Projection assumptions and scenarios

View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Archived

This page has been archived on the Web.

Skip to text

• Assumptions
• Projection scenarios
• Notes

Text begins

Assumptions

As is true of any prospective exercise, assumptions—related to population growth components, in this case—formed the basis of these projections. Assumptions were developed not only for Aboriginal populations, but also for non-Aboriginal populations, which are part of the projections. Assumptions were selected to meet the two following objectives: (1) to create scenarios comprising a plausible range of future possibilities for Aboriginal populations until 2036; and (2) to estimate the sensitivity of Aboriginal population growth and its share within the total population to key components of fertility, intragenerational ethnic mobility and internal migration.

The assumptions were developed by Statistics Canada in consultation with Aboriginal Affairs and Northern Development Canada and based on an analysis of the latest data, existing literature and consultations conducted by Statistics Canada. As well, they were submitted to the Demosim scientific committee, which is made up of researchers and specialists independent of the Demosim team and is mandated to make recommendations on methods, data sources, model content and product development.

Because of the number of components projected, many assumptions had to be developed. Therefore, the focus of this section is to describe the assumptions specifically related to Aboriginal populations; other assumptions will be described only briefly.

Assumptions specifically related to Aboriginal people

This section describes the assumptions specifically related to Aboriginal populations and the reasons for adopting them. The assumptions cover fertility; the intergenerational transmission of the Aboriginal group and registered Indian status (including registration category); mortality; internal migration, international migration; intragenerational ethnic mobility; registration on the Indian Register over an individual's lifetime; and headship rates.^Note 5 The assumptions are summarized in Table 3.

Fertility

The fertility of Aboriginal people decreased over the second half of the 20th century, in both the population with Aboriginal ancestry and the population with registered Indian status (Ram 2004; Guimond and Robitaille 2009; Maynard and Kerr 2007; Loh and George 2003). Despite the decrease revealed by these longer-term analyses, the fertility of Aboriginal people remains higher overall than that of the non-Aboriginal population, although the situation varies from one Aboriginal identity group to another (Table 4). In 2011, the total fertility rate of the Aboriginal identity population overall was 2.2 children per woman, compared with 1.6 among the non-Aboriginal population. Among specific groups, the Inuit and registered Indian populations had the highest fertility rates, at 2.8 children and 2.7 children per woman respectively. The total fertility rate of the Métis population was only slightly higher than that of the non-Aboriginal population, at 1.8 children per woman, while the rate was lower among Non-Status Indians, at 1.5 in 2011. If Non-Status Indians (whose fertility increased between 2001 and 2006 and then decreased significantly between 2006 and 2011) are excluded, the fertility of Aboriginal people has been relatively stable in recent years (Morency and Caron-Malenfant 2014; Statistics Canada 2011; Amorevieta-Gentil et al. 2013^Note 6). However, some authors (Suwal and Trovato 1998; Ram 2004) believe that the fertility of Aboriginal people will converge with that of the rest of the population. According to Suwal and Trovato (1998), this convergence may occur through the integration of Aboriginal populations into the rest of society, in particular as a result of mixed unions.

Given the relative stability of the fertility of Aboriginal people in recent years, it is difficult to anticipate if—and when—it might converge with that of the non-Aboriginal population. Because of the uncertainty associated with this issue and the importance of fertility as a component of Aboriginal population growth, three assumptions were adopted. Under the first assumption, the probabilities of giving birth to a child (as estimated using the 2011 National Household Survey (NHS)^Note 7) remain constant until 2036. Under the second assumption, these probabilities converge moderately and gradually with those of the non-Aboriginal population so that, by 2036, half of the fertility gap between Aboriginal populations and the non-Aboriginal population has been closed. Under the third assumption, the fertility of Aboriginal people converges gradually to a complete convergence by 2036, becoming identical to that of the non-Aboriginal population at the end of the projection.

