Bang for the buck

Alice at Pundits' Guide points out the surprising lack of voter turnout in a couple of the few Edmonton ridings which actually figure to offer competitive races:

One characteristic of these two ridings that really struck me just now in reviewing their history is their incredibly low turnout, both historically and in the last election. Turnout in Edmonton Centre was 51.6% last time, down from 59.8% and 62.5% in 2004-2006 and 54%-56% in its predecessor seat of Edmonton West. And this is in spite of the tight margins by which Ms. McLellan used to win her seat, defying one of the usual hypotheses for low turnout in Alberta (namely that the one party dominance has usually meant there's little to decide at the polls on election day). For more on turnout in Alberta provincial elections, see the excellent series of features last year from Jason Fekete and Renata D'Aliesio of the Calgary Herald.

Edmonton East next door chalked up just 45.4% turnout in 2008, consistent with the downward trend but overall low turnouts in this riding and its predecessors since 1988. It has consistently shown the lowest turnout of any riding in Edmonton, and perhaps not coincidentally also ranks the lowest on many census measures of income and employment. To put the 45.4% figure into perspective, it is the 13th worst turnout of the last election and 19th worst turnout of any riding in any general election since 1988.

The strategies to hold or win over low turnout ridings might look very different than those needed to win elsewhere, which undoubtedly the parties will be taking into account.

What looks most interesting to me is how the turnout level in Edmonton East actually got to its current position, and what it says about the respective chance of the opposition parties in the riding. Alice points out earlier in her post that NDP candidate Ray Martin's strong second-place finish in 2008 came despite his being significantly outspent by Con MP Peter Goldring (who in fact nearly doubled the spending of his three competitors combined). But it seems to me particularly noteworthy that Martin also managed to pick up more votes than the riding's Liberal contender Nicole Martel put up in 2006 - even though she far outspent Goldring in an election where turnout was at least somewhat higher.

And the NDP's comparative advantage against the Libs is borne out by the dollars-per-vote over the past two election cycles. In 2006, the the Libs' big-money flop naturally produced the worst result on record in the riding ($5.78), while the NDP ($2.33) finished just behind the Cons ($2.21). But that comparison on its face wouldn't seem like the end of the world for the Libs, as it's normally to be expected that parties which spend less will do somewhat better in the dollars-per-vote department based on a core level of support and the effects of national spending. (For evidence in Edmonton East, look no further than the Greens' results: consistent fourth-place finishes, but superficially impressive dollar-per-vote numbers due to the lack of spending.)

In theory, that should have allowed the Libs to appear relatively efficient once they pulled most of their investment out of the riding in 2008. And they did improve in that department, spending only $2.11 for each vote received.

But the NDP managed the impressive feat of increasing both its spending and its return on investment at $2.06 per vote in the 2008 election. That topped both the NDP's own 2006 number, and the 2008 totals for the Cons (who spent thousands more dollars to win thousands less votes than they managed in 2006) and Libs (despite their precipitous drop in spending).

Mind you, there's no guarantee that a fully-funded campaign would push the seat into NDP hands, as it's always questionable whether the next incremental dollar spent in a campaign will help a party as much as previous ones. But particularly in a riding such as Edmonton East with plenty of potential voters available to be pursued, there would seem to be obvious potential for the NDP to bridge its gap against the Cons by making full use of the spending limit next time the riding goes to the polls.