December 2005
Vol. 6, no. 12

As households age, their economic and demographic situations change. Income, savings and wealth generally decline, and household size shrinks as adult children leave or a spouse dies. Spending patterns may also change. For example, older families may spend less on transportation as they experience reduced mobility, and more on health.

The economic well-being of older households with respect to pre- and post-tax income, low-income rates, and wealth holdings has been widely discussed (Myles 2000; Gower 1998; Chawla and Pold 2003; Williams 2003). However, less is known about how their income is divided among taxes, security,¹ consumption and savings. This article looks at three household groups based on the age of the reference person: 55 to 64, 65 to 74, and 75 or over (see Data sources and definitions). Since more than three-quarters of the first group had employment earnings compared with around one-third of the second group and just one-tenth of the third, the shifts in expenditure patterns should also reflect the adjustments households make as their active attachment with the labour market diminishes.²

Household expenditure depends on factors such as income, size, composition, and urban or rural location, so any comparisons over time would at least require adjustments with respect to type and size of household. A common approach is to use per capita or equivalence scale concepts (Pendakur 1998). Since the study focuses on households at a life-cycle stage when the majority are either couples with no children or unattached individuals, it should not be affected by such concerns. In 1982, unattached men and women and couples accounted for 57% of all households in the 55-to-64 group compared with 86% in the 75-and-over group; by 2003, their proportions were 61% and 85% respectively.

Expenditure patterns change not only over the life cycle but also over time as new products and services emerge. Changes in spending patterns between 1982 and 2003 are highlighted using 'similar' rather than 'cohort' households. For example, an increase of $100 in mean expenditure on a given item by unattached men implies that they were, as a group, spending that much more in 2003 than similar men in 1982. (All money figures are in 2003 dollars.)

Socio-demographic transitions as households age

The composition of households changes notably as they age. In both 1982 and 2003, a little over one-third of households in the 55-to-64 group still contained children or other relatives, with the remainder being unattached individuals or couples (Table 1). By 65 to 74, however, households consisted largely of couples and unattached women; and by 75 plus, unattached women predominated, at a little over 40% of households. Such compositional shifts result in smaller households, causing some to downsize or move to rental accommodation. For instance, between the 55-to-64 and 75-plus age groups, the proportion renting increased from 28% to 43% in 1982 and from 24% to 36% in 2003.

Another change at this time concerns major source and amount of income. In both 1982 and 2003, three-quarters of households in the 55-to-64 group had employment earnings, accounting for more than 70% of their income. For those in the 75-plus group, on the other hand, government transfers and pensions (private and work-related) became more prevalent—constituting 59% of income in 1982 and 80% in 2003. Although a greater proportion of households in the 75-plus group reported earnings in 2003 than in 1982, the share of income from earnings fell from 12% to 9%. The share of income from investments also fell for this group—from 29% in 1982 to 10% in 2003.³

A change in the principal component of income is accompanied by a shift in the income distribution of households as they age. In both 1982 and 2003, the majority of households in the 55-to-64 group, with earnings as the major source of income, had incomes of $50,000 or more, whereas the majority in the 75-plus group, with pensions and transfers, received under $20,000.

Income, consumption and expenditure changes

As households age, their income drops (Chart). The largest decline occurs between the 55-to-64 and 65-to-74 groups, as labour market attachment diminishes and earnings are no longer the major source of income. A further drop occurs between the 65-to-74 and 75-plus groups, largely because of little or no earnings and more reliance on government transfers and pensions. Compared with a mean income of $53,100 for households 55 to 64 in 1982, those 65 to 74 received 39% less and those 75 plus, 52% less (Table 2). By 2003, even though the mean income of households in the first group had grown to $62,800, the pattern remained the same, with income falling by 34% and 52% for those in the older groups. In both years, the mean income of households 55 or older with no earnings was about half that of those with an employed member.

As household income declines, so does expenditure. The expenditure drop reflects primarily the drop in income tax and security contributions. Under a progressive taxation system, the effective tax rate drops as income decreases. Security contributions will also be less as people retire from paid employment. However, income and expenditure do not drop equally over the three age groups. Income drops much more significantly between the 55-to-64 and 65-to-74 groups, largely because of the loss of earnings, whereas expenditure drops more gradually because households take a little longer to adjust their spending. Among couples, for example, of the total drop in income over the three age groups in 2003, 68% occurred between the two younger groups (55 to 64 and 65 to 74) with 32% between the two older ones (65 to 74 and 75 plus); the corresponding drops in expenditure were 64% and 36%, with 58% and 42% for personal consumption.

