Thursday 30 June 2011

Reason to love New Zealand #57: Airports

American readers, please contrast this with your typical airport experience.

Arrived at Christchurch Airport at 7:25 for an 8:10 flight. Parked. Waved my phone at a kiosk, which printed my boarding pass. Got a coffee. Brought the coffee with me through security, where there was no particular queue because security agents here don't waste time molesting people. Walked to the gate and hopped on the plane.

My biggest delay was waiting for an excellent coffee.

Why haven't you emigrated yet? Do you like the ritual humiliation of American airports? Revealed preference says you must....

Fighting diabetes

The Greens and Labour want to ruin New Zealand's very clean GST system by putting in exemptions for "healthy food"; high NZ diabetes rates are the latest excuse.

Want to do something to reduce Type II diabetes rates? Ease up on the anti-alcohol measures! Moderate alcohol consumption reduces risk of Type II diabetes substantially; heavy consumption doesn't seem to increase risk relative to non-drinking. A very brief lit review:
  • Koppes et al 2005: Meta-analysis. Moderate drinkers (up to 4.8 std drinks per day) had relative diabetes risk about 30% lower than non-drinkers; non-drinkers and heavy drinkers have same risk.
  • Ajani et al. 2000. Male doctors who drink one drink per day have a 43% reduced risk of diabetes.
  • Joosten et al 2010: "Increases in alcohol consumption over time were associated with lower risk of type 2 diabetes among initially rare and light drinkers. This lower risk was evident within a 4-year period following increased alcohol intake."
  • Boggs et al 2010: Coffee (but not decaf) and alcohol reduce diabetes risk in a sample of African American women; tea does not reduce diabetes risk.
  • Baliunas et al 2009: Metastudy. Two drinks per day offers greatest relative risk reduction: a 13% reduction relative to lifetime abstainers. After six drinks per day, relative risk is 1.02: slightly above that of abstainers, but the 95% CI is [0.83-1.26].
  • Hu et al, 2001. Among a sample of nurses, alcohol consumption reduced diabetes risk. Protective effects strongly evident even among those with high BMI (>30).

Wednesday 29 June 2011

Ah, National Business Review, you crack me up

The National Business Review pointed last week to work by Andrea Ichino and Enrico Moretti on gender differences in absenteeism that seems utterly to vindicate Thompson's position. For readers abroad, a representative of the Employers and Manufacturer's Association was goaded in a TV interview into speculating about reasons for male-female wage differences. Among the more sensible and usual reasons - time out of the workforce, experience and so on - he noted some of his members had found higher rates of sick leave among female employees and speculated about whether menstrual cycles were to blame. [note: this is what I gather from an assortment of press reporting on the interviews, which I haven't seen]

You can guess what's happened since.

The National Business Review took the opportunity though of highlighting Ichino and Moretti's work suggesting that 28-day cycles in female sick leave are due to menstrual cycles and that increased work absence due to said cycles are responsible for a minor but significant part of the wage gap. The paper came out in one of the new AEA policy journals in 2009; here's an ungated working version. The paper's since been criticized by Rockoff and Herrmann; I'm not about to invest the time in sorting out who's right on this one.

But I love that somebody at NBR likely ran a Google Scholar search to see whether there was any lit to back up Thompson's ill-advised off the cuff remarks. That's a few notches above the rest of the media who just focused on the obviously correct point that you can't say what Thompson said in Thompson's position. Whether Thompson's remark was true is unfortunately second order relative to that some things just can't be said. I like that NBR took a bit of time to do some checking.

Traditional beers

From The Smithsonian:
Widely published in academic journals and books, McGovern’s research has shed light on agriculture, medicine and trade routes during the pre-biblical era. But—and here’s where Calagione’s grin comes in—it’s also inspired a couple of Dogfish Head’s offerings, including Midas Touch, a beer based on decrepit refreshments recovered from King Midas’ 700 B.C. tomb, which has received more medals than any other Dogfish creation.

“It’s called experimental archaeology,” McGovern explains.

To devise this latest Egyptian drink, the archaeologist and the brewer toured acres of spice stalls at the Khan el-Khalili, Cairo’s oldest and largest market, handpicking ingredients amid the squawks of soon-to-be decapitated chickens and under the surveillance of cameras for “Brew Masters,” a Discovery Channel reality show about Calagione’s business.
I tried that beer at the Dogfish Head pub on Route 7 in Falls Church with Eli Dourado. Good times.

McGovern argues that brewing helped build civilization (previously covered by Spiegel and noted here.) From the Smithsonian piece:
McGovern, in fact, believes that booze helped make us human. Yes, plenty of other creatures get drunk. Bingeing on fermented fruits, inebriated elephants go on trampling sprees and wasted birds plummet from their perches. Unlike distillation, which human beings actually invented (in China, around the first century A.D., McGovern suspects), fermentation is a natural process that occurs serendipi­tously: yeast cells consume sugar and create alcohol. Ripe figs laced with yeast drop from trees and ferment; honey sitting in a tree hollow packs quite a punch if mixed with the right proportion of rainwater and yeast and allowed to stand. Almost certainly, humanity’s first nip was a stumbled-upon, short-lived elixir of this sort, which McGovern likes to call a “Stone Age Beaujolais nouveau.”

But at some point the hunter-gatherers learned to maintain the buzz, a major breakthrough. “By the time we became distinctly human 100,000 years ago, we would have known where there were certain fruits we could collect to make fermented beverages,” McGovern says. “We would have been very deliberate about going at the right time of the year to collect grains, fruits and tubers and making them into beverages at the beginning of the human race.” (Alas, archaeologists are unlikely to find evidence of these preliminary hooches, fermented from things such as figs or baobab fruit, because their creators, in Africa, would have stored them in dried gourds and other containers that did not stand the test of time.)

With a supply of mind-blowing beverages on hand, human civilization was off and running. In what might be called the “beer before bread” hypothesis, the desire for drink may have prompted the domestication of key crops, which led to permanent human settlements. Scientists, for instance, have measured atomic variations within the skeletal remains of New World humans; the technique, known as isotope analysis, allows researchers to determine the diets of the long-deceased. When early Americans first tamed maize around 6000 B.C., they were probably drinking the corn in the form of wine rather than eating it, analysis has shown.

Maybe even more important than their impact on early agriculture and settlement patterns, though, is how prehistoric potions “opened our minds to other possibilities” and helped foster new symbolic ways of thinking that helped make humankind unique, McGovern says. “Fermented beverages are at the center of religions all around the world. [Alcohol] makes us who we are in a lot of ways.” He contends that the altered state of mind that comes with intoxication could have helped fuel cave drawings, shamanistic medicine, dance rituals and other advancements.
I'll have to make a point of looking for more of the Dogfish Head traditional beers.

Tuesday 28 June 2011

Transitional wages

I had a chat on Monday with Mike Williams from Newstalk ZB about the abolition of the youth minimum wage. The audio's here, starting around the 35 minute mark.

We didn't get beyond the effects of the prior legislative changes. And that may have been for the best since any potential changes coming from National remain pretty amorphous: rumour without confirmation or denial. Rumour has it they'll offer a transitional wage arrangement in which young workers on the unemployment benefit would be eligible for a lower minimum wage to help them ease back into work. I hope that they've thought through how these kinds of regulations would work. I can imagine perverse incentives built into the regime, but we'd need more than rumour to start working through the effects.

A mild clarification

I like the ACT Party's campaign to reinstate a lower youth minimum wage. Their ad over the weekend cited some of my numbers, though, and I'd want a footnote on there. They write:
Economist Eric Crampton estimates that Labour's abolition of the youth wage has cost 12,350 young people jobs.
I'd be more comfortable saying that the abolition of the youth wage has most likely cost between eight and twelve thousand jobs. The numbers since mid 2009 have ranged from about seven thousand to about thirteen thousand more kids unemployed than we'd expect given the adult unemployment rate. 12,350 is the number that drops out of the last quarter, but StatsNZ warned about data problems with that quarter's HLFS given data collection problems in Canterbury.

The "most likely" matters too as I don't have a design that allows for strong causal statements. I'd put 20:1 that youth minimum wage increases are to blame for current excess youth unemployment, but it's always possible that something else did it. Something else that affected kids (and not adults) and that hit with the right timing and that has a good body of theory predicting such an effect and that would have been large enough to have had that effect. I can't think of what else meets the bill, but maybe I'm unimaginative.

Academic pile-ons

Whatever you think about the quality of Satoshi Kanazawa's academic work, the folks who signed the open letter to The Times' Higher Ed damning his work sure chose the wrong time to do it if they had other than malicious intent.

To recap, Kanazawa's academic work has been subject to critique for some time. I've found a lot of his work awfully fun, and I've enjoyed his pieces on IQ and beauty. I've not gone through them with a referee's fine-toothed comb. And Satoshi could have done himself a lot of favours by answering critics more directly than he has.

But the current furore started with a blog post, as noted here. The post could have been wrong; further analysis of an updated version of the data that Kanazawa used suggested the effects went away if the fourth wave's data were included. That kind of thing happens in blog post analyses. That's why bloggers put up speculative trial balloons as blog posts - it helps to sort out where we might be getting things wrong before we go in full bore.

Kanazawa's post launched a populist campaign for his firing from LSE. On principle, even folks who disagree with Kanazawa's published work really ought to circle the wagons against this kind of populist pressure. I disagree with a lot of published work, as loyal readers might have noticed. Suppose that one of the folks at Otago Public Health - the NZ folks with whose published work I most frequently disagree - got into trouble over a blog post that got folks riled up. Suppose that, against all odds, there were mass public outrage against one of the neoprohibitionists' calls for bans on alcohol and lots of folks wanted the guy fired from his academic position. I'd hope I'd not I'd pick that moment as being the right time to launch a big public campaign against the guy's academic work. It's just not on. I'd hope instead to be writing posts in defense of academic freedom. And once the furore died down, I'd go back to putting the boot in. You shouldn't lose your job for being wrong in a blog post, and you shouldn't ride public anger fueled by same.

And so I signed an opposing letter in The Times Higher Ed in defense of Kanazawa. If it makes some folks happy to call me a racist for having done so, I've got a thick enough skin for it. Anybody who knows me, whether personally or as loyal readers, knows better. It makes me a bit sad that folks are happy to read racism into my post defending sweatshops, as I reckon the biggest force behind the anti-sweatshop movement is protectionist buy-American unions whose pro-natalism is necessarily racist in practice. Honi soit qui mal y pense.

Assortative mating

I've argued before that people are Lancasterian goods and that we all get the best spouse we can afford given our particular bundles of characteristics.

Jeff Ely finds additional evidence:
I visited the Cowles Foundation at Yale for the winter of 2006, and taught a senior elective course. Seven fortunate students took my seminar in information economics. One impressive woman student — who organized the gay and lesbian social scene — asked whether the shallow view of Becker’s model was so unrealistic. Did babes match with hunks?

We brainstormed on data sources and settled on two new web sites: and Facebook allowed users to indicate with whom they were “in a relationship with”. Facebook was still new, and not yet open to all email addresses. So the student asked her friends at various campuses across America for their logins. And so began our stealth project. Hundreds of photos of matched men and women were downloaded, and then uploaded to HotOrNot, all on the sly. HotOrNot afforded us the average evaluation of about 200 women for every man, and 2000 men for every woman.

The result: Regressing straight men’s or women’s hotness on their partner’s hotness gave a highly significant fit, with a slope of about 0.7 — so that a man rising in hotness from 7 to 8 expects his partner to rise by 0.7 points. But sorting was far closer for gays and lesbians, with a slope for each of about 0.9. As Becker implied, beauty is income in this meat market, and the “richest” men match with the “richest” women.
The fit would be even better if you could adjust for the other Lancasterian characteristics.

I love the gonzo approach to this project. The human subjects review panels there must be more sane or have fewer teeth. Excellent either way.

And Robin Hanson would (and has) asked why we care so much about income inequality when this form of inequality may matter more....

Monday 27 June 2011

More Shameless Advertising (UPDATED)

I noted here that the annual New Zealand Association of Economists annual conference was coming up (Wednesday to Friday this week at the Amora Hotel in Wellington). The final progamme is here.Tim Harford is leading off with the keynote address at 9:00 on Wednesday morning. It is possible to register on-site at the Amora Hotel, for either a single day or the full conference.

Also, as a first this year, Canterbury's Economics Department is hosting a function for former staff, alumni, and friends, at the Amora Hotel at 7:00pm on the Wednesday. This is in association with the conference but not part of it, so people not registered for the conference are welcome to attend. We don't have a good database of our former students, and so have not been able to contact all of them. So, dear friends and alumni, if you read this blog but have not been directly contacted about this event, we do love you and would love to see you there!

If interested, go the facebook link and click on "I'm Attending". And if you are a facebook user, go to our facebook page, and click "like". We hope to use this page as a way of maintaining contact with alumni and friends.

Equal pay

Auckland university economist Ananish Chaudhuri endorses the Greens' campaign for equal pay.
Women, here and elsewhere, are not asking for a hand-out. They are asking to be paid the same wage as men for the same work, which is fundamental to democratic ideals of equity and justice. Green Party MP Catherine Delahunty's bill, which proposes to amend the Equal Pay Act by allowing for gender pay comparisons, will help reduce the disparity. It is an important step forward for achieving the goal of gender pay equity.
Ananish cites some of the empirics on gender pay gaps, noting that a reasonable proportion of the pay gap remains after correcting for observables other than gender (although this depends a whole lot on which country's data you pick). But we have to remember that if a whole pile of observables are highly correlated with gender, as has to be the case if the raw pay gap is cut back a pile by correction for observables, then it would be surprising if there weren't also unobservables that varied by gender and affected pay. A reasonable coefficient on gender then can be picking up the effects of unobservables other than gender that correlate with pay. And note also that while there may be a gap in straight pay, total compensation bundles include a whole lot of non-pecuniaries valued more highly by women.

Delahunty's amendment would require firms report pay by gender, allowing easier enforcement of current equal pay legislation.
The proposed changes meant workers and unions would be able to request information on pay levels by gender in their workplaces to assess whether the Equal Pay Act was being applied.
And legal risks then increase for firms with differences in average salaries, even if such differences disappeared after correction for the usual confounds like experience, time out of the workforce, hours worked, and so on.

One risk of strengthening equal pay protection: young women become less likely to be hired because the firm then bears greater risk in case of maternity. French data shows firms show little gender bias in hiring in age cohorts where maternity risk is low but are reluctant to hire women where maternity risk is high; French firms bear high costs in case of maternity leave. Such costs are lower in NZ, but not non-existent. Firms that would otherwise let salary differences clear up the costs of flex-time arrangements or more off-job responsibilities will fear employment court action. It's cheaper and less legally risky to hire fewer women in the first place.

It's highly unlikely that Delahunty's bill makes it to the floor. But if it does, I'll have to work out terms for a bet with Ananish.


Friday 24 June 2011

Maharey on Minimum Wages

I'd take former Labour Cabinet Minister and current Massey University Vice-Chancellor Steve Maharey's NBR column "Why minimum wages are necessary" as a launching point for ripping on Massey, but I'm reluctant to talk any kind of smack about Canterbury's competitor institutions when Christchurch's future remains less than certain.

Let's take Maharey's claims in turn.
The fact is that no one really knows if the minimum wage is contributing to high levels of unemployment. In general terms, the research suggests that the minimum wage, set at a level that matches the state of the labour market, will not cost jobs.
It would be tough to find effects of the relatively minor increases we get in the overall minimum wage in New Zealand, but that isn't to say that the aggregate effect isn't large. We could similarly say that nobody really knows if carbon emissions are contributing to temperature increases if the only data we had were twenty-five years of quarterly data on changes in CO2 output and changes in temperature. The aggregate effect can be large even if year on year changes don't do much.

If we start from the premise that there are no employment effects from the minimum wage, the case for a minimum wage still isn't as clear-cut as Maharey suspects. For starters, we expect a whole lot of the cost of minimum wage increases will be passed through to consumers via product prices. If minimum wage workers disproportionately produce goods and services consumed by poorer people, or at least disproportionately relative to poor folks' relative tax burden, minimum wage increases are not nearly as progressive as income transfers to poor people funded out of the standard income tax regime.

