Adolescents' Reported Saving, Giving, and Spending As a Function of Sources of Income

Russell W. Belk, University of Utah
Clifford Rice,
Randall Harvey,
ABSTRACT - Based on survey data from 13- to 14-year-old adolescents, it appears that whether children's discretionary income is received as a gift or earned has an influence on how they allocate these funds to spending on self, spending on others, or saving. Findings suggest that giving money to children may foster saving and gift-giving more than spending on self. However, dual sources of income and greater amounts of income are associated with more spending on self. Implications for research and adolescent consumer socialization practices are discussed.
[ to cite ]:
Russell W. Belk, Clifford Rice, and Randall Harvey (1985) ,"Adolescents' Reported Saving, Giving, and Spending As a Function of Sources of Income", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 42-46.

Advances in Consumer Research Volume 12, 1985      Pages 42-46


Russell W. Belk, University of Utah

Clifford Rice

Randall Harvey


Based on survey data from 13- to 14-year-old adolescents, it appears that whether children's discretionary income is received as a gift or earned has an influence on how they allocate these funds to spending on self, spending on others, or saving. Findings suggest that giving money to children may foster saving and gift-giving more than spending on self. However, dual sources of income and greater amounts of income are associated with more spending on self. Implications for research and adolescent consumer socialization practices are discussed.


This paper investigates the general hypothesis that whether children's income is earned or received as a gift influences their use of this income. While there are several theoretical reasons to expect such a relationship and while such a finding could well have implications for later altruistic versus egoistic behaviors as well as willingness to delay gratifications as an adult, this hypothesis has been the subject of very little prior research. The present research is an exploratory investigation of young teenagers' reported income sources and their association with reported allocations of income to spending, saving, and gifts to others. Effects of sex, personal income amounts, and relative family income are also examined.


Income Sources

Prior research has provided descriptive data on how children acquire and use money. As McNeil (1979) notes, the major sources of income for children are allowances, earnings, and gifts. Conceptually, allowances are similar to earnings from outside jobs in that they are most commonly received in exchange for specified household chores. Ward, Wackman, and Wartella (1977) found that nearly all children from kindergarten through sixth grade had some discretionary personal income. Receiving money as a gift (usually from parents) was most common. While between 50 and 55 percent of children in each grade received money this way, the percentage varied from 41 percent for higher socioeconomic status children to 66 percent for lower SES children. Allowance was the second most frequent source of income and its receipt was positively correlated with age and socioeconomic status. Outside earnings were the third most common source of income and were most common for sixth graders and medium SES children. These findings support previous findings and reveal no major changes in children's sources of income over the past several decades in the United States (Dunsing 1956, 1960; Rogerson and Whieford 1960; Powell and Gover 1963; Marketing News 1983). There is some evidence that boys are more likely to have outside earnings and that girls are more likely to babysit as an outside job and are less likely to perform manual labor such as lawn mowing (Rogerson and Whiteford 1960; Powell and Gover 1963). No recent study has examined there differences however.


Children's sources of income are of less intrinsic interest than their expenditures. As Feldman (1963) points out, children can develop dysfunctional outlooks on money just as adults can. The dysfunctions include what may be thought of as miserliness, prodigality, and excessive giving to others to win their favor. By looking only at allocations of income to saving, spending, and giving to others it is not possible to determine whether children are using money dysfunctionally. Too much saving may indicate miserliness; too much spending may show prodigality; and too much giving may reflect an attempt to buy others' friendship, but what is too much is unclear.

As Herrmann (1970) observes, most of children's and teenager's purchases are discretionary items like toys, clothing, entertainment, and snacks, and since their families are expected to continue to provide the basic necessities, there is little incentive to save. McNeal (1979) reports that while preschoolers spend mainly on sweets, elementary school children buy more in the other categories and are more likely to purchase gifts for others. Gift-giving is more likely among teenagers. This is consistent with other findings that children become less materialistic with age (Marshall and Magruder 1960) as well as studies of children's wishes that show that wishes for material things decline with age (Ables 1972; Amatora 1957; Boynton 1936; Brook and Gordon 1979; Gray 1944; Guarnaccia and Vane 1944; Horrocks and Mussman 1973; Jersild, Market, and Jersild 1933; Jersild and Tasch 1949; Kuhn 1954; Speers 1937, 1939; Vandewiele 1980, 1981; Washburne 1932; Wheeler 1963; Wilson 1938; Winker 1949; Witty and Kopel 1939; Zeligs 1942). The latter studies also find some (not always consistent) differences by sex (Amatora 1957; Jersild, Market, and Jersild 1933; Jersild and Tasch 1949; Speers 1939; Vandewiele), by race (with black children showing greater desire for possessions - Gray 1944; Brooks and Gordon 1979), by socioeconomic status (with lower SES children being more inclined to wish for things--Jersild, Market, and Jersild 1933; Jersild and Tasch 1949; Witty and Kopel 1939), and by culture (with Americans being the most materialistic and foreign cultures and Amish children being less so--Kuhn 1954; Wheeler 1963; Vandewiele 1980, 1981).

