Routing Frictions and Executable Liquidity in Fragmented Markets
This paper demonstrates that while public blockchains provide transparent access to fragmented liquidity, high transaction-level routing costs on networks like Ethereum drastically erode potential multi-venue gains, revealing that economically integrated execution is an order-specific property determined by access costs rather than mere connectivity.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the world of digital finance, a new kind of marketplace has emerged where anyone can see every available price for a specific asset at the same time. Imagine a global stock exchange where every trading booth is visible through a single window, and a single order could theoretically be filled by splitting it across dozens of these booths simultaneously. This is the promise of decentralized markets built on public blockchains: a system where technology makes every option reachable. However, just because a path is visible does not mean it is worth taking. Every time a trader tries to reach out to an additional booth to get a better price, they must pay a fee for the digital energy required to make that connection. This creates a hidden barrier: the difference between what is technically possible and what is economically sensible. The question researchers have long asked is whether these markets are truly integrated, or if they remain a collection of separate islands where the cost of travel keeps traders from ever crossing over.
A researcher led by Wen-Ting Wang at National Chung Hsing University set out to measure this invisible barrier with unprecedented precision. They focused on a specific type of digital trading venue where automated systems manage pools of money, allowing users to swap one token for another. These pools often exist in multiple versions for the same pair of assets, creating a fragmented landscape where a trader might find a slightly better deal in one pool than another. The researcher gathered data from nearly 14,000 specific transactions involving these token pairs on two different blockchain networks, one of which is a major network and the other a faster, lower-cost version. By reconstructing the exact state of every pool right before a trade happened, they could calculate exactly how much better a trader would have done if they had split their order across multiple pools, and then subtract the exact cost of doing so.
The results revealed a stark and surprising reality. When the researcher looked at the raw potential for better prices, they found that in more than one-third of the cases, splitting an order across multiple pools would have generated a gross profit. However, once they accounted for the real-world cost of activating those extra connections, the picture changed dramatically. The number of trades where splitting the order was actually worth the cost dropped to less than seven percent. In other words, the fees required to reach the better prices wiped out the potential gains in about 80 percent of the cases where a gain seemed possible on paper. This means that even though the technology allows traders to see and reach every pool, the economic reality is that most of those pools remain effectively out of reach for the average order.
The study also examined what traders actually did versus what they could have done. It turned out that in the vast majority of cases, traders stuck to a single pool, even when a multi-pool strategy was theoretically profitable. When the researcher looked at the rare instances where traders did successfully split their orders across multiple pools, they found something encouraging: those traders were remarkably good at it. Among the few orders that did manage to cross the cost barrier and use multiple pools, the traders captured nearly 95 percent of the maximum possible profit available. This suggests that the main problem is not that traders are bad at splitting orders once they decide to do so; the problem is that the cost of getting started is so high that it stops them from trying in the first place. The friction lies in the decision to activate the connection, not in the execution of the split.
To understand how much of this was due to the specific costs of the network versus the nature of the market itself, the researcher ran a controlled comparison. They took the exact same orders and the exact same pool conditions but changed only the cost of the connection, simulating what would happen if the expensive network fees were replaced with the cheaper fees of the faster network. The result was a massive shift in behavior. Under the cheaper cost structure, the number of orders that became worth splitting increased by more than twenty percentage points on the expensive network and nearly nine points on the cheaper one. This proves that the barrier is not a fixed feature of the market but a direct result of the price of access. When the cost of reaching out drops, the market instantly becomes more integrated, and more traders can take advantage of the deeper liquidity available across multiple venues.
Ultimately, this research clarifies a fundamental misunderstanding about modern digital markets. The presence of visible liquidity and the ability to reach multiple venues does not automatically create a unified market. True integration depends on the specific cost of the transaction for each individual order. If the fee to reach a second pool is too high, that pool effectively ceases to exist for that trader, no matter how much money is sitting inside it. The study demonstrates that while technology has solved the problem of visibility, the economics of access still determine whether a market is truly connected. For the designers of these digital systems, the lesson is clear: lowering the cost of connection is the only way to unlock the full potential of fragmented liquidity, turning a collection of isolated pools into a single, efficient marketplace.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.