Archived Content

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.

Statistics Canada - Statistique Canada
Skip main navigation menuSkip secondary navigation menuHomeFrançaisContact UsHelpSearch the websiteCanada Site
The DailyCanadian StatisticsCommunity ProfilesProducts and servicesHome
CensusCanadian StatisticsCommunity ProfilesProducts and servicesOther links
Side menu bar Catalogue Number 75-001-XIE Table of contents Latest issue News from The Daily Latest data Survey information Back issues Feedback Studies Latest issue in PDF

March 2005
Vol. 6, no. 3

Perspectives on Labour and Income

Spenders and savers
Raj K. Chawla and Ted Wannell

The national savings rate has been oscillating around historic lows for several years. This ebb tide in savings has prompted concerns about the resilience of the macro-level economy since growth is increasingly tied to debt-financed consumer spending. Many are also troubled by the balance sheets of households, where ever-expanding debt has rapidly outpaced earnings growth. The resulting record-high, debt-to-income ratios leave households more vulnerable to income interruptions.

While macro-level indicators can give the impression of a single economic ship riding the waves of cyclical activity, it is also important to recognize the variability behind the averages. Many households do save, but increasing numbers are slipping into the red and spending more than they earn in a year. This article focuses on the distinction between savers and spenders (see Data sources and definitions). It examines how patterns of saving and spending changed between 1982 and 2001, then looks at differences in the characteristics and spending patterns of saving versus spending households (Chart). The main objectives are to highlight both macro and micro factors that have contributed to the declining savings rate, and to provide some evidence on the financial vulnerability of spending households.

A cautionary tale of two time periods

Long-run changes in spending patterns are evident in the span of a generation (20 to 25 years)—sufficient time to observe changes in demography, the labour force and technology. The period from the early 1980s to the early 2000s witnessed the continuation of long-term demographic changes that contribute to an aging of the population: a declining birth rate and increasing life expectancy. The labour force participation of women also continued to rise. And new computer, communications and consumer products progressed from curiosities to commodities in a short span of years (see New technologies, changing tastes). All of these factors contributed to immutable changes in spending patterns (see also Harchaoui and Tarkhani, 2004).

On the other hand, some changes in spending may be related to cyclical swings in the economy caused by factors such as job loss and interest rate fluctuations. The early 1980s were punctuated by a short, but steep recession in 1981-82, while the early 2000s witnessed the continuation of a long economic expansion with only a slight pause in 2001. Unemployment, 11.0% in 1982 compared with 7.2% in 2001, may affect household spending adversely, but the effect is dampened by a number of factors. Some spending is 'locked in'—for example, mortgage payments, other loans, utilities, and food consumed in the home. Employment Insurance (EI) benefits can soften the drop in income. Also savings or credit may be used to smooth consumption over short periods of unemployment.

The markets for credit and savings were very different in the two time periods. In the early 1980s, inflation, nominal interest rates, and real interest rates were all very high, while just the opposite was true in the early 2000s.1 To some extent, this sea change seems to represent a longer-run (or secular) change in environment, rather than a cyclical fluctuation. The change in environment seems to have raised the level of spending relative to earnings but its effect on the disposition of spending and consumption smoothing is less clear. Suffice it to say that enough changes occurred in the course of 20 years—whole new classes of products, changes in relative prices, income taxes and security systems, to name a few—that the article adopts a mainly empirical approach to expenditure changes, with some reference to demographic life cycles to frame the analysis.

Finally, the exact points for comparison are dependent upon available survey data. In the early 1980s, expenditure surveys were conducted every four years (1982, 1986). The current household spending survey has been conducted annually since 1997, with the 2001 data being the latest available at the beginning of this study. Thus, to meet the criterion of a generational span, 1982 and 2001 were chosen.

