Saturday, August 20, 2011

A note on changing dependency ratios and migration

The concerns I expressed about Canadian immigration patterns and policies risking the reproduction of patterns of socioeconomic inequality linked to immigration status and ethnicity are all the more salient in light of this Statistics Canada report on projected changes in the Canadian labour force.

Using a range of projection scenarios, the labour force is projected to grow to between 20.5 million and 22.5 million by 2031. In 2010, the labour force numbered about 18.5 million.

[. . .]

Between 1971 and 1976, when the large baby-boom cohorts were entering the labour market, the labour force increased at an average rate of just over 4% a year. This growth rate slowed to about 1.4% between 2006 and 2010. By 2016, growth is projected to be less than 1% on average in all scenarios. Projections show it could slow even further to between 0.2% and 0.7% in the period from 2021 to 2026.

In four of the five scenarios, the growth is projected to stop slowing after 2026, once most baby boomers have left the labour force.

The projections also suggest that, if recent trends continue, the labour force will become older and increasingly ethnoculturally diverse. Close to one person out of four in the labour force could be aged 55 or over by 2021. There would also be higher proportions of foreign-born people and people belonging to a visible minority group (as defined by the Employment Equity Act) in the labour force.

As the growth of the labour force loses momentum, the population of seniors aged 65 and over is projected to grow increasingly rapidly as a result of population aging and the entry of the baby boomers into this age range.

[. . .]

In 2010, the participation rate was 67.0%; by 2031, it is projected to range between 59.7% and 62.6%, which would be the lowest observed since the late 1970s.

The projected decline in the overall participation rate over the next two decades would be largely attributable to demographic phenomena, such as the aging of the baby-boom cohorts, increasing life expectancy and a fertility rate below the replacement level of 2.1 children per woman.

The aging of the baby boomers, which is largely behind the projected decline in the overall participation rate, has had a major impact on the aging of the labour force. Between 2001 and 2009, the proportion of people in the labour force aged 55 and over rose from 10% to 17%, an increase of 7 percentage points in nine years. The first baby boomers reached the age of 55 in 2001.

[. . .]

In 1981, there were roughly six persons in the labour force for each retiree. By 2031, or 50 years later, this ratio is projected to decline to less than three to one, according to all five scenarios. The ratio is projected to decline in every province.

By 2031, roughly one in every three people in the labour force could be foreign born. Between 1991 and 2006, the percentage of foreign-born people in the labour force rose from 18.5% to 21.2%. If recent immigration levels were to continue, that proportion is projected to reach almost 33% in 2031, according to most scenarios.

If labour forces remain largely static, it's not good at all to have a large and growing share of the national population shut out for reasons linked to ethnicity. This is the problem facing the western province of Saskatchewan, where the First Nations population is growing rapidly but remains underemployed.

"If you have a large 65-plus (population) and don't match that with an increase in the labour force size, then your dependency ratio or the health care costs or other costs associated with the retired group may be a concern," said University of Saskatchewan policy and labour expert Rose Olfert.

"On the other (hand), if those baby boomers have good retirement plans and have lots of good discretionary spending, they can also be a source of expenditure and growth for the economy as they begin to spend all their hard-earned money that they acquired over their lifetime."

Olfert warned that as bay boomers retire they will leave "a relative vacuum of experience and skills on the part of the incomers." That puts skilled and educated labour at a premium.

It's not just a numbers game, she said, but an issue of quality. Attracting and retaining workers will be a difficult and competitive process and Saskatchewan will be competing with other provinces and other countries to fill the void.

[. . .]

"We need to engage aboriginal people like never before, both as employers and as employees," he said.

High unemployment rates in the aboriginal community means that population could help meet the labour demand and, with numerous policies and programs in place, progress is being made. A recent Statistics Canada report shows offreserve aboriginal employment in Saskatchewan grew by 3,300 jobs in July from a year ago. Employment among aboriginal youth was up by 500 jobs.

"It's just that the challenge is large and even though these policies and practices are successful and are working to a large extent, they seem to not be enough," Olfert said. "It's as though you are running fast, but you have to run a little bit faster to keep up."

As for the situation in labour-exporting New Brunswick, it can be safely described as potentially catastrophic.

In New Brunswick, the labour force participation rate was at 64 per cent in 2010 and is expected to fall to between 54 and 58 per cent by 2031 - roughly six percentage points below the national average.

While the proportion of the labour force aged 55 and over is expected to reach 24 per cent in Canada by 2031, it's expected to be three percentage points higher in New Brunswick.

At the same time, the ratio of workers to retirees by 2031 is expected to be three to one in Canada and two to one in New Brunswick.

The bleak outlook is partly due to the fact the population in New Brunswick is already older than in Canada overall, Lebel said.

Seniors made up 14.7 per cent of the province's population in 2006, one percentage point higher than the proportion for Canada as a whole, according to census data.

At the same time, New Brunswick's fertility rate of 1.59 children per woman is lower than the national average - 1.68.

Almost half of the population could be more than 65 years of age by 2031, according to medium-growth projections based on historical trends by Statistics Canada.

Saskatchewan at least has a reasonable prospect of following in the path of neighbouring Manitoba.

The famous United Nations report on replacement migration of a decade ago produced some interesting results. One of these results was the confirmation that rapid population aging is irreversible. In the admttedly extreme case of South Korea, experiencing rapid aging owing to the collapse in fertility rates from the very high level of the 1960s to the very low levels of the past decade. In order to keep the exceptionally 1995 high ratio of workers to seniors, 5.1 billion immigrants would be needed. Among other things, transforming the southern half of the Korean peninsula into a vast arcology would be problematic. Declining dependency ratios are inevitable; it's occurring worldwide. Even immigration won't necessarily change dynamics, since immigrant populations would have to have exceptionally and unrealistically high birth rates to reverse secular trends (this, in turn, having its own obvious problems).

That same report on South Korea did suggest that certain relatively medium-term improvements might be sustainable; the numbers and proportions of immigrants and immigrant descended people necessary to sustain either the peak population of South Korea or the peak working-age population of South Korea seems possible. South Korea is more open to immigration than neighbouring Japan, at least, and the movement of people from Vietnam, China, and the Philippines associated with marriage-driven immigration could contain the seeds for a broader migratory movement. South Korea might come to emulate a Canada that is enacting something very much like those policies right now, you could say. For those policies to work, though, you need a labour force that is permeable to people of diverse backgrounds at every level. Besides being a waste of resources, the ethnic segmentation of a workforce carries obvious dangers for social cohesion. Creating immigrant underclasses through migration policies which don't give immigrant the chance to acquire the cultural and economic capital needed to integrate, as we've seen in many countries around the world, is--besides being foolish--self-destructive.

Migration policies really need to be thoroughly thought-out.


Colin Reid said...

The big question with dependency ratios is whether people are living longer in good health, or just experiencing a longer period of age-related illness before death. If it's the former, the issue is social/political: 65 or whatever is just a number, and there are many older people who have the capacity to work but don't want to or can't get a job. I think employers' narrow-mindedness when it comes to hiring (whether it's about age,sex,race, whatever) is a very serious problem, but it's easier to fix than the demographics, and we should expect the retirement age to increase over time (say by 1 month per year) as a matter of course and adjust our notion of dependency ratio accordingly. If on the other hand there are a lot more people who are simply too ill to work, then we have a real demographic problem and hard choices to make in how taxes and spending are distributed across the age spectrum.

I agree that if we rule out very rapid population growth, then immigration alone can't stop the change, and neither can an increased fertility rate; ultimately nothing can stop the ageing of the population except for an abrupt drop in adult life expectancy. Modern medicine is the victim of its own success, it's as simple as that.

Noel Maurer said...

I haven't looked at the numbers, but the South Korean situation seems to be a canard. It had an unusually high ratio of workers to dependents --- falling to the ratio of the United States is less than a problem.

Or am I missing something?

Michel S. said...

The title seems to be misleading -- dependency ratio is defined as the number of *dependents* per working-age person, right? In which case it's rising -- cf. the inverse dependency ratio, which declines as the proportion of working-age population shrinks.