Report on Federal Tax Expenditures - Concepts, Estimates and Evaluations 2025: part 8

GBA Plus of Federal Tax Expenditures: Considering New Identity Factors Using Data from the Census of Population^{Footnote 1}

1. Introduction

Since the enactment of the Canadian Gender Budgeting Act in December 2018, six GBA Plus studies examining the impact of federal tax expenditures have been published. These studies fulfilled one of the requirements of the Act whereby, once a year, the Minister of Finance must make available to the public an analysis on the impacts of federal tax expenditures in terms of gender and diversity that the Minister considers appropriate. In addition to gender, a number of other identity factors were considered in these publications, including the age group, family type, income group, and area of residence of tax filers. These studies were conducted using T1 tax return data, making it possible to identify these individual characteristics. As T1 returns do not allow for a comprehensive identification of persons with disabilities, a linkage exercise between tax data and data from the Canadian Survey on Disability (CSD) was completed in order to analyze the impact of existing tax expenditures on these individuals.

Other than the "non-taxation of personal property of status Indians and Indian bands situated on a reserve," none of the tax expenditures in the Canadian personal income tax (PIT) system specifically target tax filers based on their ethnocultural diversity characteristics. However, they could indirectly affect certain groups because of the specific experiences they may have in terms of access to the labour market, education, property or wealth accumulation, for example. In the United States, racial inequities that can be created by the tax system are part of the GBA Plus issues that have been examined for some time. Notably, a recent working paper published by the Office of Tax Analysis (OTA) of the U.S. Department of the Treasury concludes that there are disparities in the benefits of some tax expenditures among White, Black, and Hispanic families.^{Footnote 2} In Canada, very little research has been done on the impact the tax system may have based on immigration status or ethnicity, in part because of limited access to the data needed to undertake such research.

To overcome data limitations, the Department of Finance Canada partnered with Statistics Canada to undertake a second linkage exercise between data from the 2021 Census of Population and T1 tax return data. This linkage provides new insights on the redistributive impact of the PIT system based on identity characteristics that could not be studied until now, namely immigration history, Indigenous identity, and racialized identity. It also allows us to extend the scope of previous GBA Plus studies by examining the impacts of the tax system by gender identity (cisgender man, cisgender woman, or transgender or non-binary person) rather than by sex assigned at birth (male or female).

The next section (Section 2) provides more information on the data used in this study. It also discusses the tax expenditures considered, the various identity groups examined, and the methodologies chosen for the empirical analysis. Section 3 then presents the detailed results of the impact analysis by gender identity, immigration history, and ethnocultural diversity. Section 4 concludes the study.

2. Background information

2.1 Data used and tax expenditures considered

In Canada, a Census of Population is conducted every five years in order to provide a detailed statistical portrait of the Canadian population.^{Footnote 3} The census is the main source of sociodemographic data for several groups, including Indigenous Peoples, immigrants, and racialized groups. Over the past several years, the Department of Finance Canada has worked with Statistics Canada to ensure that a number of tax variables are matched to selected data from long-form census questionnaire respondents^{Footnote 4} who filed an income tax return for the 2020 tax year.^{Footnote 5} The new linked database contains about 40 variables taken from the census (all those enabling the identification of identity groups of interest) and about 100 variables taken from T1 returns (all those enabling the examination of the overall impact of the federal PIT system and of a number of individual tax expenditures on the redistribution of income among these groups^{Footnote 6}). The linked database was made available to the Department of Finance Canada for analysis through a special data access agreement that met all the requirements established by Statistics Canada.^{Footnote 7}

The list of the 55 federal PIT expenditures examined in this study is similar to that in the GBA Plus studies published in 2019 and 2021. It includes all tax expenditures for which claimants and beneficiaries can be identified in the linked dataset and for which a breakdown based on the new identity factors of interest is possible, in accordance with the confidentiality standards established by Statistics Canada. The four employment expense deductions included in the 2021 GBA Plus^{Footnote 8} were excluded from the analysis because the linked database does not contain the necessary variables to study them separately. The impacts of these measures are nevertheless generally captured by the deduction for "other employment expenses."^{Footnote 9} Two other tax expenditures—the disability supports deduction and the Adoption Expense Tax Credit—were excluded from the analysis because population sizes were often too small to protect tax filer confidentiality or to ensure results reliability. To avoid excluding them for these same reasons, certain tax expenditures were grouped together, namely:

1. The Lifetime Capital Gains Exemption and partial inclusion of capital gains;
2. Capital loss and non-capital loss carryovers as well as the deduction of allowable business investment losses; and
3. The tax-free amount for emergency services volunteers, the Search and Rescue Volunteers Tax Credit and the Volunteer Firefighters Tax Credit.

Conversely, three PIT expenditures were added to the list, namely the "credit for subscriptions to Canadian digital news media" and the "Canada Training Credit"—two tax expenditures introduced after 2018—along with the PIT portion of the "non-taxation of personal property of status Indians and Indian bands situated on reserve"^{Footnote 10}—a tax expenditure that has existed for a long time, but for which no data was collected prior to the 2019 tax year. Since 2019, individuals who report having earned income exempt under the Indian Act have been asked to complete Form T90 to provide information on that income. Completing this form is not mandatory in order to benefit from the "non-taxation of personal property of status Indians and Indian bands situated on reserve." However, this information is requested on the T1 income tax return so that the Canada Revenue Agency (CRA) can calculate the limit for the Canada Training Credit and the Canada Workers Benefit, as applicable, and provinces and territories can calculate certain family benefits. Since then, Form T90 includes data that can be used to estimate the tax savings resulting from this exemption. It should be kept in mind, however, that the non-mandatory nature of this form likely leads to an incomplete reporting of exempt amounts. In addition, 2021 census data are not available for 63 incompletely enumerated Indian reserves and settlements^{Footnote 11} and the tax filing rate tends to be lower among the Indigenous population. These factors may result in an underestimation of the tax savings arising from the "non-taxation of personal property of status Indians and Indian bands situated on reserve."

Furthermore, while amounts exempted and deducted under pension income splitting were considered in this analysis, the transfers of tax withholdings associated with these amounts could not be taken into account due to a lack of relevant information in the linked dataset. Without this information, any conclusions about the redistributive impact of this tax expenditure between groups are considered less relevant.

2.2 Identity characteristics considered

Table 1 presents all the identity characteristics examined in this study, namely gender identity, immigration history, and ethnocultural diversity. It also provides details on how people were grouped based on these characteristics.

