Who posts the bond in a dominant assurance contract?

In Alex Tabarrok's original 1998 design, the entrepreneur proposing the project posts the bond and personally pays every backer a bonus if the campaign fails. On Pieces, the bond is an open pool that any supporter can stake into while the campaign runs, the creator included. Whoever holds the pool when a Piece fails is who pays the backers.

How does Pieces change Tabarrok's dominant assurance contract?

One structural change: the assurance is delegated. Instead of requiring the creator to escrow the failure payout, Pieces opens a bond pool capped at the kickback percentage times the funding goal. Supporters stake into it voluntarily. If the Piece funds, the kickback, that same percentage of the raise, is split across the pool pro-rata. If it fails, the pool is forfeited to backers as their refund bonus. The backer's guarantee is unchanged from the original design; who underwrites it is what changed.

Why not have the creator post the bond themselves?

They can, and Tabarrok's design fits an entrepreneur with the capital to do it. Digital creators usually start without that capital: someone raising $500 for an EP rarely has hundreds sitting free to escrow behind their own campaign. Opening the bond to supporters brings the same fully funded guarantee to creators who have work worth standing behind but no collateral, and a creator who wants skin in the game can still stake into their own pool.

What do the people who post the bond get in return?

Bonders are paid from the kickback the creator set, split across the pool in proportion to each stake, and only when the Piece funds. If it fails, their stakes are forfeited to the backers. The terms are public before anyone stakes, and the contract settles them as written.

Who Posts the Bond? The One Change Pieces Made to the Dominant Assurance Contract

Tabarrok's dominant assurance contract asks the entrepreneur to escrow the failure payout themselves. Pieces moves that bond into an open pool any supporter can join, which lifts the capital burden off the creator and lets the people who believed earliest share in a funded outcome.

In 1998, economist Alex Tabarrok published a fix for one of the oldest failures in funding public goods. Ordinary crowdfunding punishes early believers: pledge to a campaign that stalls and your money sat frozen for nothing, so the rational move is to wait, and when everyone waits, projects everyone wanted quietly die. Tabarrok’s fix was a single added clause: if the campaign fails, every backer is refunded plus a bonus. With that clause, backing becomes the best move no matter what anyone else does, which is why he called it a dominant assurance contract.

The design is airtight, and Pieces is built directly on it. Bringing it to creative work took one adaptation, and it concerns a single question: who supplies the bond.

The bond in the original design

In Tabarrok’s paper, the entrepreneur proposing the project does two jobs. They propose the work, and they post the bond: the escrowed money that pays every backer their bonus if the goal is missed. That fits the world the paper describes, an entrepreneur with capital, providing a public good, who profits when provision succeeds and can credibly escrow the failure payout. For that entrepreneur, the bond is a cost of doing business, and the design is exactly right.

Digital creators start from a different place. A musician trying to raise $500 for an EP does not usually have a few hundred dollars sitting free to escrow behind her own campaign; if she did, she might not need the campaign. So bringing the mechanism to creators meant finding the bond a source other than the creator’s own savings. The contract works exactly as Tabarrok showed it would; the open question was who supplies its collateral.

The change: assurance, delegated

Pieces keeps Tabarrok’s contract intact on the backer’s side and changes exactly one thing on the other side: the bond is an open pool, and anyone can post it.

When a creator locks a Piece, they set a funding goal, a deadline, and a kickback percentage. Those choices fix the bond pool’s cap at the kickback percentage times the goal, and while the Piece is live, any supporter can stake into that pool: superfans, collaborators, bonders of any stripe, the creator included. Then the contract settles both endings:

If the Piece funds, the kickback, that same percentage of the raise, comes off the top and is split across the pool in proportion to each stake.
If it fails, the pool is forfeited and distributed to backers on top of their full refunds, as the refund bonus.

From a backer’s seat, nothing about Tabarrok’s guarantee has changed: fund it and the work exists, watch it fall short and you are made more than whole. What changed is who underwrites that guarantee, and that turns out to change almost everything about who can use the mechanism.

What delegation buys, seat by seat

The creator sheds the capital burden. A creator on Pieces runs a full dominant assurance contract without escrowing a dollar. A failed Piece costs them nothing extra, and the kickback is paid only out of a raise that succeeded. Running the mechanism now requires no up-front collateral at all, only work worth standing behind.

The believers get a way in. In the original design there was no seat at the table for the fan who was certain this campaign would fund. On Pieces, that certainty is a position: stake into the pool, carry a real share of the risk, and collect a share of the kickback when the Piece funds. The kickback is the creator giving back to the people who showed up first, in the only currency that means anything: a piece of the success itself. The worked example runs the exact dollars.

The backers get a second kind of signal. A bond posted by the entrepreneur is already a costly, credible commitment; that is precisely what Tabarrok designed it to be. A pool that other people can fill adds information the original could not carry: independent supporters, each risking forfeiture, publicly pricing in a success. It tells a hesitating backer that people with real knowledge of this creator have put money behind that knowledge, and every dollar of it also funds the very bonus that makes backing safe.

The original is still in there

Here is the part we find satisfying: the open pool contains Tabarrok’s design whole. Nothing stops a creator from staking into their own pool, and a creator who fills their bond alone has reconstructed the original contract exactly: the entrepreneur posting the bond, promising to personally pay backers if the goal is missed. Tabarrok’s dominant assurance contract is the special case, and the open pool is what it looks like when the assurance is allowed to come from anyone who believes.

The delegation has real teeth, and it is worth being plain about them. Bonders who back the wrong campaign lose their entire stake to the backers. That forfeiture is the collateral that keeps the guarantee honest, exactly as it was in the original design; what Pieces moved is who gets to volunteer for it.

The full mechanics are laid out in how Pieces works, the theory in what a dominant assurance contract is, or you can see a live Piece for yourself.