June 29, 2004

the collective will in Canada

Yesterday's election in Canada produced a hung parliament, a situation often as challenging for journalists as for the politicians.

(Note, for example, the WashingtonPost article which suggests that the Governor General will have a substantive role to play in the process! Yes, and Paul Martin is the second coming of McKenzie King ... But I digress)

No, the real confusion is for the journalists in parliamentary democracies who struggle to make some sort of sense out of the collective result, often implying that individual voters wanted a hung parliament and divided government.

An article in the Globe and Mail said:

The electorate decided that Mr. Martin deserved to lead the country, but gave the Liberals the message that they must be less arrogant and power-hungry. The Liberals got 36.7 per cent of the popular support compared with 29.6 per cent for the Conservatives and 15.7 per cent for the NDP.

Really? It sounds like actually about a third of the country wanted the Liberals to continue in office (some of whom might have been holding their metaphorical noses as they voted), but that the remaining 2/3 of the country, certainly the 1/3 that voted for the Conservatives would actually be quite disappointed.

(Since neither the Bloc Q. nor the NDP were realistically expected to lead a government, it's hard to tell in the absence of preferential/instant run-off voting who voters for these parties preferred between the Liberals and Conservatives).

It's a fundamental misunderstanding of how parliamentary democracy works to suggest that electorates or countries as a whole have some sort of collective wish for a hung parliament.

When one party wins a majority [of seats or votes] it's easy to write that up as the country generally preferring that party. But when no-one wins a majority most people are likely to be disappointed, and it's the gritty [pun intended] reality of parliamentary debate and coalition formation that takes over from where the electorate left off.

As to the actual outcome of the election ... I've probably paid more attention to the Canadian election than most people in the United States, even in a state bordering Canada, but that's not saying much (!) and there's certainly a lot about Canadian politics I don't get ... with that disclaimer ...

the Conservatives: One of those situations where the press seems to be saying "great campaign, not such great results" and not wondering how both those things could be true at the same time.

Electorates may or may not prefer divided governments (the evidence is that they tend to do so by voting for different parties at different levels in a federal system), but they certainly tend to give governing parties only so long before they give the other party a chance. And if the Conservatives couldn't win this time after 11 years of Liberal government, and some obvious scandal, they really are waiting for the Liberals to absolutely implode.

As for the Liberals, refreshing a party in office is notoriously difficult and generally relies on a weak opposition (which the Liberals have). So, the Liberals could be like the Menzies government in Australia in the 50s and 60s, and just go on and on for two decades, with this as their only stumble.

They have the prospect of a stable coalition partner in the NDP, but there's one slight catch -- they fall 2 seats short of a majority (assuming that the Speaker comes from the majority party as she/he does in other British influenced parliaments).

In any other country, you'd probably then think that the Lib/NDP coalition would court support from the Bloc, but here's where the peculiarities of Canadian politics get tricky (and maybe beyond my judgment).

Dealing too closely with the Bloc probably reduces the chance of the Libs making any inroads in the West, so to get a majority the Libs/NDP will probably try to either (1) peel off some members from the Bloc (or the Conservatives? One MLA switched before the election), or (2) get issue by issue support from Bloc members, who tend to be somewhat more social democratic than the Conservatives.

Neither strategy is risk free, but then nothing ever is, and it's better to be in government than out of it.

One advantage that the Lib/NDP coalition could have is that the real polarities in Canadian politics seem to be between the Bloc and the Conservatives with their regional bases, and different views of how the Canadian federal system should work.

The Libs and the NDP aspire to be truly national parties, though as the parliament is made up of members elected by plurality in geographical constituencies the parties can't craft a truly national message.

Perhaps though in this hung parliament the Liberals should call the bluff of the Bloc, offering them participation in goverment in exchange for giving up some of their agenda, or calling the bluff of both the Bloc and the Conservatives by proposing some devolution of federal functions.

Posted by robe0419 at 4:27 PM

June 28, 2004

not all deaths are created equal

Around 3000 Americans have died in terrorist attacks in the last three years, most on September 11 2001.

Despite this, the risks of dying in a terrorist attack are remarkably low compared to dying in a car accident, or a domestic homicide -- equally grisly, painful, senseless ways of dying.

No matter who wins the presidential election, the American public will demand that they waste billions of dollars reducing the risk of dying in a terrorist attack even lower than it already is. Why? And is it really such a waste?

Why? Because not all deaths are created equal; and individually (and as a society) we will value lives saved (or deaths averted) at different amounts depending on how those deaths occurred.

