Who is offering to custody what for whom?

Download the book

How is the digital asset custody industry evolving?

The origins of digital asset custody lie in the initial cryptocurrency boom of 2015-2017. Retail investors needed digital wallets to hold the private keys to their coins, and the cryptocurrency exchanges provided them. Independent digital asset custodians, and vendors of independent digital asset custody technologies, proliferated in 2017 and especially 2018, as the first cryptocurrency bubble inflated and then deflated.  At the time, these so-called “crypto custodians” were servicing mainly retail clients investing their own money in Bitcoin, Ether and other cryptocurrencies. The established global custodian banks did not come under serious pressure to provide a digital asset custody service until pioneering wealth managers, private banks and asset managers started to invest in cryptocurrencies with money that was not their own. Managing and safekeeping assets for third parties demands not only a higher standard of service but entails meeting regulatory obligations as well. This demand for digital asset custody services of institutional-grade quality, provided by regulated financial institutions to end-investors and asset managers active in the traditional financial markets already, will only increase as securities and funds are tokenised. This is a major factor in attracting institutional money to digital asset markets, because the custodian banks are already servicing end-investors and asset managers in their traditional business. Those global custodian banks that have engaged with digital assets have also identified opportunities for businesses other than custody, including payments and debt and equity capital markets. Digital asset custody technology vendors report interest from custodian banks in integrating digital asset activities with other services provided by the bank. Interestingly, this will undermine a conspicuous advantage of independent, specialist non-bank custodians, which have succeeded partly because they can offer customers access to asset services such as staking and lending as well as trading. At regulated banks, trading and lending tend to take place in siloes that are separate from custody and regulated in a different way from custody. Integration will blur those siloes.

Should regulated custodian banks offer cryptocurrency staking, lending and borrowing services?

At present, the division of labour between non-bank, specialist digital asset custodians and traditional custodians means that regulated banks experience less demand to provide cryptocurrency services. Conversely, non-bank providers lack the regulatory licences to contest asset servicing by banks in the traditional financial markets. The division of labour is likely to persist. The servicing of tokenised securities and funds will continue to be dominated by banks and the servicing of cryptocurrencies will continue to be dominated by non-banks. That is because few non-bank custodians have yet obtained the regulatory licences they need to provide services under extant securities and fund laws.

Will the banks displace the independent digital asset custodians?

There will be convergence rather than displacement. The current fragmentation of digital asset custody services, and especially the bifurcation between the independent specialist “crypto custodians” and the global custodian banks, will diminish over time.  Both will converge on a single regulated model. Regulatory measures such as the Markets in Crypto Assets Regulation (MiCAR) of the European Union (EU) are encouraging convergence already by enabling independent specialists to secure regulatory licences and allowing traditional global custodian banks to expand their services - initially into cryptocurrencies – without taking undue compliance risk.

Which custodians are securing the biggest market share?

The data needed to make a confident assessment does not yet exist. The market is not big enough for any provider to achieve genuine, measurable scale. But the opacity is also deliberate to the extent that many potential digital asset custodians have yet to go live with their offerings. When it comes to market share, the bifurcation of the industry between “crypto custodians” and global custodian banks with conventional custody businesses and ambitions to grow revenue from servicing tokenised securities and funds as well as cryptocurrencies also makes it difficult to compare like with like. Only Coinbase, as a public company, publishes a regular valuation of customer assets in custody in its quarterly and annual reports (see the Chart below). These assets consist largely of the major cryptocurrencies (Bitcoin, Ether, Solana), Stablecoins and cash used by their customers in cryptocurrency trading and financing activities. Coinbase, which has secured business from the issuers of spot Bitcoin Exchange Traded Funds (ETFs) that traditional custodians might expect to secure, is undoubtedly the market leader in the cryptocurrency custody field. But it is better compared with digital asset custodians such as BitGo and Copper (which also support clients trading cryptocurrency across multiple venues) than traditional global custodians such as BNY or J.P. Morgan (which are more inclined to support buy-and-hold investors). In the longer term, cryptocurrency clients are likely to become less important than investors in tokenised securities and funds.

However, the industry is likely to remain relatively fragmented for the foreseeable future, with different providers servicing different asset classes and market participants, if only because one-stop-shops are too ambitious at this stage in the evolution of digital asset custody services. Convergence will occur, and market shares shift rapidly, once the digital asset markets (Decentralised Finance, or DeFi) mature and become interoperable with the conventional financial asset markets (Traditional Finance, or TradFi). That is not just because global custodian banks and financial market infrastructures such as central securities depositories (CSDs) will bring existing relationships and business volumes to bear but because their involvement will accelerate the tokenisation of securities and funds.

What is the safest choice as a digital asset custodian?

