The rationale of the Flaherty "Tax Fairness Plan" (The Plan) is that corporations converting to the trust structure "don't pay their share of taxes" and that "the tax burden will shift onto the shoulders of hardworking individuals and families". Part A of our paper disputes the assertion that the government loses large amounts of tax revenue because of trust conversions, while Part B examines in detail the costs and benefits of the Plan for the government and Canadians who invested in trust units.

There have been a few attempts to quantify the alleged tax leakage. These studies however rest on incorrect assumptions or fail to take all relevant factors into consideration. They are not useable for tax policy purpose. The Plan does not offer any concrete estimate whatsoever of the alleged tax revenue loss. To say that trusts don't pay taxes is incorrect as it ignores the fact that taxes are simply shifted to their unit holders.

The alleged tax leakage has been blamed mainly on RRSP/RRIFs and pension funds. The Tax Fairness Plan wrongly treats these entities as "tax exempt". The reality is that they are not "tax-exempt" but "tax-deferred". Ultimately all capital and income accumulated in such accounts is taxed at personal tax rates when the monies are withdrawn sometime in the future. The crux of the issue is whether there is a difference between the present value of future taxes and the value of foregone tax revenue in the short run. Such a study has never been done and would be essential before policy measures affecting so many Canadians and tens of billions of their saving are decided.

The Tax Fairness Plan aimed at leveling the playing field through tax measures specifically tailored to address the alleged tax leakage will be extremely costly for both government revenue and hard working Canadians who invested in trust units. By the time the government starts taxing trust distributions in 2011 the Tax Fairness Plan will already have cost the government $9 billion in foregone tax revenue. In the following years the net annual cost of the measures announced in the Plan will run at over $1.8 billion a year while the tax on distributions will increase revenue by a mere $400 million annually, and most likely less.

Hard working Canadians with significant amounts of retirement saving invested in income trusts will be losers for the rest of their lives not only in terms of the permanent decimation of their capital and saving but also in terms of a lower flow of investment income, and ultimately retirement benefits. The increase in the Tax Credit for Age and the splitting of pensions will offer marginal compensation, if any, to the great majority of investors who are struggling with heavy losses. It is likely that these measures will benefit mostly persons with low income and/or who had no or little investment in trusts.

The gigantic cost to investors and the massive budgetary cost of the Plan compared to the very minor tax revenue gain lead to the strong suspicion that the decision to shut down the income trusts was motivated by reasons other than the alleged tax revenue loss.

Was it worth the cost and the turmoil? The answer is clearly "No, Mr. Minister".

I. INTRODUCTION AND SCOPE

There has been concern for some time that the existence of income trusts is causing a loss in government tax revenue while having a negative impact of the efficient allocation capital. This concern was heightened recently by the announcement by Telus and BCE that they intended to convert to the trust structure.

After two failed attempts in 2004 and 2005 to address the issue the federal government announced on October 31, 2006 that distributions by the existing publicly traded income trusts would be taxed at the same rate as public corporations starting in 2011. In his statement¹ Minister Flaherty placed very heavy emphasis on tax fairness and the need to stop the tax leakage. Surprisingly, he gave only minor attention to the question of efficient allocation of capital.

The purpose of this paper is not to engage into a statistical exercise but rather to consider two broad issues: a) whether the allegation that tax revenue leakage occurs has been adequately documented and proven; and b) to consider the costs and benefits of the tax measures announced² by the government in its Tax Fairness Plan.

II. ATTEMPTS TO MEASURE THE MAGNITUDE OF THE LEAKAGE

There have been a few attempts in both official and private circles to quantify the alleged loss of tax revenue. These analyses have been found by many critiques to be flawed and leading to questionable, if not, wrong conclusions. Unfortunately, the allegation of tax leakage has been so much hammered in the media and some political circles that it has become a widely accepted truth without much questioning and discussion. It was nevertheless the official rationale underpinning the decision to tax the trusts.

To simplify matters we will focus on two main documents that attempted to quantify the leakage: a) the Department of Finance Consultation Paper on income trusts (DoF Paper) released on September 8, 2005 under the authority of Finance Minister Goodale.³; and b) the October 2006 paper by Professor Jack Mintz of the Rotman School of Management of the University of Toronto⁴. We will also look at how the backgrounder⁵ attached to Minister Flaherty's policy statement of October 31, 2006 deals with this issue.

1. The September 2005 Consultation Paper

The Paper estimated that the federal tax loss was about $300 million in 2004. One of the key DoF arguments is found in Table 2 of the Paper "Integration of Personal and Corporate Income Taxes Under Various Business Structures". Table 2 was used to show that the government receives less tax revenue from the holders of trust units than it does from public corporations and their shareholders. Unfortunately, the assumptions made in that Table were flawed and led erroneously to the conclusion that a significant leakage occurs. There are other points in that paper that are questionable but let's focus on the most essential.

The fundamental problem with that Table is that it underestimated the amounts paid out by trusts to their unit holders, and therefore taxes paid by the latter, while it overestimated taxes collected from public corporations and their shareholders.

The underestimation in the case of trusts was due to the fact that the paper implied that income trusts only distribute their business income to unit holders while in fact trust distributions are significantly higher than that. Trust distributions are based on distributable cash flow per unit. Total distributions were estimated at 1.8 times earnings in 2004.

