A natural adjustment
‘Nature’ and ‘incentives’ in economics
August 30, 2007 24 comments
While unspeak.net has been quiet recently, global financial markets have been very nervous, which I can’t accept is a mere coincidence. ((Also while I was “away”: my review of William Gibson’s new novel Spook Country is posted at the newly redesigned stevenpoole.net, featuring lots of stuff. Feel free to have a look around there and let me know if the site works on whatever crazy OS/browser combination you are using.)) Anyway, my eagle eye did notice that George W. Bush said, of the US subprime mortgage crisis and subsequent stock wobbles: hey, it’s just “the nature of the market“, and there is no need to fear “a natural adjustment“. Now, IANAE, but it strikes me that appeals to what is “natural” in such contexts are somewhat troublesome. A financial market is not a phenomenon of “nature” but is an aggregate of many human decisions, bound by rules and laws that some people made up, and some people want to defend from alternatives. So I am tempted to suspect that a description of some or other market phenomenon as “natural” cannot help but be a coded defence of a particular economic ideology. It is “natural”, therefore trying to alter it would be hubristic, playing God, et cetera. To be consistent, Bush’s attitude to his own health should run along similar lines. Rather than interfering with nature by having some polyps removed, he ought surely to say to himself: “If I get colon cancer, well, it’s only natural.” Nature knows best. Why worry?
That was also the suspicion I had when reading Steven E Landsburg’s More Sex is Safer Sex: The Unconvential Wisdom of Economics, of which I wrote a brief review in the Guardian. Landsburg, too, appeals to a concept of what is “natural” and so shouldn’t be contested. Noting that child labour existed in mid-19th-century England and America, and that it no longer does very much in the more prosperous modern England and America, he argues that child labour is perfectly normal and goes away after a while. Thus campaigners against child labour in the contemporary “third world” are stupid:
Evidently, child labor is a natural response to a certain level of poverty. [p 67]
That’s not “evident” at all. It may be evident that child labour is often found in societies at a certain level of poverty, but to call it “natural” is not only to pretend you have a sample size much bigger than the one Landsburg actually invokes, it is also to make an extra, ideological claim: that it’s okay and shouldn’t be resisted. But IANAE. Perhaps some readers more familiar with economics can suggest some economic usages of “natural” that are not Unspeak, or even argue that these ones aren’t.
While I’m on the subject of Landsburg’s book: it also contains one example of the rigorous restriction of vocabulary to its narrow economics sense that is really disturbingly bizarre:
Every midsummer day, at approximately 6:03pm, the setting sun makes the traffic light on my street corner essentially invisible to westbound traffic. As a result, I’ve gotten to know the local police officers fairly well. We meet on my front lawn every week or so, where I’m delivering water, blankets and cell phones to the latest accident victims while the police file their reports.
It is an astonishing triumph of modern safety engineering that dozens of cars have been totaled on my front lawn […] without a single serious personal injury. And it is an astonishing failure of the legal system that I have absolutely no incentive to step out my front door at 6:02pm with a big red flag, directing traffic until the sun moves a little lower on the horizon. [p 114]
Here, Landsburg complains pathetically that he has no “incentive” to save people from traffic accidents outside his house for an hour or two once a year. ((Well, it’s either one day a year (“Every midsummer day”) or some unspecified number of weeks every summer (“Every week or so”). But who’s counting?)) Would a concern for their welfare not count as an “incentive”? Ah, don’t be naive. An incentive is money and nothing else. Landsburg, nothing if not a rigorous homo economicus, won’t do anything at all unless the state is promising to pay him for it, or give him a tax break, or throw him some other kind of legal bonbon.
Okay, fine. So why exactly does he deliver water, blankets and cell phones to the victims after they have crashed? Where is his “incentive” to do that? Would he not save a bit of money in water, cellphone and blanket costs if he did actually go out with a big red flag and prevent the accidents in the first place? Would those savings, indeed, not count as exactly the sort of “incentive” he dreams of?
Well, as I must repeat, IANAE, and there is probably some good theoretical reason for Landsburg’s annual decision to let people get hurt and then heroically minister to them, rather than to prevent the predictable hurt in the first place. Maybe it boils down to the fact that these accidents are caused by the sun – and, you know, that’s natural.
There is a sort of voodoo thinking that the market “needs” to have crashes every now and then – hence they are called “corrections”. It’s rather like how medievals used to believe they had a surfeit of blood and needed leeches (and is about the same level of thinking). I don’t have much of a problem with it being in the “nature” of risky securities that from time to time someone will lose a bundle on them, but you’re on to something here. If medical schools thought like the stock market, they would say “we haven’t had a tuberculosis outbreak for a while, so we are ‘due’ for one, and any attempt to prevent it will just make the eventual TB epidemic worse”.
The stuff about child labour being natural is pure Landsburg btw – thankfully for the profession, it is not normal for economists to use that form of words or anything implying it. I remember spitting a mouthful of breakfast on reading his line about how “you could phone up a manufacturer of shopping trolleys” and bitterly regretting not having done more to crush his career ten years ago.
