Tuesday, May 2, 2017

Jason Smith — Mathiness in modern monetary theory


I feel a debate coming on.

Information Transfer Economics
Mathiness in modern monetary theory
Jason Smith

39 comments:

Neil Wilson said...

What I find interesting is that Jason and Simon didn't understand the mathematics and what it meant in reality. And they didn't bother asking to clarify.

And yet they expect the rest of the world to understand their mathematics and what it means in reality.

Which just shows you how bad mathematics is as a tool for describing the real world with people in it. Which is why I never use it in my daily work.

Similarly it is interesting to see a physicist talk about economic empirical data in the same way as empirical data in physics. What that shows you is that they really believe they are uncovering some universal truth. Some 'deep parameters' that are beyond policy.

But economics is about people and they change their behaviour depending upon the circumstances in very complex ways. There is no universal truths there. Those that think they can see them are really just seeing teddy bears in the clouds.

The empirical economic data we have has been collected from a man in a straitjacket. It tells you nothing about what will happen when you take that jacket off. We simply don't know because we have never tried.

So arguing for tests against empirical data tainted and constrained by classical and neoliberal policies is actually just a political argument for keeping the jacket on.

Neil Wilson said...

I should clarify that tests against empirics won't work for *policy changes*. The tests against current reality of MMT has been done to death. It is a description of how the monetary system works.

Ben Johannson said...

I stopped reading Jason Smith over a year ago when it was clear he couldn't be bothered even to get the simplest history of Keynesian thought correct. He's far too fascinated with math for its own sake and far too attached to his assumptions.

Brian Romanchuk said...

I looked at it quickly; it looks like gibberish.

The way these things are usually framed is that government spending is direct consumption, and so it does not just affect income (a nominal variable), but also real variables. (Transfer spending might be viewed as a negative tax.) Therefore, the two sides are the government deficit equation have differing impacts on the real economy, and that is what the debate is about.

It might help if he tried understanding the mathematics used, rather than just make up stuff based on his lack of understanding.

Auburn Parks said...

Neil-

I'd like to quibble about the "never tried" part. At least in the US case because we had nobody bombing us, we have already been given real life proof of what the economy and our society is actually capable of when you take off the speed governor, that case was obviously WWII. We basically doubled the size of the economy in 5 yrs, all the while we were missing about 8 million (on average) of prime working age men in the workforce. It really is our only example of whats possible, but man are the possibilities enormous.

André said...

"Maybe MMT people don't want to say they're making unfounded assumptions just like mainstream economists (or anyone, really) and so hide them 'chameleon model'-style a la Paul Pfleiderer."

Here we go again with this "intelectual dishonesty" or "conspiracy" theories - but this time directed to us, MMTers.

Yes, Jason Smith is saying that we know we are making unfounded assumptions and that we are deliberately trying to hide it with math. I can't stop thinking he is nuts, don't you think?

But the thing is that we do the same with other theories. We are a little nuts too. When will this "intellectual dishonesty" thing stop?

Matt Franko said...

"How bad mathematics is as a tool"

Neil it is the stochastics not the deterministic functions I think you are having a beef with....

Remember also Jason as a trained physicist is trained in stochastics via the heavy doses of Schrodinger/particle in a box/quantum stuff he's been given (FD I barely passed nuclear physics I simply could not understand it vs. A in Newtonian physics, so I'm probably biased) so he will draw that stuff like a gun.... he's been trained that way...

What you and Brian are coming up with lately is deterministic functions which imo (FD I am probably biased from my training in deterministic time domain functional analysis ) is the way forward for us...

I'd encourage you and Brian to keep it up... you have that tax rate in your latest and Brian has illustrated a response function in taxes from the leading govt spending, its like you could tune Brian's function with the tax rate and get either a steeper or flatter response to the govt spending .... pretty interesting....

Matt Franko said...

Both Jason and Noah are trained physicists.... also probably neither have ever taken an Accounting course...

Matt Franko said...

The other thing is don't forget the CBO/Peterson people use that whacky Modified Accrual accounting method of theirs....

