determines this behavior?

The prevailing orthodox and even heterodox theory states that companies prefer to raise prices only when the use of idle capacity is at its maximum, something they call “full employment”. But this theory, in addition to making an explicit apology for the automatic tendency of the system towards maximum employment, is not able to explain the phenomenon of stagflation. In short, even if it is constantly refuted empirically, it continues to be admitted as an unshakable theoretical (ideological) assumption.

Inflation in Shaikh's theory depends positively on demand impulses and negatively on the profit rate subtracted from the market interest rate. Here, the capitalists are not willing to produce more to eventually earn the same or even less than before. Consequently, it positively depends on the degree of participation of investment in profits, a variable that is conditioned by the level of retention of reserves by capitalist companies. These liquid resources held by firms are generally combined with borrowing from the financial sector to finance investment.

As a result, the dynamics of commodity supply seems to be roughly determined by four variables: levels of idle capacity, retention of reserves, net profit rate and mass of profits per production period.[iii] It can be stated, therefore, that the rise in commodity prices will predominate when the profit rate is low and, above all, when the mass of profits tends to stagnate or even fall in line with demand impulses arising from the economic system itself. (consumption and investment), from abroad and from the State.

In the perspective summarized above, the inflation theory is now based on the social structure of the capitalist mode of production and on its reproduction logic. The theory that bases inflation on the expectation of inflation and that aims to serve the utilitarian practice of management – ​​relatively blind – of the system is part of the vulgarities that proliferate in the “best” economic theory. This, as is known, is produced mainly in the “top schools” of the United States, being then uncritically adopted in courses in Brazil where teaching along the lines of that practiced in that specific country predominates.

* Eleutério FS Prado is a full and senior professor at the Department of Economics at USP. Author, among other books, of Complexity and praxis (Pleiad).


[I] Rudd, Jeremy R. “Why do we thing that inflation expectations matter for inflation? And should we?” Federal Reserve Board, September 2021.

[ii] Shaikh, Anwar. Capitalism: competition, conflict, crises... New York: Oxford University Press, 2016.

[iii]See Johnson, Nick. "Modern Monetary Theory and Inflation - Anwar Shaikh's Critique". In:

See this link for all articles