Tax Design for a Northern Tiger

Jonathan R. Kesselman March 10th, 2004

Canada’s tax system, if properly designed, could play a significant role in making the country’s economy more competitive and vibrant, and thereby contribute to Canada’s emergence as a “Northern Tiger.” This paper addresses the broad archi- tecture of tax policy and offers concrete recommenda- tions for policy action. Drawing on a large body of public finance and tax economics research, the propos- als in this paper build on recent tax developments at both the federal and provincial levels.

Designs for an improved Canadian tax system should keep several goals in mind. They should create a com- petitive advantage for Canada within an increasingly integrated North American economy. They should mini- mize impediments to specialization, investment and trade, and reduce the effects of the border with the United States. They should create unique attractions to invest and to conduct business on the Canadian side of the border. And they should serve to nurture and retain skilled and talented workers in this country.

If the mix and structure of taxes are efficiently designed, the level of taxes in Canada will not hinder competitiveness. How much of national output to devote to public versus private consumption can then be decided without impeding economic performance. Yet considerations of business investment, skilled work- ers and trade flows do constrain the design of Canadian taxes relative to those in the US and elsewhere.

One way to promote efficiency and growth is to move toward taxes based on consumption, rather than on income. The priority should be to reduce taxes on personal savings, capital income and business investment without forgoing too much revenue or creating windfall gains for wealth holders. Consumption could be taxed through labour income and reformed tax bases for business. Moreover, a consumption base could be applied to direct personal and business taxes, and indirect sales taxes could be converted into more direct forms to reduce trade frictions.

For personal taxes, the highest priority is to shift further toward a consumption base, preferably by allowing tax- prepaid savings accounts alongside existing tax-deferred accounts and by removing the dollar ceiling on contribu- tions to these. Other pressing needs for personal taxation reform include broadening the base, widening tax brackets (especially at the top) and implementing modest cuts in tax rates for the middle-income brackets of federal tax and the upper-income brackets of some provinces’ taxes.

With respect to payroll taxes, employment insurance (EI) premiums should be set at rates that finance actual program costs, and employers’ premiums should reflect each firm’s EI costs. A federal general payroll tax could be instituted to raise the revenues to finance other tax cuts and reforms; part of the financing of the Canada Pension Plan could also be shifted to such a payroll tax.

Either the federal or provincial sales tax should be con- verted into a more direct form (such as a business transfer tax) in order to reduce border costs. Alternatively, the goods and services tax (GST) should be simplified and reformed, and provincial sales taxes modified to reduce the burden on business inputs or harmonized with the GST. Excise taxes on gasoline offer an economically and environmen- tally attractive way to obtain more revenues in order to lessen the reliance on other, more distorting tax bases.

The most vital improvement in Canada’s tax design would be to reduce the burden of business taxes on investment. At the federal level, this should be achieved by eliminating capital taxes, making further small cuts in statutory corporate income tax rates and sharply accelerating depreciation provisions (and possibly shift- ing to a cash flow base). At the provincial level, the remaining corporate capital taxes should be phased out and corporate income tax rates reduced; alternatively, corporate and other taxes could be replaced with a busi- ness transfer tax.

Improved tax design would strengthen the economy, raise Canadians’ real, long-run living standards and increase the country’s policy autonomy. Given the vision and political will to create a world-beating tax system, Canada could become a uniquely attractive place for eco- nomic activity – a Northern Tiger.

At this stage, we think the emphasis in pub- lic debate on taxation needs to shift from tax reduction to tax reform. Our most serious competitive problem as a country no longer flows predominantly from the overall size of our tax burden, but from what we tax most heavily…What Canada must consider now is a fundamental shift of its tax structure from an income to a consumption base. (d’Aquino and Stewart-Patterson 2001, 307)

Introduction

Recent federal and provincial budgets are remarkable for how well they have heeded the prescriptions of tax economists in academia, business and think tanks. Rarely have Canadian gov- ernments responded so faithfully to formal analysis of the need for a competitive tax system. Most ana- lysts would say, however, that these moves are only the first instalment of needed tax policy changes. Even the finance minister, just days after the 2003 federal budget, announced that further personal tax cuts were a high priority.1 So how can future tax changes in Canada contribute to a more competitive and vibrant economy? What is the tax policy agenda for making Canada a “Northern Tiger”?

Economists assessing the tax requisites for a more productive and faster-growing Canadian economy increasingly favour greater reliance on consumption rather than on income bases.2 Some suggest that this goal be pursued via greater reliance on indirect forms of consumption taxation such as the goods and serv- ices tax (GST) or sales taxes.3 There have also been complaints, emanating more from business than from academic quarters, that Canada’s overall tax burden remains too high relative to that of its main trading partner, the United States.4 Yet, despite their frequent reference to the need to be “competitive” with the US economy, these analyses have not considered all the crucial dimensions of tax policy for Canada as an open economy with a dominant trading partner. 

To become more competitive, Canada must pursue taxation and related public policies that will mini- mize impediments to specialization, investment and trade. Tax policies should complement other policies in pursuing further economic integration with the US in ways that reduce the effect of the border between the two countries.5 The parallel challenge is to create a competitive advantage for Canada within an increasingly integrated North American economy. Reducing border frictions is important, but so is cre- ating unique attractions to situate and conduct busi- ness on the Canadian side and for skilled and talented people to work here. Tax policies can play a central role in this broader policy strategy, but the goal must not be simply to create a lower tax envi- ronment. More critical is to fashion a tax system that can efficiently generate the revenues to support the public services and infrastructure that will attract high-value-added business and workers and that will increase living standards for all Canadians.

This paper addresses the broad architecture of tax policy and offers concrete recommendations for future tax policy initiatives as part of a strategy to make the Canadian economy more competitive.6 I begin by reviewing recent tax policy achievements of the federal and provincial governments and assessing whether Canada’s overall level of taxation is an impediment to a competitive economy. I then examine the comparative tax mixes of Canada and the US and how far Canada needs to go to compete with US tax levels. Next, I examine the key econo- mic issues in the design of tax policies: determining the best base for taxation, how best to implement the desired tax base and whether to rely more on direct or indirect taxes. Then, I apply the analysis to the major types of taxes (personal, payroll, sales and excise and business taxes) at both the federal and provincial levels, and I suggest directions for reform for each of the main tax areas. I conclude with thoughts on the tax strategy that Canada should pursue to become a Northern Tiger.

Tax Policy Achievements

The tax cuts and reforms that Canadian govern- ments have undertaken since 1998 should be seen in the context of the string of tax hikes that had taken place over the previous dozen years. At the federal level, these hikes included:

  • a series of increases in the rate and coverage of the (pre-GST) manufacturers’ sales tax through the 1980s, with a final rate of 13.5 percent;

  • repeated freezes on scheduled rises in contribution limits for registered pension plans (RPPs) and regis- tered retirement savings plans (RRSPs) through the 1990s (thus reducing their real value);

  • a high-income personal surtax and a corporate income surtax (1985);

  • a general personal income surtax (1986);

  • a cap on indexation of personal taxes to inflation above 3 percent (1986), which meant no changes in tax brackets for most of the 1990s and thus a stea- dily rising tax bite;

  • a “temporary” capital tax on financial institutions (1986);

  • a capital tax on large nonfinancial corporations (1989);

  • an income test on the age credit for personal tax (1995);

  • an increase in the corporate income surtax (1995);

  • increased gasoline excise taxes (1995);

  • a reduced age limit for mandatory disbursements from tax-deferred savings plans (1996); and

  • tightened tax provisions for unincorporated and corporate business through the 1990s.

The provinces undertook many parallel tax hikes over this period, and several instituted new corporate capital taxes (including Alberta’s introduction of a financial institutions capital tax in 1990).

Most of these tax increases were driven purely by the revenue needs required to address uncontrolled deficits and, as a result, often neglected concerns for economic efficiency and long-run growth. Political considerations caused many of the tax hikes to be targeted dispropor- tionately at higher-earning individual taxpayers and at business and corporate taxpayers — though even mod- erate earners were not entirely spared. When fiscal room began to emerge in the late 1990s, the first tax cuts focused on low- and moderate-income individuals and on relief for special groups such as the disabled, stu- dents and poor families with children.7 In recent years, more fiscal room has opened up, and tax policy has begun to pay proper attention to the neglected criteria — economic efficiency and long-run growth. These objectives remain paramount in designing a tax system for Canada as a nascent Northern Tiger.

Recent Federal Tax Reforms

The 2003 federal budget continued important tax policy thrusts introduced in federal and provincial government budgets since the late 1990s.8 Essentially, these poli- cies have reduced tax burdens on capital and capital income and placed greater reliance on labour income, consumption bases and user fees. Ottawa has pursued these changes through both personal and corporate direct taxes and by moving the revenue mix from direct taxes toward indirect and payroll taxes. (As this paper shows later, taxing labour income is one way to implement a consumption tax.) These changes are consistent with the prescriptions of formal public finance analysis for making the tax system more eco- nomically efficient and growth oriented.

The fall 2000 budget update is often cited as the centrepiece of federal tax cuts. That document, how- ever, represents just one step in a tax strategy that began in 1998, for personal taxes, with the elimina- tion of the general and high-income surtaxes, fol- lowed by cuts in the rates for all incomes, the expansion of rate brackets and the restoration of full indexation. In 2000, the personal “income” tax was further shifted toward a consumption base by two cuts in the capital gains inclusion rate and, in 2003, by increased limits for contributions to tax-deferred savings plans (RRSPs and RPPs). The personal tax will move further toward a labour income base if the 2003 budget’s plans to consider tax-prepaid savings plans result in policy action.

The other major strand of federal tax policy has been to reduce the tax burden on corporate income and capital and to provide more uniform treatment across industries. Central to this change was the phased reduction of the federal basic corporate income tax rate for nonmanufacturing sectors to 21 percent, the same rate as for manufacturing firms. As a result of the 2003 budget, this corporate tax rate cut will be gradually extended to the resources sector over the next several years, the small business deduc- tion for corporate tax will be raised over time from the current $200,000 to $300,000, following initia- tives at the provincial level, and the federal capital tax on nonfinancial corporations will be phased out over several years, again following the actions of some provinces. These tax changes serve to reinforce the overall move away from capital income taxation.

Further reinforcing the shift of the overall base of the tax system toward labour income was the federal government’s decision to trim employment insurance (EI) premium rates so slowly as to generate ongoing large surpluses relative to program costs. These rev- enues have provided the fiscal means for larger per- sonal and corporate rate cuts. Sharp phased hikes in premium rates for the Canada Pension Plan (CPP) from 1998 to 2003 have also increased the revenue shift toward labour income bases. Additional rev- enues to support the overall tax strategy have come from such policies as the 2001 budget’s hike in toba- cco taxes and its imposition of the Air Travellers Security Charge (a user fee, which was reduced in the 2003 budget). Over the same period, moreover, the rate and coverage of the GST have remained unchanged, thus raising the relative revenue share of indirect consumption taxes.

Recent Provincial Tax Reforms

Provincial tax changes since the late 1990s have pur- sued the same general goals as federal tax policies. Like the federal government, the provinces are shifting their overall revenue mix away from capital and capi- tal income and toward labour income; in fact, they have led in important respects. Alberta is reducing its general corporate income tax rates, eventually to 8 percent. Ontario under the Progressive Conservatives was on a similar course, but its new Liberal govern- ment is backtracking — raising the corporate tax rate from 12.5 percent in 2003 to 14 percent in 2004 (instead of the 11 percent the previous government had planned). British Columbia has pledged to keep its gen- eral corporate rate competitive with that of the other major provinces, and has begun reductions accor- dingly. Quebec, whose corporate tax rate of 9.3 percent is the lowest of any province, has initiated sharp cuts in its generous tax expenditures for business.9 Several provinces have also reduced their corporate tax rates for small businesses. All provinces from Ontario west- ward already have small business deductions for cor- porate tax of at least $300,000 (the target now for federal policy); in the 2003 Ontario budget, the Eves Conservatives set a new target of $400,000, which the new government will also pursue.

Some provinces have eliminated their corporate cap- ital taxes — Alberta abolished its tax on financial cor- porations in 2001 and British Columbia removed its general corporate capital tax in 2002. The 2003 Ontario budget pledged to eliminate, concurrently with similar action by Ottawa, the province’s nonfinancial corporate capital tax, but the new Liberal government has with- drawn that commitment. Quebec announced a multi- year plan to halve rates on both of its corporate capital taxes, but its new Liberal government is facing fiscal stringencies that have caused it to delay these cuts.

The four provinces with employer payroll taxes (includ- ing Ontario and Quebec) have left them intact, so that this source is generating a rising share of total provincial tax revenues. Some provinces have also shifted their revenues toward greater reliance on sales taxes and targeted them more directly at consumption. For example, British Columbia followed its large personal tax cuts in 2001 with a hike in its sales tax rate in 2002, and it reduced the taxa- tion of business inputs. Saskatchewan paired its personal tax rate cuts with a major expansion in the coverage of its sales tax, but it included additional business inputs as well as many consumer items. The provinces have also increased their revenues from fees and user charges — hikes in medicare premiums by both Alberta and British Columbia are examples.

In 2001 the federal government allowed the provinces to shift from a “tax-on-tax” system to a “tax-on-income” system for their personal taxes.10 Most provinces responded by reducing their tax rates and flattening their rate structures. Most notably, Alberta moved to a flat rate of 10 percent above a much higher taxable income threshold. Saskatchewan and British Columbia have significantly lowered their top marginal rates of personal tax and partially flat- tened their rate schedules. Ontario has cut personal taxes sharply at low and middle incomes but contin- ues to impose a steep surtax that keeps marginal rates on upper incomes high. The Conservative government had promised to reduce the number of taxpayers sub- ject to surtax and had even referred to its ultimate abolition, but the new Liberal government has said it will not proceed with these items. Quebec has reduced personal tax rates but increased their progressive tilt. The province’s new Liberal government entered office committed to a further reduction of personal taxes by 27 percent, or $5 billion annually, over the next five years and to having personal tax rates competitive with those in Ontario within ten years (Parti Libéral du Québec 2003). As with corporate taxes, however, Quebec’s fiscal strains are delaying the start of these cuts as well, and Ontario’s change of government may make the goal itself less ambitious.

Since the provinces, apart from Quebec, use the fed- eral definition of taxable income, they have also implic- itly adopted the federal shift of the personal tax base toward consumption and labour income. In fact, Ontario was planning to reduce the capital gains inclusion rate for provincial tax purposes to 50 percent in 2000, and was negotiating with Ottawa over whether this would be permitted within the tax collection agreement, but this was pre-empted by the federal change. In 2000, Quebec also mirrored in its provincial income tax rules the fed- eral reduction of capital gains inclusion rates.

Taxes and Competitiveness

To become a Northern Tiger, Canada requires a tax system that enhances its economy’s competitive- ness, especially vis-à-vis that of the US. To deter- mine the kind of tax system that will achieve this end, however, several questions must be addressed. First, does Canada’s overall level of taxes pose a handicap to the economy’s competitiveness? Second, how does Canada’s tax mix compare with that of the US? (I examine the comparative structures of each type of tax later in the paper.) And third, does it make sense for Canada to compete with the generally lower tax rates found in the US?

Tax Burdens and Comparative Advantage

Some business and academic analysts have asserted that Canada’s higher overall tax burden relative to that of the US represents a competitive disadvantage. In assessing this claim, it is important to distinguish between the effects of taxation on real standards of liv- ing and on international competitiveness. Moreover, it is essential to include in real living standards both pri- vate consumption (after-tax and after-transfer real incomes) and public consumption (the value placed on publicly supplied goods and services). If Canada’s rela- tively higher tax burden is borne fully by households in the form of lower net incomes, then that burden affects neither businesses’ costs of production nor their competitive position internationally.11 Whether this sce- nario reduces or raises real living standards hinges on whether Canadians value the incremental increase in publicly supplied goods and services at less than or more than their loss of real disposable income.

If part of the higher tax burden is borne by busi- nesses, it still will not reduce their international competi- tive position so long as there are additional public services or facilities (such as highways, medicare or bet- ter courts) that reduce their production costs or raise their productivity. Higher taxes could impair Canada’s trade competitiveness only if those taxes were borne in part by businesses and did not carry offsetting business benefits. Examples include pork-barrel regional spend- ing, servicing public debt incurred for noncapital pur- poses or redistributive programs. Under this scenario, production costs would rise for Canadian producers, making them less competitive with US and other foreign producers, at least initially.

A tax-induced increase in Canadian production costs is not, however, the end of the story. The resulting decline in the current account balance would cause the Canadian dollar to depreciate, thus restoring the competitive position of Canadian businesses.12 According to standard trade theory, a country’s ability to gain from specialization and trade hinges on its comparative advantage in producing some products, not on its absolute productivity or lower production costs vis-à-vis those of its trading partners. Raising taxes in Canada above those in the US, even for “wasteful” purposes, should not harm exports. True, it would reduce Canadians’ real living standards (through lower net incomes, higher import prices and other means), but it would not affect the economy’s inter- national competitiveness.

Although being an open economy per se should not affect Canada’s choice of tax level, integrated Canadian-US markets for capital and skilled labour and the high potential mobility of these factors of production will influence Canada’s optimal tax design. Canadian tax bases, structures and rates must be tailored to maximize the growth of investment, jobs, productivity and real wages. Particular care must be taken with respect to the taxation of capital income and the personal income of higher-skilled labour in Canada and to the efficiency costs of the total revenue system. As I show later, this point has important implications for the design of business taxes and for the progressivity of personal taxes.

The bottom line to this analysis is that Canada’s choice of tax level can be purely a matter of domestic priorities — whether the value of public services out- weighs the cost of the associated taxes — not a matter of trade competitiveness.13 If part of Canada’s public spending is misdirected or wasteful, it should be cur- tailed and taxes reduced solely because of the harm this does to real living standards (private plus public consumption). Still, the chosen level of spending should be financed at the lowest efficiency cost. Thus, Canada should formulate its tax design to be more efficient than that of the US. In so doing, Canada can pursue the goal of a stronger and more competitive economy even if it chooses to devote more of its national output to public consumption through higher taxes than does the US.

Canada-US Tax Comparisons

What is the relative overall level of taxes in Canada and the United States, and what are the comparative mixes of taxes in the two countries? Relative to its gross domestic product (GDP), Canada’s total tax bur- den was about 6 percentage points higher than that of the US in 2001, down from more than 9 percentage points in 1990 (see table 1).14 Significantly, these fig- ures do not include “nontax revenues” of governments — items such as fines, user charges, most licence fees and royalty payments to government for the right to extract natural resources.15 Such royalties are more important in Canada, particularly in the resource- oriented western and Atlantic provinces, than they are in the US, and are an economically efficient way for Canadian governments to obtain revenues that allow them to reduce their need to apply distorting taxes.

More economically important is the mix of taxes the two countries use to finance government (table 1). In some basic respects, the two countries have quite simi- lar tax mixes. Both rely heavily on personal taxes. As well, in both countries personal plus corporate income tax revenues jointly account for about half of total rev- enues, and property taxes account for about 10 percent of tax revenues. The most notable divergence is that Canada uses taxes on goods and services much more heavily than does the US, which, in turn, uses payroll taxes for social security much more heavily than does Canada. However, adding payroll taxes and goods and services taxes (both of which are ways to move toward taxing consumption) reveals that in 2001 each country collected about 40 percent of its revenues from these taxes jointly.

At a more detailed level, some differences between the two countries’ tax systems are noteworthy. At the federal level, personal taxes in Canada are signifi- cantly smaller than they are in the US (even as a per- cent of GDP). This is, however, offset by the personal taxes that Canadian provinces impose, which are much higher than those levied by US states and cities. Unlike in Canada, in the US cities and counties commonly impose sales taxes on top of state sales taxes, and some US cities also have a personal income tax. At the federal level, corporate income taxes were roughly comparable in the two countries in 2001, but at the subnational level they were about three times as large in Canada as in the US.

