How Effective Is Economic Theory?
nationalaffairs.com
nationalaffairs.com
This is the problem: what is the alternative? The pet theory of a madman or sociopath?
The difficult truth is that everyone has an economic theory, whether they admit it or not. The question is, whose theories do we want to pay attention to? People who spend their lives arguing and thinking about it, and trying to find some evidence in support? But who might be a bit isolated from reality? People who work close to the phenomena being explained, but who don't really rigorously explicate or defend any of their ideas, who haven't had to put them on the table, so to speak? Or have massive, critical financial conflicts of interest?
There's also the difficult problem of identifying when any theories have really been tested well. It's not like we can just run randomized controlled designs on whole civilizations--at least, not most of the time for major policies.
I guess I really don't see the alternative to encouraging the standard academic approach to economics, opening it up to public criticism and discussion, and maybe trying them out when it's ethical and feasible.
Sure, lots about classical economic theory is really ridiculous, and the source of a lot of problems, but then you change those things and then move on. It's not any different from physics theory or chemistry theory in that regard.
Regarding economics, there are some fundamental assumptions that are often criticized.
For instance, in the mid 1700s, Charles Townsend asserted the fundamental laws of economics were these: 1,) Man's want, wishes, needs and desires are infinite. 2.) Resources are finite.
But numerous anthropologists have criticized these assumptions. Margaret Mead didn't see the tendency towards infinite consumption among the tribes she studied. And see Marshall Sahlins book Stone Age Economics -- he doesn't see any evidence of infinite desires among the tribes he studies. Thus it appears Townsend's assumptions are fit only for a society in the middle of the Industrial Revolution. So his "laws" are not "laws" in the same way that Issac Newton could describe "Laws Of Motion". Rather, Townsend's laws were a convenient assumption for a certain kind of society, but then the question comes up constantly, are we still the same kind of society?
Another example, Joseph Schumpeter starts one of his books (I think Das wesen und der hauptinhalt der theoretischen nationalökonomie, 1908) "The social process is one indivisble whole! However, to make useful remarks regarding production and consumption, it is reasonable to pluck from experience those transactions that are altogether economic in nature."
There are many people who strongly agree with the first sentence and therefore they can not agree with the second sentence.
I could go on, but I don't have the time.
If you go looking, you'll find many examples. In physics, no one doubts that matter and energy are real things, but in economics, it is the most fundamental assumptions of the discipline that have always been under attack.
In spite of such major deficiencies, economic science was very lucky and promoted the free market, which was in many ways an amazing success. As a consequence economists got the mistaken idea their science was basically sound, and never set about thinking out its basic assumptions.
That's the thing, it has proven really, really hard to "change those things". I'm looking at the latest (still current?) economic/financial crisis. Lots of "economic theory tells us this should never happen"-things did in fact happen, and consistently so (there was not just a random event that you could blame in the theory not being 100% there). It matters because policy decisions are still taken based on theories that have proved themselves wrong, or at least those theories were not able to "explain" (for a lack of a better word) how our present-day economy works. Those policy decisions affect (some of them negatively) the lives of millions, even hundreds of millions.
At least when a theoretical physicist is wrong in his/her assessment nothing that bad can happen to the outside world, in the great scheme of things. But when an economist is wrong, but his theory is nevertheless taken into consideration and acted upon, then the damage can be quite substantial, its effect measurable in decades.
I don't know what needs to be done. I'm not ditching economic theory entirely. For example the Chinese authorities' current push to stop Chinese billionaires' money moving outside the country reminded me of Jean-Baptiste Say's explanation of how no Government can put a stop to the flow of currency that wants to escape a certain jurisdiction, a phenomenon that he wrote about ~200 years ago and which still seems to be in effect, almost like a "law" of economics. And there are still other "basic" economic truths/laws that have seemed to keep their relevance over the centuries. But, AFAIK, almost all of those "laws" were common sense, so to speak, they didn't involve mathematiac equations with second derivatives and the like (for example there's no "second derivative" equation with which to model the "lack of trust" in financial transactions, which once it sets in you can bet will bring any financial market down pretty fast).
