What’s in the Canada Elections Act?
A couple of weeks ago I promised to do a blog post on campaign spending. This was provoked by the Conservatives calling the election 11 weeks in advance of polling day, making this campaign the longest Canadian general election of the 20th century. The thrust of the media commentary was that by engineering such a long campaign the Conservatives have given themselves a distinct spending advantage over their opponents. I am going to consider these claims in detail over the next few weeks, setting out the rules that govern campaign spending in Canada, examining the magnitude of the Tories’ fund-raising advantage, and then summarising what a lot of political science research tells us about the impact of campaign spending. I’ll start today by going over the rules that govern campaign spending in Canada.
In Canada, spending related to federal elections is governed by rules and regulations set out in the Canada Elections Act (S.C. 2000 c.9). A copy of the CEA can be found at http://laws-lois.justice.gc.ca/eng/acts/e-2.01/. Helpfully, previous versions of the CEA are archived at the same site, so that one can figure out when and how the CEA has been amended over time… and it’s been amended a lot since 2003, 33+ times.
The CEA governs all manner of election-related spending and fund-raising, including spending on party leadership competitions, candidate nominations, third-party advertising, and political donations. I won’t consider spending on party leadership competitions or candidate nominations in any depth. Instead, I’ll focus on the rules governing campaign spending, third-party advertising, and political donations.
The CEA adjusts election expense limits for registered parties and candidates and election advertising limits for third parties by an “inflation factor” that corrects for purchasing power. The inflation factor is fraction defined by the division of the annual average CPI for calendar year prior to the election by a designated base year’s CPI (currently 1998) calculated on the basis of 1992’s CPI = 100. This year the inflation factor is 149.0/108.6 = 1.372 (http://www.gazette.gc.ca/rp-pr/p1/2015/2015-03-21/html/parliament-parlement-eng.php).
The CEA draws a distinction between spending by i) registered parties and ii) electoral district associations and candidates. Thus the CEA can be viewed as regulating a party’s national spending ad its local spending.
At the national level…
Section 430 of the current version of the CEA sets out the maximum election expenses of registered parties as the product of $0.735 x the inflation adjustment factor of 1.372 x the number of registered electors for electoral districts in which the registered party has endorsed a candidate. This gives $1.008 per registered voter. I have not been able to find the number of registered voters for this election, but there were 24,257,592 registered voters in 2011 and about 200,000 are added per annum so that puts this year’s electorate at approximately 25 million. The maximum spending limit for any party that runs candidates in all 338 ridings is therefore about $25,200,000.
Now one of the crucial changes that the Conservatives recently made to the CEA was the proviso that should the election period extend beyond 37 days, then the maximum spending limit for registered parties is increased by adding to it the product of one thirty-seventh of the maximum amount and the number of days in the election period minus 37. This means that the maximum spending limit for parties for this year’s 78 day campaign is $25,200,000 x 78/37 = $53,124,324. (To be fair, a reading of the 2nd reading debates indicates that the opposition parties were not particularly concerned about this change.)
At the local level…
The computation of the maximum spending limits for candidates is somewhat more complex. This is because the base amounts that a candidate is permitted to spend varies with the size and population density of the electoral district that the candidate is contesting. Specifically, the amount that candidates are allowed to spend is the aggregate of (a) $2.1735 for each of the first 15,000 electors, (b) $1.092 for each of the next 10,000 electors, and (c) $0.546 for each of the remaining electors. Given an average of approximately 75,000 electors per district, the base spending limit for each candidate is $70,822.50.
However, a further provision stipulates that “if the number of electors on the preliminary lists of electors for the electoral district is less than the average number of electors on all preliminary lists of electors in a general election, then,…the number of electors is deemed to be halfway between the number on the preliminary lists of electors for the electoral district and that average number.”
Further, should the number of electors per square kilometre be less than 10, then then the base amounts listed above are to be increased by the lesser of $0.31 per square kilometre and 25% of the amount calculated under subsection (3), i.e., the $70,822.50.
The upshot of these last two provisions is that candidates’ base spending limits are disproportionately higher in less populous and less dense (i.e. rural) ridings. You can see this by comparing the 2006 spending limits for Cardigan PEI (2013 population: 36,005) – $63,114.75 to that of Parkdale-High Park, ON (2013 population: 105,103) – $75,877.63. So despite having a population (and presumably an electorate) about 3 times as large as Cardigan’s, the base spending limit for candidates in Parkdale-High Park is just 20% higher. This formula reflects a presumption that i) campaigns incur high fixed costs, and ii) that campaigns in densely populated urban areas enjoy significant economies of scale.
As with the national spending limits, the base limits for candidates are multiplied by the inflation factor (1.372) and prorated for the length of the campaign. Taking the average figure of $70,822.50 this suggests that each candidate this election is allowed to spend up to $204,841. Elections Canada will, I’m sure, post the precise limits for each electoral district shortly – but even my rough calculations suggest a big jump in the limit from 2006, when it averaged just $81,200 per candidate.
There are a number of knock-on effects of this high per candidate spending limit. For example, candidates are allowed to spend up to 20% of the election expense limit on their nomination races.
Finally, the CEA limits what third-parties can spend on advertising in relation to a general election. This limit is straightforward: third-parties cannot spend more than $150,000 from the time the writ is dropped to polling day multiplied by the inflation factor and prorated by the length of the campaign beyond 37 days. So for 2015, the 3rd-party spending limit is $150,000 x 1.372 x (78/37) = $433,848.
But there are restrictions in how this money can be spent. Provisions prohibit a group from splitting itself in two to avoid the spending limit, for example. Further, not more than $3,000 (of the $150,000 base amount) can be spent to promote or oppose the election of one or more candidates in a given electoral district.
So what’s the bottom line here? Well, nominal spending limits are a lot higher for the 2015 election than they were for the 2011 election. About 2.5% of the increase is due to the natural growth of the electorate. A further 17.5% of the increase is due to an increase in the inflation factor from 1.165 in 2011 to 1.372 this year. Finally, the extended length of the campaign is responsible for doubling of the limits (i.e. 78/37 = 2.11). Of course, on a per day basis the limits remain just 20% higher than in 2011. This raises the question of whether it’s i) the sheer amount that parties and candidates spend that matters or rather ii) the rate at which and the amount of time over which parties can spend that matters.