The overestimation of tax revenue from public corporations and their shareholders resulted from two inaccurate assumptions. First, the paper assumed erroneously that all after-tax corporations profits are paid out to shareholders in the form of dividends. This is of course wrong since corporations only distribute a portion of their after-tax profits to shareholders, the rest being retained by the corporations for other purposes such as re-investment. It has been estimated that the average dividend payout ratio of Canadian public corporations was only 28% in 2004. This is very far from the much larger payout ratio (65%) assumed in the DoF Paper. Incidentally, the DoF Paper assumption is quite ironic since one of the frequent criticisms directed at trusts is that they distribute too much to their units holders and retain too little, if anything, for the financing of their growth and development.

Another source of overestimation is that the Paper assumes that all public corporations pay the (federal + provincial) corporate tax rate of 35%. However in reality it has been estimated by Statistics Canada that the effective corporate tax rate was around 27% in 2004. Moreover, this effective rate is likely to be even lower in the years ahead since the government announced in its 2006 budget a reduction in the federal general corporate tax rate would reduced progressively from 21% in 2007 to 19% in 2011. In his October 31 Tax Plan Minister Flaherty announced another 0.5% cut effective on January 1, 2011. These tax cuts would of course contribute to reduce the estimated magnitude of the alleged tax leakage if they were fully taken into consideration.

If the necessary corrections to deal with the problems of underestimation and overestimation of tax revenue outlined above were made to Table 2. of the DoF Paper, the conclusion may well be that there is very little or no tax leakage resulting from the fact that trusts are not taxed, except in the case of non-residents. It is even conceivable that taxes collected on the distributions of income trusts may be higher than taxes collected from public corporations and their shareholders.

2. Professor Mintz's Paper

Prof. Mintz states in the introduction of his paper that: "Estimating the revenue impact of income trust conversions require data that are sometime difficult to obtain". He makes it clear that most of the data he used to make his updated calculations are taken from the 2005 DoF Consultation Paper.⁶ That of course raises serious questions about Prof. Mintz's calculations, considering what we have just said above concerning the shortcomings of the DoF Consultation Paper. According to Mintz the single biggest tax loss ($500 million) occurs from one particular group of investors, namely pension funds and owners of RRSPs who own 39% of all trust units and do not pay tax on distributions at the time they are received from trusts. Mintz adds that the second biggest source of tax leakage is foreign investors ($200 million) who hold about 22% of income trust units and pay only the 15% federal withholding tax. He estimated that the BCE and Telus conversion would have brought the total to $1.1 billion. Finally, Mintz makes it clear in the conclusion of his paper that his estimates do not take into account dynamic effects through increased investment.

3. The Backgrounder to Minister Flaherty's October 31, 2006 Policy Statement

The backgrounder released by the Department of Finance on October 31, 2006 along with Minister Flaherty's policy statement does not add much to what had already been said on the cause of the alleged tax leakage in the September 2005 Consultation Paper. The justification for Minister Flaherty's decision is nevertheless based on the contents of this backgrounder. The document basically repeats the allegations made in the Consultation Paper that the cause of the problem is the unbalanced tax treatment between public corporations and trusts and that the tax leakage is due to investment in trust units by the so-called Canadian "tax exempts" (i.e., the holders of RRSP/RRIF accounts and pension funds, normally referred to as "tax-deferred") and non residents

It is interesting to note that the backgrounder does not provide much concrete evidence of the tax leakage, nor does it make any attempt to measure the extent of the actual leakage⁷ as was the case in the 2005 Consultation Paper and the Mintz study. The allegation of a high level of tax leakage is a general statement simply based on Table 1: "Simplified Comparison of Investor Tax Rates (current system)". In this Table it is shown that the tax-deferred entities (identified as "tax exempts") are not paying any tax on trust distributions. The bulk of the paper is devoted to an explanation of how things will look after the policy measures announced by Minister Flaherty take effect in the years ahead. This lack of quantification is a bit intriguing. It may well be explained by the fact that the DoF did not feel it was in a position to measure the alleged tax leakage adequately, or that it did not want to put forward numbers that would be challenged.

III. FULLY TAXABLE PORTFOLIOS ARE NOT A SOURCE OF TAX LEAKAGE

The fully taxable portfolios that hold about 39% of all trust units have generally not been considered as a source of tax leakage. This is recognized explicitly by Prof. Mintz when he writes: "By converting to a trust a business is able to avoid payment of corporate taxes and investors no longer pay personal taxes on corporate dividends and appreciation in share values. However investors will pay personal or withholding tax on income trust distributions that are typically more than the personal tax that would have been owed on preferentially-taxed corporate dividend and capital gain income."

Annex I to this paper is very interesting in this respect. The Table compares taxes payable in 2004 by an Ontario 65 plus retired investor on a $500,000 fully taxable (i.e., non-RRSP) portfolio depending on whether the portfolio is invested in income trust units (CASE A) or in dividend-paying public corporations (CASE B). In CASE A the taxpayer pays $12,872. or 23% of his taxable income. In addition he will eventually have to pay another $1,150. in capital gain tax when the units are sold due to the fact that 20% of the distributions he received were tax-deferred in the form of Return of Capital. So his total tax liability is $14,022. In CASE B the individual taxpayer pays only $758. or 2.6% of his taxable income, of which $300. is the Ontario health tax.

