I have received several E-mails over the last few weeks that suggest that the economics discipline is finally changing course to redress the major flaws in the curricula that is taught around the world and that perhaps Modern Monetary Theory (MMT) can take some credit for some of that. There has been a tendency for some time for those who are attracted to MMT to become somewhat celebratory, even to the point of declaring ‘victory’. This tendency is not limited to the MMT public who comment on social media and the like. My response is that we are probably further away from seeing fundamental change in the economics profession than perhaps where we were some years ago – after the GFC and in the early years of the pandemic (which continues). My answer reflects the incontestable fact that the make up of faculties within our higher education systems has not changed much, if at all, and the dominant publishing and grant awarding bodies still reflect that mainstream dominance. There is still a lot of work to be done and a lot of ‘funerals’ to attend (à la Max Planck)....
Summary: Nothing is going to change while the same clique remains in power and controls the educational and publishing process. Same in politics, although it is much more difficult to control the narrative that serves as an instrument of control than the narrative in terms of which the public understands economics. Heterodox economics has a long way to go in disrupting and eventually replacing this "Econ 101" narrative that firmly rules the collective mindset.
There is much more in this post than the title and lede paragraph would suggest.
Here is an example.
[Angus] Deaton then admits that “I have recently found myself changing my mind, a discomfiting process for someone who has been a practicing economist for more than half a century.”
How so?
Well,
1. He now says that the dominance of the “virtues of free, competitive markets” has meant that mainstream economics has ignored corporate power.
He wrote: “Without an analysis of power, it is hard to understand inequality or much else in modern capitalism.”
That is, without beginning with class conflict, an analyst has zero chance of gaining an understanding of the dynamics of capitalism, where capital seeks to influence outcomes in any way that advances their cause to retain their hegemony.
But if we introduced that into economic analysis there would be no mainstream elements worth retaining.
The dominance unit of analysis in mainstream economics is the individual.
Society is not considered.
Collectives are not considered.
Conflict is played down.
And when power does come up in mainstream economics the focus has been of trade unions as perverting the free workings of the labour contracting process.
Claims that mainstream economics is changing radically are far-fetched
Bill Mitchell | Professor in Economics and Director of the Centre of Full Employment and Equity (CofFEE), at University of Newcastle, NSW, Australia
This inquiry requires distinguishing two kinds of approach to economics.
ReplyDeleteThere is only one valid approach for economics: the scientific method. This leads to models that deal with facts. For example, MMT.
There is an approach for analyzing power: politics.
There is an approach for analyzing the abuse of power: justice.
This leads to models based on values rather than facts. As in science, value systems are subject to interpretation, which necessitates disagreement. Unlike theories, value judgements can't be falsified.
I don't want "economists" who extol the virtues of capitalism, socialism, etc. I don't want so-called experts pushing an agenda. I want facts. I want an objective foundation upon which politics, justice, activism and policy-making can be pursued. I accept that there will always be disagreement in those subsequent endeavors.
The scientific method is a 'for-purpose' approach. When used outside of that purpose, it contributes to the noise. Economic models serve the same purpose as lipstick on a pig. To normalize value systems as factual. To anyone paying attention, it is a vulgar agenda.
"There is only one valid approach for economics: the scientific method. This leads to models that deal with facts. For example, MMT."
ReplyDeleteAgree. And I think this is Bill's point too. Conventional economics doesn't deal with all the relevant facts.
Keynes pointed this out too, saying in effect that an equation doesn't work in an applied environment if the variables either do not include the relevant factors, or get some wrong.
Some of these relevant facts that conventional economics excludes or gets wrong are economic and others are social and political facts that have relevant economic influence.
Why is this not done? One reason is that doing it makes economic models mathematically intractable or difficult to formalize in the first place. There is a huge literature on this and some if comes from conventional economists too. Maybe things will change with LLMs but as with all digital matters, it depends on the construction of the algo. If relevant factors are not included it is still GIGO.
Another reason alleged is that economic models are designed for purpose but this purpose is influenced by class interests so that the model is rigged to provide support for desired outcomes that conform to class interest. That is, conventional economists operate under conflict of interest. This is an issue in many scientific endeavors, e.g., pharma.