So what I'm trying to say in a convoluted way is that economics should return to basics, ditch most of the mathematics with which it has become enamored and try to be a little more on the "social science"-side of things: more observation of humans and their acts, less abstract computations.
An interesting article (that was posted to HN a while back) that discusses this is "How economists rode maths to become our era’s astrologers": https://aeon.co/essays/how-economists-rode-maths-to-become-o...
Though it's been posted a few times, there hasn't been that much discussion on HN: https://hn.algolia.com/?query=https:%2F%2Faeon.co%2Fessays%2...
http://econlog.econlib.org/archives/2015/09/how_the_subprim....
But the point is the economists should have seen that was happening, so we could think about the government stopping it, or at least to put the public on guard. Instead the economics profession, with a few exceptions like Dean Baker, told the world that things were going just fine.
I mean, what is the point of even having an economics profession if it can't help us make intelligent decisions on economic matters?
A large number of economists (in academia, public institutions, and private finance firms), all across the ideological spectrum (Austrians, Keynesians, and every other flavor) did see it happening.
People didn't respond to then, probably because the existence of a bubble isn't a problem, as long as you can delude yourself to thinking you'll be able to time the market so you won't be holding the ball when it pops.
> Instead the economics profession, with a few exceptions like Dean Baker, told the world that things were going just fine.
Baker was far from the only prominent economist pointing to a bubble. In fact, economists were warning about the housing bubble and the fact that it would have to burst before the first dot-com bubble burst (or even expanded) back as far as the mid-1990s. That may actually be the real problem: the warnings had been around for so long no one took them seriously any more; as controversial as identifying a bubble can be, it's a lot easier to identify it than time when it will pop ,and the longer it is pointed to without popping, the more likely people are to convince themselves it's just a permanent feature of the market and not a bubble that will pop.
In a strange way, that's almost step down: A strong correlation between what economists say and what happens -- even a negative correlation -- suggests that theory is somehow catching up to reality.
In contrast, a weaker correlation -- even if positive -- implies that there's still a lot more problems to shake out.
What economic theory said which things should never happen exactly?
I'm not aware of an economic theory that says mispricing should be impossible, especially for a price set in advance.
> Trillions of euro-denominated deposits stood at bellow 0% for more than a year.
As far as I can tell this is referring to the an ECB policy for deposits of other banks to the ECB. This is not a market price; the ECB is actually trying to discourage its bank deposits. No contradiction of economic theory here.
> Real wages are continuously going down even though the unemployment figures are pretty damn good.
I'm not aware of real wages going down "continuously". As far as I can tell it's remarkably flat. This, however, is closer to something that actually doesn't seem to make much sense at first glance according to economic theory.
FYI I tried googling all of these things using terms you used and mostly got dubious zerohedge articles near the top.
That ECB policy was indeed a price, don't know what they want to "discourage" or not, fact is that if you wanted to deposit money to the ECB you had to pay them for the privilege. AFAIK since capitalism set in properly (about 200 years ago) it has almost been the case that you were supposed to receive money as interest when depositing it somewhere (to a king's vault, to ECB, it didn't matter).
Am on mobile, too lazy to look for the real-wage charts. I had just seen one in the FT detailing its evolution for the UK since 2005 or so, with only 3 years out of those 11-12 seeing real wage increases. I used the same source for commenting on Argentina's debt, i.e. last Friday's Financial Times. I used to read zerohedge from time to time, but I don't like their layout, I've mostly stuck with the FT and the Economist for my economy and financial-related info (there was also an interesting economics-related blog under the economist.com domain).
To not commit an entire society to a single economic theory. Which means getting rid of a lot of the centralized control mechanisms that currently require just such a commitment.
> whose theories do we want to pay attention to?
Since no one has a theory that is good enough to make useful predictions, as far as public policy is concerned, the answer is "none of them".
> There's also the difficult problem of identifying when any theories have really been tested well.