The US has no broad-based federal tax on goods and services like Canada’s GST. At the subnational level, both countries apply sales-type taxes (whether on value added or retail sales) to roughly the same degree. But because Canada also has a federal sales tax, its total relative reliance on broad-based taxation of goods and services is nearly twice that of the US. At the federal level, both countries apply comparable excise and use taxes, but at the subnational level, Canadian governments apply such taxes more heavily than do US governments. Comparative use of taxes on goods and services is of particular interest with respect to the issue of reducing border costs, as this might entail loosened border controls on the move- ment of commodities.

Table 1 Comparative Tax Mixes in Canada and the US 2001

The US relies much more heavily on payroll taxes for social security than does Canada. However, US states have little counterpart to the general payroll taxes that four Canadian provinces apply, nominally to finance health care and education but in practice to supplement general revenues. These employer payroll taxes are an important source of revenues for Ontario, Quebec, Manitoba and Newfoundland and Labrador.

Finally, the two countries rely to a similar degree on taxes on real property and wealth. Canada has no coun- terpart to US governments’ use of estate, gift and inheri- tance taxes, but such taxes generate relatively little revenue and are scheduled to be reduced in future years.16 The bottom panel of table 1 shows that taxes for both countries are levied primarily at the federal level but that the federal share of total taxes is significantly larger in the US than in Canada, reflecting the provinces’ relatively larger spending responsibilities.

Competing with US Taxes

As explained above, Canada’s overall tax level need not be a competitive concern so long as both the mix and structure of taxes are efficient, but there are further rea- sons not to pursue blindly US tax levels. These reasons include the unsustainability of US tax levels, differing visions of society’s objectives and the economic returns to smartly focused public expenditures. A key reason is that US fiscal policy is on an unsustainable path that will require either sharp future tax rate hikes or severe spending cuts — either of which will be economically and socially costly.

According to the latest estimates by the US Congressional Budget Office (CBO), US federal deficits are projected to total US$1.9 trillion for 2005–14 (United States 2004, xii). Some independent experts, however, have asserted that the CBO figures seriously underesti- mate the real fiscal challenge, projecting total deficits on the order of US$4.4–US$9 trillion (Gale and Orszag 2004). They attribute this fiscal deterioration to declining revenues as a result of economic slowdown and the 2001 and 2003 tax cuts. They conclude that the US is on an “unsustainable fiscal path” once the longer-term impli- cations of an aging population are factored in. The US fiscal problem will be further compounded by recent leg- islation to expand the scope of Medicare coverage to include prescription drugs. Even the normally staid Financial Times opined editorially about US federal po- licy: “On the management of fiscal policy, the lunatics are now in charge of the asylum…Watching the world’s economic superpower slowly destroy perhaps the world’s most enviable fiscal position is something to behold.”17

It would hardly be prudent for Canada to cut its tax rates to imitate US tax rates that, in the long run, are unsustainable and some of which are already slated to rise. US federal revenue in 2004 will be smaller as a share of GDP than at any time since the 1950s. US leg- islation specifies that many of the key tax cuts enacted in 2001 and 2003 will “sunset” in various years from 2005 through 2011, and it has been estimated that making the tax cuts permanent would reduce rev- enues by about US$2 trillion over the next decade (Gale and Orszag 2004). Moreover, state and local governments are also confronting fiscal difficulties.18 Addressing budget problems in the US will evidently require both spending reductions and revenue increas- es. While Canada faces similar long-run fiscal pres- sures, it is on a much better fiscal footing. Canada’s federal budget is currently in surplus and projected to remain so for the coming decades. Although the provinces are projected to run deficits over this period, mostly as a result of rising health care costs, fiscal surpluses are expected for all Canadian governments combined (Conference Board of Canada 2003).

Given the greater fiscal stresses in the US, the bur- den of servicing public debt will become greater there than in Canada. In 2001 all levels of Canadian govern- ment paid 5.6 percent of GDP in interest charges against 3.2 percent for US governments.19 This 2.4 per- cent differential accounts for two-fifths of the 6 per- cent of GDP differential in the tax burdens of the two countries in 2001. As the US federal government con- tinues to rack up massive deficits, adding to cumula- tive debt, while the Canadian government maintains at least modest surpluses, this differential in interest costs will diminish and then reverse. Eventually, Canada will be able to reduce its tax burden to that of the US as a percentage of GDP without sacrificing the size of its program spending relative to the economy. Alternatively, Canadians may wish to take part of the dividends from declining debt-service charges in the form of more public benefits and services.

Yet another reason for Canada not to pursue US tax levels blindly is that Canadians and Americans often have a different vision for their societies. Many Canadians place greater weight on tolerance, diversity, equality, safe and vital city centres and civil society — all of which require supportive public services and infrastructure (see, for example, Adams 2003; Jedwab 2002, 2003). Hence, Canadians may choose higher tax burdens than do Americans — that is, less private con- sumption but more public consumption. Recent research on what determines success in the high- technology and other knowledge-based sectors has focused on the role of vital urban environments in attracting talented workers; they care more about tol- erance, diversity, “coolness” and cultural amenities than about tax burdens per se (Florida 2002a, 2002b). Therefore, higher tax burdens can have a significant economic payoff if the revenues are effectively chan- nelled to creating those vital and attractive cities.

Cross-national evidence confirms the disconnect between overall tax levels and the competitiveness of an economy.20 In its latest survey of global competi- tiveness, the World Economic Forum (2003) ranked Finland, Sweden and Denmark first, third and fourth, respectively; the United States placed second. Yet the three Nordic countries have tax burdens that are 10 to 14 percentage points of GDP higher than Canada’s and are the three highest of all the 30 countries that are members of the Organisation for Economic Co-operation and Development (OECD). Their tax mixes put heavier weight on consumption-type taxes (value-added and payroll taxes) than those in Canada or the US; their personal taxes also apply relatively low flat rates to capital income alongside steep pro- gressive rates on labour income. The lesson is not that high tax burdens per se create economic strength but, rather, that high tax burdens, if properly struc- tured, need not be a hindrance to competitiveness.

Canada placed just sixteenth in the 2003 global competitiveness rankings, down several spots from the previous year. To account for its downgrade of Canada, the World Economic Forum cites issues such as distortive subsidies, favouritism in governmental decisions, bureaucratic red tape and foreign owner- ship restrictions. The remedies for these problems, however, are to be found in improved governance and regulation, not in tax policy. On the positive side, the competitiveness survey cites Canada’s budgetary surplus, sound banking system, generous paid mater- nity leaves, Internet access in schools and industry- university research collaboration. Most of these positive factors need to be supported by adequate fiscal resources, which makes the role of a well- structured, efficient tax system even more important.

Taxation Principles

The analysis of specific tax policies for Canada requires an understanding of some basic taxa- tion principles. Most of the ensuing discussion draws on the research findings of economic theory and public finance and will be familiar to tax econo- mists and tax policy specialists.21 Other readers will likely benefit from a review of this material, but those in a hurry to move on to the examination of specific tax policies in the next section — and those impatient with economic analysis — should be aware of at least the following key findings: 

  • A consumption base is superior to an income base for reasons of economic efficiency and growth; an alternative way of implementing a consumption base is through heavier taxation of labour income relative to capital income or, at the limit, a tax on labour income and zero tax on capital income.

  • A consumption base for taxation is also more equi- table in its treatment of individuals with the same lifetime labour earnings but different preferences about when in their lives they wish to consume.

  • A full and complete conversion of the tax system to a base of consumption and/or labour income would be excessively costly in terms of foregone revenues, convey large tax windfalls to the wealthy and pose severe problems of compliance, enforcement and international harmonization.

  • Partial conversion to a consumption base would avoid such problems and could fulfill most of the economic goals; such an implementation strategy is best pursued at the personal level via greater access to tax-recognized savings (especially through a new tax-prepaid format) and at the business level via greater moves toward a cash flow base (with faster deductions or full expensing of capital outlays).

  • It would be better to pursue a consumption base through the direct tax system than through indi- rect (sales-type) taxes, because of issues relating to vertical equity, tax enforcement and minimization of border effects.

  • Provincial tax policy has less leeway for excess tax burdens than national policy, since the exchange rate cannot offset the competitive disad- vantage; provincial tax policy also has less room to pursue redistributive goals than national tax policy because of the greater mobility of labour within Canada than across the border.

Transiting readers should also be aware of the wide divergence of efficiency costs related to differ- ent types of taxes. Economists use a concept called the “marginal efficiency cost” (MEC) to measure the extra loss of valued output per extra dollar of rev- enue generated by raising the rate on a given tax base. For example, an MEC of zero would indicate that a tax has no efficiency cost — it extracts the extra dollar of revenue without any economic distor- tions. An MEC of 0.10 would mean that 10 cents of real valued economic activity is destroyed in the process of raising the extra dollar. One study of the Canadian economy found the following MEC values for alternative taxes: 0.17 for a sales tax or a consumption base, 0.27 for a tax on payroll or labour income, 0.56 for a tax on personal income (labour plus capital income) and 1.55 for a tax on corporate income or capital income.22 Hence, taxing capital income is much more economically costly than taxing consump- tion or labour income.

By shifting its tax bases from capital income (at both the individual and corporate levels) toward consumption and/or labour income, a government can reduce the eco- nomic costs of raising any given amount of revenues. Using the MEC values cited above, one can illustrate the potential economic gains from a more efficient mix and structure of taxes in Canada. For example, assume that $50 billion of revenues (out of the total of more than $400 billion per year) is shifted from less efficient bases (such as capital income) to more efficient bases (such as consumption or labour income).23 The ongoing gain in the level of real consumable output would be in the range of $20 billion to $65 billion per year — a pure gain apart from any associated costs of implementa- tion.24 Moreover, if simpler forms of tax were put in place in the process, further potential saving of adminis- trative and compliance costs would be possible.

Definitions of Alternative Tax Bases

It is useful at this point to preview the alternative tax bases and their definitions; this discussion covers all the major taxes on flows of economic resources and ignores taxes on stocks of wealth (such as property taxes and corporate capital taxes). Table 2 groups the tax bases into three categories: indirect, direct and transitional. An indirect tax is one applied to the sale or purchase of goods and services and is applied on each transaction. Indirect taxes cannot be adjusted to reflect the taxpayer’s situation or tax-paying ability; they tend to be regressive, at least from an annual per- spective on tax burdens. In contrast, direct taxes are imposed on businesses or individuals and can be adjusted to reflect their specific situation (through exemptions, deductions or credits) and their tax-paying ability (through progressive rate schedules). Some direct taxes, such as corporate income or payroll taxes, are nevertheless applied at flat rates.

The principal types of indirect tax include two single- stage forms: the retail sales tax, imposed only at the point of final sale to consumers, and the excise tax, typ- ically imposed at the wholesale or distributor level. There is also a multistage form, the value-added tax, imposed at each stage of production and distribution of products but only on the value added at that stage. Canada’s GST applies the commonly used credit-invoice method of value-added taxation. There are two other ways to implement a value-added tax that are eco- nomically equivalent yet applied to the financial accounts of business (the business transfer tax) or to those accounts plus payments to labour (the direct consumption tax). These forms might be labelled “transitional” tax bases in that they are not imposed on the sale or purchase of goods and services per se but, like indirect taxes, they cannot reflect the circum- stances or tax-paying ability of the consumer.

Table 2 Major Alternative Tax Bases

For direct taxes on business, the key distinction is whether to use an income base or a cash flow base. As can be seen in table 2, the difference between the two bases relates to the tax treatment of capital inputs by the firm. An income base reflects capital costs through an allowance for depreciation (“capital cost allowance” in Canadian tax jargon) over the life of the capital plus a deduction for interest costs in financing capital purchases. In contrast, a cash flow base allows an immediate deduction of the full capi- tal purchase price (often called “expensing”) but no deduction for financing costs. Because of the time value of money to a firm, the expensing rather than depreciation of new capital makes a cash flow base more favourable to firms than an income base. The cash flow base has both operational and economic advantages. It avoids the complexity, inaccuracies and bookkeeping burdens of accounting for deprecia- tion on diverse types of capital. And it avoids the economic distortion to firms’ investment decisions, as expensing fully shields the cost of capital (reflecting a normal return to capital) from tax.

Payroll taxes are direct forms of tax that can be applied to employers, employees or both. Most payroll taxes are imposed to finance specific social insurance programs, but they can also be imposed to collect general revenues. To the extent that a close link exists between the tax paid by (or on behalf of) the individ- ual worker and his or her future benefit entitlements, a payroll tax can eliminate even the modest efficiency cost noted earlier for taxes on payroll or labour income. Payroll taxes are usually applied at flat rates but sometimes have annual exemptions and/or ceil- ings on the taxable earnings per worker. In most cir- cumstances, the burden of a payroll tax falls fully or mostly on the worker regardless of whether the tax is nominally paid by the employer or by the employee. Although payroll taxes typically are applied to wages and salaries, the base can also include fringe benefits and employer pension contributions. In some cases, a parallel tax is applied to self-employment earnings to avoid a tax bias against employees.

Many jurisdictions prefer to make heavy use of direct personal taxes because their rates and provisions can be tailored to achieve both vertical and horizontal equity. The key distinction for direct taxes is whether their base is income or consumption. A tax on income applies to the full accruing returns to capital as well as to labour income and miscellaneous receipts (such as transfer payments). A tax on consumption — imple- mented through either a “tax-deferred” method that exempts savings or a “tax-prepaid” method that taxes labour income and exempts capital income — does not distort individuals’ choices about when in their life- times to consume. It also does not distort savings or investment decisions and does not insert a “tax wedge” between the gross and net rates of return on invest- ment or savings (see Bradford 1988). Hence, any tax on capital income, such as that on an income base that includes both labour income and capital income, departs from consumption tax criteria.

Taxes can also be characterized by whether they are origin type or destination type, a distinction that is important for the freer movement of goods and services across borders at minimum cost to business and con- sumers. A destination-type tax is one applied in or by the jurisdiction where the ultimate consumer resides; for exports, this requires that the tax not be applied in the producing jurisdiction, but for imports, the tax must be collected at the border or in a manner closely related to each shipment. An origin-type tax is applied in the jurisdiction where the good or service is produced or where the productive factors are located and paid; this obviates the need for border controls or the taxing of the movement of goods and services. Only the retail sales tax, excise tax and value-added tax are destina- tion type. All other transitional and direct taxes (both business and personal) are origin type.

Implementing a Consumption Base

There is growing consensus among tax economists that shifting tax bases away from income, particularly capital income, and toward consumption and labour income would yield significant benefits. It would improve the lifetime horizontally equitable treatment of households with different savings preferences — that is, individuals with the same lifetime labour earnings would pay the same total discounted taxes despite dif- fering choices about when to consume. Shifting tax bases toward consumption and labour income would also improve incentives for savings, and to the extent that savings did increase, it would promote long-run growth of both the economy and living standards. Even if aggregate savings were not affected, such a shift would improve the efficiency of resource allocation over time, thereby raising real living standards.25 Total sav- ings would be allocated to different types of capital for- mation (such as business investment versus housing) on a more neutral basis. Additional benefits would arise from simpler record keeping and tax reporting and potentially from reduced opportunities for tax avoid- ance and evasion using complex investment schemes.

In shifting tax bases, several possible problems would need to be considered in designing the new taxes. For example, without proper formulation, the vertical equity of the tax system could be unduly com- promised. Total revenues might also be reduced unless they were offset elsewhere in the tax system. In particu- lar, revenues might be lost in the form of windfall gains to holders of “old” capital at the time the transition took place without any concomitant incentives for incremen- tal savings. Since the largest holders of capital have high incomes, this issue also relates to the preservation of vertical equity. In addition, it would be important in shifting tax bases to avoid creating new kinds of oppor- tunities for tax avoidance and undue burdens on house- holds and businesses. And any such reform would also have to be coordinated with the tax systems of other countries, principally the US.

There are two ways to implement a consumption- based personal tax. One way is the “tax-deferred” or “registered-assets” method, which permits a deduction from the taxpayer’s income of savings in the form of contributions to trusteed plans. In effect, this method provides for the deferral of tax on both the principal amount saved and the accruing investment returns to such savings; when the funds are withdrawn for con- sumption, they become taxable. RPPs and RRSPs embody the tax-deferred method within the direct per- sonal tax. If all limits were removed from contributions to tax-deferred plans, the scheme would become a “personal consumption tax,” which could retain tax rate progressivity of any desired degree. Indirect sales taxes on consumption also embody the tax-deferred method, as no sales tax is paid on earnings until they are spent, but they tend to be regressive in practice.

The second way to implement a consumption- based tax is through the “tax-prepaid” method, some- times called a “labour income” or “wage” tax. Rather than working with the “uses” of income (consump- tion versus savings) as in the tax-deferred method, the tax-prepaid method considers the “sources” of income. Labour income is taxable but capital income is exempted from tax; the capital income simply reflects the return to savings. A tax-prepaid personal consumption tax would be a progressive tax on labour and other noncapital incomes. A partial move in this direction would require the establishment of tax-prepaid savings plans with limits on the contri- bution amounts. The tax-prepayment principle is already manifested in the tax-free treatment of gains on owner-occupied housing, the preferential tax inclusion rate on other capital gains and the tax treatment of some insurance products.

A simple numerical example can be used to illus- trate the equivalence of the tax-deferred and tax- prepaid methods. Assume that the marginal tax rate is the same in two periods, t, and that all savings yield a rate of return equal to the rate used to discount future values, r. Person A earns an extra dollar, which he immediately spends, with the result that it is taxed as part of current consumption and incurs tax of t. Person B similarly earns an extra dollar, but she saves it for consumption in the next period. There is no consump- tion tax due in the first period, but the dollar grows to 1 + r in the second period, when it is spent, incurring tax of t(1 + r). However, discounting that tax back to its value in the first period yields a present value of t(1 + r)/(1 + r) = t, the same as the tax that person A paid. Alternatively, if the tax is applied to labour income rather than consumption, both individuals would pay the same tax of t in the first period, again independent of when they chose to spend the earnings.

Equivalences and Differences in Methods

Both the tax-deferred and tax-prepaid methods are economically equivalent under particular conditions. That is, they yield the same present value for a per- son’s lifetime stream of consumption irrespective of choices made about when to save and consume, so the tax does not distort either consumer or investment choices. The net rate of return to an individual saver is then equal to the gross rate of return on the tangi- ble real investment, which promotes the economy’s allocative efficiency and growth potential. However, if there are departures from either of the stated condi- tions — that the rate of return on all assets equals the rate used for discounting future values, and that the individual’s marginal tax rate is identical when saving and consuming — then the two methods are no longer equivalent, with possible consequences for equity and efficiency (see Kesselman and Poschmann 2001).

When rates of return or marginal tax rates depart from these assumptions, the tax-deferred and tax- prepaid methods will have different effects, as shown in table 3. If some assets have rates of return that dif- fer from that used to discount future consumption, the two methods give rise to differing concepts of horizontal equity — the equal taxation of people with the same level of economic resources. The tax- deferred method displays ex post equity in that assets yielding above-average returns are taxed on their supernormal returns (a consumption-based tax exempts the normal return to capital). With the tax- prepaid method, there is ex ante equity, which means that taxpayers are treated equally only in the sense that they do not know beforehand which assets will prove to yield high rates of return. The supernormal returns that accrue to some assets will escape tax under the tax-prepaid method.26

Table 3 also shows, under “individual’s position,” which method is advantageous or unfavourable depending on the divergence of marginal tax rates (MTRs) between the points of saving and consump- tion. With the tax-deferred method, there is a gain to the individual whose MTR declines between saving and the withdrawal of funds for consumption. Conversely, tax deferral penalizes those whose MTRs rise between the time they save and the time they consume. In contrast, under the tax-prepaid method, changes in MTRs have no impact on the net returns to saving for later consumption, making this method more advantageous than tax deferral for those with rising MTRs and unfavourable for those with declin- ing MTRs. These differing attributes also mean that the two methods can be combined usefully within a personal tax. Tax deferral allows individuals to undertake lifetime averaging of their tax burden in a system with progressive annual tax rates; tax prepay- ment affords “intertemporal” economic efficiency in choices about when to save and when to spend.