The audience was bearish about cryptocurrency exchanges. This is an obvious lesson of the past, from Mt. Gox to FTX, which the “prime brokerage” products of certain digital asset custodians are also designed to address by keeping encumbered assets in custody with an independent, bankruptcy-remote entity. Yet somewhat counter-intuitively, the audience still favoured independent non-bank specialists over regulated banks (see the Chart below). This bias reflects the continuing domination of digital asset markets by cryptocurrencies, where the involvement of banks in custody remains limited. In other words, regulated banks are not yet servicing the cryptocurrency industry, and especially not its more esoteric asset classes, or providing unfamiliar services such as staking, chiefly because they lack the regulatory certainty to proceed confidently. The value of the assets to be custodied on behalf of most market participants is also trivial, both absolutely and relative to the size of the portfolios they maintain in traditional markets, so anxiety about the resilience of the custodian is less intense. Again, regulated banks and financial market infrastructures with banking licences are likely to become a more popular choice as the regulatory environment becomes clearer. Ultimately, firms managing money for third parties, which treat digital assets as a commercial opportunity like any other, want to work with regulated banks because they value a bank as a counterpart in their existing business. In effect, the independent, non-bank specialists are an adequate choice for early adopters content to speculate rather than invest. Institutional money seeking consistent profits will always favour regulated banks.

What services are regulated banks developing?

Virtually every custodian bank is exploring digital asset custody, and several have launched services. By purchasing digital asset custody technology from vendors that have developed state-of-the-art systems in the cryptocurrency markets, banks are becoming formidable competitors to independent non-bank specialists. They combine technology with the assurance of a bank with a banking licence and a balance sheet. Partnerships between global custodian banks and digital asset technology vendors are commonplace already. Indeed, banks are often content to leave custody of esoteric digital assets to technology partners, which act as sub-custodians while the bank focuses on, say, the top three or five cryptocurrencies. However, this sub-custody solution is limited to cryptocurrencies. It cannot work with tokenised securities and funds, whether they are native or non-native. Custodian banks will be concerned about their responsibility to keep customer assets safely and their liability for lost customer assets if a token is issued, settled and safekept on a decentralised blockchain, and traded between parties that hold digital wallets on other blockchains. Decentralised networks of networks of that kind make it difficult for custodians to monitor the transactions that determine who owns what. In traditional markets, custodians value CSDs precisely because they provide a centralised venue where all securities are issued, all parties can settle transactions, and a register of all owners is maintained. Custodians have always struggled to support assets that are not issued and maintained in a CSD.

Could CSDs help custodians develop digital asset custody services?

If custodians value the work of CSDs in traditional markets, an obvious implication is that CSDs could accelerate the growth of the digital asset markets by making it easier for regulated banks to provide digital asset custody services. CSDs would, by definition, improve integration between the digital and traditional asset markets. They could overcome a major retardant of progress - the high cost of adopting an altered or new operating model – by providing their existing users with a familiar point of entry to the digital asset markets. In this context, the experiments with new CSD models being conducted in the United Kingdom Digital Securities Sandbox (DSS) set up by the Bank of England and the Financial Conduct Authority (FCA) are unusually important. Their long-term success will depend on adoption, and that depends in turn on ensuring any new models that emerge from the DSS are not restricted to a small proportion of the potential market in digital assets. One obvious opportunity for CSDs, which some are already exploring, is to extend their repo services to digital assets, enabling banks to raise commercial bank money and central bank money against digital asset collateral. That would create a direct and profitable link between traditional financial and money markets and digital asset markets, with the potential to enhance liquidity in both. The risk for established CSDs is that, if they do nothing, others will emerge to provide the digital asset equivalents of the services they provide in the traditional markets today: issuance, settlement, registration, asset servicing and financing. Specialist digital asset custodians are already providing the “prime brokerage” services that enable cryptocurrency market participants to post cryptocurrencies, staked cryptocurrencies and tokenised money market funds as collateral across a range of trading venues hosted on private, public and public permissioned blockchains without the assets ever leaving custody. Even if a CSD in a national market does not provide all these services it could act as the “orchestrator” or monitor – what Blockchain Law IV in Luxembourg calls the “control agent” (1) - of their provision to market participants by other service providers.

Does self-custody ever make sense?

Self-custody is sometimes the only choice. An end-investor seeking exposure to digital assets via a venture capital fund or fund of funds invested in promising but risky blockchain start-ups, for example, will struggle to find a custodian at all. It takes months to complete the due diligence on each Layer 1 blockchain or Layer 2 network and then insulate customers from the differences between them, so custodians cannot achieve the necessary scale quickly. In addition, some hedge and proprietary funds trading their own money have built custody solutions in-house. But as soon as a market participant raises external capital, or starts to manage money for third-party clients, self-custody ceases to be an option. 

What will the digital asset custody industry will look like in 2030?

By 2030 the different types of custodians will have converged on a common operating model, and the consolidation of the industry will have begun. More capital will flow into the industry. Partnerships between specialist providers will flourish. All digital asset custodians will benefit from a wider degree of regulatory clarity and consistency across jurisdictions. Services will be provided mostly to customers on permissioned public blockchains rather than wholly private or wholly public blockchains. Blockchain technology, including standards to facilitate interoperability, will be much less prominent in industry discussions and the types of digital asset being issued, traded and safekept much more prominent. By 2030, atomic settlement will also have reduced settlement risk in capital markets to zero.

---

(1) Under the fourth iteration of the Blockchain Laws passed by the Luxembourg government, a “Control Agent” can be appointed. Its duties include monitoring the digital assets held in custody and reconciling what is held with what is issued. Maintaining the integrity of an issue in this way is a classic function of a CSD.