…(Y)our looking changes it… Because our minds… our minds get in the way. Looking at something changes it. They call it the “Uncertainty Principle”.
The Man Who Wasn’t There
The position that Rosenberg and Curtain take is clearly untenable (‘What Is Economics Good For?’ The Stone, NY Times, August 24). They have no support for their claims and manage to contradict themselves.
They argue that economics looks like science, does what sciences do, talks like science, but then claim that it is not a science because its subject matter is not scientific. And the same goes, a fortiori, for practically all social sciences. That is startling in and of itself. Assuming that can even somehow make sense, how do we know this? Economics has low predictive power, so it must be the case that there are no stable regularities in economies, societies, politics – there are no laws to social behaviour. If there were laws, we could predict what e.g. a specific policy intervention would achieve, but we cannot. In physics, we know e.g. how to put a satellite in orbit around Mars, while in economics we do not know e.g. how to manage monetary policy.
Why are there no stable regularities in human affairs? This is how Rosenberg and Curtain explain: ‘(T)he domain of economics includes a wide range of social “constructions”… They are made up of unrecognized but artificial conventions that people persistently change and even destroy in ways that no social scientist can really anticipate. We can exploit gravity, but we can’t change it or destroy it. No one can say the same about the socially constructed causes and effects of our choice that economics deals with.” This, as far as I can see, is all they give in the way of an argument and it surely is wrong even though it is not clearly written.
First, they claim that social sciences cannot predict how institutions will change for the very reason that they can change. That is obviously question begging. Perhaps the argument is to be found in the claim that they, the institutions, are unrecognized, that is there is lack of information. But that would be why there are social sciences. The other argument might be that those are artificial. But that is also question begging. The fact that institutions are human made is a prima facie argument that they are predictable. The problem may be that those are conventions, but again that would work for their transparency rather than against it. So, there is nothing in the way of an argument to support the claim that economics lacks a subject matter that can be studied scientifically.
One argument could be that we do not know what we are doing. For instance, people are not perfectly or rational at all, so what they are liable to do is anybody’s guess. Perhaps it is all psychological, in which case, however, what people do would be rather predictable (this generalises to any, shall I call it, naturalistic argument). Or it is that this persistent institutional changes are unintended consequences of collective actions (consequences are unintended because, that may be the claim, they are produced by collective action). But that again does not imply anything about predictability. As an empirical matter the consequences of collective actions may routinely not be predicted by those engaging in them, but that still does not mean that they are unpredictable to the social scientist. Especially if these social, constructed institutions display causal dependences, even if they are of our choice, as the authors state. Indeed, in that case, even if there are no law-like regularities, if actions and consequences are causally connected, the latter will be predictable in principle. Indeed, predictability is usually taken to be one of the contributions of institutions and thus of sociology or sociologically inspired economics, political science, or game theory.
At the end of their piece, which is about what is economics good for, Rosenberg and Curtain suggest that ‘an economic approach’ has ‘much to contribute to the design and creative management of… institutions.’ Presumably because economic theory can predict the consequences of institutions and help the construction of those which ‘will protect us from the “knave”.’ That does not just contradict their claim about economics not been capable to predict, but it is a bit of a stretch when it comes to philosophy too. It also squarely contradicts their claim that e.g. monetary policy should be more like art or craft (presumably that is what creative here means), rather than guided by economic science. So, they do not have an argument and they contradict themselves. Is there something to the case anyway?
As they spend some time discussing Friedman’s claim that it is the predictions that matter and not the truth of the assumptions, what does that mean? What one thinks about that claim will of course depend on one’s theory of truth, but if one wants to know whether the theory will be descriptive or realistic, then the answer will be negative for some description, e.g. on anecdotal evidence, or some folk-theory of reality. That is what Friedman meant by saying that it did not matter whether the assumptions were true or theories were realistic and what Popper meant by saying that the rationality principle is surely false, but explanatory (see e.g. ‘The Rationality Principle’ in D. Miller, ed., Popper Selections, Princeton University Press1985: 357-365).
Predictions in economics as in sciences in general are indeed probabilistic. So, not necessarily testable by specific cases. Still, was it predictable that shadow banking which developed as the system of free banking would collapse? Or, which is the same thing, that the system of international finance that functions like the system of free banking will lead to crises and probably to a crash? I do not think that the answers could be anything but positive. After all, that is why there are central banks and indeed monetary policy.
But then can we predict that monetary policy will be mismanaged before the crash and that fiscal policy will not be relied on after the crash? The answer is yes. This is the contribution of political economy and political science. Is it to be expected that there will be no overwhelming support for countercyclical monetary policy before the crisis in the absence of inflationary pressures? And is it to be expected that there will be no overwhelming enthusiasm for increased public spending financed by debt? So, is the policy response, e.g. in the USA, with bailing out of the financial sector, lax monetary policy, and cautious fiscal response predictable? Yes, predictably, if the former two measures are somewhat successful and the recession does not turn into depression the push for more fiscal spending will not be strong enough to outweigh the worries about its distributional effects.
There is, however, a multiple equilibrium problem here, but that is where political science comes in and we should be able to predict with some probability that one or the other system of decision making was going to choose a particular fiscal response. It was not unpredictable how China would respond and how countries in the euro area would respond. That may not be in accordance with the best advice by the economic science, but that is more of a comment on the optimism of what can be achieved by the ‘design and management of institutions’.
For this normative task, the economic model, which is criticised as unrealistic and utopian, is actually the one usually relied on for ‘design and management’, i.e. for regulatory purposes. But that is not what it is good for. It is needed to check the consistency of theories, to enable explanations, and to support predictions. We assume that people are rational (in a very general way), that markets display some regularities, and that there is a prices system that insures consistency. Then we bring in reality and try to explain it and then rely on the explanations to predict the outcomes in changing circumstances. It is a misguided idea to try to regulate the reality in order to conform to the model or to reduce any set of data to the model. The latter is exactly opposite to the purpose that Rosenberg and Curtain assign to economics as a guide to political philosophy of institution building.
The interest in the issue of how scientific economics is and what is its predictive power comes from the colossal blunder that is current crisis. It failed to be predicted by most economists and was not prevented by policy makers like those in the central banks and in the ministries of finance whose actions are informed by economic science. The policy failure is not inconsistent with what is known in political economy, i.e. it was and still is predictable. Still, economics has to go back to the drawing board, but that is not because its subject matter is not scientific enough, whatever that means, but because some theories have been falsified and some policies were wrongly designed or implemented. That also does not support the claims that economics is not science.