The low-inflation refrain may seem at odds with reality for consumers, who have watched shelter, food, utility and tuition prices swell in recent years. Many are skeptical the official inflation rate is reflecting what they see in their monthly bills.
Accuracy is crucial because the consumer price index influences everything from interest rates to rental agreements, old age security pensions and cost-of-living rates. Last week, Ontario's government said it would tie future minimum wage hikes to changes in the CPI.
Some of the disconnect may stem from the fact that CPI is not a measure of absolute price , but rather how prices change over time. Some items, such as bread, may not have budged in the past year – but they've plateaued at a high level. We also tend to notice more when prices go up.
Another factor is that frequently purchased items tend to see higher rates of inflation, while others, such as TVs, have fallen in price, said Richard Evans, director of the consumer prices division at Statistics Canada.
And the accuracy debate goes both ways. Several studies in the past decade have found Statscan over-estimates – not under-estimates – consumer price inflation.
Housing is the biggest bone of contention. The consumer price index shows no surge in costs (partly because of low mortgage rates) – even as average home prices have risen 84 per cent in the past decade, according to the Canadian Real Estate Association. Shelter is the biggest chunk of household budgets – but house prices are not directly in the CPI. Rather, Statscan tracks the cost of running a home such as maintenance costs, property taxes and mortgage interest costs.
Some inflation experts say Statscan could do a better job at tracking shelter prices. Erwin Diewert, a leading Canadian economist and world-renowned productivity expert who has sat on Statscan's price measurement advisory committee since 1984, says he is "not in favour" of the agency's approach to housing because it is not a widely accepted one, but says he's been "unable to convince them to give it up."
He'd prefer that the cost of equity tied up in a house be added to the shelter component of the CPI, a method of measurement which would have likely increased consumer price inflation in recent years.
Finn Poschmann, vice-president of research at the C.D. Howe Institute, thinks house prices should be more directly captured in the CPI, which would make the index a better guide to monetary policy.
That approach would also have meant inflation in the past few years would've looked higher than it did.
Statscan's Mr. Evans says the agency is "very comfortable" with its methodology, though he acknowledges "on shelter, this is one of the areas where there is no consensus."
The agency is in its fourth of a five-year, $45-million project to improve the CPI that means more frequent updates to its consumer basket and an expanded range of products and stores it's sampling from. The CPI now "is better than it's ever been," Mr. Evans says.
(Editor's note: Statistics Canada would like to clarify that while house prices are not explicitly reflected in the shelter cost component of the consumer price index, house prices are used as an input to calculate homeowners' replacement costs and mortgage interest costs.)