December 31, 2012

Forecasting v. All-Else-Being-Equal policy analysis

Philip Tetlock's 2005 book Expert Political Judgment raised amusing doubts about the accuracy of professional public affairs forecasters. 

Tetlock was particularly derogatory about dogmatic "hedgehogs." Economist John H. Cochrane suggests in a Cato Unbound piece, however:
It was once hoped that really understanding the structure of the economy would also help in the sort of unconditional forecasting that Gardner and Tetlock are more interested in. Alas, that turned out not to be true. Big “structural” macroeconomic models predict no better than simple correlations. Even if you understand many structural linkages from policy to events, there are so many other unpredictable shocks that imposing “structure” just doesn’t help with unconditional forecasting. 
But economics can be pretty good at such structural forecasting. We really do know what happens if you put in minimum wages, taxes, tariffs, and so on. We have a lot of experience with regulatory capture. At least we know the signs and general effects. Assigning numbers is a lot harder. But those are useful predictions, even if they typically dash youthful liberal hopes and dreams.
Doing good forecasting of this sort, however, rewards some very hedgehoggy traits. 
Focusing on “one analytical tool”—basic supply and demand, a nose for free markets, unintended consequences, and regulatory capture—is essential. People who use a wide range of analytical tools, mixing economics, political, sociological, psychological, Marxist-radical and other perspectives end up hopelessly muddled. 
Keeping analysis “simple and elegant” and “minimizing distractions” is vital too, rather than being “comfortable with complexity and uncertainty,” or even being “much less sure of oneself.” Especially around policy debates, one is quickly drowned in mind-blowing detail. Keeping the simple picture and a few basic principles in mind is the only hope.

In other words, there is a big difference between macro and micro forecasting. Macro forecasting needs all the factors, and nobody understands all the factors. Micro forecasting, in contrasting, can benefit from theory. Micro forecasting is policy analysis, which benefits from using the economic concept of "all else being equal."

For example, in the late 1970s, the state of California passed a law allowing local municipalities to impose rent control on landlords. In microeconomic theory, this is not a good idea. And indeed, living in Santa Monica in 1981-82, as a newly minted economics major, I could observe rent control in action: my landlady treated me with all the warmth of a Serb hosting German soldiers in WWII. She invested nothing in the upkeep of the apartment in the 22 months I was there. 

On the other hand, the rent was a helluva deal for a place where I could bike to the beach a few times per week. I had no intention of moving out if I could find a job anywhere in the L.A. area. I figured that in five years the apartment building would look like a wreck from neglect, but I'd probably still be legally squatting in it, writing my $250 per month rent-controlled check. (I slowly figured out that Santa Monica had been undervalued in the past because, before antibiotics and smog, newcomers to Southern California had wanted to live inland in drier places like Pasadena to avoid TB; but by 1981 the whole climatic logic had long-since reversed, so demand for Santa Monica apartments was finally soaring.)

In the long run, it's clear that rent-control is not really a good idea. Not many municipalities in California still have it. On the other hand, ones that do, such as Santa Monica, West Hollywood, and Berkeley, are typically not post-apocalyptic wastelands, either.

Indeed, the rent-controlled municipalities tend to be ones favored by geography, and thus attract smart, well-heeled people, who figure out ways to make it work.

To make an all-else-being-equal forecast about the impact of imposing a policy of rent control on Santa Monica, you only need to know that it would cause less investment in apartment buildings than would happen otherwise.

On the other hand, to make an unconditional forecast about what Santa Monica would wind up like, you'd need to understand a lot more factors, especially about human capital, which economists are not very bright about.

And that's the bigger lesson to be learned.

11 comments:

Anonymous said...

If truth be told, economists basically have an extremely lousy and unenviably bad track record for forecasting.
Apparently they are a little better than random coin tossers (tosser being the operative word). They are no better than a general member of the public.
- Such is an academicism (I'll never ever dignify it be misusing the word noble term 'science' to describe it), that is based on opinions rather than facts.
As the old adage goes 'opinions are like @$$holes - everyone has one.'
That is all you need to know about economics and economists.
- Most of them wouldn't know a fact if flat out squarly whacked them on the face like a 10lb wet halibut.

peterike said...

Rent control effects depend on just how the controls are put into place. If you rent control a significant portion of the housing stock but not all of it, the rents on the uncontrolled portion will soar. If you impose rent control on any new buildings, you will soon find nobody is putting up new buildings, and you get a huge housing shortage and super-tight vacancy rates.

At the macro level, rent control is always a very bad idea for a city, unless you are that rare city which is a super-magnet for the ultra-wealthy (i.e. New York) and you don't mind displacing your middle-class or sticking them into shared apartments outside of the city center (I wonder what percentage of Brooklyn apartments are shared by multiple tenants?).