To determine the net take of the government in CASE B one would of course have to add to the taxes paid by the investor the taxes paid by the dividend-paying corporations from which he received his $11,000. in dividends. This can be roughly estimated as follows:

1. The $11,000. received by the taxpayer in the form of dividends represent 28% of profits (28% being the average dividend payout ratio of Canadian public corporations).
2. This means that the corporations from which the taxpayer received his dividends would have had to realize $39,285. of taxable profits to pay $11,000. in dividends.
3. As the Canadian effective corporate tax rate is about 27% the said corporations would have paid $10,607. in taxes (i.e., 27% of $39,285.)
4. The difference between the taxable profits ($39,285.) and the sum of dividends paid out to the taxpayer ($11,000.) and taxes paid by the corporations ($10,607.) would be added to the corporations retained earnings (i.e., $21,607. representing 45% of the taxable profits).

The government's tax revenue in CASE A is $14,022. or 23.4% more than the $11,365. (i.e., the sum of $758 paid by the taxpayer plus $10,607. paid by the corporations) collected in CASE B.

While it would be presumptuous to draw a firm conclusion in aggregative terms based on the facts pertaining to this single individual taxpayer, it certainly confirms the general point made by Prof. Mintz that fully taxable individuals taxpayers investing in trust units pay taxes that are "typically more" than if they invested in dividend-paying public corporations. This undoubtedly represents a very interesting windfall for the government considering that almost 40% of all trust units are held by individuals in non-RRSP accounts. This windfall offsets some of the alleged leakage attributed to RRSPs and pension funds. It may well be that the sharp increase in holdings of trust units in taxable portfolios in recent times explains in part the unusual increase in personal tax revenue that puzzles the Department of Finance.

Moreover, while it would be difficult to document, it may well be that individual taxable investors with substantial investment in income trust units tend to be wealthier than the average and pay tax at higher rates than the average. One would not expect taxpayers with low income to have substantial investment of that type. The use of the average personal tax rate to estimate tax leakage as was done in the DoF Consultation Paper may not be appropriate.

The general conclusion that clearly emerges from all of this is that if there is a problem it is not due to taxable investors.

IV. RRSPs AND PENSION FUNDS: THE MAIN CULPRITS?

All the documents focusing on the alleged tax leakage point to the tax-deferred entities, such as the RRSPs, pension funds and limited partnerships, as the most important source of tax losses.

A. RRSP Accounts

RRSPs are not tax-exempt but tax-deferred retirement saving entities. They may be tax exempt in the short run but they are certainly not tax exempt in the longer run. When RRSP holders reach 69 they must convert their RRSPs into RRIFs and start withdrawing their sheltered savings at an annual rate prescribed by the government. Such annual mandatory withdrawals are taxed as ordinary income, that is, at the highest level.

Eventually all amounts held in RRSPs are fully taxed irrespective of the types of assets or income which accumulated in such accounts. This means that all income trust distributions (which typically include interest, return of capital, dividends and capital gain realized by the trusts), the capital gain arising from the so-called return of capital, the income received from the reinvestment of the distributions, as well as the possible overall capital gain resulting from the appreciation in the market value of the units themselves will eventually be taxed as ordinary income. The highest level of taxation will therefore apply as the government denies preferential tax treatment of dividends and capital gain included in RRSP/RRIF withdrawals and pension fund benefits.

The government may have lower tax revenue at present and in the years of accumulation in the RRSPs but this is "money in the bank" as it will recoup at a later stage. The time lag can be considerable but in the end the taxman gets his due, and maybe more than his due. The key issue is whether the present value of such future tax revenues is higher or lower than the taxes foregone at the present time, or about equal.

We would now like to touch on a three points concerning RRSPs and RRIFs which might justify a number of policy changes in relation to:
a) the taxation of RRSP/RRIF withdrawals and pension benefits;
b) the annual rates at which such withdrawals must be made; and
c) the problem of the plummeting personal saving rate.

Impact of Denial of Preferential Tax Treatment The denial by the government of preferential tax treatment of accumulated capital gain and dividends at the time of RRSP/RRIF withdrawals and pension fund benefit payments is fundamentally unfair and the Flaherty tax on trust will make things worse. The usefulness of RRSPs and RRIFS may be seriously affected as it may well become much more tax efficient in the longer run for rational and savvy investors to hold all their investment in public corporation shares and trust units in taxable accounts benefiting from preferential tax treatment of dividends and capital gain. This is not caused by an unbalanced tax treatment between corporations and trusts but by the grossly unfair taxation of RRSP/RRIF withdrawals. This is an issue that has been completely ignored in the so-called Tax Fairness Plan. See annex 2 for a shocking demonstration of how retirement saving held in a RRSP/RRIF can be taxed by as much as 73% more than the same amount held in a taxable account if the tax refund is not reinvested.

Annual Rates of Mandatory RIF Withdrawals The annual rates at which mandatory withdrawals from RRIFs must be made were set by the government at a time of high interest rates. Remember the days when the no-risk GICs paid 9 - 9.5% interest? The rates of interest have fallen significantly since then but the annual mandatory withdrawal rates have remained the same. Also people live longer and longer. People are scared to outlive their capital and become dependent on welfare. This has led to a situation where the holding of a substantial amount of high yield investment including income trust units in RRIFs has become a must in order to avoid the rapid and premature depletion of the hard-earned retirement capital. The loss of capital combined with the lower rate of return on RRIF investment that will result from Minister Flaherty's decision will lead to much less income for retirees, a more rapid depletion of their RRIF savings, more costs to the government in terms of higher tax credits claimed by the affected taxpayers, and possibly higher budgetary cost to assist retirees when their RRIFs are exhausted.

