Common sense, economics, and beliefs

I see Scott Sumner has come out saying that we shouldn’t necessarily judge economic outcomes on whether they meet a “common sense” test, giving the example of an individual looking at social security.  Bryan Caplan has criticised this, claiming the common sense is the foundation of all reason and stating that in the example Scott gives by stating the Scott is only making the argument by making the assumptions sound more plausible.

I’m not going to come down the middle in this, I’m going to disagree with both of them.  The focus on an individual decision regarding something that effects only the individual (superannuation) misses the point of common sense beliefs – and isn’t the best way for Scott to make his claim regarding the difference between “truth” and “plausibility”.  Remember, beliefs include our beliefs about others – and how this corresponds to (and can be valued as) aggregate action.

I’ve written about common sense and economics before. To quote wikipedia.

Common sense is a basic ability to perceiveunderstand, and judge things which is shared by (“common to”) nearly all people, and can be reasonably expected of nearly all people without any need for debate

As a result, common sense is a theory of belief formation – not just beliefs regarding your own actions, but beliefs regarding the actions of those around you.

If we just took “common sense” reasoning, we are accepting the current base of beliefs – a base that is founded on history contingent outcomes, and descriptive elements that involve biased categorisation.  If we pin ourselves to an argument being one that involves beliefs that are “common sense” or “not common sense” then we end up with two divergent camps – one that thinks common sense is worthless, and one that thinks it is everything (note psychology and philosophy have had these debates as well).  I’ll quote my post for the rest of this:

Of course, these definitions are not actually mutually exclusive – and actual economic thought about common sense includes both of them. The clearest example of this is Rubin’s work on folk economics (REPEC).

However, when it comes to thinking about what economists do (rather than just their view on what common sense entails) the requirement to sit in one of these ‘common sense camps’ becomes even weaker. The exceedingly harsh conclusions of the two views above (common sense sucks, common sense rules) don’t actually follow from the premises involved – economics describes situations where common sense beliefs are right, and when they are wrong!

And this is the kicker. Economics is a social science – it is applying the scientific method to help describe and understand social phenomenon. It is not about saying whether beliefs are right or wrong, it is about trying to create knowledge to help us form beliefs.

Now, given that the formation of beliefs is an important part of describing social phenomenon (since choices are based in part on beliefs, and social phenomenon are the result of choices), the issue of belief formation does appear in full force. However, this does not imply that we have to make an a prori choice to state that common sense is wrong or right – instead it suggests we need to think how beliefs, and expectations, are formed and (where possible) use data to help us find what appears to be the most appropriate assumption.

This is why criticising economics as only the study of common sense, or stating that it is ignorant of common sense, is a weak criticism.

However, it is also important to note that stating that policy conclusions are ignorant of ‘common sense beliefs’ can actually be a more poignant criticism – as it indicates simply that we do no agree with the set of beliefs and/or value judgements involved in the policy conclusion.

Summary for this debate – the economic method allows us to discuss social outcomes and the existence of social beliefs in a descriptive way that helps to frame trade-offs.  The choice of the “type” of beliefs used when forming policy can be seen as another form of value judgment.

In this way “assumptions” about beliefs matter – and have to be made in a transparent way (as they are).  Instead of relying on “common sense” notions alone, which often reinforce bigotry, economists allow a great degree of respect for the individual.  It is not reducible to common sense or independent of the plausibility of the assumptions used.

And before you say “ok ok, but the ‘core’ assumptions are common sense” (as Bryan does at the end of his post) I would note this argument is a touch ridiculous.  Common sense and scientific consensus are infact two different things – the core assumptions of a discipline have been determined by debate, evidence, and a history contingent process within the discipline not broad “common sense” of beliefs.  They are beliefs formed by people within that field due to specialisation, a small subset of all people – when common sense beliefs are believed to be share by most!

Note:  None of this said “what matters if prediction not the realism of the assumptions” a la Friedman.  That argument could be made as well when considering elements of macropolicy – but it is generally one I’m not a fan of in the policy space due to the Lucas Critique.