The Great British Clubhouse Is Still Standing. Nobody's In It, Least of All Where It's Needed Most

I tried to measure the decline of British community life by counting clubs. That was the wrong place to look. What’s disappearing is participation especially where deprivation is high.

To feel truly at home in a place, I believe you have to be part of something in it. That can be a team, a club, or any other place where you meet people you’d otherwise never meet and your turning up actually matters to someone. I have strong memories of this as a teenager. I trained kids’ sport teams in my neighbourhood (shout-out to the under 13 girls at FC Castricum) and for six years straight I gave up my summers and a substantial number of weekends to teach children and adults sailing and to keep sharpening my own technique and theory at my local sailing school. On top of that, I worked at a local beach restaurant. As a result, I started being recognised on the local streets which is exactly what being a local is to me.

Then adult life happened: four countries, a great number of moves, and somewhere across all that motion I lost the part of me that was a local somewhere. I’ve become very good at arriving and fairly hopeless at belonging. To be fair, I’ve been approached a couple of times if I wanted to coach a local sports team in London (sorry Peter I could not commit since I’m in the process of moving, again, I know…).

But I’m now doing the closest thing: see if there is any way I can contribute with data and see where the bottlenecks of people joining in are. I expected to write some kind of Bowling Alone narrative which is Robert Putnam’s account of Americans who kept bowling but stopped bowling in leagues, the visible edge of a long collapse in clubs and the trust and social capital they generate. But the data told a different story.

Putnam’s image was of people who still bowled but no longer bowled in leagues. In Britain, the organisations are often still there and some new ones are still being founded. What has changed is the social form of community itself. In one Britain, civil society still looks like membership: the choir, the rowing club, the amateur dramatics society, the sports club, the place you join. In another more deprived Britain, civil society increasingly looks like provision: the food bank, the advice centre, the mental-health charity, the youth intervention project, the place you turn up to because something has gone wrong.

---

The register refused to cooperate

First I had to sort charities into the convivial kind (the bowls club, the brass band, the scout troop) and the provision kind (food banks, mental-health support, youth services), I did it three ways and only trusted what all three agreed on. A blunt keyword rule (does the description say “bowls club” or “food bank”?); an unsupervised clustering that ignores my labels and lets the text sort itself into families; and the fun one, a conviviality axis, i.e., I turned each charity’s self-description into a point in space with a language model, drew a line between a “members gather” pole and a “we provide a service” pole, and measured which way each charity leaned. (Still about a fifth of the register refuses to classify cleanly, so read everything below as direction, not decimals.)

Then I tracked the founding rate of the pure Putnam archetype such as the bowls club, the working men’s club, the sports and social club. The line over time is flat (see below). It is also tiny: the genuine members’ club is a couple of percent of the register and has barely twitched in twenty years. So, Britain didn’t stop founding the members’ club…

Zooming in, what the register does show is a reshuffle, from more community and youth sport, more arts and festival trusts, more mental-health charities; less education, training and youth work (see the figure below). I’ve seen this shape before. When I mapped Britain’s lost pubs, London looked like the one success story, until you noticed it hadn’t so much kept its pubs as swapped them: the £2.50 working men’s local for the £7.50 craft taproom. The charity register pulls the same trick at national scale, because a charity gets registered when there’s a reason to formalise, and those reasons differ wildly by field. A community sports club now has to constitute itself to get a grant; a group of friends who play bowls has never had to register anything. So the register is really a map of where it became worth incorporating, which is to say it measures the funding rules and the tax code, with extra steps.

The charity register is bad at measuring whether Britain still gathers, but it is good at showing what kinds of community infrastructure become visible to the state. That null is the first honest finding (which wouldn’t reach your academic journal!).

---

So I asked whether people actually turn up

If you want to know whether people still show up, you have to measure this directly. The Community Life Survey, the nearest thing England has to a running pulse of civic life, records formal volunteering falling from 45% of adults in 2013/14 to 28% by 2024/25 (see below) from nearly half the country to under a third, with regular, at-least-monthly volunteering down from 27% to 17%. Perhaps more surprising, the slide started before anyone had heard of covid furlough (it was already down to 37% by 2019/20); the pandemic simply shoved it off the ledge, and it hasn’t climbed back. Charitable giving has gone the same way, 82% of adults down to a low of 63%.

A small caveat: the participation surveys have seams, namely, the Community Life Survey changed method partway through, the pandemic muddies the middle, so treat the percentages as direction. And the causes of the volunteering slide are genuinely contested (time, secularisation, an ageing volunteer base, the phone in everyone’s pocket).

Sport says it in miniature. England is not less active, but Sport England now openly describes a shift away from organised, club-based sport towards flexible, individual activity: the treadmill, the solo run, the app instead of the team and the league. Fewer than one in five over 55s belongs to a sports club at all. We didn’t stop moving; we stopped doing it together (I run, cycle and swim often alone, so I have to admit that I am part of the problem - or I just blame it on Strava;)

Put the two halves together and the resolution is nearly the opposite of the piece I sat down to write. Bowling Alone did come to Britain but it’s just invisible in the place I went looking, because what’s dying is not the founding of organisations but the turning up to them. Which is the point the whole detour was really about. We measure civil society by counting its organisations, because organisations are the tidy thing the state knows how to count. But an organisation is a shell, and the shell outlives the people in it, which is what Theda Skocpol called “management without members”: bodies that still file accounts and keep a logo, and have almost nobody who belongs.