It should be noted that the recent immigrant group—including those with permanent or non-permanent resident status—has at times also been analyzed taking into account the following four admission categories: 1- Economic; 2- Sponsored by family; 3- Refugee; and 4- "Other" category and non-permanent resident.

The same applies to people of Indigenous identity who, on some occasions, were examined separately depending on whether or not they had Registered or Treaty Indian status^{Footnote 13} or identified themselves as: 1- First Nations; 2- Métis; 3- Inuk (Inuit); or 4- "other" Indigenous identity (i.e., multiple Indigenous responses and not included elsewhere).

In census products, racialized people are often categorized in the following groups: South Asian, Chinese, Black, Filipino, Arab, Latin American, Southeast Asian, West Asian, Korean, Japanese, and visible minority not included elsewhere or multiple visible minorities.^{Footnote 14} Due to statistical reliability issues and the need for additional data analyses, this type of breakdown was not highlighted in this study but will be considered in future research.

On the other hand, because the age profiles of the identity groups examined sometimes differ considerably from those of the rest of the population, the potential interactions between the impacts of belonging to these groups and the impacts of age are explored further in this study.

Other than the "non-taxation of personal property of status Indians and Indian bands situated on reserve", none of the tax expenditures considered specifically target the identity groups examined in this study. However, it can be expected that the demographic and socioeconomic characteristics of these groups will result in certain components of the system having more or less significant impacts for some of the groups.

2.3 Methodologies used to analyze the impacts of the federal PIT system

As with most of the previous GBA Plus studies, this study seeks to quantify the overall redistributive impact of the federal PIT system and its major components (i.e., adjustments, exemptions, deductions, the rate structure, non-refundable credits, and refundable credits) between various identity groups. It also aims to determine whether specific tax expenditures provide a greater or lesser benefit to some of these groups.

2.3.1 Overall redistributive impact

To have an overall redistributive impact on a particular identity group, the federal PIT system must:

• Reduce—through adjustments, exemptions^{Footnote 15}, and deductions or deferrals—the share of total taxable income that this group holds in comparison to its share of total pre-tax income; and/or
• Lower—through the progressive tax rate structure, non-refundable, and refundable credits—the share of tax payable owed by this group relative to that owed by other groups.

To determine the overall redistributive impact of the federal PIT system and its major components, we therefore compared the total amounts and shares of various concepts of income and tax payable among the various identity groups examined.

2.3.2 Redistributive impact of tax expenditures

To determine the redistributive impact of each of the federal PIT expenditures included in this analysis, we first estimated the number of beneficiaries and the total amount of benefits received for each of these expenditures for each of the groups examined. The "number of beneficiaries" refers to the number of tax filers who, in the absence of a particular tax expenditure, would have had to pay a higher amount of net federal tax^{Footnote 16}, all else being equal.^{Footnote 17} The "total amount of benefits" refers to the amount of net federal tax saved by tax filers due to the tax expenditure, all else being equal.

Next, ratios of the share of total benefits of a tax measure received by a group relative to the share of total pre-tax income reported by that group were calculated. As with our previous GBA Plus studies, we have retained these ratios as the preferred indicators since they control for pre-existing income inequality between the different identity groups examined. In other words, these ratios allow us to consider the impact of the tax system on the distribution of income between the groups, independently of the impact of observable income gaps before the application of the federal PIT system.

3. Empirical results for the various identity groups

Notes to readers

In this study, all shares have been rounded to the nearest tenth or hundredth. Population counts and amounts—which are all in 2020 dollars—have been rounded according to the random rounding standards (up or down) established by Statistics Canada's Centre for Income and Socioeconomic Well-being Statistics. As estimates are rounded separately, the sum of the values may not add up to totals shown or to 100%.

The new dataset enables the production of a very wide range of new statistics based on gender identity, immigration history, ethnocultural diversity, and several other intersecting identity factors. For practical reasons, only a limited subset of statistics is analyzed and discussed in this study. We will continue to explore ways to further tap the potential of the new dataset in future years to improve our understanding of the gender and diversity impacts of tax expenditures as required by the Canadian Gender Budgeting Act. 

3.1 Gender identity

Following an extensive consultation process with Canadians, questions on gender were included for the first time in the 2021 Census of Population. In this latest census, the question on sex specifies that "sex" refers to sex assigned at birth. A new question regarding the "gender" with which people identify was added, clarifying that gender may be different from sex assigned at birth and may be different from what is indicated on legal documents. As a result, for the first time, it is possible to examine the impact of the federal PIT system on tax filers who identify as transgender or non-binary.

3.1.1 Profile of tax filers by gender identity

Table 2 provides an overview of the adult census population who filed an income tax return for the 2020 tax year by gender identity. It shows that 0.3% of this population identified as transgender or non-binary, while the vast majority identified as cisgender women or cisgender men (51.6% and 48.1%, respectively).

The distribution of members of these three groups by demographic and socioeconomic characteristics shows that tax filers identifying as transgender or non-binary are significantly younger on average than the rest of the population. In the last census, the average age of people in this group was 38.5 years, while the average age for all tax filers was almost 50 years. Consistent with their age profile, transgender or non-binary people were also less likely to live with a partner and/or have dependent children. While their education profile did not differ substantially from that of the rest of the population, transgender or non-binary people reported average personal and family incomes that were significantly lower than those of cisgender men and slightly lower than those of cisgender women. There was also a slightly higher concentration of transgender or non-binary people in urban areas compared to the rest of the population, and they were more likely to be renters than owners.

Data released by Statistics Canada in 2024 showed a greater propensity for younger generations to report being transgender or non-binary.^{Footnote 18} To explain this trend, Statistics Canada suggests that "[i]t is possible that older people are less likely to report being 2SLGBTQ+^{Footnote 19} because of having lived a significant part of their life in a society that was less accepting of gender and sexual diversity."^{Footnote 20} 

3.1.2 Overall redistributive impact by gender identity

a) Among all tax filers

Table 3 shows that, like cisgender women, transgender or non-binary people report a lower share of total pre-tax income than the share of tax filers they represent (0.19% versus 0.27%). On the contrary, and as has been observed for male tax filers in previous studies, cisgender men report a higher share of pre-tax income than the share of the tax filer population they represent (57.7% versus 48.1%).