Ah yes, “correction” is a very good comparison, thanks. (No doubt it is “correct” that some people should lose their houses.) Belatedly, I also see you wrote something quite apposite re Landsburg a while ago:
I think “a particular assumption about the distribution of property rights” might be behind some of these types of “natural” usage, no?
Also, there is the concept of a “natural level of unemployment” (I thought Landsburg had used it himself somewhere but couldn’t find it again). IANAE so I wonder if there is a good (ie non-Unspeak) reason for that usage.
“Natural rate” was Milton Friedman’s coinage. In fairness to the profession, the confusing and misleading implications of “natural” in that context were picked up pretty quickly and hence we have “NAIRU” (the non-accelerating inflation rate of unemployment; Friedman’s original idea was that if unemployment fell below the natural rate, inflation would be rising). NAIRU turned out not to work all that well either but I think the economists caught that one quite early.
I think your #2 is right – it all goes back to Locke and the “state of nature”. The general project of coming up with weaselly little get-out clauses to justify grabbing a lot of stuff is as old as political economy itself (which makes sense, as that is like nine tenths of the subject).
Somehow I’d missed that Thomas Malthus wrote on political economy until I read this post (I looked him up to see if he coincided with Adam Smith – of course he does).
It seems dsquared’s medical/biological connection is built right into the economic language. Does the Department of Corrections have a similar ‘natural’ basis, correcting a surfeit of freedom?
I haven’t read Landsburg’s book, but as a student of economics, I think I understand his usage of the word “natural”. As you may know, economists often assume “rational” agents in theor models. This typically means that you end up assuming that agents (firms, people, etc) are doing the best they can for themselves (according to their own preference) — given their initial endowments, and other constraints. These prefernces may include many (any!) things other than money.
So the sense in which Landsburh says “natural” is that it is the households best response to the conditions it finds itself in. Herein lies a big problem — what the “household” (obviously, typically the adults) wants is not necessarily what is best for the children who are forced to bear terrible burdens. (It isnt necessarily always the case that the children would be better off not working, though).
Another was in which Landsburg might mean child labor is “natural” is that he might feel “it goes away by itself. therefore, its nothing to worry about” — which are two different things altogether. While child labor may decline with prosperity (it probably actually declined in England and the US due to compulsory schooling laws) — this does not make its presence unavoidable, nor does it make is ethically acceptable.
Is it not too much to wonder to what ends economic crashes may be orchestrated? Surely the level of economic power in the nads of the great centres of power should dictate against such nothing events with their catastrophic results; nothing in the sense that all that occurs is the fluctuation of abstract numbers- no crops fail, no fleet of ships are lost at sea, just the rise & fall of numbers.
As far as being natural, it’s almost embarrassing to have to respond to such statements; presumably if in anger a few Unspeak readers were to arrange to kidnap Landsburg & sell him to some South Seas pirates as a strange sex-xlave…well then, who’s to argue…the fact that such an event occurs shows how natural it is.
sex-slave
As you may know, economists often assume “rational” agents in theor models.
Which is actually totally unnatural given that people are not rational and there’s a vast amount of evidence now to support this. This is belatedly entering into the profession, though in practice economists largely seem to assume rationality anyway. The most significant factor in choice making is emotion. Hence the blankets rather than the flag.
There’s a lot of interesting work in anthropological economics looking at the “naturalness” of markets. Or rather, the work that goes into creating and maintaining them.
Which is actually totally unnatural given that people are not rational and there’s a vast amount of evidence now to support this.
Matters less than you’d think. Lots of economists pay religious obeisance to the totems of universal rationality and self-interest, but much less (and almost nothing important) depends on it than you’d think. If you’ve got a factory and you want some widgets, then the correct point on the widget production curve is where it is, and that’s all there is to it. The Soviet Academy managed to prove almost all the important results in Samuelson without making any such assumptions.
That’s good to know. There seems to be a real problem with people like Landsburg given economics a bad name. In the book, he repeatedly explains what people ought to prefer “rationally”, and then notes that in tests, surveys etc they don’t actually prefer it. The lesson is always: Duh, aren’t people stupid!? Which is a bracing lesson, but not necessarily always true to the extend Landsburg claims. There are often very good reasons why people assert the preference they do; it’s just that those reasons cannot be comprehended by Landsburg’s banal arithmetism.
A propos of which, we all eagerly await the next instalment of your Freakonomics review.
“Matters less than you’d think.”
But an awful lot of economists assume it in their work and then prove results of rather dubious validity on the basis of it. Public choice theory, game theory, that kind of stuff.
Public choice theory I think it’s difficult to blame economists for – a couple of economics Nobel prizes got awarded for it, but it’s not really part of economics. It’s basically just right-wing prejudice turned into a theory – the entire intellectual content is in the initial assumption “1. We assume that all public officials are venial and self-seeking”, and thence to derive the entire right-wing worldview.
Game theory I think you have a lot more of a point with. Phil Mirowski’s book “Machine Dreams” is very good on the development of game theory. He raises the of-course-very-unserious-who-would-possibly-be-so-uncouth point that a) the key game theory concepts like Nash equilibrium describe the reasoning of a paranoid schizophrenic a lot better than that of a normal person and b) a surprisingly high proportion of the key figures in the development of game theory suffered from mental illnesses, and Nash himself was of course an actual paranoid schizophrenic.
Game theory is actually very good in predicting the outcomes of things like telecom spectrum auctions, which of course really means that it is a good way of predicting the behaviour of game theorists employed to solve game theory problems.
I haven’t read Machine Dreams but I wonder about the relationship between Mirowski’s (a) and (b). Ie do you think he would have proposed (a) based on internal evidence of the theories in the absence of the biographical facts in (b), or is it possible he read (b) into (a)?
In any case (b) does not tell us very much. A surprisingly high proportion of the key figures in music, painting and literature suffered from mental illness, etc.
Natural is very ambiguous.
Lets take a big step back and start with things everyone can agree are natural like Rats and Birds. Rats and birds both make nests. I am sure MOST would not argue that the nests are “natural”.
People make nests as well, called houses. Are houses really any less “natural” then nests? I know I know, many are saying “well yes houses are man made and so they aren’t natural”, but if you really take a look at it, are they really? Didn’t people come from nature originally and so mustn’t anything we make be as natural as anything made by anything else in nature? It is mans nature to make tools and make shelter.
In my mind there is no such thing as something being unnatural. It truely seems impossible to me that we or anything else could some how break the rules of nature (physics) and do anything un-natural.
So when I look at the man made stock market and think about it, it is natural. A natural thing made by a creature of nature, just like a nest.
Well, that’s a point of view: that Apollo 11 and the hydrogen bomb and MacBooks are all “natural” too. But if you take that view — as you state it, that nothing can be unnatural — then the term “natural” no longer has any meaning, since it applies equally to everything in the universe. Ipso facto it is meaningless to apply the term “natural” to stock-markets, as to anything else.
I think it’s quite possible that he would – I was aware of the critique of Nash equilibrium on these grounds a long time before I knew about Nash’s illness and it made sense. The Prisoner’s Dilemma is squarely based on the premis that the other player is going to betray you. One of the more interesting subfields of the otherwise very, very disappointing field of behavioural economics is that people with mental illnesses behave *very* differently in economics and game theory experiments from normal people (or at least from economics graduate students, but there you go). I’d very much recommend Mirowski, by the way.
Here’s a bunch of relevant extracts. Mirowski thinks that the success of the Nash equilibrium-condition was a point where economics began to go seriously wrong in the direction that Cian indicates, and that von Neumann (and later Herb Simon) were working on much better models that were closer to “natural” modelling of human behaviour.
Just because you’re paranoid doesn’t mean they’re not likely to betray you.
Thanks for the recommendation of Mirowski, I’ll try to have a look for it. Edit: ah, thanks for the extracts.
Isn’t ‘natural’ being used here as a synonym of ‘inevitable’? If you have trading systems set up as they are, then such crashes are definitely inevitable.
If I’m right, there are two points here. One is that natural is a happy word while inevitable is a rather sad word.
The other is that we’re being told that the way markets are currently organised is itself natural and not in any sense the result of a large number of choices, many of them practical but some of them very much driven by ideology, expediency, etc. (This also serves to maintain the myth that something like pure capitalism exists in America, which I understand is an important part of the credo.)
I think ‘natural really means inevitable in the current context’ would also make the H-bomb, houses, etc. ‘natural’. So there.
I doubt anyone’ll be reading this a fortnight after the original post but that’s the Internet for you. Post 8 was a peach.
I think ‘natural really means inevitable in the current context’ would also make the H-bomb, houses, etc. ‘natural’. So there.
I think I detect some circular reasoning here: they exist, so they’re inevitable.
FWIW, I subscribe wholeheartedly to the notion that it is impossible for anything unnatural to exist in the universe: any other viewpoint suggests a rather selective or romanticised view of nature. Steven’s point about how ‘natural’ becomes a useless term (if stretched to cover all things that are) seems rather to ignore its use as unspeak, which he presumably feels we can all take for granted, since it’s covered at length in the book.
One is that natural is a happy word while inevitable is a rather sad word.
Yes, I think this is a really nice comparison. “That is the sound of inevitability,” as Agent Smith says.
On natural and happy words, are you planning to write anything on “lily-pads” – which seems to be emerging as the new preferred Pentagon term for US military bases on foreign soil (like those ones in Saudi Arabia that seem to have provoked some opposition)?
They seem to be proliferating at an alarming rate. Which might just be the time of year, I suppose. The term might not sound so idyllic to Iraqi or Egyptian ears: I understand both the Tigris and the Nile have suffered great strangling blooms of the things.
“Lily-pads”, eh? That’s wonderful. I imagine a giant mechanized frog hopping from one to the next, probably armed with nuclear weapons like Metal Gear.