Yet another example clarifying why I cannot see economics as a science. They don't even use the same words. Stiglitz refers to the theories of his fellow economists. But a theory in science is reserved for extremely well-supported explanations, ones where the evidence is so overwhelming that there is effectively no disagreement, like the theory of gravity. Economists use the word in the layman's sense of "a guess".

As far as I can see, there are no theories at all (in the scientific sense) in economics.

@Stormsinger, it's a social science, which means that ultimately you have a very severe limitation -- you can't really run experiments.

We can't roll the clock back, put strong banking regulations in place in the US, and then see whether that would have helped.

Instead you have to look at how different economies reacted to similar issues, and try to draw conclusions by way of comparison, like Krugman does in this article where he compares the US, Ireland, and Canada during the current crisis.

You also have to look at historical situations (lots of people talking about the Great Depression and how studying that can inform what we do now).

I understand all that, really I do. It's precisely -why- I don't consider the social sciences to -be- sciences. They're studies, philosophies, schools of thought, just like Aristotle's version of physics. But they simply don't meet the some of the most basic requirements (repeatable experimental tests of hypotheses) to be science. Now, many of these schools -could- run such experiments, but in most cases I don't see how you can do so in an ethical fashion and get meaningful results. Maybe in a hundred years or so, they'll have worked around these limitations, but not today.

I find it amusing that if you ask six economists/psychologists/sociologists for specific predictions about a specific situation, you're most likely going to get seven different answers.

Exactly, it's a social science, which should be treated entirely separately from traditional science which can perform experiments that isolate factors and not just rely on historical data. This doesn't change the fact that we're undoubtedly better off with economic theories than we are without them.

The trouble occurs when people who use them are not apprehensive enough of their limitations, specifically that they don't have the same rigorous backing of proper scientific theories. The assumption of perfect markets (perfect information, no barriers to entry, no anti-competitive power) is simply that, it will never exist in real life, people should treat it as such. That doesn't make the notion of the invisible hand any less valid in the vast majority of cases, as it can be readily demonstrated to occur in virtually every free market industry. Unlike traditional sciences though, there are clearly occasional exceptions or hiccups brought about by the assumptions the idea underpins and they should always be kept in mind.

I'm not wholly convinced that we're better off with them than without, since they're primarily useful in providing after-the-fact explanations, rather than before-the-fact predictions (in other words, they seem to be little more than rationalizations). But I'll agree that the biggest problems come from those who place undue faith in -any- of the schools of economic thought.

My own view is treat them exactly the same as you treat your horoscope in the newspaper...as cheap entertainment. Then ignore them and use common sense.

Common sense and history make it pretty clear that unregulated banks are disasters just waiting to happen. So regulations that separate banking from speculation and force open the banks activities are pretty much a no-brainer in my book.

...or as I like to call it, the invisible hand job. No happy ending.