Plummeting Personal Saving Rates Finally, it should be noted that the overall rate of personal saving in Canada has been declining sharply in the last 10-15 years. This often raises concerns in both governmental and private circles about the negative impact on the economy and the future retirement needs of an aging population. Pension fund under funding is another source of worry. One would think that the government would want Canadians to maximize their aggregate saving whether it is held in taxable accounts of in tax deferred entities. Minister Flaherty has however destroyed permanently between $22 and $29 billion⁸ of precious retirement saving owned by Canadians in a move to collect $400 million or less in additional tax revenue starting in 2011/12. The massive loss of capital and the lower rate of return on retirement saving resulting from his decision will translate into lesser retirement benefits.⁹

B. Pension Funds

Like RRSPs pension funds are tax-deferred retirement plans. Pension benefits paid by these funds come from their cash flow, and in some cases capital. When retirement benefits are paid to pensioners they are fully taxable as ordinary income. Therefore much that was said above about RRSPs and RRIFs also applies to pension funds and will not be repeated.

C. Are RRSPs and Pensions Funds Tax-exempt or Tax Deferred?

It is amazing how the expressions "tax-exempt" and "tax-deferred" are used interchangeably in reference to RRSPs and pension funds. The DoF Consultation Paper keeps referring to RRSPs and pension funds as "tax-exempt" entities. The backgrounder to Minister Flaherty's statement does the same.

In his excellent paper "Limiting Pension Fund Investment in Canadian Business Income Trusts" published in the December 12, 2005 edition of Tax Notes International, Martin Przysuski¹⁰ stated clearly that "RRSPs and pensions funds are not tax-exempt because their cash flow is ultimately taxed at the individual beneficiary level when it is withdrawn at a future date, presumably to fund retirement benefits." He then added: "Therefore if pension funds and other tax-deferred plans increase their participation in a growing income trust market, the fundamental issue of leakage is the erosion of the government's current tax revenues and deferral of those tax revenues rather than a complete loss of that revenue."¹¹

This leads to the conclusion that there might be a net tax leakage if the present value of future taxes did not equal the value of the tax revenues foregone at the present time, unless the value of the taxable future income rose to a point that the present value of future taxes and tax revenues foregone at present would offset each other. In the case of investment in trust units the later point may well be correct in practice considering that trust distributions to unit holders, including RRSPs and pension funds, include not only "other income" but also amounts of dividends, capital gains realized by the trusts and return of capital. These, and the possible unrealized capital gains in the market value of the trust units themselves, would likely contribute to the increase in the value of the portfolios and withdrawals of pension benefits from these portfolios in a longer-term perspective. The reinvestment over a long period of time of the substantial distributions paid by the trusts would also contribute to that rapid growth.

One must also take into account that the accumulated dividends, capital gains, and additional capital gains arising from the fact that about 30% of trust distributions are in the form of return of capital will all be taxed as ordinary income when retirement benefits are paid out, i.e., at a rate much higher that if this income had been held in a taxable accounts.

If all these factors were fully taken into consideration it may well be that there is little or no net tax leakage, or even a windfall for the government. The question of the difference, if any, between the present value of future taxes and the value of foregone tax revenues at the present time is clearly the crux of the issue of trusts and the alleged tax leakage. It would deserve to be studied in much more detail and depth than hitherto before any further consideration is given to the issue of tax leakage. Such a study has never been attempted. Only a thorough and comprehensive study taking all relevant points into consideration would permit to make sound conclusions and policy choices concerning the alleged tax leakage.

V. THE CASE OF NON-RESIDENT INVESTORS

Non-resident investors own about 22% of all income trust units and only pay a 15% federal withholding tax on trust distributions. This is all the government receives since the trusts themselves do not pay tax. Since the withholding tax is only paid to the federal government provinces get nothing. So in this case there is clearly a case of tax leakage. Prof. Mintz has estimated the net leakage is about $200 million in 2006. Specific measures to address this leakage may have been difficult to implement. Nevertheless, considering that our trust structure is unique to Canada restrictions on non-resident ownership should have been considered. There are precedents in the case of other businesses such as chartered banks and insurance companies.

The overall conclusion of Part A of our paper is that in spite of the lack of clear evidence that tax leakage occurs and the inability of DoF to quantify adequately the magnitude of the alleged leakage, Minister Flaherty made policy decisions affecting negatively millions of Canadians and tens of billions of dollars of investment and retirement saving.

VI. WHAT ARE THE COSTS AND BENEFITS OF THE GOVERNMENT'S DECISION?

In his policy statement of October 31, 2006 Minister Flaherty placed very heavy emphasis on the need to stop the tax revenue leakage arising from conversions to the trust structure. He stated that the government would lose "billions of dollars" if no action was taken. This served as the main underpinning for his decision. As we concluded in Part A. above the existence and the magnitude of such leakage, if any, have yet to be documented properly. Nevertheless, since a decision fell based on tax leakage, we will assume in the rest of this paper that a leakage does occur and that the leakage will amount to about $900 million in 2006 as guesstimated in Prof. Mintz's paper. This quantitative estimate is the only one available since the DoF did not care to produce any estimate for 2006 in its Tax Fairness Plan.

