Monday, May 25, 2015

Economics, Mathiness, and Sci-Jacking

A new schism has arisen in the economics priesthood.

Several years back, Stephen Colbert coined the term "truthiness" - assertions of specious facts which were not exactly "true" in some universal, falsifiable sense, but felt true because they reinforced certain prevailing biases of the targeted audience. Wikipedia defines it this way:
Truthiness is a quality characterizing a "truth" that a person making an argument or assertion claims to know intuitively "from the gut" or because it "feels right" without regard to evidence, logic, intellectual examination, or facts.
Now a prominent economist has accused some other economists of "mathiness," which is to say that a lot of math is used to disguise sloppy thinking and unproven assumptions. It is hiding certain assumptions not in evidence behind an impenetrable wall of equations to make yourself look smart and promote your ideas and career. It is also used to advance certain political biases in under the guise of "mathematical proof."
"In choosing to present the theory in less detail, they too may have responded to the expectations in the new equilibrium: empirical work is science; theory is entertainment. Presenting a model is like doing a card trick. Everybody knows that there will be some sleight of hand. There is no intent to deceive because no one takes it seriously. Perhaps our norms will soon be like those in professional magic; it will be impolite, perhaps even an ethical breach, to reveal how someone’s trick works."

Romer here is not saying that theory should be entertainment but instead that it has become entertainment. This is the notion of mathiness. What researchers he cites are doing, Romer argues, is dressing up an idea they want to convey in mathematics and presenting that as a logical theoretical insight...Romer’s first thesis is that mathiness is becoming more pervasive in economics...But there is a second thesis to Romer and that is that mathiness is a device being used by academics to further a political agenda rather than purely the pursuit of science...
Mathiness: A Guide for the Perplexed (Digitopoly)

Here's another good summary:
Once upon a time economists made their arguments in long, discursive, often contradictory books about pin factories and newspaper beauty contests. Verbally oriented people like me tend to extol those days, but to most modern economists they were the dark ages.

In the 1940s Paul Samuelson of the Massachusetts Institute of Technology brought enlightenment, in the form of elegant mathematical treatments of the major concepts in economics. Most of these ideas were inherently mathematical anyway, he argued in the introduction to his “Foundations of Economic Analysis,” first published in 1947, which meant that trying to express them in narrative form involved “mental gymnastics of a peculiarly depraved type.”

Samuelson’s approach gave the discipline a, well, discipline that it had previously lacked, and enabled economics to make great leaps in coherence and rigor. It also made the field incomprehensible to laypeople, but that turned out to be more a feature than a bug. Economists were seen as possessing unique scientific knowledge, and came to play increasingly prominent roles in public life in the U.S. and elsewhere. Samuelson’s economics-professor nephew even became Treasury secretary.

There are some obvious limits to this approach. In his entertaining and enlightening new book, “Misbehaving,” University of Chicago behavioral economist Richard Thaler documents case after case of “theory-induced blindness” in which economists ignored interesting and important real-world phenomena because they didn’t accord with the dominant mathematical models. 
What's Wrong With 'Mathiness' in Economics? (Bloomberg)

I've talked about how in the past, economics was once primarily institutional economics - which looked at the interaction of various social institutions to understand how the economy worked. Economics was also once called political economy - the understanding that there is nothing "natural" about markets and how they are constituted, rather it is based upon political choices; you cannot separate the workings of the economy from political decisions.

That particular approach went out of favor because of a guy named Karl Marx, who wrote extensively about how the politics and institutions of the time based around "capital" (the title of his book) were designed to reward the wealthy and screw over the poor.

What to do? Simple. Treat markets as naturally arising from of humans' inherent desire to to "truck, barter and exchange," and markets as just a reflection of human social instincts, as natural to human society as a troop to baboons, a pride to lions, or a school to fish. People were assumed to be equals making mutually-beneficial rational exchanges in a non-coercive markets headed toward equilibrium. Making certain assumptions (which are never questioned), you can describe the entire world mathematically. You can justify pretty much anything. And, more importantly, you can predict outcomes!

Washed away in this mathematical approach was any notion of political decisions, inequality, class struggle, debt, and history. You no longer needed to explain anything in terms of history, it could just be explained through extremely complicated mathematical equations that outsiders never questioned. Entry into the field became like physics - limited to math geniuses, and an jargon impenetrable to outsiders was adopted. Economists became advisers to governments rather than merely academics.

The biggest change was that once upon a time economists used to describe the economy as it was actually observed (which is what Adam Smith did). Economics was a relatively obscure discipline confined to a few "worldly philosophers."  But after the change, economists would now dictate economic management and decision making. Economics went from a relatively obscure discipline to the most popular major in American colleges today. It has indoctrinated the entire society while its inherent biases go unchallenged. Its assumptions have been internalized to such a degree that we can no longer think outside of them. It made it so that there really was no alternative.

Because of math, economists were assumed to have predictive power. Economists told governments what they could, and could not do, and usually what they could not do was anything that would help workers or redistribute income or ownership. They played up this predictive power, which is why the field was so humiliated by the economic breakdown of 2007-2008. Of course, the truth is, economists cannot predict anything.

Economics was transformed into Walrasian economics.
In 1874 and 1877 Walras published Elements of Pure Economics, a work that led him to be considered the father of the general equilibrium theory. The problem that Walras set out to solve was one presented by Cournot, that even though it could be demonstrated that prices would equate supply and demand to clear individual markets, it was unclear that an equilibrium existed for all markets simultaneously.

Walras constructed his basic theory of general equilibrium by beginning with simple equations and then increasing the complexity in the next equations. He began with a two-person bartering system, then moved on to the derivation of downward-sloping consumer demands. Next he moved on to exchanges involving multiple parties, and finally ended with credit and money.

Walras created a system of simultaneous equations in an attempt to solve Cournot's problem "(which supposedly Walras at first thought was complete merely because the number of equations equalled the number of unknowns)"
The crucial step in the argument was Walras' Law which states that any particular market must be in equilibrium, if all other markets in an economy are also in equilibrium. Walras' Law hinges on the mathematical notion that excess market demands (or, inversely, excess market supplies) must sum to zero. This means that, in an economy with n markets, it is sufficient to solve n-1 simultaneous equations for market clearing. Taking one good as the numéraire in terms of which prices are specified, the economy has n-1 prices that can be determined by the equation, so an equilibrium should exist. 
http://en.wikipedia.org/wiki/L%C3%A9on_Walras#General_equilibrium_theory

After Walras, all sorts of new math was added to economics - Coase theorems, Bayesian probabilities, Cobb-Douglas functions, Nash equilibriums, and so on (it is sadly ironic that Dr. Nash died in a car crash this weekend).

This good article - Mathiness is Next to Growthiness, quotes Nicolas Georgescu-Roegen, who was also one of the first economists to apply the concept of entropy to economics:
     "According to the temper that has prevailed for some time now in the social sciences, but especially in economics, the contributions that deserve the highest praise are those using a heavy mathematical armamentarium; the heavier and the more esoteric, the more worthy of praise. Protests against this situation have not failed to be made sufficiently often to have deserved attention. What is more, protests of this kind were made not only by "verbal" economists, such as Thorstein Veblen and Frank H. Knight, but also by some who were well familiar with the mathematical tool, for example, Alfred Marshall, Knut Wicksell, and Lord Keynes. Knight lamented that there are many members of the economic profession who are "mathematicians first and economists afterwards." The situation since Knight’s time has become much worse. There are endeavors that now pass for the most desirable kind of economic contributions although they are just plain mathematical exercises, not only without any economic substance but also without mathematical value. Their authors are not something first and something else afterwards; they are neither mathematicians nor economists. How dangerous is the infatuation with pure mathematical symbolism is proved by the fact that voices from the circle of natural scientists have also often denounced it. ...

    The fundamental reason why we cannot do without dialectical concepts is that actuality, at least as seen by the human mind, continuously changes qualitatively. …

    The most we can expect from an arithmomorphic model is to depict pure growth, or rather pure quantitative variations of qualitatively different but self-identical elements.
Now, you can guess where I stand on this issue. Romer's critique is limited to a few specific theories, not to the profession overall. but I would contend that 'mathiness' is not confined to a few bad apples or ideas, but is the core of the entire pseudoscience of economics.

The entire profession is based on a fraud. Human behavior is inherently unpredictable. Even if you wash away all the fundamental ways in which the economy is constructed (divided between owners and workers), you still cannot predict things Economics has two things at its heart 1) Human behavior, which is inherently unpredictable, and 2.) Stuff, which is described by certain physical laws such as thermodynamics, chemistry, geology, etc. Economics studies money flows and claims that it can tell all it needs to know about society from that.

