Friday, August 19, 2011

Hickey on the MMT Paradigm Shift

Tom posted this comment at billyblog that is just TOO GOOD to leave buried in comments at Bill's (sorry Tom!):

"Every system that claims to be scientific is based on a formal system (algorithm). The algorithm is a set of tautologies that say nothing about the world. MMT is a macro theory so it begins with Y = C + I + G + (X-M), and transforms this using rules of formation and transformation into the sectoral balance equation, for instance.

The theoretical aspect of the science is the interpretation of the algorithm wrt causality. Macroeconomics needs a microfoundation that reveals the causal mechanisms in terms of individual and group behaviors. This connects the algorithm to reality and enables empirical verification, which is the criterion of truth of hypotheses in science. Explanations and prediction must not only exhibit correlation wrt date, but also be based on a theoretical causal mechanism that explains the why and how underlying observed correlation.

This is where it gets tricky. The big difference between MMT and other macro theories is the mechanism and direction of causation. For example, the ISLM model is base on loanable funds and changes in interest rates, which is basic to monetary policy. Supply side macro is based on investment and changes in taxes. Austrian economics is based on the real business cycle and changes in credit. Post Keynesianism is based on effective demand and changes in income and its distribution. MMT is a sub-school of Post Keynesianism that is based on offsetting changing propensity to consume/save from income with vertical (outside) money.

Different schools of thought attack each other on this basis. This debate reveals strengths and weaknesses in the respective theories. Through debate, often intense and partisan since economics has a normative component, agreement over “laws” is reached. These laws become the norms of the “normal” paradigm that dominates the scientific universe of discourse and doing “normal” science until enough anomalies arise to warrant a scientific revolution. This is basically the Kuhnian view. Hopefully, we are in the midst of such a revolution. When mainstream economists failed to predict and cannot explain a phenomenon like the global financial crisis, it is evident that something is amiss with the paradigm they are operating in terms of.

The normal paradigm in economics is largely the neoclassical (Walrasian) paradigm, with two different schools contending, the monetarist/neoliberal school following Milton Friedman, and the New Keynesian school, following Paul Samuelson, whose goal it was to combine Keynesianism with New Classicalism. Accepting general equilibrium, both schools base their theory on the theoretical concept of “natural rate.” Heterodox schools generally reject this, as did Keynes.

These theories as scientific explanations, as well as and their predictions that are often advanced as policy recommendations, are expressed as equations involving dependent and independent variables. Testing is done by applying data (factual) and hypotheses (counterfactual) to these equations using the relevant variables. Of course, empirical testing requires data, but this is not always possible in social and behavioral sciences, including economics, due to practical considerations.

Admittedly this is a picture painted with broad brush strokes, but I think it conveys the general idea of formalization in relation to scientific theory and how this relates to economics."

One of the things that really appealed to me about Mike's show was how Mike was not content to just simply accept or simply oppose the mainstream economic dogma. Mike would often "wax Philosophic" about not just what market participants were saying, but Mike would often seek to understand why the participants were were saying it. This is uncommon in the media and had a great appeal for me anyways.

Tom here investigates and breaks down some views as to economic causation (the "why") from the points of view of the different schools of economic thought.

This is a huge, deep, scholarly look into what is going on at a basic, human psychological level in the area of macro economics in and around MMT.

For those of us already within the MMT paradigm, understanding the true nature of macroeconomics comes easy to us (it's a gift imo). I'm afraid it does not come that easy to most others, we need to be aware of this. Tom's post here will help us to reach a deeper understanding of where many intellectual opponents of MMT are coming from in their thinking.

(Tom, no one else (literally in the world) is looking at these issues at this deep a level, Resp,)


Senexx said...

This comment I'm leaving doesn't add much value but I feel that way about a lot of comments Tom Hickey leaves on various blogs.
Many of them deserve their own posts.

Y said...

For example, the ISLM model is base on loanable funds and changes in interest rates, which is basic to monetary policy.

A quibble: I think IS/LM is a reconciliation of the traditional loanable funds determination of the interest rate and Keynes' liquidity preference determination of the interest rate. So it's loanable funds AND flight to liquidity.

I don't know what RBC has to do with the Austrians - the Austrians aren't really into math and the RBC theorists think that technological hiccups we (for some reason) can't identify have made 10% unemployment optimal. Or do you mean they're quite similar to the RBC theorists in their perception of economic virtue in the slump?

Unknown said...

If Thomas Kuhn's theory of scientific revolutions is wrong (it certainly didn't stand up very well to questioning by way of historical analysis in the 1970's) what impact, if any, does this have on the rightness of MMT?