Firstly, people will pay to reduce the uncertainty surrounding death, and terrorism really does raise the uncertainty about when and how people might die.

On a cost-benefit basis we might extend the quality and quantity of people's lives more cheaply by ploughing billions of the military budget into medical research, but it's not altogether clear that extending the lives of 85 year olds is worth more than reducing the chance that younger people will die in terrorist attacks.

That is, the 85 year olds have had a "fair innings", and terrorist attacks may disproportionately target the young.

Moreover, the 85 year olds who expire in their sleep, and the car accident victims die more or less alone, whereas terrorism of the kind that threatens the United States seems likely to kill hundreds or thousands in concentrated events.

The people who are threatened by terrorist attacks are somewhat more identifiable ex ante than people who will die in car accidents or homicides. The risks are higher in certain places (like New York and Washington), creating a powerful incentive for political action amongst the potentially affected. Car accidents distributed at random across the countryside can't create political alliances.

Finally, the problem of guarding against the risks of terrorism is a true collective action problem, with the risks not easily quantifiable and highly concentrated losses when events do occur, meaning that private responses to the problem are likely to be ineffective.

Of course, the foregoing abstracts away all the political issues that motivate the threat, but that is the point -- even absent the political/religious dimensions of the current terrorism there would be powerful political forces demanding a government solution to the problem.

Add politics, religion, money and oil, and stir ...

Posted by robe0419 at 11:26 PM

June 25, 2004

government by scientists

Franklin Foer has an excellent article in The New Republic about the denigration of the hard sciences and social sciences in the Bush administration.

Foer notes that the advent of government by social scientists began in the Progressive era, and gradually expanded over the course of the two world wars, the cold war and the war on poverty.

In a Masters essay (PDF) I looked at the origins of government by social scientists in the work of Richard Ely, and Sidney and Beatrice Webb.

While I'm mostly supportive of scientifically informed policy making, I argue that its supporters can also slide into believing that politics and government can be reduced to a series of technical questions that can be rationally resolved. In part they can, but ultimately government decisions reward some people and don't reward others (tax policy, anyone?).

What's particularly egregious about the Bush administration is that they say their decisions will be "science based" when they're not, they're ideologically and religiously based. It's not a pretty combination.

Posted by robe0419 at 3:17 PM

identifying the vice president addendum

Yesterday I mentioned that to statistically identify the impact of vice-presidents on presidential races we'd need data points where the same vice presidential candidate ran with different presidential candidates, which was wrong. Actually, we'd need examples of presidential candidates running with different vice presidential candidates.

Of which we have some examples even in the modern era; Roosevelt and Nixon come immediately to mind. But this is not a lot of data, and you couldn't exactly include a lot of other variables in the model.

My supposition is that (1) at an aggregate level the vice presidential impact on the ticket is minor once you've taken out the effects of demographic and economic conditions , and (2) the impact at the level of the individual voter is low once you've accounted for partisan affiliation, opinions on the issues, and the quality of the presidential candidates (except perhaps in their home states).

Posted by robe0419 at 7:36 AM

June 23, 2004

identifying the vice president

There has been lots of speculation about who John Kerry will pick as his vice presidential candidate. The reason there's a lot of speculation, and nothing definitive is (1) the decision is all in Kerry's head (and maybe some people close to him), but more importantly ... (2) it is impossible to statistically identify the impact of vice-presidential choices on elections since the data we require don't exist.

Thus everyone can find analogies to support their argument, but no conclusive data.

With the exception of George Clinton and Rufus King who were the VP nominees in both 1804 and 1808 we have no other data on repeat candidates for the vice presidency which would allow us to control across time for the differences in presidential candidate quality.

This isn't to say that it would be theoretically impossible to estimate the impact of the VP nominees on presidential elections. Politically, however, if you are a losing VP candidate your career goes to one of four places (1) back to the Senate (2) failed attempts to be President, or your party's nominee (3) appointed to offices when your party finally wins the Presidency again, and (4) private law practice.

That said, Kevin Drum thinks the decision is consequential for the long-term future of the Democratic party, and his handy tabular summary of the VP's is indeed useful.

However,let's look at the failed VP candidates before we conclude that the VP pick is that important.