Results in Table 3 confirm that the federal PIT system is globally redistributive towards people who identify as cisgender women, as is the case with female tax filers (that is to say, tax filers who were assigned female at birth). In 2020, cisgender women's share of after-tax income was 1.8 percentage points (pp) higher than their share of pre-tax income. This finding is virtually identical to that of the 2019 GBA Plus study^{Footnote 21} and suggests that the economic disruption caused by the COVID-19 pandemic does not appear to have significantly changed the magnitude of the redistributive impact of the tax system between men and women.

Given that transgender or non-binary people represent a very small proportion of all tax filers, it is difficult to observe the redistributive impact of the federal PIT system based on the percentage point change in their share of income before and after the application of the tax system. This impact can be better observed by calculating a ratio of the share of income held by this group after the payment of federal tax relative to the share it held before this payment. This ratio, which appears in the last line of Table 3, shows that the share of total after-tax income held by this group was 1.03 times higher than its share of pre-tax income. This result indicates that the federal PIT system effectively had an overall redistributive impact towards this group in 2020. Overall, the magnitude of this redistributive impact is close to that observed towards cisgender women according to this same indicator (i.e., 1.04). Chart 1 shows that the 2020 PIT system had the effect of reducing the average income gap between the transgender or non-binary group and that of cisgender men by $5,400. This reduction is slightly lower than that observed between cisgender women and cisgender men ($5,700).

A comparison of shares held for the various concepts of income presented in Table 3 shows that, among the major components of the federal PIT system, the progressive rate structure and non-refundable credits had the most significant redistributive impacts towards transgender or non-binary tax filers in 2020. This stands in contrast with refundable credits having the most significant redistributive impacts towards cisgender women.

b) By tax filers' age group

Given that transgender or non-binary people's age profile differs considerably from that of cisgender people, it is worth asking whether the observed overall redistributive impact towards this group is more closely tied to the impact of age than the impact of gender identity. To shed some light on this question, we repeated the comparison exercise among tax filers in different age groups: first among those under age 40 (who comprised nearly two thirds of tax filers who identified as transgender or non-binary), and then among those aged 40 and over.^{Footnote 22}

This second comparison exercise (Table 4) shows that the PIT system continues to have a redistributive impact towards transgender or non-binary people when age group is taken into account, albeit to a slightly lesser degree. Specifically, the ratios of the share of total after-tax income held by these individuals relative to their share of pre-tax income are lower within both age groups considered than among this group as a whole (1.02 and 1.01 among tax filers aged 18 to 39 and 40 and over, compared to 1.03 among tax filers of all ages). This translates into a reduction in the gap between transgender or non-binary people's and cisgender men's average income of about $4,300 among the youngest tax filers (compared to a reduction of $5,400 among all tax filers), as well as a reduction of $4,100 among older tax filers. These results indicate that the specific age profile of transgender or non-binary people does indeed play a role in explaining how the federal PIT system redistributes income towards this group.

3.1.3 Redistributive impact of tax expenditures by gender identity

a) Among all tax filers

For each of the tax expenditures selected for the purposes of this study, Table A.1 (presented in Annex A) shows the share of benefits received by all cisgender women, cisgender men, and transgender or non-binary people relative to the share of pre-tax personal income reported by each of these groups. The resulting ratios allow us to identify the PIT expenditures that especially benefit each of the gender identity groups considered, and thus those that have contributed to the redistribution of income between these groups.

Table 5 below lists the top six federal PIT expenditures for transgender or non-binary tax filers in 2020, based on the ratios presented in Table A.1 and without taking into account the total cost of each tax expenditure. At the top of the list is the non-taxation of social assistance benefits, followed by the Refundable Medical Expense Supplement, the Canada Workers Benefit and three education-related tax expenditures (the Tuition Tax Credit – for self, the Student Loan Interest Credit and the Canada Training Credit).

In general, Table A.1 shows that the tax expenditures that benefit transgender or non-binary people relatively more are similar to those identified as benefiting younger tax filers.^{Footnote 23} These are associated with the recognition of certain costs related to employment, education, moving or purchasing a first home. As mentioned earlier, transgender or non-binary tax filers are much younger on average than other tax filers. It is therefore possible that their specific age profile better explains these results than the fact that they identify as transgender or non-binary (see the following section for a more in-depth analysis by age group).

Since the Basic Personal Amount (BPA) represents a larger proportion of the pre-tax income of lower-income groups, transgender or non-binary tax filers benefit relatively more from the BPA. This group also especially benefits from most refundable credits since they generally target lower-income populations. The only exception is the Canada Child Benefit, which was expected given that, overall, transgender or non-binary tax filers are less likely to live in families with children. This likely explains why, unlike for female tax filers, it was not refundable credits, but rather the progressive rate structure and non-refundable credits that, among the major components of the federal PIT system, had the most important redistributive impacts towards transgender or non-binary people in 2020.

While the 2021 study found that older tax filers benefit relatively more from the Political Contribution Tax Credit, Table A.1 shows that transgender or non-binary people are also a group that especially benefits from the credit, despite their relatively young age profile. This suggests that this group participates proportionally more in the donation aspect of the electoral process.  

b) By tax filers' age group

To determine whether the conclusions drawn in the previous section remain similar regardless of the age group of tax filers considered, ratios of the share of benefits received by transgender or non-binary people relative to their share of pre-tax personal income were also calculated among tax filers under 40, and then among those aged 40 and over (Annex A, Table A.2). Although a larger number of ratios by age group could not be disclosed due to confidentiality standards established by Statistics Canada's Centre for Income and Socioeconomic Well-being Statistics, certain observations could be made based on the ratios that could be disclosed.

Among other things, ratios by age show that the majority of the conclusions in the previous section remain unchanged, regardless of the age group of tax filers considered^{Footnote 24}. They confirm that the six federal PIT expenditures identified as benefiting the most transgender or non-binary people among all tax filers (including those recognizing education-related costs) also especially benefit this group among tax filers under 40 and those aged 40 and over.

On the other hand, the ratios suggest that for some tax expenditures, taking the age group into account somewhat changes the conclusions that can be drawn. In particular, they show that tax expenditures related to employment especially benefit transgender or non-binary people among tax filers aged 40 and over. Furthermore, while the entire population of transgender or non-binary people was not found to have especially benefited from certain health-related tax expenditures (including the Disability Tax Credit and the Medical Expense Tax Credit), these measures do appear to benefit this group when age group is taken into account. More nuanced results are also obtained for tax expenditures related to the presence of children in the family (the Child Care Expense Deduction and the Canada Child Benefit) or investments (dividend gross-up and tax credit as well as the deduction of interest and carrying charges incurred to earn investment income), since transgender or non-binary people are found to especially benefit from these measures among the population aged 40 and over. Conversely, transgender or non-binary people are found to be advantaged by the credit for subscriptions to Canadian digital news media and the foreign tax credit among the population under age 40. These results suggest that, for certain tax expenditures, the specific age structure of transgender or non-binary people may mask or amplify certain redistributive impacts attributable to gender identity.