Kuhn doesn't do very well with biology, for example, because there isn't a very clear revolution between an old and new regimen.

What I would like to see more of is a recognition of 'history' as essential to the way practitioners of MMT operate. Read Gould on history and biology and you'd see what I mean. It's the 'history thing' that permits the full force of data to be included in economic analysys within MMT influenced economics.

Steve Keen also talks about this from his perspective e.g. in Debunking Economics.

NeilW said...

For me the important point is this one.

It's just a model.

Models are abstractions, necessary simplification, not the real thing.

Therefore by definition they are wrong as a starting point. The only question is how wrong and in what way. We have to understand that and ensure that the contingencies in policy can handle the built in wrongness.

MMTs great strength is that it assesses its predictions against what happens in the real world *and then modifies accordingly*. It can handle real world feedback.

One of the reasons I like Steve Keen's dynamic models is that they clearly follow the patterns that happen in the real world. The question then is whether this is curve fitting or predictive.

Abstract mathematical models devoid of real world testable objects like banks are not suitable for simulation purposes. That's like having a weather model without clouds.

Anonymous said...

Tom needs his own blog.In that blog he should (just suggesting) explore monetary and economic theory AND policy with his unique multidisciplinary perspective.

Rogue Economist said...

What I actually like about Tom's comments is he puts his arguments in terms of people, not just in terms of concepts. It resonates better, especially compared with a lot of other discussions where people just abstract from reality.

PG said...

"Every system that claims to be scientific is based on a formal system (algorithm). The algorithm is a set of tautologies that say nothing about the world."

This comment will probably be utterly ineffective, but the following expresses the way logicians use words.

A formal system and an algorithm are DIFFERENT concepts (albeit strongly related).

An algorithm and a set of tautologies are DIFFERENT concepts.

An algorithm is a set of formal expressions that when interpreted procedurally will yield the solution for a given problem in finite time for all possible inputs.

A tautology is an expression of propositional logic that is valid, which means that it is true for all possible attribution of the true or false values to the atomic statements appearing in it.

An equation is not a tautology. Equals, '=', is NOT a logical symbol. Or, '\/', and, '/\', , implies, '->', are logical symbols.

Tom Hickey said...

Hey, I said this picture was broad brush stroke. :)

I was trying to suggest that there are key variables that different schools are concerned with because they think that these variables have priority and largely determine the way that macro should be approached. This is pretty much the way I see arguments on the blogs going wrt MMT and other schools. To some degree the parties are talking past each other because they are using different frameworks and norms, which set priorities and determine methodological approach. I have learned something from all of them, even when I disagree with them over key fundamentals.

I also realize that the philosophy of science of is a somewhat contentious field, but I see Kuhn's work on paradigms as still have value. It makes a valid criticism that unlike the hard sciences (with hard criteria and rather sharply defined boundaries), the life and social sciences do not have comparable normal paradigms to which all that practice that branch of science agree.

The problem here is that a quasi-normal paradigm gets imposed by TPTB, that is, the academics that hold the principle chairs at the top-rated universities and who control the dominant journals. This determines who is in and who is out, as Feyerabend observes, as well as how science "should be" practiced if one wished to be taken seriously and advance academically and in wider influence.

Basically, the difference between say, Popper, and Feyerabend is the distinction between structural and functional, as shown, for instance, in the difference of approach of contemporary Western medicine and Traditional Chinese Medicine. The Western approach is in terms of specific physical organs as structural mechanisms and the Chinese approach is in terns of broad, complex, and interrelated organ systems as energetic pathways. Both have their uses, advantages, and disadvantages. In the larger picture, they complement each other, since structure and function are complementary rather than antagonistic or exclusive.
(continued below)

Tom Hickey said...

So I would not argue that MMT is "right" and other schools are "wrong." All make contributions, which if pushed too far, i.e., seek to dominate discourse, end up being "wrong" in that sense, because they insert bias. No one and no school has the ocean in their bucket.

In the softer sciences there is actually a 'normal" paradigm but it is not recognized as such. The normal paradigm is fluid and shifts with the debate among the dominant school(s) and the subdominant ones as well. Even the outliers must be included in the overall context (as MMt'ers are now arguing fiercely on the blogs).

If we all looked at economic in this light, the field would be a lot healthier and make faster and more comprehensive progress.

My field is philosophy, and philosophers are just as notorious for arguing with each other and disagreeing as economists. I was trained in the historical tradition and had to encounter all major schools thought and some minor ones, as well. After graduating, I did post-doc work in the Vedic tradition and then began to incorporate other non-Western areas too.