Transmission of the Aboriginal group from mother to child

Strongly linked to mixed unions (Boucher et al. 2009; Robitaille and Guimond 2003), the likelihood of parents reporting for their children an Aboriginal group different from their own (also known as intergenerational ethnic mobility) varies from one Aboriginal group to another. According to adjusted 2011 NHS data, 95% of children aged under one whose mother is an Inuk are also Inuk. The proportions were 91% among the First Nations Aboriginal group and 69% among the Métis. Compared with estimates obtained for previous projections (Statistics Canada 2011), this phenomenon has remained highly stable in recent years. Therefore, a single assumption was used for this component, namely that the rates of intergenerational ethnic mobility remain at their 2011 levels until 2036.

Transmission of registered Indian status and registration category from mother to child

The intergenerational transmission of registered Indian status and of the registration category is governed by rules set out in the 1985 Indian Act (see the "Concepts" section). However, children entitled to registration are not registered automatically at birth; rather, the parents must complete a process with the department in charge of the Indian Register (AANDC). According to adjusted 2011 NHS data, about 71% of children aged under one with at least one registered Indian parent^Note 8 are themselves Registered Indians. This proportion varies according to the registered Indian status of each parent and whether or not the child belongs to an Aboriginal group (Table 5). The proportion is highest when both parents are Registered Indians, whether or not the child belongs to an Aboriginal group, while it is virtually nil when neither parent is a Registered Indian. In cases where only one parent is registered (mixed unions), the proportion falls in between and is little affected by the sex of the registered parent. For each type of union and registered Indian status of the mother,^Note 9 the proportion of registered children is higher when the child belongs to an Aboriginal group—which, of course, depends on the mother's Aboriginal group. Analyses conducted as part of these projections revealed that, among women in a union who had given birth to a child between 2010 and 2011, registered women living off reserve or non-registered women living on reserve were likelier to be in a mixed union. This likelihood varied slightly from 2001 to 2011, increasing among registered women living off reserve and decreasing among women, both registered and non-registered, living on reserve.

While the objective and legal nature of the rules of transmitting registered Indian status might suggest a certain stability in the propensity of transmitting registered Indian status across generations, the future evolution of this propensity will depend mainly on changes in the prevalence of women entering into unions with spouses of different status. For this reason, it is assumed that the transmission rates of registered Indian status from mother to child will remain at their 2011 level until 2036, and that trends with respect to mixed unions will gradually slow down over the next 25 years.

Mortality

The literature on the mortality of the Aboriginal population in Canada shows that it is higher than that of the rest of population, although there are differences among the groups. For example, from 1991 to 2006, Registered Indians appear to have experienced a life expectancy at birth about 5 years shorter than that of the total Canadian population (Amorevieta-Gentil et al. 2014)^Note 10 while, in regions with high Inuit concentration, life expectancy was 9 to 11 years shorter over the same period (Wilkins et al. 2008; Peters 2013). Work by Tjepkema and Wilkins (2011) on life expectancy at age 25 shows that the First Nations Aboriginal group has a shorter life expectancy than the rest of the population. The Métis group also has a shorter life expectancy, midway between that of the First Nations Aboriginal group and that of the non-Aboriginal population, even when the effects of other variables such as education and place of residence are controlled. Despite a significant increase in life expectancy over the last few decades in groups for which estimates are available (Maynard and Kerr 2007; Verma et al. 2004), neither the data of Amorevieta-Gentil et al. on Registered Indians nor the data of Wilkins et al. and Peters on Inuit regions make it possible to conclude that mortality has been converging with that of the rest of the population over recent years. Estimates produced as part of these projections show that the life expectancy of both males and females in Aboriginal identity groups appear to be shortest among Inuit, followed by Registered Indians and Non-Status Indians, and then Métis, whose life expectancy is the closest to that of the rest of the population.

