Has your custodian ever lost any of your assets? Probably not. Should you worry about such risks? Surely yes. But how much should pension fund directors get involved in custody issues?
In fact, there are normally no major problems that would require the attention of pension fund boards. Custody seems to run itself. And why should trustees search for more troubles in difficult times? Complacency could be very expensive though. Why?
Governance: Custody of assets is a primary responsibility of trustees as principals and they typically delegate the job to custodians as their agents. Like all delegated relationships, it simply needs to be governed well. All the general guidelines apply: clear objectives, good definition of roles and responsibilities, thorough risk assessment, agreed performance standards, reasonable fees, etc. In practice, however, a lot of this is missing in the real world of pension schemes.
Complexity: In addition, there are some important ito the trustee-custodian relationship. For example, custodial arrangements often involve a range of delegates (eg, fund managers) and sub-delegates (eg, sub-custodians) with often obscure reporting lines. Another factor is the technical complexity of custodial operations. To what extent do trustees need to understand margin trading or OTC settlement procedures?
Risk: Custody risks appear like earthquakes. They materialise rarely but have potentially massive consequences. But how can trustees assess the risk, eg, of losing assets held by a sub-custodian in an emerging market?
Change: Custodian banks are currently undergoing major changes which not only offer new opportunities to trustees but also pose new governance challenges.
What can custodians do for pension funds? Unfortunately, many trustees do not have more than a vague idea. The traditional services of custodians primarily are:
o the safekeeping of assets;
o clearing and settlement of financial transactions;
o income collection and allocation;
o tax recovery;
o cash and foreign exchange management;
o right issues and other corporate actions;
o proxy voting;
o data reconciliation;
o fund accounting;
o record-keeping, reports and statistics.
Since custody has very much become a commodity business with low margins, custodian banks increasingly offer ‘value added services’, such as:
o securities lending;
o performance measurement;
o risk analysis;
o compliance and investment monitoring;
o transaction cost monitoring;
o commission recapture;
o fund administration;
o transition management services.
In summary, the list of actual and potential activities of your custodian is quite long. But it is easy to get confused by details and technicalities. And the last thing pension boards should do is to suddenly move from negligence to micro-management. It is therefore important to simply focus on the main issues. What are the key questions from a trustee’s perspective? What is the current custodial arrangement? Who are the custodians? Is there a direct contractual relationship between the pension plan and custodians? If not, who is in charge and how? Is the money safe (particularly in foreign markets)? Can assets be recovered in case of bankruptcy/fraud/breakdown? Is there protection against losses arising from actions of any custodian or sub-custodian? Are the custody operations running smoothly? Are trustees informed about holdings, transactions, events? Are the custodial services cost-efficient?
Some pension plans seem to have a good picture of their status quo. For those that have not, it would certainly be advisable to go through a general review exercise.
Looking forward, however, that won’t be enough. The fiduciary duties of trustees are being increased in most countries. For example, the UK Myners review has considerably raised the standards for trustees’ governance in investment and risk matters. Custody forms an essential part of this, a fact that is often overlooked. As a result, trustees will have to work hard in particular on four areas of custody governance:
o Formal selection and more regular review processes for custodians. (A standard item for the future pension business plan!)
o Introduce benchmarking for custodians’ performance; implement better monitoring tools; demand improved reporting. (Example: How can you be sure that you are not being ripped off on the forex transactions?)
o Explicit assessment not only of custody risk in a narrow sense but the full range of risks incurred by the different activities of their custodian banks. (Is securities lending understood?)
o Increase trustees’ own understanding of the (technical) issues involved in custody and related areas. (My kingdom for a good jargon-buster!)
Many of Myners’ messages can be seen as ‘modern best practice principles of pension fund governance’ and are as such generally applicable. Of course, they will need to be applied within the particular framework of national regulation and the specific pension plan rules. In fact, custody arrangements vary considerably across pension funds in Europe. They also depend on the type of pension plan (for example, DB or DC, big or small fund), the specific investment arrangements ( in-house or external, single or multi-manager) and the asset classes invested ( alternatives).
In the past, pension schemes often had an arrangement under which their (single) investment manager either provided custody services itself, or delegated it to an internal or external sub-custodian. In recent years, segregated multi-manager arrangements have become more widespread and it has also become more common for pension fund trustees to select their own main custodian for all their assets. Going further, many international companies aim for a single global custodian.
As a result, pension plan sponsors/ trustees increasingly enter into separate agreements with custodians (or tripartite agreements also including fund manager). This can be difficult new territory. Custody agreements should not only specify the services and quality standards in some detail but also cover such areas as risk controls, insurance cover, indemnity against potential liabilities, etc. Additional services, such as securities lending, need to be clearly laid out as both financial and legal risks need to be well understood.
It is clear from this that trustees will need to develop a forward-looking policy on how to optimise their own custody arrangements. What services do you really need, and from whom? For example, moving to a single global custodian could offer advantages in terms of consistent reporting, performance/risk analytics or pricing. On the other hand, such move could disrupt long-established relationships. Some fund managers still prefer established niche players particularly for smaller markets or specialist assets.
Once Pandora’s box is opened, trustees and sponsors face difficult top-level decisions. A tricky area, for example, is shareholder activism and voting. What role will the custodians be given in future? Another example is transaction cost measurement. Should trustees use their custodian or an independent service? The list of questions is far from exhaustive.
Finally, and fortunately, the actual selection process for a custodian is an easier matter. Much of it is similar to the selection of, eg, a global fund manager and external help is easily available. Clearly, technology, economies of scale, quality and speed of administration are important ‘hard factors’ but so are ‘soft factors’ like reputation, organisational commitment and the trust in the people involved.
All the points and options raised leave pension boards with a final nut to crack: Do they have the resources to do what is deemed necessary and important? How to re-structure the pension fund in order to move things forward? What short-cut can small pension funds make?
Georg Inderst is an independent consultant based in London