In an extraordinary article, prominent economist and blogger Noah Smith has come perilously close to admitting that the entire field is a fraud!
Economics has a lot of math. In no other subject except mathematics itself will you see so many proofs and theorems. Some branches of econ, such as game theory, could legitimately be housed in university math departments. But even in fields such as macroeconomics, which ostensibly deal with real-world phenomena, math is central to everything that economists do.
But the way math is used in macroeconomics isn't the same as in the hard sciences. This isn't something that most non-economists realize, so I think I had better explain.
In physics, if you write down an equation, you expect the variables to correspond to real things that you can measure and predict. For example, if you write down an equation for the path of a cannonball, you would expect that equation to let you know how to aim your cannon in order to actually hit something. This close correspondence between math and reality is what allowed us to land spacecraft on the moon. It also allowed engineers to build your computer, your car and most of the things you use.
Some economics is the same way, especially in microeconomics, or the study of individuals' actions -- you can predict which kind of auction will fetch the highest prices, or how many people will ride a train. But macroeconomics, which looks at the broad economy, is different. Most of the equations in the models aren’t supported by evidence.
For example, something called the consumption euler equation is at the core of almost every modern macroeconomic model. It specifies a relationship between consumption growth and interest rates. But when researchers looked at real data on consumption growth and interest rates, they found that the equation gives exactly the wrong predictions! Yet it continues to be used as the core of almost every macro model.
If you read the macro literature, you see that almost every famous, respected paper is chock full of these sort of equations that don’t match reality. This paper predicts that everyone will hold the same amount of cash. This paper predicts that people buy financial assets that only pay off if people are able to change the wage that they ask to receive. These and many other mathematical statements don't remotely correspond to observable reality, nor do they have any evidence in support of them. Yet they are thrown into big multi-equation models, and those models are then judged only on how well they fit the aggregate data (which usually isn't very well).
That whole approach would never fly in engineering. Engineering is something you expect to work. But macroeconomists often treat their models as simply ways, in the words of David Andolfatto, vice president of the Federal Reserve Bank of St. Louis, to “organize our thinking” about the world. In other words, macroeconomists use math to make their thoughts concrete, to persuade others, and to check the internal consistency of their (sometimes preposterous) ideas, but not to actually predict things in the real world.
Back to Romer’s complaint. He singles out Lucas, Prescott and a few others for having tenuous or sloppy links between mathematical elements and the real world. But from what I can see, such tenuous and sloppy links are the rule in macro fields. Romer says that I am “jaded” for saying that, and that it was bad apples like Lucas and Prescott who soured me on macroeconomic theory. Well, he’s right that I’m jaded, and he’s right that it was learning about models by Prescott that I first became jaded. Romer, you got me.
But when I looked beyond those models, to older or newer models, I found what seemed to be only a difference in degree, not in kind. Macroeconomic theory is chock full of mathiness. It’s not just Lucas and Prescott, it’s the whole scientific culture of the field.
How 'Mathiness' Made Me Jaded About Economics (Bloomberg)

And speaking of Noah Smith, he has coined a neologism of his own - "sci-jacking." This means to take some difference of opinion and claim that it is settled because of "science." He defines it this way: "[the] attempt to co-opt a field of scientific inquiry for political purposes." 
For a long time I've been annoyed by the practice of political movements trying to co-opt branches of science for their own ends. ...In social science, you have lots of implications for policy, and so you get a lot more political movements trying to hijack branches of social science. Let's call this practice "sci-jacking." (I like this term for a number of reasons.)...Economics, of course, is a prime target for sci-jackers. It deals with money, and money is what most people want out of policy. It is more directly connected to policy than anthropology or even sociology.
And once again, I contend that this describes the entire field of economics! In my response to the article, I argued that economics cannot be sci-jacked because the entire enterprise IS sci-jacking. It is an attempt to to pursue a specific political agenda under the guise of "science." In a comment to the article above, a commenter writes:
As a new economics student with a previous education in philosophy and then engineering, let me inform you that economics is a joke discipline. Your formulas often don't even have a time variable, the models are oversimplified and then assumed to take the place of reality, and your first year textbooks are childishly veiled Libertarian propaganda without even any references given to original papers.
I recently re-listened to Tom O'Brien's interview with Charles Hall, author of "Energy and the Wealth of Nations," and he said something very similar - his students who study energy and ecology with him and then go on take mainstream economics courses inevitably tell him that these courses are nothing more than propaganda based on ridiculous assumptions and fairy tales.

