Delivery risk (also called settlement risk) is when the settlement process of a trade involves one party paying out cash or securities before it receives the countervalue from its counterparty. The risk is that the counterparty may default in the meantime, leaving the first to pay empty-handed.
That’s the gist of it. Still confused?
Keep reading.
Throughout this post, we will focus on delivery risk in the context of foreign exchange (FX) spot trades from the perspective of a bank trading with its counterparties, whether other banks, corporates, or funds/asset managers.
Ready? Let’s dive in:
What Is Delivery Risk
Also called settlement risk, delivery risk arises when payments are not exchanged at the same time. It is a subset of counterparty risk (along with liquidity risk).
In its essence, it is the risk that one counterparty will fail in its obligations to the other.
For instance, when a bank makes a payment to a counterparty but it doesn’t get money back until some time later. The delivery risk for the bank is that the counterparty may default before making the counterpayment.
Settlement risk is common in foreign exchange markets. Why?
Because payments in different currencies occur in their respective countries’ timezones. Thus, the cash flows can happen many hours apart.
Also, the volume of payments makes it only possible to monitor transactions after they occur.
Imagine two firms agree to exchange one currency for another.
If the two cash flows do not happen simultaneously (at the same exact time), then there is settlement risk. It’s as simple as that.
Other than timezone problems or the settlement mechanics of a given market, settlement risk also occurs when a bank with access to a central counterparty clearing house (CCP) clears trades through the CCP in name of its clients.
The client doesn’t have access to the CCP because it doesn’t want to be directly exposed to the CCP or doesn’t meet the requirements to join, so it will ask the bank to trade for them.
The bank will pay the necessary initial margin upfront to the CCP on behalf of the client. And then wait for the client’s payment.
In this case, settlement risk is the possibility that the client defaults in that period of time between when the bank pays the CCP and when the client repays the bank the initial margin.
This clearing agreement basically works as an overdraft the bank allows the client to have.
How to Measure Delivery Risk
Measuring delivery risk is straightforward.
A firm’s exposure (the amount at risk) when settling a trade is the full amount it sends to its counterparty. It lasts from the time of this initial cash flow until the time it receives the countervalue from its counterparty.
Delivery risk varies depending on the risk mitigation frameworks the firm may use:
How to Reduce Delivery Risk (2 Main Ways)
When a bank sends money to another firm expecting to receive a countervalue, and does nothing to guarantee it receives the money back, the settlement risk calculation is simply the amount the bank is owed.
So, how can firms reduce delivery risk and exposure?
Maximum Daily Delivery Risk (MDDR)
MDDR (Maximum Daily Delivery Risk) is an intraday limit set up to restrain the exposure a firm may face on the settlement date of a trade.
As we’ve seen, the settlement risk (or MDDR exposure) is all the cash flows to be received at the settlement date.
MDDR exposure is typically an intraday exposure—it won’t last for more than a day.
To control the exposure, a bank will set up a limit up until which it is ok with paying a counterparty first, and receiving the countervalue of the trade until the end of the day.
However, in practice, it’s not always intraday.
Firms may agree on MDDR limits lasting up to 2 days later (T+2), or even (T+5). Because these tenors are short, they are still treated as FX spot trades.
The longer the tenor, the lower the MDDR limit. This is because of the higher risk the counterparty defaults in the meantime (and the bank receives nothing).
A lower MDDR limit means the bank and its counterparty trade a lesser amount, limiting the potential loss.
The MDDR limit is a pure cash amount, as opposed to a future projection of potential losses (like the VaR/PFE).
Say a fund wants to trade FX derivatives with a bank. The bank agrees to do so, but it will set a limit for the MDDR up until which it is comfortable with sending money before receiving the countervalue—depending on the riskiness of the fund.
The MDDR limit is usually set up as a percentage of the fund’s NAV (Net Asset Value).
If there’s a netting agreement between counterparties, net MDDR comes into play—smaller MDDR exposure when facing an entity because of the offsetting of payments a bank is due to send to that same entity and in the same currency the entity owes the bank.
This bilateral netting is another way to reduce delivery risk, as it allows to lower the exposure. It aggregates the payments to one per currency either to or from each counterparty.
Continuous Linked Settlement (CLS) System
CLS is the global standard in FX settlement risk mitigation.
To settle an FX transaction, counterparties exchange principal (value of the trade) in two currencies. And they settle each leg of the transaction independently, at different times (with a lag), in the country of each currency.
As we’ve seen, delivery risk is the risk that one party delivers the currency it sold, but doesn’t receive the currency it bought from its counterparty. The result?
Loss of the principal.
CLS mitigates this by settling the payments on both sides of the trade at the same exact time, using a payment-versus-payment (PvP) system.
PvP is a mechanism in a foreign exchange settlement system that ensures a transfer of one currency only occurs if the corresponding transfer of the other currency also happens.
When trades are settled through CLS, there’s no delivery risk. MDDR exposure is zero. Hence, CLS is an alternative to using an MDDR limit.
Unlike the MDDR limit, there’s no limit as to how big the value the counterparties are trading because they trade the funds simultaneously. So there’s no risk the counterparty defaults while a firm waits for its countervalue, because they never wait.
With MDDR, a bank will wait up to two days or more before it knows with 100% certainty it received the currency it bought.
As a result, it sets up a prudent allocation limit on the value it is willing to trade with a counterparty to cap its potential losses in case the counterparty happens to default in those two days.
With CLS, there’s a simultaneous settlement of both sides of the FX payment, through a single, trusted, creditworthy intermediary.
Each party only transfers the net amount of all its payment obligations in each currency, unless the settlement is done through Real-time gross settlement (RTGS) systems (transactions settled as soon as they are processed rather than at the end of the day, on a one-to-one basis, without bundling or netting with any other transaction).
For example, imagine a USD/EUR trade.
Outside the CLS, the seller settles its EUR on T, whilst the buyer settles the USD on T+2. With CLS, the two flows are simultaneously settled (PvP principle).
Key Takeaways (FAQs)
What is the delivery risk of a bank?
Delivery risk (also called settlement risk) is the risk of default of one of the counterparties while a trade is live. When a bank buys an asset for a client, the client can default in the meantime before the bank gets the money/securities back.
What are the risks involved in the settlement process?
Settlement risk (risk of paying out the currency sold but not receiving the currency bought), replacement cost (risk of the price a firm would need to pay to replace an existing trade or asset is unfavorable), and liquidity risk (risk of buying or selling an asset in so much quantity that it affects the price of the asset).
What is Herstatt risk?
Settlement risk is sometimes called Herstatt risk, named after the failure of German bank Herstatt. In June 1974, the bank received payments in Euros, but was shut down by German banking regulators before paying the U.S. dollars it agreed upon, leaving its counterparties with substantial losses.
