Archived - Report on Federal Tax Expenditures - Concepts, Estimates and Evaluations 2018: part 8

Table of Contents

To encourage charitable giving, the Income Tax Act contains provisions for granting tax assistance to donors. A non-refundable tax credit is available to individuals for various types of gifts, including money, personal-use property, and certain capital property such as PLS (PLS).^[2] A similar treatment is provided to corporations, but a deduction exists in place of a tax credit.

This evaluation paper focuses on the tax assistance for donations of PLS. Over the past 25 years, the tax treatment of donations of of PLS has undergone a number of modifications. Most notably, the proportion of capital gains subject to taxation on donated securities was cut in half in 1997 and completely eliminated in 2006.

The paper begins by briefly providing background information on the tax treatment of donations of PLS. This is followed by an analysis of the measure's effectiveness, a discussion of equity considerations, and an examination of its efficiency.

The tax assistance provided in respect of donations of PLS can be divided into two components: the Charitable Donation Tax Credit (described below) applied to the fair market value of the donation, and a tax exemption on the capital gains tax that would be due if the shares were sold and not donated.^[3],^[4]

Individuals are eligible for the non-refundable Charitable Donation Tax Credit (CDTC) on gifts made to registered charities and other qualified donees.^[5] The CDTC follows a tiered rate structure: the first $200 of gifts are credited at the lowest federal income tax rate, 15%, and amounts over this threshold are credited at a higher 29% rate.^[6] Donors can claim the credit on donations valued at up to 75% of their net income in a year, and this ceiling can be increased by 25% of the value of gifts of depreciable property and capital gains on gifts of capital property made in the period, up to net income.^[7] Donors are not required to claim charitable donations in their tax return in the year of donation; they may be carried forward for up to five years (ten years for gifts of ecologically sensitive land), or simply omitted. However, gifts of securities must be reported in order to determine the capital gains eligible for a zero inclusion rate.

Corporations may also claim donations, including gifts of securities, in their tax return. Instead of obtaining a credit, eligible donation amounts may be applied as a deduction against taxable income. Rules regarding the maximum allowable deduction for corporate charitable donations are analogous to those governing individuals.^[8]

In general, the primary objective of tax assistance for charitable donations is to reduce the after-tax price of giving, which can be defined as the amount of potential consumption forgone when donating a dollar (Clotfelter, 2012). This concept can be adapted to the donation of PLS in the Canadian context. To arrive at a general expression of the after-tax price of giving, it is useful to start with a simple theoretical expression for what the economic cost, viewed from the perspective of the donor, might be on the donation of a security. Consider a taxpayer who has already made $200 in contributions, and for whom the CDTC rate matches his or her marginal tax rate. The economic cost of a marginal donation of securities can be represented as:

!$ E(f,g,m,τ,c) = [f-(f-c)gτ]+[(f-c)gmτ]-[fτ] !$ (1)

where f represents the fair market value of the security, c the adjusted cost base, g the standard capital gains inclusion rate, the marginal tax rate of the individual (which is assumed to equal the CDTC rate), and m the capital gains inclusion rate on taxable capital gains on donated securities.^[9] The first term represents the after-tax value of the security to the individual (i.e., the forgone value had the individual sold the security and kept the cash rather than donating it). The second term represents the tax that would be paid on a donated security (the value of this term is currently zero because m was reduced to zero in 2006). Finally, the third term represents the value of the CDTC, which reduces the economic cost of donating a security.

Dividing the above equation by f and simplifying, the expression for E becomes an expression for the after-tax price (forgone consumption for a donation of one dollar), represented below as P:

!$ P(r,g,m,τ)=[(1-τ)]+[rgτ(m-1)] !$ (2)

where !$ \left(\frac{(f-c)}{f}\right) !$, the gain proportion of the value of the security, is now represented by r for simplicity.

Note that when m=1 (i.e., no special tax treatment for gains on donated securities), the expression in (2) simplifies to !$(1-τ)!$ , which is the same as the after-tax price of a cash donation. On the other hand, when m=0 (i.e., the full exemption of capital gains tax on donated securities), the expression is !$(1-τ)+rgτ!$, with the extra term representing the value of the full exemption on capital gains, which is analogous to receiving a credit for the tax that would have been paid if the shares had simply been sold.

Table 1 presents a numerical example to highlight how the tax assistance changes according to the form of donation, and the inclusion rate on gifts of securities. Consider the decision to donate $1,000, either through cash or by gifting a security, and the consumption that must be forgone in each case. An individual donating $1,000 in cash would be eligible for a $460 combined federal-provincial charitable donation tax credit.^[10] Since the tax credit may be applied against other income tax owing, potential consumption has only decreased by $540 by donating. Suppose instead the individual had previously acquired a security at a cost base of $200, which has appreciated to a fair market value of $1,000. The capital gains portion of the asset subject to inclusion represents pre-tax income, and this must be converted to after-tax income to calculate forgone consumption. In column (1), the base capital gains inclusion rate is 75% and the inclusion rate on securities is 100%. By selling the asset, the individual receives $724 in after-tax income ($1,000 in proceeds less $276 in capital gains tax). If donated instead, the same amount would be paid in capital gains tax, but the individual would receive $460 through the CDTC, yielding $184 in after-tax income. The $540 difference in after-tax income between the sale and donation represents the potential consumption forgone by the donor. This is expressed more succinctly as the value of the asset less the sum of the tax credit and tax saved due to the reduced inclusion rate. Since the individual does not reduce his or her tax owing by choosing to donate when m=1, the after-tax price is equivalent to that of the cash donation. Column (2) demonstrates the effect of halving the inclusion rate on gifts of securities. Under this regime, donating the asset only incurs $138 in tax on capital gains. With the same $460 credit, after-tax income rises to $322, and the potential consumption forgone by the donor is correspondingly reduced to $402 ($724 minus $322). In column (3), the base capital gains inclusion rate falls from 75% to 50%. Selling the asset now results in less capital gains tax, and so for the same 50% inclusion rate on gifts of securities, donors save less in tax, resulting in an increased after-tax price of giving ($448). Finally, column (4) presents a scenario where the inclusion rate on gifts of securities is reduced to nil. Choosing to donate grants a tax saving of $184, and the after-tax price of giving is reduced to $356.

