Next month in Prospect we review Economist editor John Micklethwait’s new book about the rise of world religion, which reproduces a familiar argument that the success of faith in America owes much to the lack of an established church. Many churches, competitive with each other, win more converts. It was a thought, in a different context, brought up this morning at a breakfast seminar I went along to, jointly hosted by Demos and the Private Equity Foundation, on national civic service. David Willetts was the main speaker, along with a visiting representative from City Year, an American outfit who send an annual cohort of a thousand or so teenagers in striking red jackets into inner city schools, aiming to mentor and inspire children a little younger than themselves. My interest in this stems from the article I wrote with Frank Field a few months back, calling on this government (or the next) to institute a mandatory, national scheme for every young person. And it was Willetts’s point that such a programme risked ending up like the Church of England—national, tied to the state, and unloved. If your aim is to develop a national culture of service, as the Americans put it, better to do so from the bottom up—following the example of American Protestantism, where many churches (or, in his analogy, charities), compete with one another to win over the faithful. When Frank and I wrote our piece, we didn’t honestly think that a cash-strapped Gordon Brown would turn round and suddenly cancel Trident, and use the money for a new national service. Even if the money was there, there are a bunch of tricky problems to overcome—how to build the programme, make people think it was a good idea, build up the capacity of the voluntary sector, and so forth. But the piece did aim to set out a vision—one in which every child in Britain at some point spent six months serving the community. Willetts’s question, though, is a good one. Do you get there from here (where we have no such culture, and few institutions) by following a model which aims to let a thousand flowers bloom, or which aims to set a framework from above? Another question which occurred to me while listening was about the right model to start off with. Let’s say you think that it will take a generation to build support to have a national system. Even then there are lots of different ways to start. City Year, in common with many of the more famous American examples, offers an intensive period of service to a few (mostly clever) kids. The British government just went with totally the opposite approach, with plans that everyone should do only 50 hours. The Tories are currently hovering halfway between, trialling (under Boris Johnson) a scheme which would offer a few weeks of residential service, which would then form part of a longer period of part time volunteering. During the seminar I doodled the graph below on a napkin to set out these three paths. Two questions follow: which model, pound for pound, is best at changing the lives of most young people? The City Year model changes lives, but only for a few. The government model could be for everyone, but what do you get from 50 hours? But second, and related: if the aim assumes that you can’t go big at first, which model would build support for a much bigger scheme in the future? Which path will work best: narrow and deep, wide and shallow, or a mix?