About a little over two weeks ago, I came across this lovely quote by Robert Solow in the website of Arnold Kling:
I remember reading once tha t it is still not understood how the giraffe manages to pump an adequate blood supply all the way up to its head; but it is hard to imagine that anyone would therefore conclude that giraffes do not have long necks. At least not anyone who had ever been to a zoo. – Robert Solow [Link]
Then, I read a paper by Arnold Kling titled, ‘Macroeconomics: the science of hubris’ (Link). The paper is an easy read. It is easy to follow his arguments on the impossibility of macro-economic modeling. Here is a sample:
An almost limitless number of factors could affect key macroeconomic variables from quarter to quarter. For each factor, moreover, there are several potential specifications for the variable representing that factor. The variable might be entered into the equations as linear or nonlinear, de-trended or not, current or lagged, and so forth. Perhaps the effect of the level of average house prices is different from the effect of the rate of change of house prices. Ceteris paribus, a higher level might reduce demand, but a higher rate of increase might increase demand. The number of factors to be controlled for is further enlarged by ‘‘special factors,’’ such as the steel strikes or wage/price controls alluded to earlier. All things considered, there are thousands of plausible specifications of equations.
In conclusion, he compares macro-econometric models to the wrong map used by Swiss mountaineers lost in the Swiss Alps. They had the map of Pyrenees, a mountain range that borders France and Spain (I think). But, there is a big, big difference. They made it down safely from the mountain using the wrong map. The same cannot be said of macro-econometric modelers. They give wrong prescriptions, policymakers follow them and the consequences remain with societies for a very long time with no one being able to account for the costs of the wrong ideas. It is humanely impossible and for the most important reason that no counterfactual is possible to construct, in hindsight.
What the map of Pyrenees meant to the mountaineers is a different thing. That should be discussed separately. That indicates the power (or the susceptibility) of the mind to accept ‘placebos’.
Staying with maps, I must mention here the delightful review of Greenspan’s ‘The Map and the Territory’ by John Kay:
No economic model can describe “the world as it really is”; there are only provisional approximations that can be pragmatically applied in particular situations.
Many economists struggle to accept this. They suffer physics envy, believing (I think wrongly) that physicists employ models that do describe “the world as it really is”. If economics is to be a science, these economists think, those who practice the subject must identify truths that are similarly independent of time, place or context.
I would be even more circumspect of economic models. I would not venture to say that they are provisional approximations that can be pragmatically applied to certain situations. I would say that they are first approximations that are points of departure in our quest to understand economic reality.
His final blow on Greenspan is brilliant: underlying superficiality of thought concealed by the complexity of expression. WoW!
The fox, by contrast, lacks a grand vision and holds many different views about the world – some of them even contradictory….. Foxes carry competing, possibly incompatible theories in their heads. They are not attached to a particular ideology and find it easier to think contextually. [Link]
This approach to problems comes naturally to many Indians who have not crossed over to the Western mode of linear thinking. To appreciate many tenets and philosophies of Hinduism, one needs this fox-like thinking hat.
At this stage, it makes sense to link here – to put them all in one place – Emanuel Derman’s manifesto and Hippocratic Oath for economic modellers.
[Cross posted at ‘The Gold Standard’ blog here]