Here are Maharey's arguments for minimum wages:
The first reason is that it is one of the main ways of dealing with inequality. A minimum wage is an attempt to ensure those at the bottom end of the scale have at least enough money to live on.
A more efficient way of achieving the same goal is income transfers; if you want it targeted to folks in work, do it through something like the EITC (WFF). The incidence of the minimum wage isn't clear, but is almost certainly less progressive than the tax system.
It follows, and this is the second reason, that workers should not be expected to work for very low wages. This is nothing more than exploitation and a society that has any sense of fairness should not allow this to happen.
Wages would not drop massively in the absence of a minimum wage. Firms compete with other firms for workers; there isn't some big secret cartel meeting of all the potential employers where they use secret handshakes and arcane Masonic rituals to ensure that nobody chisels on a low-wage pact. It's a damned shame that Labour's notions of fairness forced a bunch of sheltered workshops for the disabled to close through ridiculous application of minimum wages; we ought to allow employers to hire workers with lower marginal product for a wage that matches their contribution.
Third, wages should be set at a level that does not require the government to provide subsidies.
I think this is a big one. EITC or other in-work income support programmes are an on-the-books expenditure; minimum wages bury the cost. If the public will is that low wage workers get higher wages, then the best way of spreading the burden of that support is through the general tax system.
Fourth, low pay is often a result of undervaluing what a worker can do or the skills they have on offer. Young workers are often paid less simply because they are young even if their contribution is the same as older workers.
Employers take greater risks hiring folks with no track record. If a kid hired on lower salary proves to be a great worker, he'd expect a decent bump up at salary review. Forcing equal pay for unequal track records forces those with no experience out of the market.
Fifth, low pay is not a good economic strategy. When times are good, low pay leads to high turnover and a weak commitment to training. ... success becomes dependent on who can pay the least to their workers not on quality, productivity, design, marketing, value-add, skill and the other elements of a desirable economy.
Efficiency wages are hardly unknown; firms wishing to avoid turnover problems may pay higher wages to get a better pool of job applicants. The other claims Maharey makes require, I think, a pretty uncompetitive market. Otherwise, some firms would start improving on the other margins to get an edge on their competitors. I don't think that Dyson would have been less innovative in vacuum cleaners had the minimum wage for product engineers been $0.02 (nor do I think their engineers' wages have ever been close to minimum).

Meanwhile, there have been burblings in favour of reintroducing a lower minimum wage for youths.

Thursday 23 June 2011

Brew Winnipeg

Manitoba is finally set to loosen up its alcohol regulations. Back in Christchurch, a new Brew Pub is just opened despite all the earthquakes; Manitoba still has only two brewers - HalfPints (excellent) and Fort Garry (meh).

It looks like Neepawa will be getting an Estate brewer: they'll grow their own hops and barley. Didn't know that the Manitoba climate was really suited to hops, but will be really interested to see how this pans out on my next visit.
The owners of Luxalune Gastropub are taking the changes to provincial liquor laws and running with them -- all the way to Neepawa.

That's where Chris and Lawrence Warwaruk are planning to build Canada's first estate brewery -- The Farmery.

"It's no different than the estate wineries in Kelowna, B.C.," Chris Warwaruk said. "We're going to be growing the raw products, barley and hops, at the farm. We shouldn't have to buy hops from (growers in) Washington state that supply Miller, Busch and Coors."

Estate breweries are different from traditional breweries because the ingredients that go into the bottles are grown on the estate or farm.

Warwaruk said changes to Manitoba liquor laws, which passed last week, have opened the door for him and his brother to explore the estate brewery concept. The two biggest changes from the new legislation, which will take effect in November, enable restaurateurs to brew beer on-site by creating brew pubs and allow customers to take their own bottle of wine to participating restaurants.

The Warwaruks -- who grew up on a farm near Minnedosa -- plan to cut the ribbon on their brewery next spring. It will include a finished building and a plant. The hope is that it will attract beer-loving tourists from near and far for tours, too.

The adjoining fields used to be home to a U-pick strawberry farm but Lawrence Warwaruk has experimented with malt barley and hops this year and has plowed under some of his strawberries.
All they needed was that the government get out of the damned way. Congrats guys, and best of luck!

Wednesday 22 June 2011

Assorted updates

  • We leave Winnipeg for Christchurch tomorrow morning, landing Friday morning Christchurch time.
  • My travel was booked through the University's travel agent; I still spent an hour on hold with Air Canada sorting out conflicting luggage allotment rules.
  • Winnipeg's a great town in summer. What was shabby and dated twenty years ago is now retro-chic. Wait long enough and all that's old is new again. I base my wardrobe on such expectations.
  • The Christchurch Airport has been iffy between regular daily earthquakes magnitude 5 and up and volcanic ash from Chile. I hope we don't get stuck in Auckland; we'll see.
  • Getting back into the daily earthquake routine fills me with dread.
  • Nobody in North America knows squat about what's going on in Christchurch. It was a headline in September and February; everybody reckons that all's back to normal. As update for folks who've lost track:
    • there was another 6.3 last week, there are between a half dozen and a dozen earthquakes every day magnitude 3-5.3, with strong potential for bigger ones
    • Literally 60% of downtown is set to be bulldozed
    • The army cordon around downtown is set to be in place for another six months
    • The boil water notice from the last quake seems to have been lifted, but there's going to be liquifaction silt all over the place, random road closures, sewerage failures.
    • All the normal bits of life you're used to in your town: the places you take the kids on a rainy day; the place you usually go for coffee; your barbershop; the swimming pools for the kids - just imagine that more than half of those are gone for a few years and the others are touch and go. And that nobody in town's getting any sleep. My twitter feed's full of Christchurch folks up at 3 in the morning with aftershocks every damned night.
    • We'd just say bugger it and stay here a while longer: Susan can work remotely. But I've got lectures to give in a couple weeks and NZ / Australian conferences ahead.
  • Blogging will continue to be very light while on the road and prepping for the NZAE meetings; I'm now giving two papers there rather than one as a wait-listed paper was drawn at the last minute. I'm only at NZAE on the Thursday as I'll be in Oz for a few days mid July and am trying to limit the number of overnights I'm away from home while the aftershocks continue.

Sunday 19 June 2011

Random Thoughts on the Quake

1. Is God A Philistine? Further damage to the Cathedral and the Basilica; the chalice still standing. ANZ chambers completely destroyed (and its dome now stolen); the Government Life Building still standing. If there is a God with influence over geological events, I have to question His aesthetic judgement!

2. Rational versus Adaptive Expectations? Friends of ours in Wellington have suggested to use that we move there, permanently, to get away from earthquakes!

3. An apt metaphor. When I was finally allowed into my office after the Feb 22 quake, I was musing to myself about what the official responses and the spontaneous volunteer work had taught us about the advantages of decentarlised decision making. On entering my office, I found that literally only two things had fallen onto the floor: my copy of The Wealth of Nations, and, from my noticeboard, a New Hampshire licence plate with its motto, Live Free or Die!

Saturday 18 June 2011

Natural Disasters and GDP for National Income Accounting Nerds (UPDATED)

The broken windows falacy has been well addressed following our recent quakes, but there is an interesting technical issue with the measurement of GDP following a disaster, brought to my attention by a graduate of ours (and a former colleague of mine at the Bank of Canada), James Yetman.  

By way of background, consider how the contribution of a lottery to GDP should be measured. From total revenue, one needs to subtract off not only expenditure by the lottery seller on intermediate goods (such as the cost of the paper lottery tickets are printed on), but the payout on prizes. In effect, a lottery provides a valuable service (it would seem, from revealed preference) for transferring money from some ticket buyers to others. It is the commission earned on this transfer that constitutes the lottery seller’s revenue from which one subtracts expenditure on intermediate goods to calculate the contribution to GDP.

Now imagine that a country runs a really large lottery in which the prize jackpots if it is not won. And imagine that the probability of any particular lottery having a winner is so low that in most years the jackpot is never won. In this case, in most years, the lottery would be appear to be making a large contribution to GDP (high ticket revenue with no prize disbursement subtracted off), and then in years when the jackpot was won, would appear to be making a large negative contribution. The lottery market, however, is providing the same lottery services each year. In this example, the appropriate way to measure GDP would be subtract off the expected level of prize payment from revenue each year, not the realised payments.

Now consider the insurance market. Just as with a lottery, one should measure the revenue in the insurance industry as the difference between income received (premium payments plus interest on accumulated investments) and payouts in claims. But there is a lottery component to insurance. In the year of a really large natural disaster, payouts on claims will be unusually high, so in normal years, the difference between income received and claim payments will need to be higher to cover this contingency. Just as with the lottery example, the true contribution of insurance to GDP (the production of peace-of-mind), does not fluctuate in this way. Apparently after 9/11, the way the contribution of insurance services to GDP is measured was changed in the U.S. to subtract off expected claim payments rather than realised payments. I have no idea what the definition is in New Zealand. Can anyone with a background in official statistics enlighten me?