In the examination of this question we will consider only the government tax measures specifically aimed at dealing with the alleged $900 million tax leakage. These are: a) the increase in the dividend tax credit announced last year by Minister Goodale and reconfirmed in the 2006 budget; and b) the four new measures announced in the "Tax Fairness Plan", namely:

- The new tax on distributions of income trusts effective January 1, 2011.
- The increase in the Age Credit Amount effective January 1, 2006.
- Income splitting for pensioners beginning in 2007.
- A further reduction in the general corporate tax rate of 0.5% effective January 1, 2011

1. Impact on the Federal Government Tax Revenue¹²

a)Increase in The Dividend Tax Credit for Dividends Paid by Large Corporations After 2005 This measureinitially announced by Minister Goodale in November 2005 and reiterated in Budget 2006 is aimed at leveling the playing field between the taxation of corporations and trusts. It will cost the federal government well in excess of $300 million a year. Table A3.1 contained in Annex 3 to Budget 2006 (Tax Measures: Supplementary Information and Notices of Ways and Means Motions) estimates the cost of that measure to be $375 million in 2006/07 and $310 million in 2007/08. The cumulative cost before the tax on trust distributions becomes effective may well reach $2 billion. This cost is not taken into consideration in the Tax Fairness Plan.

b) The Increase in the Age Credit Amount will cost a total amount of about $2.225 billion in the period from 2006/07 to 2011/12

c) Pension Income Splitting will cost a total amount of about $3.9 billion in the same period.

d) The Additional 0.5% Reduction in General Corporate Tax will cost about $725 million a year when it becomes effective in 2011

e) The Tax on the Distributions of Income Trusts will generate an estimated amount of about $400 million a year in additional revenue when it becomes fully effective in 2011/12. This compares with Prof. Mintz's estimate of an annual leakage of $900 million a year¹³. This difference may be explained by the fact that the REITs will be exempted and other factors, such as the progressive reduction of the general corporate tax. It is not very clear what calculations the DoF made to arrive at this amount. There is no explanation supporting that number. It may well be that the estimated $400 million in additional annual revenue from 2011/12 may be on the high side. If, as is likely, many trusts reconvert to the public corporation structure or are acquired by corporations, or go private, they will cease paying distributions based on their cash flow and pay much lower dividends based on their profits. The amounts involved could be significant since distributions by trusts are about 1.8 times their earnings.

By the time the tax on income trust distributions announced in the Tax Fairness Plan becomes fully effective in 2011/12 the government will already have foregone a cumulative amount of tax revenue in excess of $9 billion while the additional tax revenue from the taxation of trusts starting in 2011 will be only $400 million, and possibly much less. In the following years, the net cost of the Tax Fairness Plan to the government will likely exceed $1.8 billion a year according to numbers found in the DoF backgrounder.¹⁴ The budgetary cost of the Plan will be a very large multiple of the anticipated increase in tax revenue resulting from the "at the source" tax on trusts. This leads to the strong suspicion that the Tax Plan aimed at destroying the trusts was motivated by reasons other than the alleged tax revenue loss and the search for tax fairness.

2. Impact on Canadian Holders of Income Trust Units

Before considering the impact on groups of investors let us first discuss the general impact of Minister Flaherty's decision on all Canadian holders of income trust units. The immediate impact has been a massive and permanent reduction in the market capitalization of listed income trusts. This impact is relatively easy to estimate. While there is are still jitters in the market it has been estimated that the market capitalization of trusts has dropped by an average of 15-20% so far. This has translated into a permanent capital loss in the range of $22-$29 billion for Minister Flaherty's "hard working individuals and families". (See footnote 8.) About half of these losses was suffered by taxable investors, while RRSP/RRIF holders and pension funds lost the other half. These losses will be compounded by a lower flow of investment income, a lower rate of return on their retirement savings for years to come irrespective of whether they keep their trusts units or shift to lower yield instruments such as dividend paying stocks. Ultimately all these Canadians will receive lower pension benefits when they retire.

a) Impact on Taxable Accounts Held by Canadians

The permanent capital loss suffered by these investors is in the range of $10-$14 billion. Since they are paying tax on trust distributions at rates that are higher than the corporate rate they will likely pay less taxes than is now the case. Distributions paid to these investors will also benefit from the distribution or dividend tax credit The flow of investment income to these investors will however be reduced significantly if trusts reconvert to the corporate model or are bought by public corporations. Payment of distributions by trusts based on their cash flow will cease and trusts will start paying these investors dividends simply based on their profits at the average payout ratio of 28%. The flow of income will also diminish strongly if they decide to shift away from trust units and invest in dividend paying stocks. This concomitant drop in income along with the increase in the dividend tax credit will also result in lower tax liability for these investors, and consequently lower tax revenue for the government as is shown in Annex 1 to this paper. On a net basis they will nevertheless remain heavy losers as regards both their capital and their after-tax income.

The increase in the Age Credit Amount will reduce the tax bill of seniors only by a maximum of $155.00 a year in the case of low-income persons. The tax reduction is however likely to be much lower or tend towards zero for people with any significant amount of investment and income. This measure will provide virtually no compensation for the heavy capital loss they incurred and the sharp reduction in the future flow of investment income when trusts start reconverting to the corporate model. This very modest tax reduction will benefit mainly low-income people who suffered little or no losses on account of the government's decision.

