Flag of Canada
Government of Canada Government of Canada
 
Français Contact Us Help Search Canada Site
About Us Services Where You Live Policies & Programs A-Z Index Home
    Home >  Programs and Services > Policies, Planning and Reporting
Services for you

Applied Research Bulletin - Volume 4, Number 2 (Summer-Fall 1998) - September 1998

  What's New Our Ministers
Media Room Forms
E-Services
Publications Frequently Asked Questions Accessibility Features

  Services for: Individuals Business Organizations Services Where You Live
 

The Market Basket Measure Constructing a New Measure of Poverty

 ContentsNext

Who is poor in Canada? Ask this question and you will become part of an ongoing debate on the differing concepts of poverty and the ways poverty is measured in this country.

The Federal/Provincial/Territorial Working Group on Social Development Research and Information is developing a new measure of poverty which could achieve greater consensus on this issue. Allen Zeesman, Michael Hatfield and Stéphane Gascon of the Applied Research Branch were assigned the task of drafting a preliminary version of the "Market Basket Measure" (MBM).

Current Measures of Poverty and Low Income

The MBM is best understood in the context of current measures of poverty and low income in Canada. Although Canada has no official measure of poverty, Statistics Canada does calculate two measures of low income:

  • Low Income Cutoffs (LICOs)—This cutoff line is the income level where a household will, on average, spend on food, clothing and shelter a share of its pre-tax income that is 20 percent higher than the average family.
  • Low Income Measures (LIMs)—The LIMs are equal to half of the median pre-tax income adjusted for family size.

Canadian newspapers often use the LICOs as a measure of poverty. Because data using the LICOs are readily available and have been collected since the mid-1960s, the media have adopted the LICOs as the "official" measure. In addition, a number of advocacy groups, including the National Council of Welfare, use the LICOs to measure poverty. But not everyone agrees with the approach that forms the basis of the LICOs and the LIMs.

Differing concepts of poverty underlie the disagreement on poverty measurement. One approach sees poverty as a lack of social inclusion. According to this view, people or households are poor if they cannot maintain a standard of living which makes them, in some sense, participating members of their community. The Community Social Planning Council of Toronto periodically calculates budget guides based on this social inclusion approach to measuring poverty.

Another approach sees poverty as an inability to purchase subsistence levels of food, clothing and shelter. This is the concept underlying a measure developed by Christopher Sarlo for the Fraser Institute.

These radically different concepts serve to fuel the debate concerning poverty in Canada. Advocates of a subsistence concept of poverty believe the LICOs and the budget guides of the Community Social Planning Council of Toronto grossly overstate the level of poverty. Supporters of a social inclusion concept, on the other hand, argue that measures such as Sarlo's glaringly understate the level of poverty. It should come as no surprise, then, that consensus has not formed around any one measure of poverty or low income in Canada to date.

A second level of disagreement is whether poverty measures should be relative or absolute. The LICOs and the LIMs use a relative method. Based on a fixed percentage of average or median levels of consumption or income, they measure low incomes in relation to all incomes in the country. An absolute measure of poverty, on the other hand, defines a minimum acceptable standard of living, represented by a basket of goods and services. The cost of that basket is then priced over time.

The Federal/Provincial/Territorial Working Group is attempting to construct an absolute measure of a concept of poverty that might achieve greater consensus and serve as a more efficient tool than the existing measures. The Market Basket Measure is based on the concept of "necessaries" which was defined by Adam Smith as "whatever the custom of the country renders it indecent for creditable people, even of the lowest order, to be without."

The MBM approach is based on the lack of necessities with the key being the interpretation of the word "creditable" in today's social context. In his definition of necessaries, Adam Smith's creditable person rises above subsistence levels of food, clothing and shelter. It does not, however, imply a standard of living as comfortable as that implied by the concept of social inclusion.

The Gap Filled by the Market Basket Measure
Concept of Poverty Method of Measurement
  Absolute Relative
Subsistence Sarlo  
Creditability Market Basket After-Tax Low
  Measure Income Measure (LIM)
Social
Inclusion
Community
Social Planning
Council of 
Toronto
Low Income
Cutoffs (LICOs) 
and Low Income 
Measures (LIMs)

The Market Basket Measure

The ability to purchase the market basket will allow a family of two adults and two children to achieve a creditable standard of living. To purchase the market basket, the family must have sufficient income—a "threshold" income—to:

  • eat a nutritious diet—The family will be capable of purchasing a Nutritious Food Basket, as defined by Health Canada.
  • buy clothing for work and social occasions
  • house themselves in their community—The family will have the ability to rent a median-cost, three-bedroom apartment in their community.
  • pay for other necessary expenditures—The family will be attributed an amount equal to 60 percent of the combined cost of the food and clothing budgets to cover other necessary expenditures such as personal care, household needs, furniture, telephone service, public transportation, reading, recreation, entertainment and school supplies.

The costs of the items in the preliminary Market Basket Measure vary across the country. In addition, the cost of the basket is adjusted for family sizes and configurations other than the reference family of two parents and two children. To determine the level of income available to consume the market basket items, the following deductions should be made from pre-tax income:

  • income taxes and payroll taxes paid by members of the household;
  • child care costs incurred to enable both parents or a lone parent to work for pay;
  • child support payments made by non-custodial parents; and
  • actual out-of-pocket medically prescribed expenditures for dental and vision care, prescription drugs and prescribed aids for persons with disabilities.

In the preliminary calculations of the MBM, unfortunately only income taxes paid could be deducted due to unavailability of the other data.

Comparing the Market Basket Measure and the LICOs

Conceptually, the Market Basket Measure has a number of advantages over the LICOs. It is more sensitive to geographical differences in real living costs—particularly rents. The derivation of the MBM is more transparent and understandable to the average person. And, it compares the living standard it establishes to the income actually available to purchase the goods and services that fill the market basket. The fact that the Market Basket is related to actual purchasing power makes it sensitive to all changes in government policy including those that affect either the financial resources of or the taxes paid by each family, and those that influence the costs of services such as child care, prescription drugs, dental and vision care, and aids for persons with disabilities.

A comparison of the Market Basket Measure thresholds to the 1996 LICOs reveals a marked difference between the two measures. The national incidence of low income using the LICOs is significantly higher than the incidence of poverty calculated using the MBM—17 percent using LICOs compared to 12 percent using the MBM.

Geographical differences—Rents vary widely across the country. The variations in the costs of food and clothing are much smaller. The overall result is that the Market Basket Measure thresholds in high-rent provinces such as Ontario and British Columbia are much higher than thresholds in low-rent provinces such as Quebec, Manitoba and Alberta. A comparison of the MBM poverty thresholds to the LICOs reveals that the results are closer for the high-rent provinces than for the low-rent provinces. All of the poverty thresholds, however, are lower than the LICOs. (See table next page.)

Children and poverty—The aggregate poverty gap for children—the difference between actual incomes and the thresholds for all Canadian families with at least one child under age 18 with incomes below the thresholds—is significantly smaller using the Market Basket Measure than using the LICOs: for 1996, $3.3 billion compared to $6.8 billion.

Low Income Cutoff Levels and Market Basket Measure Thresholds
For Families of Two Adults and Two Children by Province and Community Size
  Population over 500,000 Population 100,000 to 499,999 Population 30,000 to 99,999 Population under 30,000 Rural Communities
Low Income Cutoff Levels (LICOs) $ 32,238 $ 27,651 $ 27,459 $ 25,551 $ 22,279
Newfoundland NA $ 21,234 NA $ 19,986 $ 18,714
Prince Edward Island NA NA $ 21,310 $ 20,074 $ 20,074
Nova Scotia NA $ 21,291 $ 19,899 $ 21,447 $ 19,635
New Brunswick NA $ 19,230 $ 20,742 $ 18,474 $ 17,982
Quebec $ 19,953 $ 19,077 $ 18,825 $ 18,657 $ 17,889
Ontario $ 25,194 $ 23,202 $ 22,722 $ 22,362 $ 20,370
Manitoba $ 21,746 NA $ 21,086 $ 19,970 $ 19,730
Saskatchewan NA $ 20,582 $ 19,190 $ 20,162 $ 19,226
Alberta $ 19,930 NA $ 19,486 $ 19,750 $ 17,626
British Columbia $ 25,196 $ 24,128 $ 22,712 $ 22,136 $ 20,516
Note: NA indicates that there are no communities of this size in the province.
Source: Federal/Provincial/Territorial Working Group on Social Development Research and Information, Construction of a Preliminary Market Basket Measure of Poverty, 1998, p. 20

The composition of poverty—Comparing the LICOs to the Market Basket Measure also affects the composition of the population in poverty. In order to compare the differences in composition for equal levels of poverty, the LICO thresholds were reduced to produce the same 12 percent national poverty rate generated by the Market Basket Measure. The effects on the composition of the poor at the same overall poverty rate are revealed in the table "Percentage Distribution of the Poor." The MBM shifts the composition of the poor towards Ontario and British Columbia, families with children, and families with earnings as their main source of income.

Work is continuing toward refining the Market Basket Measure in the areas outlined in this article before submitting the measure to public evaluation and comment. It is hoped that one day the MBM may serve as one of several acceptable measures of poverty in Canada.

Percentage Distribution of the Poor
By Province, Family Type and Income Source, 1996
  Adjusted LICOs (Low Income Cutoff Levels) Market Basket
Measure 
Provinces 
Ontario and British Columbia 44.9 54.2
Manitoba, Quebec and Alberta 43.0 32.8
Family Types 
Unattached individuals 30.2 25.5
Couples with dependent children 31.7 34.2
Income Sources
Families whose main source of income is earnings 34.6 36.7
Families whose main source of income is government transfers 60.1 57.7
Source: Federal/Provincial/Territorial Working Group on Social Development Research and Information, Construction of a Preliminary Market Basket Measure of Poverty, 1998, pp. 31-33
 ContentsNext
     
   
Last modified : 2005-01-11 top Important Notices