---

In richer places, community means joining. In poorer places, it means need.

There’s a second question the register can answer honestly, because it comes with postcodes attached: not just whether Britain gathers, but where. To unpack the aggregate national picture, I mapped every convivial body and every provision body to its local authority, divided by population, and lined the country’s three hundred odd districts up from least to most deprived. They pull cleanly apart. In the most comfortable tenth of England, convivial clubs are a slight majority of associational charities (about 52%) and provision the rest. By the most deprived tenth it has flipped hard: convivial bodies are barely a fifth (~22%), and provision charities such as food banks, mental-health support, youth services, are nearly four in five (see the image below). The lines cross around the third decile. Gathering, it turns out, is something the comfortable do for pleasure; provision is what grows where the state has thinned. Picture two high streets: the market town that still has its rowing club, its choir and its drama society, and the estate two postcodes over with a food bank, a youth project and a mental-health drop-in. Both are full of good people doing necessary things, but only one of them is somewhere you join.

Two checks, because a single crossing chart can fool you. The first asks whether this is a real spatial pattern or just a tidy-looking fluke. The standard tool for that is a lovely statistic called Moran’s I. It answers one question: do places that gather sit next to other places that gather, more than chance would allow? It runs from −1 (a perfect chessboard, every place the opposite of its neighbours) through 0 (random scatter) to +1 (everything clumped with its own kind). Convivial provision scores +0.11 which is modest, but with a shuffling test putting the odds of seeing even that by fluke at about one in a thousand. Thus gathering comes in geographic clumps: the leafy-suburb and market-town belt.

The second check is the objection you’re already forming: isn’t this just London, or just big cities, dragging the line? So I ran a regression that holds those things still, incorporating population size, and which region a place sits in, and asks what deprivation does on its own. It survives. Each one-standard-deviation step up the deprivation scale which is loosely, going from a middling district to a visibly poorer one, knocks seven to eight points off the convivial share, and the result is about as statistically unambiguous as this work ever gets (p<0.001). Population, once deprivation is accounted for, does nothing at all, so poverty is doing the work, not mere city size.

So the decline in showing-up isn’t spread evenly. Belonging is becoming a postcode lottery, thickest exactly where life is already easiest. A caveat first: this is a map of registration, not gathering, so poorer areas may simply incorporate their clubs less often. But the provision side is hard to wish away, and if the gap is even partly real, it matters, because clubs, teams and choirs are precisely where people meet across class lines, that is, the 'bridging' ties Robert Putnam argued let people climb and Raj Chetty's team, mining 21 billion friendships, found that the cross-class connectedness of the place you grow up in is the single strongest predictor of escaping poverty, ahead of school quality, family structure or local inequality, and worth on the order of 20% more in adult earnings for low-income children raised in well-connected places. So thinning the convivial life of poorer areas doesn't merely leave them lonelier; it strips out one of the few engines of upward mobility they had.

---

So what would I actually do about it?

The data narrows the advice three ways.

First, it says the bottleneck is not founding but turning up, so we should not make it easier to start organisations (which the register says we manage fine), but rather lower the cost of participation. The Community Life Survey’s own most-cited barrier to volunteering is time: work and life. That points at things such as paid volunteering leave, episodic and “micro” roles rather than the old time-bound committee seat, the flexible terms on which younger people are visibly still turning up.

Second, and this is where the map earns its keep since the intervention has a spatial element, and the policy actually arriving doesn’t match it. The 2026 English Devolution and Community Empowerment Act hands communities a genuinely strengthened Right to Buy the local asset such as the pub, the hall, the sports ground, now its own protected category, replacing the near-useless old Right to Bid with a real right of first refusal. Good, as far as it goes. But the Community Ownership Fund that used to put money behind those purchases has closed, and the Act’s own impact assessment expects only a minority of listed assets to change hands, because finding the money inside the moratorium is the hard part. A legal right you can’t afford to exercise is usable mainly where the money already is, and weakest exactly where the deprivation gradient says convivial infrastructure is thinnest. So, the missing half of the policy is capacity and funding aimed at deprived places.

Third, the meta-point this whole exercise kept arriving at: stop treating the count of registered organisations as the measure of civic health, it’s the number that stays reassuring while the thing it claims to track drains away. The honest dashboard already exists: the participation survey. Fund against that, target it by the geography of thinness, and stay sceptical of anyone promising community can be conjured by money alone.

I went looking for the dead clubs and found them mostly still standing. It’s the people who’ve left, and if I’m honest, I’m one of them, which is the harder thing to fix.

But these people haven't left evenly. The leaving has a geography: gathering holds where life is already comfortable and thins where it isn't, so we're not simply a country that joins less, instead we're splitting into places that still gather and places left with whatever the state can provide instead.

---

Thank you for reading! I have some cool, new ideas I’m working on. If you want to keep my Substack alive, buy me a coffee and/or subscribe:)