It is therefore important to emphasize that some of the results obtained for the entire population of transgender or non-binary people (especially results that tend to vary depending on the age group considered) may be the product of a transitory phenomenon where younger generations may be more open to report their transgender or non-binary identity in the context of a survey or more willing to reveal this identity at an earlier stage of their life. These results may change in future years if the younger age profile of transgender or non-binary people evolves towards one that is more in line with the rest of the population.

3.2 Immigration history

The 2021 Census of Population includes information on immigration history that comes directly from Immigration, Refugees and Citizenship Canada's administrative records (i.e., this is not self-reported information). This information helps identify individuals' immigration status, i.e., whether they were born in Canada, are immigrants or permanent residents, or are non-permanent residents. It also provides information on:

• Immigration year – used to determine whether or not immigrants were recently admitted to Canada.
• Admission category – designating the name of the program under which immigrants were admitted.
• Applicant type – indicating whether an individual's immigration application was submitted as a principal applicant, spouse or dependant.

In census products, recent immigrants are defined as people who obtained landed immigrant or permanent resident status up to five years prior to the census (i.e., from January 1, 2016, to May 11, 2021). In this study, this group was extended to all people who obtained this status as of January 1, 2010. This period of just over 10 years is more consistent with the definition used to study immigrant populations who are significantly more economically vulnerable, notably those facing higher poverty rates than the rest of the Canadian population.^{Footnote 25}

3.2.1 Profile of tax filers by immigration history

Immigrants—including those who have been granted permanent and non-permanent resident status—represented nearly 30% of the adult census population who filed an income tax return for the 2020 tax year. About a third of them were admitted to Canada since 2010 and are categorized in the "recent immigrants" group, while the remaining two thirds are categorized as "non-recent immigrants." Table 6 illustrates some differences between the demographic and socioeconomic characteristics of tax filers born in Canada and people in the "non-recent immigrants" and "recent immigrants" groups.

While some characteristics of non-recent immigrants are more similar to those of Canadian-born people than those of recent immigrants, including income level, education profile and/or homeowner status, non-recent immigrants are on average older and much more likely to live in an urban area than people born in Canada.  

Recent immigrants have a profile that is even more distinct from people born in Canada. In the last census, they were much more likely to live in an urban area, as were non-recent immigrants. However, unlike non-recent immigrants, the age profile of recent immigrants was considerably younger on average (about 10 years younger) than that of Canadian born, and they were much more likely to live in families with children. Furthermore, although a greater proportion of recent immigrants had more than a high school diploma, they were more likely to have pre-tax personal incomes in the bottom of the income distribution (first quartile), and their average personal and family income levels were significantly lower than those of both Canadian-born and non-recent immigrant tax filers. They were also much less likely to occupy a home that their family owned.

Chart 2 below shows some differences in the age and income profile of individuals in the recent immigrants group according to their admission category. Unsurprisingly, those admitted in the "economic immigrant" category—who represented the largest proportion of the recent immigrants group (47.2%)—had average personal and family incomes that were closest to those of Canadian-born tax filers and non-recent immigrants. Recent immigrants admitted under the "refugee" category were the most economically vulnerable, in terms of both personal and family income, followed by recent immigrants admitted under the "other and non-permanent resident" category, who represented 9.4% and 21.4% respectively of the overall "recent immigrants" group. Tax filers in this category were also the youngest on average, which is not necessarily surprising given that international students are included in this subgroup of recent immigrants. 

3.2.2 Overall redistributive impact by immigration history

a) Among all tax filers

Table 7 shows that overall, recent immigrants, who represented 9.8% of adult tax filers in the 2020 tax year, reported 7.0% of total pre-tax income. Non-recent immigrants also reported a slightly lower share of total pre-tax income (19.2%) than the share of the tax filers they represented (19.7%), while an opposite pattern is observed for people born in Canada.

A comparison of the income shares held by people born in Canada and immigrants shows that the 2020 federal PIT system had a slight redistributive impact between these groups, and more particularly towards recent immigrants. The share of total after-tax income held by recent immigrants (7.32%) was 1.05 times higher than the share of total pre-tax income reported by this group (6.98%). Among the major components of the tax system, refundable credits appear to have the greatest redistributive impact towards recent immigrants, while adjustments, exemptions and deductions seem to play a more important redistributive role towards non-recent immigrants.

Chart 3 below illustrates that this overall redistributive impact led to a $3,800 reduction in the average personal income gap between recent immigrants and Canadian-born tax filers, and a $900 reduction between non-recent immigrants and Canadian-born tax filers.

Among the group of recent immigrants, Chart 4 suggests that those admitted as refugees, followed by those sponsored by a family member, saw their average personal income gap with Canadian-born tax filers narrow the most following the application of the federal PIT system. For these two subgroups of recent immigrants, the system led to reductions in the average income gap of around $8,500 and $5,200 respectively.

Average personal income gap between recent immigrants and people born in Canada, by admission category of recent immigrants, before and after the application of the federal PIT system, 2020

		Average pre-tax income gap with people born in Canada	Average after-tax income gap with people born in Canada
Recent immigrants by admission category
Economic	$7,600		$5,500
Sponsored by family	$28,700		$23,500
Refugee	$34,500		$26,000
Other and non-permanent resident category	$28,400		$24,800

b) By tax filers' age group

The age profiles of non-recent and recent immigrants are also distinct from the rest of the population. Non-recent immigrants are older on average than people born in Canada, while the opposite is true for recent immigrants. By examining the redistributive impacts within two distinct age groups, i.e., tax filers under 40 and those aged 40 and over, it was possible to determine whether different redistributive impacts emerge when age group is taken into account.^{Footnote 26}

Dividing the population into two age groups shows that first, unlike non-recent immigrants aged 40 and over, non-recent immigrants under 40 (who are thus more likely to have been admitted to Canada at a young age) report incomes that are, on average, higher than those of people born in Canada in the same age group.^{Footnote 27} Despite this finding, Table 8 indicates that the federal PIT system remains slightly redistributive towards this group (especially due to refundable credits). It also shows that this slight redistributive impact towards the non-recent immigrant group is slightly less pronounced than among the population aged 40 and over. In this latter age group, available exemptions appear to provide a somewhat greater benefit to non-recent immigrants than to the rest of the population.

