Thursday, May 27, 2010

Blame God

Due to heavy floods in Poland, water completely destroyed graves in a small town Catholic cemetery, creating biohazard. Local parishioners blame the priest who is responsible for maintaining the graves (and who the parishioners have paid fees to do so). The priest said they should blame God, because the heavy rains were his idea.

Would this be the first Catholic priest to openly acknowledge the validity of the Argument from Evil?

The window is broken and water is seeping in

In the past week and a half or so, Poland has been hit by severe floods. The death toll is now at about 15, property damage at about $3 billion, and total amount of value lost may be an order of magnitude larger. But not to worry; as a report from Polish radio station RMF24 tells us, floods are good for the economy. In my own loose translation:
All those houses and roads that were destroyed will have to be rebuilt. This means more business for construction firms, as well as manufacturers who supply them. This situation will create some new jobs, at least for a few months.
There are some false beliefs that will just never die, and broken window fallacy is apparently one of them. Look, natural disasters are not good for the economy. The money that the flood victims will have to spend rebuilding is money they would have spent on something else. There will not be more consumption, it will just shift to other sectors. But there sure is a hell of a lot less property.

Some of the more sophisticated (Keynesian, if you will) broken-windowers may wish to point out that the money that will be spent on rebuilding might have a stimulating effect if there is an underutilization of resources. To which I'll reply: take a look at the victims. Almost all of them ar poor or lower-middle income people. They certainly were not underconsuming, because they can't afford to save money.

Wednesday, May 26, 2010

We need to be spiteful to defeat spite

A couple of days ago I wrote about a disturbing situation in Polish public healthcare: The government can only pay for some number of certain medical procedures, a number which is lower than the demand for these procedures. This means that some patients can't get them in a timely manner--they have to wait in line. Now the curious thing is that scarcity only exists on part of the government funding these procedures, not hospital capacity. The hospitals have the necessary equipment, and the doctors have the necessary time, to actually perform all procedures that are demanded--if only someone paid them for it. The government can't afford it, but some of the patients can. However, those patients are not allowed to do so, because public hospitals and doctors working for them are forbidden by law to accept any payments on behalf of the patients. Some politicians favor changing the law so as to allow waitlisted patients to pay for their procedures.

The stunning thing is that some (perhaps most) major politicians are against such a change. Why is this so hard to believe? Because allowing patients to pay would create one of those extremely rare situations that economists and ethicists call "Pareto-improvements." Pareto-improvement is a change in the social status quo such that no one is worse off, and some people are better off, than they were under said status quo. A Pareto-improvement means there are no losers, only winners (though it doesn't necessarily mean that everyone wins). Allowing waitlisted patients to pay would be a Pareto-improvement because doctors would be paid more than they are right now, and more patients would be treated than there are right now. No one would be less happy than they currently are, and some people would be happier. Ethically speaking, this really should be a no-brainer. And yet many politicians oppose it.

I think speaking out against Pareto-improvement is always profoundly immoral. Why? Because those who do so reveal that they are prepared to act on spiteful moral preferences. They want to help people they like, sure; but they also want to see people they don't like hurt--even if doing the latter means doing less of the former.

One of the ways in which societies remove immoral behavior is through social punishment. There was a time when it was quite all right to be overtly racist in speech and actions. It isn't anymore. (I realize this is a huge oversimplification but, as often, I am talking about what's true on average.) You can still have racist preferences, sure, but you can't really act on them, as this usually leads to unpleasant consequences in terms of your social standing. I wish acting on spiteful preferences would be ostracized this way as well. Here's a paradox though: one of the reasons we humans even have spiteful preferences is to ensure evolutionary stability of social punishment. In order for the society to be able to punish thieves, for example, most group members have to be ready to punish those that stole not only from them, but also from other people, even if such punishment entails costs to themselves. And that is a form of spite.

Tuesday, May 25, 2010

Perceptions and reality

Here are some quotes from a WJS article reporting on an FBI report about recent trends in violent crime:
Violent crime fell significantly in cities across the U.S., according to preliminary federal statistics (...) The incidents of violent crimes such as murder, rape and aggravated assault was down 5.5% from 2008, and 6.9% in big cities.
And further:
(...) Phoenix's most recent data for 2010 indicated still lower crime. For the first quarter of 2010, violent crime was down 17% overall in the city, while homicides were down 38% and robberies 27%, compared with the same period in 2009. Arizona's major cities all registered declines.
In other words, Arizona's voters demanded, and brought about, a change in immigration policies based on their perceptions and not reality. Perceptions that are completely and utterly wrong. Shouldn't they be at least a little ashamed of themselves?

