Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Changes in family wealth
By Raj K. Chawla
Full article in PDF format
- Debt mounts until about age 40, then declines
- Families in their 20s
- Families in their 30s
- Families in their 40s
- Families in their 50s
- Families in their 60s
- Families in their 70s
- Families in their 80s
The Canadian economy performed well between 1999 and 2005. Buoyed by rising incomes coupled with stable inflation and low interest rates, Canadians went on a spending spree.1 However, much of the increased spending was financed through credit, as the personal savings rate fell from 4.0% to 1.6% and per capita debt climbed to $28,400 in 2005. Did this additional debt support increased consumption or was it invested in appreciating assets?
Using the Survey of Financial Security, this paper compares family assets and debts in 2005 with the situation in 1999. The survey collected data on 18 financial assets, ranging from the risk-free (bank accounts and term deposits, Canada Savings Bonds) to riskier investments in stocks and mutual funds—whether tax-sheltered like RRSPs or not.2
Families are divided into seven cohorts, based on the year of birth of the major income recipient (MIR), ranging from those in their 20s in 2005 to those 80 and over (see Data source and definitions). These cohorts are matched back to major income recipients from the same birth cohorts surveyed in 1999. For example, those aged 22 to 30 in 2005 correspond with 16- to 24-year-olds in 1999. These seven groups are not true cohorts since they consist of 'similar' individuals at two points in time. Nevertheless, they provide an intuitive look at the accumulation of assets and debts across the life cycle.3
The groups together paint a portrait of the typical family as it passes along the life course: finishing their education and leaving the parental home (20s); launching their careers and starting new families (30s); amassing assets and raising the next generation (40s); paying off major debts and beginning retirement planning (50s); winding down careers and easing into retirement (60s); downsizing and drawing on savings (70s); and, finally, managing assets as the end of life approaches (80s).
It is important to remember that this approach approximates how the assets and debts of a demographic cohort progessed over 6 years, as opposed to comparing groups of the same age at different points in time.
Although the primary focus of the cohort analysis is the accumulation of wealth, it was the sharp increase in debt from 1999 to 2005 that motivated this study. Thus it begins with a look at the ebb and flow of debt across the life cycle (see Family cohorts).
The rate of indebtedness is largely a function of life-cycle stage. Young families typically start with low incomes and high expenses related to establishing a home and raising children. The imbalance is resolved by home mortgages and other forms of credit. As incomes increase over time and financial needs drop, families not only pay down their debt but also begin to invest. Indebtedness peaks at over 80% by the time the MIR is 40 and then slides below 20% after retirement (Chart A). On the other hand, the proportion of families with investment income increases steadily, from 15% for the youngest group in 1999 to 77% for the oldest in 2005.4
Although life-cycle patterns explain much of the asset and debt picture, economic trends are also important. Historically low interest rates at the beginning of the 2000s facilitated borrowing—the overall debt-to-income ratio jumped from 1.02 in 1999 to 1.21 in 2005—as average debt jumped by almost a third, from $62,700 to $82,500, while average family income increased only about 10%, from $61,600 to $68,100 (Chart B). Only families with a major income recipient in their 70s reduced their average debt load. Most of the new debt went into the booming housing market, fuelled by low interest rates, low down payment options and a strong labour market. Still, other types of debt grew in lock-step so that the overall distribution changed little.
Even though more families were indebted and carrying larger financial liabilities in 2005, they were still wealthier—mean wealth holdings rose from $281,000 in 1999 to $380,700. Half of this additional wealth was non-financial— home equity, vehicles, other real estate, home contents, and valuables and collectibles. The other half consisted of savings in employer pension plans, business equity and net financial assets.
