The changes resulting from the EU’s Central Securities Depository Regulations and the emasculation, within Europe, of their core settlement role by T2S, has led to much soul searching by the CSD population and brave talk about capturing activity from the domestic and cross border custodian service providers.
Some CSDs have sought to increase their infrastructure footprint with new CCPs and bespoke trade reporting utilities being the most favoured directions. But others believe they can capture the asset servicing role, and more, from the custodian populations.
The CSD is a generic term and covers a variety of structures ranging from the global custodian-light offering of the ICSD fraternity to a basic notary and securities only transfer function of some CSDs. Many CSDs are about as valuable as third rate country flag carrying airlines. Those have poor security, limited product, high cost and genuinely little value other than the local language service they offer to their indigenous clients. Many are well structured and have the critical mass to provide secure environments, linkage to central bank or secure commercial bank money, base themselves on sound laws protecting title and settlement finality and act as an information conduit on issues ranging through securities numbering, corporate events and governance.
I recall many years ago having a discussion with Iain Saville, then CEO of CrestCo. Ltd, the UK CSD on whose board I sat. We had agreed to allow Crest to offer services to institutional investors and had authorised them to offer a global nominee service to clients. I was asked if this would be the death-knell of the UK custodians. With hindsight, my response was totally correct. I stated that I wanted non-bank financial institutions to have direct access to the CSD as I believed in free markets and not restrictive practises. I also noted I had stated that Crest would find it hard to attract a meaningful client base to its international offering, given it was a non-bank and non-risk taker. I suggested it would be sub optimal in performance terms, limited in product scope and would expose users to timing risk. And, although at the time I rightly believed the death-knell had already sounded for many UK domestic custodian banks, the cause was not the competitive heat from the CSD but those banks’ inability to understand how the market had changed and would change even further and faster.
So why did I believe, and why do I still believe, that CSDs would be sub optimal in performance terms, limited in product scope and high risk for a broader range of users. I would stress that I believe we need consolidation of national CSDs and the creation of larger groupings in that space. I see the notary, information flow management and settlement roles of the well-structured CSD as being robust and of extreme value to the market. I just do not see them usurping the key value added of the more broadly based and better capitalised custodian banks.
Custody has three core client pillars. These are non-bank financial institutions, broker or principal dealers and asset gatherers. The non-bank financial institution uses the custodian for basic custody and settlement but also fund administration, transfer agency, banking facilities and balance sheet credibility. The broker or principal trader is looking technically for settlement services but in reality also for liquidity, credit, foreign exchange, derivative and asset finance support. The asset gatherers are, more and more, looking for basic cash, derivative and securities support in a consistent form and with clear acceptance of liability across multiple markets rather than a single market.
CSDs are not banks. Even the ICSDs are limited purpose banks which have to protect their capital by avoiding most risks beyond settlement related support. And no CSD has an adequate capital base or potential source of new capital, even where it is technically a bank, to leverage its capital for any meaningful client in any meaningful way.
CSDs are utilities with a structure that has to be uniform across clients and pays little, if any heed, to client preferences and requirements beyond its core offering. This is justified as CSDs have to be able to limit their liability, often to relatively modest levels, even where it could be argued that a loss related to errors and omissions that should be within their control. CSDs have problems giving an opinion or technical support rather than passing over information about corporate events, proxy voting and other activities that are critical to the NBFI market and others.
And CSDs have a different client role and ethos to commercial banks. First, their relationship with clients is limited. An NBFI may use its commercial bank as a network to distribute its funds, a utility to collect its premiums or regular savings from clients, a source of bridge funding, a trading venue, a taker of deposits, a clearing house and many other activities. The commercial model will be different, and more advantageous to the NBFI, than the narrow one of the CSD. A broker or principal dealer is not just a holder of assets, seeking to lend any long position and cover any shorts. Liquidity is their life blood, from intraday to overnight, and they need commitment in one form or another that such liquidity is available on a term basis. This may be a challenge under Basel III but it will remain a core component of the bank to broker or principal dealer business model. It is also a risk model that not even ICSDs can emulate. And among asset gatherers, the sheer complexity of multiple domestic CSD links is mind bogglingly complicated, when set against the standard multi-market contracts of the custodian world. And the multiple market standards, conflicting legal systems, use of local language or challenge of remote cash are further impediments to a global surge to multiple market operator or direct CSD holding models unless the asset gatherer has a meaningful presence in the relevant local market.
Most analyses of remote direct market participation, other than through the ICSDs for international securities where they are in a unique position, would point to deficiencies in the scope of offering, increases in risk and incremental cost. In the end, even if one could create a positive running operating model, it would not encompass event risk. And the first local penalty fine, the first expensive missed corporate action, the first important failed trade or the call for mandatory new funding from all direct participants of an impaired CSD would be such an event. And, unfortunately, for the decision maker involved, such an event would clearly be a sack-able offense!
Are CSDs the new custodians?
The changes resulting from the EU’s Central Securities Depository Regulations and the emasculation, within Europe, of their core settlement role by T2S, has led to much soul searching by the CSD population and brave talk about capturing activity from the domestic and cross border custodian service providers.