Table 8 also confirms an overall positive redistributive impact towards recent immigrants (more significant than towards non-recent immigrants) within the two age groups considered (mainly due to refundable credits). While the redistributive impact towards recent immigrants is more pronounced among older tax filers (ratio of 1.04 among those aged 40 and over compared to 1.02 among those under 40), they are both slightly lower than the overall redistributive impact observed among all tax filers (1.05).

3.2.3 Redistributive impact of tax expenditures by immigration history

a) Among all tax filers

To examine how specific tax expenditures contributed to overall trends in income redistribution towards immigrants, the total benefits associated with the list of the 55 selected PIT expenditures were disaggregated by immigration history. Based on our preferred indicator (ratios of the share of total benefits of a tax measure received by a group relative to the share of total pre-tax income reported by that group), each expenditure was then classified as being more or less beneficial for each of the immigrant groups considered (i.e., non-recent and recent immigrants).

The ratios presented in Table B.1 in Annex B highlight significant differences in tax expenditures that especially benefit non-recent and recent immigrants. Consistent with their particular age profile, non-recent immigrants tend to benefit proportionally more from tax expenditures that generally favour older tax filers, such as the non-taxation of the Guaranteed Income Supplement and Allowance benefits, or those related to capital accumulation and investments, health or support for dependants. However, unlike older tax filers as a whole (i.e., the age 55 and over category as defined in the 2021 GBA Plus study), overall, non-recent immigrants do not especially benefit from the non-taxation of workers' compensation benefits, the Medical Expense Tax Credit or the Political Contribution Tax Credit. The same is true for the Pension Income Credit and pension income splitting, suggesting that a relatively lower proportion of tax filers in this group reported pension income that made them eligible for these tax expenditures.

Tax expenditures that especially benefit recent immigrants differ considerably from those favoring non-recent immigrants. Recent immigrants tend to benefit more from tax expenditures related to education for self, employment, children, moving expenses and purchasing a first home, which is consistent with their younger age profile. They also benefit relatively more from tax expenditures that generally benefit lower-income populations, including most refundable credits and the non-taxation of social assistance benefits.

In contrast, certain tax expenditures especially benefit both non-recent and recent immigrants, notably the foreign tax credit and the Spouse or Common-Law Partner Credit. The Tuition Tax Credit also benefits both groups, the difference being that non-recent immigrants especially benefit from the portion of the credit transferred from a dependant, while recent immigrants benefit most from the portion claimed for self, which again is consistent with their distinct age profile.

Table 9 presents the top six federal PIT expenditures for non-recent and recent immigrants in 2020 (based on the ratios presented in Table B.1 and without taking into account the total cost of each tax expenditure). Table 10 provides an overview of the top three tax expenditures for non-recent and recent immigrant women+ and men+.^{Footnote 28}

Among other things, Tables 9 and 10 show that the Tuition Tax Credit is among the tax expenditures that benefit non-recent and recent immigrants the most, whether they be men+ or women+. In fact, our results suggest that most tax expenditures that are found to benefit non-recent and recent immigrants the most do so regardless of gender, although some tax expenditures are not necessarily among the top three measures most beneficial to immigrants. Only a few measures favor only immigrant women+, namely those related to the presence of children or dependants, including the Canada Child Benefit and Child Care Expense Deduction—two measures that include rules on who must claim in the family and that result in women, whether immigrants or not, being the main beneficiaries.^{Footnote 29} On the contrary, among the tax expenditures that are the most beneficial to immigrants, just one only benefits immigrant men+, namely the deduction for clergy residence. While among non-recent immigrants the Spouse or Common-Law Partner Credit only benefits men+, this is not the case among recent immigrants. In this latter group, both men+ and women+ benefit relatively more from this credit.

b) By tax filers' age group

Ratios of the proportion of benefits received by non-recent and recent immigrants relative to their respective shares of pre-tax income were recalculated among tax filers under 40 and those aged 40 and over. These ratios make it possible to determine whether the positive or negative redistributive impacts found among all tax filers remain valid regardless of age group.

Overall, Tables B.2 and B.3 presented in Annex B suggest that the main conclusions discussed in the last section hold among both younger and older tax filers. The six tax expenditures that benefit non-recent and recent immigrants the most among all tax filers also favor these groups within the two age groups considered.

However, for some tax expenditures, the overall findings do not necessarily apply to all age groups. Notably, Table B.2 shows that some of the tax expenditures that do not benefit non-recent immigrants among all tax filers do benefit this group when age group is taken into account. This is the case for the Canada Workers Benefit as well as the Child Care Expense Deduction, which benefit non-recent immigrants among tax filers aged 40 and over. When age group is taken into account, non-recent immigrants also appear to benefit relatively more from the Canada Child Benefit and the Student Loan Interest Credit among both younger tax filers and those aged 40 and over. A similar phenomenon is observed among younger tax filers with respect to the credit for subscriptions to Canadian digital news media and the Canada Training Credit. However, the Disability Tax Credit no longer benefits non-recent immigrants more when age group is taken into account.

Table B.3 shows that some tax expenditures that appear in favor of recent immigrants among all tax filers are, in fact, beneficial to this group only among the older population. These include tax expenditures recognizing employment and education-related costs, as well as certain measures relating to family or income support, including the non-taxation of social assistance benefits, the Child Care Expense Deduction, and the Refundable Medical Expense Supplement.

Moreover, although the deduction for clergy residence is found to favor both groups of immigrants considered (non-recent and recent immigrants) among all tax filers, this benefit only holds true among the population aged 40 and over.

