One of the recurring criticisms that mainstream economists make of Modern Monetary Theory (MMT) is that does not follow the rules of formalism that have become the norm in the economics profession. The implication is that by not following these conventions, MMT economists are unable to say anything precise and scientific. Apparently, a literary discourse cannot convey anything that is sound. Pity that some of the greatest contributions to human knowledge have come from those who could write properly. But this criticism of MMT is about something else again. Dominant academic communities develop their own rules of enquiry, which encompass perspectives of the field to be studied, procedures to be followed, methods and techniques to be used and the end goals of the analysis. If those communities become riddled with Groupthink, then a degree of uniformity in practice becomes expected and enforced either subtly through peer group pressure or more coercively through publication, grant and promotional practices, which effectively determine whether a person will advance or be cast aside. The criticism waged against MMT economists that we don’t follow the normal rules of exposition is really an attempt to enforce the discipline of the mainstream (New Keynesian) community and avoid discussion of substantive issues, such as empirical congruence or extent of anomaly. If the dominant paradigm can convince young scholars and the public that its techniques and methods are the only sound way in which to conduct scientific enquiry and highlight an emerging threat as not being up to speed then it can avoid the scrutiny.
In September 2021, the Banque de France decided to reduce its credibility by publishing a Working Paper – The Meaning of MMT (WP #833).
It was a tawdry effort and full of errors and misrepresentations.
For example, it claims MMT economists believe that social welfare initiatives have “no cost”, or that “structural policies” are always “negative” in impact, etc.
I don’t intend to discuss the paper because it is so bad, but it does raise an important issue, which I have touched upon before but which seems to keep cropping up.
That is, the role of formalism.
The Banque de France paper concludes in relation to our work:
Their academic publications are also very repetitive, lacking in empirical analysis and almost exclusively literary, which does not allow to verify their assertions or to compare them with the recommendations of other schools of thought.
The “lacking in empirical analysis” criticism is a joke – consider how many graphs and tables and related discussion I produce just in my blog.
And if you were to consult my academic work – there is a lot of heavy econometric analysis there over many years – heaps of empirical analysis.
But the focus here is that we once again we encounter this notion that MMT economists do not use the mathematical tools and methods of enquiry that define mainstream New Keynesian economics.
As if that is a weakness.
Worryingly, I also detect among younger and emerging MMT economists are sense of ‘inferiority’ about a lack of use of mathematics or a sense of ‘necessity’ for MMT to be developed as a mathematical discipline to accord with the conventions they have learned in mainstream graduate programs in economics.
I reject that outright.
And, while many heterodox economists that I have known over my career have no background in mathematics or mathematical statistics, I can say that I, personally have a strong background, especially in mathematical statistics and econometrics, and applied economics using the state of art estimation techniques.
So my rejection is one of choice rather than a lack of skill.
The same could be said about some of the most famous economists – like Alfred Marshall and John Maynard Keynes, who were both trained initially as mathematicians.
Neither author used mathematical techniques in their major works (‘Principles’ and ‘General Theory’).
The major texts before the 1930s, all avoided any systematic use of mathematics, despite their authors being well-versed in the techniques.
The shift really occurred in the post World War 2 period when in the words of Roy Weintraub in his 2002 work ‘ How Economics Became a Mathematic’:
… economics was reshaped … as a mathematical discipline …
In epistemological terms, the profession really shifted from what it sought to know, to sustaining what it wanted people believe was valid economic enquiry.
The old textbooks used to represent different schools of thought and allow the students to debate the pros and cons of each relevant to real world problems.
The same students used to study economic history (often a compulsory first-year university course) to teach them the importance of context, institutions, politics and history.
And many studied the history of economic thought to gain an appreciation of the evolution of the discipline of economics.
In the 1970s that really changed and textbooks started pumping out the ‘one true model’ approach with scant attention to competing views. Broader subjects such as history and thought were culled from degree programs and the insularity of the discipline became heightened.
In this blog post – The evidence from the sociologists against economic thinking is compelling (November 12, 2019) – I discussed research that focuses on the way mainstream economists think.
The research surveyed economists and reported that one senior and eminent professor told them that:
You are only supposed to follow certain rules. If you don’t follow certain rules, you are not an economist. So that means you should derive the way people behave from strict maximization theory … The opposite [to being axiomatic] would be arguing by example. You’re not allowed to do that … There is a word for it. People say ‘that’s anecdotal.’ That’s the end of you if people have said you’re anecdotal … [T]he modern thing [people say] is: ‘it’s not identified.’ God, when your causality is not identified, that’s the end of you.
I also reported in this blog post the way in which economic research had become standardised.
Prominent New Keynesian, Olivier Blanchard Olivier Blanchard summarised the way in which economics had to be practised in his August 2008 article – The State of Macro.
… strict, haiku-like, rules … [the economics papers] … look very similar to each other in structure, and very different from the way they did thirty years ago …
Graduate students are trained to follow these ‘haiku-like’ rules, that govern an economics paper’s chance of publication success, which then determines academic ppointments, promotion, grant success, and other elevations of influence (becoming chief economist in IMF etc).
