Not all Settlement Risk is created equal
The Bank for International Settlements (BIS) has produced its Triennial Central Bank Survey of Foreign Exchange and Over-the-counter (OTC) Derivatives Markets continually since 1986, but in 2019, for the first time, it collected settlement methods alongside its traditional FX volume dataset.
The basic maths is straightforward: to find the volume of settlements relating to FX trades, take the daily volume of spot and outright trades and double it as there are two settled payments; for swaps multiply by four. This shows a gargantuan figure of $18.7 trillion settled daily in the OTC FX market.
Adjusting for trades settled via Payment versus Payment services (PvP) - principally CLS -and those where gross exposure is reduced by bilateral netting we get a still impressive figure of $8.9 trillion that is settled daily with no protection against settlement risk.
$18.7 trillion (settled daily)
- $9.8 trillion (PvP & bilateral netting)
-------------------
= $8.9 trillion (at risk)
-------------------
How accurate is this number?
Well, many Emerging Market Economies (EMEs) are not included in the survey, generally due to size. Most of LATAM, MENA, CEE and Africa is missing, as well as a very long tail of smaller countries. Therefore, the total market size from one perspective is clearly underestimated.
However, a significant number of trades in the reporting markets do not settle gross, such as spot trades rolled with prime brokers and the extension of large FX swaps generally by larger corporate and buy-side clients. Therefore, the BIS raw data overstates the settlement volumes for these trades.
On balance, I feel that the volume of over-reported risk (developed markets rolling over trades) is probably not dissimilar to the volume of underreported risk (EME’s not included in the survey), so the likelihood is that the BIS figure of $8.9 trillion is substantially (if accidentally) correct.
The problem with this number is that the volume of settlement risk doesn’t adequately describe the true inherent risk.
An uneven distribution
With CLS settling approximately 40% of net payment obligations, it’s easy to think that settlement risk reduction is evenly distributed across the FX market. In fact, while for CLS members and their third-party clients may achieve risk reduction of 90% or more, this percentage is zero for those with no access to settlement risk mitigation - and it is here that the burden is profound.
Trades between second and third-tier banks and larger corporate and institutional clients, settled outside of CLS, in currencies with deep liquidity, efficient correspondent banking and fast payment systems have minimal perceptible risk of settlement failure. Readily available counterparty information drives accurate assessments of credit quality and desired exposure, and efficient processing ensures effective settlement.
Dip below this privileged layer and credit quality, lack of currency liquidity, technological capability, correspondent bottlenecks and payment inefficiencies are a perfect storm for EMEs.
While it may still be simple and transparent to settle a G10 pair between EME banks, or between banks and their customers, a cross-border trade involving local currencies has a considerably higher amount of risk of not being settled, or of settlement delays, than its G10 counterpart.
The fact is that not all settlement risk is equal
The real picture requires us to think about the substantial additional risk on parties and currencies with no access to PvP solutions, and to address this gap.
Settlement Circle is working with some great EME banks to overcome this, and we would like to invite banks from all markets to join us in our journey.
By Declan Clements, November 2021
Comments