Families in the youngest cohort represented 6% of all families in 1999 (Table 1). Families in this (and the next) cohort are at the stage of family formation and expansion, home purchase, and asset building. 'Family' is used in the broad sense, since as many MIRs remained single as were married by 2005. Not surprisingly, because accumulation takes time, young families have the lowest holdings of financial assets. By 2005, they had raised their share of total wealth from 0.5% to 2.0%. Their mean holdings nearly doubled (from $25,100 to $49,600), almost entirely due to changes in the rates of asset ownership and debts owed. About half of this cohort's wealth gain came from home equity and contents and savings in an employer pension plan; another third came from net financial assets. For instance, 26% owned a home in 2005 compared with just 8% in 1999, while the respective proportions in mortgaged homes were 95% and 76%. Even this early in their careers, 42% had RRSPs and 33% had employer pension plans in 2005 compared with 21% and 13%, respectively, in 1999. On the other hand, seven in ten had outstanding loans (education, vehicle or other) or credit card balances owing.
The composition of total financial assets changed the most (44 percentage points) for these young families. From having 55% of their assets in bank accounts and term deposits and 17% in RRSPs in 1999, they had switched to 50% in RRSPs and just 22% in accounts and term deposits by 2005.
This cohort represented 19% of all families in 1999. Even though these families took on more additional debt (41%) than any other cohort, their mean wealth almost doubled—from $94,600 to $180,100. One-half of this increase came from home equity alone, followed by increases of 15% in employer pension plans and 12% in net financial assets. These three components accounted for nearly 80% of the increase in this cohort's wealth.
By this stage, the majority were two-spouse families with children. More of them had a home with a mortgage, raising their rate of homeownership from 41% to 62%. And, to provide for their children's postsecondary education, the proportion with RESPs jumped from 7% to 21%, compared with much smaller increases for savings in an employer pension plan (from 38% to 49%) or an RRSP (from 57% to 63%) (Table 2).Despite the substantial gains for this cohort, their share of total wealth increased only modestly—from 6% to 8%.
This cohort consisted of the latter half of the baby boomers. They were in their peak income years and represented 25% of all families in 1999. Even though they took 29% of the additional household credit, they improved their share of total wealth from 18% in 1999 to 27% by 2005—the largest gain in wealth share of any cohort.
A little over half of these families were couples with children and/or other relatives and another one-fifth were unattached individuals. Not only did the incidence of homeownership among families in this cohort rise between 1999 and 2005 (from 63% to 74%), their holdings of other real estate also increased (from 15% to 21%). In fact, they had the highest change in the rate of ownership of other real estate. Although the proportion with RRSPs remained unchanged (65%), the proportion with RESPs more than doubled—from 10% to 22%. And their employer pension plan participation rose from 47% to 52%.
These late boomers had the largest increase in wealth, more than doubling their holdings from $210,800 in 1999 to $456,800 by 2005. Equity in a family home and other real estate accounted for almost one-half of this gain and business equity for another one-fifth. The remainder came from employer pension plans and net financial assets (Table 3).
The older baby boomers, within sight of retirement, accounted for 20% of all families in 1999. Their share of total wealth increased modestly—from 26% in 1999 to 28% in 2005—all because of the amounts of assets and debts. Since many in this cohort had become 'empty-nesters'—the proportion of two-spouse families with children dropped from 30% to 10%—they likely had more money to invest or pay off debts.5 Homeownership rose marginally from 71.1% to 75.9% as did the proportions of those with RRSPs (from 66.8% to 69.2%) or employer pension plans (from 53% to 56%).
These early boomers increased their wealth holdings by $196,300, bringing the amount to $569,900 in 2005. Accrued savings in employer pension plans alone accounted for 43% of the gain, followed by 41% for equity in home or business. Net financial assets accounted for a meagre 9% of the gain.
Families in this cohort were transitioning into retirement. In 1999, more than half of them (55%) had employment earnings as the major source of income compared with less than one-third (32%) in 2005. Overall, they represented 13% of families in 1999, comprising largely couples and unattached individuals. Their share of wealth fell from 24% in 1999 to 20% by 2005—not because their wealth declined, but because the wealth of other cohorts increased more.
Three-quarters of these families lived in an owned home and a little over half had savings in employer pension plans. Not too surprisingly, the proportion holding RRSPs fell by 11 percentage points (from 66% to 55%), counterbalanced by a similar increase in the proportion holding RRIFs (in 2005, it was still mandatory to convert funds held in RRSPs into RRIFs by age 69). Also, the proportion owning a business fell from 21% to 13% and other real estate from 26% to 21%. Apparently some families reaching their 60s opted to wrap up or sell their business or investment properties (if not transferred to the next generation) and convert the proceeds into financial or other assets.