UPDATE: James Yetman has emailed me a reply he got from Statistics New Zealand about this. The upshot:
Premium income is used as the indicator for insurance in quarterly GDP, which affected directly by changes in insurance claims. Premium income may rise as prices rise in the longer term, but unless more people actually take out insurance it won't affect GDP in constant prices.
Annual GDP in current prices is a bit trickier. We get the output of this industry by deriving a service charge that represents the service the insurance industry offers policyholders. This starts with a service charge ratio, which measures the proportion of premiums that aren't used in paying claims (with a few other adjustments for supplementary income and reinsurance). The service charge ratio is averaged over five years to smooth out volatility and then multiplied by the premiums received for the year (again with extra adjustments I won't detail here), to give the service charge/output of the insurance industry.
A big rise in claims could potentially pull down the service charge ratio significantly, even with the five year average, though it would likely be offset by reinsurance claims by NZ insurers. So the final impact would depend on the difference between insurance claims and reinsurance claims. It's also possible that we would intervene here if we didn't think the service charge ratio was realistic, as the service charge is intended to be based on the 'normal losses' you mentioned (as the insurance industry calculates its premiums based on probabilities over the long term).

Thursday 16 June 2011

Alcohol minimum pricing

The Australians are considering imposing alcohol minimum price regulation to reduce alcohol's social costs. Here's The Australian:
...decision by federal Health Minister Nicola Roxon to ask the National Preventative Health Agency to examine the idea of a uniform national floor price for a standard unit of alcohol, whether it be contained in beer, wine or spirits. Roxon described the referral as a preliminary but important step.
I'd noted some of the problems with minimum alcohol pricing last year. Unless there's really good evidence that the kinds of drinkers Roxon wants to target are really disproportionately hitting the cheap alcohol, we have to worry a lot about that moderate drinkers are far more responsive to price changes than are heavy drinkers. If the policy shifts enough "glass of cask wine per day" drinkers to not drinking, it will do reasonable harm given the health benefits of moderate drinking.

And as for that $30+ billion dollar social cost figure they're citing, recall that about $6 billion of it comes from this dodgy move alone, which I'd also noted last year. From that post:
At page 133, they provide some estimates of the economic costs of intangible harms Australians suffer as a result of someone else's drinking. Their survey respondents answer questions that give them a health-related quality of life score. A one-point drop in that score, they say, costs the individual the equivalent of $50,000. Respondents are sorted between those who cannot identify a known drinker who has negatively affected them, those knowing a drinker whose drinking has affected them "a little", and those knowing a drinker whose drinking has affected them "a lot". They take the mean quality of life score for each group, multiply the differences from baseline by $50,000, extrapolate to population averages, and come up with about $6.4 billion in costs.

There are a couple of pretty obvious problems here, if I understand their method correctly. Which I may not, but I'd love to hear if I'm wrong on this count. It really looks like they're just comparing sample means. If so, and if there are any other differences - differences unrelated to alcohol - between folks who have a close relationship with a problem drinker and those who don't, then the mean comparison is hopelessly confounded. There are all kinds of bad social circumstances that are predictors of problem drinking. Those same circumstances would lead to lower quality of life scores regardless of whether there's any drinking. Regression analysis is needed, not comparison of means.
I hope that the Australian health folks have more substantial foundations for their current policy push.

Here's Luke Malpass in The Australian [update: or, at least the journo's rendition - see Luke's comment, below]:
But Malpass demurs, saying figures on the cost of alcohol to society invariably fail to offset the benefits of the industry, including the employment generated, the export dollars created and, on some views, the positive health effects of the odd glass of wine.

"Across the socio-economic spectrum people enjoy a drink. But the unspoken assumption is that alcohol and cigarettes must be priced out of the range that these nanny-staters, these healthists, consider problematic. They assume people don't make a conscious choice to drink and not exercise. It's incredibly infantilising."

He says the so-called alcopops tax on pre-mixed drinks brought in by the Rudd government didn't achieve any behavioural change when it came to the alcohol consumption of young drinkers.

"The girls who would drink one Breezer or two Breezers don't any more. The alcohol tax changes resulted in perfect substitution, the teenagers stopped buying alcopops and instead bought a bottle of Jim Beam and mixed it with Coke."

More political intervention in this policy space is potentially politically damaging, according to former Labor senator John Black, now a political analyst.

"[A minimum price] would effectively be another tax on lower income groups and they won't like it," Black says. "It will have a disproportionate impact on people in the lowest income quintile, and it won't affect behaviour for those who are dependent and who have the problem. It's economics 101, inelastic demand."

There could well be blowback politically on the Gillard government, he says, with electorates containing a higher than average proportion of the nation's poor the political flashpoints.
I'd disagree with Malpass's the journalist's counting of employment as a benefit; the primary benefit ignored is consumption benefits to drinkers. Employment is a cost (think of it this way: we'd be better off if a genie appeared and wished every day's alcohol production simply appear and folks employed got to shift to other sectors). But Malpass is dead right about the healthists' disdain for the lower orders' pleasures. And Black is also right about the likely incidence.

Oh Australia, what's happened to you?

Wednesday 15 June 2011

Minimising liabilities

What's an earthquake event, for insurance purposes?

New Zealand's Earthquake Commission, EQC, is funded by a check-off on property insurance and covers the first $100k of property damage subsequent to an event. September, Boxing Day, February, and this past week's earthquakes all counted as separate events for EQC purposes, each with its own three month deadline for filing claims.

But what happens if EQC's assessor is slow in checking up on your property so the assessor can't tell how much damage was due to each of the events? EQC argues that the assessment then becomes the event: the first $100k of property damage on an assessment falls on EQC, with the rest going to the private insurer. EQC and the insurance companies are seeking a declaratory judgment.

And there is some sense to this. Our place was inspected post-September but we never received a quote from EQC on their estimate of the damage cost, and so we were never able to book in with a builder to get things fixed. Once EQC's assessors get around to us, hopefully before this coming September, there's no way they'll be able to tell how much damage is due separately to Boxing Day, February, or this weekend's major earthquake (for folks abroad: the city just keeps rocking). But there have been 12 separate events. If private insurers argued that damage were spread evenly across all twelve events, EQC would be on the hook for all damages - $1.2 million per home, well in excess of the "bowl it over and build new" cost for the vast majority.

On the other side, if EQC can minimise its liabilities by lumping a whole pile of separately damaging earthquakes into one event, EQC's foot-dragging in getting assessments out makes more sense. Again, using our place as example: damages from September's quake were pretty minor as we're over on the east side of town; Boxing Day provided no new damage; February was far worse for us. Our housesitter (who was also our housesitter for the Boxing Day quake) informs me that the house is currently a mess of broken glass and that the cracks in the internal wall plaster are rather worse than they've been since February. EQC footdragging then lets them count all the earthquakes against the same $100k cap, with any excess costs then falling on AMI.

Meanwhile, nothing gets rebuilt. In some cases, that's efficient; fixing cosmetic damage like plaster cracking probably doesn't make sense if the next damned earthquake will just undo everything. But each bit of structural damage that doesn't get fixed because of foot-dragging makes the next quake's damage worse. Will be interesting to see what's waiting for us on our return.

