Printed from The Vanier Institute of the Family's website at www.vifamily.ca. © 2007.

Virtual Library
     > Contemporary Family Trends
          > Trends In Canadian Family Incomes, Expenditures, Savings And Debt

Trends In Canadian Family Incomes,
Expenditures, Savings And Debt

by Roger Sauvé,
People Patterns Consulting
(1999)

Highlights

This report examines how the financial situation of Canadian families has changed over the last several decades. Several conclusions are reached easily while others are less certain due to conceptual and data considerations. The long time span and the broad level of analysis reveals that the financial life of families is both diverse and volatile. The economy, the role of dual-incomes, the size of families and their composition provide a changing backdrop for both family life and for analysis.

This table provides several, but not all, of the highlights from the findings. Readers are encouraged to read the entire report to get a deeper understanding of the trends and see how their families compare to the averages outlined here.

Have times really changed that much, eh?

"OK, I hear what you are saying. Times are tough. But are family finances really that different today than they were just a few decades ago?"

Providing objective answers to this question is the challenge of this report.

Many interesting reports, articles and opinion pieces are released each year on the subject of family finances. The focus is usually on the short-term and on a few aspects of household finances such as the latest income numbers or changes in year-to-year spending.

This paper paints a broader and longer-term picture of how families have fared financially through much of the last half century and especially during the last quarter century. The 22 charts and 9 tables that follow attempt to capture the direction of longer-term trends and highlight any permanent or temporary breaks in these long-term trends. The emphasis is on the longer-term trends rather than the details of any particular year. The nature of this report, as a resource document, means that much of the presentation is descriptive of what happened with somewhat less emphasis on trying to uncover the causes for the changes. Further study, at a broad level, are needed to help understand the evolving inter-relationships and to attempt to offer implications of the findings.

This paper examines family finances by looking at incomes, expenditures, savings and debt (or liabilities in accounting jargon). Net worth is the difference between the accumulated savings and the debt of families … some people call this wealth. The paper reviews trends covering all families, households or the personal sector. It also looks at the experience of specific types of families or households over similar time periods. The latter helps provide some understanding of who is gaining or falling behind. For example, over the last few decades the older generations have been able to increase homeownership while the younger groups have actually lost ground.

A few conceptual questions are also tackled … What is the real savings rate? … Is current spending tied to current year income and/or to accumulated savings that can be drawn upon when needed? Are families over-extended, given that 6 financial stress indicators examined in this report are now all at or very near record levels?

As much as possible, the information presented is on a per family or per household basis. This provides individual families and others with a basis of comparison, which they can link to their own experiences. Billions of dollars are not used in this report.

Some of the information presented begins from 1951 while most covers a shorter period of time. The majority of the time series begin around 1971 and thus cover roughly one-quarter century. Data availability and comparability and/or the cost of acquiring additional or compatible data are the key determinants of the length of time actually covered by any trend.

A few simple definitions

This report focuses on how "economic families" are doing. Relevant family data from Statistics Canada is used whenever it is available. In several instances, however, useful data from the household or the personal sector are used to supplement data that is not readily available directly for families. The resulting trends are still highly indicative of what is happening for families.

According to Statistics Canada, an "economic family" is generally defined as a now-married couple, a couple living common-law, or a lone-parent of any marital status plus any never-married children living at home. In some cases, relatives are also included. Members of economic families are assumed to pool their resources. Economic families comprise about 7 of every 10 households and 9 out of every 10 Canadians live in families. Other Statistics Canada studies sometimes define families using the term "census families" with results that are somewhat different. For instance, single-parent families comprise about 8.6% of "economic families" but 14.5% of "census families".

The remaining households are comprised of unattached individuals who are people who live alone or with others with whom they are not related. No data is presented explicitly for this group.

In some cases, the study refers to a broader concept called the personal sector. The personal sector includes all households plus activities of unincorporated business and non-profit groups.

The report makes extensive use of constant dollars. The use of 1996 constant dollars for all years enables a comparison of incomes, expenditures, savings and debt in terms of what its purchasing power would have been in the year 1996. This process eliminates the impact of rising prices.

Four background trends help set change in a broader context

Before examining the real subject of this paper, it is important to set the context within which long and medium-term change has occurred in Canada. Among many possibilities, four key trends have had major impacts on overall family finances. These relate to a shift to relatively high unemployment, to a growth and peak in dual-earner families and a growth in no-earner families, to shrinking family size and to a changing composition of families. Discussions about other factors such as globalization, technological change, freer capital movement, improved communication, and changing attitudes would add further insight … these are beyond the scope of this paper.

• The Canadian unemployment rate rose gradually, but with some significant variations, from the mid-1960s to the end of the 1970s. Two recessions and ensuing recoveries since then have caused the unemployment rate to soar to modern day highs in 1983 and 1993. Except for the years 1988 and 1989, the unemployment rate has been continuously above the levels of the 1960s and 1970s. During the 1980s and 1990s, changes in family incomes, indebtedness and net worth have tended to move in conjunction with the ups and downs of the unemployment rate.

• The percentage of dual-earner families rose sharply from 32% in 1967 to 55% in 1981. This was also a period of high family income growth. Dual-income families increased more slowly over the next decade and peaked at 63 % in 1989 and have declined somewhat since then. The proportion of single-earner families continues to shrink. Families with neither spouse having earnings has risen throughout the period and reached 19% in 1996. The decline in dual-income families and the increase in families with no earnings have generally held down family incomes, when calculated in aggregate terms, in the 1980s and 1990s.

• The size of the average family peaked at four in 1966 and has declined to just over 3 by the mid-1990s. The decline has been due to fewer children, more lone-parent households, more empty-nesters and more seniors. In theory and practice, smaller families "need" less income to get by.
• The composition of families has also changed over the period. Elderly families increased their share of all families from 12.4% in 1980 to 14.7% in 1996 and lone-parent families increased from 6.5% of all families in 1980 to 8.6% in 1996. A recent study ¹ found that the changing family structure from 1984 to 1993 cut market income gains by households by about half of what would have occurred if family structures had not changed. This is so because families with lower incomes, such as the elderly and single-parents, increased their share of all families while those with higher incomes shrank as a percentage of all families. This is called a compositional impact.
• The age structure also changed dramatically as the boomers moved from being teenagers in the early 1970s to being 30 to 50 years of age near the end of the period. As will be seen, people over 45 have done much better financially than those under 45.

A few interesting relationships

The pages that follow will examine incomes, expenditure, savings and debt on a step by step basis. Before going any further, it is useful to get a feel for the overall results of the trip. As you will see, the financial life of families is not a straight and simple one.

Chart 4 provides a summary view of the quarter century from 1971 to 1996. The use of indexes allows a comparison of change for three indicators from a common starting point.

A few notable and eventful time periods stand out:

• The 1970s were marked by generally rising incomes, expenditures and net worth. This occurred even though there was a mid-1970s downturn, which was largely induced by the OPEC petroleum cartel.
• The early 1980s was a period of severe recession (some people even called it the "great" recession at the time). Unemployment rose sharply. It caused significant declines in incomes, expenditures and net worth.
• The subsequent period saw a sustained recovery in all three indicators. It stopped in 1989. By 1989, spending and net worth both reached new highs … but family incomes had not.
• A new recession hit families in the early 1990s. Again all three indicators headed down.
• Something strange has happened since the new recovery. First of all, unemployment has remained relatively high as shown in chart 1. Secondly, expenditures recovered and net worth rose strongly and both continued upwards to 1997. Net worth per household actually reached a record high in 1997. Thirdly, and not in line with tradition, family incomes did not and have not recovered. The surprise relates to the fact that incomes had improved in past recoveries. In addition, the personal savings rate has now plummeted to record lows. The personal savings rate stood at 5.4% in 1967, 12.5% in 1977, 9.1% in 1987 and a record low of 1.8% in 1997. Families seem to have boosted their spending by buying on credit supported by increasing net worth.

Some long-term interrelationships seem to have changed. Many analysts are concerned. The two-handed economist has returned. Are Canadian families in bad shape (stagnant incomes and low savings) or are they in good shape (record net worth)? Some suggest that the current situation points to another recession in the near future. Some believe that we should stop looking to family income gains as the key predictor of expenditure growth and switch to net worth as a better indicator. Others suggest that we should stop looking at the old savings rate and start looking at the change in net worth from year to year. One danger of looking at net worth is the degree to which wealth is concentrated in relatively few families … over two-thirds of all wealth is held by the richest 20% of families (see chart 20) and almost half of this wealth in held by the richest 10% of families.

Regrettably, an honest answer to this puzzle is that there are no clear-cut answers. Canadians, however, are more certain about what is happening. An ongoing survey found that roughly 85% of Canadians felt that they were financially satisfied when they were asked such a question in both 1975 and 1980 … by 1995 only 72% said they were satisfied.²

Family incomes virtually stagnant since early 1980s

The longest data series available on family incomes goes back a half-century. It measures incomes before income taxes. In 1951, the average family had a total income of about $3500 in terms of actual dollars received at that time or $22,743 in terms of constant 1996$. This family income measure increased sharply to almost $56,000 by 1980. The path since then has been flat with large swings brought on by recessions and recoveries. The early 1980s recession caused incomes to dip until 1983 while the subsequent recovery then caused it to increase to almost $59,000 by 1989. The cycle started again with the early 1990s recession causing a dip to 1994 and was followed by an increase to about $56,600 by 1996. In 1996, average family incomes remained below the 1989 peak and were only $600 above the year 1980 level.

The average income per person living in these families rose from about $6,000 in 1971, to $17,000 in 1980, to $18,900 in 1989 and down to $18,400 in 1996.

There are other widely used measures of family incomes. The next chart goes back a quarter century to 1971. The middle line shows incomes before transfers from government. The top line presents income levels after the receipt of government transfers and is the same concept as used in chart 5. The bottom line shows what is left after the payment of income taxes. The size of both transfer payments and income taxes have increased as a proportion of family incomes. In 1996, transfers were equal to about 12% of the top line while income taxes were equal to almost 21% of the top line.

Most families experienced income declines in the 1990s

Table 1 provides an overview of the income experience of various types of families since 1971. The basis of measurement is income after income taxes. The total for all families is the same series used in the bottom line in chart 6.

• All families, except male lone-parent families, had significant income gains when the quarter century is taken as a whole. The gains were the largest among elderly families. Single-earner families with children had very small income gains over the entire period.
• All family types recorded income gains during the 1970s with the largest increases being among the elderly. For most non-elderly families, the only gains in incomes over the last quarter century occurred during the 1970s.
• The 1980s continued to provide improvements for elderly families and temporary reprieve for male lone-parent families. Married couples without children experienced income declines while married couples with children and female-lone parent families had small increases.
• The 1990s have not been good for families. Both elderly and non-elderly families experienced significant income declines. The hardest hit were lone-parent families with the largest drop among those headed by males. Dual-earner married couples were just able to maintain their incomes. The growth in the number of dual-earner families flattened in the 1990s (see chart 2) and thus fewer couples chose or were able to add an additional earner to increase family incomes. For married couples, a second earner added from $13,000-$15,000 of income in 1996.
• For the majority of families, the peak income year was 1989. For married couples without children, the peak income year was 1980.
• Average income changes have varied by age groups. This is shown for the 1982 to 1996 period only. Households aged under 35 and 35-44 had very negligible growth from 1982 to 1996. In sharp contrast, older households experienced significant growth in incomes.

Income distribution moving slowly to higher income groups since 1969

The way income is distributed among households is important from a public policy perspective.

• In 1969, the richest 20% of families garnered 35.7% of all the aggregate income after income taxes. By 1996, this had increased to 37.6%. The share going to the upper-middle 20% of families also increased a little.
• The three groups of families comprising the poorest, the lower-middle and middle groups all lost some ground to the two higher income groups.
• The distribution would have been more unequal if the distribution had been based on a before transfers and before income tax basis. "In short, the Canadian government cash transfer system reduces inequality, not only between but within family types. Types with relatively low average market incomes have their income levels raised by the combined effect of income taxes and government transfers, while types with relatively high market incomes pay significantly more in income taxes than they receive in cash transfers". ³

Younger families now more likely to suffer from low-incomes

Using a consistent methodology, Statistics Canada classifies some families as being in a low-income situation. (Some people like to call this poverty – Statistics Canada warns against such usage. Even so, most of the households in this group do live in poverty however defined). In 1996, a two-person household living in a city with a population of at least 500,000 was classified as low-income if it had a total income of less than $21,414 or $10,717 per person. The proportion of low-income families declined in the 1970s but there has not been any sustained improvement since then. The percentage of families with low-income was 13.2% in 1980 and 14.5% in 1996. The percentage of children under the age of 18 living in low-incomes families rose from 15.8% in 1980 to roughly 21% in each of 1993,1995 and 1996.

The more interesting developments relate to family types and age groupings.

