Specialising in client services, securities and settlements, Pike was with Morgan Stanley for almost 18 years before he joined Taskize in December 2022. He was instrumental in helping to implement the central securities depositaries regulation (CSDR), which went live in February 2022, and now heads up business development growth for Taskize, Euroclear’s workflow platform that helps the group and its clients resolve matching and settlement problems both before and after settlement date.
What impact did CSDR have and how painful has it been?
I think it’s fair to say that it would have been more painful if they had brought in the mandatory buy-in regime, but they pulled the plug on that three or four months before the regime was due to go live. Essentially, it would have forced a mandatory buy-in for anybody who was failing to deliver a security. That would have put the onus on the person receiving the security to manage and enforce the mandatory buy-in – meaning they would basically have to go into the market and execute a new trade for the same identical security, then cancel the old trade and pass the cost of the new trade back to the failing counterparty. The concern was that that might drive liquidity problems. It would have cost a lot of money as well – millions of dollars for each institution – had it come into effect.
Read More – ESMA postpones buy-in regime under CSDR for three years
Once that element was removed, however, CSDR just became a penalty regime: which is good in principle and a strong incentive for parties to settle stuff earlier. However, there are still problems with settlements in Europe, which has a notably lower settlement rate – or in other words, a higher fail rate – than the US. That’s because of some structural issues that need to be resolved.
What are the structural issues contributing to Europe’s higher fail rate?
In a nutshell, Europe has more central securities depositories (CSDs) than the US. The US just has one, DTCC. Europe has far more, around 10-15, and that in itself presents a barrier, because some people will want to settle in one, and then the party delivering might be delivering from a different one, so you’ve got the problem of settling across CSDs. That then brings about the process of realigning securities in order to settle them. You’ve also got different currencies, and it’s a much more rigid and robust framework. So we have this concept of matching trades in Europe, before you can settle – whereas in the US, you don’t have that. Now, arguably it’s a stronger system with more robust risk mitigation, but it does also become a barrier to settlement.
Is there a link between the CSDR penalties and the volume of equities failing to settle? Is CSDR actually working?
I wouldn’t say there’s a failure of linkage. But I think the problem is that the way the penalty regime works, you essentially get charged as the failing party, or the person who’s deemed to be at fault.
It’s charged at 1bp per day from the day that it fails up until the point where it settles. So if you think about the equity model, when you execute a trade on behalf of a buy-side institution, as a broker or a sell-side institution, you’re essentially charging a commission rate for equity trades that’s somewhere between 5-10%, so 5 and 10bps. If you’re charged a penalty of 1bp per day, and you’re only making 5bps on that trade, you have in theory eaten through your entire profit margin by the time you get to the fifth day. The challenge of the equity market is that most equities settle locally. So if you’re trading Royal Dutch Shell, you’re likely to settle that in the local Dutch market – or the UK market, because that’s where the liquidity is. But someone might be holding that in Euroclear, so you’ve got to realign it across two markets to settle.
But that’s not the biggest impediment. The biggest impediment is that actually, more often than not, people don’t have the full slug of inventory they need to settle an individual trade. So if you’ve got a trade for $50 million and you’re only holding $40 million of the security to settle it, that trade still fails for the whole amount. There is a concept called autopartialling, where you can essentially settle the balance that you’ve got; but that requires you to amend the trade, settle the balance you’ve got and then leave a residual component thereafter to settle it, which presents challenges for buy-side institutions as well as for the custodians that support the buy-side institutions.
For example, if you have a large asset manager who has done a trade for $50 million and they’ve allocated that across 10 funds, and you’re trying to settle $40 million out of that, they’ll then need to work out how to pro-rate that settlement across those 10 funds – or work out how to really calibrate the trade they want to do, then re-book it to settle that $40 million as opposed to the $50 million, and then leave the residual $10 million to settle.
Read More – Trade settlement penalties are high following CSDR, so should the buy-side be receiving more credits?
That’s a big problem for the buy-side. It’s not so much a problem for the sell-side organisations because they just think of it as just one trade to sell to that asset manager. But ultimately, for the buy-side firm, each of those allocations make up 10 individual trades with different legal owners of the securities. That creates a structural challenge.
What can firms do to amend or improve these issues and reduce their fail rate?
There’s a bunch of things they could do. One is simply to manage their inventory better, and communicate better. A lot of the problems stem from the fact that while firms are buying and selling a lot of securities, they usually run a pretty flat inventory – they’re not running excess inventory, because it’s inefficient. In the old days what you would have done is filled your failing trade from your existing inventory but with all the capital constraints that have come in, through the Volker Rule and so on, people are running less inventory because it’s costly. The balance sheet cost of holding long inventory means that it disincentivizes people from doing that, certainly on the broker side of the equation. So what they’re essentially doing is waiting for a trade to settle to enable that liquidity to then provide the requisite liquidity to settle the next ongoing trade that is failing. It’s all a chain – a bit like the housing market in the UK, where you can only buy or sell your house if the chain persists and everyone in the chain settles on time so the money passes down the line. The settlement market is a bit like that – each trade has to settle on time to provide the liquidity to enable the next trade in the chain to settle. So the liquidity of the asset they’re trying to deliver on is ultimately derived from a previous trade that’s settled, it’s not coming from their own inventory. They might be holding it internally, but more often than not they aren’t, and that needs to improve.
The second thing that needs to improve is this concept of autopartialling, which both the buy-side and the sell-side and the custodians should enable so that it happens automatically. That would speed up and improve the liquidity of the settlement process.
Then I think firms like Taskize can help by allowing parties to collaborate and connect with each other to identify any problems with the settlement – there are any number of things that could drive a settlement fail and being able to identify those quickly, communicate the issues and then collaborate across different firms to fix them, is really important. It compresses their timeline around failing trades, which can save a lot of money in terms of time to settlement.