The first major tax reform relating to charitable donations was the introduction in 1930 of a deduction for charitable donations, up to a ceiling of 10% of the taxpayer's net taxable income.^[11],^[12] Numerous modifications to the tax treatment of charitable donations have been made since, detailed descriptions of which can be found in Bird and Bucovetsky (1976) and Duff (2001). No separate legal provisions relating to gifts-in-kind, including PLS, existed in the Act prior to 1972. While gifts-in-kind deductible at fair market value were possible in theory since 1930, in practice such gifts were rare and it was generally understood that National Revenue would not permit the deduction in many cases.^[13] Beginning around 1972-1973, it was clarified that individuals could deduct up to the full market value of the donated security, but had to pay the capital gains tax on any gains.

In 1987, a major conceptual change occurred when the deduction in respect of charitable donations was converted into a tax credit (the CDTC). To maintain similar tax incentives under the credit, the credit rate for donations over $250 was set at 29% so that the credit for these amounts would still function essentially as a deduction.^[14] Donations of PLS continued to be taxable in respect of any capital gains and eligible for the CDTC at full market value throughout this time.

This situation persisted until 1997, at which time a preferential tax treatment for capital gains on gifts of PLS was introduced in Budget 1997. The inclusion rate applied to the capital gains on donations of PLS was cut in half for donations made to charitable organizations and public foundations (but not to private foundations). The measure was originally enacted on a temporary basis (it included a "sunset clause" set to expire after five years), with extension to depend on whether the measure was found to have increased donations overall and whether the additional donations were distributed fairly among charities. Budget 2001 announced that the measure had met these objectives, and the one-half inclusion rate was made permanent.

Budget 2006 replaced the reduced capital gains inclusion rate with a complete exemption (i.e., a capital gains inclusion rate of zero) for donations of PLS. Table 2 presents the effective capital gains inclusion rate on gifts of PLS over time as these reforms took effect.

In the following year, the exemption was extended to donations made to private foundations, while subjecting the latter to excess business holdings rules designed to mitigate potential conflicts of interest.^[15] Budget 2008 further extended the zero inclusion rate to donations of unlisted exchangeable securities, provided they are exchanged for PLS and gifted within 30 days.

The evaluation proceeds by way of an analysis of tax return data, providing more information to assess whether the policy has been effective, and taking into account considerations of equity and efficiency.

This section considers the effectiveness of the measure in encouraging charitable giving. A previous evaluation of charitable donations estimated the overall price elasticity of individuals at -1.1, indicating that a 10% reduction in the after-tax price of giving would be expected to lead to an 11% increase in charitable donations (Department of Finance Canada, 2016).^[16] The full capital gains exemption on gifts of securities effectively lowers the after-tax price of giving for such assets (as can be seen in Equation (2) above when the case of m=1 is compared to the case when m=0), and is therefore expected to encourage more giving. In addition, the literature suggests that a lower overall after-tax price for donations is associated with a positive income effect, which could also lead to an increase in donations.^[17]

However, the relative after-tax price of giving securities is also of significance. That is, for donors considering a donation of cash or a donation of securities, a decrease in the after-tax price of giving securities relative to cash would be expected to increase donations of the former relative to the latter, so long as the donor is in a position to donate securities. Very few studies have considered this issue, but a U.S. study by Eaton and Milkman (2004) concluded that there is a substantial elasticity of substitution between cash and noncash gifts among donors, i.e., that donors will adjust the type of gift made according to the relative after-tax price. If donors simply substitute donations of securities for cash donations while maintaining a constant amount of contributions, this would result in a higher tax expenditure, and would not increase the aggregate level of donations.

This section of the paper proceeds by presenting some statistics on PLS donations since 2001, which can be used to inform the extent to which overall donations have increased, or have simply been substituted from cash into securities, as a result of the elimination of tax on capital gains of donated securities in 2006. Aggregate trends, which are consistent with the price and income elasticity estimates in the literature, are presented first. PLS donations, benefiting from a lower after-tax price than cash donations, are found to have a growth rate far outstripping that of claimed donations in the sample period, even through the 2008-2009 recession. Moreover, immediately following the full capital gains exemption in 2006, the number of PLS donors and donated value of securities increased appreciably. Against this, to assess whether the change in the relative after-tax price has led to substitution of securities for cash donations, this section proceeds by examining the evolution of donors' portfolio of gifts. It is observed that gifts of securities take up a larger share of donations following the full capital gains exemption in 2006. To better isolate any behavioural change, the analysis then abstracts from the new donors appearing each year, and follows repeat donors over time for the change in the value and composition of their donations. Repeat donors first appearing before the policy change primarily add to gifts other than securities between 2001 and 2005, but switch to increasing their overall donations through additional gifts of securities after 2006.

Individuals

Chart 1 illustrates the total value of claimed eligible donations and gifts of PLS between 2001 and 2015.^[18], ^[19], ^[20] The number of individuals claiming donations in each year is also represented. In real terms, total claimed donations grew from an estimated $7.47 billion in 2001 to $9.48 billion in 2007. A decline is observed during the recession, along with muted growth thereafter; as of 2015, claimed donations stood at $9.71 billion, and the number of donors had declined by approximately 300,000 from its peak of 5.91 million in 2008.^[21] Donations of PLS, while modest relative to claimed donations, grew rapidly between 2001 to 2007, from $144 million to $643 million. Despite a near-halving of value the following year, gifts of securities recovered to reach a maximum of $822 million in 2014. Moreover, between 2001 and 2015, gifts of securities grew from 1.9% to 6.7% as a share of claimed donations.

Table 3 reports the average yearly growth rate of claimed donations and gifts of securities. Claimed donations averaged a 2.4% yearly growth rate between 2001 and 2015, but gifts of securities grew six times faster, leading to an increase in the proportion of securities among the mix of gifts over time. Partitioning between the pre- and post-2006 periods reveals that in the former period, the ratio of the growth rates was narrower, at 5.2. From 2006 onwards, however, the growth rate of gifts of securities and claimed donations both decline, but the ratio widens to 7.5. Starting from the trough of the recession, both growth rates recover somewhat, and the ratio returns essentially to its 2001-2005 level.

Chart 2 presents summary data on donors of PLS, where new donors refer to individuals making a contribution for the first time within the sample period, while repeat donors are those having gifted at least once previously. Between 2001 and 2015, donations of securities totalled $6.06 billion. There was a substantial increase in both the number of donors and value of PLS donations in 2006 and again in 2007. Moreover, in 2006, when the full capital gains exemption was introduced, the new donor share peaked at 66%. The share subsequently declined, a fall which is also observed in terms of the share of new PLS donation value. Thus, from 2008 onwards, the majority of donors of PLS are repeat donors and the majority of the value is attributable to them.

In Tables 4 and 5, the data on individual donations is separated between gifts of securities and other donations. Other donations are derived by subtracting gifts of securities from claimed donations. Although gifts of securities must be reported in order to assess the gain eligible for a reduced (zero) inclusion rate, the donor can claim the CDTC in any of six years (the year of the donation plus the five following years). Therefore, the data are limited to those observations where the value of claimed donations was at least equal to the value of donated securities.^[22]

Table 4 illustrates the mean and median value of gifts made by new and repeat donors in the pre- and post-2006 periods, sorted by type of gift. Repeat PLS donors are found to claim larger donation amounts, both in terms of securities and other donations, across both periods. Between 2001 and 2005, securities comprised on average 57.5% of the value of claimed donations for repeat donors, and 65.1% for new donors. From 2006 onwards, these proportions increased to 68.2% and 67.1%, respectively. Since the average value of claimed donations declined slightly among repeat donors, this is indicative of a substitution effect for this group in favour of securities. It should be noted, however, that the value of gifts claimed by new donors increased between the two periods.

Evidence of a substitution effect can also be found by examining gifting patterns among repeat donors of PLS over time—in particular, before and after the introduction of the full capital gains exemption. Table 5 illustrates the change in the value of donations among repeat donors, according to their cohort (i.e., the period in which their first donation of securities was made) and donation period. As donations of PLS tend to be irregular, the change is calculated as the difference in the value gifted between two periods, whether consecutive or not. Pooling all repeat donors together, an average positive change of $12,200 is observed over the period spanning 2001-2015, with securities accounting for 85.5% of the increase. Disaggregating between earlier and later donation periods reveals that previously, the increase in donation value was primarily through other gifts, while securities became the principal driver in the latter period. This pattern is also found among the 2001-2005 cohort. Meanwhile, repeat donors who first donated in 2006 or later typically increase their gifts of securities at the expense of other forms of contributions.

Overall, the T1 return data indicate that donations of PLS have increased substantially in value, and that the growth rate of PLS donations has been higher than that of other donations, particularly in the face of the 2008 downturn. This evidence is in line with the measure's aim to incentivize giving of securities. Consistent with the possibility of substitution, however, donations of securities are found to be an increasing share of the mix of claimed donations. Moreover, securities lead the growth in the donations of repeat donors from 2006 onwards and, among the cohort appearing after the full exemption, other donations are typically decreasing over time.