• Elderly families have experienced significant improvements. The low-income rate for elderly married couples declined from over 20% in the early 1980s to a low of less than 6% by 1990. Rates for all elderly families have risen recently.
• The worst year for all non-elderly families was in 1984. It is noteworthy that the rate in 1996 was just below the worst year.
• Non-elderly married couples, with or without children, were more likely to be in a low-income situation in 1996 than in 1980. This is true for both single-earner and dual-earner families. A move from a one-earner to a dual-earner family greatly reduces the risk of being in low-income. In 1996, a single-income family with children was 3.8 times more likely to have a low-income than was a dual-income family.
• The highest low-income rate is still among female lone-parent families. Almost all-female lone-parent families with no earner have low-incomes.
• Age matters greatly. The low-income rate more than doubled among families below the age of 25 and jumped by about three-quarters for those aged 25-34. Low-income among households headed by persons 65 and over fell by more than half from 1980 to 1996. In 1980, both youth and elderly households had similar low-income rates … by 1996 young households were 5 times more likely to be living in a low-income situation. This is consistent with the income gains presented in the previous section.

Fewer wages and more pensions

Wages and salaries and self-employment earnings form a shrinking share but remain the major source of family incomes.

• In 1996, wages and salaries and self-employment earnings comprised about 79% of family incomes compared to about 88% some 25 years earlier. The share coming from this source dipped most significantly during the early 1980s and early 1990s recessions. The rise in families with no earnings has contributed to this long-term decline (see chart 2).

• The second largest source of family incomes is government transfer payments, which have risen from about 6% of incomes in 1971 to about 12% in 1996. These transfers tend to rise during periods of recession. The changing distribution of transfer payments is examined in the next section.
• Other money incomes, which include such items as private pension incomes and alimony payments, are now the third largest source of incomes and have tripled from about 2% in 1971 to 6% of the total in 1996.
• Investment income has declined as a source of family incomes. This decline is largely due to the significant reduction in interest rates from the highs (near 20% for some rates) of the early 1980s to the end of the period. This decline continues even though the value of shares held by households increased sharply beginning in 1993 and 1994. (see chart 15). It is notable that about half of the investment income was earned by the richest 20% of families in each of the mid-1970s, mid-1980s and mid-1990s.

Government transfers are now a more important source of incomes

Selected government transfers have been grouped according to 4 broad categories. The data for transfers is derived from the National Income and Expenditure Accounts and refers to the personal sector as a whole.

• In 1996 dollar terms, transfer payments for the total of these 4 categories increased from about $3,700 per household in 1971 to about $7,500 in 1995. These averages include both recipients and non-recipients.
• The category where the strongest trend exists is for government sponsored pension payments. In 1971, these payments comprised about half of all transfer payments within these 4 groups and by 1995 were equal to 55% of the total. In dollars, payments rose from about $1,800 in 1971 to about $4,100 in 1995.
• The most volatile category of transfer payments is for unemployment insurance and workers’ compensation and reflects recession and growth periods in the economy and some changes in government policy. In 1983, these payments were equal to 30% of the total. In 1995, they were equal to only 19% of the total.
• Provincial government transfer payments for direct relief and other local payments have been increasing in recent years and now average about $1200 per household, which is double what they were in 1971.
• A fourth category is called "family/youth/native/disabled plus GST credits" (GST credits are included from 1993 to 1995). Transfers to this group totalled about $770 per household in 1995 and were the highest since the 1974 to 1979 period when they reached $930 per household in 1974.

Household expenditures highly sensitive to economic conditions

Consumption expenditures per household have trended upwards but have done so in a highly cyclical fashion. Consumer spending is very sensitive to the business cycle and the business cycle is not dead. The recessions of the early 1980s and early 1990s both caused average household expenditures to drop by similar percentages of just over 9%.

Expenditures per household averaged just above $34,000 in the late 1960s and then jumped quickly to reach $40,000 by 1979. The recession of the early 1980s caused expenditures to dip by 9.5%. The subsequent recovery helped establish a new peak of $41,600 by 1989. Again, the early 1990s recession caused a dip of 9%. By 1997, average expenditures per household were still 3.5% below the 1989 peak and only 0.5% above the 1979 peak. The swings in average expenditures have been more volatile than the swings in average family incomes.

Makeup of spending is changing

The changing makeup of family budgets from 1969 to 1996 is shown in table 3.