This isn't a difficult problem. You test theories by comparing their predictions to what actually happens. (Note that a "theory" that makes no testable predictions fails this test.) As the article notes, economics does not do well on this test. But that's not a "difficult problem" of how to test economics; it's just that economists, and politicians who want to use economics to justify their pet policies, don't want to admit that the result of the test is that economics doesn't make good predictions.
> It's not any different from physics theory or chemistry theory in that regard.
Yes, it is, because, as you note, you can run controlled experiments in physics and chemistry, but you can't in economics.
Isn't that just another economic theory?
Yes, in the same way we could formulate a medical theory that says "we don't know enough about disease to know whether our treatments are effective, so we aren't going to create a standard diagnostic manual". Prior to the last couple centuries, that would probably have been an improvement.
No, in the sense that, compared to most economic theories, you can't really use this theory to justify a particular group being given access to government power.
I think most of the gains we've gotten from the advent of evidence based medicine is building the taxonomy of medicine so we could reason about it effectively. That only works in a larger sense, small groups toiling with different nomenclatures end up wasting effort.
Much the same way economics is building the taxonomy of economies so we can compare and discuss.
Economics is going to become an entirely different animal in the future as we begin to automatically collect fine grained metrics from our automated economies.
Soon we'll be in the era of evidence based economism, and we'll see a huge increase in the effectiveness of it's utilization.
Not what I was trying to say; I apologize if I was unclear. Rather, my point was that during times past, in many cases it would have been better if doctors had not applied any treatment to patients. Not because doctors were fools, but simply there wasn't enough rigor around assessing the effectiveness of treatments.
So too do I advise this for the field of economics. Not to say that we shouldn't strive for rigor, but until we have it I recommend the "doctors" (i.e. economists) refrain from applying treatments to their "patients" (i.e. the nation via policy prescriptions).
> Soon we'll be in the era of evidence based economism, and we'll see a huge increase in the effectiveness of it's utilization.
An optimistic view that I do not share. I recommend the talk "Science, Knowledge, and Freedom" by Jim Manzi[0]. In short, observations in the "soft sciences" of psychology, sociology, and economics do not generalize across different times and cultures in the way that observations about physics, chemistry, and biology do.
In reality it's more like "psuedo-scientific mismanagement" vs. "mostly civilized confusion". Neither of these scenarios is particularly appealing, but I'd rather take the confusion.
A rather savvy economic-statistician once told me that economic theory is a very effective way to determine who has been able to get their theory published.
That's not to say that economic theory is all sound, but that even the soundest of theories doesn't stand a chance against the incompetence of government.
There are many economists trying to do the right thing, but overall the area has a deep problem.
Ricardo Reis argues that macro has largely moved beyond real business cycles with frictions[1]. I'm not as optimistic as him, but DSGE is no longer as prestigious as it once was.
Aside from that, I always wonder why we have never had a "Great Economic Project." In WWII we united scientists of all stripes to come together to build the atom bomb and more recently Obama launched the BRAIN initiative. Understanding how our economy works is, I'd say, as important as either one of those. Some project where the government can assemble the best minds in sociology, psychology, business, math, history and then use economists to help glue all of this together. Perhaps our way forward is not in coming across some new beautiful model but in building the ugliest, most glued-together piece of crap that happens to be (relatively) accurate and then simplifying it later.
There's a reason why Western-style market capitalism has taken over the world. You aren't anywhere near remotely worried enough about it.
The nature of politics is such that if the government funded such a thing, the only acceptable outcome for it to come to is that the people funding it already had the correct model. The alternative is that the politicians funding it would have spent a lot of political capital to get that funding only to be told that they're wrong, probably wrong about everything.
On one level or another, even politicians realize that they might as well just cut out the middleman here and operate on the vigorous and loud assumption that they are already correct about economics and anyone who disagrees is the literal embodiment of evil.
Interesting to note that the author tries to discredit Keynesian economics, but makes no mention of the Reinhart / Rogoff debacle that completely undermined their paper which purported to show a link between debt and growth - a paper much referenced by people like the author of the article who are primarily interested in using economics as a fig leaf for their ideology.