The best description I've yet read of the underlying hidden assumption of economics comes from numers 10 and 11 of Eric Zencey's Theses on Sustainability:
[10] THAT A DISTINCTION can usefully be drawn between wants and needs seems obvious. Mainstream economics, however, refuses to countenance such a distinction. (Marxist economics does, which, from the viewpoint of an ecologically enlightened economics, is one of the few ways in which it is distinguishable from its neoclassical alternative.) The work of Wilfred Pareto was crucial to this refusal. His contribution to economic theory marks a turning point in the evolution (some would say devolution) of nineteenth-century political economy into the highly mathematized discipline of economics as we know it today. Pareto’s novel idea: because satisfactions and pleasures are subjective — because no one among us can say with certainty, “I like ice cream more than you do” — there is no rational way to compare the degree of pleasure that different people will gain by satisfying desires. All we can do is assert that if an economic arrangement satisfies more human wants, it is objectively better than an arrangement that satisfies fewer human wants. This seems commonsensical until we unpack that caveat “all we can do.” An economic arrangement achieves Pareto Optimality if, within it, no one can be made better off (in his own estimation) without making someone else worse off (in her own estimation). Economic science, in its desire to be grounded on rational, objective principles, thus concludes that were we to take a dollar from a billionaire and give it to a starving man to buy food, we can’t know for certain that we have improved the sum total of human satisfaction in the world. For all we know, the billionaire might derive as much pleasure from the expenditure of his billionth dollar as would a starving man spending a dollar on food. All we can do — all! — is promote the growth of income; and if we care about that starving man, we must work to produce two dollars’ worth of goods where before there was only one, so that both the billionaire and the starving man can satisfy their wants.
[11] THUS WAS neoclassical economic theory, putatively value-free and scientific, made structurally dependent on a commitment to infinite economic growth, a value-laden, unscientific, demonstrably unsustainable commitment if ever there was one.
Note that Walrasian economics, mentioned above, is also value-free. Wikipedia again:
Léon Walras provides an amoral definition of economic utility...In Walras's theory of value, value is thus totally independent of the common meaning of value or utility. Production or increase of "value", in fact monetary amounts as for instance measured in profit or GNP, is independent of notions like a just or fair society or a better world. This is a fundament [sic] of modern "neoliberalism" or neoclassical economics.
Those biases, along with other ones like productivism, materialism, hedonism, and many others, are all hidden beneath the surface and buried in the math in economics' pretension to be a "pure" science. If that isn't sci-jacking, I don't know what is. As the above commenter says, it is "thinly disguised libertarian propaganda," and note that includes the entire field of economics, not just the ostensibly "libertarian "variety (which is even more ridiculous). Private ownership is superior; competition is efficient; a small state is best; markets when left alone head towards equilibrium as long as there is no "distortion" from government (i.e. protecting workers or the environment), and there should be no restrictions on accumulation by the wealthy whose wealth will inevitably "trickle-down" to the masses. There can be no gluts; monopolies are impossible; supply produces it's own demand; exchanges are always voluntary, as are work contracts; people are rational utility maximizers, employment is always unlimited. Income is determined entirely by one's marginal productivity. The planet is infinite; natural resources are provided free of charge; goods are infinitely substitutable; energy and debt are irrelevant, waste is ignored; externalities are exceptions, the list is almost endless.

Hopefully, these criticisms mean that the economics that justifies our current arrangements is starting to crack and fall away, and we will finally be able to start thinking outside the current system for better alternatives. It is very much analogous to when the absolute authority of the Catholic Church began to be questioned which led to all sorts of revolutions in human thought and social organization.

8 comments:

  1. Good commentary on mathiness here as well, from a guy who i think really has his head screwed on right.

    http://www.bondeconomics.com/2015/05/mathiness.html?m=1

    ReplyDelete
    Replies
    1. Especially this part:

      ...I have spent little time looking at growth theory, since it appears self-evidently useless to me. The main recommendations seem to be: "grow your population", and "be more productive, somehow!".

      The long run is made up of short runs. At present, all of the major countries have put in place policy settings that leaves a large portion of their working population in unemployment or underemployment. This obviously constrains growth, yet I have not seen any part of "Growth Theory" that addresses this problem.


      BTW, unless I'm wrong, there is no growth theory that takes into account the role of energy, debt or entropy. I believe Nate Hagens has made this point in his speeches.

      Delete
    2. So he's kind of half right, I guess? I've been doing a fair bit of reading about growth theory lately so I kind of feel able to comment with some knowledge here.

      The mainstream view of growth, both Brian and Nate are right, insofar as it descends from the Swann-Solow model, and it essentially just considers two factors, of labour and capital. You you try to account for growth using just these two factors, they fail miserably, you have to introduce a factor representing "technology", which is essentially treated as manna from heaven. The whole thing is rife with tautology and is deeply unphysical, and frankly does no better job than a simple exponential fit.