Spending changes by age

Since income is a key determinant of expenditure, a drop in income may adversely affect standard of living. Households may spend more than their income, running down savings or incurring debt in order to maintain their lifestyle. In fact, about one-third of households 55 and over spent more than their income in 1982, and almost 41% in 2003. One-sixth of households in the 55-to-64 and 75-plus groups were on the border line, with expenditure within 5% of income. The majority of those who outspent their income did so by 5% to 25%.

All households spent most of their income dollar on personal consumption—anywhere between 58 cents and 84 cents, depending on age and type of household. The remainder went for income tax, security contributions, gifts and contributions,⁴ or savings. (The shares spent on these items also varied by age and type of household.) In 1982, households in the 55-to-64 group, with earnings as the major source of their relatively higher incomes, used 64 cents for personal consumption, 16 cents for income tax, and 4 cents each for security and gifts and contributions, saving the remaining 12 cents; by 2003, such households were spending more on consumption (69 cents), income tax (20 cents), security and gifts and contributions (8 cents), and saving very little (3 cents). The situation was no different for non-working households in the 75-plus group. They spent 65 cents of each income dollar on personal consumption and another 8 cents on income tax in 1982, compared with 73 cents and 12 cents in 2003. Consequently, these households also saved much less of their income dollar in 2003 than in 1982—6 cents versus 19.

In both 1982 and 2003, unattached women in the 55-to-64 and 75-plus groups spent most of their income dollar on personal consumption (food, shelter, household operations, clothing, and the like)—much more than their male and couple counterparts. Since these women's incomes were low, they of course paid less in income tax and security contributions. Nonetheless, they spent relatively more of their income dollar on gifts and contributions and saved less. On the other hand, unattached men aged 55 to 64 spent more of their income dollar on gifts and contributions and income tax than couples.

Consumption changes by age

All types of households spent more of their income dollar on personal consumption in 2003 than in 1982. In 1982, the 55-to-64 group spent $34,200 compared with $16,700 for those 75 plus. By 2003, spending had reached $43,500 and $22,000 (Table 3A). The widening gap between working and non-working households largely reflected greater expenditures by working households—$9,300 compared with $5,300. As always, food, shelter and transportation dominated, accounting for between 61 and 68 cents of each consumption dollar. The ranking of these three items changed for households in the 55-to-64 group—from food, shelter, transportation in 1982 to shelter, transportation, food in 2003. However, the order did not change for those 75 plus: shelter, food, transportation (Table 3B).

The next three components of consumption in 1982 for those 55 to 64 were clothing, recreation, and household operations; in 2003, this group spent relatively more on recreation and much less on clothing. In both years, these three items accounted for another 18 to 19 cents of consumption. For households in the 75-plus group, on the other hand, the next three components of consumption in 1982 were household operations, clothing and household furnishings; by 2003, the last two were replaced by health and recreation. Expenditure on these three components took 17 to 19 cents. Overall, then, in both 1982 and 2003, just six components of consumption accounted for 80% of the total for households in the 55-to-64 group and 84% for those 75 plus.

The amount spent on personal consumption drops as households age. For instance, in 1982, mean consumption by couples in the 55-to-64 group was $32,400 compared with $20,800 for those 75 plus, almost 36% less; by 2003, the difference was nearly 42% as expenditures hit $44,400 and $25,900 for the respective groups. A similar pattern prevailed for unattached individuals. The picture was much the same in 2003, but with narrower gaps between unattached individuals and couples in the 75-plus group.

Since most women have lower incomes than men, they also consume less. In the 55-to-64 group in 1982, women had 35% less income but only 14% less consumption. But as women's incomes improved over time, their income in 2003 was only 13% less and their consumption matched men's. For unattached individuals 75 plus, on the other hand, both income and consumption ratios by sex dropped—from 81% to 78% for income and from 97% to 83% for consumption.

Despite increases in personal consumption between 1982 and 2003, the spending patterns of couples in the 55-to-64 and 75-plus groups were about the same. In both years, food, shelter and transportation accounted for nearly two-thirds of their total consumption. Both groups spent less on food in 2003 than in 1982, but more on shelter and transportation. More was also spent on recreation and health; for couples 55 to 64, the mean expenditure rose from $1,500 to $3,900 (157%) on recreation, and from $1,000 to $2,300 (116%) on health; the corresponding increases in the 75-plus group were from $800 to $1,000 (25%) and from $700 to $2,100 (219%).

The key spending patterns of unattached individuals were similar to couples. Like couples in the 55-to-64 group, unattached men and women allocated a little over 60% of their consumption to food, shelter and transportation. However, in the 75-plus group, women spent more on household operations while men spent much more on transportation. The gap between men and women on health expenditures narrowed in the 75-plus group—men spent a little over half the amount spent by women in 1982 but slightly more in 2003.