As might be expected, these families had the highest mean wealth—$581,900 in 2005 compared with $514,600 in 1999. An increase in home equity alone accounted for 59% of this gain, followed by 30% for employer pension plans and 29% for net financial assets. As business ownership dropped, so did the contribution of business equity.
These elderly unattached individuals and couples constituted 11% of all families in 1999. They were mostly retired, with government transfers and retirement income as their major sources of income (see Families dependent on government transfers). Between 1999 and 2005, their share of total wealth fell from 16% to 11% as their numbers dropped because of deaths and they began to use their savings to fund consumption. The proportions owning real estate, a business, vehicles, or RRSPs fell, whereas the proportions holding RRIFs or employer pension plans rose.6 Between 1999 and 2005, these families increased their mean wealth by only $69,700 (to $484,200) with 57% coming from home equity alone, 25% from financial assets and 21% from employer pension plans.
Since mortality is a significant factor in the number and size of families in their 70s and 80s, it is important to note that the population in these cohorts is becoming less comparable at the beginning and end of the period. Therefore, the increases in wealth observed towards the end of the life course in these artificial cohorts may be due to unequal probabilities of death across the wealth distribution. True longitudinal data would be required to determine whether wealth typically increases or declines towards the end of the life course.
In this age cohort, unattached individuals outnumbered couples. They dropped from 8% of all families in 1999 to only 4% by 2005. Not surprisingly then, their share of total wealth fell from 9% to 4%. The proportion of homeowners remained unchanged at 61%, but the proportion of those with an employer pension plan rose from 44% to 57% (this apparent anomaly may arise because an elderly major income recipient is living with a younger spouse or other relative). Between 1999 and 2005, mean wealth for these families increased from $323,800 to $389,500. Similar increases in employer pension plans and home equity accounted for most of the gain.
Between 1999 and 2005, Canadian families took on $215 billion of additional debt while increasing their wealth by $1,386 billion. Most of this additional wealth consisted of non-financial assets like a home, other real estate, vehicles and contents of a residence, and the actuarial value of employer pension plans. Since most of the additional debt was in mortgages, many families may have acquired assets using leverage.
Not all cohorts of families gained equally (see Changes in wealth distribution). Nearly half of the additional household wealth from 1999 to 2005 was accumulated by baby boomers in their 40s and almost another third by those in their 50s. The gain for the former consisted of increased equity in a home, other real estate, or a business, and financial assets, whereas for the latter it came from home equity and employer pension plans. Families in their 20s and 30s improved their net worth by way of homeownership and other financial assets.
A home remained a major asset for Canadian families and its equity the largest component of wealth for most. In fact, by 2005, home equity and employer pension plans constituted over one-half of total wealth for families in their 50s, 60s, 70s or 80s. Rising real estate values pushed up home equity, and the appreciation in home value, as a proportion of home equity increased in importance for older owners.
RRSPs remained the major financial asset for families from their 20s to their 60s and RRIFs for those in their 70s and 80s. Although more families in their 30s and 40s with children contributed to RESPs, amounts paled in comparison with RRSP holdings. On the other hand, the proportions of families investing in riskier assets like stocks and mutual funds outside of registered plans dropped for most cohorts between 1999 and 2005, as did the amounts in these holdings.
Overall, neither the shape of the wealth distribution nor inequality changed between 1999 and 2005. Nonetheless, general economic prosperity and rising real estate values resulted in 461,000 more families worth one million dollars or more—bringing the total to 1.1 million by 2005. On the other hand, 134,000 fewer families were totally dependent on government transfers.
The analysis is based on the Survey of Financial Security (SFS) for the years 1999 and 2005. The survey collected information on family demographics, assets and debts at the time of the survey, and income during the preceding calendar year. It covered private households in the 10 provinces. Excluded were persons living on Indian reserves, members of the armed forces, and those living in institutions such as prisons, hospitals, and homes for seniors.
