At the time this article was written Gary Moncrief was professor of political science at Boise State University in Idaho.

Some Canadian provinces were well ahead of most American state legislatures in enacting campaign finance laws. Because of the relative importance of political parties in the Canadian political system, most of the research on campaign financing concentrates on the contribution and expenditure patterns of the parties. This article focuses instead on the candidates' expenditures, and compares the situation to that in several American states. Among other things, it asks whether incumbents enjoy a financial advantage in electoral campaigns.

There are two very different issues involved in the question of campaign financing. One is the issue of campaign contributions. The other is campaign expenditures. In the United States, the individual states vary greatly in their approach to contributions. While some states impose no restrictions, most states limit the value of the contributions which can be made to a candidate or political party by an individual, a corporation, or a political action committee.¹ The actual size of that limit is subject to great variation from state to state.

While most American states regulate the contribution side of the equation, there are very few states (Wisconsin and Minnesota) which seek to impose expenditure limits on legislative races. The reason, of course, is that the US Supreme Court's ruling in Buckley v. Valeo makes such restrictions problematic. Because the "freedom to spend" is equated in the US with the "freedom of speech", expenditure limitations are only permissable if candidates voluntarily accept them. The incentive for voluntary compliance is public funding for all or part of the candidates' expenditures, and very few states have been willing to provide public funds for campaigns. Thus, in the US, the regulatory emphasis has been on the contribution side of the campaign finance equation.

In the Canadian provinces, the emphasis tends to be the opposite. While only four provinces have established contribution limits, seven impose expenditure limits.² All seven also have reimbursement provisions, providing for public funding for some of the expenditures incurred during the campaign. The precise terms of those spending limits and reimbursement qualifications vary. For example, in 1988 in Quebec a candidate could spend up to 80 cents per qualified elector, while in Saskatchewan a candidate could spend between $1.94 and $2.62 per elector, depending on the size of the electoral district. But because of the differences in population in Quebec ridings compared to those in Saskatchewan, the differences in the total spent per candidate between the two provinces is not as great as the "per elector" figures might suggest (See Table 1).

There are also differences in the qualifications for and rate of reimbursement. Five provinces have established a threshold of 15% of the vote in order to be eligible for public funding, but in Manitoba the figure is 10%, while in Quebec it is 20%. The proportion and nature of the expenditures subject to reimbursement also vary by province. Further, it should be noted that most of these provinces provide either a direct subvention or a reimbursement to the provincial party organizations, in an effort to "level the campaign playing field".

Three provinces—Alberta, British Columbia, and Newfoundland—have no contribution or expenditure limits. In the past, individual candidate financial reports were not even permanently available for public review in the latter two provinces. Given the differences that exist between the Canadian and American campaign finance systems, and the differences that exist between the provinces themselves, it should be instructive to compare the spending behaviour or candidates among the various systems.

Data

We collected data from the official summary reports issued by the Chief Electoral Officer or Election Finance Commission in six provinces. The data represented the most recent election in each province at the time we began the project. Two of the six provinces (Alberta and Nova Scotia) have since held another election. The six provinces for which we were able to obtain candidate data represent the gamut or regulatory laws—ranging from the permissive, "free-market" situation in Alberta (high ceilings on contributions and no limits on expenditures, no reimbursement provisions) to the comprehensive regulations in Quebec (detailed restrictions on contributions and expenditures, and substantial reimbursement provisions).

 

Table 1 Average Expenditure for Selected Provinces
	All Candidates				Those with 15% of vote			By Candidate Type
Province	Year	Average	No.	1991$	Average	No.	1991$	Challenger	Incumbent	Challenger/Incumbent
Alberta	1989	$12,748	266	$14,112	$14,882	233	$16,474	$11,420	$20,861	54.7%
Manitoba	1990	13,302	199	14,047	17,106	147	18,064	15,177	10,978	76.0
Nova Scotia	1988	19,525	163	23,040	24,040	127	28,387	20,922	29,825	70.1
Ontario	1990	32,394	614	34,208	51,588	354	54,477	44,238	65,773	67.3
Quebec	1989	14,238	534	15,761	27,960	362	30,951	27,362	28,224	96.9
Saskatchewan	1991	21,370	207	21,370	24,186	172	24,186	19,269	31,678	60.8

We collected data on each candidate—their party affiliation, incumbency and electoral status, total expenditures and contributions—in the six provincial elections under study. There are 1,983 candidates represented in the data set. The average expenditures are reported in real dollars and in constant 1991 dollars, to facilitate comparison between elections held in different years.

