Poverty in Canada (2nd Edition)

Table U1-1

The updated comparison between "poverty" lines and "poverty" rates for the family of four in 1993 demonstrates again the degree of exaggeration that results when high, relative lines are used. The LICO and CCSD lines are roughly twice the basic needs line, and the poverty rate about four to five times higher. This order of magnitude is approximately the same as was the case in 1988.

Click here to view Table U1-1:Comparative Poverty Lines and Poverty Rates, Family of Four, 1993

Table U1-2

The book presents the LICO and CCSD lines for the years 1988, 1989, 1990, and 1991. The LICO lines were constructed using the 1978 base. In 1986, Statscan rebased LICO and the lines presented here for 1992 to 1995 use the 1986 base. The sharp rise in the cut-off line for single persons is noteworthy. In the late 1980s and early 1990s, LICO (for large cities) and CCSD lines for singles were almost identical. With the recent rebasing, the LICO line for singles is now about $2,000 above the CCSD line.

Click here to view Table U1-2: Canadian "Poverty Lines," 1992-1995

Table U5-1

The list of nutritious foods presented in this table is somewhat larger than the original list. Several additions and some minor refinements were made in order to incorporate the changes in the latest Canada Food Guide (1992). Food prices are those prevailing in early October 1994. The cost per serving is little changed in the meat, vegetable, and bread groups, but clearly higher for the dairy and fruit groups.

Click here to view Table U5-1: Food Cost Chart, 1994

Table U5-6

Substantial improvements have been made to this table. A much larger list of foods (23 versus 13 in the original) are included in the comparison basket. This enhances the reliability of the comparison of costs in our major cities. Overall, in 1994 as in 1988, food costs are somewhat higher in the East and lower in the West, albeit with some notable exceptions.

Click here to view Table U5-6: Comparative Food Prices for Various Canadian Citiesin 1994

Table U5-7

The family composition in this table incorporates changes in Health Canada's recommended nutrient intake (energy) and is precise about the children's ages. The selection of twelve year old males attempts to capture a rough average in terms of cost. The child is still young enough that there is no expectation of any monetary contribution to the household. Food costs, for the Ontario "base" in 1994, come in at about 20 percent higher than in 1988, on average.

Click here to view Table U5-7: Essential Food Costs by Family Size for Ontario, October 1994

Table U5-8

As with the original, this table constructs essential annual food costs for Canada's major cities by comparing the Ontario base costs to the costs of the comparison basket in the other cities. The third column, the comparison basket ratio, shows that in 1994, food costs in other cities were the same or higher than the Ontario "base." None were lower.

Click here to view Table U5-8: Essential Annual Food Costs, 1994 ($), by Family Size and Major Metropolitan Area

Table U6-3

Average monthly rents in the list of cities surveyed by the CMHC (those with a metropolitan population of 50,000 or more) are displayed by bedroom type. Overall, rents in 1994 are higher than those in 1988. In some major cities, such as Toronto and Vancouver, rents are about 30 percent higher than they were six years earlier. In most others, the increase has been more modest. In at least one major centre, Montreal, rents are down somewhat from 1988.

Click here to view Table U6-3: Average Monthly Rents of Privately Initiated Apartment Structures of Six Units and Over, by Bedroom Type and Urban Area, October 1994

Table U6-4

As with the original, the key assumption in constructing essential annual shelter costs is that poor people are renters and are drawing from the bottom half of the market for rental accommodation. This assumption will not be valid for every single low-income person at any point in time but, over time, is a reasonable generalization. As long as vacancy rates are above 2 percent, CMHC argues that there exists sufficient choice for all households. Evidence on vacancy rates in 1994 is displayed in a later update table. Low-income households drawing from the bottom half of the market would pay approximately 10 percent below average rents. This gives us the shelter costs in the table. As with the previous table, costs are generally up, but the variation across cities is quite substantial.

