Why planners keep getting population projections wrong

It may be futile to try to predict the future. But that doesn’t stop local politicians and planners from trying.

Two decades ago, local planners vastly overestimated how many people would move to the Fraser Valley over the next two decades.

Now, the pendulum may have swung in the opposite direction—at least in the region’s smaller communities and rural areas. The resulting disparities, and recent population statistics, show why predicting growth is often more of a political art than a demographic science.

Projecting the region’s future population isn’t just a make-work project for the Fraser Valley Regional District and its staffers. The figures provide vital information to help the regional district and its member governments develop land use plans and determine how much money they will need to spend over the coming years and decades on infrastructure and amenities.

This spring, the FVRD adopted a brand new regional growth strategy that is informed by population projections based on BC Stats data and estimates from 2022. But new population counts released this month illustrate the difficulty—and, potentially, the impossibility—of predicting future populations based on previous trends.

Two decades ago, when the FVRD was creating its last regional growth strategy, the consulting team in charge took provincial projections and revised them sharply upwards. History showed that to be a mistake, with the projections dramatically overshooting growth in most of the valley’s cities.

This time, the FVRD seems to have more closely relied on the provincial projections, while accounting for undercounting in the census. The resulting projections, which were based on 2021 numbers, came with the caveat that they were “high estimates.” But those reports already have missed the mark for some communities.

Rural areas

New official population counts from Statistics Canada suggest that the last year has seen an explosion of growth in the Fraser Valley’s smaller communities and, especially, the region’s rural electoral areas outside of city boundaries. And in some of those areas, the population is already a decade ahead of expectations. (The regional district’s projections for those communities are particularly important since electoral areas are governed directly by the regional district, rather than a municipality.)

The FVRD’s regional growth strategy projected the electoral areas would have a population around 12,440 in 2030 and around 13,300 in 2040. But after a surge of building in the Cultus Lake and Popkum areas, the electoral areas have surged past that 2030 target number and are halfway to their 2040 future.

With a major development near Lake Errock under consideration, even more people might be on the way.

Hope, Kent, and Harrison

BC Statistics' algorithms had predicted significant growth in Hope. And after a couple years, the municipality is more or less on track to hit the 7,600 people expected for 2030.

But in the region’s two other small municipalities—Kent and Harrison—new development has the communities growing far faster than projected.

With only minimal population growth projected for Harrison Hot Springs, the recent modest influx of people has left the village with as many residents as had been expected in 2030.

In Kent, meanwhile, a surge in newcomers has the community just seven residents shy of the population projected for 2040.

Abbotsford, Chilliwack, and Mission

In the valley’s larger cities, growth hasn’t yet surged past 2030 goalposts. Both Chilliwack and Mission are roughly on pace to modestly exceed their projected 2030 populations—110,493 and 48,896, respectively. Abbotsford, meanwhile, had been projected to hit 192,000 people in 2030 but seems unlikely to come close to that mark, having grown slower than the provincial average in recent years.

The figures underlie a fundamental challenge that goes to the heart of efforts to predict populations and plan for future growth.

The projections by BC stats use demographic information, like the ages of residents, to predict future population trends. A community with an abnormally large number of 20-year-old residents can likely expect a baby boom in a decade. A certain portion of elderly residents, meanwhile, can be expected to die in the years that follow. The figures can help municipalities plan for aging populations and anticipate the needs for schools, recreational facilities, and programs aimed at people of different ages.

But the population projections come with massive limitations. The pace of home-building in cities and towns can vary incredibly over time, and previous growth patterns, which are also incorporated into projections, can mislead about future trends.

The guesswork of projections

The FVRD’s previous growth strategy, crafted in the early 2000s following a decade of rapid growth in Chilliwack and Abbotsford, used predictions that wildly overestimated growth over the following decades. The early returns suggest that in many communities they may miss the mark in the other direction. And there’s good reason to think that even where they align with growth so far, future deviations may be coming

There are a few chief reasons why it’s hard to predict how municipalities will grow in British Columbia.

The most obvious is that significant population growth tends to be linked not to organic flows of people and new residents being born, but to the number of homes being built in any one community. That home-building behaviour is itself influenced by an array of factors—the supply of land, zoning restrictions, political activity, migration and immigration, and demand for housing—that frequently change in unpredictable ways.

The BC Stats algorithm cannot, for instance, predict that the owner of a gravel mine will seek to turn their property into a residential development—as is the case in Lake Errock. They can’t forecast whether a provincial government will override local planning rules in order to spur development in single-family neighbourhoods. And they don’t fully account for land use plans, developer incentives, and changes to the pace of permit approval.

Sometimes the demand for land and homes, the availability of jobs, and the attractiveness of a community drives population growth. But recent trends suggest those factors are having a minimal impact on British Columbia’s municipalities. That can be best seen in the Lower Mainland, where adjacent communities are growing at vastly different rates. Vancouver and Abbotsford, for instance, are two of the slowest-growing large municipalities in the province while their immediate neighbours, Burnaby and Langley Township, are two of the fastest growing places. The same dichotomy is seen around the province.

Those trends suggest that growth and home-building is chiefly driven not purely by resident demand or demographic trends, but also by the politicians and local governments themselves who can make it easier (or harder) to construct new houses, townhomes, and apartments.

In other words: While population projections can be used to influence planning and policies, it’s precisely those plans and policies that usually determine just how many people arrive in a community.

A person may not be able to manifest change in their own lives. But a local government? Now that’s another story.