VP nominee What happened after losing
John Sparkman (1952) long-term AL Senator
Estes Kefauver (1956) TN Senator until death in 1963
Henry Cabot Lodge (1960) Various ambassadorial posts, including Vietnam under Kennedy Johnson
William Miller (1964) Law practice
Edmund Muskie (1968) ME Senator, then in Carter's Cabinet
Thomas Eagleton (1972) Lawyer
Bob Dole (1976) KS Senator, Presidential nominee 1996, Viagra and Pepsi advertisements after retirement from Senate
Walter Mondale (1980) Presidential nominee 1984 (lost), late replacement Senate candidate 2002 (lost)
Geraldine Ferraro (1984) Democratic appointee to UN Human Rights Commission and other appointments
Lloyd Bentsen (1988) Clinton's Treasury Secretary
Dan Quayle (1992) Ran for Republican nomination, lost
Jack Kemp (1996) Philanthropic and think-tank memberships
Joe Lieberman (2000) Ran for Democratic nomination 2004 and lost. CT Senator

So, we have 4 [Democrats] remaining in the Senate, 2 nominees who went onto gain their parties presidential nomination and lose pretty badly, and the remainder who bowed out of elective public office. Not exactly a stellar record.

Posted by robe0419 at 7:04 PM

what use is the word then?

Henry Farrell at Crooked Timber notes the difference in how Americans, and Britons/Antipodeans use the word "quite"

When I first arrived in America I was bemused by how seemingly polite Midwesterners would say "It was quite nice to meet you." As these phrases were generally the conclusion of five minute first-meetings I'm sure that quite=moderately was the accurate judgment of meeting me (maybe even an overly good judgment!).

But the conversational convention most places seems to be that you should err on the side of effusiveness in saying how nice it was to meet someone.

The OED notes that the contemporary American usage is "usually felt to be old-fashioned or stilted, and has become less common," another example of how word use in the United States reveals its seventeenth century separation from English English.

Posted by robe0419 at 2:25 PM


John Judis, guest blogging for Josh Marshall, has an articulate post about the Bush administration's tightening up of foreign student visa rules.

Now it's true that several of the 11 September hijackers entered the U.S. on student visas, and that they never showed up at the English language school they were supposedly entering. And the ability of the [former]INS, now BCIS, to track foreign students was pretty much non-existent.

But the response of the Bush administration has been to throw the proverbial baby out with the bathwater; in this case to create a lot of obstacles for legitimate students who make a great contribution to American universities and the American economy.

To wit, students from China (!) have been particularly slowed in their visa applications. Whatever the other merits or demerits of having lots of Chinese graduate students enter the country, no-one to my knowledge has ever suggested that China was sending lots of potential terrorists to the United States.

Moreover, the 11 September hijackers all enrolled in privately operated English language schools. These kinds of operations are certainly deserving of more scrutiny.

But students coming to study at major universities, and those who come from non-Islamic countries are not where the risk lies.

As Judis says, international educational exchange is so fundamental to making long-term allies in foreign countries that the U.S. should be offering more opportunities to the best students in Islamic and Arab countries to come to the U.S. to study. Of course, some oversight is necessary, but the long-term payoff of giving future opinion leaders the opportunity to see America up close, is great enough to outweigh the slight risks.

[Update: 30 June 2004]: an op-ed in the WaPo addresses the same issues.

Posted by robe0419 at 12:23 PM

June 12, 2004

who is a terrorist

Matthew Yglesias wonders whether the electoral politics of Abu Ghraib might not break in favor of Bush. After all, there's a constituency for erring on the side of doing too much to protect America by torturing terrorists.

Yes ... but what Matt's post really illustrates is how issues are framed is crucial to this election. First, the notion that what went on at Abu Ghraib is in anyway about protecting America needs to be overturned.

And the notion that the people at Abu Ghraib were "accused terrorists," well, being accused doesn't make you a terrorist. Wasn't innocence until proven guilty one of those great liberal democratic reforms we were going to bring to Iraq?

No better place to start than the justice system. By permitting/encouraging American troops [and contractors] to behave in a way that was a little too close to Saddam Hussein's regime the moral case for the war really is undercut.

The "It's OK because we're doing it" and "our hearts are good because we're American" attitude which underlies any defence of what went on at Abu Ghraib is a precursor to continuing torture.

Posted by robe0419 at 10:52 AM

June 10, 2004

seeing double

having become the lucky user of a new PC with dual monitor support recently, I have traded physical desk space for screen space, and now have two monitors on the desktop. excellent for the multi-tasking!

Posted by robe0419 at 3:12 PM

June 7, 2004

Kerry on religion!

this might prove more damaging than all the caveats about Iraq!

Essential reading? Obviously, um, uh, I mean, if you—I don't know if you're picking ten books or five books or whatever, but I happen to believe—a lot of people don't—you ought to read the Bible. People might pick up the Koran or the Torah or whatever, obviously, depending on your religion, but I think a religious foundation—I personally think—is important, whether it's Confucianism or Hinduism or whatever it is.
Posted by robe0419 at 5:34 PM

a true contrarian?

Christopher Hitchens shows that he may still be a contrarian, after all.

I take his last paragraph to mean that contempt for one's political opponents leads you astray more often than it helps you beat them. A good thing to remember when the temptation to be contemptuous is high.

Posted by robe0419 at 5:28 PM

marginal voters

It may seem a strange day to talk about voters in the abstract when Reagan has just shuffled off his mortal coil, but I have nothing much to add to the discussion of his life, so ...

Josh Marshall commented last week on the inanity [and covert racism] of commentators who claim that without black voters Democrats would be a hopeless, permanent minority party.

With Stephanie Herseth's narrow victory in South Dakota last Tuesday this meme is being repeated, though this time with the claim that Herseth's majority came from Native Americans.

It's nothing more than a tautology to say that without some of the votes a party got, they would lose.

But it's a willful misunderstanding of this special election in particular to claim that it was the Indian reservation that ensured Herseth's victory. Native Americans in South Dakota, like blacks across the country, are [as a group] among the Democrats most reliable constituencies. And getting your strong supporters to the polls is part of winning.

But in the June '04 SD special election, where Herseth reversed the result of the 2002 general election, what changed was that Herseth narrowed the margin in Pennington County (Rapid City), and reversed the results in Minnehaha County (Sioux Falls).

Given that turnout in American elections is pretty low, it's possible, but unlikely, that few individual people changed their votes. What's more likely is that among white people living in Sioux Falls and Rapid City, there were people who changed their minds; who had voted for Janklow in 2002 and now voted for Herseth.

As Ruy Tuxeria will tell you, there's a name for these people who do not consistently vote for the same party -- independents. And it's among those voters that the story of seats changing hands really lies. Those people provide the margin, not the most loyal voters.

Posted by robe0419 at 12:48 PM

June 2, 2004

a limited defence of starbucks

once upon a time, way back in, oh, early 2003, I too clung to a vague hostility to Starbucks. (see this post for some other thoughts on this issue). I don't think I could defend Starbucks as the best dominant player in the coffee market, but it's an advance on the alternatives in America.

Starbucks espresso is not great, but it's predictable, and not too bad. That is its virtue.

to be sure, it would be lovely if everywhere you went there was good coffee at independent coffee shops, but most of America is not Amsterdam, Melbourne or Wellington.

If you don't like it, and value the coffee culture that much, move to one of those places.

Or, hey, here's another thought, buy a french press and learn to use it. You know, make the coffee yourself. Not difficult.

Most of America is a place where the majority of the population apparently prefer to drink the vile swill that is Mr. Coffee. And to judge by the fetid kona pots in a lot of gas stations, there is still a market for this stuff.*

Starbucks makes its own offering in this market -- witness the pump pots of coffee they're happy to fill with 20oz of hot brown water.

But Starbucks does also serve espresso, and from O'Hare Airport to Knoxville to D.C., they seem to be make it pretty well. Strangely, the latte is more variable at Starbucks -- normally the latte and its milk is the buffer against poor coffee making skills behind the counter.

If it wasn't for Starbucks, I can just about guarantee you that you would not be able to get decent coffee in much of America, certainly not in any airport, or random off-ramp from I-40.

As for Starbucks chasing out the independent coffee shops, I doubt that its effect is as great as all that. Having been to a fair share of the large cities east of Minneapolis and north of Durham in the last three years, it seems that Starbucks and the independent stores are co-existing pretty well.

It'd be hard to prove, but I'd wager that where there are few independent coffee shops it's because there's not a market for them. For economic or cultural reasons, many Americans want [bad] coffee in a hurry -- they wouldn't have been hanging out in the "Friendly Local Organic Roastery" shooting the breeze with their friends, they would have been grabbing their 20oz polystyrene gulper from the Super America and hopping in their truck.

Americans, it seems, are not willing to wait the extra 3-4 minutes it would take to get a decent cup of coffee -- presumably this time/money/taste trade-off is the optimal one or they wouldn't be making it. It's also possible many folks just aren't aware of how much better coffee can be than the thin brown hot water that is served in many places, and without this information can't make the optimal decisions.

Posted by robe0419 at 4:52 PM