This is one of the projects in Harmony Hackathon and I would like to ask for your feedback, before we will move to much in the wrong direction.
90% of trades on the interbank market are done between market makers and market takers that have some sort of relationship with each other. Order Book-style trading is relatively limited because there are very few market participants that know where the real market is in a particular millisecond. Market Makers aggregate feeds from multiple sources and offer prices to the end customers. Thank “last look” Market Makers are able to check if the market price that they quoted has not changed to much. During the last look Market Makers can auto-hedge flow before giving final confirmation to the end customer. Every big market taker has just 5-7 very good market makers (having more would have a negative impact on the price). Smaller market takers usually just use single bank portals to trade FX. Thanks to this structure bid-ask spread on OTC FX SPOT/FX Forward/FX SWAP is < 0.001% on the most traded pairs.
With most crypto volume happening on CEX, flow on DEX platform is mostly related to arbitrage. To confirm, you play around with our P&L report of LPs on Uniswap v3: Dune Analytics - in most cases LPs would be better off, not providing liquidity to Uniswap v3 and just keep the initial ratio of tokens that they put in because Fees are smaller than loss related to the change in token ratio. From our perspective, long term it makes sense to provide liquidity to DEX only on pairs, where a particular DEX is a primary market for a particular instrument or at least if arbitrage flow is smaller than real client flow.
Our interbank experience is telling us, that even Harmony finally of 2 sec, if not enough to make on-chain DEX a leading market for most trades crypto pairs. With growing knowledge on the market about the profitability of market-making strategies on AMMs like Uniswap, it might be that liquidity on these platforms will slowly diminish.
Taking all the above into account there is a need for a new type of blockchain-based trading.
One of the solutions that we see in the long-run, is a split between the low-latency price discovery from the blockchain-based settlement. This would bring back the blockchain role to the one it was initially designed for (a double-spending problem for digital currencies described by Satoshi Nakamoto)
The general idea of how to do combine on and off-chain workflow for the above solution can be taken of 1inch limit order protocol. To achieve the smallest possible latency between market taker sending execution requests and market maker sending trade to the blockchain for settlement, there is a need for some additional authorization workflow on both ends. To achieve decentralization, front-ends should be run IPFS/ethswarm.org style hosting.
I