      Other people have recently been trying to do a better job of this, and are trying to make growht theory more physical and include energy and other biophysical measures. Ayres and Warr's 2009 book "The economic growth engine" is probably the current gold standard, certainly from an empirical point of view. And ranier kummel's recent book "The second law of economics" is also meant to be good, though I've not read it yet. The following paper is a pretty good summary of the Ayres/Warr work: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2328101

      Another interseting guy is Blair Fix, who's currently a phd student, but stuck out a book last year called "rethinking growth from a biophysical perspective". You can find a preprint pdf here: ( http://bnarchives.yorku.ca/426/2/20150000_fix_rethinking_economic_growth_theory_preprint.pdf ). It's a very itneresting book, and it covers many things including the problems with current growth theory, issues with production functions, issues with using GDP as a measure of the size of a physical system and much more besides. He's got an interesting mix of biophysical, post keynesian/mmt and marxist views, and the book is really fascinating reading. He seems to think that we pretty much need to demolish most of the current framework underpinning growth accounting, which probably isn't a bad idea really.

      Delete
    3. Interestingly, that information feeds right into what I’ve been trying to put together into a longer format. It’s based somewhat on Gregory Clark’s “A Farewell to Alms,” which describes the dynamics of a Malthus-Ricardo economy. Increasing population growth eats up any gains in living standards because the marginal contribution of each new mouth to feed necessarily declines because adding labor while land and capital are fixed gives diminishing returns. Gradually, living standards always return to what he calls the “subsistence income.” Clark contends that this means that there was no significant gain in living standards between the Neolithic and Industrial Revolutions, and that were actually better off as hunter-gatherers. He is silent on what determines the subsistence income for various societies, but we know what it is – energy capture per capita (White’s Law).

      In the Industrial Revolution things change, and Clark sees this through the traditional economic lens of land, labor and capital. He has an excellent dismissal of the “institutionalist” school of economics which I agree with, basically pointing out that the conditions institutionalists claim for the IR (which he points out are basically the Washington Consensus – low taxes, small gov’t, no trade barriers, etc.) existed at various points throughout history, including Medieval England. Instead he claims a pure Social Darwinian explanation - the virtuous and hard-working rich had more surviving offspring than the lazy, shiftless poor, thus passing on the genetic propensity to work, save money and invest rather than faff off and drink all day, and thus capitalism is born. I swear I am not making this up! This explanation has been enthusiastically embraced by the Libertarian right, who use it as an argument to make sure the poor die off as swiftly as possible for the good of the economy and to further the genetic progress of the human race (seriously).

      http://press.princeton.edu/chapters/s8461.pdf

      Clark develops something called the “technology schedule” basically arguing that better technology, along with enhanced humans, made the IR possible. But as I’ve pointed out, what is true for institutions is true for technology - it did not start being invented in 1776. What did happen? Of course we know it’s discovering the energy in half a billion years of stored sunlight. In FTA., there is no mention of energy, or even a single mention of the steam engine! The IR is purely a function of genetics and technology, in Clark’s view. But if energy is the cause, with the peaking of energy sources, the Malthus-Ricardo economy will come roaring back, rendering all of today’s economic dogma not only worthless, but counterproductive. The dogma of today’s economics is the opposite of Malthus – that growth leads to higher living standards, and that the Malthusian economy has been left behind forever. I don’t think that’s true, and listening to the economists is dooming the human race to misery and destitution.

      Delete
    4. The amount of energy relative to the population increased a hundred-fold essentially overnight. In the interim, we’ve behaved exactly as Malthus said we would – breeding up to very limits of this energy release (7 billion and counting). But now that the surplus is used up, we can no longer look forward to higher living standards. Technology is now being used mainly to keep the status quo going (e.g. desalinization, electric cars, new antibiotics), or make do with less and/or intensification.

      Since today’s economic dogma is used to justify the opulence and wealth of elites, they cannot acknowledge this truth. We’re already seeing the living stands fall for most people – we’re told we must eat less meat, chow down on insects, take shorter showers, retire later. etc. Our goods are shoddier, our food is worse, our environment is polluted. There is a precedent for this – the move from hunter-gathers to farmers. People on the top of the pyramid got much better off, while the 99 percent replaced lives of leisure and freedom with the shackles of serfdom, slavery, malnutrition, endless toil, disease, political subjugation, etc.(which Clark himself points out). If I’m right, future growth will have the exact same effect.

      Since “technology” is ephemeral and unlimited, if you use that to explain the IR, you just posit that technology will just increase forever and solve all problems. This is economic dogma. But really this exposes economic “science” as basically just a modern cargo cult.

      Delete
  2. I think the fact that an obvious ideologue like Alan Greenspan could be Fed chairman for almost 20 years tells you all you need to know about mainstream academic economics.

    ReplyDelete
    Replies
    1. Actually the Fed chairman is appointed by the president, so that tells you nothing about mainstream economists.

      Delete

Note: Only a member of this blog may post a comment.