COLUMN – OLAF RANSOME | Plans to shorten the settlement cycle are now locked in for Europe. Every operations team is going to have to do something to adapt for what is coming. This month, Olaf writes about what that is and how those in operations should prepare for the changes in Europe.
Through a series of six column articles for PostTrade 360° in 2025, banking operations veteran Olaf Ransome digs into everyday operations – seeking to help us understand some of the everyday challenges and how to master them. Find Olaf’s articles indexed here.
The T+1 thing has happened in the US. It will happen in points east of the US, too. To help me with a better view of the facts, the challenges, and to get insight on what the industry might do to help us all, I spoke to Simon Bennett. He is an industry veteran; he’s worked in big places and big consulting firms and is now an independent consultant.
What’s happening?
While T+1 has already happened in the US, in Europe, it will come in November 2027. Here’s how to think about this. Inside any bank, there is a book-of-work called Change-the-bank. That book has only two sections: mandatory and discretionary, or must-do and may-do. As far as budgets go, the second of those sections for the most part gets the leftovers after the must-do stuff has been agreed on.
Big bold statement
“There is no business case for this change, but the industry will have to play the rule of rule taker on this one,” says Bennett.
On things operations and process, like me, Simon has done more than his 10,000 hours. When we talked, he suggested that the purported savings on the funding side, allegedly coming from reduced intra-day liquidity expense, are not what we think they will be.
This makes sense: if your net settlement position is a requirement for say EUR 10mm, if that is required on T+1 or T+2, the effect is the same.
So, you know you’re going to do it. But what, and how to get the money to make the changes needed? With mandatory projects, the standard operating procedure (SOP) is three steps in three Ds: denial, doubt, and do just enough to get a tick in the box.
Denial is telling yourself, “this is never going to happen, so don’t spend time on it”. Doubt creeps in when it is apparent that something will happen, but the date may get pushed out, to which the knee-jerk big place response is, “it will be delayed, we don’t need to do anything yet”.
The final movement in the standard three-step is, “do just enough to get a tick in the box”. This is where there is a very engaged three-way between Business, IT, and Ops. Having been in denial and planned on a delay, some agreement gets found on the bare minimum needed to get a tick in the box. Broadly, this equates to being compliant – i.e. effective, but almost certainly without being efficient. Normally, this means Ops does not get the right tool, but gets consoled, or fobbed off, with a “we’ll fix all those things later, in version 2.0”. As those who have been around know, this is the third biggest lie of all time.
Bennett shared a few stats to hint at what Ops teams can expect. The US is a very homogenous market. Typically, fails, i.e. trades not settling on time, have been around the two per cent mark, versus some five to eight per cent or so in Europe. The US experience to date has been that fail costs have increased 18 per cent. In Europe, we do not have a single flavour as they do in the US, so we would rationally expect the impact in Europe to be 2 times or even three times.
There are three “suspects” or causes of fails as we shorten settlement cycles:
Broker shorts: Brokers can’t settle because they have not received the stick and are unwilling or unable to borrow it to make delivery.
Standard Settlement Instructions (SSIs): This is largely just a case of bad data management. Today’s market practice is that every individual firm has to find out its counterparts’ SSIs, validate them, and then manage them.
Place of Settlement (PSET): This is about where a trade will settle. In Europe, there are multiple options.
An extra factor in the “short” category is securities lending. In days past, before the Central Securities Depositories Regulation (CSDR) and fines for late settlement, when a stock was trading special and there were good fees to be made from lending, there was an incentive to knowingly fail a trade and simply keep collecting the stock loan fee. CSDR has put the kibosh on that one, but T+1 will force both lenders and borrowers to up their game. Lenders will need to sharpen up their intraday inventory or collateral management processes, especially if they use third party lending agents. Borrowers will have the same challenge; they too will need to sharpen their intraday inventory management processes.
I challenged Bennett on things SSIs and allocations and will admit to learning a lesson. To my process-centric way of thinking, asset managers know or ought to know which fund they are buying or selling for and how much, so the whole thing should be a simple flow. Bennett put me right: first, there are rounding issues when executions are broken down, second, many funds have a minimum buy/sell amount of shares, and lastly, simply dividing a big order into many slices is likely to be counterproductive on at least two dimensions. The first is that simple old adage that size matters; big blocks are easier to execute. The second is that too many orders to process can be a bad thing, simply overwhelming the processing systems’ ability to cope. Add to that for those using cloud solutions, where the number of messages being handled is a component of cost.
How might you help yourself?
For a tier two broker, Bennett estimates that a complete programme to cover all the implications of T+1 would need an investment of some USD15 to 20 million. To come back to the three Ds above, you know you are not going to get the money to do everything. If you are the Ops person who has to deal with the change, how might you make the case for some money and just be able to get on with things?
Understand the requirements: The UK taskforce, which is well aligned with the EU and the Swiss, has put out twelve key requirements and 26 recommendations.
Prioritise: Start with those twelve. What do you think you need and if you did it, is there any value beyond purely T+1? Figure out how to express that in terms of control, capacity, and cost. Make the case to do those things first and be very clear that they are mandatory and on what needs to be done in 2026 versus 2027. It is probably too late to win a battle for any discretionary budget in 2026; typically, the mandatory work starts out at 120 per cent of the total available change-the-bank budget.
Assume more fails: Do a “pre-mortem’. Assume T+1 happens on time. Assume fails go up, at least initially. What happens to your operational capabilities if this is 2x, 3x, and 5x? What do you need to do to be ready and which of those things would benefit the rest of your operations?
Help the industry: Every Ops team has to play the hand it’s dealt. If trades are settled by sending a messenger around on a bike to collect shares, or If we have to write to a registrar, then today at least, that is what we have to do. Today, or in the very short run, no one firm can change the process. To put it another way, you can’t fight city hall.
In financial services, the industry needs to change together. After all, as I have joked about over the years, I have never met a trade which could settle itself. As Bennett and I talked, we identified two things the industry could do to help itself and all of us meet this challenge:
A pro-forma / generic business case: Now of course, every firm is different. Even if 10 firms buy the same software, they will implement it 13 different ways. Every Ops manager would be helped if there were an indication of the general direction of travel as far as the impact goes. Even a clear statement that says the regulatory expectation is that there are 12 things which each market participant must have in place and there is no specific business case for them, this will help change managers articulate the need to support the change programme.
Legal Entity Identifiers (LEIs), not SSIs: As an industry, we spend a considerable amount of time, and therefore money, properly managing SSIs. Every firm has to source, validate, and store the SSIs for every combination of counterpart and product. Why? Because that is the way we have always done it and our messaging standards require us to say, “I want to settle this trade with X, who has instructions Y for currency / product Z”. Now, what if all we needed to know was the LEI for our counterpart and have some central oracle to manage that? Then each entity would manage their own data. It would be no different from mobile phones – all I need is your number; I don’t need to know if you have changed your provider, and you don’t need to tell me. Yet, we operate like we did when I was a kid growing up in Stevenage – when we moved house to the other side of town, we have to let everybody know our new phone number and address. Some years back, I suggested to SWIFT that they could change the industry and use LEIs – it was thought too hard. Now we have Web 3.0 wallets. We could do better. If we want to.
