Skip to main content

The Globe and Mail

When it comes to policy, the median is the message

Any progressive income redistribution policy is inspired by Robin Hood:

a) Take from the rich, and

b) Give to the poor.

Story continues below advertisement

In an earlier post, I noted that the economics of the labour market for high earners makes part a) of this agenda surprisingly hard to implement. The economics of part b) is trivial -- sending cheques to those with low incomes isn't complicated -- but it would seem that politics also makes it harder than one might expect.

As electoral politics have become more sophisticated, political parties have become better able to identify segments of the population that are "in play", and the median voter is a coveted target. This may explain why the share of government transfers received by the lowest income quintile has declined over the past 30 years, in favour of income groups in the middle of the distribution. (See here for more on this point.)

The most recent example of this trend is the enthusiasm on the part of parties across the political spectrum for reducing or eliminating the GST/HST on certain hot-button items. These measures are sometimes presented as progressive measures, but they aren't. As Kevin Milligan pointed out in the context of gasoline, reductions in the GST are a gift to those with high incomes.

This is a general point. The biggest current exemption to the GST is the zero-rating on food expenditure; the Department of Finance's estimate for the tax revenues lost to this measure in 2010 -- Table 3 here -- is $3.34-billion. The zero-rating of essentials such as food enjoys broad support in progressive circles, but it can hardly be considered to be a progressive measure.

According to the 2009 Survey of Household Spending, average household expenditures on food purchased at stores was $5,658 (food purchased at restaurants is subject to the GST). Unsurprisingly, households with higher incomes spent more; see the accompanying graph.

Households in the top 20 per cent of the income distribution that earned more than $108,000 spent more on GST-exempt food than the 40 per cent of households with incomes of $48,600 or less. The benefits from zero-rating food are proportional to expenditures, so the biggest winners are at the top end of the income distribution.

It is at this point that defenders of the current policy invariably point out that as a percentage of income, the biggest winners are low-income households. But those who think that this is a convincing argument in favour of zero-rating food must necessarily be comfortable with arguing that high-income households somehow deserve a greater share of public resources, for no better reason than because high-income households earn more money. (Suffice it to say that I do not find this to be a convincing counter-argument.)

Story continues below advertisement

Removing the GST exemption on food would produce an extra $3-billion for the federal government, and only a fraction of those revenues would be necessary to supplement the GST tax credit in order to compensate low-income households. Indeed, the GST tax credit could be considerably expanded so that those with low incomes are unambiguously better off.

But this will never happen. No political party will propose this sort of transfer from rich to poor, because it might irritate the median voter. And no party will risk losing the support of the median voter.

Follow Economy Lab on twitter

Stephen Gordon's recent posts and Twitterfeed can be viewed here.

Report an error
About the Author

Stephen Gordon is a professor of economics at Laval University in Quebec City and a fellow of the Centre interuniversitaire sur le risque, les politiques économiques et l'emploi (CIRPÉE). He also maintains the economics blog Worthwhile Canadian Initiative. More

Comments are closed

We have closed comments on this story for legal reasons. For more information on our commenting policies and how our community-based moderation works, please read our Community Guidelines and our Terms and Conditions.