Click here to view Table U6-4: Essential Annual Shelter Costs ($), 1994 by Major Cityand Family Size

Table U7-1

Combining essential food costs with essential shelter costs and adding in "other" costs, we get the basic needs poverty lines by family size in Canada's major cities. The calculation of "other" costs was improved considerably using 1992 data for my Critical Issues Bulletin entitled "Poverty in Canada—1994" (Fraser Institute). I have simply updated those costs to 1994 using the prevailing C.P.I. The list of cities for which poverty lines are calculated is larger than was the case in the original. It includes those smaller metropolitan centres in the second half of the table (Barrie to Sydney). Since these centres are not part of the Statscan food cost survey, food costs are assumed to be the provincial average. There is much less variation in food costs within a province than there is between provinces. This assumption will give us a reasonable and useful approximation of food costs in unsurveyed communities.

Click here to view Table U7-1: Poverty Lines ($), 1994, by Major City and Family Size

Table U7-2

It is useful to construct poverty lines for each of the provinces, partly because of our interest in regional disparities and partly because social assistance, which is the last resort program of income maintenance for the poor, is a provincial matter. The lines in this table are constructed as a weighted average of the poverty lines of the cities in the previous table. The methodology is broadly the same as was used in constructing table 7-2. In addition to the provincial lines, I have calculated the poverty lines for Canada by family size at the bottom of the table. The all-Canada lines are a weighted average (with provincial populations as the weights) of the various provincial lines. As with the original table, the reader should note that the lines for the provinces of Newfoundland, P.E.I., and Manitoba are biased on the high side because each of those provinces has only one major city for which there is information on shelter costs. It is almost a certainty that shelter costs in smaller communities within each of these provinces would be substantially lower and that this would normally mean lower poverty lines for those areas.

Click here to view Table U7-2: Poverty Lines in Canada ($), 1994, by Family Size

Table U7-4

The most recent microdata file of economic families available at this time was that for 1993. Therefore, poverty lines for 1993 were calculated using the same methodology described above. Applying the basic needs poverty lines to the microdata file enables us to determine the extent of poverty in each of the provinces for 1993. The results show that while there were about 50,000 more people who lived in households with reported income below the poverty line, the poverty rate is the same as it was in 1988. Tracking poverty rates since 1988 using this same methodology, I can say that the rate fell somewhat, in 1990 to 3.7, then rose in the aftermath of the recession to 4.0 in 1992 before moving back down to 3.8 in 1993.

In terms of provincial patterns of poverty, Quebec continues to have the lowest rate of poverty, and Newfoundland the highest. In terms of household size, unattached individuals continue to have, by far, the highest rates of poverty. In 1988 and 1990, families of four had the lowest poverty rates. Now, in 1993, families of two have the lowest rates. This may be due to the increasing number and proportion of elderly couples in Canadian society. The social security pensions they receive universally put them above the poverty line.

Click here to view Table U7-4: Estimating the Number of Poor and Poverty Rates by Province and Family Size, 1993

Table U7-5

This table provides a useful bottom-line comparison between prevailing relative approaches to measuring poverty and my own basic needs approach. As with the comparison using just the family of four (table U1-1), the relative lines, LICO and CCSD, result in estimates of poverty in Canada that are four to five times higher than with the basic needs lines. This is roughly comparable to the results for 1988, although the estimates using LICO, because of the rebasing and sharp upward revision of the lines, are now about four times rather than three times higher than those using the basic needs approach.

There is a small difference between the estimate of poverty in this table using the basic needs lines and those in the previous table. This table employs the all-Canada poverty lines (by family size) whereas table U7-4 uses the separate poverty lines for each province. The weighted average process results in slightly higher poverty lines and an estimate of poverty that is about 3 percent higher than with the more precise provincial lines.

Click here to view Table U7-5: Comparison of Poverty Rates and Number of Poor, 1993

Table U7-7

Sarlo poverty lines for 1992, 1993, and 1994 have been determined using the methodology described above and are presented in this table. 1994 is the latest year for which relevant data are available. The 1995 poverty lines are forecast using an inflation factor of 2 percent. This, combined with the original table, gives the reader a set of basic needs poverty lines by province and family size for the 8 years from 1988 to 1995. The lines for 1988, 1993, and 1994 have been calculated using the basic needs method outlined above. The other lines have been estimated using actual or forecasted inflation rates.

