Young volunteers aren't hard to recruit. They're hard to keep.

In the most recent national US figures, 28 percent of 16 and 17-year-olds volunteered formally through an organization. The national rate across all adults was 28.3 percent. That data covers September 2022 to September 2023, comes from AmeriCorps and the Census Bureau, and was released in November 2024.

Teenagers volunteer at roughly the same rate as everybody else. Then Gen Z as a whole sits at 24.5 percent, below the 29.4 percent for Millennials, and Gen X has posted the highest rate of any generation across four survey waves.

Something happens in between, and it is not a failure of attraction.

The sector is solving the wrong half of this

Nearly every conversation I have been in about young volunteers has been a recruitment conversation. How do we reach them. What platform are they on. Should we be making video. Is our website too old-fashioned.

Meanwhile, the clearest measured fact about young volunteers is about what happens after they show up. In NCVO’s Time Well Spent 2023, based on YouGov fieldwork with 7,006 UK adults, 18 to 24-year-olds were the least satisfied age group in the survey, at 82 percent. The over-55s came in at 96 percent. Fourteen percentage points, in the same sector, in the same year.

The same survey establishes what satisfaction is worth. Very satisfied volunteers put their likelihood of continuing at 90 percent. Satisfied at 81 percent. Dissatisfied at 26 percent. Satisfaction is not a soft measure in this data. It is the leading indicator of whether somebody is still around next spring.

And Hager and Brudney, in their 2004 study for the Urban Institute, found that charities whose volunteers were mostly under 24 had lower retention than charities with older volunteer bases. That finding is more than twenty years old and it points the same direction as the NCVO data from 2023.

So: young people arrive at a decent rate, enjoy it less than everyone else, and leave sooner. A recruitment campaign makes the first number bigger and does nothing about the other two. You end up with a bigger intake and the same attrition, running faster to stand still.

The expenses problem is not a small problem

Twenty percent of young people in the NCVO survey named expenses as a barrier to volunteering. Only 55 percent of organizations reimburse expenses at all. Sixteen percent of volunteers did not know whether their organization did.

That third figure is the one I would fix this week, because it is free. An unclear expenses policy functions as a no. A nineteen-year-old on an hourly wage who does not know whether the $14 in bus fare comes back is not going to ask a stranger about it. They are going to say they are busy that weekend.

It is also worth being clear about what this costs the organization. A volunteer doing a four-hour shift is producing labor valued by Independent Sector at $36.14 an hour for calendar year 2025. Reimbursing $14 of travel to secure roughly $145 of salary-equivalent work is not a difficult decision, and yet 45 percent of organizations are not making it.

Put the policy in the role description, say it out loud at induction, and make the claim process something other than an email to a person who might be on vacation. NCVO’s own induction guidance lists explaining expense claims alongside safeguarding and data protection, which is the right level of seriousness.

What we do to teenagers who have to be there

A large share of under-18 volunteering in North America arrives through a school requirement, and this is where the evidence gets uncomfortable.

Stukas, Snyder and Clary published a study in Psychological Science in 1999 on the effects of mandatory volunteering. What they found was that stronger perceptions of external control eliminated an otherwise positive relationship between having volunteered before and intending to volunteer in future. Students who did not expect that they would have volunteered freely showed significantly lower future intentions after being required to serve than after choosing. Students who were already inclined were largely unaffected.

Read that carefully, because it is easy to over-claim. Mandatory service does not poison everybody. It appears to do damage specifically to the people the program is supposed to convert, and to leave the already-converted alone. The variable is not the requirement itself. It is whether the young person experiences the work as controlled.

Which is manageable. Give them a genuine choice of role once they arrive. Let them decide something about how the work gets done. Do not talk about hours as a quota to be signed off, even when a form says exactly that. The sheet is an administrative fact, not the frame of the relationship.

The role is doing more damage than the messaging

There is a pattern in how organizations assign work to young volunteers, and I have been guilty of it. The young person gets the task with the least consequence attached. Carrying things. Watching the door. Being available in case something comes up.

The reasoning is usually protective, which makes it harder to see. They are new, they are young, we do not want to put too much on them. What actually happens is that you have given them work that would be fine if anyone else did it, and they know that.

National US data published in 2009 by Eisner, Grimm, Maynard and Washburn found that volunteers doing general labor returned the following year 53 percent of the time, against 74 percent for volunteers doing professional or management-level work. Nothing in that finding is specific to age. It is just that young people receive far more than their share of the general labor. I have written elsewhere about what role design does to retention, and the under-25 problem is largely that finding, concentrated.

The other NCVO number worth putting alongside this: 26 percent of volunteers said volunteering felt too much like paid work, up from 19 percent in 2019, and 24 percent reported unreasonable expectations of them, up from 17 percent. A young volunteer given a menial task under close supervision is getting the worst combination available. All of the obligation of a job, with none of the responsibility, and no wage.

Ottawa, and what I got wrong for about two years

I started in youth groups and youth ministry in Ottawa, and later in New York. My first instinct with the fifteen and sixteen-year-olds was to keep them near the edges of anything that mattered, because I did not want them to fail publicly and I did not want to have to fix it afterward.

Attendance was fine. Return rates were terrible, and I blamed the age group for about two years before I looked at what I was actually handing them.

What changed it was giving a small number of them something with a name on it. Not assisting with the program. Running one part of it, badly at first, with somebody standing nearby who was not going to intervene unless it genuinely went wrong. The ones who got a real job came back. The ones who got a helping role drifted off, and I would now say they were correct to.

Young volunteers are not less capable. They are less willing to spend a Saturday being decorative, and that seems like a reasonable position.

Four things that change the numbers

Publish the expenses policy where a volunteer will see it before they commit, not after. Pair every new young volunteer with an existing volunteer as a buddy, which is standard NCVO induction practice and costs nothing. Give at least one under-25 something they own outright inside the first month, however small the scope. And ask them, at eight weeks, what has been boring, using that word, because “how’s it going” produces nothing from anyone under about thirty.

None of this is a campaign. It is four decisions about how the work is arranged, and it is the half of the problem the sector keeps not working on. The wording of the original ask still matters, but it matters much less than what the person finds when they turn up.