Involuntary volunteering: The impact of mandated service in public schools

Monitoring the Future annual surveys, 1991-2003, 8th and 12th grade cohorts

US 8th and 12th graders, with Maryland as the treated state (the only state to mandate service - 75 hours - for all public high-school graduates from 1992).

Difference-in-differences using Maryland's 1992 statewide service mandate as the shock and other states as controls, in repeated cross-sections from Monitoring the Future. The design is reasonable and the shock is genuinely a policy discontinuity, but it is a single-state treatment with self-reported outcomes, so it is graded C rather than B. READ DEPTH CAVEAT: read via the publisher record, an EdWeek write-up and Campbell's 2019 review; coefficients and standard errors were not extracted.

Mandating service raised service while the mandate bound and reduced it afterwards. Maryland 8th graders facing the requirement increased their service activity, and the first cohort of 12th graders under the mandate had higher rates of service. But LATER cohorts of 12th graders had LOWER participation in any voluntary activity, and the mandate did not produce long-term volunteer participation. Campbell's summary of the mechanism is the memorable one: rather than giving students a taste for more service, the requirement seemed to sate their appetite. This is the clearest evidence in the cluster that a civic REQUIREMENT can be net negative on the civic BEHAVIOUR it targets.

Low to medium. Identification is off a state policy change, not individual selection, so heritable disposition is not the main threat; the threats are Maryland-specific shocks coinciding with the mandate and social-desirability drift in self-reported volunteering.

No direct replication of the Maryland mandate. It points the OPPOSITE way to Metz and Youniss 2005, a single-school cohort comparison finding that a service requirement raised civic INTENTIONS among students who would not have volunteered on their own. Weinschenk and Dawes 2022, using transcripts and family fixed effects, find service-learning coursework has no effect on adult turnout. The honest reading is that mandated service is unproven at best and possibly counterproductive for later volunteering.