Cheap shots at Don Cherry

Today's National Post has a healthist condemning Don Cherry, one of the elder gods of hockey, for a hockey culture that encourages hits to the head and concussions. Nothing could be farther from the truth. Cherry has consistently argued for a physical form of the game, but one that's sensible: no hits to the head, no cross-checking from behind, and no stick play. So unless he's changed his tune a lot since I left Canada, this is just a cheap shot.

Cherry has previously argued that players wearing face shields are a danger to others as they pay less attention to keeping their sticks down on the ice. One of my honours students a year or so ago found that Cherry was right: players wearing visors were more likely to draw stick related penalties than were other players, all else equal. I'm hoping to get time this coming sabbatical to work up the results with Tim a bit more formally and get the paper out.

Growthgate and Climategate

William Easterly notes parallels between climate research and research on economic growth.

There were three steps in the the great History of Evolving Cluelessness:

1. Economists spent the past two decades trying every possible growth determinant in sight. They found evidence for 145 different variables (according to an article published in 2005). That was a bit too many in a sample of only about one hundred countries. What was happening is there would be evidence for Determinants A, B, C, and D when tried one at a time to explain growth. But the evidence for A disappeared when you also controlled for some combination of B, C, and D, and/or vice versa. (Interestingly enough, foreign aid never even merited inclusion in the list of 145 variables.)


2. The Columbia economist Xavier Sala-i-Martin and co-authors ran millions of regressions on all possible combinations of 7 variables out of the many possible determinants of growth. Skipping a lot of technical detail, they essentially averaged out the millions of regressions to see which determinants had evidence for them in most regressions. There was hope: some were robust! For example, the idea that malaria prevalence hinders growth found consistent support.


3. This new paper by Ciccone and Jarocinski found that every time the growth data are revised, or if the sample is changed to another equally plausible one, the results vanish on the “robust” variables and new “robust” variables appear. Goodbye, malaria, hello, democracy. Except the new “robust” determinants are no longer believable if minor differences between equally plausible samples changes what is robust. So nothing is robust.

There are two possible ways to describe what had happened over the past two decades:

1. The growth research was at least partially fraudulent, in that we researchers were searching among many different econometric exercises till we got the “determinants of growth” we wanted all along.


2. There was a good faith effort by us researchers to test different theories of growth, which led to some results. We didn’t realize until later that these results were not robust.

Description (1) would be a “GrowthGate,” but since so many people would be guilty (of “data mining”), and since we really can’t tell for any individual study or researcher whether it was (1) or (2), “GrowthGate” never became a story.

It's rather worrying if the Sala-i-Martin variables prove non-robust across new iterations of the Penn World Tables. Minor errors in data seem to blow the technique apart.

Genetic distance in war and economics

It turns out that folks kinda hate their genetic neighbours.

Enrico Spolaore was one of the keynote speakers at the Australasian Public Choice Society Meetings in Melbourne last week. I'd not before seen his work on genetic distance, but it's rather interesting.

Genetic distance measures the number of generations back you have to go before two populations share common ancestors. So if two populations diverged only a very short time ago, like the Danes and the English, their measured genetic distance is short; if they diverged a very very long time ago, like the Australian aborigines and the Mbuti Pygmies of Africa, their measured distance is long. While this is related to geographical distance, it's far from perfectly correlated: Canada's Inuit are closer to Tibetans than they are to any of the other Amerindians; the English are closer to the northern Indians than they are to the Lapps (Finish); the Mongols are closer to the Japanese than they are to the Chinese; the Indians of south east India are closer to the Italians than they are to the Thai people or the South Chinese.

Spoloare and Wacziarg find that this measure of genetic distance predicts whether two populations will go to war, after controlling for the usual set of determinants of conflict like geographical proximity, shared borders, income differences, religion, language, and so on. All else equal, the more two populations are genetically proximate, the more likely they are to go to war and the less likely they are to vote together at the UN. If current patterns of war have affected measured genetic distance, then causality may be wrong, but they use genetic distance as of year 1500 as an instrument.