Savings rate down

The official System of National Accounts (SNA) savings rate is simply the difference between the amount households take in as income and their expenditures on taxes and personal consumption. For each dollar of personal income received in 1982, Canadians paid 20 cents on taxes and deductions, spent 63 cents on personal consumption, and saved the remaining 17 cents. By 2001, taxes and deductions took 25 cents and personal consumption 71 cents, leaving just 3 cents in savings. As noted, annual figures may be affected by cyclical factors, but the broad trends are clear: Canadians are now spending more on taxes and personal consumption than a generation ago and, as a result, are saving less of their income.

A drop in savings potential implies more reliance on current income and borrowed funds for the purchase of both consumption and investment items. The largest investment for most families is their home. While the proportion of households owning a home edged up from 61% to 64% over the period, the SNA shows that household mortgage debt ballooned from $174.1 billion to $447.2 billion (in 2001 dollars).2 Similarly, SNA estimates of outstanding consumer loans (the amount owed on all credit cards, other personal loans, unpaid bills, and so forth) mushroomed from $84.1 billion to $203.8 billion. As a result, the total debt owed by households rose sharply from $258.2 billion in 1982 to $651.0 billion in 2001—an increase of 152% compared with an increase of just 42% in disposable income. Thus the debt-to-income ratio rose from 55% to 97%. On a per-capita basis, debt doubled from $10,300 to $20,900.

Both the proportion of spenders and the spending gap rose

As the savings rate fell, the number of households outspending their income in the course of a year increased. Of the 8.4 million households in 1982, 39% spent more than their pre-tax income (Table 1). By 2001, the number of households had risen to 11.7 million with 47% being spenders.

Overall, the total expenditures of saving versus spending households are remarkably similar. Spenders actually spent slightly more than savers in 1982, even though they brought in 28% less income. In 2001, saving households spent about $3,000 more than spending households as the income gap expanded to 35%. So it is mainly income that separates savers from spenders.

Since savers have higher average incomes than spenders, one would expect the proportion of savers to rise with income, as is indeed the case. However, between 1982 and 2001, the proportion of spenders rose across the entire income spectrum.

In 1982, 57% of all households with incomes under $20,000 were spenders compared with 16% of those with incomes of $100,000 and over. By 2001, the proportions had risen to 66% and 23% respectively.

Among spenders, the gap between spending and income grew across the income distribution. At the high end of the scale, spenders with incomes of $100,000 or more spent 11% more than their income in 1982 compared with 15% in 2001. The corresponding excess of expenditure over income was more extreme for those with incomes under $20,000—ramping up from 34% in 1982 to 54% in 2001.

Spending up in pre-retirement years

A life-cycle approach provides a useful framework for household spending. This approach divides the life of a household into three phases:

  1. Borrowing: Newly formed households finance investment in themselves (education, training) in expectation of rising income.
  2. Accumulation: In anticipation of retirement, households save from surplus income.
  3. Retirement or dis-saving: Households draw down their savings to finance consumption in later years.

To approximate these three phases, households are classified according to the age of the reference person: under 45, 45 to 64, and 65 or older.

In 1982, the proportion of spenders peaked at 44% in households with a reference person under 45, fell to 34% among those with a reference person aged 45 to 64, and remained at 33% for senior households. There is no apparent upswing in dis-saving after age 65 since public and private pension benefits appear as income for individual households but essentially represent macro-level dis-saving.3 Furthermore, irregular spending on big-ticket items such as cars will push a certain percentage of accumulators into the red each year. Taking these factors into account, the standard life- cycle model provides a reasonable interpretation of the 1982 data.

In 2001, the life-cycle pattern flattened considerably as the proportion of spenders rose in all age groups—the biggest increase occurring in the accumulation phase of the life cycle, where the proportion jumped from 34% in 1982 to 45%. The propensity to outspend income increased to 42% for senior households and to 50% for younger households. So the change in the interest rate and credit environment seems to have increased spending across the life cycle, but particularly among households approaching or already in retirement.

Household composition is another element of the life cycle that contributes to the saving-spending balance. In both 1982 and 2001, about half of all unattached individuals and lone-parent families were spenders. However, the proportion of married couples joining the spending group increased significantly over time.

Investment in education must also be considered. Fees for postsecondary education increased markedly through the 1990s, raising the possibility that increased debt might be necessary to fund studies. Indeed, among households incurring expenditures on education of children or other members, the proportion outspending their income increased from 40% in 1982 to 48% in 2001.

Finally, homeownership also follows a life-cycle pattern, with renting more prevalent among the young, followed by homeownership with a mortgage and then mortgage freedom. In both periods, the proportion of spenders dropped across this progression, but again it also increased over time for each group. The greatest increase in spenders occurred among homeowners with a mortgage, jumping from 35% to 46%.

Higher incomes of savers associated with higher personal taxes

Despite very similar mean expenditures, patterns differ for spenders and savers (Table 2). In 1982, spenders devoted 80.7 cents of their expenditure dollar to current personal consumption, 12.8 to personal taxes, 3.6 to security, and 2.9 to gifts and contributions;4 the corresponding breakdown for savers was 71.4, 20.6, 4.9, and 3.1. Savers allocated significantly less for current consumption but more for taxes because of their higher incomes.

Two decades later, both groups were spending less on consumption and more on taxes and security. Of a dollar increase in expenditure, spenders spent 74 cents on consumption and 25 cents on taxes and security, compared with the savers' 57 and 43 cents.

Spenders out-consume savers

Even though total expenditure for spending and saving households is quite similar, spenders actually consume substantially more than savers (Table 3). This is due to the standard definition of consumption, which subtracts personal taxes, security expenditures (for example, EI and C/QPP premiums), and gifts and charitable contributions from total expenditure. Since personal taxes and security payments are higher for savers, less of their spending is devoted to consumption. On average, spenders consumed $39,000 in goods and services in 1982, 15% more than the $33,900 meted out by savers. Reflecting the greater increase in income for saving households between 1982 and 2001, their consumption increased faster than spending households (12% compared with 7%). Still, spenders continued to out-consume savers by 10% in 2001 ($41,700 compared with $37,900).

Cars push spenders into the red

The greatest difference in consumption patterns between spending and saving households is transportation expenditures—mainly car purchases. In 1982, spending households dished out 43% more on transportation than did saving households. By 2001, the differential had grown to 54%. These differentials were driven by average car purchases of $9,900 in 1982 and $15,200 in 2001 for spenders, compared with $6,400 and $9,000 for savers. About one-fifth of both spenders and savers rented or leased a vehicle and spent, on average, about $3,000 on it in 2001.5

Shelter expenses were the other major difference between spenders and savers in 1982 ($8,800 versus $8,000), but the gap had narrowed substantially by 2001. Spenders also consistently out-purchased savers, though by smaller margins, in household furnishings and equipment, recreation, education, and tobacco products and alcoholic beverages.

Similar purchasing pattern changes for spenders and savers

Economic and market forces led to similar changes in purchasing patterns for both spending and saving households. Both groups spent more in 2001 on shelter, household operation, transportation, health, personal care, recreation, education, income tax and security; and less on food, household furnishings and equipment, clothing, reading and other printed materials, tobacco products and alcoholic beverages, miscellaneous, and gifts and contributions.6

Conclusion

Between 1982 and 2001, the mean pre-tax income of Canadian households grew from $51,400 to $56,800 (11%) whereas expenditure jumped from $47,800 to $55,300 (16%). Not only did expenditure grow more than income, but households also paid relatively more of their income in personal income taxes (federal and provincial) and security (such as premiums for EI, C/QPP, and other government and non-government pensions), leaving less for personal consumption and saving.

At the same time, households increased their indebtedness for both mortgages and consumer debt. Per-capita debt doubled over the two decades. As a result, the proportion of households spending more than their income increased from 39% in 1982 to 47% in 2001. Spending households tended to be younger and to have lower incomes. They were also more likely to be renters or homeowners with a mortgage. Savers, with relatively higher incomes, tended to pay more for taxes and security.

Between 1982 and 2001, the proportion of spenders grew in all income classes and across all life-cycle phases. The incidence of outspending household income increased the most in the 45-to-64 age group. An increasing proportion of senior households also entered the spending ranks.

Spending households had expenditure levels similar to savings households, but substantially lower incomes. After netting out expenditures for taxes and security, spending households had substantially higher levels of current consumption. Spenders out-consumed savers by a small margin across a number of categories, but the main difference was spending on automobiles. These households spent thousands more per year on car purchases.

The influence of automobile purchases suggests a transitory component in the spender-saver split. In any year, a number of households that are long-run savers may borrow enough for the purchase of a car (or other expensive item) to slip into the red. This component is not trivial. In 2001, the purchase price of a car exceeded the net addition to household liabilities in 7 of 10 spending households. Nevertheless, spending households exhibit a greater propensity to consume in relation to their income and across a wide variety of goods and services. And an increasing number of households fit this mould.

Low interest rates and easy credit undoubtedly influence the inclination of households to borrow as household debt continues to rise to unprecedented levels in relation to household disposable income. But at the same time that households have been accumulating this debt, they have also seen a steady rise in their net worth. As a result, SNA estimates of the ratio of household debts to assets have remained in a narrow band of 16% to 19% over the past 14 years. Herein lies another factor related to increased consumption: the recent run-up in housing values. Bank of Canada studies (Macklem 1994; Pichette and Tremblay 2003) indicate that rising home equity has a positive effect on consumer spending.7 Thus with resale values rising by a third over the past five years, it is no surprise that homeowners are feeling good about their finances and their ability to spend.

This debt-driven consumption and associated economic growth may be subject to peril. A sudden drop-off in the housing market or a sudden spike in interest rates could throw cold water on the spending party. On the other hand, the risks associated with the household sector are somewhat dampened by increased savings in the corporate and government sectors, recently noted in the System of National Accounts. Governments and the Bank of Canada may also use fiscal and monetary policies to dampen the effects of cyclical swings.

Interestingly, the increased propensity among older age groups to consume could help to offset a longer-run economic risk. Long-run forecasts that assume a more traditional life-cycle pattern of spending associate the aging population, notably the retirement of the baby boomers, with a decrease in economic activity. Assuming the boomers can accumulate enough wealth to support their consuming ways, the long-run picture may be a little brighter.

Data source and definitions

The analysis is based on the 1982 Family Expenditure Survey (FAMEX), conducted in February-March 1983, and the 2001 Survey of Household Spending (SHS) of January-March 2002. Since the surveys were taken nearly 20 years apart, some changes in spending patterns could be attributed to changes in survey concepts, content and methods. Both surveys were conducted by personal interview, and used a multi-stage stratified clustered sample drawn from the Labour Force Survey frame that excludes population in institutions such as nursing homes, hospitals and penitentiaries and those living in the territories or on Indian reserves. However, some key differences remain. First, FAMEX, a periodic survey until 1996, asked 641 questions compared with 425 in the SHS, an annual survey since 1997. Second, the methods used to derive the final weighting factors for the population estimates were different, and much more automated systems were used by the SHS. For more details on these issues, see Statistics Canada (1984, 2000 and 2003).

The surveys collected expenditures and income from all private households in the 10 provinces. The household spending unit is defined as a group of persons dependent on a common or pooled income for major expenses and living in the same dwelling, or one financially independent individual living alone. Since the composition of a household may vary over a year, the use of part-year and full-year households would have distorted some of the comparisons. Hence, the analysis is restricted to full-year households and their composition and dwelling characteristics as of December 31 linked to details on expenditures incurred and income received during the calendar years 1982 and 2001. The usable samples were 10,938 households for 1982 and 15,899 households for 2001.