Tuesday 14 June 2011

Unions like minimum wages

Dr. Bill Rosenberg, economist* at the NZ Council of Trade Unions, didn't like my piece in the Dominion Post. Recall that in my Dom piece, I argued:
  • Unemployment rates for youths are higher than they have been since 1986; these figures are worse if you account for the drop in youth labour force participation. And youth unemployment is particularly bad when compared to adult unemployment, which is nowhere near the levels seen when youth unemployment reached its prior peak.
  • Labour unions like to cite recent US evidence that US minimum wages have little to no effect on employment, but we can't extrapolate from studies where minimum wages are a third or less of the average wage to a labour market where the minimum wage is half the average wage. This applies to both Doucouliagos's meta-study and Dube's big panel study on county border pairs; Dube's suffers the additional problem that a study of an industry with relatively inelastic labour demand doesn't extend well to a broader economy (note that I said "isn't likely to be representative of a country as a whole.")
  • The unions also cite Hyslop and Stillman's excellent piece, but forget to note that the labour market was seriously overheating during the period of their study; ability to fog a mirror was sufficient to get a job in 2005.
  • The problem wasn't with minimum wages per se, but with minimum wage levels that get into the really rather dangerous range above 45% of average wages.
So what does Bill Rosenberg say in defence of high minimum wages?
  • The ILO concluded in '98 that "Whether a minimum wage has a negative or a positive effect depends on many factors such as its relative level, the structure of the labour market, and the country concerned."
    • ...which was kinda my argument: a minimum wage of a third of average wages mightn't have much effect while one that's a half of average wages will
  • Hyslop and Stillman found no effects of youth minimum wage increases in the early 2000s
    • ...utterly ignoring the overheated labour market that followed, or the negative effects on employment Hyslop and Stillman found in the short period prior to the labour market's overheating
  • Doucouliagos's meta-study found no effects of minimum wages in the US, and neither did Dube's.
    • ...again, utterly ignoring that minimum wages there are a third of average wages while they're a half here. It's also completely hilarious that while the unions fall over themselves to cite Chris's work on minimum wages, they ignore his work showing that unions kill company profit rates and depress physical capital formation. I like Chris's work overall. His results from the US are depressing and speak poorly of journal refereeing processes, but they just don't extend to places where the minimum wage is a lot higher. The dose makes the poison - it's true in toxicology, and it's also true of minimum wages.
  • Crampton says Dube's study doesn't apply to NZ because it looked only at restaurant workers, but more than a third of 15-24 year old workers are employed in retail, accommodation and food services and "about three-quarters are in industries that similarly can't be easily outsourced."
    • I didn't say that Dube didn't apply to NZ; I did say that we can't extrapolate from Dube to an overall economy. We can't extrapolate from Dube to NZ not because of industry choice but rather because minimum wages are so much higher in NZ than in the US.
Rosenberg makes a couple of more substantial points.

First, that the 2008 minimum wage changes applied only to 16-17 year olds; 18-19 year olds have been under the adult rate since the changes analyzed by Hyslop and Stillman. This is true. And I would be surprised if 16-17 year olds were driving everything here (though I'm likely soon to be getting sufficiently disaggregated data to be able to do something on this). What seems more likely is that the prior changes became binding when the recession hit. The 18-19 year old cohort analyzed by Hyslop and Stillman didn't go through any major recession since being subject to treatment. When late 2008 hit, the previously non-binding price floor became very binding.

Second, Bill argues that youth unemployment rates are artificially high because the denominator has been dropping: absolute numbers of unemployed youths have been growing more slowly than absolute numbers of unemployed adults, it's just that youths have been switching over more quickly to study (or other options) and dropping out of the labour force. But this doesn't really help Bill's case. Why? Recall that the unemployment rate is the number of people unemployed over the number of people in the labour force. Rosenberg seems to be arguing that all the dropping out of the labour force is among the folks who'd otherwise have found jobs - they're dropping only out of the denominator. But that seems highly unlikely. Kids who are discouraged from the labour market fail to show up in the denominator but had expected to have shown up in the numerator as well. If they'd expected only to be in the denominator, they might well have entered the workforce.

* ...but the PhD is in psychology, isn't it? I'd normally just smirk to myself on this one, as he's been doing it forever. But as he called me Mr. Crampton throughout while using that somewhat misleading tag-line, I'll make an exception. I don't much care about titles; the students usually call me Eric. But I didn't go to Evil Economics School for five years (after four years at Not Nearly Evil Enough Undergrad) to be called Mr by someone with a PhD in psychology and a misleading tag-line. Gotta draw the line somewhere.

Monday 13 June 2011

Gunslinging and Shooting Yourself in the Foot

Copyright is more Eric's thing than mine, but recently I encountered what seems like an ill-advised exercise of that right.

I like to use references to classic movies in my problem sets for my intermediate microeconomics course. In recent years, I have been adding links to YouTube clips from those movies in the model answers. One of these problem sets involves a retired gunslinger named Shane. In past years, I have included a link to a YouTube upload of the final showdown between Shane and Wilson. Even more fun then the scene itself is the flame war in the comments about the true reason that Shane leaves town at the end. (Is it because he was a killler, or because of the feelings between him and Marian Starrett?)

In editing the answer this year, I find that the video is no-longer available on YouTube, due to "a copyright claim".

Which leads me to wonder, why would the copyright owners want to do that? Surely, no-one is going to not pay money to purchase the entire movie simply because the final ten minutes are on YouTube. To the contrary, the link itself, and the comments flame war, if they had any effect at all on sales, would surely raise interest in the full movie. Maybe the owners feel that they need to take a tough stand against any uploading of their IP, in order to create a reputation, but I can't see reputation mattering too much in the YouTube game.

PS to all you Shane haters: Yes, I know the acting in the lesser parts is awful, and yes, little Joey is really irritating, but beneath that the presentation of the subtexts and psychological relationships are very sublte. The move is well worth renting.

Saturday 11 June 2011

Food superpower

Dan Gardner's piece in The Ottawa Citizen contrasts Canadian and New Zealand agriculture:
“Look at us,” [Agriculture Economist] Larry Martin suggests, “and look at New Zealand, sitting out there in the middle of the ocean, not close to anything.” In the world of food, New Zealand is a “superpower.” And yet, thanks to daring reforms in the 1980s, New Zealand’s farmers owe almost none of their income to government support. “You think, ‘if we could do even half of what they have done wouldn’t we be in great shape?’”

Good for us. Good for the world. If only the politicians would talk about it.
I'd noted before just how much bigger New Zealand is than Canada in dairy exports:
For most things, if you want to get a ballpark comparison from NZ to Canada, New Zealand is about order of magnitude smaller. Not on dairy though. 2007 dairy exports for New Zealand: $6.3 billion (about $4.5 billion Cdn). On that one, we're more than an order of magnitude bigger than Canada [at $255 million Cdn], or two orders of magnitude bigger than you'd expect given everything else about the two countries.
How to get out of the mess? I'd suggested paying the dairy farmers off. But New Zealand liberalized without that kind of compensation. Instead, and I hope Seamus can help me out on the timing on this one, I think the farmers received other forms of liberalization as compensation, including freeing up the ports from pretty heavy union control.

As much as western grain farmers might rejoice were Canada's newish majority Conservative government to put in place the kind of legislation* that would break the Longshore and Warehouse Union, I doubt it would do much for the Quebec and Ontario dairy farmers who are the ones that really need to be bought out.

*Here's an excellent model; here's a perhaps less ambitious but still reasonable alternative.

Friday 10 June 2011

Check my sources

Today's Dom Post has a piece I wrote on youth minimum wages. I've copied the text of it below. Folks coming in from the Dom can check my work by hitting the links embedded in the text below. Here's my best response to minimum wage "denialists".

Stop pricing young workers out of the labour force

When youth minimum rates were scrapped in 2008, unemployment of young people rose, writes Eric Crampton.

IF THE Government said that the minimum price for a new car were $50, nobody would expect it to affect sales. Neither would an increase to $65. But it would certainly start mattering if the Government applied a minimum price of $5000 to all cars, new and used.

This is the situation in New Zealand with youth minimum wages, which were abolished in 2008 in favour of adult rates for all workers over 16 years old. This increased the youth minimum wage by 25 per cent (barring the first few weeks’ work for new entrants).

The latest youth unemployment figures are very bad. The unemployment rate for kids aged 15 to 19 is 27.5 per cent; worse, the youth labour force participation rate (those actually engaged in the job market) also has dropped.

Whereas the participation rate was above 50 per cent for the decade before 2009, it has since been dipping and now sits at 47 per cent.

If youths hadn’t been discouraged from entering the labour market because of poor job prospects, the measured youth unemployment rate would be even higher.

This isn’t just the recession. Unemployment rates for adults are higher than they were in the boom of the mid 2000s, but the recent downturn has not hit adult workers the same way that it’s hit the kids. The current adult unemployment rate of 5.6 per cent is only three points higher than its low mark in the mid 2000s. Meanwhile, youth unemployment rates are a staggering 15 points higher.

Both rates usually track each other, reflecting the overall strength of the labour market. Changes in the adult unemployment rate explain a high proportion of changes in the youth rate.
But in late 2008, this relationship began to break down. Compared with a previous trend, the current youth unemployment rate is eight points higher than we could have expected given the adult unemployment rate. That’s about 12,000 kids who, given the current adult unemployment rate, we would have expected to have jobs. The results aren’t simply due to the Canterbury earthquake – the trend started well before last September.

Neither can they simply be due to the current downturn: when adult unemployment hit 10.2 per cent in 1992, the youth unemployment rate was 23.4 per cent – three points lower than today – and youth labour force participation rates were higher. Bear in mind that adult unemployment today is nowhere near 10.2 per cent.