Decreases occurred in some areas of spending as households aged. For example, for couples in 1982, substantial decreases were noted for tobacco and alcohol, recreation, transportation, health, and clothing. Shelter and food on the other hand showed the least reduction. By 2003, the components with larger reductions remained the same but those with least reduction now included health, and reading and printed materials.⁵ Similarly, for the unattached, relatively smaller reductions were noted for shelter, food, and household operations.

Rising health expenditures

Between 1982 and 2003, household expenditures on health rose because of increased premiums for government and private health insurance, and because of higher out-of-pocket expenses for treatments and medicines not covered by insurance.⁶ Households with a reference person 55 and over spent $7.2 billion in 2003 on health compared with $2.1 billion in 1982. And in both years, health insurance premiums accounted for 30% of these costs.

Since supplementary medical coverage through a private insurance plan is often a benefit of employment, the proportion of households covered under such schemes declines between the 55-to-64 and 75-plus groups. For instance, for unattached women, it fell from 53% to 47% in 1982 and from 49% to 42% in 2003. Thus, not only are more households in the 75-plus group incurring more out-of-pocket health expenses, but also these direct costs constitute the lion's share of their health expenditure—for unattached women, the percentage grew from 78% in 1982 to 81% in 2003 while jumping from 64% to 75% for men (Table 4).

Besides health insurance, all households, irrespective of age, spent the most on prescribed drugs, and other medical equipment and appliances. After these two, the order of spending on dental services, eye care, and other health care and medical services varied across age groups—more in 1982 than in 2003. However, couples and unattached individuals in the 75-plus group in 2003 showed a consistent order of out-of-pocket spending on health: prescribed drugs, other medical and health care services, dental services, and eye care.

Spending patterns of households dependent on government transfers

For households in the 55-to-64 group, government transfers may include Employment Insurance benefits, worker's compensation, C/QPP disability benefits, or social assistance; for households 65 to 74 and 75 plus, such payments may also include Old Age Security, Guaranteed Income Supplement, the Allowance, veterans' pensions, or the C/QPP retirement pension. Households in the latter two age groups are more likely to derive all their income from government transfers—especially those with no work-related pension, investments, or other source of income.

In 2003, almost one-third of households in the oldest group received their entire income from government transfers compared with less than one-fifth in 1982; the corresponding proportions for the 55-to-64 group were 9% and 8%. In both years, two-thirds of these households were unattached individuals (more women than men) and one-quarter were couples.

The average income of households totally dependent on transfers in the 55-to-64 group rose from $11,200 in 1982 to $12,900 in 2003, while their expenditures jumped from $11,800 to $15,400. In the 75-plus group, on the other hand, income went from $12,500 to $17,000 and expenditures from $11,900 to $17,200 (Table 5). The higher income of the 75-plus group in 2003 can be attributed to the maturity of the C/QPP, resulting in more recipients as well as higher benefits, and to inflation-adjusted payments from other programs. In spite of such increases in income, 42% of these households spent more than their income in 2003 compared with 35% in 1982; the corresponding proportions for households in the 55-to-64 group were 62% and 53%.

Because of lower incomes, households with their entire income from government transfers paid very little in personal taxes or security contributions. Instead, they spent their income on personal consumption. Those 75 plus spent slightly more on gifts and contributions than those 55 to 64. In fact, in both 1982 and 2003, households in the 75-plus group spent, on average, more on gifts and contributions than they did on personal care, recreation, or tobacco and alcohol.

Most of the consumption dollar in households dependent on transfers went for food and shelter, accounting for 52 to 57 cents in 2003, compared with 58 to 65 cents in 1982. The relative share spent on shelter grew over time as rent and home maintenance went up and food dropped. Relatively similar amounts were spent in 1982 and 2003 on transportation and household operations. The major difference between the 55-to-64 and 75-plus groups was in spending on tobacco and alcohol and on health. The former spent more on tobacco and alcohol, the latter on health.

Summary

As households age, not only does their income drop but also their spending patterns change. In addition, they tend to become smaller, which may necessitate downsizing or moving to rental accommodation. The loss of earnings as the major income source means less personal income tax to pay and almost no contributions for security. This lowers expenditures. On the other hand, the proportion of spending on personal consumption as well as gifts and contributions tends to increase. Changes in spending patterns also reflect altered lifestyles.

All households 55 and over were spending more on personal consumption, income tax and security in 2003 than in 1982. As a result, saving fell from 13% of income in 1982 to only 4% in 2003. Largely because of their higher incomes, couples fared better than unattached individuals.

The key components of household consumption were food, shelter, and transportation, together accounting for 61 to 68 cents of the consumption dollar. The ranking of these components changed for the 55-to-64 group: from food, shelter and transportation in 1982 to shelter, transportation and food in 2003. For those 75 plus, on the other hand, the ranking remained unchanged: shelter, food, and transportation. Households in this group were also spending more out-of-pocket on health in 2003 than in 1982; expenses incurred were for prescribed drugs, other medical and health care services, dental services, and eye care (ranked by relative share of the health dollar).

