The case for taking oil, gas (and heat) out of equalization
Non-renewable resource revenues fuel ongoing debate

By William Watson

After the recent spitting match on this page it’s probably foolish to take up the subject of equalization again. But here goes anyway.

Suppose the tax base in your province works out to $10,000 per citizen, while in mine it’s just $6,000. The federal equalization formula compensates my province for the difference. If the average rate of tax is 20%, Ottawa sends my province a cheque for 20% of the $4,000 difference, or $800 per person.

The Constitution says that’s fair. If both our provinces levied a tax rate of 20%, then without equalization your province’s revenues would be $2,000 per person and mine would be just $1,200. Our governments would not be able to provide reasonably comparable levels of public service at reasonably comparable tax rates, which is the principle the Constitution commits them to. At the identical tax rate of 20%, you get $800 more public services than I do.

Equalization may also be efficient. Say you and I make the same income. If income is a measure of productivity, then even though I have the same productivity as you where I am, I have a strong incentive to move to your province: Taxes are lower and public services better, or both. In fact, I have an incentive to move even if moving reduces my productivity, so long as the reduction is less than $800.

In actual practice, the equalization system is much more complicated. There are 33 tax bases in total, and the equalization my province receives is the sum of the 33 differences, some positive, some negative, between my province’s tax base and a five-province average base. (The five are the middling-rich provinces: not Alberta and not Atlantic Canada.)

The principle is simple, however: A province gets the tax revenue it would raise if it levied the national average tax rate on the shortfall in its tax base.

The question at issue in the recent debate is whether revenues from non-renewable natural resources (RNRNRs) should be included among the tax bases. They’re mainly taxes on oil and gas, and, in their various incarnations, currently account for 11 of the 33 bases.

Ken Boessenkool, who may be younger than most equalization specialists but is one of the country’s most productive and insightful policy wonks, argues that natural resource revenues shouldn’t be in the formula (see www.aims.ca).

His main point is that RNRNRs aren’t like other government revenues. They’re not a tax on an infinitely renewable activity, as, say, consumption taxes are. Rather, they’re the result of transforming a capital asset from one form into another — from hydrocarbons in the ground into cash in a heritage fund or, if the revenues are used to retire provincial debt, a better fiscal balance sheet.

From this perspective, Alberta is only better off by the interest on the heritage fund, or by the interest it saves by retiring debt, and its taxes will be lower or its public services higher only by that amount, so that’s the amount that should be equalized.

That makes sense. But do provincial governments with RNRNRs only invest them in capital projects of this sort? Or do they instead blow them on spending sprees that, until either the resource runs out or its price falls, allow them to cut taxes and/or raise services to levels quite different from what’s available in other provinces?

And if the principle is that asset transfers shouldn’t be equalized, shouldn’t we look at the other 22 tax bases to make sure none is really a similar capital transfer. Moreover, why is non-renewability such a dividing line? We Boomers are acutely aware that our human capital is a wasting asset.

It’s true we can replenish it by re- investing, but why is that key? Alberta could buy gas and oil and pump them back into the ground (even if the world as a whole can’t replenish the global stock of these fuels).

A second argument for not fully equalizing RNRNRs is that they end up being capitalized in local land and house prices and therefore eventually show up elsewhere in the equalization formula — which means they’re counted twice.

Higher local prices also discourage people from moving inefficiently to an RNRNR-rich province, so equalization isn’t needed. Again, that’s true in principle. But as a factual matter, is oil and gas wealth 100% capitalized in local assets?

If the answer is yes, that weakens the third main argument for removing resources from the equalization formula, namely, that because provinces lose a dollar of equalization for every dollar of RNRNR they pick up, they have no incentive to develop their resources. With capitalization they do get a revenue boost. Of course, if it turns out the provinces are mainly blowing their resource endowments on current consumption, we probably want high taxes on development.

Ken Boessenkool may well be right on RNRNRs, but the debate isn’t over yet.

William Watson, editor of Policy Options, the magazine of the Institute for Research on Public Policy, teaches economics at McGill University.