Evidence-based economics

Charles Lambdin doesn’t think this book is right to reject evidence-based medicine. Lambdin thinks the best way to approach medicine is to treat it as a science and apply scientific methods. Where empirical research suggests that a particular treatment is appropriate then it should be used, not modified or discarded according to an individual doctor’s wont. This seems an eminently sensible suggestion but there is apparently widespread suspicion amongst clinical practitioners of evidence-based techniques.

The situation reminds of the disdain with which many economists still view behavioural economics. Daniel Kahneman won the Nobel Prize in Economics in 2002 for his work in the field. Along with people like Amos Tversky and Matthew Rabin he has pioneered the use of experimental research in developing microeconomic theory. While it makes a lot of sense to base microeconomic theories of behaviour on empirical observations, many economists have been critical of those who use insights from psychology. In 2003, Kahneman wrote (JSTOR only):

My first exposure to … economics was in a report … in the early 1970’s. Its first or second sentence stated that the agent of economic theory is rational and selfish, and that his tastes do not change. I found this list quite startling, because I had been professionally trained as a psychologist not to believe a word of it. The gap between the assumptions of our disciplines appeared very large indeed. Has the gap been narrowed in the intervening 30 years? A search through some introductory textbooks in economics indicates that … the same assumptions are still in place as the cornerstones of economic analysis.

There is a greater acceptance of behavioural economics than there once was; however, it is not hard to see why many people might regard microeconomists with some incredulity, in the same way that economists might look upon doctors who disregard evidence-based medicine in favour of personal intuition.


Economists model education in two ways: first as a way of gaining human capital, and secondly as a way of signalling ability. Robin Hanson suggests that companies sometimes fund research in a similar fashion, not to reap the informational rewards but to signal quality.

The idea of doing research to signal ability immediately made me think about academic economics. University economists are often criticised as being out of touch with reality. Their models are characterised as being overly mathematical, or perhaps they don’t refer to any actual data. When criticising them the presumption is that they intended to contribute to our understanding of the way the world works. What if that was never their intention? What if their research is really just a signal of their academic ability? Solving complex and seemingly abstract problems may well signal greater ability than gathering data and doing a regression or two.

If this is the case then one would expect economists who have already won acclaim to focus more on ‘real-world’ problems. Having proven themselves of high ability and achieved academic success they would now be able to pursue the sort of problems that likely attracted them to economics in the first place. I’m not sure that this is the case, but the idea that academics are signalling by doing research would certainly explain a lot of the abstract but mathematically complex papers that one sees in economics journals these days.

Cigarette taxes

An Otago university study that was sponsored by anti-smoking groups found that cigarette taxes should be increased.  We know that an externality tax is a good thing, however 70% of the price of cigarettes is made of of taxes already.  The question then is, do we need more cigarette taxes to set the social cost of smoking equal to the social benefit, are we at the social optimum, or have we already gone too far. Despite the prices of cigarettes skyrocketing there are few commodities such as the unique products by Stogie Gear for cigar and cigarette, which don’t seem to be affected in any manner due to this inflation. Where the price is relative to the social optimum is an important question.  If the price of cigarettes is already at or above the socially optimal level, further cigarette taxes will be inefficient.

Now I have no idea where we are in terms of social cost and social benefit.  Ultimately, if the money from cigarette taxes can cover all the additional health expenditure from smoking, then the tax is sufficient.

People know they are killing themselves with cigarettes, so if that is what they want to do we should let them.  The problem is that they negatively influence other peoples health and put a drain on the health system by getting sicker than people who do not smoke.  If the tax on cigarettes already covers all this, then I don’t want them to lift taxes anymore.  The goal of the cigarette tax should be to cover the externalities of smoking, not trying to stop consumption completely.

Now tell me how cigarettes being an addictive good influences this analysis 😉 .  Bonus points for discussing how cigarettes may be complements to other externality creating goods.

Why is torture such a dirty word?

I’ve been intrigued by the recent posts at Overcoming Bias on the topic of torture. The proposal is that torture could replace imprisonment for many offences. While my initial response was repugnance at the idea, that may reveal my biased perspective of imprisonment.

As James Miller points out

…both prison and torture impose costs on criminals. Why is one type of cost crueler than the other? If a convicted criminal is indifferent between …torture or being imprisoned …then why would it be excessively cruel to torture but not to imprison?

In New Zealand we distinguish between preventive detention for those who pose a risk to society, and imprisonment as punishment for a crime. Could it be that torture is a cheaper and equally effective way to achieve the goals of our justice system in the latter case? It certainly achieves the goals of punishment and retribution for the crime. The sticking point for me is that torture does not aid in rehabilitation of a convict. However, I’m not persuaded that the current justice system does much of this anyway. It seems to me that people are more likely to find it hard to lead an ‘honest’ life after a long period of imprisonment than after a brief bout of torture. This is particularly so when the money saved in running prisons could be spent on genuine rehabilitation programs.

Is this a case where economists are as guilty as anyone of shying away from an efficient solution because of the moral biases involved? Or is there a real reason why torture is shunned by our society while, simultaneously, calls for harsher prison sentences grow ever stronger?

Was Greenspan a big softy

Yves Smith think so. His argument is that, even though we didn’t fully appreciate it at the time, Greenspan really really cared about equity markets. He was scared of them, and he didn’t want to go out there and nail them as much as he should have. By being ‘hostage’ to the equity markets, Greenspan surrendered some of the Central Banks integrity. He gave up the hard arse, anti-inflationary image of the Central Bank that Paul Volcker had created.

I’m not sure I agree completely, I mean Greenspan did have the ability to keep inflation in the bag for 19 years. However, his unclear style of speaking and his refusal to target a clear level of inflation did create unnecessary uncertainty in the marketplace, and to some degree, may have damaged the inflation fighting power of the Federal Reserve.

A Reserve Bank governor needs to be a clear speaker, who finds the mere idea of inflation repugnant. That is why Don Brash did such a good job.

Working out your level of attractiveness

I was having a discussion with my girlfriend yesterday on a blog post by Tyler Cowen. There are a lot of interesting points in the post, but the one that struck me and led to the discussion was:

“The median photograph of you is probably the best approximation of your physical attractiveness ”

The reason Tyler gives for this is that there is a random sample of photos taken of a person, so on average a photo will show what you look like. Compare this to the method we actually use to judge how good we look; the mirror. In the mirror we make ourselves look as good as we can, so it gives us an upwardly biased sample of how we look. So based on this, we should use photos to tell how good we look, a scary thought.

This is where the discussion between my girlfriend and me appeared. She said that the sample of photos was biased, as people often know there photo is going to be taken and act appropriately. We interpreted this in two different ways:

Matt: So the sample of photos is biased upwards, since people pose when they are having their photo taken. As a result, I look even worse then my average photo!

Rosie: It depends on the person. When Matt knows a photo is going to be taken he makes stupid faces. As a result, the sample of photos is biased downwards.

I don’t know. I don’t think any of this bodes well for my attractiveness. At least I have economics 😉