We find three provinces (Alberta, Manitoba, and Quebec) where the mean expenditures are very similar. Saskatchewan and Nova Scotia share similar expenditures per candidate. By far the greatest average expenditures are in Ontario. This should be no surprise, given the population of that province, and the fact that most ridings are contested fairly seriously by the NDP, Liberal, and Progressive Conservative parties. In fact, Ontario has among provincial legislatures the only real three party-system.

The average figures are distorted by the fact that many of the candidates represent parties which are not competitive in certain districts. Such candidates tend to spend very little money (often because they cannot attract much money), and consequently the provincial average is much lower than the amount actually spent by more "serious" candidates. A more accurate picture is obtained by using figures for only those candidates who receive at least fifteen percent of the votes in the election. (Fifteen percent was chosen because it is the most common "threshold" used by the provinces to determine eligibility for reimbursement.)

By creating a minimum vote "threshold" of 15%, we eliminate almost seven hundred "minor" candidates—the vast majority of whom are in Ontario and Quebec. While about 20-30% of the candidates did not receive at least fifteen percent of the vote in most provinces, the proportion is much higher in Ontario and Quebec—indicating the existence of many candidates representing small political parties.

Screening out these less competitive candidates changes the average expenditure levels—in some cases dramatically. This procedure boosts the mean expenditure level in Alberta, Manitoba, Nova Scotia, and Saskatchewan by anywhere between $2,000 and $5,000. It increases the means very substantially in Ontario and Quebec, as we might expect. In Ontario, the average for "serious" candidates (those with more than 15% of the vote) now exceeds $50,000. In Quebec, the average expenditure virtually doubles. These averages are quite similar to those found in many state legislatures in 1988. We find, for example, means of approximately $48,000 in New Jersey, $36,000 in Oregon, $18,000 in Pennsylvania, and $25,000 in Washington.³ Even accounting for differences between the value of U.S. and Canadian dollars, these numbers are quite similar.

Table 2 Average Expenditures by Candidate Type, US Lower Chambers
State	Challenger	Incumbent	Challenger/ Incumbent
California	$154,605	$592,181	26.1%
Idaho	4,222	7,282	57.9
Minnesota	11,353	14,857	76.4
Missouri	7,899	15,050	52.5
Montana	2,506	3,325	75.4
Oregon	27,524	37,739	64.1
Pennsylvania	9,246	28,139	32.8
Washington	17,450	34,520	50.1
Wisconsin	15,803	17,841	88.7

A more important issue is the relative parity of expenditures between incumbents and challengers. Conventional wisdom in the US holds that because incumbents have greater access to contributions, they are able to outspend their challengers by substantial margins. This is certainly the case at the congressional level, but comparative data at the state legislative level have been hard to come by, until recently. Given the fact that few American states attempt to impose spending limits on candidates, it stands to reason that incumbents would generally be advantaged. On the other hand, because most Canadian provinces impose spending limits, the relative spending levels of incumbents and challengers should be more balanced. An examination of expenditures for all candidates receiving at least 15% of the vote, by incumbent or challenger status shows some important variations. The most obvious differences are between Alberta, where challengers' expenditures were only slightly greater than half (54.7%) those of incumbents, and Quebec, which exhibits virtual parity between incumbents and challengers. Given the fact that Alberta has no expenditure limits while Quebec has the most stringent regulations as well as the most generous reimbursement policy we should not be surprised at these findings. All other provinces sampled fall within a range of 60-76% for challenger-to-incumbent spending.

Several American states exhibit far more imbalance in the challenger:incumbent spending ratios. In California, challengers spent only about one-fourth the amount spent by incumbents; in Pennsylvania challengers spent about one-third as much. On the other hand, Minnesota (76.4%) and Wisconsin (88.7%) reach levels parity found in several of the Canadian provinces. Recall that Minnesota and Wisconsin are precisely the two states which offer public funding in exchange for voluntary spending limits.