Table 3 Tax Deferred vs. Tax Prepaid Method

Additional attributes of the two tax methods are summarized in table 3. In terms of vertical equity, tax deferral reduces the effective rate progressivity of the personal tax because savings are deductible at the individual’s MTR. Tax prepayment, in contrast, does not allow for the deduction of amounts saved, and hence treats savers in different tax brackets in a more uniform manner, thus preserving greater effective progressivity of annual tax burdens. Another way of viewing this difference is that tax deferral offers a form of lifetime averaging of taxes. It is also much simpler for taxpayers to comply with tax prepayment than with tax deferral, since only the latter method requires complex calculations and forecasts about future income for the individual to decide when to make tax claims or when to withdraw funds.

From the treasury’s position, the tax-deferred method entails large current revenue costs because of the tax deductibility of savings. These costs may grow for an extended period as the middle-aged population increases, but eventually they will reverse as individu- als retire and withdraw their tax-sheltered savings. The deferral of tax means that additional revenues will be generated in the future, which, in Canada’s case, could be helpful for financing public pension and health care needs when the large population of baby boomers retires (see Mérette 2002). In contrast, the tax-prepaid method entails revenue costs that are small at the outset but grow over time and are not reversed when savers retire and spend their savings. Hence, tax prepayment may aggravate future revenue problems when the baby boomers retire, unless budget- ary surpluses are generated in advance.

Another issue arises when there is a shift between the two methods of taxing consumption. Although the two methods are equivalent for individuals who live their entire lives under them, they are not the same if applied to individuals at different stages of their lives. To illus- trate this point, suppose the tax-deferred method is applied through an indirect sales tax and the tax-prepaid method through an employee payroll tax. Clearly, a shift from a sales tax to a payroll tax would favour retirees and those late in their working lives, since most of their lifetime labour earnings would escape the payroll tax, but it would be unfavourable to those who were young or early in their working lives. Conversely, a shift from a payroll tax to a sales tax would favour younger workers while penalizing older workers and retirees (who would have already paid payroll tax on most of their earnings and who would pay again, in sales taxes, when they spent their lifetime savings). This point suggests that policy should avoid sharp shifts between the two methods of taxing consumption and that it may be desirable to use a balanced mix of the two.

Partial Implementation Methods

Both methods of consumption taxation have been rec- ommended in previous tax reform proposals. In the US, the tax-deferred method was endorsed by Senators Sam Nunn and Peter Domenici in their proposed USA (unlimited savings allowance) tax (Boskin 1996); in Canada, it was supported in the Fraser Institute’s 1994 proposals for replacing the GST with a “personal con- sumption tax” that could retain a progressive tax rate schedule (Walker 1994). The tax-prepaid method was elaborated in the US in the Hall-Rabushka (1985) flat tax plan, promoted by congressman Dick Armey and presidential candidate Steve Forbes. The Hall-Rabushka scheme is a combination of a personal tax that exempts financial capital returns and a business tax based on cash flow (this is the business counterpart to consump- tion for households). In Canada, the former Reform Party favourably assessed a scheme of this kind but did not endorse it as official policy. A variant of the Hall- Rabushka scheme, Bradford’s “X-tax” (1986) combines progressive personal tax rates and a flat business tax rate imposed at the top rate of personal tax.

Major problems arise with the full implementation of a consumption-based tax. A central problem is how to handle capital or savings that were already accumulated at the time of transition. Under the tax- deferred method, such pre-existing wealth taken from undocumented sources could be contributed to regis- tered savings plans as “new” savings, thus sharply reducing the tax liabilities of high-wealth holders for many years. Under the tax-prepaid method, pre- existing wealth could be used to finance personal consumption without incurring any direct taxes. To prevent this revenue loss during the transition, it would be necessary to establish comprehensive accounting of all initial wealth holdings to ensure they were not being used for tax avoidance. The dif- ficulty of achieving such a goal, however, means that substantially higher marginal tax rates would need to be applied to taxable consumption and labour earn- ings, which would undermine the efficiency gains of the reform. This kind of tax avoidance would also sharply reduce effective tax progressivity because of the high concentration of wealth that is in the form of neither registered savings nor home equity.

One way to control opportunities for avoidance, reduce potential revenue losses and focus the incen- tives on incremental savings is to implement a consumption-based personal tax only partially, rather than fully. For example, annual contributions to reg- istered savings plans — whether tax deferred or tax prepaid — can be limited as a percentage of current labour earnings. This approach is used in the current Canadian RPP and RRSP schemes, which have both a percentage limit and an annual dollar limit, with lib- eral provisions for the carry-over of unused contribu- tion amounts. Two other methods of applying the tax-prepaid approach are annual exemptions on given amounts of financial income and partial exclu- sion of realized capital gains from personal tax.

Partial implementation methods do, however, have some limitations. One is that they restrict the forms of savings that qualify for consumption treatment. With registered plans and an annual financial income exclusion, the holding of marketable financial securi- ties is favoured relative to kinds of savings that do not qualify (for example, tangible assets, real estate and business assets other than those that are publicly traded). In addition, partial exclusion of capital gains is further biased toward equities, especially growth equities, relative to holdings of debt and dividend- paying equities, though the latter is partially mitiga- ted in Canada by the dividend tax credit.27 The tax-prepaid method is equivalent to a zero tax on capital gains — which some analysts (for example, Grubel 2003a) have backed — but it also offers neu- tral treatment of interest and dividend incomes. Moreover, tax-prepayment with limits on annual contributions avoids some of the most severe prob- lems of eliminating taxes on all capital gains, includ- ing large and economically inefficient revenue losses, windfall gains to the wealthy and a sharp reduction in personal tax progressivity.

The business tax counterpart to a consumption base for households is a cash flow base. Economists regard a business cash flow base as economically efficient and neutral for real investment decisions.28 Such a base further simplifies tax accounting by eliminating the need for depreciation accounting, as all capital pur- chases can be fully expensed (and there are no deduc- tions for financing costs). Indeed, the business component of the Hall-Rabushka plan embodies cash flow taxation. However, because the cash flow base exempts the normal return to capital, it reduces corpo- rate tax revenues and these would need to be replaced elsewhere. A cash flow base also leads to difficult tran- sition problems for firms with high debt loads, since their interest financing costs would no longer be deductible. Most of these difficulties could be avoided by allowing highly accelerated depreciation, or full expensing, only for new capital investment.

Direct versus Indirect Application

The choice of whether to apply a tax directly or indi- rectly pertains mostly to consumption-based taxes, for which there are both direct and indirect forms, but it raises similar issues in comparing an indirect consumption tax with various forms of direct perso- nal and business taxes. The first consideration is that an indirect tax is of limited use because its operation prevents adjusting rates based on the taxpayer’s abil- ity to pay or on other attributes such as family size or health status. For this reason, an indirect tax is suit- able only as a supplementary means of collecting revenues where substantial use is made of direct taxes on households. An indirect tax’s lack of dis- criminatory power often leads to the use of compen- satory provisions such as income-tested refundable tax credits for lower-income households.

Second, it has been argued that diversifying tax collection between direct and indirect methods reduces the potential for tax avoidance and evasion. Spreading taxes over multiple bases and collection points reduces the incentive to cheat on any one tax, it is claimed, and increases the cost of cheating on multiple taxes. Yet this argument is easily overstated, in that a business that cheats on its collection or remittance of GST will consistently misreport its gross revenues and costs on its income tax return. If it did not, an audit for one of the taxes would reveal that the other tax had not been properly reported. Hence, multiplicity of tax bases and collection points does not necessarily improve the tax authorities’ detection powers, but it typically increases their administrative costs and taxpayers’ compliance costs.

Third is an aspect of tax collection and enforce- ment of particular salience for Canada in choosing between direct and indirect tax formats. As noted above, an indirect tax is applied on a “destination” basis, whereby the tax is levied, paid and collected in the jurisdiction in which the final consumer of the good or service resides. In contrast, a direct tax is applied on the basis of the “origin” of the income used to purchase the good or service, and it is levied by the jurisdiction in which that income is generated and in which the factor owner resides. Hence, an indirect tax requires the means to collect taxes on purchases that individuals make across the border and consume at home — typically through controls on com- mercial and private purchases of goods shipped or car- ried across the border.

Canadian tax design must heed the last point in partic- ular if the goal is to increase the country’s competitiveness. Existing border controls to enforce the GST and provincial sales taxes impose large real burdens on businesses, indi- viduals and governments. These burdens include customs officers, brokerage and Canada Post import fees, shippers’ costs of preparing export documentation and higher couri- er and postage fees for shipments between Canada and the US than for shipments within either country. On top of these tangible costs are the time costs imposed on truckers and individuals queued at the border and on Canadian businesses waiting for the delivery of needed materials, parts and supplies.29 All these items are added to the costs of doing business in Canada, perhaps rivalling the costs of having a separate currency, which have been much- discussed in the dollarization debate.

Any plan to spur Canadian economic growth should thus seek to minimize border costs. Canada might also contemplate an eventual customs union with the US, whereby trade between the two countries is tariff free and both apply common tariff rates to trade with other countries. Such an arrangement would allow Canadian producers to operate in the North American market with as little friction to doing business as US firms experi- ence. Gains to Canadians’ real living standards also might result from reducing the use of Canada-specific distributors for many products.30 However, present sales- type taxes, with total rates averaging about 15 percent for all provinces except Alberta, are roughly double their US counterparts, a tax differential that is a major barrier to relaxing border controls on trade. Hence, Canada should consider reducing total indirect taxes toward the average US level, which would entail chang- ing the form of federal and provincial sales taxes.31

In switching from a destination-type indirect tax to an origin-type direct tax, two notable effects would arise. First, Canadian residents would be taxed on their consumption abroad, to the extent that it was financed out of domestic earnings. Unless the direct tax effect- ively captured their foreign earnings, Canadians might in part escape taxes here. Second, switching to an origin- type tax would raise domestic production costs; howev- er, as explained earlier in the discussion of tax levels, an automatic adjustment of the exchange rate would offset this effect. With flexible exchange rates, domestic auton- omy over tax rates would be preserved even with more relaxed border controls on trade.

One might question whether Canada’s direct tax system could bear heavier usage if indirect taxes needed to be reduced to minimize border costs. This approach might appear to run counter to the advice of Bev Dahlby:

We should levy both direct and indirect con- sumption taxes because they are subject to dif- ferent forms of tax evasion and avoidance… Therefore, it is preferable to have two sys- tems of consumption taxation, each levied at a moderate rate, rather than have one con- sumption tax system levied at the full rate. (2003, 100–01)

Dahlby cites two types of cross-border tax evasion: underreporting of foreign earnings for a direct tax, levied on a residence basis; and cross-border shop- ping with respect to an indirect tax, levied on a desti- nation basis. Relaxed border controls on the movement of goods would make the latter type of evasion easier without changing the ease of the for- mer type. Hence, it would dictate some shift of the tax mix away from indirect and toward direct taxes. Indeed, only a full common market, in which average workers could readily obtain US earnings, would also ease the former type of evasion and thus tilt the bal- ance toward indirect taxes.

Provincial Tax Principles

The analysis of provincial taxes differs from that of federal taxes in two important respects. First, at the national level, any differential tax burden will be accommodated through an exchange-rate adjustment to restore trade competitiveness; this process occurs regardless of whether the taxes provide good value for households and businesses. In contrast, any province that raises its tax burden without providing fully valued public goods and services for households and businesses will impede the ability of firms in that province to compete both nationally and internation- ally. There is no adjustment counterpart at the provincial level to the exchange rate for the national economy.32 Consequently, the provinces face more severe market constraints than does the federal gov- ernment on their ability to overtax relative to the value of public expenditures.

The second factor that differentiates provincial and federal tax policy is the greater mobility of labour within the country than internationally. Such mobi- lity arises across the skills spectrum but is greatest at the highest levels. As a result, a province must not only provide public goods and services that all tax- payers fully value; it must also do so with respect to the taxes high earners pay and the services they enjoy. Otherwise, some of them will leave the province, thus raising the gross pay for those occupa- tions and shifting the burden of more progressive taxes onto employers. Evidence from US states sug- gests that this mobility process prevents them from pursuing much redistribution through the tax system (Feldstein and Wrobel 1998). Accordingly, the provinces may be more constrained than the federal government in their ability to use tax and spending policies for redistributive purposes.33

Personal Taxes

Personal taxes are the largest component of the Canadian revenue system at both the federal and provincial levels, making them critical in any tax design. My analysis begins by assessing the Canadian personal tax base and comparing it with the US base. This leads to proposals for both expanding and reforming the base — including, most importantly, further moves toward a consumption base. I then examine personal tax rates in Canada as well as the top tax rates on various types of income, and compare them at both the federal and subnational levels with those in the US. This provides a basis for assessing the most economically beneficial ways to modify the per- sonal tax rate structure at both the federal and provincial levels.

The Base for Personal Taxes

In its efforts to become a Northern Tiger, Canada should seek to implement the broadest, most efficient base for personal taxes, with due regard for horizontal equity. For both Canada and the US, the personal income tax is by far the largest source of tax revenues at the federal level. Fortunately, Canada enjoys some advantages by not replicating several leakages found in the US tax base. One of these is the deductibility of mortgage interest, which moves the US base away from consumption, favours current spending and biases savings toward investment in home equity at the expense of business investment.34 In the 2000 tax year, itemized deductions for mortgage interest totalled US$300 billion, or about 7 percent of total US adjusted gross income. Another leakage in the US tax base is the federal deductibility of state and local income taxes, real estate taxes and personal property taxes, which in the 2000 tax year shrank the tax base by a further US$295 billion, or another 7 percent.35

A third type of personal tax base erosion in the US arises through the tax-free treatment of interest on state and municipal bonds (“munis”).36 As of late 2002, the total volume of munis outstanding was US$1.76 trillion, of which about 70 percent was held by individuals (either directly or through funds).37 At a tax-equivalent interest rate of 4.5 percent (almost all munis are long term), this treatment removes more than US$55 billion from the tax base of upper- bracket US taxpayers. Tax-free treatment is an ineffi- cient method for making intergovernmental transfers. It also reduces the progressivity of the tax system and encourages excessive spending by governments that have access to cheap financing. For these reasons, Ontario’s issuance of tax-free Opportunity Bonds in 2003 is a troubling Canadian precedent, one that other jurisdictions should eschew.38

Although the deductions and exclusions available in the US lessen effective tax burdens, they do so in a way that is economically inefficient. They necessitate marginal tax rates that are higher than would other- wise be needed to generate the same total tax rev- enues. These higher MTRs carry larger efficiency costs than necessary by distorting taxpayer choices with respect to work, saving and investment.39 Hence, by avoiding such base-eroding tax deductions, Canada can either raise the same relative revenues as the US with lower MTRs and lesser economic distor- tion or raise greater revenues than the US with com- parable MTRs and economic distortion — either way, a useful competitive advantage.

Canada’s personal tax base is broader than that of the US, but it still has some areas in which efficiency and horizontal equity could be improved.40 The most notable of these areas is the omission of employer- paid health and dental insurance benefits as a taxable fringe benefit for employees. This omission reduces the tax base while causing inefficient overprovision of such coverage relative to taxable compensation. It also penalizes disproportionately workers at lower earnings levels, who have little or no such coverage. Another example is strike pay, which is nontaxable, while the union dues that finance it are tax deductible, an arrangement that is inconsistent and encourages more and longer work disruptions. Moreover, some transfer payments to individuals are nontaxable, which may make sense for income-tested transfers, where taxability would exacerbate work disincentives, but transfers such as workers’ compen- sation benefits should be taxable just as are employee compensation and unemployment benefits.

To improve efficiency, growth and competitiveness with the US, however, the most critical change that could be made to Canada’s personal tax base is a further shift toward consumption. Currently, high earners in Canada can take advantage of only about one-third of the tax-recognized savings of their US counterparts, which raises their relative tax burdens and distorts their savings behaviour. The 2003 federal budget’s proposed phased hikes in the dollar ceiling for RPP and RRSP contributions from $13,500 to $18,000 over several years is long overdue, but it does not go far enough. Any dollar ceiling on contributions eliminates the effi- ciency gains for individuals who are constrained by that ceiling; for them, it is like a lump-sum tax reduction with no marginal incentives to save. Moreover, there is no reason to keep high earners from undertaking life- time savings on the same consumption-tax basis as low and middle earners.

Maintaining the link between contributions and labour earnings is an effective way both to limit con- sumption treatment to lifecycle savings and to prevent the large revenue loss of a full consumption tax base. The existing 18 percent rate is adequate for those not constrained by the dollar limit, given the ability to carry forward unused contribution amounts — a point con- firmed by the fact that very few of those individuals actually contribute their full allowance. However, along with eliminating the dollar ceiling on contributions, it might be sensible to apply a lower rate (such as 15 per- cent) for allowable contributions on earnings above $100,000 per year. Such a change would reflect the pro- gressivity of the tax rate schedule, the application of the percentage limit to gross earnings and the need for retirement savings to replace only part of net earnings.41

Further moves toward a Canadian personal consump- tion tax will invoke the choice between tax-deferred and tax-prepaid methods. The 2003 federal budget announced plans to consult on the possible adoption of tax-prepaid savings plans (Canada 2003a, 341–42). In fact, several considerations support the expansion of access to tax-recognized savings along tax-prepaid, rather than tax-deferred, lines (Kesselman and Poschmann 2001). These include lower immediate rev- enue costs, greater economic efficiency, less sacrifice of vertical equity, and greater benefits for low-income earn- ers whose higher MTRs in retirement make tax-deferred savings unattractive. Yet another virtue of the tax- prepaid approach is that high earners would not be tempted to emigrate to skirt their tax-deferred liabilities on dissaving; at present, after emigrating, such earners can reduce their tax on withdrawals from tax-deferred plans by taking advantage of low nonresident tax- withholding rates. Still another incentive for Canada to move toward treating tax-recognized savings along tax-prepaid lines is that the US administration has pro- posed converting existing tax-deferred Individual Retirement Accounts to tax-prepaid schemes.42

In addition to instituting new TPSPs that are inte- grated with existing tax-deferred savings plans, Canada could usefully reinstate annual exemptions on a limited amount of interest and dividend income. Prior to the tax reforms of the late 1980s, individuals could exempt $1,000 of such income each year; an appropriate contemporary figure would be about $2,000 per year. This provision would allow for consumption-type treatment of small amounts of savings without the need for formal registration of plans or record keeping of contributions. An exemp- tion method would not be a suitable substitute for much larger TPSPs, however, since the exemption can be used for pre-existing savings, and access to it can- not be readily linked to labour earnings. A fixed exemption for interest and dividends would also encourage families to accumulate a modest rainy-day fund outside their registered savings to help tide them over emergencies or jobless spells.

As part of the move toward a consumption tax base, changes to the rules for interest deductibility would be appropriate. Current rules require tracing borrowed funds to specific financial assets to block deductions on homeowner or consumer loans; interest cannot be deducted on loans to finance RRSP contri- butions; and Finance Canada has proposed barring deductions for interest to finance equities oriented to capital gains rather than to dividends. An attractive way to reform these rules would be to limit the tax- payer’s annual interest deductions to total income from interest, dividends and taxable capital gains. Any excess amount could be carried forward for use in future years. This approach would eliminate the complexities of tracing, prevent economically ineffi- cient tax arbitrage on leveraged equity investments and maintain the consumption tax base’s integrity. The US tax system already uses a similar method.

Reform of Canadian personal taxes relating to mat- ters in the international arena would also bring greater economic benefits. For example, the limits on foreign content in registered savings and pension plans oper- ate to the disadvantage of Canadian workers and savers with no offsetting benefits (see Fried and Wirick 1999) and should be completely eliminated. The with- holding taxes on interest and dividends paid to non-

residents impose significant burdens on the Canadian economy (see Mintz 2001b), and they should be removed through bilateral agreement with the US or else unilaterally lifted. Other US tax provisions make it more burdensome for US managers, professionals and technical workers to work in Canada,43 thus reducing the benefits to the Canadian economy from tapping specialized skills. Canada should negotiate with the US to remedy this matter.

Rates of Personal Taxes

Personal income taxes (PITs) represent a somewhat smaller share of total taxes in Canada than in the US and a slightly larger share of GDP (see table 1). Despite this rough aggregate similarity, the composi- tion between federal and subnational usage of the PIT differs sharply in the two countries. In Canada, fed- eral income taxes are lower than in the US, but provincial income taxes are nearly twice as large as their US state and local counterparts.44 In this section, I examine the comparative tax rate structures at the federal level in the two countries, then at the subna- tional level and finally the top MTRs for combined federal-subnational income taxes.

Table 4 presents the federal PIT rate schedules for the two countries in 2003, with the US taxable income brackets converted into Canadian currency using two exchange rates: a recent market rate (US$0.76) and the purchasing power parity (PPP) rate (US$0.84).45 The table includes the major US rate cuts of 2003 and presents schedules for three types of tax filers; Canada has just one rate schedule for all filers. When married couples in the US opt to file separate returns, they face tax rate brackets that are just half the width of those for married joint filers. Hence, for Canadian couples with partners having equal incomes, the US rate schedule for married separate filers (table 4e) offers the most appropriate comparison.

At incomes below $30,000 for single earners and $65,000 for couples, US federal MTRs are distinctly below those in Canada, a disparity that is augmented by larger personal exemptions and deductions in the US. At a taxable income of about $60,000, or $120,000 earned equally by married partners, the Canadian federal MTR of 22 percent compares with an MTR of 25 percent in the US. The top MTR of 29 percent in Canada applies for taxable incomes above $104,600, while the top MTR in the US is 6 percentage points higher at 35 percent and applies for taxable incomes above US$312,000 (or US$156,000 for married separate filers). In Canada, a couple with equal incomes does not incur the top rate until reach- ing a combined taxable income of $209,000. At a comparable income using PPP in the US, the couple filing jointly would pay an MTR of 33 percent.

Table 4 Comparative Canada US Federal Personal Tax Rates 2003

Tables 5 and 6 show the top MTRs for PIT at the Canadian provincial and US state levels, respectively, for 2003. In Canada, the top MTRs range from 10 per- cent for Alberta (with its flat rate PIT) to the upper teens for several provinces (after adjusting for the rebate of federal tax to Quebec taxpayers). British Columbia and Saskatchewan have top MTRs of 15 percent or just below. In most provinces, these top- rate brackets apply to incomes substantially below the $104,600 threshold for the top federal MTR, although New Brunswick matches the federal thresh- old and Saskatchewan approaches it. Overall, the rate schedules of the provinces other than Alberta display considerable progressivity of MTRs across incomes (though this is not displayed in table 5).

Table 6 classifies the US states by their top MTRs, with seven states having no PIT and another two tax- ing only interest and dividends. Six states, including most of those with the lowest top MTRs, apply flat rate schedules. For many states with progressive PIT schedules, the top MTR is attained at relatively mod- erate incomes, such as US$20,000 (or US$40,000 for couples) in New York. This pattern and the fact that several states have either no PIT or a flat rate PIT means that, overall, US state PITs are much less pro- gressive than Canadian provincial PITs; this contrasts with the greater progressivity of the US PIT than the Canadian PIT at the federal level.46 Only two states (Montana and North Dakota) have top MTRs above 10 percent, and their rates are exceeded by those of all Canadian provinces except Alberta.

Table 7 displays the top MTRs of combined federal and state/provincial PITs for illustrative jurisdictions in each country and for major types of income. For US states with no PIT, the results correspond to the column labelled “federal only.” The illustrative states include relatively low-taxed Michigan, with a top MTR of 4 percent, and high-taxed California, with a top MTR of 9.3 percent. For comparison, the table presents rates for the four most populous Canadian provinces. Of course, the table does not show the pattern of MTRs for indi- viduals who are not in their country’s top tax brackets.

As table 7 shows, the US federal top MTRs on labour income include a payroll tax for Medicare that is applied to all earnings without an upper limit; this adds 1.45 percentage points for employees and 2.9 percent- age points for self-employed workers. These figures assume that the payroll tax paid by employers is borne by them rather than shifted to employees. For labour incomes, top total MTRs are roughly comparable between low-taxed US jurisdictions such as Michigan and low-taxed Alberta and also between high-taxed California and mid-taxed Ontario. But top MTRs for Quebec and the Atlantic provinces (not shown in the table) exceed even the most highly taxed states.

Table 5 Provincial Income Taxes Top Marginal Tax Rates 2003

Table 6 State Personal Income Taxes Top Marginal Tax Rates 2003

Table 7 Top Marginal Tax Rates by Type of Income US and Canada Combined Federal and StateProvincial 2003

The top MTR patterns diverge much more for dif- ferent types of capital income. MTRs for interest income are somewhat lower in the illustrative US states than in their Canadian counterparts. And with the 2003 cut in the US federal tax rate for dividends of higher-bracket taxpayers to a flat 15 percent, the top US MTRs on dividends are now substantially lower than those in Canada. The US federal tax rate on long-term gains on assets held for more than one year has also been cut to a flat 15 percent. Yet, Canada’s 50 percent tax exclusion rate on capital gains leaves top Canadian MTRs on long-term capital gains fully competitive with those in the US except in zero-tax states; most of the other states tax long- term gains fully without any preferences. For short- term capital gains, top MTRs are sharply lower in Canada than in the US, since the US taxes such gains at full statutory MTRs while Canada does not distin- guish between short- and long-term gains.

Changing Tax Rates and Brackets

Reductions in MTRs for personal taxes are economi- cally useful for the entire income spectrum; in terms of efficiency, incentives and a competitive tax system, however, such cuts are best targeted on particular income ranges.47 They are most important at upper- middle and relatively high incomes (covering many high-technology, knowledge-based, managerial, entre- preneurial and professional workers) but not at the very highest incomes. They are also important at moderate incomes, where clawbacks of benefit programs com- pound with PIT rates to yield high total effective MTRs. Another way of cutting MTRs for particular income ranges is simply to extend the income brackets for exist- ing statutory rates so that more individuals are shifted into a lower MTR.

Further reductions in Canadian federal tax rates would be useful to offset at least partially the provinces’ high PIT rates relative to those of US states. Federally, the US PIT is considerably more progressive than the Canadian tax, largely because of the relatively low income levels at which the Canadian tax hits its upper rates. At the subnational level, Canadian perso- nal taxes are more progressive than US taxes but not by enough to offset the greater progressivity at the fed- eral level in the US and the fact that state taxes are typically small. But the goal of Canadian tax reform is not simply to compete with the lower personal tax bur- den in the US. Rather, it is primarily to reduce the dis- incentives and inefficiencies Canadians face as workers, savers, investors and entrepreneurs — and to moti- vate, attract and retain the most skilled, talented and productive workers.

Table 4b illustrates how Canadian federal PIT rates could be reduced and their income brackets expanded to achieve the desired goals. The bottom-bracket MTR could be left unchanged at 16 percent, but low- income workers would obtain relief by a 25 percent hike in the taxable income threshold. The two middle tax brackets of 22 and 26 percent could each be reduced by 2 percentage points and their brackets considerably widened;48 the top bracket threshold could be raised by at least 50 percent. Although the 2003 top-bracket threshold of $104,600 is already slated to rise to $113,800 in 2004, this is still very low compared with the US federal tax threshold for the top MTR. A large increase in the top Canadian thresh- old would bring the incentives of a large drop in MTRs for upper earners (5 percentage points, from the current 29 percent to the new 24 percent rate bracket); it would also send an important signal to productive workers that their talents were welcome in Canada.

The top Canadian federal MTR could be left unchanged, or reduced slightly to 28 percent to match the third-highest rate in the US federal rate schedule. High-income earners would gain the most if the dollar ceiling on contributions to tax-recognized savings were removed and TPSPs were introduced. The economic efficiency gains would be greater from reducing tax rates on savings and capital income than from reducing tax rates on labour earnings, a point that is particularly salient for high earners, who tend to be high savers as well. Saving via TPSPs is equivalent to a zero marginal tax rate on all capital income within the account.

For beneficiaries of the National Child Benefit (NCB) supplement, widening the bottom federal MTR bracket to $40,000 would facilitate relief of the high effective marginal rates they face.49 At present, total marginal rates, including PIT and benefit clawbacks, can rise above 60 percent for NCB beneficiaries with incomes in the $21,500 to $33,500 range, and recent and planned enrichments to the supplement will only exac- erbate the disincentives and inefficiencies such earners face.50 With a wider tax bracket, the clawback rate could be reduced by starting the benefit phase-out at lower incomes and extending it to higher incomes without crossing into the next tax bracket. Even better would be to curtail NCB supplements and redirect the saved resources to in-kind benefits for children with special needs not targeted by family income.

More generally, federal and provincial programs have used the personal tax system to deliver a range of cash and in-kind benefits, raising effective MTRs for many taxpayers at moderate to middle incomes.51 This practice worsens incentives for training, promo- tion and work, and it should be curtailed in future tax and transfer policies. Better ways need to be found to structure the benefits of such programs — particularly to link them to positive behaviour by individuals in the form of, for example, labour force participation and savings, rather than dependency.

One might ask whether there is any good policy reason for Canada to follow the US federal tax rate cuts on capital gains and dividends. As table 7 shows, Canada’s top tax rates are already competitive on long-term gains and much lower than comparable US rates on short-term gains. Moreover, by avoiding dis- tinctions based on asset-holding periods, the Canadian tax obviates the complexities of tax plan- ning and avoidance that arise in the US. With respect to the tax rates on dividends from Canadian corpora- tions, the dividend tax credit acts to reduce effective rates below those on interest and labour incomes. However, when the tax inclusion rate for capital gains was cut from 75 percent to 50 percent in 2000, no parallel adjustment was made in the dividend tax credit, which left an imbalance between effective tax rates on corporate dividends and retained earnings. This imbalance should be corrected by enhancing the dividend tax credit. This move would be justified on the basis of domestic tax policy alone, but it would also bring top effective tax rates on dividends in Canada much closer to US rates.52

Rather than attempting to undercut US tax rates on all capital gains and dividends, a more effective policy for Canada would be to pursue greatly increased access to tax-recognized savings accounts, particularly TPSPs. The effective personal tax rate on returns to assets held in these accounts is zero, but tying the contribution limits to labour earnings would avoid large windfall gains to holders of wealth outside registered plans prior to the change. This method would focus the tax incentives effectively on incremental savings and prevent massive tax reduc- tions for the very wealthy.

Since labour is more mobile within Canada than across the border with the US, one would expect the provinces to exercise less PIT rate progressivity than does the federal government. Indeed, a province that pursued greater progressivity than Ottawa would tend to push up the gross wages of its skilled workers to compensate for their higher taxes, thus driving out businesses that employ high-skilled workers. Before 2000, all the provinces (except Quebec) were restricted from pursuing flatter rate schedules by the operation of the federal-provincial tax collection agreement. With the shift to a tax-on- income system in 2001, the provinces were unleashed from the federal rate schedule and can now choose any degree of rate progressivity they like. Several provinces have exploited this new flex- ibility. Alberta has gone so far as to introduce a flat rate tax, albeit one that applies above a much higher taxable income threshold that actually raises the effective progressivity of average tax rates over low to middle incomes. Saskatchewan has also pursued rate flattening.53 British Columbia cut personal taxes considerably in 2001, but the cuts were roughly pro- portionate across-the-board, leaving progressivity little changed and higher than that of the federal rate schedule — although the province did signifi- cantly reduce its top MTRs.

Interestingly, Ontario and Quebec have departed from the predicted outcome of decreased personal tax progressivity. Quebec has increased the progressive tilt of its rates and is likely to do so further as it cuts personal taxes sharply in future years. This might be explained by the lesser mobility of Quebecers, espe- cially francophones, in response to higher progressiv- ity; it may also reflect a greater preference by Quebecers for redistributive policies. In Ontario, a surtax makes the province’s tax much more progres- sive than the federal tax, which might be explained by the high gross wages paid for top skills in Toronto, the nation’s pre-eminent business centre. The Ontario surtax may be capturing economic rents of the top earners without causing them to leave for other provinces where their net earnings would be lower even with lower tax rates.

Regardless of whether the predicted economic ten- dencies play out in further flattening of provincial PIT rate schedules, some changes would be desirable. To achieve more competitive overall tax rates for the most skilled workers, further reductions in upper MTRs would be helpful. This issue is most acute for the Atlantic provinces and Quebec but also applies to Ontario’s PIT surtax. An ultimate goal for top MTRs in all provinces should be no higher than 15 percent — near the current top rates in British Columbia and Saskatchewan. Moreover, the provinces should move the income thresholds for the application of their top rates toward the much higher threshold applied for the federal PIT, and they should follow any future increases in the federal threshold. This change would further improve the Canadian PIT climate for highly skilled workers and specialized talents. In the event that many provinces pursue much flatter PIT rate schedules, the appropriate response to maintain vertical equity would be to steepen the rate schedule of the federal tax.

Yet there are limits on the progressivity of the federal personal tax rate structure, too. Although labour mobi- lity is typically much greater interprovincially than internationally for Canada, this may not be true for some of the most skilled and highest-paying occupa- tions. The relevant point of comparison and potential relocation for some workers may not be elsewhere in Canada but, rather, in the US. An investment banker is more likely to move from Toronto to New York than to Calgary, a neurosurgeon from Vancouver to Los Angeles than to Montreal. Overall, though, the federal govern- ment is better suited than the provinces to exercise tax- ation’s redistributive function.

Payroll Taxes

The premiums levied to finance social insurance programs can carry tax-like distortions, depend- ing on how they are structured. Although they are denoted “premiums,” they are mandatory levies and hence similar to taxes. They are applied mainly to an economically efficient base of labour income; to the extent that premiums are applied to the net income of unincorporated self-employed workers, they may also strike some capital incomes. Moreover, such premiums can be structured as a kind of user charge for the asso- ciated program’s benefits. If there is a tight linkage between taxes and benefits — in an ex ante sense of “expected” benefits — then the tax distortions of labour market decisions are attenuated. To the extent that the premiums individual workers and their employers pay are disconnected from their prospective benefit entitle- ments, they become a distorting general payroll tax (see Kesselman 1996, 1997).

Canada and the US differ sharply both in the rate of premiums they levy for social security and in their dis- torting effects.54 As table 8 shows, CPP premiums have a total rate (split evenly between an employer levy and a worker levy) of 9.9 percent, while the total rate for US Social Security, including Medicare, is more than half again as high at 15.3 percent. Even more dramatic is the difference in the amount of annual earnings per worker subject to the premiums. In 2003 in Canada, CPP pre- miums are levied only on a worker’s annual earnings up to $39,900 with an exemption for the first $3,500; in the US, the levies have no exemption and apply up to annual earnings of US$87,000, which equates to $114,500 at a recent market exchange rate or $103,600 at the PPP exchange rate. Thus, total maximum annual premiums are $3,600 in Canada and US$13,300 in the US (equivalent to $17,500 or $15,800, depending on the exchange rate used).

Not only are the maximum premiums four to five times higher in the US, they also act more like a distort- ing general payroll tax. In Canada, there is a relatively tight linkage between total CPP premiums paid over a worker’s lifetime and his or her benefit entitlements in retirement. In contrast, benefits in the US system have a strongly redistributive tilt, so that US workers with higher lifetime labour earnings get a poor return on their premiums.55 Canada does provide substantial income redistribution for retirees with low lifetime earn- ings, and therefore low CPP benefits, but it achieves this through universal payments of fixed amounts under the Old Age Security (OAS) program.56 The financing of OAS through general revenues rather than through CPP pre- miums means that the tax distortions are found else- where in the Canadian revenue system.

An important point from this comparison is that, because of its sharply lower payroll taxes, Canada can remain competitive with the US even with signifi- cantly heavier burdens from other kinds of taxes. Given that the largest differential in premiums between the two countries arises at upper-middle incomes, Canada could impose heavier personal taxes in the range of $50,000 to $120,000. So long as the Canadian personal tax base is reformed toward con- sumption and labour-income bases, the total distor- tions of the Canadian system could still be less than those of the US system. Since US Medicare premiums, including both the employer and employee shares, add 2.9 percentage points to the taxes on labour incomes at the highest incomes,57 the top US federal MTR on labour income is 37.9 percent, nearly 9 percentage points above the current Canadian top federal MTR.

Yet another aspect of the comparative positions of social security systems favours Canada over the US. Canada has now completed a multiyear increase in CPP premium rates designed to ensure the future financial viability of the program by building up sub- stantial reserves against future benefit liabilities. The US Social Security and Medicare systems, in contrast, are widely agreed to have massive unfunded liabilities that will necessitate major program adjustments. If those adjustments are made primarily through financing rather than through curtailing benefits, US premium rates could rise further, making Canadian premium rates look even more favourable. (By con- trast, the social security systems of European Union countries commonly have total premium rates in the range of 30 to 50 percent.) Of course, Canada does remain exposed to some unfunded future liabilities through its general revenue financing of OAS and medicare programs.58

Much of the Canadian federal government’s suc- cess in generating budgetary surpluses in recent years can be attributed to the large excess of EI premiums over program costs. Each of the numerous EI premi- um rate cuts over the past 10 years has been quite small; cumulatively, they have been insufficient to bring revenues down to the declining costs of bene- fits. The 2003 federal budget announced that “[T]he Government will consult on a new EI rate-setting regime for 2005 and beyond, based on the principles of transparency and of balancing premium revenues with expected program costs” (Canada 2003a, 26).

Table 8 Canada and US Social Security Premium Rates 2003

Yet, there was no mention of whether the EI pro- gram’s cumulative surplus would be counted in this future “balancing” act or, more likely, simply be swallowed by the federal treasury in a one-time accounting change since the funds actually ended up as part of federal general revenues.

One consequence of these large EI surpluses is to tilt the overall revenue system further toward labour income. Although this might be viewed as a positive move toward promoting Canadian competitiveness, there are good reasons to reduce EI premiums to self- financing rates and to collect any lost revenues else- where in the tax system. Growing surpluses in the EI account are an open invitation to increase program spending through new and expanded benefits. Such benefits, however, should be carefully assessed on their merits, not pursued simply because funds are at hand. Moreover, the use of EI premiums to collect what amounts to a labour income tax provides an inefficient bias toward self-employment, a regressive tilt because of the ceiling on taxable earnings, and a disincentive to hiring, particularly lower-skilled workers.

In addition to being reduced, EI premiums should be experience rated, as Kesselman (1983, chap 9); Canada (1998, 8.7–8.11); and Poschmann and Robson (2001), among others, have suggested. Experience rat- ing makes premiums like a user charge, whereby the rate for each employer reflects its own record of gen- erating the usage of program benefits through lay- offs. This change would provide incentives for employers to find ways to stabilize their labour demand — such as inventory policies, adjusted work hours and complementary seasonal lines of activity — thereby reducing their layoffs. The US unemployment insurance system has used experience rating for decades, and it has been found to reduce rates of benefit claim and unemployment. In Canada, regional and industry opposition to such a reform undoubt- edly would arise from affected sectors (principally construction and resource industries), but political resolve will be required in order to improve the coun- try’s competitiveness.

As an interim step to experience rating, the EI pre- miums of employers could be set on an industry- rated basis. This would be administratively simpler and could mute opposition from groups concerned about a tightening of EI benefits for voluntary depar- tures. It would also follow the practice of some provincial workers’ compensation programs that use a mix of premium rating by employer and by indus- try. Industry rating of premiums would improve efficiency by allocating program costs to those industries and products generating the most layoffs. It would also remove the EI program’s cross-industry subsidies, which retard the growth of high value-added and knowledge-based sectors. It would not, however, achieve the additional gains of inducing individual employers to stabilize their employment.

Another useful element of a redesigned tax system for Canada might be a general payroll tax (GPT).59 In principle, the best way of taxing labour income is through the personal tax, since it already exists and any desired exemptions and rate progressivity can be applied through it. However, since many Canadians may perceive that personal taxes are already too high, both domestically and in comparison with those in the US, the use of a payroll tax could be a more acceptable alternative to raising personal taxes. Indeed, the US makes heavy use of payroll taxes, such as its 2.9 per- cent flat levy for Medicare, and many EU countries apply payroll taxes at high rates to compensate for their lighter relative use of personal taxes. A Canadian federal GPT could generate the revenues to facilitate other desirable tax changes and rate cuts described in this paper. At a rate of just 2.5 percent, a Canadian tax could raise more than $15 billion per year. Labelled or earmarked for a popular national purpose, such as health care, the new tax might gain public acceptance.60

The formulation of a GPT would have to address a variety of issues (see Kesselman 1997, chaps 4–7). A key choice would be whether to structure the tax as one on employers, on employees, or on both. A tax solely on employers could be operated as a tax on aggregate pay- rolls but would be deductible in their business taxes, and therefore would require a higher GPT rate than a tax on employees to collect a given amount of net rev- enues. A tax on employers, moreover, would arouse less public resistance, but it would also reduce accountabi- lity relative to a tax on employees. In addition, a tax solely on employers might run afoul of constitutional bars on Ottawa’s taxing provincial governments. To be economically neutral, the base for a GPT should encom- pass fringe benefits, employer pension contributions and self-employment earnings. A base exemption for taxable annual earnings per worker or an enrichment of refundable tax credits would also be needed to insulate the lowest earners from the impact of the tax.

If a GPT were instituted for other fiscal purposes, it would be desirable to reform the financing of the CPP. Part of the premiums (though a declining share over time) is needed to pay for benefits received by earlier retirees under the plan, who paid premiums that were well short of the cost of their benefits. This portion of the CPP’s cost is financed, along with the rest of the program, by premiums that fall on earnings of up to $39,900 per year. To spread this burden to all earn- ings levels, the program’s financing could be shifted to a GPT with no taxable ceiling (like US Medicare premiums). Such an approach would reduce the regressive impact of CPP premiums, tighten the link- age between premiums and benefit entitlements and open the potential of innovative options for workers to invest their CPP funds.

Taxes on Goods and Services

Indirect taxes on goods and services would pose the most challenging fiscal issue for Canada if it sought to form a customs union with the US or simply to reduce border costs. The essence of a cus- toms union is that countries agree to adopt a common set of customs duties with the external world and to eliminate all duties and nontariff barriers to trade with each other. This allows the elimination of border controls on the movement of goods and the contract- ing for services. Depending on other arrangements between the countries, such as whether they had also formed a North American security perimeter, the cus- toms union could allow for unimpeded crossing of the border. The two countries might still wish to maintain some checks on immigration across the border,61 but, other than for security reasons, there would no longer be a need to check for goods, either carried by private individuals or contained in commercial shipments. Short of a customs union, even incremental steps toward the freer movement of goods across the border would entail some relaxation of border controls.

The reason this poses a challenge for Canadian tax policy is that this country relies much more heavily on broadly based taxes on goods and services than does the US (as table 1 shows). At the subnational level, the two countries are roughly comparable in their reliance on indirect consumption taxes, but at the national level, Canada applies a broadly based indirect tax on consumption in the form of the GST. A gap of this magnitude — 7 percent of the total tax mix and almost 3 percent of GDP — would create major incentives for cross-border shopping by Canadians as well as telephone, mail-order and Internet purchases of goods from the US for delivery in Canada.62 Since most Canadians reside relatively close to the border and since shipping costs are small relative to the tax savings on many higher-value consumer goods, this differential would not be sus- tainable. In short, in contemplating a customs union with the US or simply a much freer border, Canada would have to find a way to replace either the federal or the subnational sales tax with another source of comparable revenues.63

In this section, I consider how Canada could replace its GST with another form of revenue that would be collected more directly and would not be open to eva- sion through cross-border purchases. I also consider the alternative, and perhaps preferable, policy of the provinces’ replacing their sales taxes with a more direct revenue source. In either case, the revenue of an indi- rect tax should be replaced with another tax that also employs an economically efficient consumption-type base. In addition, I examine ways to improve the fed- eral and provincial sales taxes that fall short of trans- forming or replacing them.

Federal Sales Tax

If the GST were abolished, one way to replace its lost revenues would be to make the direct personal tax more consumption based, as suggested above, and then simply raise personal tax rates across the board. Such an approach would, however, entail some diffi- culties. First, to maintain existing total revenues, fed- eral MTRs would have to be raised across the board by about 35 percent of existing rates. The high visi- bility of personal tax rates would make such a sharp rise politically thorny, and it might also be viewed as making Canadian personal tax rates uncompetitive relative to those of the US. In addition, the personal tax would retain elements of an income base for wealthy individuals, and higher MTRs would exacer- bate the economic distortions of their savings, invest- ment and business behaviour.

The two principal candidates for replacing the GST would be the business transfer tax (BTT) and the direct consumption tax (DCT); both were contenders when the federal government reviewed GST alternatives in 1994.64 In contrast to the destination-type GST, which is collected by the jurisdiction in which the good or service is consumed, both the BTT and DCT are origin- type taxes. That is, the tax is collected not on each transaction, but on the costs of production (in the case of the BTT) or on the sources to finance consumption (in the case of the DCT). Therefore, both these alterna- tive taxes could be applied even with a border that was open to the movement of consumer goods.

Moving from a destination-type to an origin-type tax would not affect the competitiveness of Canadian business because the exchange rate would adjust to compensate in much the same way as described earlier for differential tax levels.65

Both the BTT and the DCT would apply to the full range of consumption goods and services and would not allow for any exemptions, which would be more eco- nomically efficient than the GST. And because the base of either tax would be wider than that of the GST, the rate could be lower — a 3 percent DCT could replace the revenues of the 7 percent GST (Kesselman 1997, 329). The BTT could operate as a simple adjunct to corporate and unincorporated business income taxes — its base would be cash flow plus total compensation to labour — while the DCT could operate as an adjunct to personal and payroll taxes in the form of an employee payroll tax plus a cash flow tax on business. The BTT would be somewhat simpler to operate than the DCT, but it would also be a hidden tax for consumers, making the DCT preferable if one desires taxes to be visible for purposes of political accountability. In terms of the competitive- ness of the Canadian tax system, the DCT’s new emplo- yee payroll tax component would be counterbalanced by the US’s higher rates of payroll tax on workers.

The DCT would clearly thwart tax evasion on cross-border purchases because tax would be col- lected from a payroll tax on employee earnings and a cash flow tax on business. In effect, the tax would be prepaid, whether the net amounts were spent domestically, spent abroad or saved for future con- sumption. It is perhaps less apparent how the BTT would foil cross-border tax evasion. Prices of all consumer goods and services would rise in Canada to reflect the origin-type tax, but what about Canadians’ purchases abroad? In fact, because the shift to an origin-type tax would cause the Canadian dollar to depreciate, purchases made abroad out of Canadian-source earnings would also become proportionately costlier, thus removing the incentive for cross-border shopping.

If, instead, the federal GST were retained and provincial indirect taxes removed or reformed, the tax could still be improved in several ways:

  • The land component of building purchase prices should be removed from the tax base, as it is nei- ther newly created output nor value added. This change would reduce the tax burden on buyers in cities where housing prices are high on account of land prices, thus encouraging denser development.66

  • The tax should be removed from snack foods, baked goods and other nonrestaurant food items. Distinctions between, for example, salted and unsalted peanuts and between two cookies and half a dozen are not only arbitrary but silly, and they tarnish the tax’s public image.

  • Full rebates could be given for GST paid on purchas- es by entities in the MUSH sector (municipalities, universities, schools and hospitals) to eliminate the current inefficient bias toward in-house labour serv- ices versus contracted-out services.67

  • The application of GST to employee fringe bene- fits, especially cars, should be greatly simplified or eliminated.

  • The small trader exemption of $30,000 of annual sales, below which firms are not required to register for or collect GST, should be substantially increased to $75,000 or $100,000, in line with the levels typi- cally found in other countries’ value-added taxes. This would exclude from the GST most self- employed workers without employees, including most in the home repair and maintenance sector, where tax evasion is widespread. A higher exemp- tion would also reduce the high compliance costs small businesses face.68

Provincial Sales Taxes

Provincial retail sales taxes are likely better candidates for replacement than the GST if Canada wished to free up the border or pursue a customs union with the US, for several reasons. First, the provincial taxes bear heavily on business inputs (about 35 percent of the tax revenues), thus penalizing production efficiency and investment incentives. Second, they have a narrower base than does the federal GST, although some provinces have extended the coverage of services; broadening the base would improve efficiency and hor- izontal equity for consumers. Third, eliminating the provincial taxes would reduce the burden on business of having to deal with two levels of sales taxes with differing taxable bases and operating procedures.

As with the federal GST, the main alternatives for the provincial retail taxes are a BTT and a DCT. The direct consumption tax would be more appealing if govern- mental accountability through tax visibility is a priority, since most revenues of a DCT would stem from an employee payroll tax. But there are reasons to expect that the provinces would opt for the business transfer tax for- mat, since governments tend to prefer reduced tax visi- bility — indeed, most provinces have resisted harmonizing with the federal GST in part out of fear of adverse public reaction. Moreover, a BTT could be oper- ated as a straightforward adjunct to provincial corpo- rate and personal income taxes. The Canada Customs and Revenue Agency might agree to collect the BTT for participating provinces if they could settle on a uniform base and structure for the tax.

Although both are forms of value-added taxes, a BTT differs from the GST in being an origin-type rather than a destination-type tax, which means that the tax would be paid on exports but not on imports.69 If the GST were replaced at the national level by a BTT, this change would induce an exchange-rate adjustment to restore trade competi- tiveness, with the tax newly built into export prices. For any single province, which has no exchange rate to mediate changes in its taxation regime, the shift to a BTT might be expected to reduce competitiveness — unless all the other provinces also changed regimes. Yet, at present, the one-third of retail sales taxes borne by business is already built into the prices of goods and services sold to out-of-province and for- eign customers. Hence, any competitive disadvantage for that province would be limited.

The use of a BTT by the provinces would raise a number of practical implementation issues,70 includ- ing the method for allocating tax across provinces for firms operating in more than one and whether the BTT should be deductible from business income taxes at the federal and provincial levels. Solutions to such issues should follow from negotiation among the governments. Because of its simple operation and economic efficiency, a provincial BTT would also offer a convenient vehicle for replacing or reducing some other provincial taxes. Leading candidates include the corporate income tax, provincial payroll taxes and part of the nonresidential property tax (as discussed later in this paper). A BTT might also be offered as a base for municipalities to impose an add- on to the provincial tax rate.

Quebec, New Brunswick, Nova Scotia and Newfoundland and Labrador have harmonized their sales taxes with the federal GST. In doing so, they have already eliminated most of the problems of the remaining provincial retail sales taxes. If Canada freed up the border or formed a customs union with the US, these “harmonized” provinces would never- theless face pressure to convert their sales taxes from the GST format to a BTT suggested for the other provinces. Otherwise, their high total rates of sales tax on purchases relative to those in the rest of Canada and in US states might induce unacceptable levels of cross-border shopping, smuggling and tax avoidance. If all provinces (other than Alberta, which does not have a sales tax) converted their sales taxes to a BTT, the exchange rate would adjust to offset any competitive penalty of shifting to an origin-type tax. Conversely, if Canada were willing to foreclose future moves toward a freer border or a customs union, harmonizing sales taxes with the GST could be an economically attractive option for the provinces that have not already done so.

Developments in the US might reduce the pressure on Canada to curtail its reliance on indirect sales taxes. In 1992, the US Supreme Court ruled that retail- ers selling for delivery to customers in another state did not have to collect the sales tax of the destination state unless the vendor had a “physical presence” (such as a warehouse or retail outlet) in that state. As a result, most mail-order and e-commerce vendors do not collect sales tax on their out-of-state sales (including sales to Canadian customers). In addition to the 45 states that impose a retail sales tax, US cities and counties also may have a sales tax, with the result that there are about 7,500 distinct sales tax jurisdic- tions in the US. Within a given state, the sales tax can vary by location with respect not only to the total rate but also to the taxable base. The US Supreme Court cited this complexity in ruling that the states had to simplify their sales tax laws before requiring out-of- state vendors to collect and remit taxes. In response, in 2002 the US National Governors Association approved an agreement called the Streamlined Sales Tax Project (SSTP),71 which requires participating states to simplify tax computation for out-of-state vendors. Full implementation of the SSTP by US states would facilitate Canadian governments’ collection of their sales taxes from US vendors on items for deliv- ery to Canada, but the existing Canada-US gap in sales tax rates would still leave a big incentive for cross-border shopping by Canadians.

If the provinces choose not to shift to the GST or BTT format, they should still reduce the inefficient coverage of business inputs in their sales taxes. In fact, over the past decade, some provinces have expanded the coverage of their retail sales taxes in ways that have aggravated this problem.72 Although the single- stage method of applying the retail sales tax makes it inherently difficult to distinguish between business and final sales, steps could still be taken to remove the tax from such business inputs as legal services, office equipment and supplies, motor fuels and machinery, though it might require rebate provisions in some cases. For the building of business and industrial struc- tures, provinces could rebate the sales tax paid on the materials component, and they could extend their sales taxes to include the construction of residential proper- ties. Provinces could also expand coverage to con- sumer items such as home heating fuels, local telephone and basic cable television.73

Excise Taxes

At the federal level, excise taxes represent a compa- rable share of total tax revenues in Canada and the US and are only about one-sixth larger in Canada as a percentage of GDP. Hence, differences in federal excises are not likely to present a problem for Canada if the two countries pursued a customs union or a relaxed border with respect to trade. At the subna- tional level, though, excises as a percent of total tax revenues are about one-third larger in Canada than in the US, and relative to GDP they are about 60 percent larger in Canada (equivalent to an extra 1 percentage point of GDP). That differential might pose con- straints on provincial tax policy.

Although excise tax rates differ considerably across provinces and states, excise revenues in both countries derive almost entirely from the same three goods: gasoline, alcohol and tobacco products. Accordingly, some unique attributes of these products and how their taxes are applied need to be consid- ered. Also, one must distinguish among three types of cross-jurisdictional purchases: individual purchases taken home for consumption, individual orders by Web or phone delivered by mail or courier and legal commercial transactions by firms for resale. The fur- ther opening of the border would not affect the need for ongoing enforcement of taxes with respect to commercial transactions, nor would it present any new issues at the policy level. However, with less scrutiny at the border, commercial-scale smuggling would become more attractive.

Alcohol and tobacco products offer large tax sav- ings relative to the costs and ease of moving them to lower-taxed states or provinces. Hence, they are sub- ject to severe restrictions on purchases by individuals transporting or shipping them to consume at home. In the US, the Jenkins Act requires firms shipping cigarettes to out-of-state customers to notify the tax department of the purchaser’s home state to help it to collect excise taxes.74 As for alcohol, the US constitu- tion prohibits the cross-state movement of liquor (both commercial and private) except under policies set by each state. These policies are complex and vary by state: 13 have entered “reciprocity” agreements that allow the interstate shipment of wine without the need to pay excise taxes to the destination state; the other states have more restrictive terms that require payment of excises or that prohibit retail shipments entirely. Similarly, in Canada, the provinces require payment of excises in the destination province for shipments of liquor, while interprovincial mail-order sales of ciga- rettes have been outlawed since 1996.

Canadian federal and provincial excises on alcohol and tobacco products have already been constrained by cross-border smuggling.75 Such activity reached its peak in early 1994, when an estimated 40 percent of all cig- arette sales in Canada were contraband smuggled from the US (but mostly manufactured in Canada). Sharp, coordinated cuts in cigarette excises by the federal and Quebec governments, followed by Ontario and most of the Atlantic provinces, put Canadian rates much more in line with US rates. The western provinces and Newfoundland and Labrador have not followed these cuts, likely because pressure from cross-country smug- gling has been controlled by spot checks of vehicles and truck weigh stations. Similar controls might also be effective even with a more open border, although one might expect downward pressure on alcohol and tobacco excises in provinces with large populations near the US border.

Gasoline offers contrasting characteristics for excise taxation, because it is not economical or safe for indi- viduals to transport much gas beyond that in their vehicle’s tank. Thus, no province or state attempts to enforce the payment of its excise taxes on gas brought into the jurisdiction in personal vehicles. Moreover, individuals returning to Canada with a full tank in their car are not subject to Canadian taxes, and cus- toms officials do not even inquire. Clearly, gasoline is not a product that the consumer can order by phone or Internet for delivery in another jurisdiction. As a result of these physical limits on the movement of gas, one can contemplate higher excise tax rates since they are constrained only by personal cross-border purchases.

Much higher rates of excise tax on gasoline — at the federal or provincial level — would be an economically attractive way to replace revenues lost when other taxes are reduced. It is efficient to place high tax rates on items such as gasoline that have demands that are relatively unresponsive to price. This change would also give consumers an incentive to conserve by shift- ing to smaller, more fuel-efficient vehicles; to move closer to their workplaces, to carpool and to use public transit; and to drive less, thereby reducing traffic con- gestion, accident rates, air pollution and global warming.76 EU countries and Japan have garnered these benefits with steep tax rates that make their retail gas prices typically at least double and up to triple Canadian prices.

For example, doubling the federal excise tax on gas from the current 10 cents per litre would raise an extra $4 billion per year.77 Trebling the tax rate would add 20 cents per litre to the retail price, generating $8 billion annually, and still leave prices far below those of most other industrialized countries. The enlarged gap in gas prices vis-à-vis those in the US would cre- ate some revenue loss but not make a major dent in the total revenue gain. Although there are strong eco- nomic and environmental grounds for pursuing such a tax shift, it would require astute political strategy to sell to the public. Higher excise tax rates on gas would best be phased in over several years, perhaps timed with periods of falling energy prices, and linked clearly to reductions in the rates of other visible taxes.

Business Taxes

Taxes on business are of disproportionate importance, relative to their revenues, for the tax strategy of an aspiring Northern Tiger. These taxes consist of corporate income taxes and corporate capital taxes, as well as sales, excise and property taxes that impinge on businesses. Taxes on capital income and business capital carry among the highest efficiency costs per dollar of revenue generated of any taxes. Hence, significant opportunities for increased growth of investment, productivity, employment and real wages could be secured through appropriate reform of business taxation. In this section, I first consider business taxes on large corporations; I then turn to special issues that arise for business taxation at the provincial level and for smaller businesses.79

Before focusing on specific types of business tax policies, a summary of the basic economic findings on corporate and business taxes is useful. First, the high international mobility of capital, especially for a relatively small economy such as Canada’s, means that the burden of heavy business taxes falls more on domestic investment than on domestic shareholders, and foreign shareholders are even further insulated from the tax. To some extent, the operation of foreign tax credits means that the Canadian treasury can benefit from corporate taxes that might otherwise flow into foreign treasuries. But, in an increasingly capital-mobile world, Canada must be able to com- pete for capital with source countries that have sig- nificantly lower business taxes.

Second, it is essential to distinguish between two types of business tax rates. One is the statutory rate of tax on corporate profits from both federal and provincial income taxes plus the income-tax equiva- lent burden of capital taxes. This total statutory rate affects firms’ incentive to shift their accounting prof- its into or out of Canada through devices such as transfer pricing on intrafirm transactions, and where they situate their debt financing — whether domesti- cally or with a foreign affiliate or parent (see, for example, Clausing 2003; Bartelsman and Beetsma 2003). The other key business tax rate is the effective tax rate (ETR) on capital investment, which reflects the impact of all business taxes on the returns to incremental investment (see Chen 2000). The ETR is affected both by statutory rates and by tax provisions such as depreciation rates and inventory accounting methods; it is a critical factor in incentives for invest- ment in any industry or country.

Business Taxes on Large Corporations

The 1998 Report of the Technical Committee on Business Taxation (Canada 1998), known as the Mintz report after the committee’s chairman, Jack M. Mintz, remains the most systematic and sagacious roadmap for reform of Canadian business taxes, particularly at the federal level. Although the committee’s mandate was to suggest business tax reforms that would main- tain total tax revenues from business, fiscal conditions since the report was released have permitted tax rev- enues from business to be reduced. Further business tax cuts and reforms should be a priority area for “spending” part of the federal budget surplus.