For the record, I trained in politics and economics.
The reason economic theory is so hard is because it's very hard to observe and experiment. You can run a test and verify in physics, math, etc, but in economics this is generally impossible. Even the boom in behavioral economics that happened in the mid-2000s has come under question after more and more data has emerged. There are a LOT of things economists agree on, but there's also nuances on how a policy can achieve a result...because there are so many actors in an economy and we can't test our theory until maybe 10 years from now, and that's even if the policy is ever implemented to begin with.
Perhaps we should distinguish Economics and Rhetoric.
Economists use historical data to calibrate models. Of course data from the Reagan years was used to are for the Bush tax cuts. But simple backtesting can never tell you with certainty wherever your theories worked out or not.
For example, how good are parameter estimates and how do we know if they are good? If I was designing a controller for a plane, I can spend a lot of time getting good parameters for the plane model and design a controller that looks good in simulation. However, when I take the plane out, it might not be stable (or doesn't meet some other metric I designed for) because the parameters changed due to a variety of physical factors and/or the model is a model and thus doesn't account for anything. So I go back and forth tweaking the controller design until I get something that works. That process isn't really feasible when it comes to economies, though.
Would building high speed rail from NYC to Chicago be a net positive for consumer welfare? Well we can use estimates of how much people pay for various modes of transportation throughout the country, and how often they're used. We can also come up with an estimate of the cost of building the rail. The analogous effects to your feedback loop are equilibrium effects - for example if people start taking the new high speed rail then plane ticket prices will fall a bit. But it adds more uncertainty to the model, and we know that plenty of flights from NYC to Chicago are connecting flights, so probably the effects are pretty small, and we can just assume those prices stay fixed for the sake of tractability.
If you do this kind of calculation and find the rail project is a massive multi billion dollar money pit, you can confidently recommend against it despite the fact you've ignored some of the feedback effects.
First one is that it is based on drawing conclusions from models that attempt to simplify extremely complex interactions of billions of agents.
Second one is that although these models are imperfect (otherwise they would be unsolvable and bring no meaningful intuition), they are getting better over time.
Stop dismissing economics because 'it doesn't work.' The subject of the science simply doesn't have any determinism to it whatsoever, and things aren't reasoned through discovery as it is the case with natural sciences.
We know that there aren't any countries which rose to first world status without judicious application of tariffs, subsidies or currency manipulation.
That does not bode well for the future of the economy...
The fact that the question is being asked about Economic Theory in a respectable, prominent magazine like National Affairs provides the answer.
[1] https://newrepublic.com/article/112745/yuval-levin-rights-ne...
When it comes to economic theory, scrutiny becomes of paramount importance if you are going to make laws based upon it!
Then there is central banks which I think is a form of central planning. Most central planned system lack information that why they do not work. I argue production of goods and services are free market but money creation is centralized by banks and central banks. https://en.m.wikipedia.org/wiki/Economic_calculation_problem
Then there is the whole thing of peels bank act where the bank loans your bank account assets to others no matter how risky but still guarantees it to you. Is banks doing a form of double book keeping? https://en.m.wikipedia.org/wiki/Bank_Charter_Act_1844
The first half of this is less than clearly true, the second half is not only not true but it's not even coherent. Resources are linear with regard to what independent variable? Certainly not time, the one that is implied in exponential growth.
The amount of energy the world consumes to produce 1 unit of GDP declined by 32% between 1990 and 2015 [1]. Measuring macroeconomic material intensity is more complicated [2], but the core of your misconception derives from the subjectivity of value.
One way to make a car more valuable might be to make it twice as heavy. But another could be to make it a nicer color. Or give it better software. (Better being defined solely by price.) As our world becomes immaterial, the decoupling of material inputs and economic output makes sense. It also lays the ground for theoretically infinite economic growth. As long as peoples' utility is being increased the economy will grow.
[1] https://www.eia.gov/todayinenergy/detail.php?id=27032
[2] http://www.sciencedirect.com/science/article/pii/03014207859...