The proportion of households receiving their entire income from government transfers increased over the 1982-to-2003 period. Nearly two-thirds of these households were unattached individuals—more women than men. Most of their consumption dollar was spent on the two essentials of food and shelter: 52 to 57 cents in 2003 compared with 58 to 65 cents in 1982. The major difference over time in spending of these households by age was that those 55 to 64 were spending more on tobacco and alcohol whereas those 75 and over spent more on health.

Notes

1. Security expenditures include Canada/Quebec Pension Plan, Employment Insurance, and private pension plan contributions.

2. The analysis could have been carried out by splitting older households into only two groups: 55 to 64 and 65 and over. However, this would have masked the expenditure patterns of households in which the reference person may have opted to delay retirement to after age 65, or in which the spouse or other family members are working. Such situations, more common in 2003 than in 1982, made it desirable to examine a 65 to 74 year-old group, many of whom had some attachment with the labour force—32% had earnings in 2003 compared with 30% in 1982 . Although the tables show data for the three age groups, the text makes comparisons mainly between the 55-to-64 and 75-plus groups. In the former, the majority had earnings, whereas in the latter, the majority did not. See Chawla and Wannell (2005) for shifts in expenditures between 1982 and 2001 based on household surveys, and Harchaoui and Tarkhani (2004) and Sauvé (2005) for shifts based on personal expenditure data from the System of National Accounts.

3. Some of the drop in investment income may be due to the decline in the trendsetting bank rate, which fell from 13.96% in 1982 to 3.19% in 2003. Not all households are equally affected by interest-rate fluctuations. For households with large savings, a higher rate will generate more investment income, perhaps encouraging them to spend more. On the other hand, for households with greater consumer and mortgage debt liability, the higher rate may dampen spending as more of their income goes toward discharging debt.

4. Gifts were treated somewhat differently in the 1982 and 2003 surveys. The 1982 questionnaire contained a separate category for gifts, while in 2003 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 2003 relative to 1982.

5. Some of the reduced expenditure on food, clothing and recreation over time may be attributed to a drop in prices for these products and services. This has been brought about largely by increased competition in the retail and wholesale markets, the opening of discount outlets, and changes in tariffs and quotas on imports. Similarly, some reduction in spending on tobacco and alcohol may be due to greater knowledge of their harmful health effects.

6. The SHS does not collect information on the cost of treatment provided by doctors or hospitals under provincial health insurance schemes. Instead, it asks about expenses such as government or private insurance health premiums, prescription drugs, dental and eye care, and services provided by other medical professionals. See also Luffman (2005) for spending by households on prescription drugs.

References

• Chawla, Raj and Henry Pold. 2003. "Family wealth across the generations." Perspectives on Labour and Income (Statistics Canada, catalogue no. 75-001-XIE). October 2003 online edition.

• Chawla, Raj and Ted Wannell. 2005. "Spenders and savers." Perspectives on Labour and Income (Statistics Canada, catalogue no. 75-001-XIE). March 2005 online edition.

• Gower, Dave. 1998. "Income transition upon retirement." Perspectives on Labour and Income (Statistics Canada, catalogue no. 75-001-XPE) 10, no. 4 (Winter): 18-23.

• 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.

• Luffman, Jacqueline. 2005. "Out-of-pocket spending on prescription drugs." Perspectives on Labour and Income (Statistics Canada, catalogue no. 75-001-XIE). September 2005 online edition.

• Myles, John. 2000. "Incomes of seniors." Perspectives on Labour and Income (Statistics Canada, catalogue no. 75-001-XPE) 12, no. 4 (Winter): 23-31.

• Pendakur, Krishna. 1998. "Changes in Canadian family income and family consumption inequality between 1978 and 1992." Review of Income and Wealth 44, no. 2 (June): 259-283.

• Sauvé, Roger. 2005. The Current State of Canadian Family Finances - 2004 Report. Ottawa: The Vanier Institute of the Family.

• Snider, Bradley. 2005. Constant Dollar Adjustment of Expenditure Data from the Survey of Household Spending. Statistics Canada, catalogue no. 62F0026MIE2005005. Ottawa: Statistics Canada.

• 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.

• Williams, Cara. 2003. "Finances in the golden years." Perspectives on Labour and Income (Statistics Canada, catalogue no. 75-001-XIE). November 2003 online edition.

Full article in PDF

Author
Raj K. Chawla is with the Labour and Household Surveys Analysis Division. He can be reached at (613) 951-6901 or 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.