Corporations

Chart 3 represents the total value of claimed donations and the gain on gifts of securities. Note that in the case of corporate returns, the value of gifts made in the form of PLS is not designated, only the associated capital gain. The data are therefore not directly comparable to those of the T1 returns. The average growth rate of corporate donations between 2002 and 2014 was 5% per annum, and gains on gifts of securities as a proportion of claimed donations increased from 2.6% in 2002 to 22.5% in 2015.^[23]

The average yearly growth rates of the different components of donations are represented in Table 6. Corporate donations of securities grew 3.9 times faster than total claimed donations between 2002 and 2005. This ratio increases to 6.9 in the 2006-2015 period, indicating a more dramatic divergence in growth rates compared to the earlier period. Limiting the later period to years following the recession (2010-2015) eliminates this effect, revealing a ratio of only 2.6, as the growth rate of claimed donations rebounded while that of donations of securities slowed.

Chart 4 summarizes data on corporate PLS donations, separating new and repeat donors. Corporations claimed some $6.43 billion in gains on gifts of securities between 2001 and 2015. As with individual donors, there was an increase in the share of new donors and in the share of the value of PLS donations attributable to new donors in 2006 and 2007, both of which subsequently decline. The shift in 2006 is particularly striking in terms of the value of PLS donations generated by new donors, which jumps from 25% to 58% of the total in one calendar year. Also of note is a sharp increase in the share of PLS donation value attributable to new donors in 2014 and 2015.

As with individual donors, data on corporate gifts can be disaggregated between gifts of securities and other donations. Other donations are calculated as the difference between the value of charitable donations claimed for deduction and the gain on donations of securities (the data are limited to those observations where this difference was non-negative).^[24] Since the gain on donations of securities can be no greater than the fair market value of the securities, the calculation likely overstates the value of other donations. Nonetheless, the data reveal a change in behaviour following the introduction of a full capital gains exemption.

Table 7 shows the mean and median value of gifts made by corporations before and after the full capital gains exemption was introduced. Between 2001 and 2005, corporate donors—whether new or repeat donors—were gifting the majority through donations other than securities. From 2006 onwards, the composition is reversed, with donations of securities now responsible for most of the gifted value. This holds at both the intensive (repeat donors) and extensive (new donors) margins. Also observed is a small decrease in the real value of overall donations made by repeat donors between the two periods, and a small increase among new donors.

As with individual donors, it is possible to follow corporate donors who make repeat donations of PLS over time. Table 8 shows that, across the whole sample period, overall donations are typically increasing in value, but this is entirely driven by growth in donations of securities—other donations are, on average, decreasing in value. However, between 2001 and 2005, the average change in donations was positive for both components. It is only after the introduction of the full capital gains exemption that other donations begin declining in value among repeat donors from both the early and later cohorts. The cohorts are, despite this, not identical in their gifting behaviour. Repeat donors who first appear after 2006 typically make smaller gifts of securities over time,^[25] but this is more than offset by the large increases in security donations made by repeat donors from the early cohort.

The T2 data reveal broadly the same trends as those seen among individual donors. There was a large increase in the value of the gain on gifts of securities among corporate donors between 2001 and 2015. In addition, the slowdown of growth coincident with the recession was less marked for securities compared to claimed donations. On the other hand, the composition of gifts switches to being led by securities following the full capital gains exemption in 2006, among both new and repeat donors. In addition, repeat corporate donors are, on average, decreasing non-security gifts in this latter period while increasing their donations of securities.

The two aspects of equity addressed here relate to: (a) the vertical and horizontal equity of the measure, including measures of progressivity; and (b) the degree to which donations of securities are equitably distributed across charities—in terms of both the sector and size of charities—relative to cash donations. This section proceeds by addressing these equity issues, with the discussion grouped under the following headings: individuals, corporations and charities.^[26]

Individuals

The relevant vertical and horizontal equity implications of the measure can be illustrated by slightly modifying the after-tax price of giving securities expression from Equation (2) as follows:

!$P(r,g,s_i,τ_i)=[(1-s_i)]-[rgτ_i]!$ (3)

The expression for donating cash remains:

!$P(r,g,s_i,τ_i)=[(1-s_i)]!$ (4)

Note that the restriction m=0 has been imposed (relevant beginning in 2006 in Canada). In addition, the CDTC rate is now indexed by i=L,H, such that !$s_L!$ refers to the lower credit rate on the first $200 of donations and !$τ_L!$is the higher credit rate on additional donations. Similarly, the marginal tax rate is represented as either a low rate !$τ_L!$, or a high rate !$τ_H!$. Table 9 contrasts some special cases using the above expression for two types of donors: Donor A who donates cash or PLS, and Donor B who always donates PLS.

Horizontal equity is the idea that two taxpayers with similar incomes should pay a similar amount of tax. With respect to the measure related to PLS, horizontal equity can be measured by the degree to which two donors in the same bracket, one donating cash and one donating securities, face similar tax treatment. From Case 1 in the table above, the difference in after-tax price between two individuals with the same marginal tax rate, !$τ_i!$, can be expressed as !$rgτ_i!$ It is clear from this expression that the difference in the after-tax price between the situations increases in the value of the marginal tax rate. That is, the higher the marginal tax rate (or taxable income) of two similar donors, the greater is the horizontal inequity. The intuition behind this is straightforward: a higher marginal tax rate implies a greater benefit of not being subject to capital gains tax at this higher rate.