So even if we point to the area under Brian's curve as the current sum of govt "Taxes Receivable" so to speak and treat the PV of it as an accrued asset for govt, those CBO/Peterson people can't do that via Modified Accrual they can only use Cash Basis on the left for Federal govt accounting... so we see them constantly come up short...

So we have to also state up front that we are intending to use a normal GAAP Accrual Basis of our accounting so we would always be in balance... then model Brian's time domain response function of govt spending with Neil's different tax and savings rates as the accrued asset "Taxes Receivable" value declines as taxes are paid and the TGA increases....

Neil Wilson said...

"I'd like to quibble about the "never tried" part."

Beveridge has a similar conclusion about the impact of WWII on the British economy. Reading "Full employment in a Free Society" is jaw dropping. The arguments laid out in there are the same ones we are debating today. We are no further forward after 70 years.

What I'm talking about are the specific policies that arise from analysing from the MMT viewpoint - proscribing bank lending, the Job Guarantee, using control functions other than taxation to free up real resources, external sector trading limits, even the Debt jubilee.

We don't have the data for anything other than aggregate demand management - and that tends to cause inflation. The Job Guarantee deals with the inflation part, and because it is automatic it goes someway towards addressing the Liberal objections to state action - that you can't rely on politicians to do anything sensible.

AXEC / E.K-H said...

ICYMI Where MMT got macro wrong
http://axecorg.blogspot.de/2017/04/where-mmt-got-macro-wrong.html

Ignacio said...

He has some good points though, and I like his conclussions:

There seems to be a substitution of mathematics for understanding. In fact, the Medium article seems to think the derivation it goes through is necessary to derive its conclusion. But how can a series of definitions lead to anything that isn't itself effectively a definition?

This is how mathematics work, and this is way mathematics are NOT a description of the real world per se. Mathematics are a series of tautologies proved by the means of logic. Nothing else, nothing more. It's a self-contained, closed (and incomplete), symbolic system that can be manipulated in a serial manner.

There is way too much fucking "under the hood" Platonism going around in the scientific world nowadays re. mathematics (not just economics, but is particularly bad in economics).

He continues: This really tells us nothing about the real world unless we make some connection to it however.

Right, but this is where he goes wrong thinking MMT (labeled incorrectly, the the "Theory" in the MMT is out of place and a marketing mistake, it should be MM-Operations or something like that) makes not such connections and is making whatever assumptions and compares it to ISLM nevertheless, give me a break.

There ar eparts of MMT which are underdeveloped and subject to criticism, but when it comes to the internal description of a national economy as a closed system, it's pretty much on point. When it comes to international trade dynamics, foreign exchange and the political economy, it's all a lot weaker, but the monetary basics are ok.

AXEC / E.K-H said...

Mathiness is NOT the problem — scientific incompetence is
Comment on Jason Smith on ‘Mathiness in modern monetary theory’

“The economist Paul Romer at New York University has recently begun calling attention to an issue he dubs ‘mathiness’ …”. This is somewhat beside the point. First of all, Paul Romer rediscovered a phenomenon that is as old as neoclassical economics: “Mr. Stanley Jevons, and M. Walras, of Lausanne, without communication, and almost simultaneously, have worked out a ‘mathematical’ theory of Political Economy; — and anyone who thinks what is ordinarily taught in England objectionable, because it is too little concrete in its method, and looks too unlike life and business, had better try the new doctrine, which he will find to be much worse on these points than the old.” (Bagehot, 1885, PE. 27)

Secondly, Paul Romer messes the whole issue up with a false analogy: “Romer believes that macroeconomics, plagued by mathiness, is failing to progress as a true science should, and compares debates among economists to those between 16th-century advocates of heliocentrism and geocentrism.”

As a matter of fact, the debates of the advocates of heliocentrism and geocentrism were strictly PHYSICAL, i.e. about whether the earth stands still and the sun moves or just the other way round, and mathematics/geometry was only insofar an issue as it has been assumed since Plato that planetary orbits must be perfect circles.

