Thank you, Mr. Chair.
Thanks to the committee for inviting me to speak with you today and actually, more generally, for conducting a comprehensive study of tax issues in Canada. I think it's a good moment for this in Canada. It's necessary to do these things on a regular basis. The world changes. We need to rethink our tax system on a regular basis.
As members of the committee are undoubtedly aware, tax policy has important implications for the quality of life that Canadians enjoy, both because of its effects on the efficiency and competitiveness of the Canadian economy, but also because of its impact on the distribution of economic resources in Canada.
In my comments today I'd like to address two sets of issues, considering first the goals or objectives of the tax system, especially at the federal level in Canada, and second, some of the implications of these goals for the kinds of taxes that the federal government should collect and the specific design of these taxes.
First of all, I'll consider the goals of taxation. The preamble, of course, to the mandate of the committee rightly identifies as an important goal “to collect sufficient revenue to provide required services in the least costly manner”. Other things being equal, of course, a tax system should be as efficient as possible, so as not to distort market behaviour in an inefficient manner; it should be as simple as possible, so as not to consume resources unnecessarily in administration and compliance; and of course, it should be as competitive as possible, so as to encourage economic activity in Canada and not to encourage the relocation of economic activities to other jurisdictions.
That said, we also know that a tax system must be fair and must be perceived to be fair. It's important to remember that one of the most efficient and least costly kinds of taxes to collect is a poll tax. It would impose the same amount of tax on each individual, regardless of their economic behaviour. It's also instructive to note that the attempt to introduce such a tax in the United Kingdom led directly to the downfall of Margaret Thatcher. As a result, tax scholars typically suggest that a tax system should seek to collect revenues in a manner that is fair or equitable as well as efficient, simple, and competitive.
In addition to revenue raising, moreover, a tax system has two other important goals, which tax scholars generally refer to as the so-called “allocation function” and the so-called “distribution function”. Consistent with the allocation function, certain taxes should be designed not only to raise revenue to finance public goods and services, but also to correct for market prices where these do not reflect the true social costs to produce a given good or service. A classic example of a corrective or so-called Pigouvian tax—this is how economists refer to it—along these lines is an environmental tax such as a carbon tax, which would correct for market transactions that do not currently take into account the environmental costs from carbon emissions. Although such a tax would inevitably raise revenues—and these revenues, of course, could be used to reduce other taxes or public debt or to finance public goods and services—the primary purpose of a carbon tax, or an environmental tax, is not actually to raise revenue, but rather to correct for the market failure that results from not putting a price on the environmental harm.
In addition to the allocation function, taxes can also be used, and are used, to redistribute economic resources, moderating inequalities in economic outcomes that result from market transactions as well as from transfers of property from one generation to the next. Although some tax scholars suggest that this distribution function is best accomplished on the expenditure side after raising revenues in the most efficient manner, others, myself included, contend that distributive fairness is best accomplished not only by a so-called “end state approach” that looks at the ultimate result and redistributes towards those who might be most needy, but also through a process-based approach that defines individual entitlements to income, as well as gifts and inheritances, in terms of after-tax amounts that moderate extreme inequalities that might otherwise prevail.
What implications then do these three goals—efficient, simple, and competitive revenue collection; market correction; and redistribution—have for the design of the tax system, particularly at the federal level?
Beginning with revenue collection, of course, experience demonstrates that one of the most efficient, simple, and competitive kinds of taxes that most countries do impose, and can impose, for revenue collection is a value-added tax like the GST. These taxes are levied by almost all developed countries around the world, generally to a larger extent than Canada, with the exception, of course, of the United States. They are effective taxes for raising revenues. They have very few implications for competitiveness if, as in Canada, they're imposed on a destination basis rather than an origin basis, so that you tax imports and you strip away the tax on exports.
As such, I think it's unfortunate that the federal government chose to lower the rate of GST while initially increasing, and then subsequently leaving untouched, personal income tax rates, particularly of course since the leverage from the reduction in federal rates could have been used more effectively to encourage provinces that have yet to harmonize their retail sales taxes with the GST.
I think there are lots of good arguments that the provinces should abandon their retail sales taxes where they exist. There's this full story of cascading effects on businesses that purchase business inputs and don't get the credit that the GST would provide. And something has to be done to encourage the provinces that haven't harmonized to harmonize. It's now much more difficult, unfortunately, to do that.
Turning to the allocation function of the tax system, I endorse the arguments, first of all, that Professor Boadway made, but also those that I notice Professors Kesselman and Davies made earlier to a session of the committee, that Canada should introduce a carbon tax in order to put a price on carbon emissions. Now, such a tax can raise competitiveness concerns. That's one of the biggest concerns about introducing a carbon tax. On the other hand, it's quite likely that the U.S. is going to be moving in this direction in the near term.
Furthermore, there are ways to try to design a carbon tax that would take into account the competitiveness concerns, generally by imposing the tax on a destination basis like the GST, imposing a tariff on imports based on the carbon content, and unfortunately--but I think this is probably necessary, it's a short-term step--trying to strip away the carbon tax on exports. Otherwise, of course, you simply encourage the flight of those industries to countries that don't levy the carbon tax themselves. So you could try to levy a carbon tax on a destination basis, like the GST, and address a lot of the competitiveness concerns.
Finally, with respect to the distribution function, I believe this is best accomplished by moderately progressive income taxes, like the current federal income tax, that moderate the results of market returns and, I believe, by a progressive gift and inheritance tax that would moderate the inequalities in wealth and the opportunities that wealth generates resulting from the transfer of wealth from one generation to the next.
Although the federal government currently levies the progressive income tax at rates that are not significantly out of line with comparator jurisdictions like the U.S., which will probably be increasing income tax rates after the next presidential election, it does not tax the transfer of wealth from one generation to the next. It hasn't done so since 1972, except to the extent that it taxes capital gains through a deemed realization at death. My view is that that's not an adequate substitute for a comprehensive gift and inheritance tax.
As well, I believe a number of changes over the last several years have caused the federal income tax to increasingly take the form of what scholars call a personal consumption tax, through increases to the capital gains exemption, reduction in the capital gains inclusion rate, increases in RRSP contribution limits, and now the introduction of tax-free savings plans. This kind of personal consumption tax effectively exempts income from savings from tax.
Economists often tout the alleged efficiency of personal consumption taxes over personal income taxes, but I believe the sufficiency case is seriously overstated. Experience suggests that, for the most part, savings are responsive to changes in income levels rather than changes in the return from savings. In fact, you see that in evidence about who contributes to RRSPs. Lots of low-income people with potential contribution room don't contribute to RRSPs, simply because they don't have the ability to contribute. I think the movement in this direction doesn't necessarily increase savings, but rather shifts savings from a taxed form to an untaxed form.
And I think that's going to be the case with tax-free savings plans, which by the way creates an amazing intergenerational transfer of wealth issue. You start transferring $5,000 a year to kids once they turn the age of 18, and if you contribute for 10 years to your kid--a colleague of mine did some numbers with standard rates of return--by the time they're 65 they could have $1.5 million that is not subject to tax. That's just through the rate of return on the tax-free savings plans.
So the conclusion on this, then, is that the shift toward a personal consumption tax not only is not necessarily required by efficiency considerations, but has significant implications for the fairness and redistributive function of the tax system, and I think it is a disappointing direction that we've headed in.