Vertical equity is respected when higher-income taxpayers pay at least as great a proportion of tax on their marginal income. Both proportional and progressive taxation respect the idea of vertical equity. Regressive taxation, on the other hand, violates vertical equity. It turns out that the expression for Case 1 can also be used to measure vertical equity where Donor A donates cash in excess of $200 (because the value of P for these cash donations is independent of the Donor A's tax rate).^[27]

Case 2 considers a different type of potential vertical inequity created by the measure: this time, both Donor A and Donor B donate securities, and A is assumed to have a lower tax rate than B. In this case, the difference in the after-tax price is !$rg(τ_H-τ_L)!$. So long as is sufficiently greater than zero, this expression is smaller than the expression in Case 1, and thus there is less of an inequity between A and B. This is due to the fact that A now also gets some benefit for not being subject to capital gains tax.

The final case presented in Table 9—where Donor A is in a low tax bracket and donates less than $200 in cash, and Donor B donates securities and is in a higher tax bracket—is relevant empirically, as about 40% of tax filers claiming the CDTC claim total donations of less than $200 (represented by !$s_L),^2!$),^[28] and about three-fifths of donors of PLS have taxable incomes greater than $150,000 (see Table 10).^[29] In this case, the difference in after-tax price can be represented as !$(s_H-s_L)+rgτ_i!$, the greatest value of the three expressions in the table. Assuming r tends towards a value of 1(to approximate the case where f is large relative to c), and taking the 2015 federal values of the other parameters, this after-tax price differential is equal to 0.285. In other words, for a given dollar donated in Case 3, the after-tax price of giving is 28.5 cents lower for Donor B.

There are other dimensions by which the measure is seen to benefit individuals unequally. Table 10 also provides a disaggregation by age group, which reveals that 65.3% of donors were above the age of 65, and a further fifth were between 56 and 65 years of age. Together, these two groups accounted for 86.9% of the donated value and exempt capital gains on gifts of PLS.

Income and age breakdowns are also provided in Tables 11 and 12, respectively, but are complemented by a disaggregation by sex. Overall, PLS donors are 58.6% male and 41.4% female, and their shares of donated securities are 68% and 32%, respectively (with nearly identical shares for the exempt capital gains). Table 11 indicates that the number of donors is similar by sex for those earning up to $150,000, but increasingly tilted towards men in the higher income categories. Turning to Table 12, there is clear increasing trend of donors by age group for both sexes. There are approximately two to three times as many men donating PLS shares in all but the highest age category. Among donors more than 65 years old, the shares narrow to 54.1% male and 45.9% female, as does the value of donated securities (to 52.7% and 47.3%, respectively). In all age groups, men are seen to have a higher median donation, which likely reflects their larger presence among high-earning PLS donors.

Corporations

Chart 5 presents the distribution of the capital gains eligible for exemption on donations of PLS made by corporations, according to their non-farm revenue group. In 2015, corporations with non-farm revenue exceeding $10 million accounted for 71% of the exempted capital gains on donations of PLS. In most of the years under review, only a negligible amount was claimed by non-Canadian-controlled private corporations (CCPCs).

Charities

Registered charities report the total value of gifts-in-kind received in a year for which tax receipts were issued, and must identify each type of gift-in-kind received (e.g., books, machinery, PLS). However, since only the total value across all gifts-in-kind is specified, it is not possible to determine the value attributable to any particular class of gift. It is nonetheless possible to construct lower and upper estimates for the true value of gifted PLS. The lower estimate is limited to charities' gift-in-kind value where the only type of gift identified is PLS. The upper estimate also includes the gift-in-kind value for charities which recorded other types of gifts alongside securities.^[30]

The distribution of gifts of PLS by recipient organization size is shown in Table 13.^[31] In 2015, approximately two-thirds of the value of these gifts was received by charities with annual revenue exceeding $10 million. Moreover, the median donation received by charities of this size was eight times larger than the overall median. By comparison, charities with revenue no greater than $250,000 received roughly 1% of the value of gifts of PLS.

To summarize, the results presented show that the associated tax assistance has benefited a narrow subset of taxpayers, whether individual or corporate, and that the donated securities are largely directed to charities with high annual revenue. However, as discussed in the following section, a large proportion of these charities are foundations, and to the extent that foundations allocate funding to smaller charitable organizations, the distribution presented above underestimates the benefits received by these smaller entities.

As with most tax measures, the capital gains exemption on PLS is cost-effective, or efficient, if the cost of generating a dollar of donations is minimized, or is the lowest cost among a set of alternative policies. This cost of forgone revenues, or fiscal cost, is a direct function of the effectiveness of the measure; the more the measure leads to increased donations while minimizing the loss of otherwise collectible revenue (which includes minimizing substitution from cash donations), the lower the fiscal cost. The first part of this section contains estimates of the fiscal cost for the 2006-2015 period.

Apart from the fiscal cost, another efficiency consideration is in respect of the supply chain linking donations to their ultimate beneficiaries. A supply chain will be less efficient where there are more intermediary steps before funds arrive at the program delivery destination. As demonstrated in the latter part of this section, this is a salient concern because it is foundations—which operate differently from charitable organizations—that ultimately receive most donations of PLS.

Fiscal Cost

Table 14 shows the fiscal cost of the full capital gains exemption measure since 2006, which is comprised of two parts: the non-taxation of capital gains and the tax credit (deduction) associated with donations of PLS.^[32] It is estimated that the non-taxation of capital gains represents approximately $109 million in forgone revenue per year.^[33] In addition, the average annual cost attributable to the CDTC resulting from gifts of PLS is estimated at $146 million.^[34] The tax credit cost applies whatever the form of the donation, while the forgone capital gains revenue is specific to donations being made using PLS.