Each year used a regular area sample supplemented by a small sample of 'high income' households in order to improve wealth estimates at the upper end of the income distribution. Financial data were sought from the family member most knowledgeable about the family's finances. Although the sample size of the 2005 SFS was about one-third of that in 1999, the surveys were otherwise identical. This simplifies not only the comparability of wealth by components, but also measurements of change over time. Nonetheless, two adjustments were made to the 1999 data: first, the sample was re-weighted following the procedure used for the 2005 sample, and second, all money data were converted to 2005 dollars in order to remove the effect of inflation—acknowledging that it may not have affected all assets uniformly. The analysis is based on a sample of 15,933 families in 1999 and 5,103 in 2005.
Family refers to economic families and unattached individuals. An economic family is a group of persons sharing a common dwelling and related by blood, marriage, common law or adoption. An unattached individual lives alone or with unrelated persons.
The major income recipient is the family member with the highest income before tax. If two persons had exactly the same income, the older one was selected.
Pre-tax family income is the sum from all sources during the calendar year received by family members aged 16 and over. Sources include wages and salaries, net income from self-employment, investments, government transfers, pensions, scholarships and alimony. Excluded are income in kind, tax refunds, and inheritances.
Government transfers include all direct payments from federal, provincial and municipal governments to individuals or families. These include Child Tax Benefits, Employment Insurance, Canada/Quebec Pension Plan benefits, Old Age Security, Guaranteed Income Supplement, Spousal Allowance, Goods and Services Tax credit, workers' compensation, social assistance, provincial tax credits, and training allowances.
Financial assets consist of liquid and non-liquid assets. Liquid assets include deposits held in chequing and savings accounts, term deposits, guaranteed investment certificates, Canada Savings Bonds (including accrued interest), and other bonds. Non-liquid assets comprise registered retirement savings, registered education savings, registered retirement income funds, deferred profit sharing plans, treasury bills, stocks, mutual funds, mortgages owned, loans to others, annuities, trust funds, and other miscellaneous financial assets.
Non-financial assets are the market value of the owner-occupied home, other real estate, market value of owned vehicles (including recreational), value of the contents of a residence, other valuables and collectibles, and other non-financial assets.
Business equity is the market value of business assets less the book value of debt outstanding.
Savings in employer pension plans at the family level are the sum of accrued savings that can be claimed by members covered under such plans on termination of their job. Among retirees, these reflect their current entitlement. In both surveys, such pension savings were estimated on the basis of information collected on the type of plan, yearly contribution, and the number of years contributed, etc.7 Unlike conventional assets like a home or business, savings in such plans are not transferable except to a surviving spouse.
Total debt comprises any mortgage on an owner-occupied home or other real estate and all non-mortgage debt; the latter includes amounts owing on credit cards, secured and unsecured loans (including lines of credit from banks and other institutions), car loans, and other unpaid bills.
Wealth is total assets less total debt. It is based on marketable assets (with the exception of savings in employer pension plans) that are in direct control of families. It does not include future claims on publicly funded income security programs or any potential returns on human capital (like employment income or the ability to generate investment income).
To keep tables to a manageable size, wealth was examined in terms of eight components: savings in employer pension plan, business equity, home equity, equity in other real estate, and equity in vehicles, value of contents of residence, other non-financial assets, and net financial assets (total financial assets less total non-mortgage debt).
Mean wealth is aggregate wealth divided by the total number of families, whereas median wealth is the value at which half the families have lower values and half have higher values. The mean value is affected by extreme values whereas the median is not.
The Gini coefficient is a measure of inequality in a distribution. It lies between zero (no inequality) and one (total inequality)—the closer it is to 1.0, the greater the inequality in the distribution.
A family is treated as a debtor if it owes any money on a mortgage or other debt, and as an investor if it has non-zero investment income for the reference year. Investment income includes interest earned on deposits and bonds, dividends from stocks or mutual funds, and net rental income.
