The 2-Point Block Party

A block party permit is worth about two percentage points at the polls. That is the finding from a 2024 study by Tanika Raychaudhuri, Joshua Davidson, and Michael Jones-Correa, published in Urban Affairs Review. The authors merged Philadelphia's voter file, municipal block-party permits, and Census data for a decade of elections, then asked a simple question: do people who live on a block that holds a block party vote at higher rates than people who live on blocks that do not?

The answer, holding income, age, and prior turnout constant, is yes. Residents on blocks that held at least one block party in the 2012 cycle were nearly two percentage points more likely to vote in that year's presidential election. The boost was larger in Black neighborhoods than in predominantly white, Asian, or Latino ones.

Two points is not a big number. Two points is a small number. But two points is also larger than the turnout effect of most of the expensive get-out-the-vote operations that campaigns run, which academic research has generally found to do very little. Two points in a close primary, multiplied across a ward, can be the margin that decides a citywide race. Two points in a city council district, multiplied across the right set of divisions, is the same kind of margin.

The authors do not claim to know exactly why. The leading theory is that block parties are one of the few remaining civic rituals at which neighbors who do not already know each other well end up in casual conversation, and casual conversation is how political information actually moves through a neighborhood. Registration deadlines. Polling place changes. Which candidate is actually showing up at the ward meeting. None of this is the kind of thing a campaign mailer conveys. All of it is the kind of thing you hear standing over a folding table with a paper plate of hot dogs.

For a Philadelphia politician, the implication is direct. You do not need a TV buy. You do not need a consultant's slide deck. You need a block permit application, a few folding tables, and a portable speaker. The cost is small. The effect, in a city where Democratic primaries are often the only election that matters, can be the size of a citywide margin. This is not a secret. It is just a fact about how turnout actually works at the neighborhood level. The broader pattern — what a councilmember's ward power actually buys, and why so many of them end up in court — is the subject of a piece called Six Out of Seven Convicted.

What is interesting about the study is less the political implication than the data infrastructure that made it possible. For decades, the question of whether casual civic events moved votes was a question you could not really answer, because the underlying records did not exist in a form that could be linked. A city permit here, a voter file there, a Census tract somewhere else. Three different agencies, three different schemas, no way to put them on the same map. The fact that they can now be put on the same map is itself a story — and a fairly recent one. The underlying permit dataset, by the way, is mapped permit by permit in a companion piece called Every Block Party in Philadelphia. And what those block parties actually look like on the ground — a folding table, a portable speaker, a three-month-old in an Eagles onesie being passed around the room — is the subject of a piece called The Eagles Onesie.

The study does not say block parties are good or bad. It does not say politicians who fund them are good or bad. It says something narrower and more useful: a folding table, a permit, and a few hours on a Saturday in June are worth about two percentage points at the polls in November. That is a number worth knowing. It is not a verdict. It is just the data.