The forgone revenue from the non-taxation of capital gains is estimated under the assumption of no behavioural change on the part of donors (as is the case for all tax expenditure estimates). It may therefore be overstated, insofar as the measure has encouraged the donation of "long-term" investments, that is, assets which would otherwise have been held for a substantial period of time.^[35] The capital gains accrued up to the present on such assets would only be triggered for tax purposes in the future, at a lower present value, and this would represent the appropriate basis for calculating forgone revenue. On the other hand, donated assets which would otherwise have been disposed of by sale within a short time frame entail forgone capital gain revenue approximately equal to the usual calculation. In particular, substitution of donations of PLS for cash donations will incur less excess fiscal cost than indicated by the forgone capital gains revenue if the assets being substituted are of the "long-term" variety (though this cannot be assessed empirically). Generally, these considerations indicate that the values in Table 12 should be treated as an upper estimate.

Donations to Charitable Organizations vs. Foundations

Efficiency may also be considered in terms of which charities receive donations of PLS, and how these are deployed toward charitable activities. The Canada Revenue Agency differentiates charities by designation: public foundation, private foundation or charitable organization. In 2015, charitable organizations accounted for the vast majority of registered charities (87.6%). By contrast, private foundations represented 6.4% of all charities, and public foundations made up the remaining 6%. Chart 6 shows the share of gifts of securities received by designation between 2003 and 2015, for both the lower and upper estimates of the true value of gifted securities (as described in the previous section). At the lower estimate of gifted securities, the share accruing to either public or private foundations exceeded 80% in most years. At the upper estimate, the proportion increased from a low of 37% in 2004 to a high of 81.4% in 2014, and averaged 63.8% over the period.^[36] According to either measure, the preponderance of gifts of PLS were directed to public or private foundations.

The distribution of gifts of securities within the charitable sector frames the full capital gains exemption (and, previously, the one-half inclusion rate) as a measure which, to the extent that it encourages gifts of securities, largely directs these funds to foundations. This is of potential concern as expenditures on charitable activities, expressed as a share of a revenue, were substantially lower among public and private foundations (13.1%) compared to charitable organizations (74.8%) in the period under consideration.^[37] Gifts of PLS made to foundations therefore represent a less direct pathway to expenditures on charitable activities. However, it should be noted that foundations operate differently than charitable organizations. Foundations tend to invest donations in permanent endowment funds to ensure funding to the charitable sector over the longer term. The income from the endowments subsequently flows out over time as sizeable gifts to other registered charities: between 2003 and 2015, public foundations devoted on average an amount equivalent to 42.4% of their revenue to gifts to qualified donees; private foundations similarly gifted an average of 36.7%.^[38] These patterns of expenditure support the notion that foundations can have a role as an automatic stabilizer for overall charitable funding (Payne, 2012).

It could be argued that foundations are the most appropriate destination for gifts of PLS, on the assumption that these charities operate with a longer-term perspective which best positions them to manage securities as part of a broader and evolving portfolio. Foundations, in this sense, could be contrasted with charitable organizations, which are more oriented towards immediate expenditures on charitable activities. Yet monetizing donations of securities should not be onerous even for small charitable organizations, and foundations will sell PLS as appropriate to maintain a balanced portfolio. That donations of securities are disproportionately directed to foundations, then, either reflects a real preference for gifting securities to foundations (e.g., because the donor wishes to contribute to an endowment fund), or a possible misapprehension that charitable organizations are unable to effectively utilize securities.

This paper presents an evaluation of the non-taxation of capital gains on donations of PLS from the perspectives of effectiveness, equity and efficiency. A review of the measure's effectiveness indicates that observed trends are consistent not only with the possibility of a real increase in the value of overall donations, but also with the possibility of a substitution effect, whereby some donors may be replacing cash donations with gifts of securities. In terms of equity, the evaluation indicates that donations of PLS are primarily made by individuals in higher-income brackets and corporations with high non-farm revenue. The donations primarily benefit large charities which fall under the public and private foundation designations. With respect to efficiency, the true fiscal cost is sensitive to assumptions about the extent to which donors have replaced cash donations with donations of PLS, and about when the asset would have been disposed of in the absence of the measure. It is also observed that, since foundations spend proportionally less on charitable expenditures than charitable organizations, directing gifts of PLS towards foundations generally entails a longer delay before the funds reach their ultimate beneficiaries. On the other hand, through the build-up of endowment funds and granting of gifts to charitable organizations, foundations play a significant role in supporting the charitable sector over the long term.

Andreoni, James (2006). "Philanthropy", in Serge-Christophe Kolm and Jean Mercier Ythier (Eds.), Handbook of the Economics of Giving, Altruism and Reciprocity (Vol. 2), Elsevier.

Andreoni, James and Abigail Payne (2013). "Charitable Giving" in Auerbach, A.J., et al. (Eds.), Handbook of Public Economics (Vol. 5), Newnes.

Batina, Raymond G. and Toshihiro Ihori (2005). Public Goods: Theories and Evidence. Springer Science & Business Media.

Bird, R.M. and M.W. Bucovetsky (1976). "Canadian Tax Reform and Private Philanthropy". Canadian Tax Paper No. 58, Canadian Tax Foundation.