Vote, shout, riot or pay

From the Economic Logic blog:
Surajeet Chakravarty and Todd Kaplan use (...) arguments to compare simple voting and shouting matches. In the latter, those who care more about the outcome put more effort into shouting. Thus, if there is a lot of variance in opinions, shouting better reflects marginal utility and yields something closer to social optimum. How is this shouting concretely expressed? It should be a signal that is costly and in some way wasteful. In France, it is demonstrating on the streets. In the United States it is donating to political campaigns. In Thailand it is erecting barricades. Usually seen as major inefficiencies, all these can actually be good.
Well, some of those forms of shouting actually are major inefficiencies. Striking and rioting do play a role of costly signals and may in this way bring about outcomes closer to optimal, but they have huge negative externalities. Violent protests and strikes are costly not only to those who care about the issues, but also impose significant costs on innocent bystanders (i.e. people who do not share the protesters' preferences). Simple shouting matches proposed in Chakravarty and Kaplan's paper are much better in that respect.

Here's one curious thing though: there is a criterion for determining intensity of preferences that is much more efficient than screaming your brains out. It's willingness to pay. Why don't the writers propose a model in which voters simply bid for votes? Is it because suggesting that people pay to vote is great sacrilege against one of our sacred values, and therefore it's bad politics to submit a paper like that?

Sunday, May 23, 2010

Contradiction? Who cares!

People can profess a belief while at the same time saying things that directly contradict it. Here's a vivid example of this fact: a recent press interview with the former Polish secretary of public health. Given who the interviewee was, the interview topic was of course public healthcare system in Poland, so to give the proper context I'll say a few words about how this system works. The system is funded through payroll taxes; a certain fixed percentage of earned income is withheld and then deposited into the budget of a central government agency called National Health Fund (Narodowy Fundusz Zdrowia). This agency employs all the doctors working in the public sector, manages all public hospitals and clinics, and provides funding for all publicly available medical services; consumers of public healthcare are not required to pay any out of pocket fees.

Dr. Balicki is currently managing one of the largest public hospitals in Warsaw (and in Poland, for that matter). He uses an example from the day to day affairs of his hospital to illustrate problems with Polish public healthcare. Balicki says his hospital performs 700 cataract removal surgeries a year. That's how many the National Health Fund is able to pay for. However, the demand is much higher: each year, 1600 patients sign up for this surgery. Balicki says his hospital actually has the resources to perform 1600 cataract removals without much problem, but it can't because the law forbids public hospitals to accept any money on behalf of the patients, so those 900 patients a year who cannot have their surgery paid for by the Fund have to get in an ever growing line.

A few moments later, Dr. Balicki briefly talks about the idea of creating a "basket of government-guaranteed free medical services." As of now, the law "guarantees" every citizen that the state will provide any available medical service free of charge. Of course, no government in the world can afford to keep such a promise; hence the need to ration public healthcare in Poland. The idea of a basket of free goods is basically to assess how many and which medical services the government is actually able to pay for, and guarantee those. The rest would have to be paid for via the private sector, either by insurance premiums or fee-for-service payments. Balicki is opposed to the idea, and his argument is that "the government cannot restrict access to medical services by providing some of them for free while charging for others."

Think about this for a moment. If the government says it only has money to provide certain explicitly enumerated medical services for free and that other services will have to be paid for by someone else, that's restricting access to healthcare. But if the government guarantees to provide all services for free even though it cannot afford to do so and is forced to ration them (just as it happens in Balicki's own hospital with cataract surgeries), that is somehow different than restricting access to care. It seems that, to Balicki, "restricting access" means not "causing a situation where some people who want certain services have to go without" but "explicitly charging fees for services." But wait; in the very same interview, Balicki himself gave an example of a situation where 900 people a year can't access a service, even though surely some of them would be willing and able to pay for it. In other words, a situation where explicitly charging a fee for a service would actually increase access to it.

The human brain is an amazing device. It's capable of holding so many blatantly contradictory beliefs...

Snake oil tax

In his recent NYT column, Thomas Friedman writes:
[After 9/11] I suggested a $1-a-gallon "Patriot Tax" on gasoline that could have simultaneously reduced our deficit, funded basic science research, diminished our dependence on oil imported from the very countries whose citizens carried out 9/11, strengthen the dollar, stimulated energy efficiency and renewable power and slowed climate change.
Wow; that's amazing. A single policy could do all that? Well, to be sure, it could do some of the things on the list. It could reduce trade deficit (though not by any significant amount); it could increase research funding; it could lead to developing more renewable energy sources; and it would definitely slow climate change. But some of the things on the list are on the list because... who knows why, exactly. A gasoline tax would not diminish our dependence on foreign oil, at least not in the short run. Paying more for gas would decrease our consumption of it, which would make us buy less oil--but there's no reason why it should make us buy less foreign oil relative to ours. The tax would also not stimulate energy efficiency (again, at least not in the short run): the very reason why gas is cheaper than other energy sources right now is because it is more energy efficient.

And then there's the issue of strengthening the dollar. This one left me sort of dumbfounded. How exactly would a gas tax accomplish this feat? What's the proposed economic mechanism behind it? And while we're at it, what else would the tax do, cure cancer? Oh wait; we already know it could. It's number two on Friedman's list.

Overheard in new york

A conversation in a NYC coffee shop at 7.30 on a Sunday morning between a customer (female in her early twenties) and a clerk (female in her early thirties).

Customer: You are so friendly and happy, how do you do that this early in the morning?

Clerk: I got a good night's sleep; I went to bed at eleven.

Customer: Oh my God! How did you do that?

Clerk (still smiling, but with an ever so sightly condescending look): When you work all day you get real tired, you know.

Wednesday, May 19, 2010

Out of sight, out of mind

Here's a quote from one of the major candidates in Polish presidential elections (translation mine):
The biggest advantage of our public healthcare system is that it doesn't categorize patients as rich or poor. Both the poor and the rich are being treated with the same professionalism and respect.
This is true, but it's far from the whole truth. While Polish public healthcare does not categorize people as rich or poor, it does make distinctions, and it doesn't really treat everyone equally. Public healthcare in Poland is completely nationalized; the government runs both health insurance and provision of medical services. The system is based on strict price controls: the government regulates prices of basically everything that has to do with public healthcare. In practice, of course, it means that said prices are artificially kept at levels much lower than market clearing rates. This creates severe shortages of supply. Healthcare in Poland is rationed (quite severely when it comes to certain services). Given that, how can you possibly say that public healthcare treats all patients with "the same professionalism and respect?" It doesn't. Some patients it does not treat at all.

Thursday, May 13, 2010

Tail-heavy aircraft, tail-heavy normal distributions

Air Midwest Flight 5481 from Charlotte/Douglas International Airport to Greenville-Spartanburg International Airport is a short (30 minutes) commuter flight, operated by a very small turbo-prop Beechcraft 1900D aircraft. On the day of the accident in January 2003, the plane was filled to capacity (19 passengers and 2 crew members). Few seconds after takeoff, just after landing gear was up, the plane's nose pitched extremely steeply upwards and the pilots were not able to counter it as they found they didn't have enough elevator movement to force the nose down. The plane stalled, lost lift, and dropped to the ground like a rock, killing everyone on board.

NTSB investigation concluded that the crash was caused by a conjunction of two unrelated problems. By itself, neither would have led to the accident, but unfortunately on that day they paired up. First, a flawed maintenance repair of the plane's elevator left it with compromised down movement, which meant the pilots were not able to stop the upwards pitching of the aircraft. Second, the plane was too heavy, which is what caused it to pitch up in the first place. The maximum load of the plane was determined by the airline based on averages: 175 pounds per passenger and 20 pounds per piece of luggage. Based on those directives, the crew's calculations showed the plane to be within limits to take off. The problem was that those averages were determined based on surveys conducted in 1936. When NTSB investigators painstakingly calculated the actual weight of passengers and bags aboard Midwest 5481, it turned out that unbeknownst to the crew the plane was overloaded by 580 pounds.

Of course, using weight averages from 1936 when everyone and their mother knows people have consistently been getting heavier is outrageous negligence on part of the airlines. Due to the finding of the investigation into Midwest 5481 crash, NTSB started lobbying FAA to force airlines to use actual weights for determining whether or not planes can take off. FAA ordered the airlines to do a new survey and update their averages. The results were quite shocking: the average weight of an airline passenger A.D. 2004 turned out to be 195 pounds. (As a side note: this is interesting in itself. The average weight of a male adult in the U.S. is 191 pounds; of a female adult--164 pounds. Why is the flying public so much heavier than the general public?) At any rate, Air Midwest updated their average passenger weight to 200 pounds, which meant that the capacity of a Beechcraft 1900D had to be adjusted downwards to 17 passengers. NTSB investigators felt this was not enough, and are still advocating for regulation requiring to use actual weights.

Which finally brings me to the main point of this post: does it make sense to use acutal weights instead of averages? In other words, is the increase in safety worth the increase in costs? The answer of course depends on two things:

1) How much would it cost to use actual weights; and,

2) What is the probability of overloading a plane if only averages are used?

