Friday, October 4, 2002
The Globe & Mail

PR sounds good but isn’t: Why I am a recovering electoral reformer

By Brian Lee Crowley

Recently electoral reform, including proportional representation (PR), is back on the national agenda. The BC government has just commissioned a thoughtful former provincial Liberal leader, Gordon Gibson, to look at major changes to the way elections are run in that province. Earlier this week former NDP leader Ed Broadbent teamed up with IRPP President Hugh Segal to make the case for PR in a Globe op-ed.

I once shared this passion for PR, and helped write the Quebec government’s Green Paper touting PR in the late 70s. But over the years I’ve come to have growing doubts that PR would in fact improve our democracy.

PR’s supporters believe that elections are supposed to produce a legislature that is a faithful representation of all the currents of policy opinion in the electorate at election time. Not so: elections are about choosing a government. Under our current system, voters get to do that directly, but are robbed of that ability by PR. Look at New Zealand or Germany or Israel or most of the other PR systems: the choice of government is often the unpredictable outcome of a mad few weeks of behind-closed-doors horse trading after the election is over. It’s Meech Lake every four years.

It is no vice of first past the post (FPP) that it usually boosts the seat count of the most popular party; on the contrary, this “winner’s premium” is a way of boosting the chances that the party with the strongest support can form a stable government. A stable government able to resist the short term shifts of public opinion and unambiguously answerable for its stewardship of power at the next election is a great strength of our democratic system.

Friends of PR believe that elections are essentially policy-based and forward-looking. They are neither. For most people policy is far too complicated and uninteresting. What elections allow them to do is to form judgements about people – something most people are pretty good at.

An election is a contest between competing teams for the confidence of voters. And the first thing to be judged is the performance of the outgoing team. In other words, because actions speak louder than words, past performance counts for more than future promises. That is why incumbents enjoy an advantage. Elections are first and foremost a judgement on the past.

Only where past performance has been disappointing do other parties really have much of a chance. And then the electorate’s judgement is chiefly about intangibles like personalities, trust and confidence. If that is the case, then most votes are only a weak expression of preference for one party’s “programme” or “principles”, and yet Broadbent and Segal, for instance, clearly think that this is the central point of elections.

PR supporters also tend to believe that party platforms ought to bind future decisions by parties, and indeed that the platform ought to trump all other considerations in policy making. Yet this is not at all the way politics works. George Bush didn’t have a policy on the war on terrorism when he was elected; there wasn’t anything to have a policy on. Ditto for the Enron collapse and a series of other matters. And many of the things that people *did* vote for with George W. have been pushed aside by events. That is not a failure of politics or our institutions, but a recognition that politics is a struggle to impose will and order on unpredictable and shifting events and people.

The future is not bound by party platforms. That’s why democracies confer a rather unrestricted ability to make decisions about unpredictable future events on specific individuals and party teams. And voter judgements are relative, not absolute. When Stockwell Day was Alliance leader, a lot of anti-Grit votes went to Joe Clark because, relatively speaking, he inspired the most confidence. Now that Stock has been replaced by Stephen Harper, the situation may be reversed. And it looks like both Jean Chretien and Paul Martin will continue to trump both of them in the confidence sweepstakes, but in shifting proportions as specific issues emerge and strike the public’s consciousness.

What democracy requires is that each election produce a party which is clearly accountable at the next election for its administration of the necessarily unpredictable affairs of the country. The population can then pass a judgement on whether that team should be returned to office, or hand that responsibility to others in whom they have greater confidence.

In other words, PR focuses too much on mathematical representation of parties, on binding politicians to programmes, on principles and on the future, when elections are really about letting the population choose a government, the confidence that potential leaders inspire in people, dealing with the unpredictable and accountability for the past. First past the post, for all its faults, simply does it better.

Brian Lee Crowley is the President of the Atlantic Institute for Market Studies, a public policy think tank based in Halifax.