Friday, 29 June 2007

David Card - An Emperical Economist

I have a big problem with economics as it's currently taught. The 'Nobel' prizes are won by abstract mathematical economists that have no real world value what so ever. They plug in ridiculous axioms and make so many assumptions to just make their theories work. The sad thing is that these prescriptions are lapped up by policy makers because it turns political choices into market realities. Thus, outside democratic choice.

Take one example, the Coase Theorem. I came across this theory in a Tort class. Law professors love to analyse legal problems with economic tools, because the majesty of purported objective formalism gives a good rationale for an overall encompassing theory of tortious jurisprudence. Thus the Theorem becomes "The starting-point for [The] economic analysis of law" [Markesins & Deakin's:26].

I was intrigued at such an ambitious claim, rarely in jurisprudence does one concept shed light on everything. The only exception I can think of is Hart's distinction between primary and secondary rules, which does a pretty good job of exposing the rest of his legal theory - but even that is general theory, not based on an actual area of practiced law. But much like Hart's primary and secondary rules, which have no evidential basis[p44] - nor does the Coase Theorem, which assumes there are no transactions costs and externalities are quantifiable. This is ridiculous, and free market proponents of the Chicago school advocate this as a viable alternative to Pigovan Taxes.

I was pleased to read an interview by David Card, an economist I hadn't heard of. To surmise his findings in his career

  • The minimum wage does not have an effect on employment levels.
    • "small raises in the minimum wage won't have much of an effect [on employment levels].
  • "Skill-based technical change' does not explain wage inequality [in the US].
    • "Like a lot of other ideas in economics, I think that “skill-biased technical change” can be pulled off the shelf and used to explain inequality in a very superficial way"
    • SBTC holds that people with lower skill should have slower wage growth than people with higher skill. Women, have lower wages, evidence of their lack of skills. However, during the 1980's women did far better than men. Over the 90s women's relative wages were stable.Basically
    • Basically, the trend in the female labour market shows trends SBTC predict.
  • Immigration
    • imigration has minimal effect on unskilled domestic labour
    • "the best available evidence is the effect is on the order of a couple of percents nationwide over 25 years"

Card's summation of the economics profession as a whole is also bang on:


Economics as a whole is really a combination of two kinds of people: those who are very practically oriented and those who are more like mathematical philosophers. The mathematical philosophers get most of the attention. They deal with the big unanswerable questions. Labor economists try to be more scientific: looking for very specific predictions and trying to test these as carefully as possible. The mathematical philosophers get very frustrated by labor economists. They come up with a broad general theory, and we tell them it doesn't fit the evidence.

And this is the problem I have with anarcho-capitalist proponents. Take one example off a message board


it is a well known law of nature (economics) that a monopoly can neither efficiently nor effectively provide goods or services because of the lack of market input in the determination of production values and proper allocation of resources.

This is absurd. Theories of economics and laws of nature are not the same thing. Einstein's theory of relativity is a law of nature, but even physicists aren't so arrogant to call this theory, that has been empirically proven thousands of times, a law. Off hand I can think of examples of efficient monopolist state supplied healthcare, which is far more efficient than the private market.

Furthermore, if theories of Chicago school economists were theories of nature, they would have an essential property of a theory of nature - symmetry. Which means that relativity applies on earth as it does on the moon as it does in a distant galaxy. Also mere mortals cannot break laws of nature, the keyboard I'm typing into cannot teleport over to mars (microscopic quantum tunnelling aside!), but I can easily be an efficient monopoly supplier - if I disobey the first commandment (axiomatic assumption) of free markets - the profit motive. Healthcare is a good example; people are motivated by more than money.



References,

Markesins & Deakin's, Tort Law 5th Ed.



2 comments:

Chris Cella said...

Dush -

Nice post. Just a few quick thoughts though:

1 - David Card is a great economist. However, his work on minimum wage, especially the study he performed with Alan Krueger (sp?), has been scrutinized and criticized by both the theoretical as well as the empirical economists you talk about.
http://www.nber.org/papers/w5224
http://www.nber.org/papers/w6996
http://www.nber.org/papers/w10656
http://www.nber.org/digest/jan98/w6127.html

Even left wing economist Paul Krugman – certainly no libertarian – has levied criticism on the study. http://www.pkarchive.org/cranks/LivingWage.html

Additionally, he is somewhat flippant in the way that he dismisses inequality/skills based change and the effect of immigration. Here, as with the minimum wage, these are complex questions and I don't believe that Card has yet 'answered' them.