Among the report’s most important recommenda- tions were that the corporate tax rate be levelled across business sectors, that the total rate for larger corpora- tions be reduced to around 33 percent (20 percent fed- eral rate plus a typical provincial rate of 13 percent) and that the federal corporate surtax be removed. Policies either announced or already implemented have gone some distance toward meeting these goals. Ottawa has delivered in full on levelling tax rates across sectors: in 2004, the federal general corporate tax rate will be reduced to 21 percent for all sectors except resources, which will reach that figure in 2008. A federal surtax rate of 1.12 percent remains for large corporations; this surtax should be eliminated, as it was originally introduced to combat the deficit.

Progress on reducing tax rates for large corpora- tions has been more uneven at the provincial level. Quebec’s 9.3 percent rate and Alberta’s targeted 8 percent rate will put their total general corporate income tax rates into the 30 to 31.5 percent range — even lower than that advocated in the Mintz report. However, the new Liberal government in Ontario has reversed course on large corporate tax rate reductions and raised its rate from 12.5 percent in 2003 to 14 percent for 2004. Ontario would be wise to renew its earlier commitment to reduce its corporate tax rate to 8 percent, and the other provinces should follow suit.

Total tax rates on corporations have been further reduced by recent cuts to corporate capital taxes at both the federal and provincial levels, and those rates will fall still further with impending cuts. According to official calculations, capital taxes on nonfinancial corporations added 3.6 percentage points to the effec- tive corporate income tax rate in 2000 (Canada 2000, 159). Alberta, Newfoundland and Labrador and Prince Edward Island have never had a capital tax, while British Columbia abolished its capital tax in 2002. However, Quebec is delaying, and Ontario is perhaps shelving, earlier commitments to phase out or reduce capital taxes for nonfinancial corporations; accordingly, this item should be a high priority for action as soon as the fiscal resources of those two provinces permit.80 Capital taxes on financial corpora- tions by both levels of government also warrant reducing or eliminating.81

Another of the Mintz report’s recommendations was to trim capital cost allowances (CCAs) to rates that more closely approximate true economic depreci- ation rates, with a view to introducing greater neu- trality across types of capital assets and providing an expanded tax base with which to finance corporate tax rate reductions. This proposal, however, was made in the context of the committee’s mandate to maintain total tax revenues and warrants review now that the federal fiscal situation is far rosier than when the committee deliberated in 1996 and 1997. Moreover, subsequent changes to rates of capital write-offs in the US tax make it imperative to revisit this issue. Notably, the 2003 US tax package increased first-year “bonus” depreciation deductions from 30 percent to 50 percent for investments made prior to 2005;82 for small business, it raised the amount of investment that can be immediately deducted from US$25,000 to US$100,000.

The corporate tax rate cuts already planned by both levels of government will bring statutory rates in Canada below most US rates, but will still leave them above those of some other countries. Significantly, planned rate cuts will leave the ETR on corporate investment in Canada well above that on most sectors in the US (Chen and Mintz 2003) and in such countries as Ireland, Sweden, the UK and Italy (Wilson 2003). The ETR is particularly important for investment incentives, and the provisions for capital expensing and “bonus” depreciation in the US have a big impact on it.

Given the improved fiscal outlook, Canada should pursue a more ambitious corporate tax strategy than the Mintz report recommended. Two alternative directions warrant careful study followed by policy action. First, further cuts could be made to the statutory rates of cor- porate income tax to bring Canadian rates not only below US rates but also closer to rates that other coun- tries have successfully pursued. The shining example of this scenario is Ireland, whose stunning rate of econo- mic growth since the 1980s has been explained, in part, by its very low corporate tax rates. Canada would not need tax rates as low as Ireland’s to become the pre- ferred location for businesses wishing to serve the North American market; in any case, it is doubtful whether Canada could pursue such low rates without some repercussions from the US, just as the EU has pressured Ireland to raise its rates.83

The second approach would be to move Canadian cor- porate taxes from an income base toward a cash flow base.84 A cash flow tax would be the most economically efficient way to tax business, as it would remove the nor- mal rate of return to capital from the tax base, and the ETR would become zero (apart from the impact of other taxes on investment such as sales and property taxes).85 Business taxes would then be collected only on “economic” or “supernormal” profits. With a cash flow base, all capi- tal purchases would be fully expensed, thus eliminating all problems in measuring tax depreciation. Since the capital cost would be fully recognized for tax at the time of purchase, no deductions would be allowed for interest costs. Such a change would yield a balanced tax treat- ment of the debt and equity costs of corporate finance.

Of course, using a cash flow base would reduce total revenues from the corporate tax unless higher statutory rates were applied. One would thus have to examine whether the gains in economic activity and productivity — which would yield at least partially offsetting revenues from other taxes — could justify the switch to a cash flow base. Key issues in such a reform are whether other coun- tries would allow foreign tax credits with respect to a corporate tax of this format and what transitional provi- sions would be needed for businesses with large out- standing debt and interest costs when the new base was implemented. Even if a cash flow base were not adopted for corporations, it might be allowed for smaller unincor- porated businesses, both to simplify their accounting and to give them some tax relief without incorporating.86

An intermediate strategy between significant rate cuts in the corporate tax and moving to a cash flow base would be to combine further modest cuts in the general corporate tax rate with aggressive changes to accelerate capital cost allowances. Accelerated CCAs would fall short of full expensing, and some of the transition problems of a full cash flow tax could be avoided. One potential hazard of this approach, how- ever, is that biases might be introduced between short- and long-lived capital and across capital types such as machinery and equipment, structures and inventories. In addition, such an approach would not eliminate the deductibility of corporate financing costs, as under a cash flow tax, so variants with lim- ited interest deductibility would be worth exploring.87

There are economic reasons for preferring a tilt toward cash flow taxation or some form of accelerated CCA to further sharp cuts in statutory rates of federal corporate tax. So long as total statutory rates in Canada are below those in the US, they limit incentives for corporations to engage in transfer pricing and income shifting, although some such incentives would remain vis-à-vis other countries. A cut in the corporate income tax rate would reduce the tax on the return to all capital — both “old” or pre-existing capital and capital newly created after the tax cut. The revenue cost of this cut, therefore, would be relatively large and dispersed. In contrast, increasing CCA rates only on newly created capital would focus the revenue cost on incremental investment. This goal could be achieved through an investment tax credit or by adopting the US method of “bonus” depreciation, which applies only to new investments in their first year.

Two other recommendations of the Mintz report war- rant brief comment here. First, the report suggested that a new corporate distributions tax would be a better way to deal with the double taxation of dividends and at less revenue cost than the existing dividend tax credit. This approach might be superior to adjusting the dividend tax credit for neutrality of corporate distributions. Second, the report recommended denying federal corporate tax deductibility for provincial capital taxes. Phased applica- tion of this change might induce the provinces to pro- ceed more quickly in cutting their capital taxes.

Provincial Business Taxes

In 2001 corporate income taxes levied by the provinces accounted for 3.6 percent of Canada’s total tax mix compared with 1.2 percent for US states.88 As a percentage of GDP, subnational corporate taxes were nearly four times as large in Canada as in the US. Since 2001, provincial corporate income taxes have been declining as the larger provinces institu- ted major rate cuts (although the new Liberal gov- ernment in Ontario has announced a course reversal for 2004). Further rate cuts by the provinces would be desirable, and any future acceleration of deprecia- tion in the federal corporate tax will be mirrored in the taxes of provinces whose corporate tax falls under federal-provincial tax collection agreements. Alberta, Ontario and Quebec, which collect their own corporate income taxes, would be well advised to expedite CCAs similarly.

Beyond further incremental changes, there is room for major improvements in the structure and effic- iency of provincial corporate income and other busi- ness taxes. Moving to a business transfer tax would be an attractive way to replace provincial corporate income tax revenues, just as it would be to replace their retail sales taxes. A BTT or other form of accounts-based value-added tax could collect the same revenue as current corporate income taxes at much lower rates of tax, since the BTT base is equiv- alent to total labour compensation plus business cash flow,89 and labour costs are typically several times larger than a firm’s cash flow. In effect, a BTT would shift much of the tax burden on capital income to labour income, consistent with the efficiency princi- ples described earlier. A BTT has all the economic advantages of a value-added-type tax, in that it is neutral with respect to the production, investment and financing decisions of firms. Moreover, a BTT is much less costly for firms to comply with and for governments to administer, an advantage over a credit-invoice type of value-added tax such as the GST, which is applied to each transaction.90

In addition, a BTT could be used to subsume provincial general payroll taxes and part of the bur- den now borne by business through high rates of nonresidential property tax. In short, a BTT could serve as a multipurpose, major tax at the provincial level. Even with relatively high tax rates, a BTT would still be neutral and conducive to unbiased business decisions. The provinces that would need the highest rates of BTT to replace their sales, corporate income, payroll and part of property taxes would be the four that currently apply a general payroll tax — Ontario, Quebec, Manitoba, and Newfoundland and Labrador.91 A BTT could be operated as an adjunct to the federal corporate income tax and, for unincorpo- rated businesses above a threshold, as an adjunct to the personal tax. A general payroll tax could also be applied to financial institutions and other sectors for which a BTT would not be well suited.

Bird and Mintz’s paper (2000) — the most exten- sive analysis of replacing selected provincial taxes with a value-added tax — advocates the use of an income-type measure of value-added called a “busi- ness value tax” (BVT). That tax has a base that is similar to the tax definition of net income — but without allowing labour costs to be deducted — so that it would continue to deal with the cost of capital through deductions for depreciation. This approach con- trasts with the expensing of capital allowed by a BTT (or by the GST, implicitly), which is a consumption-type measure of value added. Both a BTT and a BVT would work on the accounts of business and would be much simpler to operate than a transactions-type tax. On economic principles, however, a BTT’s consumption- type base appears to be superior to the income-type base of a BVT.92 Moreover, a BVT would bear relatively more heavily on capital income, since it would strike the normal return to capital, while a BTT would bear more heavily on labour income, since it would tax only supernormal returns.

Taxes on Small Business

The Mintz report noted that “Canada’s income tax treatment of small business…[is] among the most generous in the world” (Canada 1998a, 5.7). Incorporated smaller businesses, through a small business deduction, obtain a preferential tax rate on a limited amount of active business income. This deduction cost the federal government an estimated $3 billion in 2003, the largest of all the tax expen- diture items tracked for corporations (Canada 2003b, 26). The Mintz report recommended against raising the amount of income eligible for the lower tax rate and suggested only small cuts to corporate tax rates for small business.93 The report further suggested that the differential taxation of large and small cor- porations be narrowed mainly by lowering the gen- eral corporate tax rate on large business. It argued that this approach would reduce existing economic distortions in the tax system, as well as incentives for tax planning associated with the small business deduction.

Since the Mintz report was released, the provinces have been actively competing for boasting rights to the lowest small business tax rates. In 1998, the average provincial rate was 8 percent; by 2003, all provinces except Quebec and Prince Edward Island had rates of 6 percent or lower, and only Quebec applies the same corporate tax rate, 9.3 percent, to both small and large business. In 1998, the upper income limit for the small business deduction at the federal and provincial levels was $200,000. Since then, all provinces from Ontario westward have raised their limit to at least $300,000, Ontario has committed to raising its limit to $400,000 and the 2003 federal budget announced that Ottawa would gradually follow suit over several years.

Although small business has the reputation of being the economy’s wellspring of job creation and new prod- ucts, this image may be overstated.94 In fact, there is extensive job loss as well as job creation among small and medium-sized enterprises (SMEs), and the sector’s contribution to overall net employment growth is unre- markable. Moreover, relative to large corporations, SMEs typically lag in their research and development, productivity growth, export penetration and employee compensation (both cash and benefits). Unless small business activity creates special external benefits, there appears to be a strong economic case for significantly reducing the favourable tax treatment of small firms relative to that for large business,95 which would result in more neutral treatment and a more level playing field for the efficient allocation of resources. Such a course of action would, of course, be politically challenging — small business ranks alongside farmers, seniors and stu- dents for sympathetic treatment that often leads to poor public policy — but Canada needs bold action if it is to become a Northern Tiger.

Despite inevitable opposition, the corporate tax rate differential between SMEs and large corporations should be reduced. If SMEs can claim deductions worth $3 billion per year (plus corresponding provin- cial revenue losses), the implication is that large busi- ness is taxed more heavily to keep business tax revenues constant. The probable consequence of this arrangement is a negative net impact on research and development, exports and the creation of high- quality, well-compensated jobs. I do not offer specific recommendations to remedy this major bias other than to suggest that the differential tax advantages for SMEs be narrowed by, for example, leaving the corporate surtax on small business while eliminating it for large business. Shifting corporate tax to a BTT at the provincial level would also offer a way to reduce the tax differential; an exemption or regis- tration threshold could also be offered for a BTT, but this would relate to a base that included labour compensation as well as cash flows.

The Mintz report also proposed replacing the life- time capital gains exemption for small business and farms with increased access to RRSPs. This proposal remains valid, although there is less reason to offer special treatment now that capital gains tax rates have been sharply cut. Eliminating the lifetime exemption would also simplify the personal tax sys- tem. An enhanced dividend tax credit, lower rates of personal tax, accelerated depreciation and expanded access to tax-recognized savings, all of which are recommended in this paper, would provide further reasons to terminate or replace the lifetime capital gains exemption.

Tax Design for a Northern Tiger

If Canada aspires to become a Northern Tiger, the goals for tax policy are clear. In an increasingly integrated North American and global economy, Canadian tax policies must operate to minimize obstacles to specialization, investment and trade and to reduce the effect of the border. These policies must seek to create areas of competitive advantage in nurturing and retaining skilled people and busi- ness investment. Indeed, Canadian tax policies must exploit the strong complementarities between skilled people and the creation and transfer of advanced technology and managerial practices. Those policies must further work to create unique attractions to situate and conduct business on the Canadian side of the border.

These goals require a comprehensive approach to tax design, one that reduces taxes in areas that will have a strong positive impact on growth and that shifts the tax mix to raise revenues in ways that minimize the nega- tive impact on growth. Building on reforms of federal and provincial tax policies in recent years, Canada should pursue a judicious mixture of consumption, labour income and efficient business tax bases, struc- tured so as to limit their natural tendency to reduce tax progressivity. Simultaneously, changes should improve horizontal equity and simplify the tax system, thereby reducing administrative and compliance burdens.

Table 9 summarizes the key elements of tax design suggested by the analysis in this paper. Some of these changes would transform tax bases (such as a cash flow tax or a business transfer tax) or introduce a new tax (a federal general payroll tax). Changes of this magnitude would be difficult to sell to policy makers and politicians fearful of resistance from taxpayers, and they would be rightly cautious in view of uncer- tainties about the distributional burdens, operational matters and unintended effects of new or transformed taxes. Thus, there would be a natural preference for smaller, incremental changes that offer the prospect of achieving at least some of the desired goals, even if they fall short of achieving the benefits of more sweeping reforms.

Implementing all the changes proposed in this paper would absorb large amounts of revenues, likely more than policy makers or politicians would other- wise entertain over the next several years. A federal general payroll tax and higher federal and provincial excises on gasoline could offset the lost revenues if policy makers wished to proceed quickly with major transformative tax changes.

If policy makers preferred a more gradual course of action, then some prioritization of tax policy measures based on revenue costs would be needed. The highest medium-term priorities include the following steps:

Federal Tax Policies

  • Expedite the scheduled elimination of the corpo- rate capital tax on nonfinancial entities, the rev- enue cost of which has already been built into budgetary projections; consider the phase-out of capital tax on financial institutions.

  • Proceed with the scheduled reductions in corporate income tax rates for the resources sector, which have already been built into budgetary projections, remove the surtax from large corporations (but not from small businesses) and halt the scheduled increase in the small business deduction.

  • Sharply accelerate capital cost allowances on new investment for business (both corporate and unin- corporated), and make this a priority over further cuts in federal corporate income tax rates (beyond removal of the surtax).

  • Reduce EI premium rates to cover the cost of pro- gram benefits over the business cycle and move toward experience rating the premiums or, as an interim step, make them industry rated.

  • Introduce tax-prepaid savings plans, with contribu- tion limits linked to earned income but without a dollar ceiling; the revenue costs would be minimal at the outset and would grow gradually over time.

Table 9 Summary of Major Components of a New Tax Design

  • Reconfigure the dividend tax credit so as to restore neutrality vis-à-vis the effective tax rate on the capital gains of large corporations; intro- duce a $2,000 exemption per taxpayer for inter- est and dividend income from nonregistered accounts.

  • Expand PIT tax rate brackets substantially, particu- larly to provide a wider income range in the bottom bracket to allow reduced clawback rates for National Child Benefit supplements and to raise sharply the income levels at which the upper tax rates begin.

  • Halt or reverse enrichment of NCB supplements and devise ways to provide in-kind benefits targeted to children in need but not tied to family income.

  • Raise the taxable threshold for personal tax by about 25 percent to $10,000.

  • Cut marginal tax rates for the two middle brackets by 2 percentage points each and cut the top-bracket rate by 1 percentage point; this would yield a four-bracket system with rates of 16, 20, 24 and 28 percent.

Provincial Tax Policies

  • Alberta should proceed with scheduled corporate income tax rate cuts; Ontario should make good on the previous Progressive Conservative govern- ment’s promise of scheduled corporate income tax rate cuts toward an 8 percent target; and British Columbia should deliver on its promise to adjust its corporate income tax rates to remain competi- tive with other provinces — indeed, all provinces should make similar moves.

  • Halt or reverse increases in the income limits for the small business deduction.

  • Alberta, Ontario and Quebec, which collect their own corporate income taxes outside the federal tax collection agreement, should match the recom- mended sharp acceleration of capital cost allowances for federal taxes.

  • Provinces that operate a retail sales tax should move vigorously to reduce the burden of the tax on the intermediate and capital inputs of business, whether by reforming their sales taxes or by har- monizing them with the federal GST, and to broaden the coverage of consumption items.

  • Increase sharply the income thresholds at which the top provincial personal tax rates apply, to match the top federal rate threshold and any hikes to it.

  • Reduce top marginal rates of personal tax and fur- ther flatten rate schedules, particularly in Quebec, Ontario and the Atlantic provinces, so that top marginal tax rates do not exceed 15 percent.

The proposed changes in federal and provincial per- sonal tax rates and brackets would improve incentives for productive behaviour by workers, entrepreneurs and investors across the income scale. The largest cuts in marginal tax rates would be in ranges where the economic payoffs are largest — for highly skilled and talented workers who now face unduly heavy burdens on their efforts. For example, in 2003 Ontario workers with a taxable income above $67,300 face a total MTR of 43.4 percent; those with an income above $104,600 face an MTR of 46.4 percent. My proposal, when fully implemented, would reduce the MTR for the first group by about 10 percentage points and for incomes between $104,600 and $160,000 by 7.4 percentage points.96 For those with incomes above $160,000, the MTR would decline by 3.4 percentage points to a top rate of 43 percent. Moderate earners with children would also receive significant cuts in their effective MTRs as a result of tax changes and reforms to the National Child Benefit supplement phase-out.

Pursuing these near-term tax changes would help to make the Canadian economy more efficient, pro- ductive and prosperous. Many of these changes are needed simply to keep Canada in line with what is available for businesses and workers across the bor- der. To slash border costs and advance economic integration with the US, however, more dramatic changes are required. Most useful would be a large reduction in Canada’s reliance on indirect forms of consumption taxes. The GST could — as was proposed in the 1994 GST review — be converted to an origin- type business transfer tax or a direct consumption tax, either of which would also bring major savings in administrative and compliance costs. Alternatively, provincial sales taxes could be converted into origin- type business transfer taxes, a format that could con- veniently subsume other inefficient provincial and local taxes. Excise tax rates, particularly provincial rates on alcohol and tobacco, would also be under pressure from a border that was much more open.

To achieve the even more critical goal of attract- ing and expanding business on the Canadian side of the border, the business taxation system should be transformed. Canada should follow the path of Ireland, the Celtic Tiger, by pursuing much lower effective corporate tax rates than those contained in current official plans. If the results for Canada were anything like those achieved in Ireland, the potential payoff in terms of investment, productivity and eco- nomic growth could be huge.97 An even better way to achieve this goal might be to move dramatically away from business income taxation and toward cash flow taxation. Another candidate to replace corporate income taxes, perhaps more suitable at the provincial level, is a business transfer tax. Research into the economic, operational and legal aspects of these approaches to business tax reform should be a priority — and the optimal design should be implemented.

The changes to the design of Canadian taxation that I have suggested here may raise concerns about their adverse effects on the overall progressivity or vertical equity of the tax system. Indeed, changes to the taxation of capital income at the personal and corporate levels would, at least initially, reduce effec- tive progressivity. Yet, much more vital for house- holds with low and moderate incomes than big tax cuts for themselves is the adequacy of aggregate tax revenues to sustain and expand public services and benefits, on which such households rely dispropor- tionately.98 A revitalized economy, facilitated in part by an efficient and growth-enhancing tax system, offers these income groups the best long-run prospects. They would benefit not only from better- paying work opportunities, but also from the strengthening of Canada’s social safety net that a rapidly growing economy can finance.

Moreover, the proposed reforms would shift only incrementally toward a tax base of consumption, labour income and cash flows. There would be no large windfall gains to current wealth holders and no resulting massive revenue losses. By making new investment more attractive, these changes would actually depress the returns to current holders of wealth and “old” capital. The distributional shift would be far less severe than under various propos- als for a flat tax scheme (see Kesselman 2000). A revitalized economy would benefit high-skilled workers as well as less-skilled workers through more and better-paying jobs.

In redesigning Canadian taxes, it is imperative to remember that the tax system is a tool, not a goal in itself. A well-designed tax system would generate higher levels of real income, which Canadians could allocate as they chose between more private con- sumption and expanded public services and benefits. So long as public expenditures were efficiently con- figured and responsive to public needs, they would raise real living standards despite their associated tax burdens. Hence, it is essential that the tax system be designed as a tool to raise revenues with minimal hindrance to the working of the economy. An efficient and growth-oriented tax system would thus play a central role in enhancing Canada’s policy autonomy. Only a lack of vision and resolve can keep Canada from instituting a world-beating tax system, one that could help turn this country’s economy into a Northern Tiger.

The author thanks Bev Dahlby, Don Drummond, Finn Poschmann, Bill Robson, France St-Hilaire, Daniel Schwanen and David Stewart-Patterson for their exceed- ingly helpful comments. The support of the SSHRC through the MCRI project on “Equality/Security/Community” is gratefully acknowledged. All findings and any errors are those of the author alone. This study reflects developments through December 2003.

  1. Simon Tuck, “Tax Cuts Next for Manley,” Globe and Mail, February 21, 2003, A1. Based on a public speech and a meeting with the Globe and Mail’s editorial board, the finance minister was reported to be aiming for lower tax rates than those in the US and an increased threshold for the top income tax bracket.

  2. For early precedents in the analysis of and recommen- dations for moving the Canadian tax system toward consumption bases, see Boadway et al. (1987); Davies and St-Hilaire (1987); and Economic Council of Canada (1987).

  3. Mintz suggests “a sharp increase in sales tax revenues (sales and excise) to reduce income taxes” (2003, 49). Grubel (2003b) proposes hikes in the federal GST and provincial sales tax rates. Both authors recognize, how- ever, that the shift toward consumption taxation can be achieved through direct personal taxes. See also Eric Beauchesne, “Raise GST to 10 Percent: TD Bank: Income Taxes, Not Sales Taxes Drive Away Workers, Expert Says,” Ottawa Citizen, November 22, 2001, D1. Dahlby (2000) and McKenzie (2000a) estimate signifi- cant efficiency and growth benefits for Alberta in replacing its provincial income tax with a GST-type sales tax; note that this reflects the gains of moving from a partial income base to a consumption base, not solely the shift between direct and indirect tax formats.

  4. Some commentators from the business community, such as d’Aquino and Stewart-Patterson (2001), are surprisingly muted on this point (see, for example, the quotation that opens this paper). Mintz (2001a) argues for reduced Canadian taxes, but mainly on the grounds that marginal public expenditures produce little of value relative to the economic costs of the taxes needed to finance them.

  5. Dobson (2002) assesses the requisites and conse- quences of further integration between the economies of Canada and the US under different strategies; Goldfarb (2003) focuses on issues related to the poten- tial formation of a customs union (but does not address the taxation dimension).

  6. Many of the elements examined here have been sup- ported in previous research by various authors; for example, see Brown (2000); Dahlby (2003); Duclos and Gingras (2000); Fortin (2000); Kesselman (1999, 2000); Mintz (2000, 2003); and Mintz and Poschmann (1999). Kent (2003) recommends a curious combination of a comprehensive income tax and a personal consump- tion tax, both applied at progressive rates with a top total federal marginal rate of 51 percent (up from the current 29 percent).

  7. The most notable relief for poor families with children was the conversion of the Child Tax Benefit into the National Child Benefit and its subsequent enrichment.

  8. See Dahlby (forthcoming) and Kesselman (forthcom- ing) for detailed assessments of the tax policy aspects of the 2003 and earlier budgets.

  9. The initial changes will increase revenues by $760 mil- lion over a full year and include the elimination or reduc- tion of tax credits for “designated sites”; accelerated depreciation in manufacturing; and general tax credits, deductions and holidays for businesses (Quebec 2003).

  10. Under the tax-on-tax system, each province specified its personal tax rate as a percentage of the individual’s federal tax liability; the tax-on-income system allows each province to specify a full schedule of marginal rates at various income levels. Quebec has always set its own tax rate schedule, as it does not participate in the federal-provincial tax collection agreement.

  11. Still, the distribution of the tax burden vis-à-vis public benefits across the income spectrum may matter. If the fiscal system drives away “star” workers, it could reduce business competitiveness.

  12. For example, assume that half of Canada’s differential tax burden vis-à-vis that of the US (6 percent of GDP) reflects “wasteful” public spending in which, by con- trast, US governments do not engage. That 3 percent of GDP, if fully borne by business, would cause a depreciation of about 2 US cents per Canadian dollar. See Mintz (2001a, 101) for the exact formula.

  13. For further analysis along these lines in the context of an alleged “brain drain,” see Kesselman (2001).

  14. In preliminary statistics for 2002, Canada’s total taxes as a share of GDP fell by a sharp 1.6 points from 35.1 percent to 33.5 percent. All of the decline was accounted for by changes in personal and corporate income taxes. No preliminary 2002 statistics were pro- vided for the US. In a broader ranking of tax burdens with all 30 countries that are members of the Organisation for Economic Co-operation and Development (OECD 2003, 72), Canada placed below both the median and the mean, where the latter assumed a value of 36.9 percent on an unweighted basis in 2001. The top three countries and their respec- tive taxes, including social security levies, as a percent of GDP, were Sweden (51.4), Denmark (49.8) and Finland (46.1); the bottom three were Mexico (18.9), South Korea (27.2) and Japan (27.3).

  15. As the OECD analysis states, “Royalty payments for the right to extract oil and gas or…to exploit other mineral resources are normally regarded as non-tax revenues since they are property income from government- owned land or resources” (OECD 2003, 286).

  16. Although Canada applies capital gains tax on assets held at death (unless bequeathed to a surviving spouse), the US completely exempts these gains through a “step-up” in cost basis for such assets.

  17. “Tax Lunacy,” Financial Times, May 23, 2003.

  18. Many US state governments have been facing fiscal crises, with governors in 29 states responding with proposals for revenue increases in fiscal year 2003/2004. The revenue increases include rate hikes, credit cuts and base broadening for sales taxes (15 states); personal income taxes (10 states); corporate income taxes (11 states); and cigarette, tobacco and alcohol taxes (14 states). (National Governors Association and National Association of State Budget Officers 2003.)

  19. The Canadian figures are computed from Statistics Canada sources (three levels of government excluding the Quebec and Canada Pension Plans); the US figures are from the US Treasury Board and the Bureau of Economic Analysis. Canadian public debt-service charges as a percentage of GDP have been rapidly declining, reaching 4.3 percent in 2003.

  20. For a review of empirical research on the relationship between national tax burdens and economic growth rates, see Kesselman (2000).

  21. For a nontechnical presentation of similar material, see United States (2003a).

  22. OECD (1997, 85) reports on a study by the Department of Finance. Similar rankings of the tax bases by MEC but with different estimates are reported in other stud- ies, such as Jorgensen and Yun (1991); see also the lit- erature review in Kesselman (1997, 39-49). Some studies have found more dramatic differences between the MEC of capital income taxation and those of labour income and consumption taxation.

  23. For example, shifting one dollar of revenue from the corporate income tax to a tax on labour income would save $1.55 of efficiency and would cost $0.27 of effi- ciency, for a net efficiency gain of about $1.28.

  24. This gain is pure in the sense that all individuals could potentially be made better off to that aggregate extent; in practice, however, some groups would benefit more than others, and there may be losers as well. In concept, all losers could be compensated from the aggregate gains, but this would be unlikely to occur in reality.

  25. Cross-country statistical analyses support the theoreti- cal finding that heavier relative reliance on consump- tion and labour income taxes would raise the growth rate of real incomes. See, for example, Kneller et al. (1999) and the discussion in Kesselman (2000).

  26. However, the tax-prepaid method should not produce any lower expected tax revenues in aggregate, since the rate used to discount future taxes should be an average that includes both above- and below-normal rates of return.

  27. Canada offers an offset against these savings biases that excludes private business investment via the life- time capital gains exclusion of $500,000 for shares in small business corporations and farm properties.

  28. See Boadway et al. (1983); for an operational analysis, see Shome and Schutte (1993). For a related economic analysis that supports the efficiency of business cash flow taxation and that addresses the treatment of loss carry forwards, see Diewert (1988).

  29. Note that the increasing use of customs preclearance in some sectors, such as automotive parts, reduces congestion at the border but does not eliminate the real costs to the economy.

  30. These benefits would be further augmented by the adop- tion of uniform testing and labelling standards for Canada and the US, including the elimination of manda- tory dual-language labelling (which a province such as Quebec could retain if it wished).

  31. One could also consider legislative changes in both countries to facilitate the collection of sales taxes on Internet and mail-order sales based on the locale of the purchaser, but this would pose major problems. Such changes would not address cross-border purchases where delivery is taken in the other jurisdiction and the item is transported privately.

  32. However, a province that squandered its tax revenues might so depress its economy that prices of labour and land decline, which would offset some of the loss of competitiveness. Also, the Ontario economy is a large enough part of the entire Canadian economy that its policies alone might influence the exchange rate.

  33. McKenzie uses this line of analysis to conclude that “the use of the personal tax system as a redistributive mecha- nism should be confined largely to the federal govern- ment, and…provincial governments should not use the tax system to redistribute income” (2000b, 362).

  34. Ontario’s May 2003 Speech from the Throne promised a scheme of mortgage-interest tax deductibility, but this was not implemented.

  35. Author’s calculations from US Individual Tax Statistics, available at the US Internal Revenue Service Web site, http://www.irs.gov/taxstats/index.html

  36. This situation followed from the US Supreme Court’s landmark decision on taxation (McCulloch v. Maryland, 1819), which prohibited state governments from taxing the federal government and its agencies. Historically, this decision evolved to yield tax exemptions not only for state and local taxation of interest paid on US Treasury bonds, but also for federal taxation of interest paid on munis as well as state and local taxation of munis in the state of issuance.

  37. Derived by the author from United States (2002a, 89).

  38. Because of the federal-provincial tax collection agree- ment, Ontario cannot simply make the interest tax free but must rebate the Ontario personal taxes paid on the interest.

  39. One exception here is the US federal deductibility of state and local income taxes for filers who itemize. The result is to lower the total effective MTR, since incre- mental income creates federal deductions that partially offset the MTRs of the state and local taxes. A similar result would arise if the deduction were abolished and the associated revenue gains were used to trim statutory federal marginal tax rates.

  40. For more detail concerning these matters, see Kesselman (1999, 2000). That work further suggests converting some of the nonrefundable tax credits (for medical expenses and employee social insurance contributions) back to deductions, as they were before the tax reforms of the late 1980s. Of course, some of the suggested changes would face political resistance from groups such as insurers and unions, and the benefits would be mod- est relative to the other major changes recommended in this paper. The largest of the cited items — nontaxa- tion of employer-provided health benefits — is also mirrored in the US tax system.

  41. At higher income levels, personal taxes and social insurance premiums account for one-third or more of gross income, so that 15 percent of gross earnings equates to more than 22 percent of net earnings.

  42. See United States (2003b) for a description of the proposed TPSP-style Lifetime Savings Accounts and Retirement Savings Accounts. The Bush administration backed off from these proposals to secure passage of its 2003 tax package, but these proposals are being renewed in 2004.

  43. Examples include application of the US alternative minimum tax to prevent full use of foreign tax credits and, more recently, heavy paperwork burdens for US citizens with Canadian RRSPs and RPPs.

  44. PITs and payroll taxes on employees or employers are also applied by some cities, counties and school dis- tricts in the US, but due to their diversity these taxes are ignored in this analysis.

  45. Note that taxable brackets in the two countries are not exactly comparable, as the US offers relief via personal exemptions and a standard deduction, while Canada uses nonrefundable personal credits.

  46. The combined effects of federal and subnational PITs yield greater overall rate progressivity in the US than in Canada; see the review of findings in Kesselman and Chung (forthcoming).

  47. I refer later in this study to the effective tax rate (ETR) on capital; an analogous ETR on labour includes the impacts of personal taxes, payroll taxes (employer plus employee) and sales taxes. Mintz (2001a, 86) computes the ETR for labour and finds it significantly higher in Canada than in the US. Here, I stress reductions in the ETR on capital because of the relatively high efficiency costs of taxes on capital. The aim is to make Canadian marginal rates of personal tax on labour income more competitive with those in the US even if the Canadian ETRs on labour remain higher than the US ETRs.

  48. Note that the second and third bracket thresholds are slated to rise to $35,000 and $70,000, respectively, in 2004.

  49. Several analysts, including Poschmann and Richards (2000) and Lefebvre and Merrigan (2003), have noted and assessed this deficiency. Despite some possibilities for relief, there will be unavoidable incentive problems with enriching the NCB supplements. This suggests that future policies to address child poverty should rely much more heavily on in-kind programs that are delivered either universally or targeted at children or neighbourhoods with special needs, rather than by family income. See Kesselman (1994); Sayeed (1999).

  50. Both the basic and supplementary child benefits reduce average tax rates, but they do so in a way that also raises effective MTRs. Thus, they are examples of a tax reduction that increases inefficiencies and disincentives.

  51. In addition to the NCB, examples include the GST tax credits, the Guaranteed Income Supplement and clawbacks of both Old Age Security and EI benefits for higher earners, as well as provincial in-kind and cash benefit income-tested programs.

  52. It appears, however, that the US special tax rate for dividends will be extended to payments from foreign corporations, whereas the Canadian dividend tax credit is limited to payments from Canadian corporations. Additionally, the US federal tax rate on dividends and capital gains for taxpayers in the 10- and 15-percent tax brackets is being cut to 5 percent in 2003 through 2007 and to zero in 2008, something not matched by the Canadian provision.

  53. Rate flattening is gauged here by the ratio of the province’s top MTR to its lowest positive MTR. See Kesselman (2002, 911) for a table of figures.

  54. This comparison ignores the payroll taxes used to finance other programs, mainly workers’ compensation and EI. The Canada-US differential in premium rates for unemployment insurance does not offset much of the differential noted in the text. The employer payroll taxes that four provinces apply are discussed separate- ly in the section “Business Taxes.”

  55. Both the Canadian and US social security systems also contain some element of distorting general payroll taxes in that part of their revenues is needed to finance bene- fits for early retirees who did not pay their full costs.

  56. In addition, both countries have income-tested support programs for seniors — the Guaranteed Income Supplement in Canada and Supplemental Security Income in the US.

  57. Unlike the figure shown in table 7, this assumes that the employer’s portion of the tax is fully shifted to its employees, consistent with empirical findings.

  58. This burden has been partially limited through the tax- based OAS clawback, but it would have been limited more substantially by a Seniors Benefit, which was proposed in the 1996 federal budget and abandoned after much criticism and opposition.

  59. Several analysts have advocated removing the taxable ceiling on existing federal payroll taxes or creating a general payroll tax, especially an employee tax; see, for example, Robson (1996); Duclos and Gingras (2000); Fortin (2000); Mintz (2000); and Poschmann and Robson (2001). The proposal for a direct consump- tion tax, discussed in the next section, would contain a general payroll tax on employees.

  60. The pros and cons of labelling and earmarking of payroll taxes are examined in Kesselman (1997, 93-94, 174-76).

  61. Note that a customs union, unlike a common market, does not involve the free movement of labour across borders.

  62. To the extent that the coverage of services and utilities is broader in Canada’s indirect taxes than in those of the US, these figures overstate the problem. Most serv- ices and utilities must be purchased locally, where they are consumed, and thus cannot be easily substituted through cross-border purchases.

  63. It might be possible to obtain agreement among all the relevant governments to ensure that sales taxes of the destination jurisdiction were collected by all businesses, but this would still leave exposed individual cross-bor- der purchases; see further discussion in the section on provincial sales taxes.

  64. For detailed analysis of the BTT and DCT proposals, see Kesselman (1997, chap 8). Note that when the BTT was considered in Canada in the 1980s, it was ques- tioned whether the provinces could use that tax for- mat. A BTT might be construed legally as an “indirect” tax, and the Constitution restricts the provinces to “direct” taxation (the use of these terms in the legal context is different from that in the economic context).

  65. For a review of economic analysis of the trade compet- itiveness effect of destination-type versus origin-type taxes, see Kesselman (1997, 85-86).

  66. As a quid pro quo for this change, the GST rebate for new housing could be eliminated, thus adding another simplification.

  67. More than half of the cost of this change — currently about $900 million per year in total — would redound to the benefit of municipalities. Note that the MUSH sector does not have the same constitutional immunity from federal taxation as do the provinces.

  68. Some of these firms might still opt to register for GST in order to obtain input credits, at least in cases where their sales were primarily to other GST-registered busi- nesses rather than to final consumers.

  69. One of the principal putative benefits from introducing the GST was to get the federal sales tax off exports and thereby improve the country’s trade competitive- ness. As the earlier discussion of exchange rate adjust- ment indicates, this was a fallacious argument; the real deficiency of the manufacturers’ sales tax was that it applied at differential, distorting rates for various goods that were exported or imported.

  70. See Bird and Mintz (2000) for an analysis of these issues; see also the discussion below in the section “Provincial Business Taxes.”

  71. For details of the SSTP, see http://www.nga.org. As of March 27, 2003, 38 states had adopted the requisite legislation, and California had a bill pending approval. The National Governors Association cites a study that forecast total state revenue losses of US$45 billion from Internet sales by 2006. Greve (2003) criticizes the SSTP as a “tax cartelization” scheme, advocating instead that state sales taxes be applied on an origin basis, in part to spur competi- tion for lower tax rates.

  72. For example, British Columbia extended sales tax cover- age to include legal services and installation and repair of machinery, but in more recent years has made moves to reduce the coverage of business capital. Saskatchewan financed its PIT rate cuts in part by expanding sales tax coverage to include more business inputs.

  73. Enriched refundable sales tax credits could be used to offset the impact on the lowest-income households.

  74. Since violation of the Jenkins Act is deemed only a misdemeanor, Internet sales of cigarettes to out-of-state customers are growing rapidly (see United States 2002b).

  75. See Harrison (2003) for a historical account and insight- ful analysis of Canadian tobacco excise tax policy.

  76. The idea of a “green tax shift” has been promoted in works such as Durning and Bauman (1998), which takes an overly optimistic view of the potential for shifting tax revenues in this direction.

  77. The Sales Tax Division, Tax Policy Branch, Finance Canada, provided the author with an estimate that each 1 cent per litre increase in the excise tax on gas raises revenues by about $400 million annually. This estimate does not encompass behavioural responses by consumers to higher gas prices, the additional impact on GST rev- enues or the somewhat offsetting reduction of business taxes from deductible purchases of gas for business use.

  78. As shown in table 1, federal corporate income tax gener- ates only about one-eighth of total federal tax revenues; adding corporate capital and excise taxes on business still leaves this share at less than one-sixth. Although business pays more than half of total social security con- tributions, their burden is believed to fall mostly, if not entirely, on workers via lower compensation.

  79. I do not deal with the issue of income trusts, which are rapidly eroding the Canadian corporate tax base; see Hayward (2002); Aggarwal and Mintz (2003). However, my proposals for an enriched dividend tax credit, lower provincial corporate income tax rates and faster depreciation provisions would partially blunt incentives for the creation of such trusts.

  80. The Mintz report suggested that corporate income tax rate cuts should have priority over capital tax cuts in a revenue-constrained environment but stated that the capital tax “would appear to be more a matter of fiscal necessity rather than a permanent feature of a well- designed and integrated tax system” (Canada 1998, 4.21). There has been very little formal economic analysis of the corporate capital tax, but the analysis by Dahlby (2002) implies low rates of capital tax under certain con- ditions. For a detailed chronology and comparison of Canadian capital taxes, see Clemens et al. (2002).

  81. Dahlby (forthcoming) argues that the capital tax on banks entails undue economic efficiency costs. To date, the most detailed analysis of this issue, by Dancey (1998), was undertaken for the Task Force on the Future of the Canadian Financial Services Sector.

  82. The “bonus” depreciation (introduced in 2001 at a 30 percent rate) is allowed for capital with a life under 20 years in addition to regular depreciation in the year of asset purchase. An example is given by the John Deere Company for prospective purchasers of a $100,000 piece of equipment with five-year depreciation using 200 per- cent declining balance and a half-year convention; regu- lar depreciation in the first year would be just $20,000, but with the addition of “bonus” depreciation, the total becomes $60,000. See http://www.deere.com.

  83. Actually, the EU targeted the discriminatory nature of Ireland’s corporate tax, which initially favoured exports but after 1990 favoured profits from the manufacturing industry and internationally traded services (see Walsh 2003).

  84. Fortin (2000) advocates moving to a corporate cash flow tax; the Mintz report examined but rejected this approach for reasons of administration, implementation and international tax harmonization (Canada 1998, A.7-A.11).

  85. The elimination of provincial retail sales taxes on capi- tal inputs, noted in an earlier section, would reduce the ETR on capital investment by 3 percentage points (Chen and Mintz 2003, 5).

  86. This approach would parallel the US allowance for small businesses to expense up to US$100,000 of capital purchases annually.

  87. Limitations on interest deductions would additionally reduce incentives for the creation of income trusts.

  88. Nevertheless, Harden and Hoyt (2003) found that even US state corporate income taxes were set too high relative to sales taxes and personal income taxes if the states are concerned about employment levels; corpo- rate taxes were uniquely adverse to employment.

  89. This result can be seen by comparing the definitions for the business transfer tax and the business cash flow tax in table 2.
  90. Of course, if the federal government kept the GST in place, those operational costs would remain.

  91. This outcome can be seen in the tables of Bird and Mintz (2000, 282-83), which show Quebec needing nearly an 8 percent rate and British Columbia a rate less than 4 per- cent for a similar BVT to replace the revenues of corpo- rate income and capital taxes, payroll taxes and 5 percent of property tax. The requisite rates would be correspond- ingly higher if sales tax were added to this list.

  92. The only obvious advantage of a BVT over a BTT is that the former would more likely be granted foreign tax credits of other nations; see Bird and Mintz (2000); Bird and McKenzie (2001).

  93. It recommended a cut of just 1 percentage point on average for the provinces and an average cut of only 0.6 of a percentage point for the federal tax, the latter using a credit against the employer’s EI premiums to reward job creation (Canada 1998, 5.10).

  94. The Mintz report, after briefly reviewing the relevant evidence, gingerly concluded that “many recipients of the small business deduction provide little or no employment [for example, the self-employed], or have not grown despite the availability of the small business deduction” (Canada 1998, 5.10).

  95. Carroll et al. (2001) estimate, however, that decreased tax rates do stimulate the growth of small firms.

  96. Note that many of these MTR reductions are the conse- quence of workers being shifted into lower brackets, in addition to the reduction in the rates applying to each tax bracket. Recall also that the federal brackets are slated to expand in 2004 to $35,000, $70,000 and $113,800, respectively.

  97. Of course, Ireland also had other factors to its advan- tage, such as low initial wages, convergence with more prosperous economies and large subsidies from the EU. It also pursued ambitious expansion of advanced edu- cation and offered free tuition. See Walsh (2003).

  98. For a parallel message in the provincial context, in particular for British Columbia, see Kesselman (2002). 

Adams, Michael. 2003. Fire and Ice. Toronto: Penguin Canada.

Aggarwal, Lalit, and Jack Mintz. 2003. “Income Trusts and Shareholder Taxation: Getting It Right.” Toronto: University of Toronto, Capital Markets Institute. Draft.

Bartelsman, Eric J., and Roel M.W.J.Beetsma. 2003. “Why Pay More? Corporate Tax Avoidance through Transfer Pricing in OECD Countries.” Journal of Public Economics 87, nos. 9-10: 2225-52.

Bird, Richard M., and Jack M. Mintz. 2000. “Tax Assignment in Canada: A Modest Proposal.” In Canada: The State of the Federation 1999/2000: Toward a New Mission Statement for Canadian Fiscal Federalism, ed. Harvey Lazar. Montreal; Kingston, ON: McGill-Queen’s University Press.

Bird, Richard M., and Kenneth J. McKenzie. 2001. “Taxing Business: A Provincial Affair?” C.D. Howe Institute Commentary 154. Toronto: C.D. Howe Institute.

Boadway, Robin, Neil Bruce, and Jack Mintz. 1983. “On the Neutrality of Flow-of-Funds Corporate Taxation.” Economica 50, no. 1: 49-61.

——. 1987. Taxes on Capital Income in Canada: Analysis and Policy. Toronto: Canadian Tax Foundation.

Boskin, Michael J. 1996. “The Nunn-Domenici USA Tax: Analysis and Comparisons.” In Frontiers of Tax Reform, ed. Michael J. Boskin. Stanford, CA: Hoover Institution Press.

Bradford, David F. 1986. Untangling the Income Tax. Cambridge, MA: Harvard University Press.

——. 1988. “What Are Consumption Taxes and Who Pays Them?” Tax Notes, April 18, 383.

Brown, Robert D. 2000. “The Impact of the U.S. on Canada’s Tax Strategy.” ISUMA 1, no. 1: 70-78.

Canada. 1998. Department of Finance. Report of the Technical Committee on Business Taxation. Ottawa: Department of Finance.

——. 2000. Economic Statement and Budget Update. Ottawa: Department of Finance. October 18.

——. 2003a. The Budget Plan 2003. Ottawa: Department of Finance.

——. 2003b. Tax Expenditures and Evaluations 2003. Ottawa: Department of Finance.

Carroll, Robert, Douglas Holtz-Eakin, Mark Rider, and Harvey Rosen. 2001. “Personal Income Taxes and the Growth of Small Firms.” In Tax Policy and the Economy, vol. 15, ed. James M. Poterba. Cambridge, MA: MIT Press.

Chen, Duanjie. 2000. “The Marginal Effective Tax Rate: The Only Tax Rate that Matters in Capital Allocation.” Backgrounder. Toronto: C.D. Howe Institute.

Chen, Duanjie, and Jack M. Mintz. 2003. “Taxing Investments: On the Right Track, but at a Snail’s Pace.” Backgrounder 72. Toronto: C.D. Howe Institute.

Clausing, Kimberly A. 2003. “Tax-Motivated Transfer Pricing and US Intrafirm Trade Prices.” Journal of Public Economics 87, nos. 9-10: 2207-23.

Clemens, Jason, Joel Emes, and Rodger Scott. 2002. “The Corporate Capital Tax: Canada’s Most Damaging Tax.” Public Policy Sources 56. Vancouver: Fraser Institute.

Conference Board of Canada. 2003. Performance and Potential 2003–04: Defining the Canadian Advantage. Ottawa. October.

Dahlby, Bev. 2000. “Tax Reform and Economic Growth in Alberta.” Calgary: Canada West Foundation.

——. 2002. “Globalization and the Optimal Taxation of Capital in a Small Open Economy.” University of Alberta, Department of Economics. Draft.

——. 2003. “Restructuring the Canadian Tax System by Changing the Mix of Direct and Indirect Taxes.” In Tax Reform in Canada: Our Path to Greater Prosperity, ed. Herbert G. Grubel. Vancouver: Fraser Institute.

——. Forthcoming. “The Chrétien Government’s Legacy: A Tax System for the 21st Century?” In The 2003 Federal Budget: Conflicting Tensions, eds. Charles M. Beach and Thomas A. Wilson. Kingston, ON: Queen’s University, John Deutsch Institute for the Study of Economic Policy.

Dancey, Kevin J. 1998. Impact of Taxation on the Financial Services Sector. Research paper for the Task Force on the Future of the Canadian Financial Services Sector. Ottawa: Department of Finance.

d’Aquino, Thomas Paul, and David Stewart-Patterson. 2001. Northern Edge: How Canadians Can Triumph in the Global Economy. Toronto: Stoddart.

Davies, James B., and France St-Hilaire. 1987. Reforming Capital Income Taxation in Canada: Efficiency and Distributional Effects of Alternative Options. Ottawa: Supply and Services Canada for the Economic Council of Canada.

Diewert, W.E. 1988. “On Tax Reform.” Canadian Journal of Economics 21, no. 1: 1-40.

Dobson, Wendy. 2002. “Shaping the Future of the North American Economic Space: A Framework for Action.” C.D. Howe Institute Commentary 162. Toronto: C.D. Howe Institute.

Duclos, Jean-Yves, and Julie Gingras. 2000. “Tax Mix and Tax Size: Directions for Federal Tax Reform.” C.D. Howe Institute Commentary 126. Toronto: C.D. Howe Institute.

Durning, Alan Thein, and Yoram Bauman. 1998. Tax Shift. NEW Report no. 7. Seattle: Northwest Environment Watch.

Economic Council of Canada. 1987. Road Map for Tax Reform: The Taxation of Savings and Income. Ottawa: Supply and Services Canada.

Federation of Tax Administrators. 2003. “State Individual Income Taxes: Tax Rates for Tax Year 2003.” http://www.taxadmin.org/fta/rate/ind_inc.html

Feldstein, Martin, and Marian Vaillant Wrobel. 1998. “Can State Taxes Redistribute Income?” Journal of Public Economics 68, no. 3: 369-96.

Florida, Richard L. 2002a. The Rise of the Creative Class: And How It’s Transforming Work, Leisure, Community and Everyday Life. New York: Basic Books.

——. 2002b. “The Economic Geography of Talent.” Annals of the Association of American Geographers 92, no. 4: 743-55.

Fortin, Pierre. 2000. “Less Taxes and Better Taxes: Principles for Tax Cuts and Tax Reform.” Canadian Tax Journal 48, no. 1: 92-100.

Fried, Joel, and Ron Wirick. 1999. “Assessing the Foreign Property Rule: Regulation without Reason.” C.D. Howe Institute Commentary 133. Toronto: C.D. Howe Institute.

Gale, William G., and Peter R. Orszag. 2004. “The Budget Outlook: Updates and Implications.” Washington, DC: Brookings Institution. January 29.

Goldfarb, Danielle. 2003. “The Road to a Canada-U.S. Customs Union: Step-by-Step or in a Single Bound?” C.D. Howe Institute Commentary 184. Toronto: C.D. Howe Institute.

Greve, Michael S. 2003. Sell Globally, Tax Locally: Sales Tax Reform for the New Economy. Washington, DC: AEI Press.

Grubel, Herbert G. 2003a. “Why There Should Be No Capital Gains Tax.” In Tax Reform in Canada: Our Path to Greater Prosperity, ed. Herbert G. Grubel. Vancouver: Fraser Institute.

——. 2003b. “Don’t Just Cut Taxes — Reform the Tax System.” Fraser Forum (July), 14, 17. Vancouver: Fraser Institute.

Hall, Robert E., and Alvin Rabushka. 1985. The Flat Tax. Stanford, CA: Hoover Institution Press.

Harden, J. William, and William H. Hoyt. 2003. “Do States Choose Their Mix of Taxes to Minimize Employment Losses?” National Tax Journal 56, no. 1: 7-26.

Harrison, Kathryn. 2003. “Follow the Leader and Dominoes: Games Provinces Play in Tobacco Taxation.” University of British Columbia, Department of Political Science. Draft.

Hayward, Paul D. 2002 “Income Trusts: A ‘Tax-Efficient’ Product or the Product of Tax Inefficiency?” Canadian Tax Journal 50, no. 5: 1529-69.

Jedwab, Jack. 2002. “Diversity Popular among Canadians.” Association for Canadian Studies, Montreal. http://www.acs-aec.ca/polls/poll45.pdf

——. 2003. “Canada: Bastion of Individual Liberties?” Association for Canadian Studies, Montreal. http://www.acs-aec.ca/polls/poll49.pdf

Jorgensen, Dale W., and Kun-Young Yun. 1991. “The Excess Burden of Taxation in the United States.” Journal of Accounting, Auditing and Finance 6: 487-508.

Kent, Tom. 2003. The Repair of Taxation. Ottawa: Caledon Institute of Social Policy.

Kesselman, Jonathan R. 1983. Financing Canadian Unemployment Insurance. Toronto: Canadian Tax Foundation.

——. 1994. “Public Policies to Combat Child Poverty: Goals and Options.” In A New Social Vision for Canada? Perspectives on the Federal Discussion Paper on Social Policy Reform, eds. Keith Banting and Ken Battle. Kingston: Queen’s University, School of Policy Studies.

——. 1996. “Payroll Taxes in the Finance of Social Security.” Canadian Public Policy 22, no. 2: 162-79.

——. 1997. General Payroll Taxes: Economics, Politics, and Design. Toronto: Canadian Tax Foundation.

——. 1999. “Base Reforms and Rate Cuts for a Revitalized Personal Tax.” Canadian Tax Journal 47, no. 2: 210-41.

——. 2000. “Flat Taxes, Dual Taxes, Smart Taxes: Making the Best Choices.” Policy Matters 1, no. 7. Montreal: Institute for Research on Public Policy.

——. 2001. “Policies to Stem the Brain Drain—Without Americanizing Canada.” Canadian Public Policy 27, no. 1: 77-93.

——. 2002. “Fixing BC’s Structural Deficit: What, Why, When, How? And For Whom?” Canadian Tax Journal 50, no. 3: 884–932.

——. Forthcoming. “Unfinished Business: A Tax System for the 21st Century.” In The 2003 Federal Budget: Conflicting Tensions, eds. Charles M. Beach and Thomas A. Wilson. Kingston: Queen’s University, John Deutch Institute for the Study of Economic Policy.

Kesselman, Jonathan R., and Ron Cheung. Forthcoming. “Taxation, Progressivity, and Inequality in Canada.” Canadian Tax Journal.

Kesselman, Jonathan R., and Finn Poschmann. 2001. “Expanding the Recognition of Personal Savings in the Canadian Tax System.” Canadian Tax Journal 49, no. 1: 40-101; abbreviated as “A New Option for Retirement Savings—Tax-Prepaid Savings Plans.” C.D. Howe Institute Commentary 149. Toronto: C.D. Howe Institute.

Kneller, Richard, Michael F. Bleaney, and Norman Gemmell. 1999. “Fiscal Policy and Growth: Evidence from OECD Countries.” Journal of Public Economics 74, no. 2: 171-90.

Lefebvre, Pierre, and Philip Merrigan. 2003. “Assessing Family Policy in Canada: A New Deal for Families and Children.” Choices 9, no. 5. Montreal: Institute for Research on Public Policy.

McKenzie, Kenneth J. 2000a. “Replacing the Alberta Personal Income Tax with a Sales Tax: Not Heresy but Good Economic Sense.” Canada West Foundation, Calgary.

——. 2000b. “Provincial Tax Priorities in a Global and National Economy: What’s Good for the Goose Is Good for the Gander.” Canadian Tax Journal 48, no. 2: 356-73.

Mérette, Marcel. 2002. “The Bright Side: A Positive View on the Economics of Aging.” Choices 8, no. 1. Montreal: Institute for Research on Public Policy.

Mintz, Jack M. 2000. “Reforming the Tax Cut Agenda.” Canadian Tax Journal 48, no. 3: 689–709.

——. 2001a. Most Favored Nation: Building a Framework for Smart Economic Policy. Policy Study 36. Toronto: C.D. Howe Institute.

——. 2001b. “Withholding Taxes on Income Paid to Nonresidents: Removing a Canadian-US Border Irritant.” Backgrounder 49. Toronto: C.D. Howe Institute.

——. 2003. “Taxation with the Least Pain — A New Tax Structure for Canada.” In Tax Reform in Canada: Our Path to Greater Prosperity, ed. Herbert G. Grubel. Vancouver: Fraser Institute.

Mintz, Jack M., and Finn Poschmann. 1999. “Tax Reform, Tax Reduction: The Missing Framework.” C.D. Howe Institute Commentary 121. Toronto: C.D. Howe Institute.

National Governors Association and National Association of State Budget Officers. 2003. The Fiscal Survey of the States. Washington, DC. June.

OECD. 1997. OECD Economic Surveys 1996–1997: Canada. Paris: OECD.

——. 2003. Revenue Statistics, 1965–2002. Paris: OECD. Parti Libéral du Québec. 2003. “Raising Quebecers’ Standard of Living: Priorities for Political Action in Order to Reach Our Full Potential for Economic Prosperity.” Montreal.

Poschmann, Finn, and John Richards. 2000. “How to Lower Taxes and Improve Social Policy: A Case of Eating Your Cake and Having It Too.” C.D. Howe Institute Commentary 136. Toronto: C.D. Howe Institute.

Poschmann, Finn, and William B.P. Robson. 2001. “Reprogramming the EI Cash Machine: Matching Employment Insurance Premiums and Payouts.” Backgrounder 51. Toronto: C.D. Howe Institute.

PriceWaterhouseCoopers LLP, Canada. 2003. Tax Facts and Figures for Individuals and Corporations, Canada 2003. Available from website: http://www.pwc.com/ca.

Quebec. 2003. Ministère des Finances. Budget Plan: 2003- 2004 Budget. Quebec City.

Robson, William B.P. 1996. “Putting Some Gold in the Golden Years: Fixing the Canada Pension Plan.” C.D. Howe Institute Commentary 76. Toronto: C.D. Howe Institute.

Sayeed, Adil. 1999. “Improving the National Child Benefit: Matching Deeds with Intentions.” C.D. Howe Institute Commentary 125. Toronto: C.D. Howe Institute.

Shome, Parthasarathi, and Christian Schutte. 1993. “Cash Flow Tax.” International Monetary Fund Staff Papers 40: 638-62.

United States. 2002a. Board of Governors of the Federal Reserve System. Flow of Funds Accounts of the United States: Flows and Outstandings, Fourth Quarter 2002. Washington, DC.

——. 2002b. General Accounting Office. Internet Cigarette Sales: Giving ATF Investigative Authority May Improve Reporting and Enforcement. GAO-02-743. Washington, DC: General Accounting Office.

——. 2003a. Council of Economic Advisers. “Tax Policy for a Growing Economy.” In Economic Report of the President. Washington, DC: US Government Printing Office.

——. 2003b. Department of the Treasury. “President’s Budget Proposes Bold Tax-Free Savings and Retirement Security Opportunities for All Americans.” Press release KD-3816. Washington, DC. January 31.

——. 2004. Congressional Budget Office. The Budget and Economic Outlook: Fiscal Years 2005–2014. Washington, DC. January.

Walker, Michael. 1994. In Canada, Minutes of Proceedings and Evidence of the Standing Committee on Finance. 35th Parliament, 1st Session, March 1, issue 13: 16–28.

Walsh, Brendan. 2003. “Taxation and Foreign Direct Investment in Ireland.” In Tax Reform in Canada: Our Path to Greater Prosperity, ed. Herbert G. Grubel. Vancouver: Fraser Institute.

Wilson, Thomas A. 2003. “An Evaluation of Business Taxes in Canada.” In Tax Reform in Canada: Our Path to Greater Prosperity, ed. Herbert G. Grubel. Vancouver: Fraser Institute.

World Economic Forum. 2003. Global Competitiveness Rankings 2003–2004. Geneva.