I'm spending today writing a review of Caren Grown and Imraan Valodia's new book Taxation and Gender Equity.The review is for the journal Feminist Economics, but I'll give you an uncensored sneak preview of the good bits here.
Vertical equity is a standard concept in taxation: people with a greater ability to pay taxes should do so, both in absolute terms and as a percentage of their income. Dispensing entirely with careful conceptual understandings, let's just say a tax system promotes gender equity if it's good for women.
Gender equity can conflict with vertical equity. India provides perhaps the most interesting example. Since 2005, women have had a higher personal income tax exemption than men. Men pay no tax on the first Rs. 150,000 they earn; women can earn up to Rs. 180,000 before paying tax (2008/9 tax brackets, $1US=40.24 Rupees). Unfortunately, “the number of women within the income tax net is a miniscule proportion of the total number of income taxpayers [3 percent] and an even more miniscule proportion of the total number of adult women in India [0.1 percent]” (the paper can be downloaded here). Even if this provision has some success in enhancing gender equity, the women who gain will be among the most privileged in Indian society, suggesting that this attempt to support gender equity compromises vertical equity.
Argentina gives a more complex example of the interaction between gender and vertical equity. Under its income tax act income from any marital property, that is, assets owned by the couple, shall be “totally allocated to the husband” unless certain conditions are fulfilled, for example, in specific cases where the asset was declared to be the wife’s separate property at time of purchase. In Canada such a law would be declared unconstitutional, as it differentiates amongst taxpayers on the basis of gender.
Yet my guess is that the primary intent of the law was not to discriminate against women. Many countries with individually based tax systems, including Canada, have some kind of "attribution rules" to prevent income splitting. These rules prevent from couples reducing their total tax burden by transferring income to the partner with the lower marginal tax rate. The rules are usually justified on equity grounds. In Argentina, as in Canada, a relatively small proportion of the population have any significant amount of property income. Forced income aggregation in the hands of men increases the tax liabilities of a relatively wealthy group, thus enhances vertical equity -- even as it discriminates on the basis of gender.
And here's another complication: it's not obvious that this rule will always be bad for women. An Argentinian couple can still achieve some measure of income splitting, if assets are declared to be the wife's property, and not marital property. By encouraging female property ownership, the overtly discriminatory Argentinian tax legislation could conceivably be good for women, thus enhancing gender equity.
Indeed, gender equity and vertical equity often go hand in hand, because women are disproportionately represented among the poor. The Grown and Valodia volume has a really neat example about the taxation of salt in Uganda, but it's a bit complicated to explain here. Instead, I'll give you a simple example from Canada Revenue Agency's Taxation Statistics. In 2008, women filed 51 percent of all income tax returns, earned 38 percent of all reported employment income, and paid 33 percent of all taxes. Canada's personal income tax system is progressive, with higher income earners paying higher rates of tax. Since men, on average, have higher incomes than women, they pay more tax than women - vertical equity and gender equity working together.
Perhaps this is one reason why reduced taxes don't cut it with women voters - paying fewer taxes in the first place, they have less to gain from tax cuts. And since there are more women than men receiving old age security and working in government-funded sectors such as health care and education, women have more to lose from cuts in government spending.