Almost Right

From the WSJ

"The Wealthiest 5% Grabbed Created Most of the [sic] America’s Gains"

There, fixed it for you.

Metaphors for No Free Lunches

From the excellent 'The House Wins', by OK GO, a wonderful metaphor for the strong form of the No Free Lunches principle* :

Ice age upon catastrophic ice age of selection and only one result has trickled in...

The house wins.

Oh the house always wins.

If evil were a lesser breed than justice after all these years the righteous would have freed the world of sin.

The house wins.

Oh the house always wins.

The house wins, and you lose. No matter the game, no matter the circumstance, no matter if you're sure you're figured out a system - doesn't matter, the house will win.

I love it! It's almost as good as Bob Dylan's metaphor for opportunity cost.



*Granted 'No Free Lunches' is already a metaphor, so this is more of a meta-metaphor.

A life well-lived

John Harsanyi was one of the founding fathers of Game Theory, a fascinating part of economic theory that examines how agents make strategic choices when the value of their actions depends on the choices of others. In 1994, he won the Nobel Prize in Economics for his work, along with John Nash and Reinhard Selten. Although most people would rate Nash's contribution as the greatest (in discovering Nash Equilibria, the foundation of all Game Theory) Harsanyi added a lot to the understanding of games of incomplete information, greatly expanding the range of situations that can be analysed using Game Theory tools.

Long before this, he also came perilously close to be killed in the Holocaust. He was a Hungarian Jew in Budapest in 1944, which was a pretty damn hazardous situation to be in. As he describes it:

In that November the Nazi authorities finally decided to deport my labor unit from Budapest to an Austrian concentration camp, where most of my comrades eventually perished. But I was lucky enough to make my escape from the railway station in Budapest, just before our train left for Austria. Then a Jesuit father I had known gave me refuge in the cellar of their monastery.

Before winning a Nobel Prize, John Harsanyi escaped both from the mutha-f***ing Nazis in World War 2. But it gets better. He stayed around in Budapest, and became a professor in Sociology:

But in June 1948, I had to resign from the Institute because the political situation no longer permitted them to employ an outspoken anti-Marxist as I had been.

He hated communism, and wasn't afraid to say so. In 1948. In Budapest. Think about that. Would you have had the balls to say that stuff?

But [Anne, his eventual wife] was continually harassed by her Communist classmates to break up with me because of my political views, but she did not. This made her realize, before I did, that Hungary was becoming a completely Stalinist country, and that the only sensible course of action for us was to leave Hungary. 

Actually we did so only in April 1950. We had to cross the Hungarian border illegally over a marshy terrain, which was less well guarded than other border areas. But even so, we were very lucky not to be stopped or shot at by the Hungarian border guards.

John Harsanyi also escaped from the mutha-f***ing Communists in 1950. Holy Hell! 

He got to Australia, didn't speak the language and his credentials were worthless.So what did he do? Complain to a diversity consultant? Protest that the government wasn't supporting him enough? Lobby for Hungarian language education and civic notices?

As my English was not very good and as my Hungarian university degrees were not recognized in Australia, during most of our first three years there I had to do factory work. But in the evening I took economics courses at the University of Sydney. 

No, instead of bitching, he worked his ass off in a factory and studied economics at night.

And then won a God Damn Nobel Prize in the subject. 

If Milton Friedman and Chuck Norris had a son together, that son would be John Harsanyi.

John Harsanyi, for being a wicked economist, an opponent of tyranny and a thoroughly hardcore dude, you are hereby post-humously inducted into the Shylock Holmes Order of Guys Who Kick Some Serious Ass.

I was reading about this, and two things occurred to me.

The first is how close the world came to never having heard his insights. It's probable that someone else would have figured out similar ideas eventually, but the economics profession and the world benefited greatly from having John Harsanyi pass through this vale of tears. Any small number of mishaps, and he would have been one more unknown death statistic for the Nazis or Communists.

And that is the second point is this - how many John Harsanyis didn't make it? How many more Nobel prize winning insights were lost to the butchers of Auschwitz, the cowardly scum of Nanking, and the rest of World War 2's parade of horrors? The Holocaust in particular killed millions of Ashkenazi Jews, a group notable for their unusually high intelligence. It is a virtual certainty than many future prodigies were killed over those years, whose insights might have advanced human knowledge and welfare in ways we'll never know. What a horrible, criminal waste of humanity. The deaths of the many ordinary people in World War 2 are no less tragic for their lack of extraordinary potential. But it's hard not to wonder at what might have been. 

Business Models I Can Respect

Every now and then, I'm struck by a company where my overwhelming response is 'Damn, I can't believe they're selling things that cheap.'

One of those is White Castle (most famous for the 'Harold and Kumar go to White Castle' movie). I had a friend in Chicago who thought this place was awesome (he also had a penchant for Michelin star restaurants, so he spanned the whole culinary universe). I went there with him once, and the burgers there were nasty. They were so bad, in fact, that it taught me something I didn't imagine was possible - that there is in fact fast food of a sufficiently low quality that I wouldn't want to eat it. Everything else - McDonalds, KFC, whatever - retained a feeling of tasty indulgence. It's low quality and bad for you, but boy is it tasty. Not White Castle. I don't plan to ever go back.

But that's not what's most important. What's truly amazing is that their base hamburger costs 63 cents. 63 frigging cents! (At the time it was 50 cents). And they're making a profit selling it for 63 cents. A profit large enough to keep them in business. Truly astonishing.

Reader, I stand in awe of a business model that can profitably produce burgers at scale for 63 cents. Sure, it's tiny and low quality. But suppose you exempted me from every health code and labor regulation. Suppose you let me make the meat out of diseased offal and scrapings from the abattoir floor. Suppose you let me make the bun using moth-eaten, moldy bread. Suppose I didn't even have to keep the quality high enough that anyone would actually purchase it.

Even then, I still can't see me way to making a hamburger for less than 63 cents.

White Castle can, and for that I take my hat off to them. I don't want to eat there, but they retain my strong respect and regard.

Marginal Costs to the End-User* of various items in Sweden

-Triple Heart Bypass:  0 Kroner.

-13 years of primary and secondary schooling: 0 Kroner

-Undergraduate university degree: 0 Kroner

-Taking a piss in a mall in Stockholm: 10 Kroner.

From this, I can only conclude that as a society they're far more worried about the moral hazard problems with urination than with, say, wasting years on a degree in gender studies. If we make urination free, people will just be going back to toilets every 5 minutes, flushing them over and over for fun!

Talk about a strange preference ordering.

(from a conversation with The Greek, which also ranged over the question of whether 'Stockholm Syndrome' was so-named because when you get to Stockholm you initially think it sucks, but gradually get used to it).

*!= "Free"  -  Weak Form No Free Lunches applies, as always.

Asian Marriage Rates: Who? Whom?

The worst assumptions are those that you don't even realise you're making.

The Economist recently had an article discussing how marriage rates in Asia are dropping. The sub-heading tells you pretty much all you need to know:

Women are rejecting marriage in Asia. The social implications are serious

That would be serious.

But the first sentence actually hides two claims, not one. These are:

1. Marriage rates are dropping, and

2. This is primarily the result of women actively deciding to avoid marriage.

So what is the evidence the article marshals in favour of each of its claims?

The first one seems on fairly solid ground:

Marriage rates are falling partly because people are postponing getting hitched.The mean age of marriage in the richest places—Japan, Taiwan, South Korea and Hong Kong—has risen sharply in the past few decades, to reach 29-30 for women and 31-33 for men.

Almost a third of Japanese women in their early 30s are unmarried; probably half of those will always be. Over one-fifth of Taiwanese women in their late 30s are single; most will never marry.

Okay, this probably doesn't surprise too many people - it's happening everywhere. But how about the second claim? How do we know this is a choice by women?

That’s partly because, for a woman, being both employed and married is tough in Asia. Japanese women, who typically work 40 hours a week in the office, then do, on average, another 30 hours of housework. Their husbands, on average, do three hours.

I'd want to see where these numbers came from - are the women who are working also doing the housework, or are they disjoint sets ( i.e. the average woman does both work and housework, but this is made up of some who only work and some who only do housework). But I'll give them the benefit of the doubt - let's assume married Asian women who also work still do a lot of housework. Anything else?

Not surprisingly, Asian women have an unusually pessimistic view of marriage. According to a survey carried out this year, many fewer Japanese women felt positive about their marriage than did Japanese men, or American women or men.

Okay, so women who are married report being unsatisfied. From this the author concludes that women who aren't married are avoiding marriage based on anticipating the same feelings. Let me translate this into a metaphor in a different context to see if any alternative hypotheses might more readily present themselves:

My friend Timmy got a new bicycle for his fifth birthday, but now he doesn't play on it much. I don't have a bicycle, but seeing that Timmy doesn't use his much any more, I stopped wanting one.

See the problem? How about buyer's remorse as an alternative? In other words, it's entirely plausible that a lot of women desire marriage beforehand, but only once they get there do they realise it's not all it's cracked up to be. Does this sound like human nature to you?

Now, I'm not claiming massive evidence in favour of this proposition either. But let's be clear - the article doesn't even countenance the possibility that marriage rates in Asia are dropping because of choices by men, not women. 

