I find it rather strange that even predictions that should be fairly clear-cut and straightforward (e.g. "A ceiling on rents reduces the quantity and quality of housing available.") only get around 9/10 agreement. As a non-economist, I would have thought that a seemingly simple (and testable!) question like that would be like polling physicsts with the question "Does F = ma?". And I would sure hope that more than 93% of physicists would say yes.
Well, unlike with physical laws, economic "laws" rarely describe direct causal mechanisms, but rather generalizations about what happens when you poke a multi-agent system, whose behavior is often rather complex. It's hard to test, tests often throw up quirks, it's nearly impossible to control for all confounding variables, etc.
In the rent-control case, for example, here is one (paywalled and somewhat dated) dissenting view: http://www.springerlink.com/content/k75072652rkxut61/. Pointed out mostly as an example--- it's quite possible that particular paper is now obsolete and adequately responded to. It's a common approach to dissenting from economic conclusions, though: you admit that a particular widely held belief is true in a particular model, but then argue that the model doesn't sufficiently correspond to reality, e.g. because it doesn't incorporate certain important effects.
Indeed. In particular, I've noticed that nearly all of the cities I've lived in and love (SF, Berkeley, NYC, Montreal) have rent control, whereas many cities that don't have rent control (Dallas, Houston, Phoenix) I've really had no interest in visiting.
The historical justification of rent control is that it allows for security of tenure: hence, even renters can put down roots, and a community can have time to develop. The communities I love seem to have required more time to develop than the cycles of the real estate market would allow. Even if I personally pay more rent, I benefit from those communities.
How do the economists model that? They probably don't. They do answer one question: the housing stock in Dallas and Houston is dramatically more affordable and probably built to a higher standard than the equivalent in SF and NYC, but the quality of life is significantly different (to my personal tastes, sacrificed.) The economists are answering the wrong question.
Rent control does seem to create quite an affordable housing shortage, though. One wonders, however, if ultimately it spurs on a lot of upscale development, since the premium you can charge for new units is so much higher, relatively. (giving you SF's South Beach...)
Edit:
Cato's article on the subject describes the change of Boston and Cambridge from a rent controlled to a free market system, on January 1, 1997.
Rent control creates a shortage of affordable housing. Sure, it might make cities nicer because with rent control only wealthy people can afford housing, so all of the low-income people live elsewhere. Think about living in SF on a teacher or fireman's salary, and compare that to anywhere else.
The problem with arguing about economics is that for some people, this is a desirable outcome -- rent control creates nice cities. For others, the outcome is undesirable -- rent control makes it more difficult for the poor to afford housing.
Nobody can argue that rent control is good or bad as a matter of fact; this is an opinion. That's why economists don't model that, it's like trying to calculate whether chocolate or vanilla is a better flavor.
What can be argued is the stated intention vs. the outcome; when a politician argues that rent control will keep housing affordable for low income people, we can state with a very high degree of certainty that he's wrong.
The ugly side of this is that most people don't care whether the politician has got his facts straight. They want the outcome he's promising.
A ceiling on rents reduces the quantity and quality of housing available. (93%)
Counterpoint: surprising to me is the number of economists who "agree" with this statement.
First of all, it does not compute, intuitively, to me as a logical "AND" statement. "Quantity" and "quality" are two completely different things.
"Quality of housing" would be a function of renting vs. owning status (e.g. paying rent vs. paying a mortgage), further subjected to the degree of separation between the tenant and the actual mortgage/title holder. "Quantity of housing" would be a function of information asymmetry between "buyers" and "sellers", and obviously price (determined by the kind of information asymmetry).
>"Quantity" and "quality" are two completely different things.
They both cost to maintain or increase though - and rent control limits the returns of increasing either; which reduces the owner's incentive to maintain the quality of existing housing or to build new.
That question is a tad bit more subjective than "F = ma" (an economist from a Marxist school of thought would probably disagree that is is entirely clear-cut), and it would depend entirely on the specific way it was asked and exactly what counted as an "agree".
Professionals will tend to hedge their answers when they can conceive of any kind of scenario that would contradict an absolute like "A ceiling on rents reduces the quantity and quality of housing available.". A ceiling of 2 trillion dollars/sq. foot would likely not impact on the quality or quantity of housing in the US, to make up a wild counter-claim off the top of my head.
Are you sure more than 93% of physicists would say yes? I would say "what if the mass changes" or "what if the speed of the mass is relativistic" -- then you have to use something else, like F = Integral[dp, dt], and F = ma no longer holds. Rather, it turns out to be F = (gamma)^3 m a.
Assumptions, assumptions! They plague everybody, not only economists. F = ma only in the non-relativistic limit with a constant mass.
Before you say "of course we are thinking in the non-relativistic limit", be mindful most of the rest of the universe, moving around at sizable fractions of c compared to us, disagrees with you.
The 1/10 perspective makes a strong claim about the price elasticity of supply of housing. It would seem to be testable, but while it's easy to measure elasticity in a controlled environment, for something complex like the market for housing, doing so might entail some assumptions that 1/10 economists don't choose to get on board with.