Knowing precisely nothing about how the airline industry is run, i can't even begin to answer (1). Would using actual weights require installing scales and actually weighing each passenger or not? If so, how much would that cost? Would it mean ticket prices would have to go up and if so, by how much? Etc-these are all questions I can't answer. So I'll just answer (2), as that one is just a bit of trivial probability. But in order to do this, we need to know one more thing: standard deviation of weight. I take it to be 40 pounds. (I couldn't find this statistic, so I'm just making it up as I go along at this point.)
So the official passenger weight capacity of a Beech 1900D is 17 x 200 lbs = 3400 lbs. By central limit theorem, total weight of a group of 17 people is a random variable that follows a normal distribution with mean 3400 lbs and standard deviation of 40 x sqrt(17) = 165 lbs. Now we need to make an additional assumption: how much over 3400 pounds do we have to go to consider the plane to be potentially dangerously overloaded? Surely 10, or even 50 pounds over the limit will not make a whole lot of difference. Let's assume then that exceeding the prescribed limit by 5% or more is unacceptable. This means that we can't allow total passenger weight to go over 3570 pounds. So the question now is: given that we're not weighing anyone and just using averages, what's the probability that, if 17 pax board a Beech 1900D, their total weight will exceed 3570 pounds?

Calculating the z-score and plugging it into the normal distribution we've just found, we can see that this probability is over 15%. That is way too high; completely and utterly unacceptably high. NTSB investigators are absolutely right: with aircraft so small that they board under 20 pax, averages simply won't do; variance is too high. Something else needs to be worked out.

However, as planes get bigger, central limit theorem does start taking care of us. Suppose we're trying to find out the probability of overloading a 60-seater. Sticking with the assumption that the weight limit needs to be exceeded by 5% to consider the plane too heavy, we get the overload probability of just under 3%. Still too high for my taste. In a 100-seater, however, it's 0.6%, and in a 200-seater it's 0.02%. That's probably acceptable. And the bigger they get, the more acceptable it'll be; plus, the bigger the plane, the higher the costs of the alternative (i.e. of actually weighing passengers).

Monday, May 10, 2010

Put your money where your candidate is

A lot of Polish people who are interested in internal politics think that major Polish polling companies are politically biased and are "cooking" their results to help their favored politicians. Specifically, they believe that polling companies favor the liberal party called Civic Platform (Platforma Obywatelska), and are actively trying to damage their opponents, a conservative party called Law and Justice (Prawo i Sprawiedliwosc). One of the reasons behind this belief is what happened in 2005, a year in which Poland held both parliamentary and presidential elections. Law and Justice won the parliamentary elections, and their candidate (Lech Kaczynski) won the Presidency. All three major polling agencies predicted otherwise; worse still, the actual results were outside their reported margins of error. This suggests the polling companies do indeed have problems. Those could include any combination of intentional bias (which I don't think is actually the case), dishonest voter response, or flawed sampling. At any rate, when polling agencies drop the ball, prediction markets could step up to the plate. I would really love to see a political prediction market in Poland; for if polling companies are indeed biased, and if the rough magnitude of that bias is common knowledge, then candidate option prices would provide a much better prediction of their electoral performance than polling numbers.

June 20 this year, Poland will hold presidential elections (they were actually scheduled for November but had to be expedited due to President Lech Kaczynski's tragic death). There are several candidates but only two serious ones: Bronislaw Komorowski representing the Civic Platform, and late President's twin brother Jaroslaw Kaczynski representing Law and Justice. The polls currently show Komorowski's support to be at about 48%, whereas Kaczynski's to be at about 37%. The world's largest prediction market, Intrade, has opened betting on Polish presidential elections. Now, on the assumption that the poll numbers are again incorrect, I don't actually expect Intrade to do better. The main traders would be people outside Poland, who will be getting most of their information from Polish poll numbers, so Intrade prices will most likely track polls instead of correcting them. But, (again, if the polls are wrong), Intrade will give those who think polls are wrong a chance to make some money. If you think Polish polls are biased in favor of Komorowski, go on Intrade and buy Kaczynski right now.

In fact, I may do so myself. Regardless of what the polls say, the current price of the Kaczynski option seems way too low (the price is $2.10, which means the market estimates his probability of winning to be 21%).

It's not that you're wrong; it's that you're stubborn

Aristotle maintained that women have fewer teeth than men; although he was twice married, it never occurred to him to verify this statement by examining his wives' mouths.
That's Bertrand Russell speaking. The greatest sin is not to hold wrong beliefs, but to refuse to verify them empirically. In its own domain, science has burned through this prejudice thoroughly and with extraordinary results. Controlled experiment is one of the most important ideas humankind has ever had.