My conclusion is that all great contributors were describing the same context from different points of view, and all of these points of view contribute to an understanding of how human beings approach ultimate questions about that context in which they find themselves. Even though there are many sub-contexts and many differences in human individuals and cultures, there is a uniformity of concern and approach because we are human. Philosophy, the humanities, and arts, especially, but all human endeavor provides insight into what it means to be human and what that implies for, e.g., cognition, volition, affect and behavior. Economics is a piece in this fascinating puzzle, too.

BTW, a lot of economic thinking is starting to emerge from the developing world now. It will play an increasingly important role in shaping the economic universe of discourse, as well expanding appreciation for global context from points of view that are as yet unfamiliar in the West.

Tom Hickey said...

BTW, I should add that the point of the original comment at Bill's was to counter an argument in the comments there that MMT is just "formal,"
consisting of a bunch of identities (tautologies) to which all economists agree, and that MMT is trying to make a big deal of it, as well as trying to get water out of a stone, since identities say nothing other than certain relationships always hold in a particular system.

The main point of my comment there about the difference between formalization of model and interpretation involving causality was articulated nicely in terms of MMT in Peter Cooper's post on Sectoral Balances and Keynesian Causation, which distinguishes between the formal aspet of the sectoral balances SFC model and the imputed causation that characterizes MMT as a theoretical explanation. Peter's post is developed much more in detail regarding MMT than my brief remark, and It's a must-read, IMHO, especially since this is a criticism of MMT that is being thrown up more often as if it were lethal. Unfortunately, I didn't think to cite it when I wrote the original post. Apologies to Peter, since this goes right to the heart of the objection.

Tom Hickey said...

"Tom needs his own blog.In that blog he should (just suggesting) explore monetary and economic theory AND policy with his unique multidisciplinary perspective."

Hey, I am trying to get some work done in my own field, too, and have a life as well. Lots of things I'd like to do if there were 48 hrs in a day. :)

Greg said...

Well done Mike (and Tom too!)

Tom is one of the MMT troopers out there. It been in the comment sections that Ive learned a lot the last three years and Toms comments are always a stopping point.

Detroit Dan said...

I think having Tom as a poster at Mike Norman's blog is working well. There are several posters here, and that keeps the blog lively.

Thanks to you all!

Tom Hickey said...

PG, I responded to you at Bill's here.

Tom Hickey said...

Y: "I don't know what RBC has to do with the Austrians - the Austrians aren't really into math and the RBC theorists think that technological hiccups we (for some reason) can't identify have made 10% unemployment optimal. Or do you mean they're quite similar to the RBC theorists in their perception of economic virtue in the slump?"

Right, I should have written Austrian business cycle theory. My bad.

RBC theory holds that any failure of markets to clear is due to some exogenous shock rather than markets failing to clear. The problem is said to be "real" since it is neither fiscal nor monetary, but structural, so if government is going to be involved it should only be in response to structural shifts in the economy.

Austrians say the failure is due to malinvestment owing to imprudent credit extension due to excessively low rates for too long. Austrians also hold that the problem is structural, and in that sense "real", since it results from institutional arrangements that depart from sound money. They hold that the solution is a return to sound money in banking to eliminate boom-bust by changing these arrangements rather than using monetary or fiscal policy. Their preferred solution is to get government out of the picture, since they think that government is responsible for the institutional issues.

What I was thinking of when I wrote real business cycle in the comment was the distinction between ordinary business cycles usually followed by a V-shaped recover as unplanned inventory clears and long financial cycle that are followed by L-shaped stagnations as deleveraging takes place and savings are replenished. Neither RBC or Austrians make this distinction made by Fisher and Minsky, and adopted by Post Keynesians. MMT proposes a fiscal approach.

The ISLM model is based on an assumption of loanable funds, that is, that demand and supply of loanable funds are equal. It was first suggested by John Hicks, who later said that it doesn't represent the position of Keynes and is really only useful as a classroom device. Tell that to New Keynesians though, who prefer a monetary approach to a fiscal one, except in a so-called liquidity trap.

Unknown said...

The second thing I'd say is 'politics'.

Keynes is a really interesting thinker because he understood this as one of the determinents of economic activity. And haven't we had some wonderful examples of this since the economic world refused to behave itself under pressure from the political detminants that lead to the great recession.

And politics most definitely is not of the micro behavioural kind that some economists see as populating and animating the macro world.

Finally, the word 'uncertainty' of Keynesian etymology also springs to mind!

mike norman said...

Tom is a really smart guy and I value his contributions to this blog immensely. He's also incredibly generous to commit his time here as well as other places. If he leaves to start his own blog I would be sad to see him go, but I would definitely be an avid reader of whatever he puts up.

Anonymous said...

It's a struggle just to get people to read MMT literature. Krugman syndrome :(