So we get a formulaic approach to publications in macroeconomics that goes something like this:
- Assert without foundation – so-called micro-foundations – rationality, maximisation, rational expectations.
- Cannot deal with real world people so deal with one infinitely-lived agent!
- Assert efficient, competitive markets as optimality benchmark.
- Write some trivial mathematical equations and solve.
- Policy shock ‘solution’ to ‘prove’, for example, that fiscal policy ineffective (Ricardian equivalence) and austerity is good. Perhaps allow some short-run stimulus effect.
- Get some data – realise poor fit – add some ad hoc lags (price stickiness etc) to improve ‘fit’ but end up with identical long-term results.
- Maintain pretense that micro-foundations are intact – after all it is the only claim to intellectual authority.
- Publish articles that reinforce starting assumptions.
- Knowledge quotient – ZERO – GIGO.
This is why the publication bias problem in mainstream macroeconomics is so significant.
One of the major issues with this approach is that it is inherently dishonest – persuading us that it is scientific and worthy of authority yet being so compromised as to be worthless.
In this blog post – Mainstream macroeconomic fads – just a waste of time (September 18, 2009) – I discussed the reasons why I would make such an accusation.
The summary point, to reinforce the dot points above, is that the mainstream macroeconomics paradigm claim they are rigourous because they begin with a set of first principles based on stylised assumptions of human behaviour that no psychologist or sociologist would recognise.
These assumptions about rationality and maximising strategies, individually pursued (devoid of social influence) are mathematically specified to allow a ‘solution’.
So the economists claim that their ‘macroeconomics’ is ground in microeconomic behaviour – which they, in turn, claim allows for consistency, a failure in their eyes of the economics of Keynes.
The mathematical framework they produce is by definition highly simplistic because otherwise they would not be able to ‘solve’ it for the unknown variables of interest (output, inflation etc) using standard mathematical approaches.
That then creates an issue.
To move from a theoretical mathematical rendition to a statistical model that econometricians estimate (unknown values) using regression techniques applied to real world data requires a statistical ‘specification’ – estimating equations with dependent variables and explanatory variables.
But the problem is that the simplistic theoretical models provide poor specifications if one is interested in capturing real world data movements.
What to do?
Solution – introduce ad hoc new variables (such as lags between variable responses) to allow the estimating equation to achieve desirable statistical properties – known as ‘goodness of fit’ and sound residual (the unexplained) performance.
The primary justification for the ad hocery is empirical.
But that creates a further issue.
If we believe that starting from first principles – micro optimising foundations – is rigourous then it is usually impossible to relate the econometric specification used to ‘fit the data’ with anything that can be derived from these foundations.
Which means that any results obtained from the data-fitting exercise cannot be used to justify the foundations.
The ‘rigour’ always fails to deliver anything remotely consistent with reality and the ad hoc (non rigourous) tack ons are then used to salvage the empirical exercise.
So at the end of the process there is no rigour at all.
And any policy analysis that comes from that sort of approach is typically useless for helping us understand real world problems.
Moreover it actively undermines the capacity to solve these problems by coming up with stupid conclusions that governments can run out of money or cause inflation and destroy their currencies if they run deficits etc.
The point is that there is no automatic virtue in using mathematics in economic analysis.
Keynes and formalism
Various writers made the same accusations of Keynes’ 1936 book – General Theory of Employment, Interest and Money.
Apparently, it needed economists trained in mathematics to reconstruct the General Theory into formal terms before it made any sense.
Even in recent times, this criticism continues to be made – for example, the 2010 article by Roger Backhouse – An Abstruse and Mathematical Argument: The use of Mathematical Reasoning in the General Theory.
Further, Keynes was warned by his peers that by adopting the conceptual apparatus, tools, and language of the classical school, which he aimed to attack, he would be blurring the argument and leaving himself open to having his ideas hijacked by the mainstream of the day.
Keynes made matters difficult for himself early on in the General Theory (Chapter 2), when he accepted the neoclassical postulate that the real wage is equal to the marginal product of labour and that the two move in opposite directions as a result of the assertion of diminishing marginl returns in production to labour (with short-run capital fixed).
The Stockholm School economist, Bertil Ohlin advised against the use of this ‘competitive’ assumption and Keynes wrote back to him in 1937 saying:
I have always regarded decreasing physical returns in the short period as one of the very few incontrovertible propositions of our miserable subject!
You can find the full discussion in Volume 14 of the – The Collected Writings of John Maynard Keynes – edited by Donald Moggridge and published by Macmillan in 1973. The ‘Letter to B. Ohlin, April 29, 1937 appears on pages 187-191.
By 1939, after two economists John Dunlop and Lorie Tarshis and published several articles (separately) criticising this assumption, Keynes published a follow up article (‘Relative Movements in Real Wages and Output’), which admitted he had erred, effectively ratifying the initial observations of Bertil Ohlin.