But outside of the domain of science, anti-empiricism reigns, and its rule is deadly. Contempt for experiment kills innocent people. I'm not exaggerating, and I don't mean it in the sense of "It used to kill innocent people in senseless wars in Medieval Europe;" I mean it in the sense of "It's killing innocent people right now in the United States judicial system."

Scores of people have been imprisoned, and yes, also executed, based on beliefs, professed by forensics "experts," which were no better than urban legends. Beliefs that, had they been actually tested in controlled experiments, would have been discarded once and for all. Experiments show that flaking of concrete in a puddle-like shape, or multiple ignition points occur frequently in accidental fires, and therefore cannot constitute evidence of arson. Tell that to all those people who have been convicted of arson, and sentenced to long prison terms, or death, based on precisely those myths. Experiments can show that one cannot tell with much precision at which time bruises were inflicted just by looking at them. Tell that to people wrongfully convicted based on that ludicrous belief.

Et cetera, of course. There are many more such cases and such myths. The idea that beliefs need to be verified empirically, no matter how strongly they are held, or how commonsensical they may sound, is held in deep contempt by our entire justice system, by its each and every branch. Which of course means that the lofty notion that everyone is "presumed innocent until proven guilty" is really just a (rather cruel) joke.

Tuesday, May 4, 2010

There is no money tree anymore

I understand why Greek citizens protesting spending cuts implemented by their government are upset, but I don't quite get what their protests are supposed to accomplish. What exactly do they think is supposed to happen? There is no money. It's not like the Greek government has a stash hidden away somewhere; it is actually bankrupt. Sure, it's getting huge foreign grants, but that money has to be used to pay down at least part of Greece's gigantic liabilities; no one is giving Greece money so that it can continue on its old consumption path. Which, as shown by the bankruptcy, was pretty much unsustainable.

Sunday, May 2, 2010

Sampling, shmampling. Science, shmience

Here, via God Plays Dice, is a link to an article about the census and statistical sampling. The article talks about random sampling done after census data is collected in order to estimate the direction and magnitude of measurement error involved in census data collection. The broad issue is that it's possible, through sampling, to discover said measurement errors, but it's not possible to correct them, because the law forbids to use statistical sampling in the data collection process. Scientists have for a long time been lobbying to change the law in that respect, but their arguments have invariably been met with the same response. Here, an example of this response is provided by Senator Judd Gregg (R-NH):
You take guesses based on what you think is the best political outcomes that you want, rather than counting people who actually exist.
One sentence, two very ignorant misconceptions. Misconception one: statistical sampling is "just a guess." Misconception two: it's physically possible to "count people who actually exist." Infinite accuracy in observation is a very naive illusion; "counting people who actually exist" is riddled with measurement error, which is actually worse than sampling error. Why? Because, if sampling is done correctly, sampling error is easily quantifiable, which means that estimates come with known margins of error. (Here is a truly excellent blog post explaining why sampling works, i.e. why, as random sample gets "large enough," sampling estimates approach true population values.)

I may be wrong in my observation, but it seems to me that this particular objection to statistical sampling is more likely to be heard from Republicans rather than Democrats. It seems to stem from the Republicans' general distrust of the scientific method: sampling is "just guessing," evolution is "just a theory;" basically anything that isn't the result of direct observation but involves inference of any kind is somehow suspect. Exactly why this sort of distrust is more of a Republican phenomenon I do not know.

To wrap things up, a truly hilarious line from the post at God Plays Dice:
As a combinatorialist I admire the theoretical elegance of our country's once-a-decade exercise in large-scale, brute-force combinatorics.

Saturday, May 1, 2010

Regressive innumeracy

This example of innumeracy comes from an article published in a 1995 issue of a Polish right-wing magazine called Fronda (don't ask me how I found it). In the article, the author argues for a regressive income tax (i.e. income tax such that the higher your taxable income is, the lower tax rates you have to pay) on the grounds that progressive taxation punishes for trying to get rich, and even flax tax does not reward it because when income tax rate is flat, high income taxpayers still pay more money (in absolute terms) than low income taxpayers.

The belief that rich taxpayers should pay less money than poor taxpayers shows profound commitment to a certain social philosophy. The belief that regressive income tax rates guarantee that high income taxpayers will always pay less than low income taxpayers shows profound confusion about middle school-level algebra. I'll leave it to the readers to construct a numerical counterexample.