In addition, as other scientists and philosophers of science have pointed out, economics lacks a framework for theory construction so it cannot resemble the natural sciences faithfully as conventional assume it does because it resembles the life sciences and social science much more that the natural sciences, and neither of these categories have frameworks like the natural sciences do.
This has been investigated and debated ad nauseam for decades at least and the discussion shows that conventional economics is pseudoscientific other than in very limited applications and contexts. This applies in particular to macro, and macro is used to formulate policy that affects the lives of huge numbers of people globally.
To illustrate this Lars Syll, IIRC, cited the example of a general telling his weather people that he needed a 30 day forecast to make some important decisions. They came back with the response that that they could only provide a scientific forecast for 10 days. The general's response was that he needed a 30 day forecast. What do you think the forecasters did?
Incorporating concepts like “power” or “class”‘isn’t Science…
ReplyDeleteSo you’re going to have to continue to use the Art Degree methodology for THAT part(s) of it…
“ the make up of faculties within our higher education systems”
ReplyDeleteIt’s not the people… it’s the methodology that the people are educated/trained under…
It’s like Bill is saying “well, if we could just get rid of the current people and just replace them with people that agree with me then all our problems go away”
It’s not going to work as then there will be new people appear that will disagree with him and the arguments start all over again…
STEM training doesn’t allow that to happen…
In addition, as other scientists and philosophers of science have pointed out, economics lacks a framework for theory construction so it cannot resemble the natural sciences faithfully as conventional assume it does because it resembles the life sciences and social science much more that the natural sciences, and neither of these categories have frameworks like the natural sciences do.
ReplyDeleteHence the obsession with mathematics.
Even the social sciences don't rely on math as a framework.
Because equations are not evidence of anything. They're descriptive.
Economists may as well use Shakespearean English to 'formalize' the stories they tell.
To illustrate this Lars Syll, IIRC, cited the example of a general telling his weather people that he needed a 30 day forecast to make some important decisions. They came back with the response that that they could only provide a scientific forecast for 10 days. The general's response was that he needed a 30 day forecast. What do you think the forecasters did?
They should reiterate that his request is beyond the scope of current analysis. Meteorology is a field that mitigates risks to economic activity. In the real world, inaccurate 1-day forecasts are not tolerated; 3-day forecasts are a work in progress, and 5-day outlooks are done for entertainment purposes. A 30 day 'forecast' will amount to 'above or below normal temperatures etc.' Good luck fighting a war on that amount of information.
If economic models were of any value, supercomputers would have time allocated towards their calculation, as is the case with meteorological models.
ReplyDelete
ReplyDeleteSo you’re going to have to continue to use the Art Degree methodology for THAT part(s) of it…"
That is right, Matt, although there is work being done on quantifying those concepts. See Turchin's work in cliodynamics.
This condition undercuts economics as a science, unless one admits that social sciences are science, which many do not wish to do, especially economists that think that they are doing natural science. The closest the economics can get to science is social science. It is not even as close as life science with its basis in biophysics and biochemistry.
Conventional economists struggle with that and many even refuse to debate economic methodology because they assert that "the methodological debate has already been settled" in their favor, generating an orthodoxy that is really dogmatism. So they continue on their way. That would be OK if there were not perverse effects but there are, e.g., though the use of conventional economics in policy making. As a result of shortcomings, the policy prescriptions they generate get put into policy that is guaranteed to fail. How many times has austerity been prescribed with disastrous results. Even more basic, they still haven't figured out why the cb raising the interest rate is not having the effect they expect. They can't even figure out the easy stuff.
The hard problems? Forget it. Their answer is that these are excluded from the model. Remember when the queen asked her economists why they missed the onset of the GFC. That was their answer. It was a "shock" and shocks are surprises in that the conditions that result in them lie beyond the model assumptions and so they cannot be foreseen.
Not only that, conventional economists simply assume a lot of things called into question by the social and psychological sciences. Heterodox economists point out that many of these assumptions, which are assumed without examination with respect to fact, are unrealistic, which accounts for why the models built on them are not representative and fail.
Furthermore, humanity is not yet at the stage that every issue can be solved scientifically. Yet, many of these issues are vital for individuals and societies, like medicine wrt individuals and policy for societies. Even if MMT were accepted by the profession overnight, there would still be policy questions that are not decidable based on fact since societies are complex adaptive systems subject to emergence.
