NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) and National Research Council (US) Committee on an Aging Society. Productive Roles in an Older Society. Washington (DC): National Academies Press (US); 1986.

Cover of Productive Roles in an Older Society

Productive Roles in an Older Society.

Show details

The Economics of Volunteerism: A Review

Carol Jusenius Romero

The role of the volunteer in America has become a topic of public interest, mainly because of the confluence of three trends. First, all levels of government are encouraging volunteer work as a substitute for government's declining role in the provision of social services. Second, women—historically the source of much volunteerism in America—have been entering the work force in increasing numbers. Finally, growth in the size of the retired population has led to discussion of ways to use the time, talents, and energy of older people. But although there is interest in understanding and encouraging volunteerism, relatively little is known about the activity.

Volunteerism is often perceived as donations of time or labor but it can also include donations of money or goods, and in each of these forms it varies widely along several dimensions.

Volunteer activities take place in settings that have different degrees of structure and formality. Such activities may be quite informal, occurring in a casual, unstructured way; or they may be organized but not performed within a formally structured group; or both the activity and the group may be structured and formally organized. To illustrate: volunteer activities range from a neighbor helping a neighbor, to a group of parents organizing to clean a playground, to a formal volunteer group, such as the PTA or Volunteers of America, performing formally specified functions.

Volunteer activities also vary in their time requirements. People may participate in them from 1 to 52 weeks in a year, any day of the week, and for any number of hours. Some activities are spread out over weeks or years; others are concentrated (for example, within communities after a disaster strikes).

Finally, volunteering, like other activities, usually has some unpleasant aspects, and for people to be willing to undertake unpleasant tasks they must be compensated in some fashion. Compensation for volunteering can take a variety of forms. Sometimes a volunteer receives only psychic, personal satisfaction, or there may be some public recognition. A range of monetary compensations exists as well. Some people receive income tax deductions for their donations. There are also more direct means of monetary compensation. For example, the Foster Grandparent program gives participants a stipend to supplement their income, and people who work for the Salvation Army earn their entire income from this activity. To the extent that the earnings of these individuals are less than could be obtained in another type of work, this differential is equivalent to donating money or time.

A key question of this paper is why people volunteer. Is altruism the primary motivation, or do people volunteer out of some form of self-interest? Are people motivated ''to volunteer" in general, or are they motivated to volunteer for particular types of activities? Such questions highlight an important policy area. If volunteerism is to be promoted to meet social needs, then the mechanisms that encourage volunteerism must be understood.

Survey results suggest a range of reasons for volunteering. According to the data in Table 1, nearly 30 percent of people who volunteer time "thought [they] would enjoy doing the work; feel needed." Almost 25 percent volunteered because a child, relative, or friend was involved, and 11 percent reported that they wanted work experience. Generally, motivating forces include children, relatives, and friends; religious beliefs; political or social concerns; personal history or interest in the activity; social pressures; and a desire to keep busy, feel productive, meet new people, or interact socially.

TABLE 1. Reasons for First Becoming Involved in a Volunteer Activity.


Reasons for First Becoming Involved in a Volunteer Activity.

The large percentage of multiple responses shown in the table suggests that individuals volunteer for a combination of reasons: No one motive, taken alone, is likely to be sufficient. After all, not every parent volunteers for a child-related activity and not every religious person volunteers for a church-synagogue or social welfare cause.

This paper stems from my work dealing with people's willingness to volunteer for formal organizations. The data and research cited here reflect that focus, although as indicated later many of the issues are applicable to other forms of volunteering as well. The paper is divided into four sections. The first presents data on Americans' unpaid work for formal volunteer organizations; it provides a backdrop to the analyses that follow. The next section reviews a number of economic analyses of this type of volunteering to determine how the motivation to volunteer has been analyzed conceptually and empirically and why motivations matter for public policy. The third section suggests areas for future research on volunteerism, and the final section gives a brief summary and offers some concluding thoughts.

Work Without Pay for Formal Organizations

About 75 percent of the work people do without pay is for formal organizations.1 The amount of time people spend in this way and the types of organizations for which they volunteer are discussed below.

Amount of Time

The proportion of Americans who volunteer has risen somewhat over the past 20 years. In 1965, 21 percent of women and 15 percent of men did some volunteer work. By 1981, participation in volunteer activities had risen to 28 percent among women and 30 percent among men.2

Although these figures may suggest a fairly widespread involvement in volunteer activities among Americans, a few cautions are in order. First, the proportions are small compared to the percentages of men and women who are either working or looking for work (77 percent of men and 53 percent of women are currently in the work force). Second, few people volunteer on a continual basis. While cross-sectional data indicate that about 25 percent of middle-aged women volunteer in a given year, longitudinal data reveal that fewer than 15 percent volunteer year after year.3 Looking at it another way, fewer than 50 percent of women volunteers in one year had volunteered more than once in the recent past.

Other data on middle-aged women who volunteer illustrate some of the range of variability in the amount of time spent volunteering. Of all women who volunteer, the average participation time is 24 weeks per year; approximately 25 percent of those women, however, spend between 49 and 52 weeks in this activity. The average volunteer time per week for middle-aged women is 6 hours. More than 60 percent of women spend between 2 and 7 hours, 10 percent spend 1 hour, and another 10 percent spend 15 or more hours.3

Data on men close to or at retirement age reveal similar variations in the amount of time spent volunteering.4 The 20 percent who volunteer spend, on the average, 24 weeks per year in the activity. About one-third of the volunteers participate 49 to 52 weeks and another one-third participate for 4 or fewer weeks. During the weeks of volunteering the men average 7 hours in this activity. More than 60 percent of the men spend between 2 and 7 hours, about 10 percent spend 1 hour, and more than 10 percent spend 15 or more hours.