Could the differences in mortality among these groups remain the same in future years, particularly given that certain mortality causes are more prevalent among Aboriginal populations and that the groups' geographical distribution—which differs from the rest of the population—often limits their access to health care services? Or, will these differences instead decrease as a result of other factors, such as, for example, converging living conditions or lifestyles? While this question may be difficult to answer, the effect of either possibility on populations projected over a 25-year period would be limited, as the convergence of mortality would occur over a longer period. In fact, most Aboriginal people who would benefit from higher life expectancy because of this convergence would die long after the end of these projections. For this reason, a single assumption was used for these projections. Under this assumption, the mortality rates by age and sex of the total population gradually decline in accordance with the middle mortality assumption presented in Statistics Canada’s latest national population projections (Statistics Canada 2014), while the gaps between Aboriginal people and non-Aboriginal people estimated as part of these projections will remain unchanged until 2036 (Figure 1).

Description for figure 1

Internal migration

There are relatively few studies that looked at internal migration of Aboriginal populations. Of those available, a study by Dion and Coulombe (2008) analyzes migration in the 2006 Census and shows that migrating Aboriginal people are less likely than non-Aboriginal people to settle in Toronto, Montréal or Vancouver; they more often chose rural areas and the territories as their destination. Clatworthy and Norris (2014), who also examined the migration of Aboriginal people, indicate that migration generally contributes very little to changes in the proportion of Aboriginal people living in certain types of regions, in particular census metropolitan areas. They also showed that internal migration for each of the five-year periods from 1966-1971 to 2001-2006 led to net gains in the number of Registered Indians on Indian reserves. Analyses conducted for the previous projections (Statistics Canada 2011) and the current projections also show a different migration profile for Aboriginal people, in particular among Registered Indians, who continued to record net migratory gains on Indian reserves in the most recent period. While internal migration shows some stability in its link to certain characteristics (for example, the migration of Registered Indians to Indian reserves or the migration of young people to major metropolitan areas), internal migration is nevertheless a volatile component; it is likely to vary over time, and is thus hard to project (Smith 1986).

Because of the uncertainty regarding the future evolution of this component, two assumptions were used. The first assumes, on one hand, that the future contribution of internal migration to total regional demographic growth reflects the contribution observed during the 1996-2001, 2001-2006 and 2006-2011 periods, and, on the other hand, that the composition of the migratory flows, notably relating to Aboriginal identity, is in line with that observed during the 2000-2001, 2005-2006 and 2010-2011 periods. Each part of this assumption was developed using one-year and five-year mobility variables from the 2001 and 2006 censuses and the 2011 NHS. Since the five-year mobility variable is considered more robust because of its longer coverage in time, it was used to determine interregional migration patterns. The one-year mobility variable is considered to be more appropriate for measuring migrant characteristics,^Note 11 and was therefore used to determine the composition of migratory flows. According to the second assumption, there would be no internal migration during the projected period. In comparison to the first assumption, this assumption makes it possible to estimate the overall contribution of internal migration to the growth of Aboriginal populations and to determine the combined contribution of other components to this growth. It also makes it possible to evaluate the evolution of the population living on reserve if it were to stop experiencing net migration gains.

International migration of Aboriginal people

According to the 2011 NHS, only 13,800 people of Aboriginal identity were born outside Canada, and the vast majority of them were born in the United States. Data from the American Community Survey, a large-scale survey conducted annually in the United States, show that the American Indian and Alaskan Native population born in North America outside the United States was estimated at 10,200 from 2010 to 2012.^Note 12 Assuming that the vast majority of them were born in Canada^Note 13 and that most Aboriginal people who leave Canada settle in the United States, the resulting 'net' migration is very low, almost nil. For this reason, a single assumption was used for this component, namely that immigration and emigration are nil among Aboriginal populations during the entire projection period.^Note 14