As far as individuals, if one lives to be over 80, one finds that evidence-based medicine has ended in one's case, there being little evidence gathered and few studies done wrt to the group that one finds oneself in. So one becomes dependent instead on the "art" of medicine — and luck. At this point some people become their own physician by shopping for an opinion that meshes with their confirmation bias.
If economic models were of any value, supercomputers would have time allocated towards their calculation, as is the case with meteorological models.
ReplyDeleteYes, and experiments with LLMs are being explored in many fields now. This is still at the stage of infancy.
Conventional economists do not undertake it with their tools, since formalization (Platonic") is a strong methodological assumption. So models have to be tractable. That's just not possible other than for the simple stuff that is not all that useful for much decision making.
STEM training doesn’t allow that to happen…
ReplyDeleteSo how do you convince engineers to get PhDs in economics, especially since they have to pass in conventional economics, which any engineer should recognize as inadequate pretty quickly.
Actually, academic economists are well trained in M, but not so much in S, E, or T. That's why formalization — which is Platonic rather than Aristotelian, so speak — is a foundational requirement in economic modeling as it is in the hard sciences.
Yet, they object to incorporating accounting-based models even though accounting is organized formally. Then Treasury secretary Larry Summers reportedly said to Warren Mosler when Warren brought reserve accounting into the conversation, "I don't understand reserve accounting."
Here is an interesting MMT-friendly article on the three body problem in economics. It's an interesting analogy.
Bruce Dickson, Economics is in transition between “two-body problem” and “three-body problem” thinking
“ So how do you convince engineers to get PhDs in economics”
ReplyDeleteThey don’t have to they can just keep working with their materials trying to increase efficiency and effectiveness of our material systems… increasing mpg, thermal efficiency, software development process etc..
Economics can then just examine the financial and real results of any policy to see if it is working to achieve the goals established in the laws ..
Like right now Biden Art Degree economists are trying to generally reduce their figure of speech “inflation” by imposing the higher risk free rate and reducing FRS reserve balances … is this working ?
Are the regressive results ideal? Or in conflict with other lawful mandates?
There are plenty of analysis for each to do but they just have to stay in their own lanes..
Here it works both ways:
ReplyDeletehttps://x.com/libertarianduty/status/1767226287150526978?s=46&t=4xUfv9EMbAb_JJs9EMjKYw
Massie (allegedly MIT EE can you even believe it?!?! ) not understanding Finance and Accounting abstractions and thinks THEY ARE REAL!
I would bet the farm he never had any rigorous training in Finance and Accounting Science … none…
So he should just stay out of it..
Yes, and experiments with LLMs are being explored in many fields now. This is still at the stage of infancy.
ReplyDeleteLarge Language Models parrot whatever is fed into them. This is what most people do.
It's a frivolous waste of computer resources for something that can only serve as a tool.
Zero decision making power, 100% circular inference. But it's seen as profitable, so none of that matters.
As far as individuals, if one lives to be over 80, one finds that evidence-based medicine has ended in one's case, there being little evidence gathered and few studies done wrt to the group that one finds oneself in. So one becomes dependent instead on the "art" of medicine — and luck. At this point some people become their own physician by shopping for an opinion that meshes with their confirmation bias.
ReplyDeleteI had a 2nd cousin who died at the age of 23 of an abdominal tumor. I had three uncles who died in their 60s of cancer. My parents passed away at the age of 72 and 73 of cancer and complications from diabetes. When it comes to individual physiology or pathology, medicine has very little to do with how long you live.
The world would be a better place if medicine and economics were limited in scope to what is factual or evidence-based. The pronouncements of so-called experts have to be curtailed.
So the next time a policy is made, the policymaker is forced to admit that their decision is not based on facts or evidence. That it is not backed by science.
This reduces the damage that agenda driven policies do. People will have reason not to tolerate policies that don't have scientific backing.
“ But it's seen as profitable, so none of that matters.”
ReplyDeleteIt’s profitable to the ones selling the picks and the shovels…
“ where capital seeks to influence outcomes”
ReplyDeleteAnthropomorphism..