Many of us have always had doubts about the accuracy of the figures published by Statistics Canada, particularly in relation to food inflation. In recent years, it had become difficult to understand why the food inflation announced by the federal agency in Ottawa was so different from what consumers perceived at the grocery store. Thanks to the investigative work of a journalist, we now have an explanation.
Paul Webster wrote a great article last week for the Toronto Star on the consumer price index (CPI). He was the first journalist in Canada to challenge and question the data presented by Statistics Canada on food costs. The federal agency even granted her an interview, an extremely rare occasion, where she admitted that her data collection strategy in the field required a different approach.
The article revealed that some experts believe the Canadian CPI in no way reflects what is happening at the grocery store. BetterCart, a Canadian firm that collects some 20 million pieces of data a week to better understand food inflation across the country, shows how stark the price disparity can be.
For example, while Statistics Canada recently told us that the price of ketchup has fallen by about 5.9% since January, BetterCart tells us that the price of that same ketchup has increased by 7.3%.
Frozen fries are up 5.9%, but for BetterCart, the increase is more like 26.2%. The same goes for butter: an increase of 2.8% compared to 35.5%. The buttergate and the palm oil crisis in the dairy industry may have forced dairy farmers to increase their cost of production, as expected.
In short, there is a variety of products for which the price increase is underestimated. In some cases, the opposite is true, but for about a quarter of products, the increase recorded by BetterCart exceeds that of the national body. Statistics Canada appears to have difficulty tracking market changes, especially when prices show great volatility.
Previously, no one could upset the federal agency, but new technologies and hyper-communicative information networks now make it easy to monitor retail prices, virtually in real time. First, Statistics Canada only works with three food chains. The agency tries to increase the scope of the data it receives each month, but to no avail. She therefore probes few chains, few brands, few stores. Indeed, we can ask ourselves which brands the sampling is made up of. Consumers know there is a difference between a discount chain and a store that typically has a higher grocery basket cost.
And, of course, to collect this data you need employees, but the lack of manpower is already creating a huge problem for our economy. We have no idea if the price reading in the field is being done as rigorously as it should.
On the other hand, Statistics Canada only relies on about 100,000 data points for its monthly report. The agency must therefore use averages and approximations which can give us a somewhat misleading picture. Each store can hold 8,000 to 10,000 products, and in some cases that number can even exceed 30,000. For greater clarity, Statistics Canada needs to increase the scope of data collection across the industry.
Reduced quantities
To add to this, “reduflation” is also another factor that is difficult to measure in the long term. Most processors will reduce quantities without changing prices, especially during an inflationary period like the one we are currently experiencing. Instead of scaring customers off with higher prices, companies are reducing quantities. However, in Statistics Canada’s grocery basket, approximately 70% of the products selected showed quantities that no longer exist on the Canadian market. Most formats have changed. Even though Statistics Canada mentions that it takes into account the effects of “reduflation”, there is every reason to believe that by comparing the CPI with BetterCart, quantity conversions are not being measured adequately.
But there are solutions. Instead of relying on industry, the United States Department of Agriculture turned the tables by creating a consumer panel of approximately 120,000 American households. Retail data collection is done by consumers, not grocers. An interesting option for our own federal agency to consider.
In sum, no one is to blame for this failure, but accurate measurement of food inflation is of crucial importance. Statistics Canada has an excellent reputation around the world, but the agency must now think about its strategy and the market. To do its job well, it must review the way it operates and its budget in order to fully understand the effects of food inflation on the lives of Canadians.
What do you think? Express your opinion