Intragenerational ethnic mobility

Studies on changes in the Aboriginal group reported over an individual's lifetime—known as intragenerational ethnic mobility—have shown that this component was responsible for a significant share of Aboriginal population growth in Canada from 1986 to 2006 (Guimond 1999; Guimond et al. 2007; Lebel et al. 2011). This phenomenon was more prominent among the Aboriginal group of Métis than that of First Nations people, while it was not observed conclusively among Inuit and populations living on Indian reserves during this period. The phenomenon was also observed in the United States (Passel 1996; Perez and Hirschman 2009), Australia (Ross 1999; Australian Bureau of Statistics 2013) and New Zealand (Brown and Gray 2009). According to a recent study based on linked data from the 2001 and 2006 Canadian censuses, (Caron-Malenfant et al. 2014), the net gains due to ethnic mobility in the Métis and First Nations groups in fact resulted from multidirectional changes between the two groups and the non-Aboriginal population; ethnic mobility was, therefore, not a unidirectional phenomenon. Analyses conducted as part of these projections reveal that this phenomenon continued from 2006 to 2011. However, the First Nations group benefited more than the Métis group from changes in reported identity during the period, a reversal from previous periods. From 2006 to 2011, ethnic mobility appears to have contributed 64% of growth in the First Nations population, while it represented 52% of growth in the Métis population. In contrast to past periods, the phenomenon appears to have been a significant growth factor for the Inuit population, especially outside Inuit Nunangat. The adoption of the Gender Equity in Indian Registration Act and the recognition of the Qalipu Mi'kmaq First Nation in 2011 may have favoured transfers to First Nations people.^Note 15 Changes in wording to the question on Aboriginal groups between the 2006 Census and the 2011 NHS may also have contributed to the changes observed during the period, although it is impossible to know for certain.^Note 16 In addition, methodological differences between the 2011 NHS and the 2006 Census could partly account for the increase in the Inuit population residing outside Inuit Nunangat.^Note 17

The changes in intragenerational ethnic mobility from 2006 to 2011 reveal the uncertainty associated with the future evolution of this component. One may also wonder if a possible decline in the number of people likely to make a transfer to an Aboriginal group would slow the phenomenon in future decades. Because of these uncertainties, two ‘extreme’ assumptions were adopted in the projections. Under the first assumption, the net rates of intragenerational ethnic mobility maintain the average levels observed for the 1996-to-2011 period (Table 6) until 2036.^{Note 18, Note 19} As ethnic mobility was not observed conclusively in the Inuit population before the 2006-to-2011 period and as the phenomenon observed during this period may be linked to changes in the 2011 NHS, this group’s ethnic mobility was assumed to be nil from 2011 forward. Under the second assumption, ethnic mobility is nil from 2011 to 2036 for all groups.^Note 20

Registration on the Indian Register and reclassification of registration category over an individual’s lifetime

Amendments to the Indian Act over time (see Box 2) have had a significant impact on the registered Indian population. Clatworthy (2001) estimates that, from 1985 to 1999, 174,500 people became entitled to registration under the 1985 amendments to the Indian Act (Bill C-31) and that 114,700 of them were in fact registered on the Indian Register in 1999. Although most of these registrations were recorded in the years immediately following the enactment of the new legislation, AANDC continues to record a small number of new registrations each year (600 on average from 2007 to 2014). The adoption of the Gender Equity in Indian Registration Act (Bill C-3) in January 2011 resulted in 29,200 new registrations as of August 12, 2014, and 12,500 reclassifications from registration category 6(2) to 6(1) as of July 18, 2014, with the vast majority occurring in 2011 and 2012. In addition, as of August 12, 2014, close to 24,000 people were registered on the Indian Register as members of the Qalipu Mi'kmaq First Nation, a band that was created on September 22, 2011. However, the total number of Registered Indians in the Qalipu Mi'kmaq First Nation may change as the Government of Canada and the Federation of Newfoundland Indians reached a Supplemental Agreement in June 2013, which aims to resolve issues that emerged in the implementation of the enrolment process.^Note 21 Through the Supplemental Agreement, "all applications received during all phases of the enrolment process, except those previously rejected, will be assessed or reassessed."^Note 22 Lastly, other people register late each year on the Indian Register (children in their first few years of life representing a large share) or have their registration category changed for various reasons.