I'd worried that results might be drawn from a few places that have been strategically important going back well before 1500. For example, if the bridge from Africa to Asia Minor via Sinai and Israel has been strategically important for thousands of years and if similar populations have lived around there for a long time, then the correlation of genetic distance and conflict could have things the wrong way round: frequent conflicts in strategic regions bring genetic mixing, and if the regions' strategic importance continues from well prior to 1500 to present, then it wouldn't be that folks want to fight with their genetic neighbours, but folks who fight a lot become genetic neighbours. Controlling just for having a common border or just for geographical distance wouldn't quite cut it. But they find that the effect also holds for country pairs that do not share a border.

Why might we fight more with our nearer than our more distant cousins? Spolaore suggests that genetic closeness makes it more likely that we'd be in conflict over rivalrous resources. I wonder whether we couldn't imagine a pleistocene explanation: if there's a fertility advantage to outbreeding but not too far (the sweetspot between inbreeding depression and outbreeding depression), then our ancestors 90,000 generations back on the Savannah who raided more closely related neighbours would have had a slight advantage over those who raided groups too genetically distant. Run the mechanism for 90,000 generations, and you've a population that's keyed to want to raid folks who are more like them.

In the May QJE, Spolaore and Wacziarg found that genetic distance from the United States explains cross country income differences after correcting for geographical distance, climate, transportation costs, and measures of social distance (historical, religious, linguistic). Again, they argue that genetic distance may be the best measure of "slowly changing genealogically transmitted characteristics, including habits and customs" - the bits of culture we can't adequately otherwise measure. That's certainly possible, and cuts against my evolutionary biology explanation above, mostly because it's hard to come up with an ev bio explanation of why genetic distance from the US would correlate with income differences. The best explanation I'd have would be that it's proxying for differences in average IQ: also somewhat genetic, but at some of the more depressed ends of the scale almost certainly highly environmentally influenced). But that would be a bit of a wash: there are genetically distant places above the US (Hong Kong) and far below the US (Equatorial Guinea) in reported average national IQ. Spolaore's culture explanation seems the more plausible.

Spolaore gave one of the best plenary addresses I've ever seen. If you get a chance to see him give a talk, go.

DeLong, then and now

DeLong and Lang's 1992 "Are all economic hypotheses false?" was a very nice contribution. Long story short: given the ease with which researchers can pick variables to ensure their chosen specification has significant results, and the near impossibility of publishing things that show statistically insignificant results, all published standard errors in economics (or any discipline where this happens) are wrong: they're all biased downwards, and a proper accounting for publication and researcher bias would have a hard time rejecting the null that all of them are insignificant.

And so all the recent shenanigans at Hadley had me thinking of DeLong and Lang.

Today DeLong approvingly cites Mark Kleiman.

Assume some climate model predicts that... temperature would rise 3° C by 2100. If the model were very accurate and precise, that might be 3°± 1°. If the mechanisms involved remain obscure and the data unclear--as is the case today – that might be 3°± 5°.... Given how bad a 3° increase would likely be, if we knew for sure that would be the outcome in the face of inaction there would be a strong agument for making big and expensive policy changes to prevent it from happening.

But following DeLong and Lang, it isn't that the confidence interval is just wider, it's that our current uncertainty (coming from Hadley) is to what extent the estimates have been picked from the top rather than the middle of the interval. Are all climate hypotheses false?

Wind and water are complements

Wind power can make sense if you also have a lot of hydroelectric power generation. Wind is of course highly variable and somewhat unpredictable, and the grid requires realtime constant matching of demand and supply to maintain stability. It's much easier to open or close the sluices a bit depending on wind than it is to ramp up or down a thermal plant.

If you don't have the flexibility of a hydroelectric system...well, here's some evidence from Europe, from Motl:

Every year, a huge excess of wind-generated electricity from Northern Germany causes problems to the grids in Czechia, Poland, Austria, Slovenia, the Netherlands, Belgium, and Switzerland.

A year ago, the wind was really strong and the first problems occurred. So they established a warning system. The problems are repeating in 2009, too. For example, it was planned for the last week that there would be a total of 130 MW of electricity flowing from Czechia to Germany. However, the actual budget was 1300 MW in the opposite direction. The balance was almost 1500 MW different than planned.