Clotfelter, Charles T. "Charitable Giving and Tax Policy in the U.S.", CEPR Conference Proceedings: Charitable Giving and Tax Policy—A Historical and Comparative Perspective, Paris, 2012. Edited by Fack, Gabrielle and Camille Landais, 34-62.

Department of Finance Canada (2002). "Special Federal Tax Assistance for Charitable Donations of Publicly Traded Securities", Tax Expenditures and Evaluations—2002.

Department of Finance Canada (2014). "Evaluation of the Federal Charitable Donation Tax Credit" Report on Tax Expenditures and Evaluations—2014.

Department of Finance Canada (2016). "An Evaluation of the Response of Individuals to Changes in the Price of Charitable Donations", Report on Federal Tax Expenditures: Concepts, Estimates and Evaluations—2016.

Duff, David G. (2001). "Charitable Contributions and the Personal Income Tax: Evaluating the Canadian Credit", in Bruce Chapman, Jim Phillips and David Stevens (Eds.), Between State and Market (Montreal: Published for the Non-profit Sector Research Initiative by McGill-Queen's University Press, 2001) 407.

Duff, David G. (2003). "Special Federal Tax Assistance for Charitable Donations of Publicly Traded Securities: A Tax Expenditure Analysis", Canadian Tax Journal, 51(2), 925-936.

Eaton, David H. and Martin I. Milkman (2004). "An Empirical Examination of the Factors that Influence the Mix of Cash and Noncash Giving to Charity", Public Finance Review, 32(6), 610-630.

Hossain, Belayet and Laura Lamb (2012). "Price Elasticities of Charitable Giving Across Donation Sectors in Canada: Is the Tax Incentive Effective?", ISRN Economics, Vol. 2012, Article ID 421789.

Innes, William I. (2003). "The Case for Tax Incentives on Gifts of Publicly Traded Securities", Canadian Tax Journal, 51(2), 905-912.

O'Neil, Cherie J., Richard S. Steinberg and G. Rodney Thompson (1996). "Reassessing the Tax-Favored Status of the Charitable Deduction for Gifts of Appreciated Assets", National Tax Journal, 49(2), 215-233.

Payne, A. Abigail (2012). "Changing Landscapes for Charities in Canada: Where Should We Go?" SPP Research Papers, 5(34), 1-23.

Peloza, John and Piers Steel (2005). "The Price Elasticities of Charitable Contributions: A Meta-Analysis", Journal of Public Policy & Marketing, 24(2): 260-272.

Schwartz, Robert A. (1970). "Personal Philanthropic Contributions", Journal of Political Economy, 78(6), 1264-1291.

^[1] The analysis presented in this paper was prepared by Maxime Dufournaud-Labelle, Economist, Tax Policy Branch, Department of Finance Canada, with assistance from Scott Legree, Economist, Tax Policy Branch, Department of Finance Canada. Enquiries regarding Department of Finance Canada publications can be sent to finpub@canada.ca.

^[2] For consistency with the terminology used in the Estimates and Projections section of this report, the term "publicly listed securities" is employed. However, other commentators have referred to the assets targeted by this measure as publicly traded securities (e.g., Innes, 2003).

^[3] The exemption is provided for in paragraph 38(a.1) of the Income Tax Act, R.S.C., 1985, c.1 (5th Supp.), as amended (herein referred to as "the Act").

^[4] A donor making such a contribution is considered to have disposed of the property for proceeds equal to its fair market value. The eligible amount of the gift, i.e., the difference between its fair market value and any advantage received, may be claimed for the Charitable Donation Tax Credit. For example, on a gift of real estate property with a fair market value of $300,000, but which is still subject to a $100,000 mortgage (the assumption of which by the charity is an advantage for the donor), the eligible amount is $200,000.

^[5] See section 118.1 of the Act. Both natural persons and trusts are included in the definition of "individuals" within the Act. In what follows, however, we use individuals to refer to natural persons filing a T1 return.

^[6] Since 2016, donors with annual taxable income greater than $200,000 (indexed yearly to inflation) are credited at 33% on the lesser of the portion of donations exceeding $200 and taxable income subject to the new rate.

^[7] In the year of death and the year preceding death, the limit is increased to 100% of net income.

^[8] See section 110.1 of the Act.

^[9] This Canadian-adapted expression is analogous to that in one of the original papers on the topic of the after-tax price of giving: Schwartz (1970). Note that the theory underlying this formulation of the economic cost assumes that the donor had been planning to sell the securities in the short term. This assumption is challenged by Innes (2003), who argues that the economic cost is overstated in the case of securities that would have otherwise been held for the long term. This issue is revisited in Section 3.4.

^[10] In this simple illustrative example, the federal-provincial CDTC rate and the top marginal tax rate are assumed to take the same value of 0.46. In practice, the values of these parameters will vary by province and by tax filer, and the CDTC rate can be different from the top marginal tax rate. The values used in the calculation of the tax expenditure contained in the Estimates and Projections section of this report use the observed tax filer values.

^[11] There were targeted provisions allowing for deduction of donations to the Patriotic and Canadian Red Cross Funds for a limited time between 1917 and 1920 (Duff 2001).