No, the sharp increase in youth unemployment from late 2008 appears to have been caused by the abolition of the youth minimum wage in early 2008. Such a result isn’t surprising. Economist Stephen Gordon summarised Pierre Fortin’s work on this effect in relation to minimum wages: when minimum wages are below about 45 per cent of the average wage, they have little effect on employment; above that, they present a danger to employment.

By contrast, New Zealand’s minimum wage of $13 an hour is about 50 per cent of the average hourly wage – well into the range in which we expect negative employment effects, particularly for young workers.

Internationally, only 13 per cent of labour economists surveyed disagree that minimum wages increase unemployment among young and unskilled workers, although some studies find contrary results.

Of these studies, the best New Zealand study, by Dean Hyslop and Steve Stillman, found that youth minimum wage changes in the early to mid 2000s had no effect on youth unemployment. It was a quality study, but it spanned a period during which overall unemployment was at record lows and businesses were desperate for any workers.

The increase in the youth minimum wage during that period had little binding effect. New Zealand labour unions also like to cite an excellent paper by Arindrajit Dube and co-authors showing no effect of minimum wages on workers in the American restaurant industry.

But it’s a lot harder to outsource waitresses than call centre operators, or factory machinists, so this finding isn’t likely to be representative of a country as a whole. Also, in this case, minimum wages in the United States were low compared with average wages.

SO WHY do economic studies find differing results on the employment effects of minimum wages? One reason is that, in many places, minimum wages are well below the level at which we would expect that they would have serious effects, like the $65 dollar minimum price for a car. But when minimum wages for youth jump to New Zealand’s $13 an hour, you get a problem.

The effects of minimum wages on young and unskilled workers are well known. It is a shame that minimum wage ‘‘denialists’’ let the few contrary studies outweigh the dozens of papers finding that minimum wages hurt employment prospects for the people they’re intended to help.

In 2010, the Organisation for Economic Co-operation and Development published research supporting the majority theory and pointed out that half its member countries (including Australia) had minimum youth rates.

Reinstating a youth minimum wage well below the adult rate wouldn’t eliminate youth unemployment. But it would let employers start creating new jobs that young workers could productively fill while gaining experience. It’s time to stop pricing young workers out of the labour force.

Eric Crampton is senior lecturer in Economics at Canterbury University and Visiting Fellow at the Centre for Independent Studies, Full references to reports in this piece can be found at his website

Thursday 9 June 2011

Expressive voting

Hamlin and Jennings survey the expressive voting literature. It's a nice piece, and I like how they've brought Kuran's preference falsification arguments. I'll quibble on two points, mostly because I like the word quibble and need excuses to use it. They write:
Caplan is careful to distinguish between [rational irrationality and expressive voting]: 'In expressive voting theory, voters know that feel-good policies are ineffective. Expressive voters do not embrace dubious or absurd beliefs about the world ... In contrast, rationally irrational voters believe that feel-good policies work.'

Therefore, a further condition would need to be fulfilled in order to judge a vote to be expressive of true preferences rather than rationally irrational ones; we would need to check how well informed the voter is. One suspects that this issue may be similar to social pressure. If voting is both expressive and 'rationally irrational', making information available might be expected to result in a rapid and significant shift in the political equilibrium. If, by contrast, voting is an expression of truly-held expressive preferences, the political equilibrium will be much more stable.
I doubt that Caplan would argue that rationally irrational voters need only be provided more information in order to improve outcomes. If information provision were the sole problem, voters wouldn't be openly hostile to the provision of information with which they disagree. And the rise of the Econoblogosphere would have quickly led to substantially better economic policy.

Here would be a rather better test of Caplan's rational irrationality model. I've not seen it conducted, but more experimental economics applications have been melding voting and markets. Here goes. Set up an experimental double-auction environment framed in a salient way - buyers and sellers of labour, for example. Run a few rounds of the experiment as baseline. Then, let folks vote on whether they'd like to make a change to the trading environment: policy changes that either improve or reduce overall efficiency. A price floor, for instance - a minimum wage. Set treatment groups that vary in individual expected decisiveness: the odds that any player's vote will determine the trading structure for the next round, with the sum of all player odds being less than or much less than one. Then run a few rounds with the (potentially) changed trading environment before offering other votes - some which augment and some which attenuate efficiency, with varying expressive framing. If traders make better choices when more decisive, that would be consistent with rational irrationality. It wouldn't distinguish between expressive voting and rational irrationality, but I'm more interested in testing the broader concept anyway.

Hamlin and Jennings later discuss the implications of expressive voting for constitutional choice. Brennan and Hamlin worried that constitutionalism exacerbates expressive voting problems and suggested that constitutional questions be left to small but statistically representative groups in order to avoid the Veil of Insignificance. They write:
Perhaps these proposals should be decided by small (but representative) groups, which might be more likely to take an all-things-considered view. Crampton and Farrant make explicit the potential problem that such a small group might design institutions that enrich themselves if they are not fully representative in a relevant sense. Therefore, a trade-off may exist between the problem of expressiveness, on the one hand, and allowing too much room for the narrow self-interest of unrepresentative groups, on the other.
It's probably semantics (a lot can be packed into "in a relevant sense"), but our main worry (ungated) was that the perfectly statistically representative group would have, by virtue of being the constitutional committee, a newly granted interest in enriching the members of the committee. If the group is small enough to overcome expressiveness problems, it may also be small enough to solve internal collective action problems and set itself up as effective dictator post the constitutional phase. The only way of breaking past the Veil of Insignificance is by reintroducing the problem that constitutional political economy in the Buchanan sense was meant to solve: separating individuals at the constitutional level from their particular interests in order that the constitution foster the general interest. Absent the Veil of Insignificance, the constitution serves the general interest of those writing the constitution.

Paper gated permalink below:
Expressive Political Behaviour: Foundations, Scope and Implications

Wednesday 8 June 2011

Altruism and government

On the three days a week I'm an anarchist, I'm sometimes asked what would become of the poor or those in severe hardship if the state weren't there to protect them.

Without the State, the homeless could be fed:
Members of Orlando Food Not Bombs were arrested Wednesday when police said they violated a city ordinance by feeding the homeless in Lake Eola Park.

Jessica Cross, 24, Benjamin Markeson, 49, and Jonathan "Keith" McHenry, 54, were arrested at 6:10 p.m. on a charge of violating the ordinance restricting group feedings in public parks. McHenry is a co-founder of the international Food Not Bombs movement, which began in the early 1980s.

The group lost a court battle in April, clearing the way for the city to enforce the ordinance. It requires groups to obtain a permit and limits each group to two permits per year for each park within a 2-mile radius of City Hall.

Arrest papers state that Cross, Markeson and McHenry helped feed 40 people Wednesday night. The ordinance applies to feedings of more than 25 people.
And tornado victims could be helped:
Mike Haege owns a tree-trimming business in Hastings, Minnesota. After a tornado hit northern Minneapolis, he decided to help out. On May 23, the day after the tornado, he signed up as a volunteer and brought some equipment to help people without insurance to dig out from the damage. Mike and his fellow volunteers removed fallen or damaged trees from driveways and doorways, all free of charge. He probably made a lot of friends that day.

Regulators were not among them. While he is licensed to work in many Minneapolis-area cities, he isn’t licensed in Minneapolis proper. So they kicked him out of the city.
HT: @MitchellHall on the first, @TPCarney on the second.

Tuesday 7 June 2011

Go Canucks!

CBC Vancouver interviewed me last week about prices and ticket scalping; the resulting feature aired Monday morning and is here. The CBC used a lot more of their interview with Mike Munger than they did of mine, but that was efficient.

A few bits that didn't make the cut:
  • Economists are usually puzzled by scalping; why would the venue be leaving money on the table?
  • In the music industry, the puzzle was largely solved by that the venue was in cahoots with the scalpers and earning a good chunk of those rents; but, I'd be surprised if the Canucks were selling blocks of tickets via StubHub.
  • The Canucks are doing very well financially. Why might they be selling tickets lower than market clearing rates? To keep the loyalty of long term season ticket holders for starters: those folks might fail to renew tickets if they thought they'd been hard done by. And, the mix of fans at the game changes in ways potentially detrimental to television revenues and to merchandise sales; the Canucks have a more complicated maximization problem than just "get the highest possible ticket revenue."
We'd also talked a bit about the benefits of price gouging in Christchurch after the earthquake, but that went far too far afield for this piece.

I did chuckle that Mike was introduced as being a wacky libertarian while no such disclaimer was put on my bit.

Kanazawa [updated]

I've been a fan of Satoshi Kanazawa; I've long had regrets that I didn't know he was at Canterbury for the couple of months we overlapped there.

