May 7, 2007

Show me the mathematics.

Filed under: Economics — duncan @ 4:56 pm

So I’m reading a dummies guide to economics (“a bridge between the transitions from high school to college”) by Tony Cleaver, lecturer at the University of Durham. On page four, he scolds us thus:

“It is always easy to ask important questions in economics. It is easy also to make colourful and outrageous claims about the nature and conduct of economic affairs. (Have certain people really squandered the riches of Earth?) It is not always easy, however, to give balanced, objective and accurate responses to such questions and assertions. That is nonetheless the challenge of positive economics.”

Soon he’s bringing in Milton Friedman:

“One comment on theorising before we begin: Milton Friedman, probably the most famous economist entering the twenty-first century, has said that no matter how abstract and seemingly unreal the assumptions on which a theory is based, ‘the only relevant test of the validity of a hypothesis is comparison of its predictions with experience’. As you will see, though many assumptions in economics make perfect sense (e.g. we assume consumers and producers behave rationally) they are not necessarily true for all time. These assumptions do not matter so much, however. In positive economics we are most of all concerned about whether or not a theory’s predictions are confirmed by hard evidence. So long as it produces workable results, even the most unrealistic hypothesis must be taken seriously.”

His example is, to put it mildly, curious. Surely the assumption that consumers and producers behave rationally is itself barking mad. (Let’s mention the cigarette industry, as just one example among infinite others.) (Although, of course, the question of what constitutes rationality is central here; and we can take nothing for granted). But the broader point is compelling. It reminds me of a passage from Keynes’s General Theory.

“To say that net output to-day is greater, but the price-level lower, than ten years ago or one year ago, is a proposition of a similar character to the statement that Queen Victoria was a better queen but not a happier woman than Queen Elizabeth – a proposition not without meaning and not without interest, but unsuitable as material for the differential calculus. Our precision will be a mock precision if we try to use such partly vague and non-quantitative concepts as the basis of a quantitative analysis.”

I wonder whether Keynes had Lytton Strachey in mind when choosing his Queen Victoria example. But obviously it’s the general point that’s of interest. Put these three quotes together in a blender, and I think you have a pretty coherent attack on the project I’m pursuing here, or that I’m hoping to try to pursue. My sound and fury about the false premises of mainstream economics is all very well. But what concrete results could it possibly have? Economics is a quantitative science, or pseudo-science, but quantitative nonetheless. The false premises I rage against may be less than ideal, but they have one inestimable advantage: they produce results. And in economics, unlike literature or philosophy, if you don’t have results, you don’t have nothing.

I wouldn’t want for a moment to deny the force of this argument. But I want to make a few moves to defend my little economically and mathematically illiterate corner of the blogosphere; or at least to try to stop my project being killed before it’s learnt to walk. All I’m saying is: the things I’m pursuing are worthy of pursuit.

1) Okay, so these lines of enquiry may not produce material suitable for quantitative analysis. But truth does not, after all, have to be suitable for quantitative analysis. Assuming we recognise the substantial limits of my enquiry, this need not diminish its relevance to economics proper. For if we choose an unrealistic assumption as the basis for our economic predictions, it is of the greatest interest and relevance to know just how it is unrealistic. Even if that assumption remains the only useful one, we have, if we fully understand its failings, expanded the scope of our predictive powers – because we know more about the reasons why our predictions are likely, under certain circumstances, to fail.

2) But who’s saying that such enquiries will not result in material suitable for quantitative analysis? After all – the only way to really be sure of that is to see where we end up. I hate the analogy, but I’ll use it anyway, out of foolishness and haste. When Bolyai, Gauss and Lobachevsky were creating non-Euclidian geometry, no one thought for a moment that their work would have any practical applications. But, of course, it did. Similarly, for all we know economists who reject economic investigations that do not result in quantitative analysis may be sensible in the short term; but the last laugh could be on them.

3) And, finally, are the assumptions we’re talking about really just geared towards results? This is, perhaps, the most important question. For consider again the example Cleaver gave to illustrate a useful economic assumption: rational consumers and producers. Cleaver says: this assumption makes sense; it turns out to sometimes be wrong; but that doesn’t matter, because results are what count. But, of course, the assumption doesn’t make sense – not to me, anyway. We’re not rational at all, are we, brothers and sisters? I would argue that far from being assumptions that are meaningless apart from the results they generate, all too many economic claims have conceptual and (yes – I’ll say it) ideological functions wholly separate from whatever quantitative analyses they produce.

All obvious, sure, but let me expand anyway. Friedman says ‘the only relevant test of the validity of a hypothesis is comparison of its predictions with experience’. But, of course, that’s not all that economic hypotheses do. Economics cannot be separated from politics. Economic hypotheses are used to justify political action – by Friedman as much as by anyone. If we’re not attentive, a dishonest conceptual manoeuvre can be perpetrated. A claim that is supposedly only justified by the predictions it generates can be used to justify political choices and values that are utterly devoid of predictive power. This might even have happened, once or twice.


  1. Hello — I’m enjoying the blog. My simple-minded take on this post is that maybe you have nothing to worry about. The claim you are considering is that the assumptions may be wrong taken by themselves, but that they get good results in economics, and that’s what matters. But economics is everywhere. If the assumptions get good results in economics, they get good results full stop, and therefore they are correct in the fullest sense.

    Maybe the point to force is that the assumptions do not yield particularly good results, after all…

    Comment by RobS — May 8, 2007 @ 12:54 pm

  2. Hey Rob, glad you like the gibber. Thanks for posting.

    I would definitely want to force the claim that certain pervasive economic assumptions don’t yield very good results. But I’m afraid my gut instinct is to run like a screaming bat out of hell 2 from the idea that economic results = results period. Economics is everywhere, yes; but everywhere ain’t economics.

    Of course all this depends on just what we mean by ‘results’, ‘economics’, ‘yield’, ‘good’, ‘Milton’, ‘Friedman’ and so on. My sense of it (though of course I’m lifting the quote from Cleaver’s primer) is that Friedman is talking about predictive results; and obviously that’s not all economics amounts to (as he’d be the first to emphasise).

    I guess I’m just trying to pre-empt the obvious (slightly positivist) criticism that non-obviously-concrete economic claims are worthless. But I’d be suspicious of doing so in the way you suggest, because saying economic results are results full stop seems dangerously close to falling into the positivists’ hairy grip. My inclinations run more toward the ‘more in heaven and earth, Horatio’ end of the spectrum.

    All depends on how capacious our sense of economics is, of course. Perhaps we should order some pizzas and crates of ale, and spend a cheerful weekend clarifying our terms.

    Comment by praxisblog — May 10, 2007 @ 8:27 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Create a free website or blog at

%d bloggers like this: