Dr Geoff Davies, keen to crack on with solving the world’s problems, makes a critique of neoclassical economic theory, seeking to explain why the results of its application are so bad (see previous post on this subject). His second post is not as straightforwardly incorrect as the first in his series, because several of the criticisms he makes are ones with which I agree, or are true in a trivial sense, or are true but don’t have the significance he attaches to them. In addition, discussions about theory aren’t as amenable to simple refutation like his first post about empirical evidence.
“It is no crime to be ignorant of economics, which is, after all, a specialized discipline and one that most people consider to be a ‘dismal science.’ But it is totally irresponsible to have a loud and vociferous opinion on economic subjects while remaining in this state of ignorance.” -Murray Rothbard
That quote is simultaneously arrogant and elitist and true to some degree. I think anyone who can use basic arithmetic can have a valid opinion about economics, but I do find it frustrating when people don’t bother to inform themselves about the basic concepts in economic thinking, then proceed to dismiss centuries of thought on the subject. Unfortunately, Dr Davies is a member of the latter group.
To maintain mathematical tractability, the [neoclassical] theory makes simplifying assumptions about people and firms.
It would seem that Davies, “an experienced real scientist” (his words), chose to forget that all theories of complex systems make simplifying assumptions. In his own field of mantle dynamics, as in all physics, models based on simplifying assumptions are used as a matter of course. This isn’t a weakness, but a necessity. If one is concerned with understanding fundamental processes or dominant patterns, there is no point in developing a model that can mimic every eddy of molten rock, or every economic transaction. And that’s assuming it is even possible – such a fine-grained model would be mind-bogglingly complex, as convoluted as the real thing, and thus would defeat its own purposes as a model!
The misconceptions continue when Davies moves from the metaphysical level to the actual composition of the neoclassical model of economic decisions:
It assumes we are narrowly rational and that we can foretell the future. It assumes we have access to all relevant information for free, and can assimilate its implications immediately. It assumes we are brute materialists. It assumes there are no social interactions. It assumes there is a limit to economies of scale, based on constraints peasant farmers used to face.
Some of these I can map to genuine assumptions, others just seem to be made up on the spot, or are perhaps the result of a profound misunderstanding of economic principles (as per the Rothbard quote). Let’s take them one at a time.
- The assumption of rationality is a perfectly reasonable starting point for a critique of economics. This is typically the first thing people object to, and you can see why – all of us perceive a multitude of seemingly irrational decisions by others, and might even acknowledge the occasional irrational decision by ourselves. However, it can be defended on two fronts. Firstly, the judgement call of ‘rational’ or ‘irrational’ is inherently subjective. What might be rational for you, might not be for me, because of our differing ideas about what methods and aims are appropriate for our circumstances. At its heart, the assumption of rationality is an assumption that people will pursue their self-interest, and that must perforce be individual in nature. Secondly, I will freely admit that people appear to behave irrationally sometimes, but even if they are indeed doing so, most of us are acting rationally most of the time. It is hard to imagine a society functioning where this was not so. For the purposes of modelling human behaviour, rationality (as opposed to antithetical options) would seem to be the only possible assumption to make.
Further into Davies’ post, he expands on this critique, citing “herd behaviours” as one example of supposedly irrational behaviour. There will be plenty of “experienced real scientists” who would contest that claim, such as biologists and psychologists. In a world of limited information, paying attention to what other people are doing and following the crowd can be a successful strategy. Indeed, there is a lot of empirical evidence to suggest that aggregated knowledge can be more accurate than individual judgements.
- “and that we can foretell the future” – this is flat-out false. In trying to understand where the heck Davies got this particularly bizarre idea, the only thing I can think of is rational expectations theory. Contrary to Davies’ laughable interpretation, all that this assumption says is that people will try to make predictions of the future, and in aggregate won’t be far off. That is very different to being able to “foretell the future”! I actually hope that I have misunderstood Davies’ meaning here, because otherwise it would mean that he is worse than ignorant.
- ‘Perfect information’ is the typical way of describing the next assumption. Davies is right in this instance – it is assumed that all the actors in the model know all the prices (or whatever), and can access that knowledge at zero cost. However, this is only true of the more basic models. Apparently Davies is unaware that they do get more complex than the first-year undergraduate examples found in textbooks. It is the equivalent of the frictionless surface (or infinite plane, uniform density, perfect vacuum, etc) of physics problems. Transaction costs, as economists call them, are a prominent feature of the vast majority of models actually in use.
- I don’t know why “brute materialists” merits an entry. It is true that economic models assume we want more for less (a corollary of the rationality thesis), but there’s nothing to suggest it must be money or material goods that are sought. Typically economists lump people’s preferences into a category called ‘utility’, which encompasses everything that humans might consider part of the Good Life, including social interactions. One need only read pop-economics books like Freakonomics to realise that economic tools can be applied to infinitely more things than just money.
- The reference to economies of scale is particularly obscure. Reading further into Davies’ post, he alludes to the presence of “oligopoly or monopoly” as somehow proving an unspecified neoclassical assumption wrong. Without more detail about what his ideas really are, I can only surmise that he is referring to (and dismissing) the concept of diseconomies of scale.
We are all familiar with economies of scale, the natural growth of the division of labour. We live it every day, by specialising in producing one set of goods (broadly speaking) and trading for others we need, and organising ourselves into groups that do the same thing. The efficiency gains are truly enormous, and are the basis of modern economies. Diseconomies of scale become evident when organisations get so large that counter-acting effects kick in – lines of communication multiply, slowing decision-making and response times; incentives change as people become more disconnected from what are supposed to be the central aims of the organisation; identifying necessary work from unnecessary work gets more difficult, leading to duplication of effort; and the phenomenon so well-known it has its own cliché – “too many cooks spoil the broth”.
Dr Davies appears to suggest that these are not important effects, because… well, because large corporations exist. I’m struggling to come up with a generous interpretation of his proposition, but it’s not easy. To be clear, the existence of large corporations is not proof that those effects are negligible. Diseconomies of scale are only one of many factors influencing size and market power. Davies’ first example, Microsoft, essentially owes its dominant position to first-mover benefits due to the unique history of PCs, and network effects.
Thus, for example, in 1987 stock markets fell thirty or forty percent in a day, though thirty percent of the world’s factories had not been bombed overnight.
If stock prices were simply a measure of future company earnings, this would be a valid criticism. But they’re not. Dr Davies exhibits a perhaps wilful ignorance (for rhetorical effect) of the true nature of stock markets, which are primarily a measure of what people think other people think about future company earnings, a very important difference.
But this is a minor problem compared to what Davies is using stock markets as an example of – disequilibrium. If I may backtrack, Davies contends at the beginning of his post:
With enough assumptions like this, you can deduce, using clever mathematics, that a market will balance all supplies with all demands and the economic system will come to an equilibrium.
The simplest and most simplistic of economic models – graphs of supply and demand curves – do have an intersection that is the implied equilibrium. Rational market participants with perfect knowledge will inevitably drive supply and demand towards that point. However it would be a profound misunderstanding to think that this is the end of the story. If nothing else, economists emphasise the dynamic nature of economies. The static picture painted by that kind of abstracted models are in fact no more than a snapshot in time of an ongoing, ever-changing process. In reality (and in the models actually used by economists), the forces pushing demand and supply towards equilibrium are always over- and under-shooting it in a continuous off-kilter balancing act. Those forces have to contend with what are called frictional factors, which brings me to Davies’ last central claim.
If any one of those assumptions is violated you predict very different behaviour of the economy. If the behaviour is very different then the central theoretical conclusion, that a free-market economy comes to an optimal equilibrium, is lost. Lost with it is the basis for all the free-market rhetoric.
The concept of frictional factors is as old, if not older, than neoclassical theory, dating back to at least Walras. This fact is a blow to Davies’ assertion about simplifying assumptions, but let’s leave that for now. What I am concerned with is the total lack of evidence for his claim that the slightest deviation from the (supposed) assumptions means radically different outcomes. He does give a whole paragraph of possible reasons why equilibrium might be disturbed, all of which I have addressed earlier in this post, but he makes no attempt to either calculate the models with the deviations (very hard, so that’s understandable) or to find out if other people have done so (reasonably easy, so quite damning).
As it turns out, this is basically what economists do for a living. There are thousands of scholars, publishing in dozens of reputable journals, who – contrary to the insultingly dismissive opinion of Dr Davies – spend most of their time thinking about the sorts of problems that he apparently believes are devastating original insights. Their results show that modifying the underlying assumptions can change the output of the models, but most definitely does not invalidate the broader conclusions. The data gets messier, but the patterns stay the same.
Yet again Davies’ arguments share many characteristics with those of global warming deniers, something I’m sure he will be mortified to realise. It’s all depressingly familiar – the ignorance of basic concepts, the refusal to acknowledge that experts are aware of and grappling with the obvious concerns, cherry-picking, the danger of a little learning, and liberal doses of the Dunning-Kruger effect.