3.3 Ethnocultural diversity

The information on ethnocultural diversity characteristics collected in the census includes Indigenous identity and racialized identity. The "Indigenous identity" classification refers to whether the person identified with the Indigenous peoples of Canada. This includes those who identify as First Nations (North American Indian), Métis and/or Inuk (Inuit), and/or those who report being Registered or Treaty Indians (that is, registered under the Indian Act of Canada), and/or those who report being a member of a First Nation or an Indian band.^{Footnote 30} The term "racialized group" is based on the concept of "visible minority" in the Census. Visible minority refers to whether a person is a visible minority or not, as defined by the Employment Equity Act. The Employment Equity Act defines visible minorities as "persons, other than Aboriginal peoples, who are non-Caucasian in race or non-white in colour".^{Footnote 31},^{Footnote 32}

In the census products, these two ethnocultural diversity characteristics are presented separately. There is a classification according to a person's Indigenous identity (Indigenous identity or not, regardless of racialized identity), and another classification according to a person's racialized identity (people identifying as being part or not of racialized groups regardless of their immigration history). According to census definitions and concepts, persons of Indigenous identity and persons who are Caucasian in race or white in colour are classified as non-racialized people. This is also the case in this study, except that they have been divided into two categories, namely "persons who are Caucasian in race or white in colour" and "people of Indigenous identity." This classification into three mutually exclusive groups (i.e., "people who are Caucasian in race or white in colour," "people of Indigenous identity" and "racialized people"^{Footnote 33}) allows for an analysis that distinguishes between the impacts of two ethnocultural diversity characteristics available in the census data.

3.3.1 Profile of the tax filer population by ethnocultural diversity

In the 2021 Census, 3.8% of adult tax filers self-identified as a person of Indigenous identity and 24.6% as belonging to one or more racialized groups. The remaining 71.6% self-identified as neither racialized nor Indigenous and were classified in the "Caucasian in race or white in colour people" category.

Compared to people who are Caucasian in race or white in colour, racialized people were on average slightly younger and more likely to be in a couple with children or to live alone (Table 11). In the last census, over 95% of racialized people lived in an urban area, compared to just under 70% of people who are Caucasian in race or white in colour. While racialized people were slightly more likely to have more than a high school diploma, their pre-tax income level was on average lower than that of people who are Caucasian in race or white in colour. This group also included a greater proportion of renters.

Like racialized people, people of Indigenous identity were on average slightly younger than people who are Caucasian in race or white in colour. Their pre-tax personal and family incomes were also generally lower, and a smaller proportion of them owned their own home.

However, their education profile and distribution by area of residence and by gender differed from those of racialized people. Unlike racialized tax filers, people of Indigenous identity were much less likely to have more than a high school diploma and to live in an urban area, compared to people who are Caucasian in race or white in colour. In 2021, over half of people of Indigenous identity lived in a rural area, compared to under a third of white people and under 5% of racialized people. It is also interesting to note that, among people of Indigenous identity, women represented a slightly larger proportion of adult tax filers (54.6%) than among people who are Caucasian in race or white in colour (51.4%) or racialized (52.5%) people.

In the last census, individuals could disclose their Indigenous identity in a number of ways. In particular, they could do so by self-identifying as being part of the following identity groups: First Nations, Métis and/or Inuk (Inuit), and/or by self-identifying as having Registered or Treaty Indian status. Chart 5 shows that of the 962,000 adult tax filers who self-identified as Indigenous, just under half (42%) reported having Registered or Treaty Indian status. It also shows that the majority of Indigenous people self-identified as being part of the "First Nations" identity group (52.6%), followed by "Métis" (39.8%). Less than 8% of tax filers of Indigenous identity primarily self-identified with the "Inuk (Inuit)" group (3.8%) or with an "other" Indigenous identity (3.8%).

This chart also highlights a certain heterogeneity in the age and income profile of people of Indigenous identity according to Registered or Treaty Indian status and/or identity group. In fact, those identifying with the "Inuk (Inuit)" group are the youngest, on average. They also report the lowest average personal and family incomes. In contrast, Indigenous people who identify as "Métis" and/or who do not have Registered or Treaty Indian status report the highest average personal and family incomes.

Like people of Indigenous identity, racialized people are not a homogeneous population. This population consists of several subgroups of racialized minorities with distinct characteristics and realities who may potentially be impacted differently by the federal PIT system.^{Footnote 34} Due to confidentiality and statistical reliability issues, among other things, this group was only examined at the aggregate level in this study. Ways to better use available data and overcome limitations will be explored to improve future GBA Plus studies on tax expenditures by different subgroups of racialized people.

Distribution by Registered or Treaty Indian status and/or identity group

Age and pre-tax income profile by Registered or Treaty Indian status and/or identity group

	Average age	Average personal income	Average adjusted family income
Registered or Treaty Indian status
Non-registered	45.7 years	$48,700	$55,600
Registered	44.2 years	$42,400	$45,600
Identity group
First Nations	44.6 years	$42,700	$46,700
Métis	45.9 years	$50,600	$57,800
Inuk (Inuit)	33.0 years	$36,100	$37,600
"Other" Indigenous identity	47.7 years	$46,900	$53,800
Source: 2021 Census of Population data linked to 2020 T1 returns.

3.3.2 Overall redistributive impact by ethnocultural diversity

a) Among all tax filers

Table 12 shows that, in contrast with people who are Caucasian in race or white in colour, the proportions of total pre-tax income held by people of Indigenous identity and racialized people were lower than the proportions of tax filers they represented in 2020. However, a comparison of the total income held by these different groups before and after the application of the federal PIT system shows an overall redistributive impact towards these groups. This comparison indicates that the respective shares of total income held by people of Indigenous identity and racialized people are 1.06 and 1.03 times higher following the application of the federal PIT system. 

By comparing different concepts of income, Table 12 also shows that, among the major components of the federal PIT system, adjustments, exemptions, and deductions have the most pronounced redistributive impact towards people of Indigenous identity, followed by refundable credits. In contrast, the redistributive impact towards racialized people appears to be due first to refundable credits and second to the progressive tax rate structure and non-refundable credits.  

Table 12 also shows that the overall redistributive impact of the federal PIT system towards people of Indigenous identity and racialized people appears to correspond with respective reductions of $99.8 billion and $84.9 billion between the total income held by these groups and the total income held by people who are Caucasian in race or white in colour. This equates to reductions in the average income gap observed between these same groups of around $4,500 and $3,000, respectively, when comparing pre- and after-tax income (Chart 6).

Showing some of these results in more detail, Chart 7 suggests that the overall redistributive impact of the federal PIT system towards people of Indigenous identity is most pronounced among those with Registered or Treaty Indian status. Specifically, the average income gap between this subpopulation of Indigenous people and people who are Caucasian in race or white in colour is reduced by $5,800 following the application of the federal PIT system, compared to a reduction of $2,200 for all non-registered Indigenous people. It is worth noting that the overall redistributive impact of the federal PIT system towards people of Indigenous identity would probably be even greater if the tax filing rate for this group was closer to that observed among non-Indigenous people.

Average personal income gap between people of Indigenous identity and people who are Caucasian in race or white in colour, by Registered or Treaty Indian status (labelled as Non-registered and Registered) and Indigenous identity group, before and after the application of the federal PIT system, 2020

	Average pre-tax income gap with people who are Caucasian in race or white in colour	Average after-tax income gap with people who are Caucasian in race or white in colour
People of Indigenous identity by Registered or Treaty Indian status
Non-registered	$9,000	$6,800
Registered	$15,300	$9,500
People of Indigenous identity by identity group
First Nations	$15,000	$9,800
Métis	$7,100	$5,300
Inuk (Inuit)	$21,600	$17,500
"Other" Indigenous identity	$11,000	$8,400

b) By tax filers' age group

Since people of Indigenous identity and racialized people are also somewhat younger on average than all tax filers, this section aims to control for the impact of age, focusing on the redistributive impact towards these groups within different age groups (i.e., tax filers under 40 and tax filers aged 40 and over).^{Footnote 35}  

Table 13 shows a positive change in the share of disposable income of people of Indigenous identity and racialized people following the application of the federal PIT system, regardless of the age group considered. As is the case among all tax filers, the redistributive impact observed among the population under 40 and 40 and over is more pronounced for people of Indigenous identity than for racialized people. Table 13 also shows that the redistributive impact of the PIT system towards racialized people is slightly more pronounced among the population aged 40 and over, while people of Indigenous identity benefit from a more pronounced redistributive impact among tax filers under 40. 

3.3.3 Redistributive impact of tax expenditures by ethnocultural diversity

a) Among all tax filers

Table C.1 in Annex C lists the federal PIT expenditures that, based on the ratio described above, had a positive, negative or neutral impact on the ethnocultural diversity groups examined. Table 14 shows the six federal tax expenditures that benefited people of Indigenous identity and racialized people the most in 2020.

Among other things, Table 14 indicates that the "non-taxation of personal property of status Indians and Indian bands situated on reserve" was the federal PIT expenditure that benefited people of Indigenous identity the most. Nearly 98% of the benefits from this measure were received by tax filers who self-identified as people of Indigenous identity in the census. Table 15 denotes that this measure was the top tax expenditure benefiting people of Indigenous identity only among those who also report having Registered or Treaty Indian status. This result was expected since this measure specifically targets status Indians and Indian bands situated on reserve. One of the reasons that may explain why people of Indigenous identity with Registered or Treaty Indian status do not fully benefit from this tax expenditure could be the fact that those who are not registered but are entitled to be registered may be eligible for the measure. There may also be individuals whose status changed during the study period, given the significant number of court decisions that may impact eligibility requirements. The fact that some eligible individuals may have preferred not to self-report their Indigenous identity or omitted do so in the census may also explain this result.

For the Indigenous identity group (Table 14), the second and third most beneficial tax expenditures are "Northern Residents Deductions" and the "non-taxation of social assistance benefits." While this group reported 3.0% of total pre-tax income in 2020, it received nearly 20% and 15% of the total tax savings attributable to these two tax expenditures. These results are intuitive, given the relatively higher concentration of people of Indigenous identity who are northern residents and their generally lower income level. The Canada Child Benefit and the Canada Workers Benefit—two refundable credits targeting low- and moderate-income populations—are also among the top six federal tax expenditures in favor of Indigenous people, as well as the Eligible Dependant Credit, which is non-refundable. For these three measures, the share of tax benefits received by people of Indigenous identity was at least twice as high as the share of pre-tax income that this group reported.

It is also interesting to note that when people of Indigenous identity are divided into two subpopulations based on their Registered Indian status (Table 15), the "tax-free amount for emergency services volunteers, Search and Rescue Volunteers Tax Credit, and Volunteer Firefighters Tax Credit" are among the top three measures that were most beneficial to non-registered people of Indigenous identity. Further disaggregation by gender (Table 16) shows that among non-registered people of Indigenous identity, only men+ benefit from these tax expenditures.

Examined as a whole, the racialized people group is found to benefit more than proportionately from a list of tax expenditures that is quite similar to the list of tax expenditures most beneficial to immigrants and recent immigrants in particular (Table 14 versus Table 9). The list of the six federal tax expenditures that were most beneficial to racialized people in 2020 includes the Tuition Tax Credit – for self or transferred from a dependant, the Spouse or Common-Law Partner Credit, the First-Time Home Buyers' Tax Credit, as well as the Canada Child Benefit and the Canada Workers Benefit. This was also the case for people in the recent immigrants group.

When the racialized people group is disaggregated by gender, the results are also similar to those for immigrants (Table 17 versus Table 10). Although in a slightly different order, the top three federal PIT expenditures that were most beneficial to racialized men+ are the same as those for recent immigrant men+ (i.e., the Spouse or Common-Law Partner Credit, First-Time Home Buyers' Tax Credit, and Tuition Tax Credit). Among women+, only the third-ranked tax expenditure differs between these two groups. Among racialized women+, the Eligible Dependant Credit ranks third in the list of beneficial measures, while it is the Canada Training Credit that ranks third among recent immigrant women+.

Looking at Table C.1, which shows in more detail the list of federal PIT expenditures that are more or less beneficial for the different ethnocultural diversity groups examined, it can also be noted that, unlike non-recent immigrants, racialized people are generally less advantaged by tax expenditures related to capital accumulation (notably measures related to capital gains, interest and investments, stock options, dividends received, and investing). They also benefit relatively less from tax expenditures related to older age (such as the Age Credit, Pension Income Credit and pension income splitting). These tax expenditures tend to benefit more people who are Caucasian in race or white in colour, reinforcing pre-existing income inequalities between different ethnocultural diversity groups. The same is true for the Charitable Donation Tax Credit, Political Contribution Tax Credit, credit for subscriptions to Canadian digital news media, and Medical Expense Tax Credit. Of course, the fact that racialized people—a very heterogeneous group—were examined as a whole likely masks some of the redistributive impact towards various subpopulations of racialized people. The slightly younger age structure of members of these two ethnocultural diversity groups may also have had an impact on these results.

b) By tax filers' age group

To determine whether the redistributive impact observed among all tax filers is present regardless of age, Tables C.2 and C.3 in Annex C compare the redistributive impact (positive, negative or neutral) towards people of Indigenous identity and racialized people among tax filers from different age groups (all ages, under 40 and 40 and over).

