Another train journey – this time heading down to London – and another opportunity to have a few uninterrupted hours to focus on work. These journeys have proved to be quite fruitful in terms of producing a number of my blog entries, so here we go again.
It’s quite hard to blog and hope to stay interesting. The choice of topic is always difficult, with the main objective being to write something that people want to read. If you’ve read my previous blogs, you’ll know that I can be a bit ‘ranty’. One of my friends recently called me a ‘grumpy old man’ on Facebook, and I think the label fits! However, my ‘grumps’ tend to occur when I think people are being short-changed in some way. And one place where I think this has happened has been in the custody arrangements for most pension funds.
Custody related matters are relatively close to my heart; I know, I’m strange. I’ve always found this area to be of interest, since it tends to be ordered and neat when compared with the huge number of variables that come in to play on the investment management side, and that order and neatness appeals to my structured side. But it’s changed just as much as the investment management world in the last 20 years, and the potential has emerged for pension funds to be short-changed in a variety of ways:
1) Processing Asset Transactions: The physical holding and settling of assets has changed a lot, even over the relatively short period of time of my own financial career. In my first job, I remember being responsible for settling UK stock sales by sending an actual share certificate and stock transfer form off to the broker. With the advent of CREST, the UK Central Securities Depository or CSD, most UK equity and gilt paper certificates were ‘dematerialised’, Dr Who style, to reappear on the electronic register of assets. So, custody of these assets changed from physically holding them in a vault and manually settling them in the manner described above, to it all being done electronically. The cost of undertaking this function fell dramatically but, in my view, it’s taken a long time for this reduced cost to be reflected in the charges passed on to asset owners. I used to manage a direct access to CREST, so I knew what it actually cost to settle a UK trade – and pence rather than pounds was the charge. It’s true that, for some markets, it remains expensive to hold and settle assets, but this usually relates to exotic overseas markets where there is no CSD. So, most LGPS Funds shouldn’t have been paying as much as they were for pure custody, since the assets were mainly held in ‘non-exotic’ or CSD enabled markets where transactions were cheap.
2) Holding Assets: Whilst there are a lot fewer share certificates for custodians to hold these days, there are still lots of assets to ‘hold’ and account for on a daily basis. The electronic nature of most of the holdings makes life a lot easier, but the way that they are typically held is more for the convenience of the custodian, than for the benefit of the client. Many custodians still hold assets in commingled ‘omnibus’ accounts – where the assets of many clients in a single security are held together. There are a few reasons why some custodians might want to do this, but operational efficiency when it comes to securities lending and voting tend to be quite high up that list. However, as a client, there are risks associated with custodians taking such an approach that are not necessarily immediately obvious – for example, if an omnibus holding of a single stock is ‘over-voted’ i.e. a ballot is submitted to the appropriate registrar for a number of shares that exceeds the registrar’s expectation for that account, then the whole ballot will be rejected, and all votes will be lost. How often this happens is anyone’s guess, since most custodians don’t report performance in this area, and you’d need to phone a lot of registrars to get a big enough picture as to the scale of the problem. It’s probably best that I don’t get started on the nature of the outsourced provider of voting services for many custodians – I can feel a huge rant coming on!
3) FX – an area where a number of custodians have, speaking bluntly, been caught with their hand in the till. A simple Google search can deliver some of the background to this charge, but essentially some global custodians have been less than fair in their treatment of clients who asked them to undertake FX activity on their behalf. Whether it was an actual FX trade related to settling an asset purchase, or an ‘automatic’ FX trade to facilitate the conversion of overseas income or corporate action proceeds back into domestic currency, the spread applied to the FX rates have been very generous to the custodian, and less so to the client. How do you ensure that your fund is getting a fair deal when it comes to FX transactions with your custodian? That’s a good question to ask them directly, or indeed your investment consultant for them to give a view on FX ‘best-execution’.
4) Securities Lending: Most pension funds escaped any kind of problem in this area in the immediate aftermath of the Lehman Brothers’ collapse, but then again the non-cash collateral aspect of the lending system worked in the instances where Lehmans were the borrower of assets. Once they were declared in default (i.e. unable to return the asset borrowed), the collateral provided by them against the original loans was sold, and replacement assets were bought for those clients affected. However, some funds had elected to also accept cash collateral as part of their lending programme guidelines, as well as non-cash. This cash collateral had to be invested somewhere to generate a return, and some cash reinvestment vehicles employed by custodians were invested in asset or mortgage backed securities. This resulted in a loss in the cash collateral, which fell on the funds to ‘manage’.
5) Branches vs Sub-Custodians: Does your custodian have mainly branches in its worldwide network, or does it have sub-custodian agents? And do you care? It’s arguable whether there is more comfort in having one rather than the other, but to me having a branch network in place should be more reassuring, since the link back to the parent organisation is very clear. Having a network of sub-custodian agents might seem like a more cost effective approach, but when the crunch comes, how safe are your assets? And how much time and effort can you as an investor spend ensuring that your main contracting party has put in place a rigorous and detailed sub-custodian selection and monitoring system? What liabilities will they accept, and what will they try to exclude, for your assets held by their sub-custodian agents rather than those held by a branch?
I’m currently doing some work with a custodian on an ongoing basis, so – if you already knew this – the blog might seem like an advert to point out the problems with most custodians in an attempt to generate some interest in my ‘client’. Those of you who know me, and know of the custody work I did in my previous role, will recognize the points above as being ones I’ve been concerned about for a number of years, and any good investment consultant or custodian selection company will know about them, and factor them in to any review that a client might undertake.
Why the blog then? Because I’m irritated that things have not moved on in ways that they should, for the equitable treatment of pension fund clients. A bit of pressure every so often might actually bring about change in these areas where clients have been short-changed in the past, and indeed the present.
Don’t get me wrong – as a capitalist (albeit a grumpy one these days!), I fully understand and support the need for custodian banks to make money somewhere – our assets need to be safe after all, and the financial health of the organisations that look after them is important to us all. However, that doesn’t mean pension funds should be seen as a soft touch, and it’s important that funds understand the areas above when reviewing their custody arrangements.
Grump over…until next time!
I'm busy working on my blog posts. Watch this space!