Click here to view Table U7-7: Povety Lines in Canada by Province and Family Size ($)

Poverty profiles: tables U8-2 to U8-7

The poverty profiles using 1993 data are quite similar to those using the 1988 information. For example, the poverty rates for young people are well above those of any other age group. As table U8-2 shows, poverty is, to a very large extent, a problem of youth. For family heads, the rate declines precipitously after age 24 and continues to decline steadily to old age with the exception of a small uptick in middle age. Except for the very high youth rate, the pattern for unattached individuals is less clear. The poverty rate is in the 6 to 11 percent range for singles between 25 and 64. Only after that does the rate fall sharply to about 2.5 percent in the later sixties and to below 1 percent after age 70. This age pattern of poverty, for both families and unattached individuals, conforms quite closely to the 1988 results.

Click here to view Table U8-2: Poverty Rates by Age, 1993

Similarly, there are no real surprises in table U8-3. As was the case in 1988, the poverty rate for family heads declines steadily as educational attainment increases. Of particular interest, however, is that the poverty rate for families whose heads have a university degree has crept up considerably to the point where, in 1993, there were twice as many families [in that category] with below-poverty incomes as there were in 1988. Curiously, both the number and the percentage of unattached individuals with university degrees living in poverty has declined sharply. While the relationship between poverty and educational attainment is weaker for singles than for families, singles with a university education have, by far, the lowest poverty rates.

Click here to view Table U8-3: Education profile of the poor, 1993 Families and Unattached Individuals

In table U8-4, we see that, once again, the group with the lowest poverty rates is immigrants who arrived in Canada prior to 1980. In this table, I have subdivided this group into two: those who immigrated before 1966 and those who arrived between 1966 and 1980. The first group has poverty rates of 2.27 and 2.75 for families and unattached individuals respectively. The poverty rates for Canadian-born persons is fully twice and three times (respectively) those rates. This is difficult to explain given the background of many of the older immigrants to Canada, i.e., unskilled, different language, visible minority, etc. Even the group of immigrants arriving between 1966 and 1980 have lower poverty rates than those of people born in Canada. Only the most recent immigrant group (post-1980) have higher rates of poverty—and they are substantially higher. These results are consistent with the findings in the original table.

Click here to view Table U8-4: Immigration Profile of the Poor, 1993 Families and Unattached Individuals

The results in table U8-5 are also very close to the 1988 table. Employed persons, whether family heads or unattached individuals, have the lowest poverty rates. The rates of poverty for the unemployed are quite a bit higher. While the unemployment rate in Canada in 1993 was about 10 percent, the unemployed formed about 18 percent of the poor. Those not in the labour force had lower poverty rates than the unemployed, but, due to their large number, they comprise fully half of the poor. It is worthy of note that while the number of working poor singles in 1993 was about the same as in 1988, the number of working poor family heads increased by over 30,000 and their poverty rate increased by 48 percent.

Click here to view Table U8-5: Employment Profile of the Poor, 1993

We see from table U8-6 that husband-and-wife families continue to be the dominant pattern in Canada. They made up 84 percent of all families in 1993, down slightly from 86 percent in 1988. The poverty rate for husband-and-wife families was a meagre 2.77 percent. However, because of their large numbers, they comprised fully half of all poor families. The percent of single parent families was up from 1988, increasing from 9.83 percent to 11.96 percent of all families, and the number of poor, single parent families increased by 28,000. Female single parent families living below the poverty line—a virtual cliché of poverty in our time—made up about 41 percent of all poor families in 1993, down slightly from 1988.

