Solving the Coincidence of Wants in Digital Markets

The coincidence of wants is the requirement that two parties in a direct exchange must each possess something the other desires. This constraint is the fundamental barrier to barter and the historical reason money was invented. In digital asset markets, trade coordination offers a different solution: discovering trades across a preference graph that bypass the need for both bilateral coincidence and monetary intermediation.

Jevons and the Original Problem (1875)

In 1875, William Stanley Jevons published Money and the Mechanism of Exchange, laying out the case that barter economies inevitably fail at scale. His central argument was elegant: for a barter transaction to occur, you need a “double coincidence of wants” -- Party A must want what Party B has, and Party B must simultaneously want what Party A has.

Jevons illustrated the improbability with a vivid example. Suppose a baker wants meat, but the butcher does not need bread -- he needs candles. The baker cannot trade with the butcher directly. He must first find someone who wants bread and has candles, trade for candles, then return to the butcher. This chain of intermediate trades is inefficient, unreliable, and only works when the baker can identify the correct chain of intermediaries.

The problem scales poorly. In a market with many participants offering different goods, the chance that any two people have a reciprocal match drops sharply. The more diverse the marketplace, the less likely a direct swap becomes. Jevons saw this as an unsolvable limitation of barter. His answer was money.

Money: The First Solution

Jevons's solution, and the consensus of economists for the next 150 years, was money. A universally accepted medium of exchange decouples the act of selling from the act of buying. The baker sells bread for coins, then uses coins to buy meat. No coincidence of wants is needed because everyone accepts money.

Money works brilliantly for fungible goods with efficient price discovery. When you sell 100 shares of Apple stock, the market instantly prices your sale and matches it with a buyer. The currency intermediation adds negligible friction because the bid-ask spread is tight, the market is deep, and settlement is near-instant.

But money as a coordination mechanism has a structural limitation that Jevons did not anticipate: it requires that every asset have a reliable price in terms of the common medium. For unique assets, this assumption breaks down. What is the fair price of a rare digital collectible? It depends entirely on who is asking and what they are willing to give up in exchange. Price discovery for unique assets is inherently subjective, noisy, and slow.

Why Money Fails for Unique Digital Assets

Consider the state of NFT marketplaces. Millions of unique digital assets are listed for sale at owner-determined prices. The vast majority never sell. The structural data tells a consistent story: most NFT listings expire without a transaction. This is not because demand is absent -- it is because the currency-mediated exchange model forces every transaction through a price bottleneck.

The problem compounds in several ways. Floor price compression drives a race to the bottom within collections. Sellers who price above the floor get no visibility; sellers who price at the floor destroy the value premium of rare traits. Buyers, meanwhile, must hold liquid currency (ETH, SOL, USDC) to participate. A collector who owns a valuable NFT but no liquid crypto is locked out of new acquisitions -- even if another collector would gladly swap.

The economic term for this situation is market failure. Not because the market lacks participants or assets, but because the matching mechanism -- bilateral, price-mediated exchange -- is structurally inadequate for the asset class.

Trade Coordination: The Third Solution

If barter fails because of the double coincidence constraint, and money fails because of the price discovery problem, what is the alternative? The answer lies in expanding the matching dimension from bilateral to coordinated.

Trade coordination does not require any two participants to want each other's assets. It only requires that a path exists in the preference graph -- a connected set of participants where each one gives their asset to the next and receives an asset from the previous. The “coincidence” shifts from pairs to networks, and the number of viable coordinated trades grows dramatically as more participants join.

Consider the scaling dynamics. In a bilateral marketplace, adding more participants makes it harder for any individual to find a direct match -- the demand fragments across more unique assets. But in a preference graph, adding more participants creates exponentially more potential coordinated trades. The same marketplace data that yields a handful of bilateral matches can produce hundreds or thousands of viable coordinated trade opportunities.

This is the fundamental insight: scale works against bilateral matching for unique assets, but scale works for trade coordination. More participants, more preferences, more assets -- all of these increase the density of the preference graph and the number of trades SWAPS can discover.

After discovery, a scoring system selects the highest-quality non-overlapping trades, optimizing for fairness, value balance, and execution feasibility. The abundance of candidates is the engine's advantage: it gives the scoring system many options to choose from.

Why Real Marketplaces Are Ideal for Trade Coordination

Real marketplace preference graphs are not random -- they exhibit small-world properties. Participants within a niche (say, pixel art NFTs) form densely connected clusters, while bridge participants who collect across genres create long-range connections between clusters.

This structure is ideal for discovering trades. Dense local clusters mean there are many short paths between participants who share similar tastes. Bridge collectors connect disparate communities, which means SWAPS can find trades that span seemingly unrelated collections -- a pixel art collector, a photography collector, and a generative art collector, all connected through a chain of preferences.

This network structure also provides resilience. If a participant drops out of a potential trade, the dense local clustering ensures alternative paths exist. SWAPS can rapidly discover replacement trades that route around the missing participant. A bilateral match, by contrast, has no fallback -- if one side drops out, the trade is dead.

Three Exchange Paradigms Compared

Historically, there have been two exchange paradigms: direct barter and money-mediated trade. Trade coordination introduces a third that combines the advantages of both while avoiding their core limitations.

From Theory to Implementation

The idea that coordinated exchange can outperform bilateral matching was recognized long before it was computationally feasible at scale. Kidney exchange programs -- where incompatible donor-patient pairs are matched so each patient receives a compatible kidney from another pair's donor -- have used this approach since the early 2000s. These programs demonstrated that coordinated matching could meaningfully increase the number of successful transplants.

What makes digital asset markets different is scale and speed. A kidney exchange program matches dozens of pairs per quarter. A tokenized marketplace may need to match thousands of participants continuously. SWAPS achieves this through the living graph architecture: a preference graph that updates incrementally as inventory and wants change, discovering new trade opportunities in real time rather than in periodic batches.

Blockchain settlement adds the final piece: atomic execution. In kidney exchange, the transfers execute sequentially over days, with the risk that a donor backs out mid-chain. With SWAPS, onchain batched transfers ensure that either all assets in a coordinated trade move simultaneously or none do. The coincidence of wants is solved not just in discovery but in execution.