To study changes in family wealth over time, the ideal source would be longitudinal. However, using surveys conducted at different times allows the creation of groups of families (cohorts) sharing a common characteristic. The usual classifying characteristic is the age of a person—in this study, the major income recipient at the time of the 1999 survey. While other characteristics such as the type of family, area of residence, or income may change over time and contaminate the concept of a cohort, a person's age is least volatile and easy to use.
To avoid the problem of a family of two or more changing over time into two or more unattached individuals or vice versa, families and unattached individuals are used collectively as a unit of analysis. Given the range of age groups, the major income recipient may have changed, especially if one spouse retired and the other continued to work. Families with a major income recipient who was under 22 or who immigrated to Canada after 1999 were excluded from the 2005 data (accounting for 5.2% of families and 1% of the total wealth).8 No adjustment was made for emigrants who left after July 1999, or for those who may have been temporarily away between 1999 and April 2005. (Table)
Families in their 20s, 30s and 40s took most of the household credit between 1999 and 2005 and also experienced major shifts in their wealth distributions. For example, the proportion of families in their 20s with a net worth of less than $10,000 dropped from 70% to 45%, whereas the proportion worth between $50,000 and $249,999 jumped from 7% to 24% as these families increased their financial assets or bought a home. Overall, the distribution of wealth shifted by 26 percentage points for families in their 20s, 24 points for those in their 30s and 23 points for the 40s cohort. The shift was minimal (7 points) for families in their 60s. For instance, 14% had a net worth of one million dollars or more in 1999 compared with 15% in 2005. On the other hand, relatively more baby boomer families in their 50s and 40s increased their wealth to one million dollars or more (see Millionaire families for more details).
Overall, the distribution of wealth shifted by 8 percentage points—all at the upper end of the distribution—as families increased their wealth. However, the shape of the curve remained unchanged as median wealth stayed at 43% of the mean, and inequality measured by the Gini coefficient remained at 0.678. Statistically, the situation was not much different by cohort with the exception of families in their 20s and 30s whose wealth was slightly more equally distributed in 2005 than in 1999 as more of them owned a home. Median wealth rose from 7% to 26% of the mean for those in their 20s and from 35% to 51% for the 30s cohort—indicating reduced skewness in their wealth distributions. (Table)
The proportion of families with a net worth of one million dollars or more rose from 5% in 1999 to 9% in 2005. Almost all of the increase was concentrated among the baby boomers—for those in their 50s, the proportion jumped from 6% to 16%; for those in their 40s, from 2% to 9%. One in three millionaires were baby boomers in 1999 compared with about 6 in 10 by 2005. Among the oldest cohort, the proportion fell from 9% to 4% as a result of deaths, business wind-ups, home downsizing, or use of financial assets. The median age of the major income recipient in millionaire families fell from 58.2 to 56.9, but increased from 43.7 to 46.4 among non-millionaires.
On average, millionaire families held 10 times more wealth than non-millionaires ($1.9 million versus $190,000 in 1999 and $2.1 million versus $222,000 in 2005). While non-millionaires derived most of their wealth from home equity and an employer pension plan, millionaires' wealth came mostly from net financial assets, followed by business and home equity. The mean pre-tax income of millionaires, on the other hand, was only 2.5 times that of non-millionaires—$135,000 versus a little over $50,000. Despite their higher incomes, the proportion of millionaires carrying debt increased from 51% in 1999 to 58% in 2005, while non-millionaires with debt inched up from 68% to 71%.
Wealth was more equally distributed than income for millionaires, but the reverse for non-millionaires. (Table)
In both 1999 and 2005, about one million families drew their entire pre-tax income from government transfers. Compared with families receiving no transfers, these families were much older—the median age of the major income recipient was 49.9 in 1999 and 54.2 in 2005. Their mean income was only about $12,000 compared with $100,000 for other families. Because of their lower income coupled with age, less than 40% owed money compared with over 80% of those without transfers.