Click here to view Table U8-6: Family Characteristics Profile of the Poor, 1993

Finally, a new table has been added to this update to more formally present the evidence on child poverty in 1993. In table U8-7, children living in families whose reported incomes are below the basic needs poverty lines are subdivided into four age groups. We see that, overall, there were about 313,000 poor children below the age of 18 in Canada in 1993. Of those, about 287,000 were below the age of 16. This compares with about 225,000 in 1988. This result is consistent with the general finding that families have been hit rather hard by the recent recession. Almost half of poor children in 1993 were age 7 or less. Overall, the child poverty rate was 4.6 percent, somewhat higher than the poverty rate for the general population.

Click here to view Table U8-7: Child Poverty, 1993

Welfare recipients: tables U10-1 to U10-6

An important use of basic needs poverty lines is as a basis of comparison to assess the adequacy of social assistance benefits. The purpose of social assistance is to provide, as a last resort, "income to meet the cost of basic requirements" for people when all other resources have been exhausted. Tables U10-1 to U10-6 update provincial welfare rates for 1994 and show how the incomes of recipients compare to basic needs poverty lines.

The tables are self explanatory and, for the most part, confirm the results of the 1988 comparison. That is, those classified as unemployable receive, in virtually all cases, enough from welfare and tax credits/benefits to be at or above the poverty line. In contrast, those classified as single employables find themselves below the poverty line in half of Canada's provinces. In the case of families with employable heads, only in Newfoundland, New Brunswick, and British Columbia do recipients end up with less income than the poverty line. Positive poverty gaps show the extent to which the household receives more than is required to cover basic needs. These tables confirm that the reputation of the province of Ontario as the most generous provider of welfare benefits is well deserved. Ontario, given the rate increases in the early 1990s, had by far the largest positive poverty gaps in every category. This continued until October of 1995 at which time most welfare rates were cut by 21 percent. In 1988, Ontario was the most generous in only two of the six major categories.

Click here to view Table U10-1: Welfare recipients—Income and poverty gap calculations, 1994: Single employable

Click here to view Table U10-2: Welfare recipients—Income and poverty gap calculations, 1994: Employable couple, 2 children

Click here to view Table U10-3: Welfare recipients—Income and poverty gap calculations, 1994: Unemployable couple

Click here to view Table U10-4: Welfare recipients—Income and poverty gap calculations, 1994: Single parent and 1 child

Click here to view Table U10-5: Welfare recipients—Income and poverty gap calculations, 1994: Single parent and 2 children

Click here to view Table U10-6: Welfare recipients—Income and poverty gap calculations, 1994: Single parent and 3 children

Table U10-8

I have already commented on the question of changes in the affordability of rental accommodation since 1988. Rents are up by about 30 percent or about 5 percent per year in Canada's most expensive cities. For the most part, however, rental increases have been far more moderate, and, in some cases (such as Montreal and Regina) rents have not increased at all. The information on vacancies in Canadian cities presented in this table is perfectly consistent with the trends in rents. In sharp contrast to 1988, where over half of our urban population lived in communities with "tight" rental markets—defined as vacancy rates below 2 percent—in 1994, only 36 percent did. Only 7 out of 36 urban areas had tight rental markets in which the renter might experience some difficulty or delay in obtaining the preferred accommodation.

Click here to view Table U10-8: Average Vacancy Rates in Apartment Structures withSix or More Units, 1994

Table U10-9

Since 1988, minimum wages have increased most sharply in Ontario, British Columbia, and Quebec. This table shows the annual income in 1994 of a person working full time on minimum wage in each of our ten provinces. If we compare these values to the poverty lines in table U7-2, it is clear that, except in Ontario, a minimum wage job cannot support more than two people at the level of basic needs. For most families, a single minimum wage job will mean poverty.

Click here to view Table U10-9: Hourly Minimum Wages and Annual Income by Province, 1994

Table UA-1

The distribution of income among households in Canada as evidenced by the distribution by quintiles in this table has changed very little over the past forty years. The 4-40 rule is still useful, i.e., roughly 4 percent of income going to the bottom quintile and roughly 40 percent to the top.

Click here to view Table UA-1: Distribution of Total Canadian Income by Household Quintile Shares, 1951-1993