The spending studies show that while girls and boys may be equally willing to spend money on themselves, girls are more likely to buy clothing and are less likely to instead spend money in the other purchase categories noted above (Powell and Gover 1963; Ryan 1966). The exact percentages spent, saved, and given to or spent on gifts for others varies considerably from study to study, but it generally appears that children of all ages from 5 through 20 spend more than they give and give more than they save (e.g., Dunsing 1956; Rogerson and Whiteford 1960).

Income Source Relationships to Expenditures

While various authors have suggested that giving children an allowance is likely to make them manage money better (usually interpreted as saving more and spending less - e.g., Dunsing 1956), the only direct evidence of allowance effects suggests that whether or not the child receives an allowance has no affect on expenditure patterns (Marshall 1964). However, the matching procedure used in this study is suspect (Campbell and Stanley 1963), and leaves the question open. Although no significance testing was done, it appears from Dunsing's (1960) data that high school girls use their outside earnings more than their allowances for gifts, savings, and purchases of clothing. However, the dearth and age of these available data suggest that more testing is needed to examine relationships between sources of income and expenditures. This was the PurPOSe of the present study.



What effect does giving children money as a gift (i.e., with no reciprocal tasks or gifts stipulated or understood as conditions for the money) have on their attitudes toward and use of this money? One possible hypothesis is that either because the adult giver acts as a role model or because such one-sided (unreciprocated) transfer causes guilt, children receiving income in this way would tend to be more generous in giving to others. It is also possible to argue however, that for the child who only receives income in this way, there is little appreciation of its scarcity or value to the giver and therefore little incentive to do anything but spend it. But considering the special nature of certain money gifts given to the child on occasions such as a birthday, Christmas, or graduation, this hypothesis seemed less plausible to us than the initial suggestions that gifts beget giving. Furthermore, as Hyde (1983) and Caplow (1982) demonstrate, such giving is likely to extend beyond simple two-way reciprocal exchange and involve gifts to others from whom money has not been received.

Children's earned income, whether from an allowance, odd jobs, regular part-time job, or summer job, may be seen as an embodiment of the Protestant Ethic. Frugality is a part of this ethic and suggests that saving should be most likely among children who earn their discretionary incomes. Frugality is also an egoistic trait and is therefore unlikely to be associated with altruism and giving to others. Even apart from questions of frugality, earning an income is more likely to allow children to feel that they deserve their money, that it is not easily replaced, and that there is less compulsion for them to spend St on others. Thus we hypothesized that when the adolescent ' s income is earned it is more likely to be saved and less likely to be used for gifts.

A third condition may exist when a child both earns and is given discretionary income. Because they earn part of their money, these children are apt to feel that they deserve that money. But because they also receive part of their money as a gift, this group is less likely than the preceding group to see such money as hard to replace. We would therefore expect this group to be most likely to spend money on themselves and somewhere between the other two groups in propensities to save and Rive to others.


The sample consisted of all 122 seventh graters in a single junior high school in suburban Salt Lake City, a community of approximately one million people. Sixty of these students were 13 and sixty-two were 14 years old during the Spring when the data were collected. Fifty-seven were males and sixty-five were females. It was not possible to collect socioeconomic data on the sample, but families in the area are heavily blue collar. The homogeneity of the community is shown in responses to a relative income question to which 72 percent said their family's income was about the same amount of money as the families of others they knew versus 14 percent each saying their family had more or less money.