Household: A person or group of persons occupying one dwelling unit. The number of households, therefore, equals the number of occupied dwellings. A full-year household has at least one full-year member; a part-year household is composed entirely of part-year members.

Head/reference person: Despite some differences, the two concepts are used here synonymously. The 1982 data are classified by age of the head of household and the 2001 data by age of the reference person. The husband was treated as the head in families consisting of married couples with or without children, as was the parent in lone-parent families and normally the eldest in all other families. On the other hand, the reference person was chosen by the household member being interviewed as the person mainly responsible for the financial maintenance of the household. Also, this person must have been a member of the household on December 31 of the reference year. The head/reference person can be either male or female.

Tenure: Households are classified by tenure (homeownership status) into three groups: renters, homeowners without a mortgage, and homeowners with a mortgage.

Expenditure on shelter: Data on this component are not comparable. In 1982, they included mortgage interest on a home or vacation home whereas the principal was included under 'net changes in assets and debts.' In 2001, this component included information on regular mortgage payments (principal and interest).

Pre-tax household income: Sum of incomes before taxes and other deductions received during the reference calendar year by all members of the household. Sources include wages and salaries, net income from self-employment, rental and investment income, government transfers (EI benefits, Child Tax Benefits, GST credits, provincial tax credits, social assistance, Old Age Security, Guaranteed Income Supplement, C/QPP benefits), private and employer pension plans, scholarships, alimony, child support payments, and so forth. Income in kind, windfall gains, and capital gains/losses are excluded.

Disposable income: Pre-tax income less federal and provincial income tax less premiums/contributions paid on components pertaining to security (such as EI, life insurance, C/QPP, and other government and non-government work-related pension plans). Contributions to registered retirement savings plans are not treated as a component of security.

Expenditures collected: With some minor exceptions, the survey includes spending on all goods and services received during the reference calendar year. All expenses attributable to an owned business are excluded. On the other hand, taxes such as the GST, provincial sales tax, duties, customs and excise on all goods and services purchased are included in expenditures.

Total expenditure: Sum of expenditure on current consumption of goods and services, federal and provincial income tax paid, payments pertaining to security, and gifts and contributions made.

Current consumption (also referred to as total consumer spending): Includes expenditure on broad components: food, shelter, household operation, household furnishings and equipment, clothing, transportation, health, personal care, recreation, reading material and other printed matter, education, tobacco products and alcoholic beverages, and miscellaneous (including union dues and games of chance). For a detailed breakdown of the components, see Statistics Canada (2003).

Spender/saver: For analytic purposes, households are classified into two groups: those whose total expenditure exceeds income as spenders, and those whose expenditure equals or is less than income as savers. This classification by no means implies that the former group was more extravagant or spent more frivolously than the latter, or had no savings or wealth. This statistical divide, based on total income and expenditures during the reference years, is made simply to look at the two groups by life cycle, compare their spending patterns, and highlight any changes over time.

Data in constant dollars: To remove the effect of inflation or rising prices over time on consumption, all expenditures and incomes are in 2001 dollars. While the prices of all 1982 goods and services may not have moved up at the same pace as the all-items CPI, the use of one conversion factor simplifies the analysis.

Average expenditure by item: Two averages are used; the overall and for reporters only. Tables 2 and 3 use the overall averages.

New technologies, changing tastes

In addition to the macro-economic factors that seem to be spurring spending, the array of new products and services may also be loosening purse strings. Technologies that were rare or non-existent in 1982—home computers, cellular phones, VCRs, CDs, DVDs, the Internet, game consoles, digital cameras—are now commonplace. Many of these electronic devices follow development paths where features and quality increase rapidly relative to prices, leading to short product life spans. This generates incentive to upgrade frequently.

Spenders and savers do not differ greatly in their ownership of these technological devices. Rather, both reflect the rapid growth in such technologies over the course of a generation. Desktop computers, at the heart of the technological revolution, were in the realm of hobbyists in 1982 but could be found in the majority of homes in 2001. Similarly, connecting these computers to the Internet has progressed from a rarity (not even measured in 1982) to just about half of all households in 2001. Cell phones were owned by almost as many 2001 households.

As of 2001, some newer technologies have begun elbowing their way into categories that were saturated by an earlier wave of technology. For example, DVDs are rapidly supplanting video cassettes. Just less than one in five households owned a DVD player in 2001, jumping to more than half in 2003. Although not on the same rapid trajectory as DVDs, satellite dishes could be found in 18% of households in 2001, providing significant competition to cable TV services.

In addition to these new toys, the taste for comfort and convenience in the home has also increased markedly in the past 20 years. Household ownership of microwave ovens increased sevenfold and central air conditioning fivefold, while twice as many kitchens had dishwashers. Even the stalwart washers and dryers experienced appreciable growth in this period. (New technologies, changing tastes - Table)

Notes

  1. The rate of inflation in 1982 was 10.9% compared with 2.6% in 2001, and the trendsetting bank rate was 13.96% versus 4.31%. The bank rate, set by the Bank of Canada, affects not only the rate households pay on personal loans, mortgages, lines of credit and other consumer loans, but also what they receive as return on their savings and investments.

  2. All of this mortgage debt may not necessarily be owed by first-time home buyers; it includes debt owed by those who may have remortgaged their home and used funds for business, investment or consumption.

  3. The System of National Accounts treats pension benefits and RRSP withdrawals as dis-savings and thereby presents a more coherent picture of the life-cycle model. In this respect, senior savers as measured by their SHS total income might more properly be thought of as senior households with a positive cash flow.

  4. Gifts were treated somewhat differently in the 1982 and 2001 surveys. The 1982 questionnaire contained a separate category for gifts, while in 2001 respondents were directed to include them under the relevant subject category (furniture, toys, and so forth), except for clothing. This creates a small upward bias in personal consumption in 2001 relative to 1982.

  5. Since no separate data for expenditure on renting or leasing of automobiles or other vehicles were captured in 1982, this inference should be viewed with some caution. Averages here are for reporters only.

  6. Households in the U.S. also spent less on food and clothing in 2000 than in 1990; like their Canadian counterparts, they spent relatively more on shelter, transportation and health. For details, see Weiss (2002).

  7. Financial asset appreciation was also found to have a positive, but much weaker effect on consumer spending.

References

  • Harchaoui, Tarek M. and Faouzi Tarkhani. 2004. "Shifts in consumer spending." Perspectives on Labour and Income (Statistics Canada, Catalogue no. 75-001-XIE). June 2004 online edition.

  • Macklem, Tiff. 1994. Wealth, disposable income and consumption: Some evidence for Canada. Technical report no. 71. Bank of Canada: Ottawa.

  • Pichette, Lise and Dominique Tremblay. 2003. Are wealth effects important for Canada? Working paper no. 2003-30. Bank of Canada: Ottawa.

  • Statistics Canada. 1984. Family expenditure in Canada, 1982. Catalogue no. 62-555-XPB. Ottawa.

  • ---. 2000. Note to former users of data from the Family Expenditure Survey. Catalogue no. 62F0026MIE-00002. Ottawa.

  • ---. 2003. Spending patterns in Canada, 2001. Catalogue no. 62-202-XIE. Ottawa.

  • Weiss, Michael J. 2002. "Inconspicuous consumption." American Demographics 24, no. 4 (April): 30-39.

Full article in PDF

Authors
Raj Chawla and Ted Wannell are with the Labour and Household Surveys Analysis Division. Raj Chawla can be reached at (613) 951-6901, Ted Wannell at (613) 951-3546 or both at perspectives@statcan.gc.ca.


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.



Home | Search | Contact Us | Français Return to top of page
Date Modified: 2014-05-14 Important Notices