Let's go out on a TOTALLY CRAZY LIMB HERE, and propose the following meth-and-LSD-induced alternative hypotheses, purely to play devil's advocate:

-Women are more attractive in their 20s than in their 30s. The decision to pursue education and careers in their 20s makes them seek out marriage later only when they are less physically attractive, at which point men are no longer interested. Such women who miss out on marriage are filled with regret.

-Men in Asia are avoiding marriage because divorce law favours women too much, and thus they see marriage as a raw deal for them.

Are these right? Who the hell knows?! But ask yourself the following - do we really have strong reasons to prefer the author's 'every trend is the result of informed and rational choices by women'  hypothesis over either of the above? The unstated assumption, which the author probably doesn't even realise they're making, is that women are always the who, and men the whom, in Lenin's famous formulation.

If you wanted to tell the hypotheses apart, wouldn't you start by surveying men and women who aren't married, and asking them if they're actually looking for marriage? Or asking them if they're actively avoiding marriage, and if so, why?

Given how flimsy the evidence, let's evaluate the article's conclusion:

Relaxing divorce laws might, paradoxically, boost marriage. Women who now steer clear of wedlock might be more willing to tie the knot if they know it can be untied—not just because they can get out of the marriage if it doesn’t work, but also because their freedom to leave might keep their husbands on their toes. Family law should give divorced women a more generous share of the couple’s assets. Governments should also legislate to get employers to offer both maternal and paternal leave, and provide or subsidise child care.

If this trend is all the result of women's time-consistent choices to avoid marriage, then yes, you would need to make marriage more attractive to women to reverse this trend, and these policies might accomplish this.

If this trend is the result of women's time-inconsistent choices to inadvertently avoid marriage, then most of these policies would do very little. The only one that might work (and highly ironically) is forcing unsubsidised maternity leave and child care on employers  - the most immediate effect would be businesses avoiding hiring women of child-bearing age. The resulting female unemployment may end up pushing women away from education and jobs and into marriage. I don't think this is what the author had in mind though.

If this trend is the result of men's time-consistent (or time-inconsistent) choices to avoid marriage because they think it's a raw deal, then these policies (especially giving more assets to women in divorce) would be disastrous.

Repeat this type of article dozens of times, and you start to realise why the endorsement of a particular policy position by The Economist does not fill me with reverence and awe.

The Weak Form and the Strong Form of "No Such Thing as a Free Lunch"

The expression 'there's no such thing as a free lunch' is one of the best summaries of free market economics. It was popularised by Robert Heinlein in his book 'The Moon is a Harsh Mistress', and was also the title of a book by Milton Friedman.

With such an illustrious pedigree, it may seem churlish to suggest a slight expansion of this principle. Nonetheless, I propose breaking the idea into two parts - the weak form, and the strong form. I think this is because people use the term to describe two separate cases.The Weak Form of No Free Lunches states that there is no lunch that you can eat that is without cost to someone. The Strong Form of No Free Lunches states that there is no lunch that you can eat that is without cost to you.

The weak form is particularly true in the case of governments, who notionally try to balance lots of people's interests. There is no such thing, for instance, as 'free healthcare'. At the margin, it may be free to you as the end user, but it is paid for in taxes. Even if you don't pay any taxes yourself, there's still opportunity cost - less can now be spent on schools, or roads, or national defense. Everything is paid for by someone, even if it's not you. The weak form is especially good for attacking pie-in-the-sky idealists and people fond of misleading labelling of social programs. The weak form, if true, injects sober realism into a lot of debates. Stimulus programs will not pay for themselves. Then again, neither will tax cuts. The biggest violations of the weak form would seem to be comparative advantage and gains from trade - if we both specialise and trade with each other, we really are both better off.

The strong form says that even when something seems free to you, it's actually imposing a cost, most likely in the form of something that the lunch provider is getting out of you - a favour, a chance at selling you something, a chance at guilting you into paying more than the cost of the lunch, a  chance to waste more of your time with a discussion that you wouldn't have otherwise undertaken. This is stronger than the weak form, because it says that you should be wary of accepting anything that seems free until you understand what the other person is getting out of you. Welfare, for instance, is an apparent violation of the strong form but not the weak form. It's paid for by someone, but not by you. Or is it? Being on welfare for long periods tends to be psychologically very damaging in ways that people don't anticipate. Which is why communities where everyone is on welfare tend to be drug-addled, violent hellholes that nobody in their right mind would want to live in. It's also why rich parents worry a great deal about giving too much money to their children in case it turns them into entitled brats. Gifts too - if you're Australian cricketer Shane Warne and an Indian bookmaker wants to give you thousands of dollars for providing freely available information about weather and pitch conditions, you'd better believe that ain't a free lunch.

The strong form is useful for combating fools about to part with their money. 'Money Back Guarantee' != No Risk. Church Soup Kitchen Lunch -> forced attendance at sermon. Backroom deal with politician to get tax break for your company -> being strongarmed for donations and political support later on.

Like most weak and strong form laws, there are more apparent violations of the strong form than the weak. The weak form is true in the vast majority of cases. The strong form is true more often than you might like to think.

A proxy for the wealth of a society

The number of stray dogs walking around.

The more there are, (and the more emaciated such dogs look), the poorer the society.

By this metric, India is about the poorest place I've been to, which might not be exactly right but is probably close. Delhi had tons of stray dogs, and they were some of the sorriest-looking creatures I'd ever seen: mangy, rib cages sticking out, and devoid of energy, lying as if dead by the side of the road. It made me realise how incredibly healthy virtually all of the 1st world pet dogs were that I'd seen - I hadn't actually seen a truly neglected dog before.

Belize , Mexico and Honduras were all also bad, but with slightly fewer and not as decrepit-looking animals, but they were still fairly ubiquitous.

Funnily enough, this measure would also place this one Indian (as in Native American) Reservation as the poorest place I've seen in America - the dogs there looked thin but not unhealthy, and there were a few of them around. I've never seen stray dogs in noticeable numbers anywhere else in the US that I've travelled. I'm not sure how accurate this would be as a measure of cross-sectionl wealthwithin the US, but it's probably not too far off.

Correlations, man. Though you throw them out with a pitchfork, yet they return.

The Most Frustrating Local Monopoly

Economists typically have an ambiguous attitude towards monopolies. When the monopoly arises because one company is providing the single most popular good or service, opinions often differ between educated and well-meaning people as to what to do. Some people worry about the potential for predatory pricing and other schemes to shut down new entrants, and want the government to step in. Others figure that the the market will eventually take care of it - you can sue Microsoft for bundling Internet Explorer with Windows to shut down Netscape, but eventually IE will be replaced by Chrome and Firefox anyway.

But the worst kinds of monopoly are those granted by government fiat. Because then there is little possibility of new entrants displacing the artificial monopoly.

My entry for the category of 'most frustrating and insidious government-granted monopoly' that I've come across is currency exchanges at airports.

At certain airports (Sydney being one that comes to mind), the government allows only one company to be the exclusive currency exchange for the whole airport. I think in Sydney it's Travelex. So you can walk around and find multiple currency exchanges, but they're all Travelex.

And this is an incredibly devious thing to do.  Travelex makes extra money by charging a higher spread (buying at lower prices and selling at higher prices) than they could sustain if they had competition two metres away.

The reason they can get away with this is twofold. Firstly, most people need foreign exchange pretty soon after leaving the airport. But even more tricky is the fact that most people have very little idea what is a competitive exchange rate at any point in time. Even if you check the exchange rate on the internet, this will tell you the midpoint. But what's a reasonable spread if you want to exchange $200? Unless you've been paying attention to what other places in previous airports were offering that day, it's not at all clear.

And this is how Travelex launders its monopoly profits. It's one of the few cases where most people don't have a strong sense of the fact that they're being ripped off.

So why does the government do this?

Simple - they charge Travelex higher rents in order to be the monopoly provider for the airport. In essence, they split the monopoly rents with Travelex. If they're auctioning off the right to be the monopoly provider, my guess is that the government ends up getting most of the rents.

So here's a hearty up yours to the Government of New South Wales for ripping off citizens. Again.

[Update]: Loyal reader BW pointed out to me that Sydney airport is actually owned by MAp Airports, a subsidiary of Macquarie Bank. In addition, some reading around informed me that this privatisation was carried out by the Australian Federal Government, not the NSW Government. So the main post unfairly besmirches the reputation of the New South Wales Government, which is deeply dysfunctional for a number of reasons, but this is not one of them.

BW is quite right, of course. Monopoly pricing is not any more appealing when carried out by a private corporation (especially in an arena, like airports, where there's not exactly a viable threat of competition).

It turns out I'm not the only person to notice this trend at Sydney Airport. According to Wikipedia:

In March 2010 the Australian Competition and Consumer Commission released a report sharply critical of price gouging at Sydney airport, ranking it fifth out of five airports. The report noted Sydney Airport recorded the highest average prices at $13.63 per passenger, compared to the lowest of $7.96 at Melbourne Airport, while the price of short-term parking had almost doubled in the 2008–09 financial year, from $28 to $50 for four hours. The report also accused the airport of abusing its monopoly power.

Let the record stand corrected - up yours, MAp Airports!