Mr Petr Zeman, the CEO of the Czech Power Grid (ČEPS), explained that they have survived so far. In his opinion, the problems are caused by the uncoordinated development of the wind turbines in Germany. They have stated that 20% of the electricity should come from renewable sources but these slovens no longer care what it does with the grids.

It is clear to Mr Zeman that if Germany were not connected to the rest of Europe, it wouldn't be able to survive these moments. The grids could be built to sustain such irregularities. However, it takes something like 10 years to build new power lines (which includes 7 years of the hugely complex EU paperwork).

The blackouts are being avoided by rather ad hoc methods of turning individual stations on and off in an internationally semi-coordinated and semi-predicted fashion. However, the chaotic description makes it likely that the system may collapse at some moment. The tasks for the regulation will be increasingly difficult as Germany wants to inflate the current 25 GW of pinwheels to 50 GW of pinwheels.

A commenter at Motl's blog notes that coal and oil thermal plants that have to stand down because of wind have to be put into "hot standby": they're still burning fuel to keep the plant's temperature up, but they're not generating electricity.

Sounds like a great deal: pay for wind turbines, keep paying for fuel for the oil and coal plants, and get the added bonus of grid instability.

I'm a creative artist

In response to my email of a while back, ACC has decided to classify me as:

92420 Creative Artists, Musicians, Writers and Performers

I've always thought that economics was best placed in the Arts rather than Commerce anyway.

Still no word on what the earnings threshold is for having to start paying a separate ACC levy.

A friend tells me that he had a similar experience a while back, though he was above the reporting threshold. On calling ACC to confirm that he wasn't involved in manufacturing (the default category, it seems), they asked what he was involved in. He asked what's the cheapest category. They said "data entry". He said "data entry". And they were happy with that.

If I ever wind up actually having an invoice, I may petition to be reclassified as a data entry guy. Transcribing results from Stata into Excel counts, right? (and, no, Outreg won't work easily for what I'm trying to do...)

Come to think of it, if data entry has the lowest risk, what does that say about theories of low status leading to stress and bad health? If depression is work-related, and ACC covers for mental health issues related to work.... We have good reason to expect that ACC's premia aren't fully risk adjusted, but I don't think anybody's claimed they have the rank order wrong, just that it's too compressed. Hmm.

Data entry...the high school guidance counselor, on seeing the results of my standardized tests around the 10th grade, in which my "clerical speed and accuracy" score had me at the 100th percentile (none of the other bits at all shabby, save mechanical reasoning, which was only around 75th and was only that good because I kept thinking back to playing with the Lego Technix set...), suggested I might wish to become a clerk. Transcribing Stata results makes me wonder whether she wound up being right after all about actual outcomes....

Ok, back to the clerical work....

Masnick on Music

Mike Masnick, who earlier put together the rather nice explanation of Trent Reznor's business strategy, continues to be one of the most sensible analysts of the current state of the music industry. Today he writes:

...selling music is just not a good business model, but it doesn't mean there aren't good, music business models. It's just that selling music isn't a very good one. Instead, you need to learn to use the music (which still needs to be good, and is still the central reason why these other business models work) to sell something else -- something scarce, which can't easily be copied. That can be attention, access, time, creative ability, cool physical products, whatever. All of those things are made more valuable the more popular the music is, and you can build all sorts of powerful and immensely profitable businesses once you recognize that.

But if you still think that selling the music or making money directly from the music has to be at the "center" of any music business model, you're shutting yourself off to the largest opportunities out there. But, the thing is, music has always been a product that makes something else more valuable. While there was some disagreement on the panel from someone about how record stores were profitable in the 70s, that's a case where the music was making the vinyl (and later, plastic) more valuable. Today, it makes iPods more valuable. As the big box retailers know, it acts as a loss leader to bring people in to buy higher margin goods. Music is great at selling other, higher margin things. If you ignore that in the music business model, you're missing the big opportunity.

This isn't to downplay the importance of music, or say that the quality of music doesn't matter. It absolutely does. But the music is not the scarcity, and you don't make money off of selling something that's abundant. You use the abundance to figure out what other scarce goods it makes more valuable and you sell those.