Not entirely unsurprisingly, he's gone and annoyed some folks. This time it's looking serious. The Add Health data series interviews high schoolers in three waves, making a nice panel data set. One question asks the interviewer to rate the respondent's attractiveness on a five point scale. Satoshi ran some regressions on attractiveness and found racial differences in means after correcting for possible confounds like weight; black women, but not men, were found less attractive in the surveys. He then speculated about whether testosterone levels might account for the result. His blog post, as usual, was pretty blunt about what he'd found; it's mirrored here as it's now been pulled.

The pile-on has been pretty brutal. He's been called a racist for finding data suggesting black women are less attractive than white or asian women; I'm not sure whether he's a racist also for finding data suggesting that there are no big racial differences in attractiveness among men.

Here's Huffington calling him a racist.

Lindsay Beyerstein is less than charitable in her interpretation of Kanazawa's stats. She gets the last wave of Add Health data and says that the difference disappears by Wave Four, raising troubling questions about Kanazawa's bias. I'd say rather more likely, he just had the first three waves' data sitting on his hard drive; getting the fourth wave would have been a pain in the arse for a blog post, so he just used the data at hand.

Hank Campbell is no more generous, with lots of snarky scare quotes about what factor analysis is. Because three interviewers rated respondent attractiveness at different points in time, you need to draw some summary stat out of the three observation. I'd have just gone with a straight average, maybe weighted towards the latter waves when the respondents were older. Kanazawa ran a factor analysis instead. The difference between the two isn't going to be great - factor analysis will try to extract some common underlying measure from the three observations, making the weights across waves endogenous. But Campbell likes to say 'factor analysis' with the scare quotes to make it seem dodgy.

The first thing I'd thought when I saw the controversy was that OK Cupid recently put up data noting that black women get far fewer messages from other OK Cupid members than they ought to; this was potentially consistent with Kanazawa's story (or with other equally plausible ones). But Campbell calls Hontas Farmer a racist for citing that data.

Now, Huliq reports Kanazawa's lost his blog slot at Psychology Today (one wonders how long Walter Block will last).

Scientific American wonders whether Add Health should be collecting data on interviewer-rated attractiveness:
I am disturbed by the fact that the Add Health study's adult researchers even answered the question of how attractive they rated these youth.
Never mind that a ton of research on kids' social capital would draw on measured attractiveness as a potential explanatory variable; apparently it's better to make things unknowable than to risk disturbing findings.

The Daily Mail insinuates that Kanazawa's a racist for his prior work suggesting IQ might be responsible for some poor health outcomes in Africa, and cites an LSE colleague calling for his firing.
It is not the first time that Dr Kanazawa, 48, a lecturer within the department of management at the LSE, has been accused of peddling racist theories.
In 2006 he published a paper suggesting the poor health of some sub-Saharan Africans is the result of low IQ, not poverty.
Professor Paul Gilroy, a sociology lecturer at the LSE, said: ‘Kanazawa’s persistent provocations raise the issue of whether he can do his job effectively in a multi-ethnic, diverse and international institution.
‘If he announces that he thinks sub-Saharan Africans are less intelligent than other people, what happens when they arrive in his classroom?’
He added: ‘The LSE risks disrepute if it fails to take a view of these problems.’
Here's Linda Gottfredson on IQ and health; here's Garrett Jones on IQ and economic outcomes.

Britton at Scientific American, linked above, raises a lot of better questions about whether Kanazawa's findings would stand up to more thorough investigation; so does Robert Kurzban. But it was a freaking blog post! Blog posts are where you put up initial data exploration and speculation to bat things around and see whether it's worth more thorough investigation. If you disagree with the analysis on a blog post, you write up your own post on why you think it was wrong or how it could be done better (like Kurzban); calling for Kanazawa's firing borders on witch hunt.

The LSE beclowns itself if it sanctions Kanazawa for this particular blog post.

I take more seriously Andrew Gelman's critique of Kanazawa's published work. It's fine to be wrong; it's a bit worrying that Kanazawa posted subsequently on that study without noting or addressing the critique (I'll take Gelman's word on this). [See below] As for how much weight I'd put on the soon to be released open letter of a bunch of psychologists castigating Kanazawa, well, I'd want to know what kind of Sneetches they are.

Michael Mills's "Seven Things Satoshi Kanazawa Cannot Blog About" is a must read...

Update: I've read Gelman's critique in more depth. Gelman's an excellent statistician. But some of the criticisms there lodged would apply to a reasonably high proportion of published empirical work. Endogeneity issues are everywhere; damning everyone who's ever had potential endogeneity / reverse causality problems in their published work would be a bit broad. And failing to adjust significance tests for the potential number of comparisons (as a guard against data mining) - I have a hard time thinking of many published pieces that have done that other than the metastudies that say we can't trust any empirical work.

Gelman's specific (and not at all unreasonable) worry on datamining is that Kanazawa's work on whether more attractive couples have more daughters tests whether the most attractive couples have more daughters than all others; equally plausible would be tests of whether the least attractive couples had the fewest daughters, the top two categories of attractive couples against the rest, and so on. XKCD summarized the problem here. But subsequent work with a different data set found the same result; matching the prior paper's result via datamining would then have taken mining across different datasets until finding the one that gave the best match, and I'm not sure there are all that many datasets that include attractiveness data.

Finally, Gelman (2007) critiques Kanazawa's earlier (2005) work for missing that there are potential problems in using number of daughters on the right hand side of a regression equation and number of sons on the right if some couples use a stopping rule that aims at particular ratios. But Kanazawa's 2007 piece recognizes that issue. I'm not sure whether the prior pieces' results were sensitive to this specification issue, but I'm also not sure it's right to say, as Gelman implies, that Kanazawa then went on to do other work without taking due account of critics' view of prior work.

Monday 6 June 2011

Hacking the city

Thanks to @robyngallagher for the pointer, a must-read on Newcastle's redevelopment - cities as software.

According to the piece, hackers routed around city regs to make abandoned old buildings useful again.
Renew Newcastle started by hacking how much spaces cost and the terms they were available on. While there were over 150 empty buildings in Newcastle few if any of them were cheap or simple to access. They were bound up in complex rules – from bad tax incentives to complex, costly and long-term commercial leases that made it difficult to access them flexibly. Renew Newcastle traded cost for security. We created new rules, new contracts, and convinced owners to make spaces available for what was effectively barter – we would find people to clean them use them and activate them and they could have them back if and when they needed them. We stepped outside the default legal framework in which most property in Australia is managed and created a new one. We used licenses not leases, we asked for access not tenancy and exploited the loopholes those kinds of arrangements enabled. While such schemes are institutionalised in many European countries they have little precedent in Australia – in Newcastle, the entire scheme was devised, brokered and implemented directly from the community without the involvement of a government or formal development authorities still grasping at hardware based solutions. Only after the first dozen buildings had been activated did any funding appear. More than two years later any changes to rules and regulations – to the operating system – are yet to transpire.

Yet cheap space is not in itself enough. It is not enough to simply change how much space costs, it is also vitally important to change how it behaves in the face of initiative. Renew Newcastle created a whole system to lower barriers to initiative and experimentation. We created another layer – between the operating system and the users to make it simpler and easier to enable experimentation and risk.

Again we followed the path of least resistance. We decided to make things simple that could be made simple and not butt up against what would remain impenetrably hard. We managed to do what is easy rather than get caught up in waiting for the ideal – to find spaces that were usable and use them. Renew Newcastle designed systems – an API in programming terms – that made activation simple. We took spaces, brokered cheap access to them and gauged what could be done in them easily – what they were already approved for – and set out to find it and plant and water it.

In doing so we effectively made a whole system to make space behave as quickly and responsively. To allow people with enthusiasm and passion to direct it into the city. We made it quick for people to try and cheap for them to fail. We removed capital and complexity from the equation and in doing so we seeded more than 60 experiments – unleashing the energy of hundreds of people.
Christchurch's version, GapFiller, faces a rather worse problem. But they're doing a lot to help make the city liveable until the hardware issues are sorted out.

Sunday 5 June 2011

Escalation in the police state

The videos Alex posted (I caught them via BoingBoing), along with MR comments on the video, induce despair. [See also TechDirt]

Subsequent to a judicial decision that it is right and proper that the US government use force to prevent people from dancing at the Jefferson Memorial, a flash-mob tested it by wearing earphones and dancing while at the Memorial. And some were thrown to the ground and handcuffed.

