The world is moving to T+2. The ECB, under the auspices of T2S, has mandated it. And DTCC has produced enough evidence in its research work to validate the proposal.
Nobody doubts that shorter settlement make sound risk management sense. Given the average trade to settlement lapsed time of three days, we reduce the open position in the market by a third if we move to T+2. And we further reduce the funding costs in the market, a much-needed benefit to brokers in those markets where they carry the margining costs of CCPs during the open period.
It is important though that we consider two major challenges of the T+2 environment. The first is in the cross-border markets and reflects the idiosyncrasies of the foreign exchange markets. The second is in the troubled path between trade and settlement.
I have long argued that the traditional two-day lag between trade and settlement in the foreign exchange markets is ridiculous. It dates back to the era when foreign exchange matching was a manual, time-consuming, telex-based process and is out of place in today's much more automated environments. Although same-day foreign exchange or next day is possible in all the major currencies, and especially those settled over CLS, logic dictates that the standard settlement period should move to T+1 for all transactions. T+1 has the advantage over T+0 of being less time sensitive to either side of a transaction, for we cannot avoid the ramifications of the world time zones even in this day of extended trading hours.
The far more serious challenge lies in the chain between execution and settlement of a trade, and especially a cross-border one. This gives rise to two key issues, namely the problem of poor communication between investors and their agents and the challenge of matching and allocating a trade.
Looking at the settlement life cycle of a complex securities trade, ISO messaging standards has improved radically communication between global custodians, their agents in local markets and those many infrastructures that have now adopted this standard or at least the data components that it uses. The major problem is matching and allocating the trade between broker and investor before it can be passed on to the global custodian for settlement.
There is no reason for this. There are, for those trades that are not internalized within the universal banks of Europe, excellent trade matching facilities. The most visible is Omgeo, with the SWIFT Accord platform gaining some traction. Regulators cannot mandate one or another of these systems, but they do have a critical role in the matching space. A capital requirement for all trades unmatched by close of business on trade date would focus minds, automate processes and could well be the catalyst, further downstream, for the elimination of those risky, faxed contract notes or similar communiqués that remain the preferred communication vehicle for a sizeable minority of fund managers.
And then we have a further challenge. How can the matching engines forward matched trades to the appropriate custodians? Such facilities exist, but most parties are wary of them for risk concerns. Given that the matched trade needs to find a compensating and matched instruction in the settlement system, there is an inbuilt safety check in the process as long as certain parameters are considered. Perhaps we need a database of approved counterparties by fund? Or custodians could send, on receipt (mainly on T+0), a message to clients of pending settlements received from the matching engines rather than wait for the inbound from their client? Or alternatively we could add another capital haircut for funds failing to advise their custodian of their trades within a given time frame that is compatible with T+2? Thus, the solutions can vary from the appropriate matching engines acting as source of information for the custodian through to incentives to ensure a prompt routing from the fund manager through to the custodian.
And, whilst looking at the T+2 environment, we also need to consider whether action is needed to ensure efficient settlement, especially in the major settlement hubs of the future. The CSD Regulation, proposed by the EU, will include the concept of settlement discipline. In Europe, especially, with its decision to settle in higher-cost central bank money, we need performance guidelines. Logically these should target a high settlement percentage in T2S' overnight batch to avoid bunching of settlement during the settlement day, driven by settlement agents or T2S participants who wait for inbound funds from sales before settling their purchases.
T+2 is a valid aim, but it needs to be accompanied by measures that ensure there is no deterioration in the quality of markets. Regulators should discuss the options with the stakeholders. None of us wants to see a decline in due-date settlement or a scramble to complete settlement during the dying moments of the settlement day.