A single assumption was used for each part of this component, which complements the component for registration at birth. According to this assumption, registrations under the 1985 amendments to the Indian Act, now few in number, will continue to decline over the projected period at the same pace as that observed from 2007 to 2014. Registrations resulting from Bill C-3—most of which appear to have already occurred—will follow the AANDC projections, which assume that registrations will first decline, then stabilize before being completed in 2019/2020.^Note 23 It is also assumed that the number of reclassifications from registration category 6(2) to 6(1) associated with Bill C-3 will follow the same trends as registrations under the same bill until 2020. Users should note that it is the observed numbers of C-31 and C-3 registrations and C-3 reclassifications until 2014 that are used at the start of the projection. Registrations resulting from the legal recognition of the Qalipu Mi'kmaq First Nation are more uncertain because of the reasons mentioned previously. In the absence of alternative estimates, the assumption was based on the only number available, namely the approximately 24,000 individuals who registered as Qalipu on the Indian Register between 2011 and 2013. Although potential fluctuations in this number are not likely to significantly affect the projection results at the national level and in most regions, this is not the case for the province of Newfoundland and Labrador, where most registrations originate. The reader is therefore advised to exercise caution when interpreting the projection results for this province.^Note 24 Lastly, under the assumption, rates for the late registration of children and adults and for reclassifications from 6(2) to 6(1) for various reasons^Note 25 will remain constant at their recent level.

These registrations on the Indian Register over an individual's lifetime will affect not only the registered Indian population, but also the non-status Indian population as the assumption is made that people within the non-status Indian population will be most likely to register over the course of their lifetime.

Headship rates

In Canada, existing analyses of Aboriginal headship rates are few in number, having been primarily associated with past projections of Aboriginal population (Clatworthy 2012, 2006; Ng and Perreault 1998; Kerr and Kopustas 1995). However, these analyses have revealed that Aboriginal headship rates are, as expected, lower than those of the total population overall, as a result of the greater average size of Aboriginal households. These analyses have also underscored the difficulty of establishing trends in headship rates since, as stated by Clatworthy (2012), it is difficult to determine whether changes in headship rates over time among the various Aboriginal populations (especially Registered Indians, Non-Status Indians and Métis) are the result of actual changes in behaviour or simply changes in the composition of the population (for example, ethnic mobility, legislative amendments and specific agreements that move a large number of individuals from one group to another who do not necessarily have the same behaviours in terms of household composition as their new group).

As part of these projections, trends affecting the evolution of headship rates from 2001 to 2011 were analyzed, revealing a slight increase in the headship rate over time. However, when these rates are standardized to account for the age of the household head, place of residence, Aboriginal identity, marital status and household size, the trends recorded between 2001 and 2011 and between 2006 and 2011 disappear almost completely. This means that the trends are largely the result of compositional effects related to the variables used in these projections. For this reason, headship rates are kept constant throughout the projection.

Other assumptions

The other assumptions developed for these projections deal with either components that contribute indirectly to the growth of Aboriginal populations (for example, education and marital status) or components that more specifically affect non-Aboriginal populations (for example, international migration). There is a single assumption for each component. The assumptions were adopted to reflect the latest conditions and trends. They were developed to be as consistent as possible with the two other sets of projections produced by Statistics Canada: (1) Population Projections for Canada (2013 to 2063), Provinces and Territories (2013 to 2038) (Statistics Canada 2014), the assumptions of which were developed through extensive consultations^Note 26; and (2) Projections of the Diversity of the Canadian Population, 2006 to 2031 (Statistics Canada 2010), the assumptions of which were also developed through consultation. Assumptions for international migration, total population fertility and total population mortality are as close as possible to the medium-growth scenario M1 of Population Projections for Canada (2013 to 2063), Provinces and Territories (2013 to 2038). Assumptions for the other components are as close as possible to the reference scenario of Projections of the Diversity of the Canadian Population, 2006 to 2031, which was updated to reflect the recent demographic context.