Commenter Ben at MR writes:
The whole point of liberty is that you can do what ever you want until it infringes on other peoples ability to do what they want. These people set out to deliberately confront the police and make a scene, infringing on the ability of everyone else to enjoy the monument.

I think these officers behaved entirely appropriately, and removed a bunch of people who were causing a disruption, resisting arrest, and causing an unsafe environment for normal Americans trying to enjoy a day at the monument.
He's joined by others of similar mind at TechDirt.

I wonder what fraction of the US population, or the Canadian one for that matter, would be willing to pay to get to watch a hippie be beaten by police, or share Homer Simpson's fantasy of driving a plow truck through a crowd of protesters.

I wasn't there so I don't know great a disruption was caused by the dancing prior to police intervention. But wouldn't a more sensible police approach have been to just dance with the flash mob for fifteen minutes 'till the crowd decided to disperse? Unfortunately, that option was closed once the park police asked folks to stop and people continued quietly and nonviolently dancing.

The point of civil disobedience is to force the cops' hand to get exactly the kind of video Alex linked: a disproportionate police response to nonviolent behaviour. There was little chance the cops would back down after the request to disperse was ignored, and no way that the cops wouldn't come out looking terrible. When peaceful behaviour is made illegal, this is about the best way of showing the law to be an ass. When enough folks side with Ben, above, in cases like this or in far far worse cases of police abuse, peaceful folks ought to consider their exit options.

I'd hoped to take the kids to the Jefferson Memorial as we're in town. But the three year old and the one year old dance spontaneously. Going home to earthquakes is seeming less bad...

Saturday 4 June 2011

A thesis waiting to be written

The university administration here recently asked for information about any research currently being undertaken here relating to the Christchurch earthquakes.

I am sure that the quake will provide lots of scope for future research, in geology, engineering, and social science. I am not aware of any research in economics, currently, but one thesis I would love to see written is on the volunteer work in the aftermath of both the September 4 and Feb 22 quakes.

A major part of this story, of course, is the sheer magnitude of the volunteer work. (My favourite anecdote: My daughter had her clarinet exam, originally scheduled for the week after the Sept 4 earthquake last year, postponed by a week because of the quake. The examiner, a middle-aged woman who had been brought out from England for this round of exams, was already in Christchurch. I asked whether she had left the city during the week exams were cancelled. Oh no, I was told, she went out to the eastern suburbs to help shovel silt!)
Being a social science geek, however, the thing I find most fascinating about the volunteer activities was the high levels of coordination that existed with activities that originated in the bright ideas of lots of people, such as the lunchpacks made for members of the student army, prepared by volunteers in Dunedin, and driven up to Christhchucrh by other volunteers overnight. The student army originated as a single student’s idea after the September quake, communicated to others via Facebook. Is this an example of a Hayekian spontaneous order, with social networking providing the platform for a coordinating network? Or did civil defence and other official organisations have an important role to play in the coordinating mechanism? To what extent was the scale of the volunteer activity this time possible only because the system evolved from the smaller-scale activity last year?

Obviously, whatever the local coordinating mechanism, the whole activity would not have been possible without the global price mechanism allocating resources (think I pencil, with everyone adjusting their behaviour to help the Christhchurch volunteer work), and clearly it is relatively easy to have spontaenous volunteer work when there is so much work that can be helpful, allocating the available help to the most pressing concern is less important than simply increasing the total amount of help. But even acknowledging those points, there is an interesting question seeking an answer here.

Hopefully, there is a lead player in the student army out there with a background in social science (ideally including economics), who would be keen on taking this on as a thesis topic.

Friday 3 June 2011

Alcohol reform

Luke Malpass surveys alcohol reform in New Zealand, providing a decent history of the regulation and deregulation of alcohol and of the current controversies about the social cost of alcohol in New Zealand.
Liquor reform has become a political hot potato in New Zealand. Social problems caused by alcohol have meant successive governments have come under pressure to regulate sale and consumption of alcohol. One factor driving public outrage has been reports that give outrageously high ‘social costs’ of alcohol, that discount benefits, and that have little economic basis in fact. The NZ Law Commission’s report into reregulating the sale of liquor was overly reliant on public health literature that made a) a series of erroneous economic claims and b) holds values based views parading as objective evidence. A Liberal alcohol regime should be continued, in which personal responsibly is emphasised.
On Luke's request, I provided a short summary of my work with Matt on the BERL study and subsequent critique by Marsden Jacob; it appears as a two-page inset to the monograph. Enjoy!

Thursday 2 June 2011

Voting is for chumps

Would this post's title be illegal were it posted on election day?
Social media sites including Twitter and Facebook will be monitored on election day to ensure electoral rules are not breached, Chief Electoral Officer Robert Peden says.
It is illegal to campaign on an election day, a prohibition which covers the publishing or broadcasting of anything intended to influence votes.
The Electoral Commission says the rules apply to social media sites including Twitter, Facebook and blogs - and any breaches will be followed up.
With tweeting becoming increasingly popular, Mr Peden says, the commission will keep an eye on electronic media communications on polling day, which is on 26 November.
"If people tweet on election day in a way which is trying to influence how somebody votes, then that's a breach of the act and we'll be following it up."
Mr Peden says the Electoral Commission will consider whether it needs to do anything more to ensure people are aware of the rules.
It sounds like pages that were up prior to election day can stay up, but no new content can be added:
Chief Electoral Officer Robert Peden said material posted on social media websites was covered by strict rules which prohibit electioneering on election days.

"People should be aware that if they tweeted on election day to influence how somebody votes they will be breaching the [Electoral] Act and the [Electoral] Commission will take action."

He said while people could leave websites with campaign material up on election day, they could not add further material or advertise the website.

"For a long time, the law has allowed for campaign-free election days, and my sense is that New Zealanders like it that way and so it's not really in people's interest to do things like tweet and breach the rules."

The sites would be monitored on November 26, and people caught breaking the rules could face fines of $20,000, Mr Peden said.
I have no interest in electioneering - I abstain from voting. But I could imagine posting the latest odds from iPredict along with strategic voting suggestions for those who feel compelled to vote. Would that be illegal? How about encouraging people not to vote? That may affect whether people vote rather than the content of the vote.

What's worst about the rule? New Zealand no longer has bragging rights over Canada, where regulations prohibit the publishing of election results from eastern Canada before polls close in the West.

Wednesday 1 June 2011

Over and underestimating effects of minimum wages editor David Chaston blames raising the youth minimum wage for the loss of 30K jobs.

I think that's overstated: we can't just look at how much more the youth unemployment rate rose than the adult unemployment rate; we have to look at how much more it rose than we would have expected it to rise given the increase in the adult unemployment rate. We expect youth unemployment to be worse hit in a recession than adult unemployment; this time, it's been rather worse than in prior recessions. I've pegged it at around 10-12k more kids unemployed than we'd expect given the prior worst performance of youth unemployment. The method's described here; results here. It remains possible that something else happened that threw a great big wedge between the adult and youth unemployment rates, but you'd need something that happened about the same time as the change in minimum wage legislation and that has persisted from 2008 to present and that has a plausible theoretical reason for causing the effect and is big enough to have done the job.

On the other side, the New Zealand Herald columnist Tapu Misa says minimum wage critics are falsely painting Labour as folks ignorant of economic reality, citing Card & Krueger and Hyslop & Stillman as evidence against disemployment effects of minimum wages. But the American studies over an era where minimum wages were about a third of average wages just don't carry over well to New Zealand, where the minimum wage is half the average. And the Hyslop and Stillman paper, while very nice, spanned a period during which ability to fog a mirror was sufficient to get a job in New Zealand; increases in the youth minimum wage over that period were unlikely to be binding.

It's a bit ironic that Misa writes:
Since the global financial crisis, I've wondered if even economists understand what Thomas Carlyle called "the Dismal Science".
As all economists know, or ought to know, Carlyle called economics the Dismal Science because economists like JS Mill opposed slavery; Carlyle thought that slavery was necessary to turn blacks from subhumans to being fully human. Economists were dismal for wishing to deny blacks the beneficent lash, in contrast to the gay science of poetry that recognized blacks' inferiority. And the minimum wage was introduced by early twentieth century progressives deliberately to disemploy blacks and women, rendering them unemployable.

Those wishing to ally with Carlyle in deeming economics as dismal ought consult the work in which he coined the term: his "Occasional Discourse on the Negro Question", or the US version entitled "Occasional Discourse on the Nigger Question". And their research on the economics of minimum wages really ought to go beyond literature reviews from the Council of Trade Unions.