Pension Income Splitting will certainly benefit many eligible retirees and offset some of their losses. The number of persons able to avail themselves of this opportunity may however not be all that great and they may well not be those who need compensation for their losses. First, this measure will do absolutely nothing for a very long time to come for people who are not retired but have and will suffer from capital and investment income losses. They may have to wait for as much as 20-25 years before they become eligible. Moreover, it is highly discriminatory since singles, divorced/separated persons, widows and widowers (of which there are plenty in the ranks of the retirees) will not be in a position to benefit. The same comment applies to the increasing number of workers who work for employers who do not offer pension funds. These workers have to rely on alternative methods of retirement saving such as RRSPs that have just been decimated by the Tax Plan. (In 2005 less than 40% of all workers participated in pension funds.) In the same way the measure will be of very little or no benefit to households where both spouses work and contribute to pension funds. In contrast very wealthy individuals with very rich pension entitlements and a stay at home spouse will likely benefit handsomely. It is also probable that a large number of beneficiaries of this measure had no or very little investment in income trusts. Finally, this new measure does not concern CPP pensions as the splitting of these pensions have already been allowed for many years.

The increase in the dividend tax credit is a mixed bag for retirees with taxable accounts. The grossing up of dividends to 145% from the present 125% will boost further the artificial swelling of their taxable income and lead to the further erosion of tax credits, such as the age credit, and a higher degree of Old Age Security (OAS) claw back for many middle income persons. The investors most likely to benefit are those with very large taxable investment portfolios and income, and who do not qualify for other tax credits, such as the credit for age, and who do not receive the OAS due to their high income.

The additional 0.5% cut in the general corporate tax rate will only benefit these investors to the extent that they have significant amounts of dividend paying stock in their portfolios and that the cut leads to increases in dividend payments by corporations. Given the low dividend pay-out ratio (about 28%) the pickings may well be very lean.

One can conclude that the personal tax reduction measures announced in the Tax Fairness Plan will not do very much to offset the capital losses and the future reduction in investment income of investors with taxable accounts, as well as the significant shortfall in their retirement benefits. These measures are not only of very limited value for most individual investors with significant investment in income trusts but they are also likely to benefit a very large number of persons who suffered little or no financial damage as a result of Minister Flaherty's decision. The latter will undoubtedly applaud the Minister's Plan as it will be all gain and no pain for them.

b) Impact on RRSP/RRIF Holders and Pension Funds

These investors are the most seriously affected. Their permanent capital loss is in the range of $11-15 billion so far. Moreover, the tax on trusts will have a serious negative impact on their investment income and rate of return, and ultimately their retirement benefits as they are forced to shift their investment to lower yield instruments during the period leading to 2011. After 2011 it will be impossible for them to keep trust units in their portfolios because of the extremely high rate of taxation as is explained in the next paragraph. While the younger RRSP holders may be able to recoup some of their capital loss over time by shifting to investment instruments more promising (and possibly more risky) in terms of capital appreciation this will be virtually impossible for the older ones approaching or having reached the age of retirement.

The unfairness resulting from the denial of preferential tax treatment of dividends and capital gain to these tax deferred entities will be compounded. The increase in the dividend tax credit will be of no benefit whatsoever to these investors as it does not apply to them. Assuming that some trusts survive after 2011 trust distributions will be taxed first at the source and then re-taxed at the standard personal rate at the time of withdrawals or pension benefit payments. For example, if in 2011 a trust has $100. to distribute to a RRSP/RRIF holder it will first have to pay $31.50 in corporate tax (18.5% federal + 13% provincial) and then give the investor $68.50 Eventually when the RRSP/RRIF holder withdraws his/her money he/she will have to pay $26.03 in personal tax (i.e., $68.50 X 38% -- the average personal tax rate used in the MoF Consultation Paper). The total tax rate will be an astonishing 57.53% compared to 31.5% if the same units had been held in a taxable account benefiting from the dividend tax credit. In the case of a person paying the top rate of 46% the total tax paid on $100. of distributions would amount to $63.01 (i.e., a total tax rate of 63.01%). Minister Flaherty would no doubt agree that this is most unfair for hard working Canadians.

In the case of RRIF holders some of the negative impact of the proposed tax on trusts will be felt very quickly. The decimated value of their portfolios as of end December 2006 will serve as the base to determine their mandatory withdrawals in 2007. This means that next year their withdrawals will already be lower than would have been expected otherwise and so will the government tax revenue on these withdrawals.

The usefulness of RRSPs and RRIFS may well be affected further by the Flaherty decision as it will become much more profitable in the longer run for rational and savvy investors to hold all their investment and retirement saving in public corporation shares and trust units in taxable accounts benefiting from preferential tax treatment of dividends and capital gain. This is certainly what a good financial adviser should recommend to his/her clients. See Annex 2 showing the extent to which RRSP/RRIFs can be drastically tax inefficient.

The comments made above concerning the very limited usefulness of the increase in the Age Credit Amount and pension splitting are more or less the same for this group of investors and will not be repeated.

3. Impact on Provinces

The additional federal tax revenue from trusts is estimated in Table 4 of the DoF backgrounder at about $400 million a year starting in 2011/12. At that time the combined level of the general corporate tax rate will be 31.5% (18.5% federal + 13.0% provincial). This means that the ten provinces' additional tax revenue from the same source should be in the order of $280 million, with the bulk of this going to Alberta and Ontario where most of the trusts and trust capitalization reside.

