Indirect Taxation and Progressivity: Revenue and Welfare Changes

When debating an existing or proposed indirect tax, particular emphasis is often given to the disproportional effect of the tax. This effect is intended to show how the welfare loss from the tax is distributed across households with various income levels. Typically, the disproportional effect is assessed using a tax-revenue-based measure that computes how the amount of tax paid, expressed as a fraction of household expenditure, varies as household expenditure increases. However, the amount of tax paid by a household does not always form a good measure of the welfare loss experienced by the household. Hence a more appropriate assessment of the disproportional effect of a tax is a welfare-based measure that computes the variation in the welfare loss (from the tax) as a fraction of household expenditure, as household expenditure increases. This paper shows that when the households' preferences, in respect to their choice of goods, are allowed to vary among households, tax revenues provide a poor measure of disproportionality, producing results that directly conflict with those produced from a welfarebased measure. The New Zealand indirect tax system is used to show that such cases of conflict are frequent, reinforcing the importance of using a welfare-based measure when assessing the disproportional effects of an indirect tax.