By Hugh Madden & Liam Bussell
In regard to cryptocurrencies, the prevailing reference to “Centralized Exchanges” is arguably misleading and inaccurate as there is no transfer of legal ownership or title taking place when participants (i.e. account holder) act to buy or sell BTC on a digital “exchange”. There is also no record of the participants’ actions on the blockchain and there is no back office settlement of fiat between the participants. Rather, when selling a BTC that was previously deposited in an account with an “Exchange”, the seller has operationally and economically traded a “BTC IOU” issued by the Exchange for a “fiat IOU” issued by the same Exchange, and the transaction is recorded internally in the Exchange’s accounting ledger. Existing “Centralized Exchanges” in a sense are not exchanges at all.
With current/existing centralized crypto exchanges, users hold fiat and BTC IOUs issued by the exchange in the form of their account balances and do not have actual possession of the underlying assets. Depending on the terms and conditions of the Exchange and whether it operates segregated accounts, in some instances the BTC IOU is not even for a claim on a BTC but rather for the fiat value of the BTC in a liquidation; and as previously seen the Exchange may not even have the BTC on hand to settle all of the outstanding BTC IOUs that it has effectively issued to account holders.
The counterparty credit risk of these IOUs is the crux of the problem with unregulated centralized exchanges. The lack of operational and financial transparency with regard to these exchanges prevents the users to properly assess and evaluate the counterparty risks and more importantly to appropriately price this assumed risk. As it has been shown time and time again, in black swan events, bankruptcies etc., when there is a lack of information and transparency, market participants generally tend to significantly underprice the risks they are taking on and therefore are not making optimal decisions and are not being adequately compensated for those suboptimal decisions. On the other hand, in return for assuming these risks, these exchanges provide valuable functions to users — they provide a platform for price discovery, varying degrees of liquidity and a means for the exchange of economic exposure. Participants also avail themselves to a number of critical and practical services — they are able to transact in fiat, store fiat and crypto with varying degrees of security, execute foreign exchange transactions, send fiat remittances, resolve disputes, access help desks, make withdrawals or purchases via debit cards, comply with AML/KYC requirements, etc. And while these services are critical to a functioning exchange ecosystem, it is likely that participants are paying too high an economic price for these services as they have underpriced the counterparty risk of the IOUs they have assumed due to lack of transparency.
Beginning in 2016 and accelerating through 2017 there have been an increase in the number of initiatives to launch “decentralized exchanges” with the primary and oftentimes sole objective of eliminating the counterparty credit risk exposure participants assume with a centralized exchange. The premise of many of these initiatives is that through the use of smart contracts, participants are able to maintain possession of assets and by being able to transact via the smart contracts, counterparty credit risk is eliminated — and this is true for the trading of crypto pairs. But the experience of the decentralized exchanges that have been launched to date is that there a significant lack of activity and liquidity on these platforms as not everyone is desirous, or is able, to trade only crypto pairs — these exchanges don’t work and are broken. In the utopian quest of eliminating counterparty credit risk what has been ignored and shunted aside is that the critical and practical services provided by centralized exchanges are needed for a functioning exchange system. A functioning exchange system is a microcosm of an economy, and all economies are networks — while originally applied to telecom networks Metcalf’s law states that “The value of a [telecommunications] network is proportional to the square of the number of connected users of the system.” As network externalities are critical for all networks to be functional and of value, decentralized exchange initiatives that limit or ignore the non-trading functionalities and ancillary services otherwise provided by centralized exchanges are doomed for failure.