Much of this additional revenue will however be eroded by the increase in the dividend tax credit as most provinces have agreed to follow the federal lead in this respect. Also, as in the case of the federal government, the anticipated additional tax revenue may not materialize fully as trusts reconvert to the corporate structure or are acquired by public corporations or go private. In such cases they will stop paying distributions based on their cash flow, and start paying their shareholders much smaller amounts in dividends based on their profits. The huge capital losses suffered by investors and the lower flow of investment income to both taxable and tax deferred accounts starting in 2011 will lead sooner or later to lower tax revenue. In the end the net impact on the tax revenue of most provinces may well be negative.

It should also be added that under the Flaherty Plan trust distributions will be taxed at the source by provincial governments, that is by the province where the trusts reside, and not in the province where the investors reside as is now the case. This means that significant amounts of tax revenue will be lost in the case of provinces that host few trusts but where taxpayers are actively investing in trusts. We would opine that larger provinces such as Quebec and British Columbia would be the most affected. As for Alberta one can only wonder how the destruction of the oil and gas trusts will benefit the province.

VII. CONCLUSION

The issue of the alleged tax leakage in the case of income trust is obviously complex. Attempts made so far to demonstrate its existence and measure its magnitude are highly questionable except in the case of the non-residents.

In the case of taxable accounts the taxes on trust distributions are already taxed at a higher average rate than the corporate tax rate. This difference will increase as the corporate tax rate is progressively lowered in the years ahead.

The bulk of the leakage has been attributed to RRSPs and pension fund investment. These entities are tax deferred and it is totally inappropriate to classify then as "tax exempt". Since the government outlives everybody all trust distributions to these entities are ultimately taxed at the personal tax rate. The only way to demonstrate tax leakage in their case would be to determine whether there is a difference between the present value of future taxes and the value of foregone tax revenues at the present time. This has never been done. Without an in-depth and comprehensive study of this question it is not possible to draw any kind of firm conclusion useable for policy purpose.

The Tax Fairness Plan that aims at leveling the playing field through tax measures specifically tailored to address the alleged tax leakage will be extremely costly for both government tax revenue and hard working Canadians who invested in trust units. By the time the government starts taxing trust distributions in 2011 the Tax Fairness Plan will already have cost the government $9 billion in foregone tax revenue. In the following years the net annual cost will run at well over $1.8 billion a year, while the tax on distributions will increase federal tax revenue by a mere $400 million annually, and probably less.

Canadians with significant amounts of investment in income trusts will be losers not only in terms of the permanent decimation of their capital and savings but also in terms of a lower flow of investment income and ultimately retirement benefits whether they hold taxable or tax-deferred accounts. The increase in the Tax Credit for Age and the splitting of pensions will offer marginal compensation to seniors who had made significant investment in income trusts and are now suffering and will continue to suffer heavy losses. These two measures will likely benefit mostly persons with low income and/or who had no or little investment in trusts.

The rationale of the Flaherty Tax Plan is that corporations converting to the trust model "don't pay their share of taxes" and that "the tax burden will shift onto the shoulders of hardworking individuals". "This is simply not fair", the minister added in his statement. To use this emotional and inflammatory but faulty and incorrect argument is very clever from a populist point of view and good for political support but it hardly makes sense when all the relevant facts are considered and analyzed properly.

The Finance Minister had other options if he wanted to curb the proliferation of conversions into trusts. He chose the worst one for the hard working Canadians trying to save for their retirement.

In final analysis the gigantic cost to investors and the massive budgetary cost of the Plan compared to the minor tax revenue gain lead to the strong suspicion that the decision to shut down the income trusts was motivated by reasons other than the alleged tax revenue loss.

Is it worth the cost and the turmoil? The answer is clearly "No, Mr. Minister".

References:
¹ Department of Finance of Canada. "Statement by the Honourable Jim Flaherty, Minister of Finance." October 31, 2006.
² This includes the increase in the dividend tax credit announced in November 2005 by Finance Minister Goodale.
³ "Tax and Other Issues to Publicly Listed Flow-through Entities - Income Trusts and Limited Partnerships."
⁴ "Income Trusts Conversions: Estimated Federal and Provincial Revenue Impact." This paper updates the estimates made in "Income Trusts and Shareholder Taxation: Getting It Right", by Lalit Aggarwal and Jack Mintz, Canadian Tax Journal, 52(3).
⁵ Department of Finance of Canada "Canada's New Government Announces Tax Fairness Plan". October 31, 2006
⁶ Prof. Mintz noted that his revised estimates incorporate tax changes.
⁷ The backgrounder only makes reference to an estimated provincial loss of $400 million made by Alberta.
⁸ The market capitalization of income trusts (minus the REITs that will be exempted from the tax) and limited partnerships was about $186 billion as of October 20, 2006. Of this about 78%, or $145 billion, was owned by Canadian investors. Since the Flaherty announcement the market value of trust units held by Canadians has fallen by about 15-20% on average, or well between $22 and $29 billion.
⁹ The case of the Canada Pension Plan is instructive in this regard. Some years ago the viability of the CPP was threatened. One of the main reasons was it had been limited to invest in low-yield instruments. The problem was solved by allowing the CPP to invest in higher yield instruments and by forcing workers to pay higher contributions.
¹⁰ Martin Przysuski is a former senior tax official in the Canadian Customs and Revenue Agency. He is a leading specialist in income tax, commodity tax and transfer pricing. He has written numerous articles on these topics.
¹¹ Przysuski attributes these points to Gordon Tait, "Income Trusts - A Perspective on Trusts and Taxes," a BMO Nesbitt Burns research paper, October 3, 2005.
¹² Figures mentioned in this section have been excerpted from Table 4: Impact of Proposed Measures of the backgrounder attached the Minister Flaherty's policy statement of October 31, 2006.
¹³ Does not include the BCE and Telus conversions.
¹⁴ This number includes the impact of the increase in the dividend tax credit.

(1) Resident of Ontario over 65 receiving CPP and Old Age Pension
(2) Calculations made on the basis of 2004 tax forms, except for Ontario health tax for which the full 2005 rates were applied.
(3) No calculation is made for the Ontario property and sales tax credits

The huge difference in tax paid is explained by the following factors:
1. The taxable income in CASE A is much larger even though the taxable income in case B has been grossed up by 25% for the calculation of the dividend tax credit.
2. The taxpayer in CASE A does not receive the federal and provincial dividend tax credit.
3. The taxpayer in CASE B is in a lower tax bracket at both federal and provincial levels.
4. The taxpayer in CASE B receives almost the full amount of credit for age while the taxpayer in case A receives no credit for age due to "high" income.
5. The taxpayer in CASE B pays only $300.00 for Ontario health tax while the one in CASE A pays double at $600.00.
6. The taxpayer in CASE B even receives a small ($69.00) for Ontario income tax reduction.

It is should also be noted that:

1. If the Ontario property credit ($625.00 for persons over 65 of age) were included the taxpayer in case B would most likely end up paying no Ontario income tax at all except $300. for the health tax..

2. The taxpayer in CASE A will also have to pay eventually capital gain tax on the $10,000. tax-deferred distributions received from income trusts as Return of Capital.
This means that in addition to the $12,812. paid for the current year he will also have to pay another $1,150. in tax as soon as he sells his trust units.

i.e.,$10,000. of tax-deferred income (capital gain)
X 50% (capital gain inclusion rate) = $5,000.
X 23% (his effective taxation rate) = $1,150.

1. Ten years ago in January 1996, 60-year old twin brother and sister, John and Mary, each had $7,200. to invest for retirement.

2. Based on the level of their taxable income both paid income tax at the average personal tax rate of 38%

3. While John invested in a RRSP because he thought that RRSPs were tax efficient, Mary decided to put the money in her taxable account.

4. They each bought 200 shares of IPL Energy Inc. (now called Enbridge) at the then market price of $36.00 each for a total cost of $7,200. each.

5. They both joined the Enbridge Dividend Re-investment Plan (DRIP) so that their dividends would be re-invested automatically in new shares of Enbridge.

6. John deducted the $7,200 from his 1996 taxable income and, based on his personal tax rate of 38%, he happily received a tax refund of $2,736. He used the refund to make repairs on his house. As Mary invested in a taxable account she did not received any tax refund.

7. Enbridge proceeded with a 2-for-1 stock split in May 1999, and another 2-for-1 stock split in May 2005. As a result John and Mary each had 800 shares.

8. Between 1996 and 2006 John and Mary each received a total of $6,504. in dividends, all of which was reinvested in Enbridge shares through the DRIP.

9. By November 2006, both John and Mary had acquired an additional 168 shares of Enbridge through the DRIP. Their holdings increased to 968 shares.

10. Having his money in a tax deferred RRSP John did not pay tax on the dividends he received between 1996 and 2006 while Mary paid tax in a total amount of about $750. since she benefited from the dividend tax credit.

11. In 2005 they both had to convert their RRSPs into RRIFs as they were now 69.

12. On November 10, 2006 the Enbridge shares closed at $38.58 The market value of John and Mary's shares was therefore $37,345. (968 X $38.58).

13. On November 10, 2006 John and Mary decided to sell their Enbridge shares and to withdraw the proceeds from their RRIFs.

14. As RRSP and RRIF holders are denied preferential tax treatment of dividends and capital gain by the government, John will have to pay $14,191. in income tax as the whole amount of his RRIF withdrawal is taxable at his 38% personal tax rate. Since John had received a tax refund of $2,736. when he invested his $7,200. in his RRSP in 1996 his net tax bill will amount to $11,455.

15. Mary realized a capital gain of $30,345. (proceeds of $37,345. minus the base cost of $7,000.) of which half ($15,173.) will be added to her taxable income.

16. Since Mary pays income tax at the average personal tax rate of 38% her capital gain will be taxed at the tune of $5,765. Adding this amount to the tax she paid over the years on her dividends ($750.) her total tax bill will amount to $6,516.

17. Poor John who thought that RRSPs were tax efficient will end up paying $4,939. (or 76%) more income tax than Mary.

Would John be better off if he had re-invested his tax refund rather than spending it on his house? The answer is no. Even if he had re-invested in more shares of Enbridge and quadrupled the value of his refund, Mary would still be slightly better off. If he had reinvested in a 4-5% GIC it would only have had a marginal impact. And if John had reinvested in Nortel shares he would have lost his tax refund, and his situation would be exactly as it is #17 above.