Even though their mean wealth rose from $35,000 to $57,000, it was still only about 10% of the level for those without transfers. Since one-fifth to one-fourth of transfer-dependent families owned their home, this equity plus the value of the contents of residence constituted around 60% of their wealth compared with 30% for those with no transfers. Transfer-dependent families also had relatively more equity in other real estate and very little in the way of net financial assets or employer pension plans.9
In both years, wealth was much more unequally distributed among transfer-dependent families. Part of this may be attributed to the low proportion of homeowners in this group.
- Between 1999 and 2005, per capita income of Canadians rose from $32,300 to $42,600 (or 31.9%) whereas the rate of inflation, measured by the change in the all-items Consumer Price Index, varied between 1.8% and 2.8%, unemployment rate between 6.8% and 7.7%, and the trend-setting bank rate, that determines interest rates charged on a variety of personal loans including mortgages, between 2.50% and 5.77%.
- Compared with the National Balance Sheet Accounts of the personal sector, a household survey collecting data on assets and debts usually provides underestimates of financial assets and slight overestimates of non-financial assets resulting in fairly comparable estimates of wealth. Under-reporting in a survey is primarily due to the poor recall capability and/or refusal of respondents. All of the missing data on components used to compile estimates of wealth are imputed.
- A similar approach was used in an earlier study on wealth (Chawla and Pold 2003).
- The current analysis is restricted to families by cohort based strictly on the age of the major income recipient rather than classifying families further into debtors and investors. Since the latter two concepts are much more volatile as families within a cohort may change status from debtor to investor and vice-versa, any further discussion based on these concepts is beyond the scope of this paper.
- All other things being equal, the monetary needs of a family drop when children leave home, and consequently, that family has the opportunity to improve its wealth situation by using the spare funds to acquire more assets and/or pay off any outstanding debts. On the other hand, if the departure of children encouraged that family to change its lifestyle and tastes and spend more on goods and services, then the situation would be different.
- An increase in the proportion holding savings in employer pension plans in this cohort may be attributed to a situation where an elderly major income recipient is likely living with a younger spouse and/or other family members. Data are analyzed at the family level. Different mortality rates between those with and without employer pension income may also be a factor.
- A detailed description of the methodology used to estimate savings in employer pension plans can be found in Survey of Financial Security – Methodology for estimating the value of employer pension plan benefits (Cohen, Frenken and Maser 2001).This paper and the SFS questionnaires are available on the Statistics Canada website ().
- In 1999, there were 12,216,000 family units with a total wealth of $3,432 billion. By 2005, there were 13,348,000 families with a wealth of $4,862 billion. Excluding 694,000 families with a major income recipient under 22 or who immigrated to Canada after 1999, there were 12,654,000 families remaining for the analysis. The difference of 438,000 families between 2005 and 1999 can be attributed to the re-weighting of the 1999 sample as well as to the dissolution of two-spouse families into lone-parents and unattached individuals and formation of new two-spouse units since some unattached individuals married by 2005.
- Transfer-dependent families, who were mostly renters, may have acquired real estate other than a home when their incomes were higher. Although incomes of families change as they dissolve or members become unemployed, withdraw, or retire from the labour market, some may have kept their assets intact. Income pertains to a given calendar year, whereas when an asset was purchased is not known.
- Chawla, Raj K. and Henry Pold. 2003. "Family wealth across the generations." (PDF)Perspectives on Labour and Income. Vol. 4, no. 10. October. Statistics Canada Catalogue no. 75-001-XIE. p. 5-15. (accessed June 11, 2008).
- Cohen, Michael, Hubert Frenken and Karen Maser. 2001. Survey of Financial Security – Methodology for estimating the value of employer pension plan benefits (PDF). Statistics Canada Catalogue no. 13F0026MIE – 01003. Ottawa. 47 p. (accessed June 12, 2008).
Raj K. Chawla is with the Labour and Household Surveys Analysis Division. He can be reached at 613-951-6901 or email@example.com.
You need to use the free Adobe Reader to view PDF documents. To view (open) these files, simply click on the link. To download (save) them, right-click on the link. Note that if you are using Internet Explorer or AOL, PDF documents sometimes do not open properly. See Troubleshooting PDFs. PDF documents may not be accessible by some devices. For more information, visit the Adobe website or contact us for assistance.
- Date modified: