The Free-Rider Problem: Why Good Things Stay Underfunded (And How to Fix It)

Your neighborhood has a community garden. Tomatoes all August, herbs you can raid on the way home, a bag of basil that turns up on somebody's doorstep. Sixty households live on the block. When the soil and seed order comes due, six of them chip in. The other fifty-four aren't bad neighbors. They assume someone else has it covered, and for a while someone does.

This is the free-rider problem, and it is the main reason good things in communities stay underfunded and understaffed.

The Logic of Not Contributing

Free-riding is a rational response to a structural incentive, not a defect in anyone's character. Economists call a good non-excludable when you benefit from it whether or not you paid: clean air, safe streets, a well-lit park, a neighborhood watch, a mutual aid fund. Because nobody can be locked out, the individually sensible move is to let someone else cover it. Multiply that sensible move by sixty households and the garden has no budget.

What follows is predictable. Everyone waits for everyone else, contributions trickle in, and the six people carrying it burn out. Then the garden withers, or the mutual aid group stops meeting and nobody sends the email saying so. People haven't stopped caring. The structure just charges them for caring and hands out the alternative for free.

Economists have chewed on this for decades. The textbook fixes are privatization, government provision, and moral appeals, and each one costs you something. Privatize the garden and only dues-paying members get tomatoes, which defeats the point. Wait for the city and you will wait a long time. Guilt-trip your neighbors and you get resentment instead of seedlings.

There are better options. Three mechanisms in particular work well in combination, and the rest of this piece takes them one at a time.

Mechanism 1: Quadratic Funding

Quadratic funding starts from one idea: what a fundraiser should measure is how many people gave, not how much any one of them gave.

Run a traditional fundraiser and a project with one $100 donation raises $100. A project with a hundred $1 donations also raises $100. The totals match, so the two look identical on the ledger, and that is a mistake. A hundred people each went to the trouble of pulling out a card. One person wrote a check they may not have noticed leaving the account. The second fundraiser has proved something about public value that the first has not.

The fix is a matching pool that pays out based on the number of contributors. The math is less intimidating than it sounds: take the square root of every contribution, add those roots together, then square the sum. Square roots are what do the work, because they flatten large gifts while leaving small ones nearly intact. Whatever that squared total exceeds the raw donations, the matching pool covers.

Run the numbers on the garden. Twenty-five neighbors give $1 each, twenty-five square roots of 1 sum to 25, and 25 squared is 625. The pool tops up the missing $600. Meanwhile a single donor putting $25 into a pet project gets a square root of 5, squared back to 25, and no match at all. Nobody had to decide which projects deserved support. The formula reads the community's attention and routes money toward it.

Gitcoin has run quadratic funding rounds for digital public goods since 2019, distributing over $60 million.^[1] Across those rounds, projects with hundreds of small backers have consistently pulled more than projects backed by a handful of large ones. None of this makes free-riding vanish. What changes is the arithmetic in your head: a dollar costs you almost nothing and the multiplier makes it land like twenty, so holding onto it stops feeling clever.

Mechanism 2: Community-Backed Incentives

Money is the easier half. Somebody still has to pull the weeds. Mutual aid networks need drivers and cooks and someone willing to sit on hold with a landlord for forty minutes, and none of that gets solved by a matching pool. Time free-rides exactly the way money does.

Community-backed incentives, or CBI, work on this side of the ledger by making steady participation visible and worth something. The mechanism is recognition over time rather than surveillance or obligation.

Picture a community currency that accrues to people who keep showing up. You volunteer at the food pantry three Saturdays running, and that record goes on-chain, where anyone can verify it and nobody can quietly edit it out. Over time the record carries weight in community decisions and standing in the next funding round.

What separates this from a volunteer sign-in sheet is that no central authority validates anything. The community attests. You say you cleaned up the park, three neighbors confirm it, and that is the record. No manager, no time clock, no one person whose spreadsheet is the official version of events.

That matters because free-riding thrives on anonymity. Once contributions are legible, and legible in a way that reads as recognition rather than a leaderboard of who is slacking, the social pressure takes care of itself. People like being seen as the sort of person who shows up. The garden never has to guilt-trip anyone.

Mechanism 3: Transparent Governance

There is a third dimension to all this that gets overlooked, and it is trust. People withhold contributions when they doubt the money will be handled well, and that doubt is usually earned.

Think about the groups you have been in. One person holds the bank login. The balance is whatever they say it is at the meeting. Decisions get made in a group chat you are not in. This is almost never malice; keeping honest books is real work, and running things democratically is exhausting, so power quietly concentrates and the rest of the membership drifts off.

Transparent governance addresses that without building a surveillance apparatus, because what gets exposed is the institution rather than the individuals. Every treasury transaction is visible to every member, proposals are voted on in the open, and allocation follows rules the group wrote down together. Trust stops depending on any one person's reputation.

Compare the two asks. One is “trust me, we spent it wisely.” The other is a ledger with every dollar in and every dollar out. The second one does not require you to trust anybody, which is precisely what makes it durable.

Participatory budgeting experiments, from Porto Alegre to New York City, point the same direction: when people can see where the money goes and have a real say in it, participation climbs. The incentives did not change. People simply started believing their contribution counted for something.

Accountability Without Surveillance

Worth flagging: none of these three mechanisms require tracking individuals or building profiles. Quadratic funding counts contributors. CBI records attestations people chose to give. Transparent governance exposes what the institution did, not what any member did.

That distinction is easy to lose. Most proposed fixes for free-riding amount to watching everyone more closely: put a camera on the park, score people on how often they turn up. Those methods probably do reduce free-riding, and they cost you the thing that made the community worth joining, which is that participation was yours to offer.

The goal is not to stamp out free-riders. It is to make contributing cheap enough and trustworthy enough that most people just do it.

Putting It All Together

Now run the garden again with all three in place. It posts a funding proposal on Goodkeep. Forty neighbors put in $2, quadratic matching turns that $80 into $1,600, and the season's supplies are covered. Volunteers who keep turning up earn attestations that give them a say in how the plot is run. Every purchase and every vote sits where members can look at it.

The six who used to carry it now have forty co-contributors and an actual budget. The fifty-four found out that $2 and the odd Saturday morning was easy, and that it counted for something. Nobody was coerced or watched. Cooperation just became the path of least resistance.

That is the whole shift, and it never required better neighbors. The free-rider problem is a design problem, and design problems can be designed around.