Campaign Finances and Provincial Elections in Ontario, Canada

ABSTRACT

This study analyses the relationship between provincial election outcomes and campaign spending. This study is novel as it utilizes only those expenses that should have a causal link with election success. OLS regression controlled for candidate quality, incumbency, and economic conditions reveals a positive (negative) relationship between campaign (challenger) spending and election succession, while logistic regression results in a 93% correct prediction rate. Two-staged least-square regression corroborates the findings. The results suggest that although campaign spending is useful, incumbency status and experience are more important. Additionally, campaign spending is less important during close elections and has a diminishing marginal return.

KEYWORDS:

• Provincial elections
• campaign finances
• candidate income statement

Acknowledgments

The authors would like to acknowledge the insights provided by Crystal Caputo (chief financial officer for the Honourable Michael Gravelle, Thunder Bay-Superior North, Ontario, Liberal Party). In addition, the authors would like to acknowledge the contributions of Luciana Cavalcanti, Gustavo PaivaNogueira, and Nathalia DaSilveira Lima who provided valuable research assistance.

Notes

¹ Gerber (2004) provides a more robust discussion of the methodological issues and alternatives.

² The panel data methodology is not utilized as there would not be a sufficient number of observations of repeat elections.

³ For example, a Liberal party candidate’s challenger spending is the sum of the campaign spending incurred by the PC and NDP candidate’s in the same riding association.

⁴ The other expense category includes Appreciation Notices, Bank Charges, Candidate’s Personal Expense, Candidate’s Child Care Expense, Credit Card Fee, Furniture and Equipment, Insurance and Utilities, Interest, Social Functions, and Web and Internet.