But by then the hijacking was underway and the outcome was the neoclassical-Keynesian synthesis, which essentially declared the insights of Keynes to be special cases where wage ridigity was found – and, by construction, an ephemeral situation.
This framework dominated the way macroeconomics was taught in the Post World War II period to the detriment of our discipline.
The current dominance of the New Keynesian paradigm is also the result of that hijacking and has no ‘Keynes’ or ‘macroeconomics’ in it.
The point is that in developing a new approach, it is unwise to use the language and concepts of the degenerative paradigm.
Paradigm shift requires new ways of doing things
Next year marks the 60th anniversary of the publication of Thomas Kuhn’s book – The Structure of Scientific Revolutions – which discusses the way in which scientific activity is conducted and the way in which “normal science” is disrupted by “revolutionary science”, which, from time to time, leads to paradigm shifts where the old orthodoxy, the dominant approach, is replaced rather than augmented by a new, more consistent and empirically tractable knowledge base.
In Chapter VIII The Response to Crisis, Kuhn discusses the transition to a new dominant paradigm as the existing orthodoxy gives way (pages 84-85):
The transition from a paradigm in crisis to a new one from which a new tradition of normal science can emerge is far from a cumulative process, one achieved by an articulation or extension of the old paradigm. Rather it is a reconstruction of the field from new fundamentals, a reconstruction that changes some of the field’s most elementary theoretical generalizations as well as many of its paradigm methods and applications. During the transition period there will be a large but never complete overlap between the problems that can be solved by the old and by the new paradigm. But there will also be a decisive difference in the modes of solution. When the transition is complete, the profession will have changed its view of the field, its methods, and its goals.
That is a very important insight to what happens.
It also highlights the point that once a new paradigm usurps the old, the practitioners will have quite different perspectives, use different methods and seek to achieve different goals.
Modern Monetary Theory (MMT) economists are clearly developing a new paradigm, which we believe is incommensurate with the old New Keynesian approach.
Our questions are different.
Our starting point is different – we seek to understand the uniqueness of being a currency issuer in stark contrast to the entitites that are currency users (households, firms, banks).
We seek to understand the actual way in which the central banks, commercial banks and treasuries interact rather than specifying how we might want them to act in some stylised fictional world and then inferring things from that fiction as if it applies to the world we live in.
And out techniques of analysis are different.
There is no point assuming the tools and frameworks of the degenerative paradigm are functional and worthy of iterating from.
That sort of approach leads nowhere.
The mainstream economists might want all the dialogue to be on their terms because they maintains their relevance.
But from where I sit they are irrelevant for those who seek to understand reality and solve real world problems.
I also discussed the issues of formality in these blog posts (among others):
1. GIGO (October 7, 2009).
2. OECD – GIGO Part 2 (July 27, 2010).
I remind readers of the observation by American (Marxist) economist Paul Sweezy who wrote in the 1972 – Monthly Review Press – article entitled Towards a Critique of Economics that mainstream economics:
… remained within the same fundamental limits … of the C19th century free market economist … they had … therefore tended … to yield diminishing returns. It has concerned itself with smaller and decreasingly significant questions … To compensate for this trivialisation of content, it has paid increasing attention to elaborating and refining its techniques. The consequence is that today we often find a truly stupefying gap between the questions posed and the techniques employed to answer them.
If you get hold of our new textbook – Macroeconomics – you will see mathematical formality when it helps to simplify the argument and advance the understanding.
Otherwise, words more than suffice.
Post Keynesian economist Paul Davidson wrote the following in a Chapter he contributed for the book by Bell and Kristol The Crisis in Economic Theory (Basic Books, 1981, p.157). It describes how mainstream economics uses methods and approaches that renders it unable to embrace real world problems:
There are certain purely imaginary intellectual problems for which general equilibrium models are well designed to provide precise answers (if anything really could). But this is much the same as saying that if one insists on analyzing a problem which has no real world equivalent or solution, it may be appropriate to use a model which has no real-world application. By the same token, if a model is designed specifically to deal with real-world situations it may not be able to handle purely imaginary problems.
He went on to note that heterodox approaches that “are designed specifically to deal with real-world problems … may not be very useful in resolving imaginary problems” that mainstream economists obsess over.
He said that heterodox economists are unable:
… to determine how many angels can dance on the head of a pin. On the other hand, models designed to provide answers to questions of the angel-pinhead variety, or imaginary problems involving specifying in advance the optional-allocation path over time, will be unsuitable for resolving practical, real-world economic problems.
In this tradition, Modern Monetary Theory (MMT) has never concerned itself with computing “how many angels can dance on the head of a pin”.
It is not an imaginary approach that deals with imaginary problems. It is about the real world and starts with some basic macroeconomic principles like – spending equals income.
We leave the angel-counting to the New Keynesians and their Dynamic Stochastic General Equilibrium (DSGE) models.
Young economists who are interested in MMT should never feel that pursuing their enquiry in non-mainstream terms is inferior to the way the mainstream zealots do things.
That is enough for today!
(c) Copyright 2021 William Mitchell. All Rights Reserved.