The median reported amount of money these children had to save or spend each month was $16.50. The most frequent source of this income was as a gift from parents (70.5%), followed by babysitting (52.5%), allowance (41.8%), summer job (33.6%), gifts from relatives (23.0%), and present job (14.8%). Cross-tabulating these sources of income shows that 16 (13.1%) receive money only from parents or relatives (exclusive of allowances), 32 (26.2%) receive money only from the other (earning) means, and 74 (60.7%) receive both gift and earning types of income.

There were no age differences in sources or amounts of the child's income, but sex was related to sources of income and relative family income was related to amount of the child's income. According to Chi-square tests (1 d.f., corrected for continuity, alpha = .05), girls were less likely than boys to receive money from relatives (15% vs. 32%) or a summer job (22% vs. 48%), but were more likely to receive money from babysitting (78% vs. 23%). While relative family income did not significantly affect sources of income, it was significantly positively related to amount of reported income (F 6.14, p < .0029). Those who reported that their family had less money than friends' families had an average personal income of $7.97 per month versus $17.47 for those whose relative family income was seen to be about the same as friends' and $21.18 for those reporting greater family income than friends'. Income amounts were not significantly different depending upon whether income was earned. received as a gift, or both.


In terms of expenditures, the overall sample reported spending 38.9% of their incomes, saving 31.5%, and giving others (mostly via non-monetary gifts) the remaining 29.6%. These allocations did not differ by age or sex, but the allocation to spending was marginally different between the three relative income groups (F = 2.77, p< .0667). Using an alpha level of .10, Scheffe's multiple contrast test shows that the lower income group's allocation of 17.8% of income to spending on self is significantly lower than either the average income group's 39.7% or the higher income group's 38.7%. This would suggest that allocations to one or both of the remaining categories of saving and giving should be higher for the lower income group. But while it is nominally higher in both of these categories, the analyses of variance for these allocations to not differ among the three income groups at alpha = .10. Since we are analyzing proportional allocations rather than dollar allocations, the lesser personal spending by the lower relative income group suggests that they may be more driven by the self-abnegating Protestant Ethic in their income allocations.

Expenditures were also measured by asking in which of 22 product or service categories the child had made purchases with personal income. These responses as well as any significant sex or relative income effects (there were no significant age effects) are shown in Table 1. In general, the most popular expenditure category was food, followed by entertainment, clothing, and major durables. Females were more likely to report spending money on candy or gum, shoes, and school activities, while males reported spending more money than females on videogames (home or arcade), bicycles, skiing and camping equipment, Walkman-type stereos, skate boards, and motorcycles. Those who reported having less family income than friends, also reported a lesser amount of purchasing of candy or gum, shoes, and skateboards, but are more likely to have spent money on a bicycle, perhaps because they were less likely to receive one as a gift.



The association between type of income and proportional allocations of income by these adolescents is shown in Table 2. While giving did not differ between income source groups, both spending and saving did. Scheffe multiple contrast tests (alpha - .05) reveal that those with both types of income spent a greater proportion of their income and that those who earned all of their income saved a greater proportion of their income compared to the other two groups.



Although income amount was not significantly related to income source, it is also possible that amount of income effects allocations to giving, spending and saving. Correlations between amount of income and proportional allocations of that income show reasonably strong and significant (p < .001) relationships in each case. Income correlates .48 with allocation to spending, -.33 With allocation to saving, and -.44 with allocations X giving.


It was hypothesized that those with earned income would tend to save more and give less than others. The observed pattern of expenditures supports this hypothesis, although differences between groups in giving allocations were not significant. It was also hypothesized that those relying on gift income would be more inclined to allocate income to gifts and less inclined to allocate income to saving than others. As just noted, giving did not significantly differ between groups, even though mean proportional allocations displayed a pattern consistent With this hypothesis. The gift income group did save proportionately less money than the earned income group. The dual income source group was hypothesized to be most likely to spend money on themselves, and this was Found to clearly be the case.

Several cautions are in order in interpreting these results. In Addition to the limited nature of the present sample, it is important to recognize that no causal inferences are warranted from these results. For instance, the dual income source group may not spend more on themselves because of their income sources. It may be instead that because of their desire to spend more on themselves they are more likely than those relying solely on gift income to seek additional sources of income. However, because receipt of money from parents and relatives Is less discretionary than earning money, it is less likely that such a reverse causal flow could explain the relationship fount between earned income and saving. that is, it seems less plausible that those with a greater desire to save will seek earnings and reject gifts.

Because there was no significant difference in the dollar incomes reported for the three income source groups, the conclusions drawn from the proportional allocations also apply to dollar expenditures by the three groups. The relationships of sex and relative family income to income sources and expenditures should be considered however. Although males and females differed in the types of jobs and specific sources of gift income they received, they did not differ in regard to whether their income was earned, a gift, or both. The same is true of the three relative income groups, but the lower perceived family income group spent less of their own incomes an themselves. This may appear to indicate that the lower perceived family income group, at least at these ages, is less self-indulgent. But the finding that this group is also the most likely to have spent money On a bicycle, may suggest that their savings are to buy products for themselves that the other groups may be more likely to receive as gifts. The intended purpose of savings as well differences in non-monetary gift receipt are dimensions that need to be investigated in future research. In any case, the fact that relative perceived family income is related to expenditures but hot to sources of income, suggests that the findings in Table 2 are independent of relative family income effects. In addition, absolute child income amount is associated with greater proportional allocation to spending and lesser proportional allocation to both Saving and giving. But since income source is also independent of child income amount, the findings in Table 2 cannot be explained by these income effects either. The non-hypothesized effects of amount of children's incomes are also worthy of further investigation to assess the causality of these relationships.

With these cautions in mind, several tentative conclusions emerge from the present study:

1. Giving money to children does not foster greater spending on self. Dunsing's (1956) pejorative reference to such as income as "dole does not appear warranted. If anything, such giving seems to beget more giving by the adolescent.

2. Earning money as a sole source of income is associated with a greater tendency to save. It seems that such earnings may encourage the Protestant ethic to work and save for the future.

3. The most self indulgent group consists of those who both earn and receive income. Further research is needed to determine whether the hypothesized explanation that this group feels they deserve money and that it is not scarce, is correct.

4. Higher personal income children tend to spend a greater portion of income and save as well as give a smaller proportion than lower income children. Further research is needed to determine whether this is a social class effect related to the higher relative family incomes of children with higher personal incomes or an effect of the amount of their personal income per se.

In addition to broader investigations of these hypotheses, future research should examine age-related differences over a broader range of ages. All findings should be examined in a broader geographic context as well.

If these hypotheses receive further support, a curious dilemma exists. Assuming that saving and giving to others are more rewarding and socially beneficial than spending on self (Rimland 1982), the three hypotheses together suggest that children should either earn or receive money as a gift, but not both, and not too much. However, it is undoubtedly too simplistic to assume that spending money on self is the least desirable of the three choices investigated. Ultimately the goals behind such expenditures should be investigated to determine whether the expenditure is seen as an end in itself (in order to have the object or experience - "terminal materialism" in Csikszentmihalyi and Rochberg-Halton's [1978] words) or as a means to other ends (in order to do things using the object or experience acquired-"instrumental materialism-). Much remains to be done in investigating these Issues (see Belk 1984), but hopefully this study is a beginning in the examination of childhood influences on how we choose to utilize our money.


Ables, Billie (1972), "The Three Wishes of Latency Age Children, Developmental Psychology, 6, 186.

Amatora, Sister M. (1957), "Free Expression of Adolescents' Interests, Genetic Psychology Monographs, 55, 173-219.

Belk, Russell W. (1984), "Three Scales to Measure Constructs Related to Materialism: Reliability, Validity, and Relationships to Measures of Happiness," in Thomas Kinnear (ed . ), Advances in Consumer Research, 11, Ann Arbor: Association for Consumer Research. 291-297.

Boynton, Paul L. (1936), "The Wishes of Elementary School Children, Peabody Journal of Education, 13, 165-174.

Brook, Judith S. and Ann S. Gordon (1979), "Wishes of Adolescents, Psychological Reports, 44, 593-594.

Campbell, Donald T. and Julian C. Stanley (1963), "Experimental and Quasi-Experimental Designs for Research," in Normal L. Gage (et.), Handbook of Research on Teaching, Chicago: Rand McNally & Company.

Caplow, Theodore(1982),"Christmas Gifts and Kin Networks," American Sociological Review, 47 (3), 383-392.

Csikszentmihalyi, Mihalyi and Eugene Rochberg-Halton (1978), Reflections on Materialism,- University of Chicago Magazine, 70, 6-15.

Dunsing, Marilyn (1956), "Spending Money of Adolescents," Journal of Home Economics, 48 (June), 405-409.