At a micro level, rent control is a great deal for anybody on the receiving end, like Mia Farrow and her $2,900 a month, 11 room apartment overlooking Central Park at 73rd street (the apartment used in "Hannah and Her Sisters").

alonzo portfolio said...

before antibiotics and smog, newcomers to Southern California had wanted to live inland in drier places

This was the shock when I read Raymond Chandler: the lowlifes lived at the beach, though he never connected it to consumption. And by the way, Berkeley is a pit. One of the more amusing Letters to the Editor occurred after The New Republic pointed out that University Avenue looked like something out of Eastern Europe.

MQ said...

Cochrane's article is total ideological bullshit. He's basically saying that you should just pay attention to the conclusions of the tinker-toy econ 101 model and ignore the complexities of the real world. For example, in the real world, countries have had some fantastic successes in boosting exports and industrializing by using tariffs, currency manipulation, and other trade barriers -- this is the story of the U.S. up through WWII, Germany in the 19th century, and Japan/China/Korea more recently.

This statement by Cochrane --

"We really do know what happens if you put in minimum wages, taxes, tariffs, and so on."

is totally absurd, there is no definitive statement that can be made on the effect of these things as some kind of general law of nature.

Anthony said...

Rent control has done an amazingly good job of ethnically cleansing Berkeley of blacks without moving in large numbers of hispanics.

The effect is actually quite predictable, because landlords who have limits on the rent they can charge will prefer to rent to wealthier (and generally less-destructive) tenants. But nobody predicted this beforehand. People did start to notice after about a decade of rent control, but it's much easier to predict the past than the future.

Ed said...

I agree with Steve's and peterike's comments about rent control. Generally a bad idea, but not unmitigated bad and it makes a huge difference how, where, and why its imposed.

I live in a rent-stabilized apartment in Manhattan. People tend to make lots of comments about "rent control" in New York, but they miss things like the difference between rent control and rent stabilization (all the shocking stories about wealthy people living in cheap apartments are about rent control, not rent stabilization), how few apartments in New York are subject to this, the fact that both are being phased out slowly, plus the laws were put in place due to a post-World War II housing emergency that has never been resolved. At worst, you can say that the laws prevented a real policy resolution to the shortage of housing in New York, but repeal by itself won't do anything to alleviate the shortage, which existed before the laws and actually was the original reason for their enactment.

These laws seem to keep around low income high education people in a city that otherwise would be priced out, but hit middle income people.

I agree in general that microeconomics offers genuine, though somewhat limited, insights into how economies work, but that macroecnoomics is almost entirely useless.

beowulf said...

Its probably relevant to note that John Cochrane is to economics as astrology is to astronomy.
http://delong.typepad.com/sdj/2012/01/matthew-yglesias-and-paul-krugman-read-john-cochrane.html

headache said...

Here in Germany about 60% of the population rent. Rent controls only exist indirectly in that yearly increases are capped. Generally there is a conspiracy between municipalities and developers/large landlords in that land for new developments is restricted driving up development costs and thus rents, securing investments and returns. AN the municipalities make a cut through tax and land prices. All this at the expense of the renters.

So I am in favour of rent controls, but this conspiracy between politicians and developers/large landowners also needs to be busted.

Engineer Dad said...

I agree that California rent control best serves 'first come' renters while harming later arrivals.

My childhood friend, a Cal Computer Science graduate, moved into a rent-controlled Berkeley apartment in 1976 near University avenue and kept it into the mid 1990s to store his music collection and gear while living in Europe.

Anonymous said...

Forecasting would be a lot easier and closer to the mark if there were fewer taboos. Because of taboos, many forecasters have to pretend to be blind to certain inconvenient facts.

For example, with the rise of black power, an honest forecaster would have foreseen the danger of black violence and corruption. But the taboos forced the 'experts' to prop up the grand narrative of the rainbow nation led by wonderful Nelson Mandela.

So, there's a lot of forebanning or forecensoring or forecastrating or forecastigating going on.
And this is true even in a free society. In a way, it could be more insidious in a free society because the notion that we are 'free' makes us blind to all the subtle and not-so-subtle means that prevent us from saying what needs to be said.

And it's not just in the West, and it's not just about the future but everything. Take Japan. It's supposed to be a 'free democracy', but there are so many cultural taboos on behavior.

This scene in TAMPOPO is funny but its implications are serious.

http://www.youtube.com/watch?v=RRVLqUpHDJE

In a truly free society, the guy who knows more would be applauded, praised, and promoted, but in Japan, what matters most is maintaining the rank system. So, even though the 'lowly' aid knows more, his boss tries to suppress the demonstration of superior knowledge. So, even those who know more are supposed to act like they don't know nothing. It explains why rigid Japan lost out in the high-tech game. So much initiative and energy are lost.

ben tillman said...

I think your readers would understand you if you said, "ceteris paribus".