So, if the current macroeconomic debates resemble anything medieval then the debates about how many angels can dance on a pinpoint. This debate was indeed vacuous because angels are NONENTITIES. In contradistinction, the debate about heliocentrism and geocentrism had REAL content. Likewise, the question whether planetary orbits are circles or ellipses had REAL content. And this is why the measurements of Brahe became decisive. Physicists NEVER had a mathiness problem, only economists have.

As an economist, Romer does not understand what science is all about. This is not a big problem, though, because his fellow economists have no idea either.

See part 2

AXEC / E.K-H said...

Part 2

Non-science or pseudo-science is easily recognizable. It is much harder with what Feynman called cargo cult science: “So I call these things cargo cult science, because they follow all the apparent precepts and forms of scientific investigation, but they’re missing something essential.”*

What is missing? Clearly, macroeconomic theory lacks content. The fatal defect is the vacuousness of fundamental concepts like utility, optimization, equilibrium, production function, capital, perfect competition, demand function, etcetera. ALL these concepts are NONENTITIES like angels or the Easter Bunny. It is NOT formalization that has to be criticized in the first place, it is the all pervasive green cheese assumptionism which has become the hallmark of current macro.

Ultimately, nonentities cannot be formalized. With superficial and faulty formalization economists abuse mathematics: “When very sound and proper mathematics is misused and misapplied to fairyland problems without any basis in the real world, that fact that the mathematics itself is impeccable makes the whole obnoxious game just that more offensive.” (Blatt, 1983, p. 173)

But things are even worse. Mathematical economists do not really understand the math that they apply. This, too, is not news: “The so-called ‘mathematical’ economists in the narrower sense — Walras, Pareto, Fisher, Cassel, and hosts of other later ones — especially, have completely failed even to see the task that was before them. Professor Hicks has to be added to this list, which is regrettable because he wrote several years after decisive work had been done — in principle — by J. von Neumann and A. Wald. (Morgenstern, 1941, p. 369)

The project of the proper formalization of economics, though, had one fatal drawback: von Neumann left the underlying theory untouched: “But this [establishing the analytic mother-structure] required one very crucial maneuver that was nowhere stated explicitly: namely, that the model of Walrasian general equilibrium was the root structure from which all further work in economics would eventuate.” (Weintraub, 2002, p. 121)

This means that von Neumann swallowed all the vacuous concepts of Walrasianism from utility to equilibrium: “In any event, it seems that Morgenstern finally convinced von Neumann that they must proceed tactically by means of the conciliatory move of phrasing the payoffs in terms of an entity called ‘utility’, but one that von Neumann would demonstrate was cardinal — in other words, for all practical purposes indistinguishable from money …” (Mirowski, 2002, p. 127)

So, we arrive at the proper definition of mathiness as FORMALIZATION OF NONENTITIES. As genuine cargo cult scientists economists habitually apply mathematics without deeper understanding: “The mathematical language used to formulate a theory is usually taken for granted. However, it should be recognized that most of mathematics used in physics was developed to meet the theoretical needs of physics. ... The moral is that the symbolic calculus employed by a scientific theory should be tailored to the theory, not the other way round.” (Wittgenstein, quoted in Schmiechen, 2009, p. 368)

From calculus to set theory, economists DID IT ALWAYS THE OTHER WAY ROUND.

With mathiness, Paul Romer sloganized a pseudo-problem. For a mathematician’s assessment of the ingrained mathematical incompetence of economists see Jonathan Barzilai’s Open Letter to the President of the American Economic Association.**

It is prettey obvious that Jason Smith has zero competence in economics, mathematics, and physics.

Egmont Kakarot-Handtke

References see
http://axecorg.blogspot.de/2016/04/mathiness-is-not-problem-scientific.html

* Wikipedia https://en.wikipedia.org/wiki/Cargo_cult_science
** Scientific Metrics, Open Letters and papers
http://scientificmetrics.com/publications.html

Matthew Franko said...

"substitution of mathematics for understanding"

That is what stochastic analysis is no?

You said we typically use stochastics when we dont understand what is going on... ie its a tip-off to competent people...

so you see all of these people who dont know what is even going on using the stochastics... we should not be surprised by this...

Brian Romanchuk said...