Dunsing, Marilyn (1960), "Money-Management Experiences of High School Students," Journal of Home Economics, 52 (November), 756-762.

Feldman, Frances L. (1963), Money: an Index to Personal Problems in Adolescents, Marriage and Family Living, 25 (August), 364-367.

Gray, Susan (1944), "The Wishes of Negro School Children," Journal of Genetic Psychology, 64, 225-237.

Guarnaccia, Vincent J. and Julia R. Vane (1979), Children's Wishes," Journal of School Psychology, 17, 126-130.

Herrmann, Robert 0. (1970), Today's Young Adults as Consumers," Journal of Consumer Affairs, 4 (Spring), 19-30.

Horrocks, John E. and Milton C. Mussman (1973), Developmental Trends in Wishes, Confidence, and the Sense of Personal Control from Childhood to Middle Maturity, Journal of Psychology, 84, 241-252.

Hyde, Lewis (1983), The Gift: Imagination and the Erotic Life of Property, New York: Vintage Books.

Jersild, Arthur T., F. V. Market and C. L. Jersild (1933), "Children's Fears, Dreams, Wishes, Daydreams, Likes, Dislikes, Pleasant and Unpleasant Memories," Child Development Monographs, 12.

Jersild, Arthur T. and Ruth J. Tasch (1949), Children's Interests and What They Suggest for Education, New York: Columbia University Bureau of Publications.

Kuhn, M. H. (1954), "Factors in Personality: Socio-Cultural Determinants as Seen Through the Amish, in H. L. Hau (ed. ), Aspects of Culture and Personality, New York: Abelard-Schumann.

McNeil, James U. (1979), Children as Consumers: A Review," Journal of the Academy of Marketing Science, 7 (Fall), 346-359.

Marketing News (1983), "Projective Research Techniques Extract Valuable Market Data from Children, Marketing News (January 21), 19-20.

Marshall, Hellen R. (1964), "The Relation of Giving Children an Allowance to Children's Money Knowledge and Responsibility and to Other Practices of Parents," Journal of Genetic Psychology, 104, 35-51.

Marshall, Hallen R. and Lucille Magruder (1960), Relations Between Parent Money Education Practices and Children's Knowledge and Use of Money, Child Development, 31, 253-286.

Powell, Kathryn S. and David A. Gover (1963), "The Adolescent as a Consumer: Facts and Implications, Marriage and Family Living, 25 (August), 359-364.

Rimland, Bernard (1982), The Altruism Paradox," Psycho- logical Reports, 51, 521-522.

Rogerson, Lannie C. and Emma B. Whiteford (1960), Money Experiences of Ninth-Grade Pupils," Journal of Home Economics, 52 (1), 44-45.

Ryan, Mary S. (1966), Clothing: A Study in Human Behavior, New York: Holt, Rinehart and Winston.

Speers, George S. (1937), Wishes, Fears, Interests, and Identifications of Delinquent Boys, Child Development, 8, 289-294.

Speers, George S. (1939), "Oral and Written Wishes of Rural and City School Children," Child Development, 10, 151-155.

Vandewiele, Michel (1980), Wishes of Senegalese Secondary School Students," Psychological Reports, 46, 256-258.

Vandewiele, Michel (1981), Wishes of Adolescents in Senegal, Psychological Reports, 48, 695-698.

Ward, Scott, Daniel B. Wackman and Ellen Wartella (1977), How Children Learn to Buy: The Development of Information Processing Skills, Beverly Hills: Sage Publications.

Washburne, John N. (1932), "The Impulsions of Adolescents as Revealed by Their Written Wishes," Journal of Juvenile Research, 16, 193-212.

Wheeler, D. R. (1963), "Expressed Wishes of Children," Journal of Genetic Psychology, 102, 75-81.

Wilson, Frank T. (1938), Verbally Expressed Wishes of Children and College Women, Journal of Psychology, 5, 91-105.

Winker, James B. (1949), "Age Trends and Sex Differences in the Wishes, Identifications, Activities, and Fears of Children, Child Development, 20, 191-200.

Witty, Paul and David Kopel (1939), The Dreams and Wishes of Elementary School Children," Journal of Educational Psychology, 199-205.

Zeligs, R. (1942), Children's Wishes, Journal of Applied Psychology, 26, 231-240.