Total current consumption, as measured by the Family Expenditure in Canada survey, have shrunk from 79.5% of total household budgets in 1969 to 69.1% in 1996 or by 10.4 percentage points. The portion not allocated to current expenditures is absorbed by income taxes, investment in financial security and gifts and contributions. In 1996, income taxes were equal to 22.4% of total family budgets compared to 13.6% in 1969. Income taxes per household in real terms increased from about $5,700 in 1969 to $12,500 in 1996.

The percentage of the household budget allocated to food declined from 18.8% in 1969 to 12.2% by 1996 or by 6.6 percentage points. The portion of total spending earmarked for clothing was cut by more than half between 1969 and 1996. Other very significant declines were evident for the purchase of furniture and equipment, smoking and alcoholic beverages and medical care. Smaller declines occurred for transportation and personal care.

Significant increases occurred for shelter, household operation and recreation. The near doubling of the miscellaneous category is due to greatly increased spending on games of chance.⁴

Poor households spending a lot more on housing

Higher incomes bring about different spending patterns. The poorest 20% of households now allocate almost one-third of all expenditures directly to obtaining shelter and this is up by 8.5 percentage points since 1969. The richest 20% of households allocate about 13% to shelter, up only a bit from 1969.

Personal income taxes now comprise over 30% of the total expenditures of the richest 20% of households and are up from about 20% a quarter century earlier. The proportion allocated to taxes by the poorest 20% increased over the period but remained under 3%.

Both groups now allocate more to financial security with the richest 20% of families putting aside about 6% of their budgets compared to 1.6% for the poorest 20% of families.

Both the poorest and richest families reduced the share of expenditures going to food, furniture and equipment, clothing, medical and health care and tobacco and alcohol.

Lone-parent families also allocate more to housing

Female lone-parent families spend about one-quarter of their expenditures on housing (shelter and household operation) compared to about 16% for two-adult families, with or without children.

Food is also a bigger item for lone-parent families. The proportion of the total budget going to food declined for all three types of families in the table with the largest declines being among both types of two-adult families.

Two-adult families allocate about one-quarter of their expenditures to income taxes, which is double the proportion for female lone-parent families. All three groups are now paying more in income taxes compared to 1969.

Estimating the savings rate is not as easy as it used to be

The personal savings rate has been and still is the traditional way to measure savings by the personal sector. It measures savings as a percentage of personal disposable income, which is income after income taxes. The personal savings rate has tended to rise and fall in line with the unemployment rate. This effect is due to declining savings among those who are unemployed but increased savings by those who still have jobs but who become fearful of losing their jobs. According to a Bank of Canada study "In fact, this precautionary behaviour more than offsets any dissavings by the unemployed in the aggregate. Thus, economy wide, when unemployment rises, the savings rate tends to rise as well; when unemployment falls, so does the savings rate." ⁵

As the economy improved following the recession of the early 1990s, the traditional personal savings rate did fall (as theory would suggest). However, the decline in the personal savings rate has been much more pronounced than expected … the rate was down to less than 2% in 1997 and was the lowest on record. The 1996 rate was the third lowest on record. The record low savings rates suggests that the flatness in income growth so far in the 1990s may not have allowed households to increase savings. As such, a slowdown in the economy could cause consumers to pull back more deeply than in past recessions given that the cushion of savings is very small.

The low level of savings has generated discussions about the usefulness of the traditional measure. The CIBC Economics Department has suggested that other measures may be more appropriate.⁶ One of these measures is the annual "change in net worth" which includes changes in financial and non-financial assets and liabilities and which captures capital gains or losses from asset appreciation, from homeownership, stock holdings, etc. As with the personal savings ratio, these changes refer to the personal sector as a whole and not specifically to households.

Chart 13 plots this "change in net worth ratio" as a percentage of disposable income. Certain trends seem to be evident for this ratio. The ratio was at it highest levels in 1973, 1974 and 1980 when it averaged about 60% of disposable income. The ratio has had its two lowest levels in 1990 (15%) and 1995(18%). The ratio tended to move up in the 1960s and 1970s and down since then. This is similar to the trend in the personal savings rate. Recessions caused the ratio to drop while recoveries caused it to move up. The ratio has always been higher than the traditional personal savings rate. The ratio rose in 1996 and 1997 while the personal savings rate fell. (The accumulated net worth in chart 14 is based on the accumulated changes in this ratio.)

A new experimental measure, the "household specific savings ratio", has been developed for this report from the Family Expenditure in Canada survey. It is also expressed as a percentage of the disposable income of households. This new measure has the advantage of being the only rate that can also be calculated for various types of families or households. (see table 6).

The ratio is based on the net change in assets and liabilities when such transactions occur. This includes transactions causing net changes in bank balances, money on hand, money owed to and by the household, purchase and sale of stocks and bonds, personal property and real estate, home related expenditures, and contributions to and withdrawals from registered retirement savings plans. Potential asset gains or losses are not counted until they become part of a transaction.

The "household specific savings ratio" has followed the general trend in the personal savings rate except for the 1996 survey. The "household specific savings ratio" has been less than the traditional personal savings rate, except in 1996. It was at 2% in 1969, rose to over 7.5% in 1978, increased to 9.2% during the recession of the early 1980s, declined to near 5.4% in 1986, fell to 4.4% in 1992 (when it should have begun to rise) but it then improved sharply to 8.8% in 1996.

And so? Are people saving as much as they used to? Probably not … but it is difficult to say so with certainty. The traditional "personal savings rate" suggests that annual savings are at record low levels. The "change in net worth ratio" suggests that net worth is somewhat better even if it remains low by historical standards. The "household specific savings ratio" is at its highest level since 1982. (More research needs to be done on this new savings ratio.)

Savings rates differ by household type

In 1996, according to the "household specific savings ratio", the richest 20% of households saved over 18% of their after tax income while the poorest 20% has negative savings of 13%.

The "household specific savings ratio" for 1996 suggests that all of the types of households in table 6 were saving more or had smaller negative savings in 1996 than they did in both 1992 and 1986. Three household types were actually saving more compared to the previous records set in 1982. These record savers were households made up of two-adults with children, owners with mortgages and the richest 20% of households.

Net worth is trending upwards but is sensitive to business cycles

Aggregate measurements of net worth are estimated annually for the personal sector as well as for Canada as a whole. These annual numbers are obtained through a survey of financial institutions. They are published as the National Balance Sheet Accounts. Detailed estimates based on surveys of households are prepared infrequently with the last survey conducted for the year 1984. Some results from the latter are examined in charts 18, 19 and 20 in the next section. The next household wealth survey is scheduled for 1999 with results available during the year 2000.

Net worth represents the sum of the annual savings produced over time. The annual change in net worth was shown in chart 13.

Charts 14 to 17 summarize the trends in assets, liabilities and net worth at the aggregate level. All estimates are expressed in constant dollar terms. Ratios of liabilities to assets are shown in chart 22.

• Net worth (assets minus liabilities) has improved over time with declines associated with periods of recession and increases with periods of recovery.
• Net worth per household was equal to $148,000 in 1965 and $211,000 in 1997 or an increase of 42% over the period. Over the same period, assets (what households own or are owed) have risen by 47% while liabilities (what households owe) have increased by a much faster 76%. (As such, the ratio of total liabilities to total assets has risen over time as shown in chart 22.)

• Non-financial assets now comprise a smaller share of all assets. In 1965, both non-financial and financial assets each had a value of roughly $87,500 per household for a total of $175,000. By 1997, non-financial assets averaged $142,000 per household with financial assets valued at $116,000.
• Chart 15 reveals that life insurance and pensions have become the largest element ($46,000 per household) of financial assets compared to 1965 when they were smaller than both shares and currency and deposits. Life insurance and pensions include the liabilities of life insurance companies to policyholders, the liability of trusteed pension plans to individuals and federal government annuity liabilities.
• The value of shares per household was relatively flat from 1965 to the early 1990s. Share values jumped from $27,000 per household in 1992 to over $40,000 in 1997 or an increase of about 50%.
• Canada Savings bonds and other financial assets are a declining share of financial assets.

• Residential structures and land now comprise over three-quarters of all non-financial assets. This compares to about two-thirds in 1965.

• The market value of residential structures peaked at $54,600 per household in 1994 while the market value of land peaked in 1980 at $36,000 per household. The value of consumer durables owned by households peaked at $27,000 per household in 1979. As in other charts and tables, these estimates are measured in constant 1996 dollars.
• Total liabilities increased from $27,000 per household in 1965 to over $47,000 in 1997.
• As shown in chart 17, mortgages per household have doubled over the period. In 1997, mortgages per household rose to over $31,000 compared to under $15,000 in 1965. On average, Canadian households now hold about $10,500 in consumer credit which is the same as during the late 1970s but higher than the average of $7,000 in the mid-1960s. Consumer credit covers credit extended to persons for the purchase of consumer goods and services. Chart 22 indicates that the ratio of consumer credit to the value of consumer durables reached a new high in 1997.