^[12] The limit on deductions for charitable donations was extended to 20% as part of the 1971 Tax Reform Act. Two further increases, to 50% and 75%, were introduced with Budget 1996 and Budget 1997, respectively.

^[13] See Bird and Bucovetsky (1976).

^[14] As under the deduction regime, the amount credited continued to be limited to 20% of income at the time. See Duff (2001) for more information on this reform.

^[15] The Income Tax Act requires a private foundation holding more than 2% of all outstanding shares in any class of shares of a corporation to monitor its holdings and those of related persons. Sections 149.1 and 149.2 of the Act require the private foundation to divest itself of excess holdings if it and related persons together hold more than 20% of outstanding shares in a class.

^[16] This result is in line with Hossain and Lamb (2012), who estimated an overall price elasticity of -1.7 for Canada, and a cross-country meta-analysis by Peloza and Steel (2005), in which the authors report a mean weighted price elasticity of -1.44.

^[17] Positive income effects are ubiquitous in the empirical literature on patterns of giving. See Batina and Ihori (2005) for a review of results up through the early 1990s and, more recently, Hossain and Lamb (2012) and Department of Finance Canada (2016), which estimated positive coefficients on income using Canadian data.

^[18] Values are expressed in constant 2014 dollars throughout.

^[19] The value of donations from 2001 to 2005 is estimated from the total value of the CDTC.

^[20] Donations made through tax shelters are omitted from the data, as individuals participating in these arrangements are issued donation receipts (for the purposes of the CDTC) with a value in excess of what was actually gifted.

^[21] This trend was also observed in the 2014 edition of this report. It was noted therein, however, that the proportion of filers with net tax payable donating has remained essentially flat. The decline is therefore attributable to a lower number of individuals paying tax (that is, fewer individuals have an incentive to report donations in their return). It may also reflect an increased tendency to pool donations among spouses and partners.

^[22] 75.2% of observations on individual donors of PLS met this criterion.

^[23] For these statistics, we omit the value obtained in 2001 as it was largely the result of a single donation. This donation is also omitted in Table 7 and Table 8.

^[24] 94.5% of observations among corporate donors of PLS satisfied this restriction.

^[25] It should be noted, however, that the negative values in the final row of Table 8 are largely driven by a single donor. If this observation were omitted, the mean change in the gain on PLS donations would turn positive, and the mean change in overall donations and other donations would be closer to zero.

^[26] This paper does not consider the wealth transfer effects of the donated funds, including impacts on overall income and wealth distributions. These effects are a source of important debate in the literature on charitable donations. Reduced capital gains taxation on PLS could have an inequality-reducing impact if it results in: (a) increased donations, and (b) donations that are well-targeted towards lower-income individuals. In Section 3.2, some evidence of the first criterion being met was presented. The second criterion, however, is difficult to measure using available charitable return data. A major reason for this, as discussed in Section 3.4 below, is the majority (and perhaps upwards of 90% in some years) of donations of PLS pass through foundations. Unfortunately, data are not readily available on the extent to which funds disbursed by foundations tangibly benefit low-income households. For this reason, it would be difficult to estimate wealth transfers from high-income to low-income individuals by tracing the path of donations of PLS.

^[27] Note that this illustrative example abstracts away from the introduction of the new income-tested 33% rate introduced in 2016.

^[28] Imputed from Table 5 in Department of Finance Canada (2014).

^[29] Note that $150,000 was near the cut-off for the top federal tax bracket in 2015, the year on which the data in Table 10 are based. Individuals with income over $150,000 in that year accounted for over 90% of the value of donated securities and the associated exempt capital gain. In fact, 62.3% of the value was accounted for by individuals with taxable income exceeding $1,000,000. Only 2.7% of the value of donations of PLS is made by donors with taxable incomes below $50,000,

^[30] In other words, where other types of gifts are listed, the upper estimate assumes that all but a negligible amount of the gift-in-kind value is due to the securities, while the lower estimate assigns this gift-in-kind value to the other gifts. The true value of gifted securities will lie somewhere in between.

^[31] A breakdown by the charity categories defined by the CRA (education, health, etc.) is not provided because, as shown in Section 3.4, gifts of PLS are largely directed to either public or private foundations which fall under the "Welfare" heading. However, the "Welfare" category is not sufficiently representative of the recipient foundations' area(s) of activity as they give to a variety of qualified donees.

^[32] It should be noted that any measure which encourages "net new" charitable giving will increase the costs associated to the CDTC in proportion to the value of new gifts.

^[33] This figure assumes that individuals and corporations would have disposed of the assets in the same calendar year.

^[34] The cost associated with the CDTC assumes the entire value of donated securities was claimed for the deduction in a calendar year. Estimates for the cost associated with the charitable donation deduction for corporations are not available.

^[35] See Innes (2003) for a more detailed exposition, and in which the author suggests that most of the PLS donated as a result of the non-taxation of capital gains are of the "long-term" variety.

^[36] A large amount of the additional value at the upper estimate comes from charitable organizations, increasing their share while lowering that of foundations.

^[37] If we instead consider expenditures on charitable activities as a share of assets, the percentages become 3.1% for foundations and 61.2% for charitable organizations.

^[38] These figures are derived from T3010 data. Expressed instead in terms of assets, the percentage gifted to other charities by public and private foundations averaged 13.7% and 6%, respectively, over this period.

Table of Contents