What Banks Can Learn from the Crypto Flash Crash

View Demo Get Started
What Banks Can Learn from the Crypto Flash Crash

Institutional involvement in digital assets is growing rapidly. But as the dust settles on the recent crypto slump, what are the lessons for banks? In particular, how can banks that offer digital asset trading ensure best execution for their clients at all times, even in periods of high market volatility?   

Over the past two months, two very different narratives have emerged in the digital asset space. On one hand, a growing number of major financial institutions have launched digital asset products or have announced plans to engage in the sector. Yet May’s crypto flash crash demonstrated that institutional adoption is still at a relatively early stage and that the sentiment of retail investors still strongly influences digital asset markets.  

Thus, against a background of strong institutional interest and adoption of digital assets, cryptocurrency markets collectively lost almost $1 trillion in a matter of days. For banks and institutions which provide digital asset trading services to their clients, such a mass liquidation event poses unique challenges.  

Why it happened: Major volatility leads to transaction congestion 

When so many market participants are trying to sell blockchain-based assets simultaneously, the underlying decentralized networks become slow and congested. According to analytics site YCharts, for example, the average confirmation time for transactions on the Bitcoin network went as high as 56 minutes on some days in May. To compound matters, the APIs of almost all exchanges were overloaded at various points in time, leading to service outages during which orders could not be placed.  

This type of volatility could easily lead to serious reputational damage for banks if customers logged in to their trading portal and found it inoperable. Whether the trader wishes to buy or sell, major movements in markets provide big opportunities and threats, so it is even more important that trading functionality remains robust during such events.  

The Solution: Ensuring best execution in unsettled markets

With values rapidly fluctuating and exchanges intermittently going offline, achieving best execution becomes increasingly challenging. Thus, a clear game plan for such a scenario is essential. Banks need an integrated, systematic approach to analysing the liquidity landscape so that they can buy and sell assets at the best possible price, with the lowest possible risk, at each discrete point in time. To adopt this unified approach to best execution, six key technical and operational requirements must be considered:    

  1. Establish market-wide connectivity 
    The most basic priority for a bank at times of market congestion is to ensure that client orders can still be executed. Just as with conventional traffic congestion, the more transport options you have access to, the faster you can arrive. And when it comes to digital asset trading, there are many ways to reach your destination as liquidity is highly fragmented between myriad venues.  

    Rather than relying on one or two exchanges and executing orders manually on an ad-hoc basis, banks need automated, market-wide access to a diverse array of liquidity venues. This means that if one venue is temporarily down, orders can automatically be rerouted to alternatives. A key technical challenge in this respect is that there is no common API standard for digital asset exchanges. Thus, to ensure reliable connectivity and eliminate API maintenance, secondary market solutions exist, which provide connectivity to a range of exchanges over a standard FIX connection.  
  1. Automate trading  
    Even after you establish access to a wide array of liquidity venues, human error remains a significant risk. Indeed, so-called “fat-finger” errors, where an employee mistakenly executes a trade or mishandles a deposit or withdrawal, can cause embarrassing reputational damage. In April, for example, £3 billion was temporarily wiped off Barclays’ market cap due to a typo, when a trader reportedly executed an order at 168p rather than the intended 186p.  

    In the digital asset space, asset transfers to and from exchanges still tend to be executed manually, which increases the risk of similar errors. And unlike traditional markets, once transactions have been validated by the blockchain network, fat-finger errors cannot be cancelled. Thus, banks should aim to automate as much of the transfer process as possible, and incorporate in-built, pre-trade risk checks. This not only reduces operational risk, but also decreases operational overhead and personnel costs.   
  1. Minimize execution risk 
    Particularly if you are an agency trader, speed is of the essence when executing client orders. If trades cannot be executed on time due to occasional blockchain congestion or exchange downtimes, it increases market timing risks and the execution price to deviate from that intended by the client. In the worst case, an execution delay could prevent a client benefiting from a predicted price movement merely due to execution timing.  