Types of Volunteer Activities

There are many different types of formal organizations for which people volunteer. Unfortunately, however, the available data do not permit distinguishing among organizations according to measures of either their social usefulness or the extent to which volunteers must engage in unpleasant tasks to carry out the organizations' goals. Still, these data do illustrate the broad range of volunteer activities and the fact that different activities seem to appeal to different types of people.

Formal volunteer activities range from the PTA to soup kitchens to political and social causes. For instance, in 1981, 10 percent of the adult population volunteered for religious activities, 12 percent volunteered at hospitals or in other health-related activities, another 12 percent volunteered for school-related activities, 1 percent worked as poll watchers, 4 percent worked as campaign workers—the list goes on.5

It appears that sex and age play a role in the types of organizations for which people volunteer.6 Men are more likely than women to participate in recreational and work-related activities; women are more likely to participate in health, educational, and religious activities. Further, adults are more likely than teenagers to volunteer for political and work-related activities, while teenagers are more likely to volunteer for health, recreational and educational activities.

Longitudinal data confirm that as people age, they change the type of organization for which they volunteer.7 Among women 37 to 51 years of age, 28 percent of the volunteers participated in school activities and in such groups as Boy Scouts or Girl Scouts; 33 percent did church-related work; and another 30 percent volunteered for hospitals, clinics, major community drives, and other social welfare or civic causes. Five years later, the women volunteers (aged 42 to 56) had somewhat shifted their patterns of volunteering. Fewer volunteered for schools and other child-related activities (less than 20 percent), and more volunteered for both church-related activities (over 40 percent) and for hospitals, clinics, community drives, and civic causes (over 30 percent).

Taken together, these data suggest a richness and complexity in the amount and type of unpaid work that occurs in this society. The next section reviews analyses of the motivations that underlie the decision to volunteer.

Economic Anaylyses of the Reasons for Volunteering

The economist's approach to analyzing the motivation to volunteer is twofold: first, develop a behavioral model, and, second, derive hypotheses from the model and test them empirically. The empirical tests—typically some form of regression analysis—seek to disentangle the various determinants of people's actions. For example, a text might ask how parents' participation in volunteer activities is influenced by the presence of children in the family, while simultaneously taking into account the parents' educational levels, incomes, and work patterns.

The research reviewed here has been selected to illustrate the economist's approach to the issue of why people volunteer. And as such, these illustrations admittedly are not without limitations. Nevertheless, they can be used to stimulate thought and discussion about areas of study and suggest directions for future research.

In the studies described below, two types of reasons for volunteering are considered: (1) satisfaction derived from helping others and (2) direct personal benefit, independent of the extent to which others are helped. These research efforts examine whether altruism or personal benefits alone underlie an individual's motivation to volunteer, how family considerations influence a person's volunteerism, and the reasons why motivations of volunteerism are important to public policy.

Volunteering Motivated by Altruism

One of the first modern-day economic discussions of volunteerism is found in an article by Kenneth Boulding.8 Boulding suggested that although philanthropy—or volunteerism—is often viewed as a transfer payment, for which there is no obvious reciprocation, from one individual to another, it may be more useful to look at the benefits received by the giver for being philanthropic. Boulding proposed two ways in which the giver may be compensated: (1) ''a certain glow of emotional virtue" for having been philanthropic and (2) recognition that there is a "common identity in humanity."9

These two ways would now be called "altruism" by economists.10 The reward for volunteering motivated by altruism is simply a good feeling for having been helpful to others. However, more direct rewards for volunteering also exist (as indicated earlier by the data in Table 1 on the reasons people say they volunteered). Some of the research efforts involving more direct rewards for volunteerism are discussed below.

Experience As Compensation for Volunteering

Two studies11 tested the hypothesis that volunteerism is a form of personal investment; that is, people volunteer to enhance their future job and income prospects. Under this view, the fact that volunteerism benefits others is coincidental; it is not an important motivation.

The first article argues that women volunteer as a way of preparing for (re)entry into the job market—to gain experience and information about possible future jobs.12 This hypothesis was tested on a sample of women who had been in graduate school between 1945 and 1951 using two dependent variables: hours of volunteering in a year and hours of volunteering for organizations other than professional societies.

Two of the study's independent variables were intended to represent "a desire for training" in preparation for (re)entry into the work force: (1) whether or not the woman was planning to work in the near future and (2) the average comparable market wage of the volunteer work that was done (a measure of the skill content of the activity). The results of the study showed that both variables were positively related to volunteerism and statistically significant.13

The author interpreted her results as follows:

Women, at least those in the education, income and age group most actively participating in volunteer work, appear to be doing so in part for their own families but also to a great extent for themselves—to build and maintain their human capital to aid in the job search.

. . . the volunteer organization remains one of the few institutions accessible to these women for the important activity of maintaining or building human capital.14

The results seem plausible for women of this particular age and educational level. At the same time, however, there are some limitations to the empirical work that call the results into question.

First, the variable "comparable market wage of volunteer work" need not be interpreted as an indicator of a desire for training, but instead may indicate the status content and desirability of different types of volunteer work. Women volunteering for desirable, high-status activities may spend more hours volunteering than women who are in less desirable, lower-status organizations or positions.

Second, the test of the variable "plans to work in the future" is not "clean." This variable has been constructed in such a way that women who are out of the labor force and plan to look for work are compared to two groups combined: those who do not plan to work and those who are already working. A ''cleaner" test of the hypothesis would have been to include only the former—those not employed and not planning to look for work—in the reference group.

The inclusion of women who are already employed confuses the issue. Because they are employed, they may supply fewer hours of volunteer labor, but the reason for this reduction is not clear: either they may not desire to (re)build their skills or they simply may have less time to volunteer. There is also no information about the relative number of employed and unemployed women. Because of these limitations, it is impossible to state with certainty that volunteerism is used by women as a vehicle for easing (re)entry into the job market.

In the second article, Menchik and Weisbrod tested volunteerism as a form of investment by positing that people may volunteer because it "raises one's future income by providing work experience and potentially valuable contacts."15 An alternative hypothesis was also posed: People may consider volunteer time to be a type of consumer good, something to be enjoyed or consumed. In contrast to the investment hypothesis, the consumption hypothesis posits that people volunteer because they enjoy either the activity or the product of the activity. Under this view, there are no negative aspects to volunteering for which compensation is necessary.

These hypotheses were tested with a sample of working people drawn from a national survey. Nearly two-thirds of those surveyed were married men with a wife at home full-time; the remainder were single heads of households, many of whom were likely to be women. To allow for possible differences among types of volunteer activities, there were several dependent variables, all assessed in terms of hours: (1) total volunteer work, (2) volunteer work for elementary and secondary education, (3) volunteer work for higher education, (4) volunteer work for welfare, and (5) volunteer work for natural resources (environment).

The competing hypotheses were tested empirically by including a person's wages as an independent variable in the regression. Wages measure the opportunity cost of volunteering; that is, the money forgone by spending time not working for pay. If people view volunteerism as a form of consumption, an inverse relationship between wages and volunteering would be expected. People make trade-offs between working (earning wages) and volunteering; the higher the wage rate, the less likely a person is to spend time volunteering rather than working. By contrast, if people view volunteerism as a form of investment, a positive relationship between wages and volunteerism would be expected. The reasoning is as follows: "If those with greater ability to benefit from volunteer work also earn more per hour in the absence of volunteer work, we have a situation in which higher wage workers may volunteer more hours than lower wage workers. . . ."16

The results of the study indicated that the relationship between wages and hours of volunteerism is positive, which is consistent with the hypothesis that volunteerism is a form of investment. However, this interpretation of the results is questionable, because there are problems with the underlying theory. First, rather than expecting a positive relationship between wages and volunteerism under the investment hypothesis, it would seem more reasonable to expect an inverse relationship. People with low wages might have a greater desire to enhance their future earnings than people with high wages. To the extent that volunteering is a way of increasing future earnings, low-wage earners could be expected to volunteer more (not less) than high-wage earners.

A second problem concerns the empirical test of the alternative hypotheses. Using a wage variable assumes that the only way working people spend time volunteering is by giving up time for paid employment. However, casual observation suggests that people also volunteer after work, in the evenings, or on weekends. In other words, time for volunteering can also come from time that would otherwise be devoted to either housework or leisure.

The Effect of Family Obligations on Volunteering

This author's research raised more general questions about the factors influencing volunteerism. One concern was to determine the circumstances under which a person would volunteer, notwithstanding household obligations.17 Specific research questions focused on the relationship between volunteerism and the movement of older Americans into retirement.

The research built on previous work analyzing how individuals, operating within a family setting, make decisions about spending their time (or money).18 Each family member divides his or her time among three major categories: (1) work outside the home, (2) work inside the home, and (3) leisure. The precise way time is allocated to these categories mirrors the entire family's preferences for goods bought in the market, goods produced at home, and leisure. By selecting the combination that maximizes its utility (or happiness), the family simultaneously allocates its members' time.

Maximizing household utility may require that a husband and wife use their time differently over the family's life cycle, particularly with regard to their division of market and nonmarket (or home) work and depending upon their comparative advantages in these two activities. For example, over its life cycle a family will generally be willing to have the woman spend proportionately more time working full-or part-time at home, since men are typically more productive—in the sense of earning more—in the job market than in home work.

Including volunteerism as a way people may spend their time required the addition of detail to this model. The first argument to be considered was that people must be compensated for volunteering. Such compensation may be:

  • direct increased personal satisfaction due to "a feeling of virtue," meeting new people, interacting socially, feeling productive, or keeping busy;
  • increased welfare of another family member (This case recognized that a person operates within a family and that the output from one person's work increases the goods and services that another family member may enjoy.), and
  • increased welfare of people outside the family. (Of the three motivations for volunteering, this is the least tangible and represents the closest approximation to altruism.)

Second, it was posited that not all household decisions regarding the use of time (or money) are equally important. Those involving smaller portions of a household's time (or money) are usually less important than those involving larger portions and are made independently of major household decisions. (For example, a decision to buy shoes or a tablecloth will be made independently of a decision to buy or rent a house.)

Decisions about the amount and timing of work outside and inside the home (including child care) are of major importance to a family. Entering the job market—or raising children—typically requires a substantial time commitment on the part of household members. Also, the lack of flexibility usually found in the number and timing of hours of paid employment and the continual nature of child care make these decisions more complex.

By contrast, the decision to volunteer may be either of major or minor importance to a family. It need not consume a sizable portion of a household's time (or income), and the amount and timing of volunteerism is generally flexible.

Volunteerism, then, was hypothesized to assume a smaller role in a family's life when the benefits accrue only to people outside the family. Families would be unlikely to reduce their income, their home-produced goods, or their leisure by sizable amounts in order to help people not in the household. Only after the demands of household members have been satisfied would families volunteer in this way.

However, if a family member (such as a child) benefits from the volunteering, it was hypothesized that volunteerism would assume greater importance and in fact would become a form of homework. In this case, a family may be willing to (1) alter its original combination of market-and home-produced goods and leisure and (2) consume either fewer market goods or less leisure in order to have more home-produced goods.

This view of volunteerism suggested some specific questions concerning the reasons why people volunteer:

  • In which situations are people willing to give up work time to volunteer?
  • Are people more likely to volunteer when a family member (including the volunteer) receives some of the benefits from that volunteerism?
  • Are people who live in areas with concentrations of people outside the family who need assistance (such as in cities) more likely to volunteer than those who live outside such areas?

Regarding retirement issues specifically:

  • Are retired people who have leisure time more likely to volunteer than those who are fully employed? (Economists would argue that because the opportunity cost of their time is smaller than among those who are employed, they would be more likely to volunteer.)
  • Are people who have an independent source of income sufficient to meet household obligations—such as retired people with sizable pensions and savings—more likely to spend time volunteering than those whose major source of income comes from their current job?

The data used to address these questions were collected in the National Longitudinal Surveys of older men and older women.19 For the men, there was one dependent variable: whether or not they volunteered in 1978. The women's data permitted differentiating among types of volunteer activities. Four dependent variables were used: whether or not the women volunteered (1) generally, (2) for a church activity, (3) for a child-related activity, and (4) for a social welfare or civic activity.20

General results for the men indicated that volunteerism is greater among the more educated and among those who live outside urban areas and in places with low unemployment rates; older men do not appear to volunteer more in areas with greater social needs. Also, at this stage of their lives (when they are in their late 50s to early 70s), men are not influenced to volunteer by the presence of children in the family, possibly because the children are usually older.

General results for the women indicated that volunteerism is greater among the more educated and among those with a history of volunteering, but it does not differ between urban and nonurban areas or between areas of high and low unemployment. The influence of children on women's volunteering differs by the type of volunteering that is done. Children raise the probability that a woman will participate in child-related activities, but children do not affect the probability that a woman will volunteer for other types of activities.

The detailed investigations of volunteerism among women indicated that it is important to distinguish among types of volunteer work. The relationship between volunteerism and several of the independent variables (that is, in addition to the children variable already mentioned) differed by the type of activity. Of special importance, given the questions raised earlier, was the result that the effect of employment varies by type of activity. Women who volunteer for child-related activities are likely to give up work time to do so. Women who volunteer for church-related activities tend to work fewer hours during the week, but they do not necessarily work fewer weeks in a year. Finally, the amount of time women work throughout the year and during any given week does not appear to influence their participation in social welfare or civic activities.

Specific results on the retirement issues indicated that policies affecting the amount of time people work are likely to have different effects on the participation of men and women in volunteer activities. Men are unlikely to be affected; their volunteerism was not found to depend upon their current employment or retirement status. Policies that permit moving from full-to part-time work, however, may increase women's participation in church-related activities.

Other results on the retirement issues indicated that policies concerning the amount of income people have during retirement may affect men and women in different ways. Older men with lower levels of income from assets volunteer less. Older women with lower levels of such income are less likely to volunteer for social welfare causes; however, income was not found to be related to women's participation in either church-or child-related activities.

These results raise several questions. For example, distinguishing among types of volunteer activities is important for studies of women volunteers; it is therefore plausible that such distinctions are equally important for men. Also, it would be useful to explore why several factors seem to have different effects on the participation of women in the various volunteer activities. For instance, is volunteer time more flexible in social welfare activities so that people do not need to trade off paid employment to participate? Alternatively, does volunteering for church-related efforts carry with it a special form of gratification? Are people willing to make some trade-offs with paid employment to gain this kind of gratification, whereas they would be unwilling to make the same trade-offs for social welfare activities?

Policy Implications of the Reasons for Volunteering

The question of whether people volunteer to gain some benefit for themselves or to help others outside the family carries with it certain policy implications. The reasons why people volunteer determine in part the extent to which volunteerism could substitute for reduced government provision of social services. If people volunteer because of social needs, then governments could reduce expenditures in many areas with the expectation that volunteers would offset this reduction, at least to some extent. However, if people volunteer for some personal benefit, then a reduction in government expenditures would elicit more volunteerism only in those areas that offer the greatest benefits to the volunteers. In this case, governments would need to be selective in their actions if adequate levels of all social services were to be maintained.

Menchik and Weisbrod tested this issue empirically through the question: Do government expenditures "crowd out" volunteerism?21 (Crowding out implies that there is a finite amount of "need" and that government involvement effectively substitutes for involvement on the part of private citizens.) The authors argued that if individuals volunteer because they expect to derive personal benefits in the future as a result of the activity (the investment hypothesis given earlier), then government involvement has no predictable effect on volunteerism. However, if volunteerism is undertaken for its own sake (the consumption hypothesis), then governments could conceivably crowd out private sector volunteerism. Under the consumption hypothesis, individuals would be less likely to volunteer for a cause if at least some of the needs of that cause were being met by government expenditures. To test for the possibility of government crowding out volunteerism, Menchik and Weisbrod included an independent variable in the regression: per capita state and local government expenditures in different program areas in the state where the individual lived.

The results on the crowding out issue were mixed. In the areas of elementary and secondary education and natural resources, greater government expenditures were associated with fewer hours of volunteering. But there was no significant relationship between government expenditures and volunteering in the areas of higher education and welfare. The inconclusive nature of the test results led the authors to suggest that the crowding-out issue merits further study.

It should be noted, however, that there is a limitation to the theoretical approach that was taken in this study. Under the consumption hypothesis, people may volunteer either because they enjoy the process—the activity itself—or because they enjoy the product of that activity. (For example, people may volunteer to clean a park because they enjoy being out-of-doors or because they enjoy a clean park.) Even under the consumption hypothesis, the extent to which government involvement could crowd out volunteerism might differ depending upon why people volunteer. (If the park is already clean and people enjoy being outside, they may find other ways to both volunteer and be out-of-doors simultaneously.) The test of the crowding-out issue considered the case in which people volunteer to gain the product; it did not consider that people may enjoy the activity itself.

The results of the empirical test on crowding out are puzzling. No explanation was given as to why crowding out might occur in some program areas but not in others. Also, the authors did not explain how crowding out occurs given that (1) crowding out is only possible under the consumption hypothesis, and (2) their earlier results indicated that volunteerism is a form of investment, not consumption.

This seemingly contradictory result may be explainable. Menchik and Weisbrod neglected to consider that crowding out could occur even if people view volunteerism as a form of investment. Government involvement would presumably reduce the probability that either volunteer organizations would exist or that they would require large numbers of volunteers. And it is these organizations and these volunteer positions that provide the institutional setting for volunteerism.

In sum, the specifics of this approach are not particularly satisfying, but the question the authors have posed remains important. Do government expenditures substitute for volunteerism. and will a reduction of government involvement in social welfare and other areas call forth more volunteerism?

Lessons From Past Research

It is apparent from this review that economists do not agree on the motivations for volunteering. On some points their results are inconclusive. For example, while some studies may find that volunteerism is a form of investment, there are a sufficient number of questions about the research to cast doubt on the findings. On other points, such as the trade-off between work time and volunteering among women, the evidence is insufficient to state findings with certainty. Like most research questions, the issue needs to be investigated with more than one approach and one data set before it can be considered to be resolved.

On two points, however, the research is consistent. First, participation in formal volunteer organizations is largely a middle-and upper-class phenomenon. Most people who volunteer in this way have higher levels of education and income than those who do not. Second, an underlying view presented in the articles is that people must be compensated personally for volunteering.22 Altruism is unlikely to be a sufficient motivation, and it is not reasonable to expect people with more leisure time, such as those who are retired, necessarily to be interested in using their free time to volunteer.

Where Do We Go From Here?

This section suggests three categories of research in the area of volunteerism. Two focus on the individual volunteers: descriptive data on the nature and extent of people's participation in volunteer activities and analyses of their motivations to volunteer. The third category concerns volunteer organizations and includes analyses of the role of the ''voluntary sector" in the total economy.

Data on Volunteerism

At the outset of this paper, there was a description of the three ways people volunteer—donations of time, money, and goods—and of the range of variability possible in carrying out these activities (i.e., volunteerism occurs in settings that have different levels of organization and structure, its timing is often flexible, and there are different ways that people are remunerated).

Several data sources give an empirical content to this picture: for example, the 1973 Survey of Giving, the 1981 Gallup survey, and the National Longitudinal Surveys. These data bases and their findings, however, are not always readily available to researchers, policymakers, or program operators. The findings of these sources could be summarized to provide a useful inventory of volunteerism. This inventory could include documentation of the range of variability in the amount and types of volunteerism, analyses of the reasons for different findings among data bases, and a resulting rough set of summary statistics on volunteerism. To provide a complete picture some new data may also be required.

The purposes of this work are twofold: (1) to suggest both the areas in which volunteerism could potentially substitute for government expenditures and the groups of people who would most likely respond to an increased need for volunteers; and (2) to assist more sophisticated analyses (the data would clarify "volunteerism" when it is used as a dependent variable in conceptual or empirical research). This section offers some comments on the type of information that could be contained in such an inventory.

Information is available from several sources on the number of hours people spend volunteering, especially for formal organizations. Less information appears to be available on the amount of money and on the amount and type of goods they donate. For example, questions about money and goods were not asked in the National Longitudinal Surveys. In the 1973 Survey of Giving they were asked of a limited set of people; in other research the amount of a donation was not specified (as in the case of the 1981 Gallup study). Another neglected area seems to be informal donations of time, money, and goods. Although there are some data on the amount (in terms of time) and the kinds of unpaid labor that are donated to friends and neighbors, a complete picture would include data on the amount of informal donations of money and goods as well.

How donations of time, money, and goods are used is another issue that data collection might help to address. For instance, only a few sources document the extent to which people's volunteer time is devoted to (1) fund-raising events (so that others can be paid to do the work) and (2) the work directly (see, for example, the 1973 Survey of Giving and the 1981 Gallup). Less information is found on the ways in which donated goods are used, donations that range from old clothes given to Volunteers of America to residences left to universities after a person's death.

Another body of information concerns the timing of volunteerism. People volunteer during the weekdays, in the evenings and on weekends; they volunteer for short but concentrated periods of time and for longer, less intensive periods. Further, different types of activities seem to have different time arrangements. For example, participation in Little League activities occurs after school hours and is somewhat concentrated in time; donations to churches often occur once a week over the course of a year; and donations of time, money, and goods for Toys for Needy Children programs are concentrated in the holiday season. Detailed information on the timing of volunteerism in the form of unpaid labor specifically would improve estimates of the extent people trade off paid employment, homework, and leisure for volunteerism.

Finally, it has been mentioned that people are remunerated in different ways for their volunteering. Documenting the range of this variability could be useful if it can be proved more conclusively that people respond differently to different rewards. In addition, however, a special emphasis of data collecting should be the earnings of people who work for pay for organizations that rely on voluntary contributions. To the extent that these workers earn less than they could earn elsewhere, the pay differential is conceptually equivalent to a donation of money or time.

Motivations to Volunteer

Outlined below are some directions for research on the reasons why people volunteer. Four areas are considered: (1) the volunteer decision, (2) the amount and timing of volunteerism, (3) the interrelationship among the different ways people volunteer, and (4) the differences between formal and informal volunteering.

Some of the issues raised here could be investigated using the existing National Longitudinal Surveys cohort of older women. In 1974, 1976, and 1979, these women were asked a series of questions about their participation in formal volunteer organizations—the type of activity, the number of weeks per year and hours per week volunteered, and the type of positions held. Although this data base has been used extensively to address other issues, it has not been employed for investigations of volunteerism with the exception of this author's work.

The data set has a number of advantages: It is longitudinal and contains detailed information on the women's personal and family characteristics and their current and past work experiences, as well as some data on their attitudes toward, for example, work and appropriate roles for women. It also has some disadvantages—lack of information on donations of money and goods and the limited age/sex sample—but potentially these could be overcome. Also, additional questions could be added to future surveys of the same women, and the same questions could be asked in surveys of the other cohorts of the National Longitudinal Surveys: older men, younger men and women, and youth. REMOVE19

The general approach of this research should be oriented toward policy. It should consider what factors motivate people to volunteer and which of these factors could be influenced by public policy to encourage greater participation in voluntary activities. The initial goal would be to understand whether people decide to volunteer and then select an activity or whether people decide to volunteer for particular activities. Past research suggests that people decide to volunteer for specific activities. If certain volunteer services are more urgently needed than others, it is important to understand what motivates people to volunteer for different activities in order to provide those particular services.

A reasonable hypothesis is that although people may be ready or willing to volunteer, they will actually volunteer only if they can choose their own activity; ''being ready and willing" may be a necessary but not sufficient condition. As suggested by the literature review earlier in this paper, the sufficient condition may be a personal benefit that a volunteer derives from the activity. More work is needed to understand these motivations, work that uses the combined expertise of economists, sociologists, and psychologists.

This same issue exists for donations of money and goods. The existence of tax write-offs may make people willing to donate, but whether or not they do donate may depend upon the existence of organizations that meet some personal need or carry out some personal goal or belief. Although tax gains are one part of the compensation package, they are unlikely to comprise the entire package.

Another research issue—and an important policy area—concerns the amount and the timing of volunteerism. Alleviating or reducing some of society's problems, such as illiteracy and the medical needs of the elderly, will require a long-term effort; if volunteerism is to substitute for government expenditures and paid workers, then it must be sustainable over time. Such prolonged support could be achieved by relying on the same volunteers for long periods of time or by using an ever-changing work force of volunteers. But these "solutions" may in turn prove questionable. To the extent that volunteer organizations have high turnover among their volunteers, they will also require ongoing training and retraining capacities. This raises an organization's costs and thus reduces the efficiency with which it could carry out its activities. Analyses about the amount and timing of volunteerism would indicate whether continued reliance upon volunteers is an efficient way to provide services over long periods of time.

Data cited earlier indicate that few people volunteer year after year and that there is a broad range in the amount of time they donate. Also, casual observation suggests a greater willingness among people to do volunteer work for short, concentrated periods of time (such as in an emergency or during the holiday season) than for long, sustained periods. Or is this an illusion? Are the people who do the work during concentrated periods of time the same ones who volunteer over the course of a year? Alternatively, are there people who are only willing to volunteer at special times?

The amount of time people are willing to volunteer depends upon how much and what type of other activities they are willing to forgo—leisure pursuits, paid employment, or work inside the home. Similarly, for people to donate money or goods, they must be willing to give up alternative uses of that money and those goods. The amount and timing of a donation are likely to depend upon a household's income and the demands that are placed on it at different times of the year. For example, one could hypothesize that, among most families, sizable donations are unlikely to be made the day before a tax payment is due. Among low-income families in particular, even a small contribution is less likely to be made at Christmas or before school begins in the fall—times when there are desires and needs of one's own family to be met.

All of the discussion thus far on motivations has dealt with research that focuses on each form of volunteering (donations of time, money, and goods) individually. There is also a need for research that looks at these forms collectively and asks about their interrelationships. Do people interchange donations of money, time, and goods, or are these forms complementary?23

One way in which this issue is important is in the context of the increasing participation by women in the work force. Historically, the volunteer time of women has been a major part of volunteerism in America. But what happens when women move into paid employment? Do they continue their volunteer role to some extent, or do they change to giving money and goods?

One research result cited earlier was that the participation of women in the work force does not affect the likelihood that they will volunteer for social welfare or civic activities. This finding, which takes a cross-sectional view, is somewhat surprising; it goes against conventional wisdom. A next step would be to see if the finding holds in longitudinal analyses as well.

Another study might investigate the proposition—an implication of conventional wisdom about the impact of the movement of women into the work force—that a reduction in their volunteer time has negative consequences for volunteerism in general. Efficiency considerations suggest that a switch from donations of time (or goods) to greater donations of money actually may be preferable. Organizations can use money in any way (to buy stationery, pay workers, or purchase food, for instance) depending upon which resources are scarcer than others. In general, donated goods and time do not have this flexibility. Although an excess supply of goods can be sold to obtain money, this involves a transaction cost. Also, volunteers may not provide a stable work force (as mentioned earlier), and they may only be willing to undertake the more attractive tasks, leaving less desirable but still necessary ones unfinished.

The extent to which people interchange donations of time, money, and goods depends in part upon how volunteering time is viewed. If it is viewed as another form of work, a fair amount of substitution (donating money or goods rather than time) is likely to occur as women enter the labor force. Less substitution may occur if the volunteerism is viewed as a way of using leisure time.

Complicating the matter further, different volunteer activities may be viewed differently. High-status activities, such as heading a charity ball, may be considered leisure; lower-status activities, such as canvassing for the Heart Fund, may be more akin to work. This and previous discussions have focused on donations to formal volunteer organizations. There is evidence that participating in formal volunteer groups is largely a middle-and upper-class phenomenon. Does this imply that lower-income people are less likely to volunteer? Or do they volunteer more in casual, unstructured ways, such as helping friends and relatives or working for pay for formal volunteer groups?

In a more general vein, why do people help other people in informal ways? Are the reasons similar to those that lead people to volunteer in structured settings? A change in volunteerism from informal to formal settings may not have a significant impact on the amount of volunteerism if the same reasons hold. If the reasons differ, however, then a movement to more formal ways of volunteering may reduce the amount of volunteerism that occurs. It would be important to understand how society could compensate for such a decline.

One way of handling this question is to look to history. In the past many activities were carried out on an informal, sometimes ad hoc, basis. Some of these same activities—lending books and putting out fires, for example—are now handled within formal, organized systems (although informal volunteering is still with us; especially in times of emergency, people continue to band together, almost as a mutual insurance policy). Much of the past movement from informal to formal ways of carrying out activities may have been a response to increasing population densities and technological sophistication. Within a complex society there may be certain classes of problems that are more efficiently handled by formal, organized structures. Learning about these classes of problems in the context of historical changes in formal and informal volunteerism would help us understand more about the general direction volunteerism generally is likely to take in the future.

The Role of the Voluntary Sector

The research cited in this paper and the two previous sections on research questions have dealt with individuals and the circumstances under which they may volunteer. An equally important area for future work is the volunteer organization: the way it operates, the efficiency of its operations, and its organizational capacity. Volunteer organizations presumably have no profit motive; how then do they measure their own performances? And how do these measures affect their efficiency, that is, the way in which they distribute their own scarce resources? There is evidence that at least some volunteer organizations are notoriously inefficient and that only a small portion of the donations they receive are ever seen by the intended beneficiaries.

Further, from a historical perspective, how flexible have voluntary organizations been in adapting to changing social needs and circumstances? For example, to what extent have they adapted to the increasing participation of women in the work force? One response of these organizations might be to continue to rely on women who are not employed, an ever-shrinking pool. Another response might be to alter the timing of volunteering so that people could both work and volunteer.

A final broad area of research is the way in which the voluntary sector as a whole operates and the role it plays in the total economy.24 On the one hand are social needs, and on the other freedom of choice among people to volunteer however they wish. It may be possible to rely upon these volunteers only for some social goals; in other areas, government expenditures and paid workers may be necessary.

Earlier sections of this paper suggest that the voluntary sector comprises thousands of individuals, all making separate decisions: whether or not to volunteer in the first place, what to volunteer for, what form the volunteerism should take, how much to do, and when to participate. This characterization suggests that the voluntary sector could be viewed as a "market for volunteerism." As in any market, there are people who supply the activity—the volunteers—and there are people who demand it—the beneficiaries of the volunteerism.

The key difference between this market and others, however, is that there is no pricing system in the volunteerism market. In other markets, prices serve a critical function; they determine the relative amount of each good that is both supplied and demanded. When high prices occur, more goods are supplied than may be demanded; in response, the supply declines and prices fall until an equilibrium price and quantity are established between those who demand goods and those who supply them.

Because there is no mechanism equivalent to a price system for regulating supply and demand within the voluntary sector, it is difficult to determine whether too little or too much of an activity is being supplied relative to demand (e.g.,., Are too many or too few Girl Scout cookies being sold?). The lack of a price system would not matter if resources were limitless. But in the voluntary sector, as elsewhere in the economy, resources are scarce and must somehow be allocated among competing uses.

It is quite probable that the suppliers of volunteerism do not systematically offer the range of goods and services that beneficiaries of volunteerism would like to be offered. And there are some goods that may not be produced if supply decisions are left solely to the workings of a marketplace. The beneficiaries of volunteerism must have a way of registering their demand so that a supply can be forthcoming. Normally, to demand a good requires having command over some resource (such as money) that can be traded; but in the case of the volunteerism market, the beneficiaries have no easy way of registering their demand. For example, we only know about the demand for food and shelter by the lines that form outside churches and other charitable organizations. Is this a sufficient measure of the demand for these services?

The volunteerism market may not generate the range of desired goods and services for another reason as well. As in other cases, it is likely that there are differences between the services that people, as individuals, are willing to supply themselves and those that they believe should be supplied as part of their social responsibility. For example, families may not be willing to house indigent persons themselves, but taken together they may wish such housing to be supplied. Identifying areas in which discrepancies such as this occur would suggest where governments could appropriately target their resources.

In sum, the supply of volunteerism, the demand for it, and the way in which voluntary contributions are allocated among competing social goals should be modeled both conceptually and empirically. Also, an understanding of the nature and extent of the differences between individually supplied and socially demanded services is a critical policy area. Because governments cannot solve all social ills, it is important to know where, and to what extent, they can rely upon the voluntary contributions of Americans.


Major changes are occurring within the American economy, changes that require increased volunteer activity. As a consequence of the reduction in expenditures of all levels of government, increasing demands are being placed on the voluntary sector to meet both immediate and long-term social needs. Furthermore, the aging of the population will lead to an increasing need for health care and other volunteer services for the elderly. The ways in which volunteers and volunteer organizations can, and will, respond to these changes are not really known. (For example, there is an expectation—or perhaps more accurately, a hope—that older, retired people will become a major source of volunteers. It may well be, however, that many individuals will not be eager to volunteer. Particularly if the retirement age continues to increase, they may prefer instead to relax and enjoy themselves after a lifetime of work.) Currently, little is understood about the extent and nature of volunteerism: the reasons why people volunteer, how they volunteer, and for what kind of activities they volunteer. Because of this information gap, it is not possible to suggest either ways to stimulate volunteerism in general or areas in which it is most likely to arise on its own.

It is also not quite clear whether an increase in volunteerism would be the most efficient way to supply needed services. How the voluntary sector allocates its own scarce resources remains obscure. We also do not know the extent to which the sector's activities do, or could, correspond to those that society at large considers valuable enough to undertake.

Today volunteerism is being promoted to meet social needs, a view that is based on assumptions about ways in which individuals could, or should, productively use their time. This view is also based on assumptions about the efficiency of the private voluntary sector as compared to the government sector in providing needed social services. It remains to be determined, however, whether people wish to use their time volunteering and whether volunteering is the most efficient means of achieving social goals.


Burkhard von Rabenau and Sara B. Toye provided helpful comments and suggestions on earlier drafts. The views expressed here are not necessarily those of the National Commission for Employment Policy or of the National Academy of Sciences and, of course, any errors are the author's.


1. Gallup Organization, Inc., Americans Volunteer. Survey for Independent Sector (Princeton, N.J., June 1981).

2. ACTION, Americans Volunteer, 1974 (Washington, D.C.: Independent Sector and the Gallup Organization, Inc., February 1975).

3. Carol L. Jusenius, Retirement and Older Americans' Participation in Volunteer Activities (Washington, D.C.: National Commission for Employment Policy, June 1983).

4. Ibid.

5. For a list of many of the ways in which people volunteer, see Gallup Organization, Americans Volunteer.

6. Gallup Organization, Americans Volunteer.

7. Jusenius, Retirement and Older Americans' Participation.

8. Kenneth Boulding, "Notes on a Theory of Philanthropy," in Frank Dickinson, ed., Philanthropy and Public Policy (New York: National Bureau of Economic Research, 1962).

9. Ibid., pp. 57, 61.

10. Gary S. Becker, "A Theory of Marriage: Part II," Journal of Political Economy 82, no. 3 (April 1974): S11–S26.

11. Marnie Mueller, "The Economic Determinants of Volunteer Work by Women," Signs: Journal of Women and Society 1, no. 4 (Winter 1975): 325–338; and Paul L. Menchik and Burton A. Weisbrod, "Government Crowding Out and Contributions of Time—Or Why do People Work for Free?" Unpublished paper. Dec. 8, 1982.

12. Mueller, "The Economic Determinants."

13. Other independent variables included husband's income, woman's current or last wage in the job market, youthfulness and number of children in the household, population size of the woman's place of residence, and woman's religion (Catholic, Protestant, Jew, or all other). The results of studies of these other independent variables indicated that women who identify with a religion and live in rural areas are likely to do more volunteer work than those who do not identify with a religion and live in urban areas.

14. Mueller, "The Economic Determinants," p. 334.

15. Menchik and Weisbrod, "Government Crowding Out," p. 5.

16. Ibid., p. 11. Other independent variables included measures of the population size of the person's place of residence; presence and age of children; the individual's age, sex, marital status, and income from other sources (such as interest and dividends); a measure of the amount of time the individual has spent helping friends and neighbors; and background characteristics of the individual (such as the parents' regular attendance at religious services, father's education, and regularity of parents' donations to charitable and religious organizations).

The results of the study indicated that more hours of volunteerism are undertaken by working people who live in larger rather than smaller cities, who are younger, and who give more help to friends and neighbors. For the area of natural resources specifically, working people with children in the family volunteer more hours than people without children; they volunteer fewer hours than those without children in the areas of education (all levels) and welfare.

17. Jusenius, Retirement and Older Americans' Participation.

18. Within this setting, household members' time can be conceptually divided into different activities. Some time is spent working to produce a household output of goods and services. This work is done both in the home and in the job market. For work done in the home, the compensation is the goods produced. For work done in the job market, the compensation is income, which can be used to buy goods and services in the market or to ''buy time" to spend in activities other than work in the job market. For example, the income of one family member may be used to purchase household necessities as well as the time of a second family member so that he or she can be at home (taking care of children or taking early retirement).

Another part of household time is spent in leisure, the time devoted to consuming (or enjoying) the previously produced goods and services. More generally, leisure is time not working.

The statements above are a nontechnical description of the theory of the household's allocation of time. For technical versions, see Gary S. Becker, "A Theory of Marriage: Part I," Journal of Political Economy 81, no. 4 (July/Aug. 1973): 813–846; and Becker, "A Theory of Marriage: Part II."

19. The National Longitudinal Surveys sampled five separate cohorts: (1) men who were 45 to 59 years of age in 1966, (2) women who were 30 to 44 years of age in 1967, (3) young men aged 14 to 24 in 1966, (4) young women aged 14 to 24 in 1968, and (5) men and women aged 14 to 21 in 1979. In the first four cohorts blacks were oversampled so that they could be studied separately; there is about 15 years' worth of data for these groups. In the fifth cohort, "youth," whites from low-income households, blacks, and Hispanics were oversampled and persons in the military were included; these groups were most recently surveyed in 1982. For additional information, see the Center for Human Resource Research, The National Longitudinal Surveys Handbook, 1982 (Columbus, Ohio: Center for Human Resource Research).

20. Independent variables included the individual's race, ethnicity (Latin or other origin), health status, employment status, place of residence and unemployment rate of that area, income from assets, and whether or not there were children in the household. For the women, an additional variable measured past participation in volunteer activities.

21. Menchik and Weisbrod, "Government Crowding Out."

22. This view was confirmed by program operators participating in a panel discussion entitled "Volunteerism: Can It Work in Your Community?" The panel was part of the 1982 Congress of Cities Exposition at Los Angeles, Nov. 27–Dec. 2, 1982.

23. For a further discussion of this issue, see, for example, James N. Morgan, Richard F. Dye, and Judith H. Hybels, "Results from Two National Surveys of Philanthropic Activity," in Research Papers, Volume I: History, Trends and Current Magnitudes, sponsored by the Commission on Private Philanthropy and Public Needs (Washington, D.C.: U.S. Department of the Treasury, 1977).

24. For another way of conceptualizing the voluntary sector, the reader may wish to see Burton A. Weisbrod and Stephen H. Long, "The Size of the Voluntary Nonprofit Sector: Concepts and Measures," in Research Papers. Also, for empirical estimates of the size of the voluntary sector, see the Weisbrod and Long paper, others in that volume, and Burton A. Weisbrod, "Assets and Employment in the Nonprofit Sector," Public Finance Quarterly 10, no. 4 (Oct. 1982): 403–426.

Carol Jusenius Romero is staff economist at the National Commission for Employment Policy, Washington, D.C.

Copyright © National Academy of Sciences.
Bookshelf ID: NBK216829


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (1.7M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...