Mainstream: MMT has no math!
Jason Smith: MMT is hiding everything behind assumptions buried in mathematics!

My head hurts even thinking about how that can be responded to.

Noah Way said...

Any real scientific analysis of economics would return the simple fact that economics is not a science. Thus to discuss the failure of economics from a mathematical point of view is an excerise in absurdity.

Economic science is the intellectual equivalent of intelligent design.

Ignacio said...

Matt, IMO stochastic analysis is a tool we have nowadays (due to well developed probability theory) to approach a problem, preliminarily, we don't understand that well. We cannot derive clear causality from it, but we can explore regularities with it. It's useful if used correctly Is important to understand that an stochastic process is not necessarily truly random, but nonlinear (I don't think most economists really get any of this tbh, as they are mystified). In the other thread below from Jason I just posted something regarding this so won't repeat myself.

The problem here is that there isn't really THAT much to understand when it comes to macroeconomics, is not like we are exploring some difficult natural phenomena which is hard to observe. I like Neil's straitjacket analogy... you control the freaking straitjacket, there is no "invisible hand!" controlling it, you made it, you decide how strong you can tighten it, or soften it, you decide who wears it or who doesn't. There is nothing to figure out: you control the freaking thing like if you were God. Deal with it and move on to real problems on how to use that, do not keep pretending you don't control or have invented it stupid man.

So yeah, here we are...

Matthew Franko said...

"you control the freaking thing like if you were God. Deal with it "

Well that is textbook Determinism 101... so a lot of people dont accept that... they think "free will!", libertarianism, justification thru works, "St Peter at the gate!" metaphor, Hell Doctrine, etc..

A lot of biases to overcome... but I think we have the best shot thru an approach using deterministic functions from which we make predictive statements that are testable... this should appeal to competent secular STEM people of which there are many out there...

Tom Hickey said...

Matt, no MMT economists I have ever heard or read thinks that the economy can be known or managed deterministically.

Keynes certainly did not.

The notion that an economy can be known deterministically and therefore managed with precision is the basis for a command economy.

Ignacio said...

Tom macroeconomics is not "the economy". That perception is one of the things that has to be overcome.

And the first ones who believe that the economy can be "commanded" are mainstreamer and central bankers (which is rather ironic if you ask me), by manipulating interest rates and QE, and what not magic they come with next.

Most of the "macroeconomic" parameters can and are tuned by policy choice, there is no "invisible hand" at work here. Is all bullshit.

Tom Hickey said...

A lot of biases to overcome... but I think we have the best shot thru an approach using deterministic functions from which we make predictive statements that are testable... this should appeal to competent secular STEM people of which there are many out there...

Please illustrate this with a simple example so I can understand what you are talking about specifically. Please specify how the data are measured.

Jason Smith said...

"What I find interesting is that Jason and Simon didn't understand the mathematics and what it meant in reality."

Neil buddy -- when you start thinking a theoretical physicist and a macroeconomist didn't understand some math, it's time for a little introspection.

"I looked at it quickly; it looks like gibberish."

Brian, you had a problem understanding first order conditions, so your expertise here is a bit questionable. In any case, I re-derived the result Neil gives so if that's gibberish, you should probably take it up with Neil.

"Yes, Jason Smith is saying that we know we are making unfounded assumptions and that we are deliberately trying to hide it with math."

Andre, that was one of four reasons offered, but I said I didn't know why. I still don't know why. The mathematics in Neil's article are totally unnecessary to the conclusion which is contained in the assumptions. Maybe the math was fun on its own? I admit, I enjoyed putting together the Mathematica notebook I showed in my post (also, Brian -- there's a Mathematica notebook in my post).

I agree that "government spending can pay for itself" because it can boost growth. But that is basic Keynesianism, and there's no reason to come up with a series of mathematical operations that end up obscuring the original assumptions.

Jason Smith said...

Egmont -- still insane, I see.

Jason Smith said...

Neil, I also can't quite understand how you could say I didn't understand the math when implemented your model in Mathematica (see the footnotes) and re-derived your result. I actually solved your recursive model in closed form.

Maybe -- just maybe -- you're the one who doesn't understand the math or what it means in reality?

André said...

"Mainstream: MMT has no math!
Jason Smith: MMT is hiding everything behind assumptions buried in mathematics!"

Brian, I guess that's progress, isn't it? I see it with good eyes hehe. But Jason was talking about Neil's post specifically, and not about MMT in general, I guess...

Jason, you offered four reasons, but the "intellectual dishonesty" one is your favourite, in your own words. Don't make it your favourite. That's a mistake everyone make, including here in the MMT community. People like this sort of conspiracy theory, but it's simply not true.

"Egmont -- still insane, I see."
Hahaha I laughed a lot about that

And Jason, unfortunately MMT people don't like math and they are not as brave as Neil. I wish they could do more math (for didactic purposes) but I guess they believe it could harm economic discussion instead of enhancing it. It's a shame...

Ignacio said...

There is nothing inherently wrong with recursion. Maybe is because economists are not used to deal with recursion? But in many disciplines and fields recursion is a common tool (including mathematics itself) and brings clarity many times.

Is probably the computer scientist (and is common in maths too) in me but I many times have an easier time expressing mathematical functions recursively than otherwise, for me is simplifying, but maybe is because I use it all the time and my brain is wired to understand recursion naturally.

To André point: people assume way too much. People see things in different things and express them in different ways. Is all in the mind of the beholder.

I find all this way more interesting and puzzling from the cognitive and psychological pov than economics tbh.

Neil Wilson said...

"Neil buddy -- when you start thinking a theoretical physicist and a macroeconomist didn't understand some math, it's time for a little introspection."

Is it. When anybody uses appeal to authority as a defence I get the impression they are flanneling.

You imparted a savings rate into the equations. Which isn't there either in my little set of equations or in reality. Why did you do that?

So it fitted your preconceptions.

You didn't understand what the maths was trying to do in my article - and you haven't the humility to admit it. Nor is your response even remotely helpful.

"I actually solved your recursive model in closed form."

Very good Jason. Go to the top of the class and give out the pencils.

So what? Are you expecting a medal or something?

"Maybe -- just maybe -- you're the one who doesn't understand the math or what it means in reality?"

Maybe, just maybe you've got completely the wrong end of the stick and misunderstood the purpose of what I was doing? Because I've noticed that you haven't suggested that anything I was doing was incorrect in the context in which it was written.

And why do you think a different solution to the mathematics, which is just a rewrite given you don't know the values of the parameters - particularly the time series of the savings amounts, means anything at all?

It's just argument for arguments sake - all because somebody wrote some maths down about the financial circuit.

Matt Franko said...

I think Brian and Neil have moved the ball forward this week....

Magpie said...

Jason Smith said...

Egmont -- still insane, I see.

Bow down, repent, and embrace the True Theory, as revealed by the prophet; convert, heathen, or you'll turn to dust. :-)

Magpie said...

Egmont,

Show him what you've got. No mercy.

Bob can help you with the "excerpting and bolding".

-------

Man, I love comment threads!

Bob said...

Battle of the Bloggers

AXEC / E.K-H said...

Brian Romanchuk

You say: “Mainstream: MMT has no math!”

When you drive a car you have velocity and acceleration/deceleration. The latter is the first derivative of the former. So, by driving a car you have the NATURAL math of calculus. It is just there, whether you have ever heard of calculus or not is irrelevant. Whether you actually measure the velocity is irrelevant. The math is just there.

The same holds for economics. The NATURAL math of economics is the elementary math of accounting. Whether you can do the math or whether you actually measure flows or stocks is irrelevant. The math is just there.

The problem with economists is (i) that they do not know what the NATURAL math of a given economic phenomenon is, (ii) that they grab a piece of math from the math department and apply it without deeper understanding. A familiar example is the application of calculus to utility maximization, which is downright idiotic because utility is a NONENTITY. And here you have it in a nutshell: “Walras approached Poincaré for his approval. ... But Poincaré was devoutly committed to applied mathematics and did not fail to notice that utility is a nonmeasurable magnitude.” (Porter)

Neither Walras nor the rest understood this and because they were not immediately fired for their manifest scientific incompetence they simply carried on.

The proper definition of mathiness is (i) not to understand what the natural math of an economic phenomenon is, and (ii), the formalization of nonentities (demand functions, production functions, IS functions, equilibrium, etcetera). Economists got it wrong on both counts from Jevons/Walras/Menger via Keynes to DSGE. With Jason Smith’s information economics mathematical incompetence has actually reached a maximum.#1

The natural math of macroeconomics is the elementary math of national accounting. Fact of the matter is that Keynes, Post Keynesians and MMT got it provable wrong.#2

Keynes’s two-liner is simply false: “Income = value of output = consumption + investment. Saving = income - consumption. Therefore saving = investment.” (GT, p. 63)

Neither pro- nor anti-Keynesians have realized in 80+ years that the two-liner is false according to the natural mathematics of accounting.#3 The snag is that stupid people do not understand the unassailable mathematical proof of their stupidity and this is why IS-LM and MMT and Post Keynesianism and Krugman and Jason Smith and all the other half-wits are still around.

Scientific self-governance or what has been reintroduced by Paul Romer as Feynman integrity simply does not work in economics. Economics is since 140+ years a cargo cult science and orthodox mathiness and heterodox antimath is only the most conspicuous symptom.

Egmont Kakarot-Handtke

#1 See ‘IS-LM ― a crash course for EconoPhysicists’
http://axecorg.blogspot.de/2017/04/is-lm-crash-course-for-econophysicists.html

#2 See ‘Where MMT got macro wrong
http://axecorg.blogspot.de/2017/04/where-mmt-got-macro-wrong.html

#3 See ‘The Common Error of Common Sense: An Essential Rectification of the Accounting Approach’
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2124415

Brian Romanchuk said...

Jason,

I have no problem with first-order conditions, the issue is "first-order conditions." if you cannot see the distinction, even after I illustrated it in plain English -- with a simple example -- you are not exactly broadcasting your expertise.

Have a nice day,
Brian R

Brian Romanchuk said...

E. K.-H.,

National accounting is the one thing that all schools of thought (roughly) agree on. If you have a different theory, and everyone looks at it and tells you are wrong, then perhaps you missed something. I read one of your articles, wrote a response on my site. I have forgotten what I said then; if you wish, write a comment on my blog that responds to my take. In other words, I am not going to go through your logic from scratch again.

I am not particularly a fan of utility, so I am not going to argue about that.

AXEC / E.K-H said...

Brian Romanchuk

So, we agree on these concrete results:

• The statement ‘MMT has no math’ is false.

• Every economic phenomenon has its natural math.

• The natural math of macro/MMT is accounting.

• MMT got the elementary math of accounting provable wrong.

• The error/mistake is the same as with Keynes.#1

• The problem of MMT is NOT mathiness but insufficient math competence.

• Because the formal foundations of MMT are false the whole analytical superstructure of MMT is false.#2

• MMT is refuted according to the scientific criteria of formal/material consistency.

• The further propagation of MMT violates scientific standards.

Egmont Kakarot-Handtke

#1 ‘How Keynes got macro wrong and Allais got it right’
http://axecorg.blogspot.de/2016/09/how-keynes-got-macro-wrong-and-allais.html

#2 ‘The final implosion of MMT’
http://axecorg.blogspot.de/2016/10/the-final-implosion-of-mmt.html

Brian Romanchuk said...

No.

Like I said, I am not going to debate your theories here.

AXEC / E.K-H said...

Brian Romanchuk

Thank you for the clarification. All others are invited to proceed to the future of economics. See

True macrofoundations: the reset of economics
https://mikenormaneconomics.blogspot.de/2017/05/jason-smith-reason-for-proliferation-of.html?showComment=1493971318738#c3385606989202999352

or here
http://axecorg.blogspot.de/2017/05/true-macrofoundations-reset-of-economics.html

Egmont Kakarot-Handtke

Bob said...

I did a search of Brian's blog with the tag "Egmont" and found a single entry:

http://www.bondeconomics.com/2016/11/fun-with-accounting-identities.html