    Furthermore, as liquidity is fragmented over many venues and each has relatively shallow liquidity in isolation, banks need to ensure that larger orders are split and distributed so that they do not cause slippage. This can be achieved using a smart order router, which acts like a digital logistics hub for orders, dispatching them in a way that minimizes costs.  

    A related execution risk, which is particularly prevalent in the digital asset market, is frontrunning. This occurs when other market participants gain knowledge of your trading intentions by analysing information such as exchange order books or on-chain transaction data. To mitigate this risk, banks should always use execution algorithms such as TWAP, VWAP, ICEBERG, or SNIPER.   
  1. Minimize counterparty risk 
    Counterparty risk is one of the biggest institutional entry barriers to the digital asset market. As the market is relatively immature, with many recently established exchanges, sometimes the lowest fees come at the highest cost in terms of counterparty risk. This creates a catch-22 situation: should the bank play it safe and pay more fees by routing client orders through brokers while potentially failing to deliver best execution for clients, or go with the lowest fees and accept the long-tail risk of a defaulting venue?  

    This issue is exacerbated by the fact that trading with exchange typically requires orders to be pre-funded through exchange-owned hot-wallet accounts. This magnifies counterparty risk because the bank gives up custody of its assets to the exchanges as long as the keys to the assets are stored on the exchange’s hot wallets. In addition, to use smart order routing to get best prices across exchanges, this risk gets multiplied with every exchange a bank needs to pre-fund trades on. Although a bank can opt to use OTC desks with credit lines as a less risky alternative, this typically involves minimum trade sizes, making it less flexible for agency trading. 

    Fortunately, custody and settlement solution providers such as Fireblocks have now emerged, working on the creation of a unified settlement layer between banks and a wide range of liquidity venues. This allows to implement strict trade pre-funding rule sets that minimize on-exchange exposure through automated settlements, thus marginalizing counterparty risk.   
  1. Minimize liquidity cost 
    Another factor closely related to the pre-funding requirement is the cost of accessing liquidity. Imagine a scenario where a bank wished to access 5 of the largest crypto exchanges. To achieve this, it would theoretically need to set up 5 separate exchange accounts and constantly keep each one funded. Such an approach tyies up funds at different venues and increases liquidity requirements. This is another benefit of using a secondary market service that provides a unified custody and settlement layer. By funding trades just before executing them, it eliminates the need to tie up funds in exchange accounts, and thereby reduces the cost of liquidity.  

    When trading digital assets, it is important to remember the implications of the underlying technology. All transactions of crypto assets like Bitcoin and Ethereum are recorded on a public blockchain ledger. Wallet addresses on such networks are pseudo-anonymous: while the identity of the owner is not disclosed, it could be inferred by examining the transaction history. 
  1. Ensure execution privacy 
    On-chain transactions between banks and exchanges to pre-fund or settle trades are publicly broadcasted. When preparing to execute large orders, this can increase the risk of frontrunning by other market participants which spot an on-chain transfer from a bank’s address to an exchange. In order to ensure execution privacy and prevent frontrunning, workflows are needed that either avoid the broadcasting of trade pre-funding transactions or eliminate the need for trade pre-funding altogether.  

A unified approach to digital asset banking

For banks making their first foray into the digital asset market, the fragmented liquidity landscape poses some significant challenges, risks and costs. During a mass liquidation event like the one that occurred in late May/early June 2021, these issues become even more acute. To avoid significant reputational damage and always ensure best execution for their clients, banks need a solid infrastructural foundation coupled with a clear game plan to manage such situations.  

This is where mission-critical infrastructure providers like AlgoTrader and Fireblocks can play a key faciliatory role. They provide a unified digital asset platform to automate and optimize the entire trading life cycle. This unified, integrated foundation enables banks to respond to growing client demand for digital assets without being exposed to the costs and risks associated with an immature, fragmented market. 

AlgoTrader Logo

Stay informed about AlgoTrader