Real Versus Notional Income Splitting: What Canada Should Learn from the US 'Innocent Spouse' Problem
Canada’s individual tax unit historically has allowed a spouse to report only income over which she or he has legal control. Income splitting therefore typically requires transferring ownership of income earning property in some fashion. I refer to this as ‘real’ income splitting to contrast it with the purely notional assignment of income to a spouse’s tax return, first introduced in 2007 for pension income only. This ‘notional’ income splitting allowed spouses for the first time to shift income between their individual tax returns to achieve a lower tax rate, without any obligation to share the underlying pension entitlement. The federal government has promised to expand notional income splitting to benefit all couples with dependent children once the federal deficit is eliminated (currently projected for 2015-16). This paper challenges advocates of income splitting to address its differential impact on men and women and argues that notional income splitting worsens gender economic inequality. It highlights the egregious risk that under notional income splitting some individuals, overwhelmingly women, will be targeted for enforcement actions to collect unpaid taxes on income they have never received or controlled. Known in the U.S. as the “innocent spouse” problem it is the most acute, but not the only way in which women are systematically disadvantaged by notional income splitting. Of broader concern are the economic incentives created by notional income splitting for second earners to be financial dependents, rather than income earners or property holders in their own right. The paper urges Canadian policy makers, if they are intent on liberalizing spousal income splitting, to condition it upon a transfer of legal control over income or property to the lower-earning spouse.