Wednesday, April 22, 2009

Immigration and Inequality

David Card, perhaps the preeminent practicing academic economist holding a Canadian passport, was here today to give the annual McKenzie lecture, titled, well, "Immigration and Inequality". Fairly interesting. Basically, it argues that high school dropouts and high school grads with no postsecondary are perfect substitutes to employers, and grouping these categories together makes the education profile of immigrants match up with the education profile of Americans. So you do some econometrics and as that hypothesis suggests, American wage inequality is basically unchanged relative to the counterfactual of no immigration. Ungated working paper here.

However, as the guy sitting to my right with a zillion AERs kept gossiping to me about during the speech, the identification strategy wasn't being bought into all that much. Basically, it revolved around using the past history of immigrants to migrate to ethnic enclaves to IV for the supplies of different skill levels of labour with wages on the left hand side, since presumably immigration responds to economic opportunity, i.e. wages.

Anyway, it was nice to get to a seminar. Hausman gave a talk last Friday that I slipped into as well, extremely authoritative in front of an audience. I hear he's vicious from the audience. Next year should bring a lot more of these things, I'll probably be at all the macro bits. I'll talk about courses for next fall post-exams, which start Monday and end the next Monday.

POSTSCRIPT: No disrespect to David Card, but the most memorable part of the presentation was watching two 20-something graduate students muck with the control panel for the lights unsuccessfully, prompting 90-year-old Lionel McKenzie to come down to the floor and set the lights perfectly in approximately two seconds. Pretty good.

Wednesday, April 15, 2009

Terrible think tank reports, a neverending series

Stumbled across a story this morning on the "substantial benefits" Canadians derive from their tax-funded public services. Tracked down the report hoping to get some chuckles out of the cost-benefit methodology. But no, it wasn't that simple. It's also probably a bad sign when the lead author's biography mentions their construction of the "gas gouge meter", something on which I have blogged previously, instead of any professional or educational qualifications, but I digress.

You can look at it if you like, but let me distill it for you: "redistribution is awesome!"

Sure, you can think that the Canadian income distribution is not "fair", whatever that means. However, that's an entirely personal and subjective opinion. The authors probably realized that at some point and figured out that maybe it would be better if they had some numbers to argue their point, e.g. consider the title: "Canada's Quiet Bargain: The benefits of public spending".

Reading that, you'd think the authors would argue that government spending has some benefits, right? This seems like a logical implication. But you'd be wrong. To quote:
[W]e are following the convention in public accounting of valuing public services at their cost. To the extent that public programs are supported by a cost-benefit analysis, our implicit assumption is that the net benefit from public services is zero[.]

Yes, we are starting a study to demonstrate the benefits of public services by assuming that the benefit of public services are zero.

As such, the entire study is nice graphs illustrating that households with lower incomes derive more benefits from public services than the taxes they pay, because in our progressive taxation scheme, the rich pay a disproportionate share of the taxes to fund said services. As such, a majority of Canadians derive (their number is 80%) net benefits from governmental tax-and-spend.

Factually, they're right. But I can do one better. If we collectively decide there's one person nobody really likes, we can drag them out back, lynch them, and redistribute their assets to everyone else. By my calculations, 33,617,546 of 33,617,547 Canadians - that's a whopping 99.999997% - will be better off! Clearly, what a great policy.

The point I'm trying to make here is when economists talk about 'net benefits', they mean 'a potential Pareto improvement', that is after the implementation of policy X, there exists a potential redistribution such that everyone is better off than before the policy. In the report and in my ad absurdium, no such possibility exists, so economists are typically not comfortable using the term 'benefits' to describe the effects of redistribution.

Further, I'm not trying to argue that progressive taxation is a bad thing (I don't think it is at all) or that there aren't lots of legitimate public goods the government should provide (there are), but arguing that we're collectively somehow better off by reallocating the pie is both wrong and short-sighted. Remember, redistribution has costs - see any introductory microeconomic textbook for a primer on the deadweight loss that occurs when taxation distorts incentives, nor are taxes free to collect. So under the assumption that government services are valued at cost, then taxation unambiguously makes us collectively worse off, not better.

Population clock here. Lynching that one guy is probably an even better deal now!

Monday, April 13, 2009


Wrote Geanakoplos: “Who can remember the interest rate that Shylock charged Antonio? But everybody remembers the pound of flesh that Shylock and Antonio agreed upon as collateral."

The post is overall somewhat ignorant on just how much literature there had been out there on the importance of collateral and leverage in real economic effects, one canonical piece that comes to mind is Kiyotaki-Moore (1997) JPE "Credit Cycles", predating anything mentioned in the article. The Kocherlakota piece I mentioned the other day also generates bubbles through collateral-based effects.

Macroeconomists do generate a lot of good ideas, however it's often difficult to tell that they are good until after something blows up. Correct me if I'm wrong (I probably am), but if we accept the housing price bubble was a major contributor to this whole mess (as opposed to the notion that the weird-acronym-finance-jazz would've spontaneously combusted anyway), I'm hard pressed to think of other postwar macro events in developed countries where asset prices and household debt played a large role.

POSTSCRIPT: It's the Merchant of Venice, in case you were wondering.

Saturday, April 11, 2009


Given that there are only three micro classes left before finals, even I can calculate that blogging has negative expected value. I want to put a post up at Macleans about some of the fraternity hijinks I've seen with the nice weather breaking out, particularly a bunch of drunken guys hitting tennis balls off beer bottles with sand wedges. Security didn't let that go for long, to be fair.

Lost the basketball match to the profs 17-11 in a forty-minute game. Yes, we're collectively that bad - and it's a heck of a lot easier to play defense than offense. Hoping for revenge in volleyball on Tuesday.

Had to read a 2009 working paper for macro, feels good to be that current. Returning to MWG to finish micro, feels like easy street. Metrics remains the letdown, but I'll supplement that with the Wooldridge/Imbens summer course in Toronto before the CEA conference. Just have to grind out three weeks and it's all over. Not that I have anything to do over summer. Not that that's a bad thing.