2 - I'm a little confused about your assertion that the Coase theorem has no evidentiary basis. This theory has been test empirically and results are mixed depending on the situation. However, there is certainly a body of research that reinforces the validity of the Coase Theorem in some respects (I will gather this up for you if you'd like to get into it further - provided I can get the time!) The problem with transaction costs is not necessarily fatal. The real question is how high can the transactions costs be before they render the transaction infeasible.

That said, on some major issues, such as the environment, I would prefer pigouvian taxes.

3 - I definitely understand your thinking in the way you criticize certain aspects of economics and economists. However, the empirical economists that you seem to favor deserve some criticism of their own.

The problem is that economics is, by its nature, a broad subject. In biology or chemistry, theories can be tested by isolating the subject, and controlling the circumstances. This is not so with economics. One cannot break up two identical communities, alter, for example, just the minimum wage, and then wait to see the results. Instead, economic theories and there results are tested through econometrics. This involves constructing mathematical models which seek to account for outside effects or 'noise' while examining the effect of a single policy.

And while this can be a powerful tool, it brings us back to your original criticism. The empirical economists are still using mathematics and theory when they attempt to measure economic policy. Econometrics can become a religion as well. It is not always clear how much faith can be placed in a model and its results.

This is not necessarily a reason to begrudge the economics profession. There is no way of really preventing the broad nature of its studies. It is not so dissimilar to law, in my opinion. When people are maddened by the grey area and ambiguity of the law, it could be that they are incorrectly associating simplicity to accuracy. Broad subjects require broad methods of study, and a discussion of theoretical underpinnings is unavoidable.

As such, the application of economics warrants a skeptical approach that balances theory with the available empirical evidence, while evaluating the validity of each. One should merely avoid economic theories simply because of the conclusions they reach.

4 - As a follow up to that point, I don't think that its fair to dismiss the scholarship of the Chicago school. It is definitely frustrating when its proponents are blindly devoted to the theories without regard to evidence. However, in the quest to criticize these views, people sometimes run in the other direction and right into the arms of its critics with almost equal blindness (I am not saying that you are doing this.) David Card is a fine economist, but to support his research simply because one agrees with his conclusions is to be no better than his opponents. Skepticism is a two way street and the econometrically-heavy methodology of Card warrants its own scrutiny.


Those points aside, this is a very thought provoking post. I am sympathetic to your concerns and think that all sides need to consider these questions more thoroughly.

- Chris Cella

Dush said...

Thanks for those links, I'll read them today.

My minimum wage causing unemployment/inflationary pressure stems from the experience in the UK. When I was studying econ at school, I was told the abstract simple model that meant a minimum wage = inflation and unemployment.

However, in the UK the evidence hasn't born this out. We have a minimum wage yet the lowest unemployment and inflation in decades.

My personal feeling is, of course I cannot back this assertion up without a proper research, is that labour markets simply do not come to equilibrium prices. Corporations exploit their monopoly purchasing power and act in a cartel like manner to keep wages down. I'm not arguing they actively do this, perhaps something like how the kinked demand curve for oligopolies leads to non explicit consortation to fix prices.

2) The angle I was coming at in tacking the Coase Therom was that if property rights were defined strongly, pollution problems are handled. I'm aware that some economists brought up bee keepers as an empirical example and in this very limited sphere, yes Coase seems to apply. But the Chicago boys attempted to implement their policies in Chile (including Coase), and they had awful pollution problems. The Reignites used Coase as a justification for no environmental legislation, this is so damaging. But I suppose you can’t blame a profession for how others abuse their research.

I wasn't aware that transactions costs were not fatal though.

3) I understand that empirical economists need scrutiny. The main issue with economics is that abstract mathematical notions are actually presented as policy suggestions. This leads to devastating consequences for the environment and people round the world. Take for example the neo-liberal economic policies of the World Bank and IMF. They justify no protectionism on grounds of these abstract theories. As a result, food leaves countries with starving populations where it commands higher price. The Chicago School is responsible for many of these policies, including the devastation to Chile.

Every single country that is currently an economic power does not adhere to these free market principles. They are protectionist in nature and have strong state supported economies.

I agree that broad models have their place but all too often they are justified to keep rich countries rich, poor countries poor and take political choices outside the democratic arena and transform them into ‘market realities’.