Sunday, January 29, 2023

‘Tombstone for a Tombstone’: Dealing with the ‘bad science’ behind mainstream criticism of MMT — Phil Armstrong

The problem with conventional economics is that it requires that the "settled methodology" define both the questions and how they are answered, insuring that their mathematical models will be tractable. That is inappropriate for complex adaptive systems. All life systems and especially social systems are complex adaptive systems that are appropriately modeled using organic models rather than mechanistic models suitable for the natural sciences.

While MMT does not use organic modeling based on the life sciences, neither does it use mechanistic modeling based on the assumption that economics is comparable to the natural sciences and can be modeled on similar principles. But this assume non-existent conditions such as homogeneity and ergodicity, which do not apply to social systems that are historical and influenced by culture and institutions.

Why haven't organic models replaced mechanistic models? We aren't there quite yet although much work is being done in this directions. Economics is data-based and therefore limited by path dependence and hysterisis whereas complex adaptive systems involve synergy and therefore emergence due to reflexivity.

Therefore, MMT analysis is based on ex post data from accounting on the one hand and counterfactuals with respect to future contingency. While the future cannot be predicted, it can be forecast to some degree based on the presence or absence of certain conditions that can be specified. 

But this approach does not result in models based on "settled methodology" so it is rejected by the mainstream as deviant. Krugman: "Where is your model?" The retort is, "Your modeling doesn't work and here is why. Ours does to a useful degree and therefore it should be adopted."

The Gower Initiative for Modern Money Studies
‘Tombstone for a Tombstone’: Dealing with the ‘bad science’ behind mainstream criticism of MMT
Phil Armstrong

13 comments:

Peter Pan said...

Calling political agendas 'bad science' is disingenuous.

Tom Hickey said...

Calling political agendas 'bad science' is disingenuous.

Not when a political agenda is being presented as being science instead of being acknowledged as political ideology.

Peter Pan said...

Call it for what it is. Then sack the 'scientists'.

Matt Franko said...

“ Not when a political agenda is being presented as being science”

Yo none of the people have Science degrees…

Tom Hickey said...

They think economics is a science that is methodologically comparable to physics unlike the "social sciences." This is why, according to them, the methodological debate is "settled" in their favor.

Matt Franko said...

Where in Physics are you ever taught that you can run out of the units of abstraction?

And methodology is independent of the discipline… so you can’t say ‘methodologically comparable to Physics”…

You can get an Art Degree in Physics from liberal art schools …

A friend of mine has a BA in Physics from Bucknell… he’s very concerned US is on verge of bankrupt… very concerned about all the US debt…

Tom Hickey said...

They assume a simple mechanistic model of individuals as atoms related by the economic forces of the law of supply and demand — individual preferences relative to goods scarcity in markets as the force field. Then they reduce individual agents to a representative agent, namely, a representative household and a representative firm. This is what microfoundations is about as the basis of macro.

The approach is further hobbled by the requirement that the assumptions of the model result in an algorithm that is mathematically tractable. This is called "formalism" and "Platonism" in the derogatory sense.

OK, this is admittedly a rather simplistic caricature of the conventional approach, but it does captures the essence of econ 101. See Philip Mirowski's work to fill in the details.

Matt Franko said...

“ They assume a simple mechanistic model of individuals as atoms”

That’s an ANALOGY… ie FIGURATIVE LANGUAGE… ie textbook Art degree 101..,

Employment of figurative language is not Science…

Tom Hickey said...

To be fair to conventional economists, that is not an an analogy they use themselves, that is, they do not liken individuals with atoms although they do speak of "economic forces" and "economic laws," as well as perfect markets as "atomistic markets."

Rather, this is a criticism leveled at them by people like Mirowski who argue that conventional economics is modeled like physics, presumably due to "physics-envy."

