How to measure changes in Canada's grocery bill

June 23, 2017

Say you're at the grocery store and you notice that your favourite brand of peanut butter is 50 cents cheaper than the last time you bought it. But when you look carefully, you see that the jar is 25mL smaller. The next biggest size costs a little less per mL, and as far as you can tell, it is the same size and price as the month before. So, is your favourite peanut butter getting cheaper, or more expensive? What about the price of peanut butter in general?

Price check!

Statistics Canada's price collectors have that information. This team records the prices of approximately 100,000 consumer items each month, in 40 cities across Canada. Of these items, almost half (45,000) are food. Price collectors visit approximately 500 food retailers, from small local shops to big box supermarkets, and methodically record the prices of items on the shelves.

They also record any changes in size, weight, number of units in a box and other features that determine what consumers are actually paying for, and note when an item is discontinued.

Richard Evans, Director of Consumer Prices Division at Statistics Canada, describes how they work: "The price collectors operate very independently. They go down the aisles, most people think they're other shoppers. They just go around, check prices, and they're out."

Creating an index

Collectively, these data are used to create the Consumer Price Index (CPI), Canada's main measure of changes in price levels. The index is also used to inform monetary policy and adjust tax brackets, rent controls, transfer payments and pensions.

The CPI basket contains the vast majority of goods and services that Canadians buy. Each item in the basket is weighted according to its relative importance in the consumer's overall spending. For instance, in any given month, people spend less on chocolate compared with vegetables. As a result, vegetables are weighted more heavily in calculating the index.

"The CPI is in the best shape it's ever been," explains Mr. Evans. "We update the basket every two years. We've significantly increased the quality of the sample, the monthly pricing that we do. We've done that by adding more prices. We also have a better method for selecting stores, and for selecting products in stores, than we used to have. We're world leaders in this area."

To make sure that the data are free from manipulation, Statistics Canada is tight-lipped about which items are on the list and which stores its price collectors visit. Individual collectors also have some latitude in choosing what brands to include in their data. (Chances are your favourite peanut butter is being measured several times.)

What matters most in the data collection is that the price collectors are consistent: they visit the same roster of stores, at the same day and week each month, and record the price of the same brands. This way, the data can be compared month over month, and will not be affected by inconsistent methodology.

More expensive groceries...than in 2015

What the data show is that, from a historical perspective, overall food prices have been increasing at a relatively steady rate for many years. With very few exceptions, across all types of food purchased from stores, and all provinces and territories, consumers are paying more than they did in the past. In fact, in May 2017 they were paying about 41% more than in 2002. At the same time, the All-items CPI increased 30%, so during those 15 years, food prices climbed noticeably faster.

However, Canadians caught a small break in 2017. The overall price of food purchased from stores was slightly lower in May 2017 than in May 2016.