An old but useful measure of net worth

Three old Statistics Canada surveys, covering the years 1970, 1977 and 1984, collected wealth data directly from households. The household surveys did not collect information on assets relating to pensions. As indicated earlier, the next household net worth survey will not be available until the year 2000. These old surveys provide some idea of the distribution of wealth by age, income and other characteristics.

Chart 18 compares estimates of net worth from the three household surveys with the annual survey from the National Balance Sheet Accounts. In 1984, net worth was equal to about $125,000 per household (in 1996$) according to the household survey compared to a National Balance Sheet estimate of $150,000 excluding pensions and about $175,000 including pensions. (The last bar is the same series as that shown in the middle line in chart 14).

Net worth peaks for households age 55 to 64

The age distribution of assets, debt and net worth is shown in the next chart taken from the 1984 household survey. In that year, net worth and assets per household peaked at over $200,000 (in 1996$) for households headed by persons aged 55-64. In contrast, total debt peaked at $31,000 for households aged 35-44.

Net worth is very unevenly distributed

The household survey enables an analysis of the distribution of wealth.

• In both 1970 and 1984, the richest 20% of families held about two-thirds of all of the net worth with a small decline between both years. (Follow-up studies suggest that high-income families tend to under-report their net worth). The poorest 20% of families had more debt than assets in 1970 and just broke even in 1984.
• In 1984, the lower-middle, middle and upper-middle 20% of families held 4.3%, 10.7% and 19.7% of the net worth respectively.

Are households over extended?

Section 17 indicated that liabilities and debt have grown much faster than assets. Several indicators suggest that the degree of financial stress is increasing for Canadian households. A July 1998 survey by Angus Reid found that 54% of Canadians were finding it harder to make ends meet compared to 38% who thought this was so in 1989.⁷

• For many years, a standard measure of financial stress has been total debt/liabilities as a percentage of disposable income. This measure is a proxy of a householder’s ability to meet credit payments. In the mid-1960s, total liabilities were equal to about 72% of disposable income. This ratio rose during the early 1970s, declined in the mid-1970s and then peaked at 88% in 1979. The recession of the early 1980s caused households and lenders to firm up their financial situations and by 1984 the ratio was back to the record lows of the mid-1960s. Since then, the ratio has shot straight up. It set new record highs beginning in 1988, in1993 it passed the 100% mark and in 1997 it hit 114%.

• Chart 22 provides alternate measures of financial stress. The top three measures all reached record levels in 1997. Consumer credit as a percentage of the value of consumer durables reached 47%. Mortgages reached 35% of the value of residential structures and land. Total debt/liabilities as a percentage of "liquid" financial assets increased to 22%. The ratio of total debt/liabilities to total assets at just over 18% was very near but not at the record set in 1990. A correction in the stock market from 1997 levels could easily push this ratio to record levels… this may have happened in 1998. As seen earlier (chart 15), much of the increase in total assets has been in stocks and pensions, which are highly dependent on stock prices.
• Still another measure points to increasing amounts of family budgets going to pay interest. In 1969, mortgage interest plus interest on consumer credit was equal to 3% of after tax expenditures compared to 5.6% in 1996.⁸

The latest situation seems to be precarious and could worsen any slowdown in the economy. Section 11 indicated that household spending drops sharply during recessions. Higher debt loads could may be the next decline even larger.

Whether or not the current debt levels are at crisis levels may depend on who holds the debt. The 1984 household survey found that debt, as a percentage of assets were highest among younger households. At that time, debt as a percentage of assets was 37% for households under 25 years of age, 29% for those aged 35-44, 10% for those aged 45-54, 5% for those aged 55-54 and less than 2% for households 65 and over.⁹ The weaker income trends among younger families suggests that an even greater share of the total debt may now be borne by those under the age of 45.

Facilities ownership another measure of assets

Another way to estimate what households have in assets is to look at what kinds of homes they occupy and what items are included as part of these households. Section 21 indicates that an increasing level of debt is being used to acquire these assets.

Table 7 clearly shows that Canadians now have more of almost everything. Tables 8 and 9 suggest that not all households have shared equally in the acquisition of assets. This is especially so for younger households

Several trends are evident from 1977 to 1997 for all households and from 1982 for various types of households.

• The size of homes has increased by about half a room. The fastest growth between 1982 and 1997 occurred among older households and couple families.
• Home ownership has remained flat. Even so, since 1982, ownership has increased among households aged 55 and over but declined for all other age groups. Ownership increased for families without children but fell for those with children and especially for lone-parent females.
• The percentage of homes in need of major repair has fallen significantly.
• The ownership rate for all of the selected items has increased with the only exception being automobiles.
• Almost all households now have refrigerators, colour televisions, telephones and smoke detectors (but only half have a portable fire extinguisher).
• Over 8 out of 10 households have microwave ovens, videocassette recorders, vehicles, and 2 or more radios. The percentage owning automobiles has declined and has been replaced by a growing number of vans or trucks. The ownership of vehicles has declined among households under the age of 55 and increased for those above that age.
• Washers and dryers are in over three-quarters of homes.
• About half or more households now own compact disk players, freezers, gas barbeques, portable fire extinguishers, and automatic dishwashers.
• About one-third of homes already have a home computer and about 30% have air conditioning. Less than 20% have a camcorder.

Concluding comments

This report is intended to be a foundational document that lays out the major family financial trends over the last several decades and up to the present. Hopefully, it will be seen and used as an objective background document for policy discussions. Policies aimed directly at families or which have indirect impacts on families should be based on the changing reality in all its complexities.

The table on page 1 provided summary highlights by topic and time period.

A few key trends and issues stand out.

• From a financial perspective, the 1980s and 1990s have become more difficult for most types of Canadian families.
• The 1950s, the 1960s, and the 1970s were a period of rapid growth in incomes and sharp declines in low-income rates. Average family incomes jumped from about $23,000 in 1951 to $56,000 in 1980. All family types had income gains in the 1970s.
• The long-term trend of strong family income growth came to an abrupt end in the early 1980s. The early 1980s and early 1990s were both marked by severe recessions that kept unemployment rates relatively high. Family incomes in 1996 were only $600 above their level in 1980. This flat income picture coincided with little improvement in low-income measures and lower savings rates.
• Household spending has been more volatile than incomes. Sharp expenditure declines occurred during the last 2 recessions. Expenditures are rising once again but these seem to be increasingly financed through lower savings and more borrowing against an increasing net worth. The increase in debt has been so significant that several indicators of financial stress are now at record high levels. Whether or not these debt loads are at crisis levels depends on their distribution. Incomplete data suggests that younger families may now be holding a greater portion of the debt relative to their incomes.
• Spending patterns have been changing over the long-term. Budgets are now much less heavily weighted to food, furniture and equipment, smoking and alcohol and medical care. The share of the budget allocated to clothing has been cut in half. Income taxes now take up a much larger share of family budgets that they did 3 decades ago. Shelter costs are also up.
• Households now own more of almost everything. Even so, home ownership is down for households aged under 35, 35-44 and 45-54 and up for those aged 55 and up.
• The changing distribution of incomes, expenditure, savings and debt remains a significant issue. The distribution question goes far beyond a simple comparison of the rich and the poor but also addresses age differences and different types of families. The long-term trends suggest that the rich (the top 40%) have indeed been getting a slightly bigger piece of the pie relative to all the groups below them on the income scale. The longer-term trends also suggest that young families have been losing out. Single-earner families and single-parent families have lost the most ground in the 1990s.
• Certain issues need more analysis. More work needs to be done on savings, on the distribution of savings, debt and assets and what constitutes low-income or poverty.

It is a fact of life that not all change happens at the same pace. Some changes take place quickly. Some changes reflect cyclical movements in the economy. Some changes take years to become evident and even longer to become accepted by Canadians. This reality means that certain issues tend to be analysed and debated for years before firm conclusions are reached. Actions to alter undesirable directions of change may take even longer. This report will help identify certain areas where action is needed.

This report should be viewed as another step in trying to understand the dynamics of family finances. Regular updates are needed to ensure that significant changes are captured early and that emerging issues can be examined and discussed before they become even more difficult to handle. Most importantly, trends and forecasts of family finances should be placed in the context of a changing demographic makeup of Canadian families some 10 to 20 years into the future.

Families are worth it.

The Vanier Institute of the Family
94 Centrepointe Drive
Ottawa, Ontario, Canada, K2G 6B1
Tel: 613-228-8500 Fax: 613-228-8007
E-mail: