Volunteer corps programs — AmeriCorps, Peace Corps, City Year, Teach For America, and their domestic and international equivalents — are designed to accomplish a public purpose: education, environmental conservation, community health, disaster response. What they also do, as a structural byproduct of how they are organized, is produce friendships. Whether this byproduct is understood as incidental or essential shapes whether it is designed for and whether the broader social argument for volunteer corps adequately captures what they actually deliver.
The conditions of service in a typical volunteer corps program are the conditions that developmental and social psychology identifies as optimal for friendship formation: shared challenge, repeated close contact, clear group membership, common purpose, modest but real risk, and removal from the ordinary social structures that maintain existing social hierarchies. William Rawlins's foundational work on adult friendship identified "doing things together" as the primary mechanism of friendship formation in adulthood — not disclosure, not conversation, but shared activity that creates the incidental contact and mutual knowledge from which bonds develop. Volunteer corps programs structure this kind of activity intensively, over months or years, with people who are not previously known to each other. The result is a friendship formation environment more potent than almost any other that adults typically encounter after college.
AmeriCorps serves approximately 75,000 members per year on full-time programs and another 700,000 through part-time and VISTA programs. Peace Corps has returned over 240,000 volunteers since 1961. City Year, founded in 1988, has engaged over 35,000 alumni. Teach For America has over 60,000 alumni. Across these programs and their international equivalents, the cumulative population that has passed through organized volunteer service — and through the friendship formation environment it creates — is in the tens of millions. The friendships formed in these programs are, anecdotally, among the strongest and most durable that participants report across their lives. They are also systematically understudied.
The reason this matters beyond the personal is civic. Robert Putnam's analysis of social capital in Bowling Alone distinguished between bonding social capital — ties within homogeneous groups — and bridging social capital — ties across social difference. Bridging social capital is what democratic institutions most require: it is the social infrastructure of people who can work with and trust those who are not like them, who have friendships across race, class, geography, and political identity, and whose social networks therefore span rather than reinforce existing social divisions. Volunteer corps programs, by structuring service in contexts of genuine social diversity — economically and racially mixed communities, cross-regional assignments, teams drawn from nationally heterogeneous applicant pools — are among the most reliable institutional producers of bridging social capital in American life.
This argument has not been fully made in the political discourse around volunteer service. The case for AmeriCorps rests primarily on its direct outputs: the tutoring hours, the trail miles, the houses rebuilt after disasters. These are real and their value is not in question. But they are also, from a measurement perspective, the least distinctive outputs of the program: the same number of houses could be rebuilt by paid labor; the same miles of trail could be maintained by park service employees; the same tutoring hours could be delivered by expanded school budgets. What volunteer corps programs produce that cannot be replicated through other means is the cohort of people who have served together, whose social networks now span the social distances American life normally maintains, and whose civic engagement data — voting, volunteerism, organizational membership — consistently outperforms comparable non-serving peers over decades.
The friendship that forms in service is a form of national infrastructure that is not measured in any existing accounting of program value.