As is the case for the other diversity groups previously examined (transgender or non-binary people as well as non-recent and recent immigrants), the main tax expenditures identified as being most beneficial to all people of Indigenous identity and racialized people (see Table 14) are also in the list of measures favoring these groups among younger and older tax filers.

However, once again, the results presented in Tables C.2 and C.3 make it possible to add some nuance to the results obtained among all tax filers. In particular, some of the tax expenditures related to employment and education costs as well as family support that especially benefit people of Indigenous identity and racialized people overall, in fact benefit these groups relatively more only in the older population. In addition, while the non-taxation of social assistance benefits especially benefits people of Indigenous identity regardless of age group, this is the case only for older racialized people (as was observed for recent immigrants). In contrast, the Tuition Tax Credit (for self and transferred from a dependant other than a spouse) benefits racialized people in all age groups while the portion of the credit claimed for self is found to benefit people of Indigenous identity only in the older population, and the portion transferred from a dependant only in the younger population.

Some nuance can also be provided with respect to the redistributive impact of the Charitable Donation Tax Credit and the credit for subscriptions to Canadian digital news media. Overall, these two tax expenditures are found to be of less advantage to people of Indigenous identity and racialized people. However, among younger tax filers, racialized people appear to benefit more than proportionally from these two measures.

4. Conclusion

This study is part of a series of GBA Plus research on federal tax expenditures aimed at fulfilling the requirements of the Canadian Gender Budgeting Act. Taxation, the allocation of public resources, and other public policy decisions can have different impacts on various groups of people, which can create, maintain or reduce social inequalities. The Canadian Gender Budgeting Act therefore requires that better information be made available on the impacts of public policies—including tax expenditures—in terms of gender and diversity.

To meet the public reporting requirements of the Act, many data limitations have had to be overcome over the past few years, including in relation to the diversity factors that can be examined. Since administrative tax data include little information on tax filers' identity characteristics, a linkage exercise between data from the latest census (2021) and 2020 T1 tax return data was undertaken in partnership with Statistics Canada. As a result, it was possible to examine, for the very first time, the extent to which tax filers benefit from the federal PIT system based on their gender identity, immigration history, and ethnocultural diversity characteristics. This study therefore provides new insight into the impacts of various elements of the federal PIT system on income redistribution between cisgender men, cisgender women, and transgender or non-binary people; between people born in Canada, non-recent immigrants, and recent immigrants; and between people who are Caucasian in race or white in colour, people of Indigenous identity, and racialized people.

The study first presents the descriptive profiles of these groups. These highlight important differences in their demographic and socioeconomic characteristics which may drive claims of the various available federal tax expenditures and impact the amounts of benefits from these claims.

Overall, the results obtained confirm that the progressive nature of the federal PIT system reduces pre-existing income inequality between these different identity groups. In fact, as the 2019 and 2021 GBA Plus studies demonstrated, the current system tends to benefit economically disadvantaged groups (i.e., groups whose share of total pre-tax income is less than the share of tax filers they represent) and to redistribute a portion of reported income towards them. As had already been observed for female tax filers (based on sex assigned at birth), as a result of the 2020 tax system, the shares of after-tax income held by tax filers identifying themselves as cisgender females and transgender or non-binary people were higher than their respective shares of pre-tax income. The same is true for immigrants, especially those admitted to Canada within the last decade (including people with permanent and non-permanent resident status), as well as for people of Indigenous identity or members of racialized groups.

Since people in these groups earn pre-tax incomes that are on average lower than the general tax filer population, they tend to benefit more from tax expenditures that generally favour lower-income populations, including most refundable credits, the non-taxation of social assistance benefits, or the non-taxation of the Guaranteed Income Supplement and Allowance benefits.

Aside from their generally disadvantageous economic situation, these groups have their own realities which make them benefit more from a distinct list of tax expenditures. For example, the distribution of the benefits of the various tax expenditures between groups shows that the "Northern Residents Deductions" are particularly favorable to people of Indigenous identity. It also indicates that both non-recent and recent immigrants, as well as racialized people, benefit relatively more from the Tuition Tax Credit, and that transgender or non-binary people especially benefit from the Political Contribution Tax Credit.

The results of this study also demonstrate that, due to the specific age profiles of these groups, some of the redistributive impacts towards them may vary depending on the age group considered. For example, while recent immigrants and racialized people appear to particularly benefit from the non-taxation of social assistance benefits among all tax filers, these advantages are confirmed solely among the population aged 40 and over. The same is true for the deduction for clergy residence. However, despite some nuances, the vast majority of tax expenditures identified as being most beneficial to the different identity groups examined in this study remain so regardless of the age group considered.

Of the 55 tax expenditures considered in this study, almost all were found to benefit either transgender or non-binary people, non-recent or recent immigrants, or Indigenous or racialized people in 2020, contributing to reducing income inequality linked to gender and the other diversity characteristics examined.

However, it is important to keep in mind that the groups examined in this study are not homogeneous. Disaggregating some of these groups (e.g., analyzing racialized groups according to the visible minority group tax filers primarily identified with in the census), or even taking certain other intersecting identity factors into consideration (such as income level), could provide more exhaustive information regarding the impacts that the federal PIT system may have on these groups. The Department of Finance may consider further studies in this regard.

It should also be mentioned that the profile of some of the groups examined in this study is not static. Recent immigrants in 2020, who by definition are younger on average, over time will become non-recent immigrants and older, and be replaced by new cohorts of recent immigrants. Consequently, from one period to another in their lives, the same groups of immigrants may be able to benefit relatively more from a distinct set of tax measures. Furthermore, the younger age profile of transgender or non-binary people may shift to one that is more in line with the rest of the population if the greater willingness of younger generations to identify as part of this group, relative to older generations, diminishes over time. This change could potentially affect how certain components of the federal PIT system redistribute income towards this group.

As tax expenditures can be used to achieve objectives, including vertical equity (i.e., the idea that tax filers who have the ability to pay more taxes should contribute more), this analysis can help identify the expenditures that support or work against such objectives. It must be reiterated, however, that the tax system has various objectives and that tax expenditures cannot be assessed solely on the impact they have on the distribution of income. A thorough assessment of tax measures requires the consideration of a much broader range of possible effects than the ones examined in the current study.

Annex A: Specific redistributive impacts by gender identity

Annex B: Specific redistributive impacts by immigration history

Annex C: Specific redistributive impacts by ethnocultural diversity