The anomaly in Table 2 is Montana, which does not have spending limits, but yields relative parity. We think this is best explained by the fact that very little money was spent by either incumbents or challengers. The "money chase" simply has not hit Montana legislative races yet.

Nonetheless, a few American states without expenditure regulations show ratios similar to Canadian provinces with expenditure limits. The best example of this is the comparison between Oregon (64.1%) and Saskatchewan (60.8%).

It is apparent from the data that the campaign finance system in the Canadian provinces provides a challenger a better chance to approach spending parity than does the system in most American states.

It should be noted, however, that one problem in comparing US data with that of Canada is that in most provinces a third party is strong enough to gather at least fifteen percent of the vote in many ridings. Thus, we are left with comparing challengers and incumbents in a two-party system to challengers and incumbent in a two-plus-party system. If we refine Canadian figures by calculating challenger:incumbent ratios for candidates from only the two top vote-getting parties in each province, the parity ration improves—especially in Saskatchewan. Under these conditions, the challenger expenditures reach at least 70% of incumbents' expenditures in all provinces except Alberta (where there are no expenditure limits). In Quebec, the challengers actually outspent incumbents slightly.

The province of Alberta and the states of Minnesota and Wisconsin are exceptions best explained by the fact that Alberta has a campaign finance system more like most American states, and Minnesota and Wisconsin have laws more like most Canadian provinces.

A final question to be addressed here is related to outcomes. We are interested in the relationship between spending and winning. Table 3 shows the percentage of cases, in each province, in which the biggest spender won the election.

 

Table 3 Percent of Cases in which Winning Candidate Outspent Opponent
Province	All Races	Open Seats Only
Quebec	39.2%	41.9%
Saskatchewan	38.9	46.9
Ontario	54.6	40.0
Nova Scotia	57.7	33.4
Alberta	69.9	92.8
Manitoba	73.7	36.3

Data are presented for all elections, and for open seat contests only. In two instances (Quebec and Saskatchewan), the contestant who spends the most, lost most of the time. In two other cases (Nova Scotia and Ontario) the big spender was only slightly more likely to win than to lose. In the other two cases (Alberta and Manitoba) the largest spender was likely to win about 7 of 10 races. Most surprisingly, in open seat races (where incumbency is obviously not a factor), the biggest spender lost most of the time in all provinces except Quebec. Of course, since there are spending limits imposed in most provinces, the "biggest" spender usually did not spend significantly more than the other contestants. The implication is that spending limits have helped "level the playing field" in provincial elections.

Further Research Required

Much needs to be done before we can draw definitive conclusions about Canadian and American system of campaign finances. First, Canadian provincial elections are relatively volatile, and we do not want to rely too heavily on data from a single election period in six provinces. Campaign finance data from a larger sample of elections would be helpful. Further, a more sophisticated statistical treatment of these data is warranted. Specifically, a multi-variate technique will help clarify the relative importance of factors such as expenditures, incumbency, party affiliation, and cabinet positions in determining individual electoral outcomes.

Notes

Research reported herein was conducted with the support of a grant from the 1992-93 Faculty Research Grant program of the Canadian Embassy. I wish to acknowledge the data collection assistance of Jennifer Carrington, Danette Morgan, and especially Randi McDermott. I also wish to thank the Electoral Finance Officers from the provinces represented in this study, as well as the Ontario Commission on Election Finances, and Mary Dickerson and the staff of the Ontario Legislative Library for their assistance.

1. Alexander, Herbert. Reform and Reality: The Financing of State and Local Campaigns. New York: Twentieth Century Fund Press, 1991.

2. Seidle, F. Leslie. Provincial Party and Election Finance in Canada. (Volume 3 of the Royal Commission on Electoral Reform and Party Financing Research Studies). Toronto: Dunden Press, 1991. Also see: Ontario, Commission on Election Finances, A Comparative Survey of Election Finance Legislation, Toronto, 1988.

3. Gurwitt, Rob. "Searching for the Cheap Seats." Governing, 1992, p. 50.