Moreover, the assumption that the methodological debate is settled in favor microfoundations, maximization, etc. is the basis for the claim that conventional economics is a real science like the natural sciences unlike the social sciences and psychology that are only science-like. They really so think they are "doing science" as applied math. They reject heterodox economics as not having tractable mathematical models like conventional economics.

The criticism of this is that conventional economics just has the appearance of being like physics but that it cannot be since their subject matter is categorically different.

Matt Franko said...

“ The retort is, "Your modeling doesn't work”

They will just ignore that.. they don’t understand what you are talking about..

You’re implying testing (“doesn’t work”) which is not trained under the Art degree methodology…

“28 And according as they do not test….. God gives them over to a disqualified mind” Rom 1:28

They don’t test anything… they are trained to just robotronically advocate for their thesis like drones…

This should disqualify them…

Tom Hickey said...

They don’t test anything… they are trained to just robotronically advocate for their thesis like drones… This should disqualify them…

This is the heterodox criticism of conventional economics. Two reasons have been offered. The first is the level of investment is so high in gaining academic credentials and the penalty so onerous for going off the reservation that most economists just do what is expected of them institutionally and in their subculture. The Upton Sinclair quote, 'It is difficult to get a man to understand something, when his salary depends on his not understanding it,' likely applies here.

Secondly, TPTB profit can be viewed as profiting from the conventional model that justifies what the TPTB do as "scientific," and therefore TPTB not only support it but also suppress opposing voices. Of course, this view is criticized as "conspiracy theory" but it is widely believed — probably falsely since there is no smoking gun evidence — that the TPTB initially launched that line of criticism, e.g., through the CIA and still wield it effectively in the info war even though the "red scare" is over. But "Russia" and "Putin" still used with similar effect.

Conspiracy aside, my experience in academia showed me that the donors do wield power over what gets promoted and what is considered "off the reservation." Moreover, academia itself is constructed as a reservation, at least for the most part. One can say more or less what one wishes in the boonies, but for advancement in the profession one has to toe the line to get published in top journals, get promoted, and get hired by institutions that are influencers. MMT is tolerated at UMKC but there are still no MMT economists in positions of influence in the profession.

So training is one thing and politics is another. They may both be operative here.

Tom Hickey said...

Lars Syll has a related post today to which I just linked.

One of the confusions about economics as a science is the use of the term "science" to mean both a body of knowledge and a type of methodology. It appears to me that many confuse these.

Conventional economics is a body of knowledge contained in the vast literature, some of which is categorized as conventional or academic economics and which is claimed to be scientific with the implication that it is "rock solid" like physics and chemistry. Owing to the difference in subject matter this is impossible. To paraphrase Richard Feynman, physics would be different if electrons had feelings, (think preferences).

Part of this is the canonized methodology along with methodological assumptions that are actually substantive in addition to methodological. This allows conventional economists to dismiss heterodox economists as "heterodox," that is off the reservation.

Scientific method involves two major aspects, theoretical and empirical, that is, an axiomatic hypothetical-deductive approach to mathematical modeling on one hand, and being data-based on the other.

Conventional economists seem to emphasize the first but sometimes only pay lip service to the second, since economic data is looser that data obtained from controlled experiments, which are difficult to do in econ, e.g., for ethical reasons in addition to difficulties with the data itself. A lot data is based on aggregation and estimation. As Lars points out, Keynes cautioned about the epistemological weaknesses inherent in this type of subject matter, so that conclusions need to be qualified, which they are often not but presented as "scientific" when they are only "sciency."

Peter Pan said...

Hard science = science
Soft science = pseudoscience

Psychology has a body of knowledge (i.e. the literature). But that doesn't make it a science.

Psychology fails as a hard science because experiments can't be repeated under the same conditions. The test subjects do not remain in the same mental state. Therefore there are little to no controls to isolate the factors from the results.

Unlike psychology or medicine, mainstream economics fails as an art.
Its practitioners live in ivory towers, which are devoid of street pragmatism.
Except of course, for career/salary pragmatism.