Skip to main content

With LifePaths, Statistics Canada can create life histories for millions of virtual Canadians to create socio-economic and demographic features that replicate Canadian society, allowing researchers to forecast the impact of policy decisions in a wide array of areas, such as social insurance, pensions or health care.Sean Kilpatrick/The Canadian Press

Former Statistics Canada official Michael Wolfson is criticizing the agency's decision to stop funding its LifePaths database, saying the program has been essential for researching the long-term impacts of policy decisions.

Mr. Wolfson, a professor at the University of Ottawa who was previously assistant chief statistician at Statistics Canada, has written a new paper on retirement adequacy, which includes comments about his disappointment over the decision to cut funding to the LifePaths database, which he has used in his own work on retirement income.

"As a result, discussion of multibillion-dollar policies – discussions that could be informed by far smaller investments in statistical infrastructure – can now be pursued in ignorance," he writes.

Statistics Canada stopped supporting the LifePaths modelling tool at the end of 2014, which means the database is not being updated with new data.

Statistics Canada spokeswoman Nadine Lacroix said the resources required to update and maintain the model were too great, and it "was no longer feasible" to continue the program.

She said the agency is developing a new "dynamic socio-economic" modelling tool that will be structured to ensure "sustainability, efficiency and responsiveness to client needs." Statistics Canada expects to solicit feedback from stakeholders on the proposal next year.

LifePaths is a complex modelling tool developed in the 1990s that contains data on Canadians starting from 1971. It was started during Mr. Wolfson's time at Statistics Canada to project demographic trends for Canadians decades into the future. It was intended to help shape public policy in numerous areas – including pensions, education and health care – by modelling the impacts of various policy alternatives.

The decision to stop maintaining the model comes amid broader criticisms over cuts to Statistics Canada research, most notably the 2010 decision to eliminate the mandatory long-form census in Canada and replace it with a voluntary survey.

In an interview, Mr. Wolfson said he has not published comments critical of Statistics Canada in the past, and worries he is being disloyal to his former colleagues by speaking out now.

"But I felt it was sufficiently important that I really felt I had to do it," he said Monday.

In his paper, which will be published this week on the website of the Canadian Centre for Policy Alternatives, Mr. Wolfson said federal and provincial pension policy decisions are expensive and require a decades-long planning horizon, so they need strong supporting analysis. He said LifePaths "provides the best available estimates" of retirement income adequacy by taking into account a full array of demographic factors, such as employment trends, birth rates and life expectancies.

"Fundamental national statistical resources for sophisticated analysis of pension and related policy questions have been seriously crippled," he argues.

Mr. Wolfson wrote his paper to rebut two recent reports published by consulting firm McKinsey and the C.D. Howe Institute, both saying there is no looming retirement crisis in Canada because most people are saving enough to maintain their standard of living when they stop working.

He said he remains convinced that a substantial portion of middle-income Canadians will be worse off in retirement, and his own research using LifePaths data suggests 50 per cent of middle-income Canadians will experience a significant drop in their living standard in retirement. He defines a significant decline as a drop of 25 per cent or more in their net income by age 70 compared with their pre-retirement net income.

He said nothing in either of the two recent studies "seriously challenges" conclusions in other reports that suggest many middle-income people are facing retirement problems because they aren't saving enough and are increasingly unlikely to have workplace pension plans.

Mr. Wolfson said in the interview that Finance Minister Joe Oliver recently cited the McKinsey study and noted it found most people are saving enough for retirement, but he doesn't believe the data are rigorous enough to shape policy decisions about reforms to Canada's pension system, including decisions about whether to expand the Canada and Quebec Pension Plans.

"For those Canadians who actually risk facing a substantial drop in their living standard after retirement – Canadians who stand to benefit from CPP/QPP expansion – flawed, opaque studies tilt the policy playing field against them," he wrote in his report. "And that would be another stake in the heart of evidence-based policy decision-making."

He argues the McKinsey report contains too little detail about its methodology to allow the findings to be critiqued or duplicated.

"Anecdotes do not do justice to this debate, nor do analyses that fail to meet the fundamental requirements of transparency and replicability – a hallmark of truly open government," he said.

What is lifepaths?

The database, operated by Statscan, allows researchers to simulate the impact of long-term changes. Its data goes back to 1971, and it can generate projections decades into the future.

How does it work?

It creates life histories for millions of virtual Canadians to create socio-economic and demographic features that replicate Canadian society.

What can it do?

Researchers can forecast the impact of policy decisions in a wide array of areas, such as social insurance, pensions or health care. The data allow researchers to examine policy impact on individuals and cross-sections of the population, and between different cohorts and generations.

Follow related authors and topics

Authors and topics you follow will be added to your personal news feed in Following.

Interact with The Globe