April 22, 2004

Powers of prognostication

Posted by Henry

I’ve been reading Anthony Grafton’s Cardano’s Cosmos: The Worlds and Works of a Renaissance Astronomer which is a lot of fun. Grafton has a lively writing style, well exemplifed by the following (unfair but funny) dig at the dismal science.

At the most abstract level, astrologers ancient and early modern carried out the tasks that twentieth-century society assigns to the economist. Like the economist, the astrologer tried to bring the chaotic phenomena of everyday life into order by fitting them to sharply defined quantitative models. Like the economist, the astrologer insisted, when teaching and writing for professional peers, that astrology had only a limited ability to predict the future. … Like the economist, the astrologer proved willing in practice, when powerful clients demanded it, to predict individual outcomes anyhow. Like the economist, the astrologer generally found that the events did not match the prediction; and like the economist, the astrologer normally received as a reward for this confirmation of the powers of his art a better job and a higher salary.

Posted on April 22, 2004 07:19 PM UTC
Comments

Brecht’s “Leben des Galilei” has a good moment when Galileo’s daughter is getting married and her nurse says “Right, now let’s go consult a real astrologer.”

Posted by John Isbell · April 22, 2004 07:48 PM

Which part was unfair?

Posted by msw · April 22, 2004 10:19 PM

The last bit is a bit harsh, I thought. I’m quite partial to economic reasoning myself, and probably biased. Although by sheer coincidence, I’ve just come out of a talk by Douglass North where he describes macroeconomics as having “elements of astrology,” so if someone who won the Nobel for economics is saying these things …

Posted by Henry · April 22, 2004 10:48 PM

Nothing unfair there… Robertson Davies makes the comparison somewhere too, but I’m blanking on just where.

Posted by rxb · April 22, 2004 11:21 PM

I think that the divide between academic and business economists is worth mentioning. The academics consider the shills on CNBC to be hacks, and the business folks with their tricked-out econometric models think that the eggheads are worse than useless.

Put concisely, this is the difference between explaining and predicting. The dig is well-written though, and to be fair, most people encounter the worst of the shills but not the academics.

Posted by Chris · April 22, 2004 11:29 PM

Following up on Chris, the orthodox ‘efficient markets’ view is that large numbers of time series of interest (including nearly all those on which the CNBC types pontificate) follow a random walk and are therefore entirely unpredictable.

So an economist who makes a prediction in relation to variables like this is asserting some form of market failure.

Posted by John Quiggin · April 23, 2004 01:41 AM

As a practising economist, I have to say this is very unfair unless by economist you mean macroeconomist. In my 4 years in the profession I’ve never been involved by forecasting of any kind unless you count cost-benefit analysis model where what is esimated are potential social losses based on a framework of assumptions that is very transparently laid out. We (micro)economists perform very useful social functions e.g. in designing appropriate auction systems for governments, frameworks for competition, guidelines for licensing of IP, etc. Economics is basically a toolbox with great use in statecraft. Spiv financial markets macroeconomists which are the only exposure to economics thay laypeople get give the rest of us a bad name performing their useless tea leaf readings based on ‘gutfeeling’on how much some governor is going to raise interest rates.

Posted by Jason Soon · April 23, 2004 03:33 AM

“Robertson Davies makes the comparison somewhere too, but I’m blanking on just where.”

‘What’s Bred In The Bone’ - the bit where the English governess and spy in the German schloss is explaining astrology to Francis Cornish while he’s taking a break from forging old masters and counting concentration camp trains.

For those of you who haven’t yet encountered Davies (R.), that should whet yer appetite

Posted by Nabakov · April 23, 2004 09:29 AM

If you’re interested in surprising affinities between science and alchemy, astrology, and prophecy, you might check out some of Matt Goldish’s writing on Isaac Newton or the first chapter of “The Sabbatean Prophets.” Newton not only had interests in the Kabbalah, but, apparently, had messianic tendencies as well.

Posted by andrew kinney · April 23, 2004 11:29 AM

“We (micro)economists perform very useful social functions e.g. in designing appropriate auction systems for governments, frameworks for competition, guidelines for licensing of IP, etc.”

We astrologers perform very useful functions in helping farmers prepare crops for next seasons weather, determining whether or not a woman is a witch by looking at the signs of her birth, providing guidelines for auspicious moments on which to sign treaties or behead traitors, etc. Also, when called in to observe monstrous births, we can predict how bad omens can be averted in the kingdom.

So what’s with this comparison to economists? Astrology, need I remind you, is a real science.

Posted by roger · April 23, 2004 08:15 PM

“the orthodox ‘efficient markets’ view is …”

Could you say more about this? I thought there were many versions of “efficient markets”, and I didn’t know any of them were “orthodox.”

For example, you might think that current market prices account for (1) all past prices, or (2) all public information, or (3) all public and private information.

Someone who believes (1) would still engage in the analysis of companies, just not their past prices. Someone who believes (2) might look at what insiders are saying.

Your comments about “asserting market failure” only seem to apply to people who believe (3); people who believe (1) or (2) might easily give analysis on CNBC, it seems to me (although a (2) would believe that their making the information public would immediately change the price).

Posted by Bill Carone · April 23, 2004 08:29 PM

You have to understand that wall street economist
forecasting what an economic variable will be this month is strictly a suckers game developed by the bond houses first and later stock brokers to generate volume.

The big houses do not care what the data will be, they make their profits on volume so by
getting suckers to bet on the monthly economic report they generate additional volume.

But there is a sucker born every minute, so despite the fact that I know of no investment institution that has consistently made money betting on the economic data, there are still enough traders willing to bet on anything that the game is well worth the trouble for the “house”.

Posted by spencer · April 24, 2004 05:50 PM

Well, I thought Grafton and/or Cardano were intentionally slurring the difference between science and politics; “politically challenging documents” appears in there somewhere, of horoscopes. Even respectable economists are willing to be politically challenging.

Posted by clew · April 25, 2004 10:01 AM
Followups

This discussion has been closed. Thanks to everyone who contributed.