Section 5: Analysis of previous editions of the projections
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.
Using the components method, Statistics Canada has produced population projections of Canada, the provinces and the territories seven times since 1974. From one edition to another, the results vary considerably, insofar as each edition revises the preceding assumptions in light of more recent trends.
The current projections (2009 to 2036) revise upward the projected population levels of the preceding edition (2005 to 2031). These differences are largely attributable to increased fertility, the increase in the number of immigrants and positive net figures for non-permanent residents in the early years of the period. Already in 2009, the estimated population exceeded the projected population in the high-growth scenario of the 2005 to 2031 edition.
The same is true for the 2000 to 2026 edition. Proposing a total fertility rate of 1.8 that would be reached only in 2026, the high-growth scenario of the 2000 to 2026 edition underestimated births and non-permanent residents while it overestimated deaths.
This said, some caution is required in making comparisons with relatively recent editions of the projections, since such comparisons involve judging how long-term projections match actual values soon after their publication. But what about projections published longer ago? Clearly, even if the period covered by those projections has ended, it is instructive to study the results of the edition which was published in 1974 1 and which projected the population for the period 1972 to 2001. That edition was relatively successful in estimating the growth of the population (Chart 5.1). The medium-growth scenario suggested that the Canadian population would reach 31,050,000 in 2001, which is very close to the estimated population of 31,020,000. 2 However, care should be taken not to attribute this result to the accuracy of the assumptions underlying that scenario. Those assumptions called for a total fertility rate of 2.20 children per woman to be reached in 1985 and an annual number of immigrants of 120,000. In 2001, the total fertility rate was instead 1.54 children per woman and the number of immigrants was more than double. These divergences were important: the age structure that existed in 2001 differs greatly from the one projected for the same year by the 1972 to 2001 edition, in which, among other things, the very young population was much more heavily represented (Chart 5.2).
In subsequent editions, the growth of the Canadian population closely followed the levels projected by the high-growth scenarios. The range suggested by the 1976 to 2001 edition just manages to capture the level reached in 2001, but this is not the case with the 1984 to 2006 and 1989 to 2011 editions, which underestimated growth (Chart 5.3). The underestimation of future immigration levels, which increased significantly in the early 1990s, largely explains why these editions aimed below the actual levels.
Produced after a period during which fertility and immigration underwent major increases, the 1993 to 2016 edition stands out: in its medium-growth scenario, it proposed a total fertility rate of 1.7 children per woman and an annual number of immigrants of 250,000, which is very similar to the assumptions adopted for the medium scenario of the current edition of the projections. In the short run, fertility rates and immigration did not remain at those levels, with the result that those projections overestimated the growth of the Canadian population over the period 1991 to 2009 while suggesting a younger age structure. However, the range offered by those projections encompasses the actual levels starting in 2004.
In short, past editions have often tended to underestimate Canada's population growth. Moreover, even in the high-growth scenarios, they most often showed growth falling off in the long run. One reason for this slowing of growth is that for immigration (a component that continues to play an expanding role in the Canada's population growth), the number of immigrants annually projected was fixed, and thus it generated a contribution to growth that grew proportionally smaller as the population increased. 3 In the current edition of the projections, immigration is projected according to a fixed rate (rather than a fixed number), which results in an annual contingent of immigrants that evolves in proportion to the size of the population. Added to relatively high fertility reaching 1.9 children per woman starting in 2013 and, at the end of the period, the near or total disappearance of the baby boom generation, immigration contributes to a much higher growth curve in the high-growth scenario than was seen in previous editions. Overall, it is perhaps this trend that distinguishes the current edition of the projections from those published previously.
- Date modified: