BOWLING ALONE

by Robert D. Putnam

  As an empirical matter, however, social networks provide the channels through which we recruit one another for good deeds, and social networks foster norms of reciprocity that encourage attention to others’ welfare. Thus, as we shall shortly see in more detail, volunteering and philanthropy and even spontaneous “helping” are all strongly predicted by civic engagement. As a matter of fact in contemporary America, those of us who belong to formal and informal social networks are more likely to give our time and money to good causes than those of us who are isolated socially. For this reason, altruism (and honesty, discussed in the next chapter) is an important diagnostic sign of social capital. Thus any assessment of trends in social capital must include an examination of trends in volunteering, philanthropy, and altruism.

  Giving time and money to help others is a long and distinguished tradition in American society. Both philanthropy and volunteering are roughly twice as common among Americans as among the citizens of other countries.2For the first several centuries of our national experience the social context for volunteering and philanthropy was primarily religious. Caring for others is a central tenet of all our faiths. Toward the end of the nineteenth century a new theme became a more prominent part of the rationale for altruism—helping the less fortunate was a part of our civic duty. As Andrew Carnegie, one of the new millionaires who emerged from the period of rapid growth following the Civil War, proclaimed in his 1889 essay “The Gospel of Wealth,” wealth was a sacred trust which its possessor was bound to administer for the good of the community.3

  During the twentieth century both volunteering and philanthropy became more organized and professionalized. Modern philanthropy began at the turn of the century, not merely with the accumulation of new wealth spawned by the Industrial Revolution, but also with the invention of new techniques for stimulating financial giving by ordinary Americans—the “community chest” (forerunner of the United Way), the community foundation, and the gradual professionalization of fund-raising and volunteer management. The number of community chests exploded from 39 nationwide in 1920 to 1,318 by 1950, covering 57 percent of the U.S. population.4 While the church remained the single most important locus of volunteering and philanthropy, it was joined by new institutions for organized altruism—the foundation, the corporation, and community organizations of all sorts.

  Many of the civic associations whose growth and recent decline we chronicled in an earlier chapter—the Scouts, the Red Cross, “service clubs” (Rotary, Kiwanis, and Lions), the PTA, and so on—were active in mobilizing volunteer energies. Throughout the twentieth century new organizations of collective altruism continued to emerge in response to new needs and renewed idealism—from the March of Dimes in the 1930s to World Vision in the 1950s to Habitat for Humanity in the 1970s to Teach for America in the 1990s. Between 1989 and 1994 the number of public charities in America grew nearly six times as fast as the U.S. population, and by 1996 a total of 654,186 public charities (not counting churches) were registered in the United States.5

  Americans are a generous people. Nearly half of us claim to undertake some sort of volunteer work, including both volunteering in organized settings, like churches and hospitals, and informal helping behavior, like baby-sitting a neighbor’s plants. According to one widely cited estimate, ninety-three million of us volunteered a total of twenty billion hours in 1995. Moreover, we give an impressive amount of money to good causes. In 1997 American individuals, corporations, and foundations gave $143.5 billion to charity, of which more than three-quarters ($109 billion) was donated by living individuals. In 1992 Americans gave 1.5 million gallons of blood, and the overwhelming majority of blood donors say that their main motivation is simply “wanting to help others.” In 1989, 74 percent of Americans reported giving money (not counting contributions to religious and political organizations), 35 percent reported volunteering, and 23 percent reported giving blood. We seem to be living up to Tocqueville’s observation more than a century and a half ago:

  Americans enjoy explaining almost every act of their lives on the principle of self-interest properly understood. It gives them great pleasure to point out how an enlightened self-love continually leads them to help one another and disposes them freely to give part of their time and wealth for the good of the state.6

  Amid the pressures of everyday life, giving time and giving money often seem alternative avenues for generosity. If we lack one, we can give the other. Generally speaking, however, volunteering and philanthropy are complements, not substitutes. Some of us give lots of both, while others give little of either. In 1995 volunteers contributed two or three times as much of their household income to charity as did nonvolunteers. Conversely, 63 percent of all financial donors also volunteered, as compared with only 17 percent of noncontributors. Volunteering is among the strongest predictors of philanthropy, and vice versa. Analogously, active blood donors are more likely to volunteer time and give to philanthropy than nondonors. Altruistic behaviors tend to go together.7

  Who among us are most generous with our toil and treasure?8 Not surprisingly, well-to-do, highly educated people—those who have more personal and financial resources—are more likely to volunteer, to donate money, and to give blood. In particular, education is one of the most powerful predictors of virtually all forms of altruistic behavior, even after controlling for other possible predictors. College graduates, for example, are twice as likely as people with a high school education or less to have volunteered in the past year (71 percent compared with 36 percent) or to be blood donors (13–18 percent compared with 6–10 percent). On the other hand, material resources are not the most important predictor of altruism. As a matter of fact, because of their relatively active church involvement, the poor give no less a fraction of their income than the wealthy.9

  Size of community makes a difference: formal volunteering, working on community projects, informal helping behavior (like coming to the aid of a stranger), charitable giving, and perhaps blood donation are all more common in small towns than in big cities.10 Age makes a difference: volunteering and blood donation generally follow an inverted U-shaped life cycle pattern, reaching a peak in one’s late thirties or early forties. Volunteering is especially common among parents of school-age children, and youth activities are second only to religion as a focus of volunteering. Philanthropy, on the other hand, typically accelerates with age, as disposable wealth accumulates.11 Employment increases the likelihood of volunteering, probably because it exposes workers to diverse social networks, but among volunteers there is a trade-off between time spent working and time spent volunteering, so the highest rate of volunteering is among part-time workers.12

  More important than wealth, education, community size, age, family status, and employment, however, by far the most consistent predictor of giving time and money is involvement in community life. Social recluses are rarely major donors or active volunteers, but schmoozers and machers are typically both.

  In 1996, 73 percent of members of secular organizations and 55 percent of members of religious groups said that they volunteered, as compared with only 19 percent of other Americans.13 As figure 28 shows, Americans who regularly attend both church and clubs volunteer an average of 17 times per year, ten times as often as those who are involved in neither church nor club, who volunteer on average 1.7 times per year. Secular involvement seems to have an even greater effect than religious involvement, for “pure” churchgoers volunteer an average of 5 times per year, while “pure” clubgoers average 12 times per year. Moreover, involvement in secular organizations is closely associated with participation with community projects, while involvement in religious organizations is not.14 People active in religious organizations volunteer for ushering in church or visiting shut-in parishioners, whereas people active in secular organizations are most likely to work on cleaning up the local playground.

  Schmoozing is also closely associated with volunteering.15 For example, as figure 29 shows, Americans who entertain friends at home
are much more likely to work on community projects and to volunteer in other ways. Moreover, people who are actively involved in community and social networks are more likely not only to volunteer in the first place, but also to stick with volunteering over a period of years, whereas people who are isolated socially are more likely to engage in purely episodic volunteering.16

  Figure 28: Volunteering Fostered by Clubgoing and Churchgoing

  Philanthropy is also tied closely to organizational involvement. In 1996, 87 percent of members of secular organizations and 76 percent of members of religious organizations made some charitable contribution, as compared with only 37 percent of nonmembers. Members of religious organizations donated an average of 1.9 percent of their annual household income ($802) to charity, and members of secular organizations gave an even more impressive 2.3 percent (or $1,167), compared with 0.4 percent ($139) for other Americans.17 In round numbers, joiners are nearly ten times more generous with their time and money than nonjoiners. Social capital is a more powerful predictor of philanthropy than is financial capital.

  Altruism of all sorts is encouraged by social and community involvement. Churchgoing and clubgoing, for example, are among the strongest predictors of giving blood, controlling for other background factors, such as age, education, sex, and so on. (See figure 30.) Americans active in community affairs are twice as likely to give blood as their stay-at-home neighbors. Even informal helping, like providing emotional support in the aftermath of a natural disaster or keeping an eye on a neighbor’s house, is strongly correlated with the size of one’s network of friends and acquaintances.18 To predict whether I am likely to give time, money, blood, or even a minor favor, you need to know, above all, how active I am in community life and how strong my ties to family, friends, and neighbors are.

  Figure 29: Schmoozing and Good Works

  Social connections encourage giving for many reasons. Joiners may be generous souls by nature, but involvement in social networks is a stronger predictor of volunteering and philanthropy than altruistic attitudes per se.19 As fund-raisers and volunteer organizers know well, simply being asked to give is a powerful stimulus to volunteering and philanthropy. When volunteers are asked how they happened to get involved in their particular activity, the most common answer is, “Someone asked me.” Conversely, when potential blood donors are asked why they haven’t given blood, the most common response is, “Nobody asked.”20

  Fund-raising typically means friend-raising. So the more involved I am in social and community networks, both formal and informal, the more likely I am to be asked. And I’m more likely to agree if the recruiter is part of my network of friends. Community organizations need time and money, and members call upon one another to pitch in, not only for that organization, but also for others. If I join the PTA, I’m very likely to be asked to volunteer for the fund-raising picnic, and someone I meet there may well invite me to help with the Cancer Society walk-a-thon. Once on the list of usual suspects, I’m likely to stay there.

  Volunteering fosters more volunteering, in both formal and informal settings.21 Organizational involvement seems to inculcate civic skills and a lifelong disposition toward altruism, for adult volunteers and givers are particularly distinguished by their civic involvement as youth. Those of us who were involved in youth groups or did youthful volunteering are half again as likely to donate to charity as adults and twice as likely to volunteer as those of us who were not so involved as youngsters. Finally, careful studies have shown that (with other social and personality traits held constant) people who have received help are themselves more likely to help others, so that simple acts of kindness have a ripple effect. In short, giving, volunteering, and joining are mutually reinforcing and habit-forming—as Tocqueville put it, “the habits of the heart.”

  Figure 30: Blood Donation Fostered by Clubgoing and Churchgoing

  AGAINST THIS BACKDROP, what have been the trends in giving and volunteering over recent decades? Let’s begin with philanthropy. Each year American charities brag about new records in money raised and spent for worthy causes. For as long as records have been kept, total giving in current dollars has risen steadily. Even in dollars adjusted for inflation and population growth, the trend has been generally upward, with only temporary dips around recession years. According to one enthusiastic cheerleader for American generosity, per capita charitable giving in constant 1993 dollars nearly doubled between 1960 ($280) and 1995 ($522).22

  On the other hand, the growth in charitable giving, even in constant dollars, is hardly surprising, for our income has also risen and, along with it, our spending on practically everything. For example, over those same years (1960–1995) in which real per capita giving doubled, real per capita spending on flowers, seeds, and potted plants almost tripled, and real per capita spending on all recreational goods and services combined—from daffodils to Disneyland and from toys to TV repairs—nearly quadrupled.23 To measure our phil-anthropic generosity, we need to know how our giving compares to our income, not merely how many dollars we are handing out. If my income quadruples, while my weekly church offering increases by only a quarter, most reasonable people would say I’m becoming stingier, not more generous. “Tithing,” after all, is about relative, not absolute, numbers.24

  Trends in American philanthropy relative to our resources are dismaying, for in the 1990s Americans donated a smaller share of our personal income than at any time since the 1940s. The long-run trends in personal philanthropy are reminiscent of the evolution of other aspects of American civic engagement, as figure 31 reveals.25 The first half of the twentieth century was an era of increasing national generosity. As a share of income, personal philanthropy nearly doubled in the three decades between 1929 and 1960. After brief disruptions associated with the Great Depression and World War II, American giving, relative to our means, rose sharply and steadily after the war, increasing by nearly 50 percent between 1944 and 1960. (Since this was a period of rapid economic growth, in real terms the increase was even sharper.) Beginning in 1961, however, philanthropy’s share of Americans’ income has fallen steadily for nearly four decades, entirely erasing the postwar gains. Total giving by living individuals as a fraction of national income fell from 2.26 percent in 1964 to 1.61 percent in 1998, a relative fall of 29 percent. In 1960 we gave away about $1 for every $2 we spent on recreation; in 1997 we gave away less than $.50 for every $2 we spent on recreation.

  The parallel between the timing and direction of trends in philanthropy and the nearly simultaneous ups and downs of American community involvement and social connectedness we reviewed in earlier chapters is uncanny. By contrast, the long-run ups and downs of philanthropy are quite disconnected from the ups and downs of the economy. Hit by the Great Depression, American real per capita income fell by 3 percent between 1929 and 1939, while the share of income that we gave to charity rose by more than a quarter. Over the next two decades real per capita income soared by 74 percent, yet personal giving as a fraction of income continued to rise at just about the same long-run pace as during the Depression years. Through bad times and good, Americans grew steadily more generous. Conversely, after 1960 our generosity has steadily shriveled. Through the booms of the sixties and eighties, as well as the busts of the seventies and early nineties, this inexorable retreat was disrupted only briefly during the middle eighties in response to passing changes in the federal tax code. In short, the waxing and waning of American generosity over the last seventy years coincides closely with the ups and downs in our stock of social capital and not at all with the ups and downs in our stock of financial capital.

  Figure 31: The Rise and Fall of Philanthropic Generosity, 1929–1998

  The subsidence of American philanthropic impulses since 1960 has been very pervasive, affecting many different communities of givers and many different recipients. In round numbers, half of all charitable giving in America is religious in nature, so we can get some additional insight (and confirm the general picture) by focusing se
parately on trends in giving to the major faiths, as well as trends in giving to secular community activities. Figure 32 arrays the best available evidence on long-term trends in giving to major Protestant denominations, to Catholic causes, and to United Way, the most extensive community-based fund-raising operation in America and a good proxy for secular giving.26 The rhythms of giving in the first half of the century vary across these institutional settings, but the postwar boom in giving is apparent, as is the timing and extent of the post-1960 plunge in generosity relative to means.

  After falling sharply from 1960 to 1972, Protestant giving per member has stagnated since the early 1970s. On the other hand, as we noted earlier, membership itself in Protestant denominations has continued to fall steadily throughout this period, so Protestant giving as a fraction of national income has continued to fall; in that sense, figure 32 understates the decline in Protestant philanthropy. In other words, if a Protestant deserts her church entirely, as many have in recent years, the financial repercussions of that apostasy are not reflected in figure 32.

  Figure 32: Trends in Protestant, Catholic, and United Way Giving, 1920s–1990s

  John and Sylvia Ronsvalle, two of the leading students of church finances in America, point out that the decline in Protestant giving is not limited to a particular portion of the theological spectrum. Evangelicals give a larger fraction of their income to the church, but their contribution per member has fallen even more rapidly than the figures for mainline Protestants.27 Still more revealing, among both mainline and evangelical Protestants, giving for “benevolences” (that is, external charity) has fallen more rapidly (down 38 percent since 1968) than for “congregational finances” (down 12 percent since 1968).28 In other words, a growing slice of the shrinking pie has been consumed by internal church operations, leaving even less for ministering to the world.