The main assumptions are as follows:

• The annual immigration rate is 7.5 per 1,000 and the composition of immigrants by country of birth represents the period from 2006 to 2011^Note 27;
• The net rate of emigration maintains the level recorded from 2002/2003 to 2011/2012 and emigration differences observed from 1995 to 2010 remain the same;
• The net number of non-permanent residents declines to zero in 2021/2022 and the composition of the new non-permanent resident population represents the composition of that population in the 2011 NHS;
• The fertility rate reaches 1.67 children per woman at the national level in 2021 and the gaps between the projected groups are maintained^Note 28;
• Life expectancy increases moderately at the national level, following the trends noted from 1981 to 2010, and the gaps between the projected groups are maintained;
• The upward trend in population education gradually levels off and the gaps between Aboriginal and non-Aboriginal populations are maintained;
• Trends in marital status for both Aboriginal and non-Aboriginal populations gradually slow down;
• The intragenerational religious mobility rates, based on estimates for 2001 to 2011, are maintained until 2036;
• The intragenerational linguistic mobility rates, estimated using linkages between the 2001 and 2006 censuses and the 2006 and 2011 censuses, are maintained until 2036; and
• Rates of intergenerational transmission of religion, languages and visible minority group, estimated using 2011 NHS adjusted data, are maintained until 2036.

Projection scenarios

Following the objectives of these projections, five scenarios were adopted. Scenario 1, the reference scenario, combines the assumptions of complete fertility convergence in 2036 with an assumption of constant intragenerational ethnic mobility and an assumption of internal migration consistent with adjusted estimates from the 2001 Census, the 2006 Census and the 2011 NHS. Scenario 1 is called the reference scenario because each of the other scenarios differs from this one by a single component, thereby making it possible to analyze the specific effect of fertility, ethnic mobility and internal migration on the future evolution of the Aboriginal population. Scenario 2, or the moderate convergence of fertility scenario, differs from the reference scenario in its fertility assumption; in this scenario, the fertility of Aboriginal people converges toward that of the non-Aboriginal population until half of the gap has been closed in 2036. Scenario 3—the constant fertility scenario—differs from the two previous scenarios by assuming that fertility does not converge. Scenario 4, or the no ethnic mobility scenario, is identical to the reference scenario except that it assumes that there is no intragenerational ethnic mobility as of 2011. Lastly, Scenario 5, or the no internal migration scenario, differs from the reference scenario only in that it assumes that there will be no internal migration as of 2011 (Table 7).

Scenarios 3 and 4 will generate the most and least Aboriginal population growth, respectively, at least at the national level.^Note 29They provide a plausible range of Aboriginal population growth with regard to past trends that reflect the uncertain future evolution (previously discussed) of fertility and ethnic mobility in the Aboriginal populations. Projection users are advised to consider this range rather than a single scenario. Note that Scenario 1 was called the "reference scenario" not because it is considered most likely to occur than others, but because it provides a reference for comparing the sensitivity of projection results for the other scenarios, that differ by only one component.

In addition, using only one assumption for each component that affects the non-Aboriginal population does not mean that the future evolution of non-Aboriginal populations is not uncertain. This choice was made to estimate the percentage of the total population represented by Aboriginal people under the five scenarios adopted, the growth of the rest of the population being equal.  For more information on possible alternative growth for the rest of the population, readers may refer to Statistics Canada's projections cited above. Other projections based on the new version of Demosim will be more specific to the potential growth of other groups within the rest of the population, under scenarios different from the ones used here.

Notes