Chart 1 - Price indexes for all items and food purchased from stores, May 2012 to May 2017
Description for Chart 1
Price indexes for all items and food purchased from stores, May 2012 to May 2017
  All-items Food purchased from stores
Source: Statistics Canada, CANSIM table 326-0020.
May 2002 99.7 100.3
June 2002 99.9 100.7
July 2002 100.5 100.6
August 2002 100.9 99.7
September 2002 100.9 98.4
October 2002 101.2 97.7
November 2002 101.5 99.9
December 2002 101.1 100.7
January 2003 102.0 101.8
February 2003 102.8 101.5
March 2003 103.1 101.7
April 2003 102.4 101.7
May 2003 102.5 102.1
June 2003 102.5 102.4
July 2003 102.6 102.3
August 2003 102.9 101.0
September 2003 103.1 100.0
October 2003 102.8 99.6
November 2003 103.1 101.0
December 2003 103.2 102.2
January 2004 103.3 102.2
February 2004 103.5 101.9
March 2004 103.9 102.1
April 2004 104.1 101.7
May 2004 105.0 102.8
June 2004 105.1 104.2
July 2004 105.0 104.0
August 2004 104.8 103.6
September 2004 105.0 102.8
October 2004 105.2 103.4
November 2004 105.6 104.8
December 2004 105.4 105.4
January 2005 105.3 104.8
February 2005 105.7 104.4
March 2005 106.3 105.0
April 2005 106.6 106.2
May 2005 106.7 106.4
June 2005 106.9 106.7
July 2005 107.1 106.3
August 2005 107.5 106.0
September 2005 108.4 104.5
October 2005 107.9 104.9
November 2005 107.7 105.6
December 2005 107.6 106.5
January 2006 108.2 107.8
February 2006 108.0 107.4
March 2006 108.6 107.2
April 2006 109.2 106.9
May 2006 109.7 107.7
June 2006 109.5 108.2
July 2006 109.6 108.5
August 2006 109.8 108.4
September 2006 109.2 107.8
October 2006 109.0 108.2
November 2006 109.2 109.1
December 2006 109.4 108.9
January 2007 109.4 110.2
February 2007 110.2 112.5
March 2007 111.1 111.9
April 2007 111.6 111.7
May 2007 112.1 112.0
June 2007 111.9 111.9
July 2007 112.0 111.4
August 2007 111.7 110.6
September 2007 111.9 109.2
October 2007 111.6 108.9
November 2007 111.9 109.8
December 2007 112.0 110.1
January 2008 111.8 111.4
February 2008 112.2 111.8
March 2008 112.6 111.6
April 2008 113.5 112.7
May 2008 114.6 114.1
June 2008 115.4 115.3
July 2008 115.8 116.2
August 2008 115.6 116.4
September 2008 115.7 116.5
October 2008 114.5 116.8
November 2008 114.1 119.7
December 2008 113.3 120.0
January 2009 113.0 121.0
February 2009 113.8 121.8
March 2009 114.0 122.2
April 2009 113.9 122.1
May 2009 114.7 122.5
June 2009 115.1 122.7
July 2009 114.7 122.7
August 2009 114.7 121.5
September 2009 114.7 119.7
October 2009 114.6 119.3
November 2009 115.2 121.2
December 2009 114.8 121.5
January 2010 115.1 122.1
February 2010 115.6 122.5
March 2010 115.6 123.1
April 2010 116.0 122.6
May 2010 116.3 122.8
June 2010 116.2 122.8
July 2010 116.8 123.4
August 2010 116.7 123.0
September 2010 116.9 122.3
October 2010 117.4 121.8
November 2010 117.5 122.4
December 2010 117.5 123.2
January 2011 117.8 124.4
February 2011 118.1 124.9
March 2011 119.4 127.6
April 2011 119.8 127.1
May 2011 120.6 128.0
June 2011 119.8 128.7
July 2011 120.0 129.7
August 2011 120.3 129.2
September 2011 120.6 128.2
October 2011 120.8 127.8
November 2011 120.9 129.4
December 2011 120.2 129.4
January 2012 120.7 130.5
February 2012 121.2 130.8
March 2012 121.7 130.1
April 2012 122.2 130.1
May 2012 122.1 131.2
June 2012 121.6 131.0
July 2012 121.5 132.2
August 2012 121.8 132.1
September 2012 122.0 130.0
October 2012 122.2 130.2
November 2012 121.9 131.3
December 2012 121.2 131.0
January 2013 121.3 131.3
February 2013 122.7 133.1
March 2013 122.9 132.3
April 2013 122.7 132.3
May 2013 123.0 132.9
June 2013 123.0 132.7
July 2013 123.1 132.8
August 2013 123.1 133.0
September 2013 123.3 131.4
October 2013 123.0 131.1
November 2013 123.0 132.7
December 2013 122.7 132.0
January 2014 123.1 132.6
February 2014 124.1 134.4
March 2014 124.8 134.6
April 2014 125.2 134.6
May 2014 125.8 136.2
June 2014 125.9 137.0
July 2014 125.7 137.0
August 2014 125.7 136.0
September 2014 125.8 135.3
October 2014 125.9 135.1
November 2014 125.4 137.1
December 2014 124.5 137.6
January 2015 124.3 139.7
February 2015 125.4 140.2
March 2015 126.3 140.2
April 2015 126.2 140.0
May 2015 126.9 141.8
June 2015 127.2 141.9
July 2015 127.3 141.8
August 2015 127.3 141.4
September 2015 127.1 140.6
October 2015 127.2 141.3
November 2015 127.1 142.2
December 2015 126.5 143.2
January 2016 126.8 146.1
February 2016 127.1 146.4
March 2016 127.9 145.8
April 2016 128.3 144.6
May 2016 128.8 143.8
June 2016 129.1 143.1
July 2016 128.9 143.4
August 2016 128.7 142.0
September 2016 128.8 139.3
October 2016 129.1 138.4
November 2016 128.6 139.4
December 2016 128.4 139.2
January 2017 129.5 140.3
February 2017 129.7 140.4
March 2017 129.9 140.6
April 2017 130.4 140.9
May 2017 130.5 142.1

So why does my bill seem so high?

Several factors influence how much individuals and families pay for food—and how much they feel they are paying for food. The CPI accounts for some of these factors, such as regional differences in price, changing prices of different food products, and seasonal price fluctuations for certain items.

There is also the very human tendency, called the negativity bias, to be more affected by negative changes—in this case, higher prices—than positive ones. "We tend to remember bad news," says Mr. Evans. "Bad things strike us. And we tend not to focus so much on price declines."

Indexing the prices of so many consumer goods in Canada is a tricky art, and one that involves a great deal of knowledge. But Statistics Canada has been refining the CPI for 90 years, perfecting its methods and formula to create an increasingly precise measure. This makes the CPI Canada's most authoritative source of changes in price levels.

Please note that comments are moderated. It may take some time for your comments to appear online. For more information, consult our rules of engagement.

Date modified: