Reviewing the impact of transition to SOFR and managing legacy contracts

Oskar Rogg, Head of Treasury Americas, Credit Agricole

Below is an insight into what can be expected from Oskar’s session at Treasury & ALM USA 2023.

The views and opinions expressed in this article are those of the thought leader as an individual, and are not attributed to CeFPro or any particular organization.

What are the benefits and challenges of banks switching to SOFR and other RFR’s?

The benefits side of the RFR transition is a little bit like Y2K –the transition does give you a chance to rethink and revamp things like funds transfer pricing and some of your operational processes.

On the challenges side, over and above the need to update systems and processes and to transition substantial numbers of existing transactions, your primary challenges are communication and risk concentration.

In terms of communication, LIBOR represented a very rich message – it communicated not only interest rate expectations, but also systemic credit and liquidity expectations.

In contrast, SOFR is a very sparse message – how much did it cost to finance US Treasury securities overnight yesterday (SOFR), or is it expected to cost over the next few months (Term SOFR) – a very minimal version of interest rate expectations. Since the message is so parsimonious, in the context of commercial loans, a great of important information has to be communicated, and thought about, separately for the first time.

In terms of risk, SOFR takes the systemic credit and liquidity risk that was formerly dispersed across the entire system and concentrates it on to financial intermediary balance sheets. If banks were gas stations, it would be as if the price at the pump was tied to the consumer price index (CPI), rather than to the wholesale price of gasoline – the same amount of risk in the economy is concentrated on fewer entities.

In a SOFR-based world, a dislocation such as occurred in 2008 or 2011 or 2020 will likely have very disparate impacts on the asset and liability sides of FI balance sheets.

What implications will the shift to static CSA from a dynamic CSA have on banks?

In the last 9 months, we’ve seen market spreads on 1yr wholesale bank borrowings oscillate between SOFR+28 and SOFR+95 – it’s tempting, but dangerously wrong, to assume that SOFR plus a static spread is an analogue to LIBOR, which incorporates a dynamic CSA.

“Dynamic” vs “static” sounds dry and arcane, so another metaphor may be useful. If you are floating in the ocean and you are tied to a buoy, you move up and down with the waves. If you are instead tied to a cement block, life becomes a bit more challenging, particularly when seas get rough. Even if the length of the chain is precisely calibrated such that your head is anticipated to be above water 50% of the time over the next five years, it may still be problematic.

False equivalence is a very real risk – you might assume that 3mo term SOFR + a CSA is a “close enough” substitute for LIBOR and not pay enough attention to the fundamental differences.  Financial market participants need to reprogram not only their systems and processes, but also their collective thought processes to accommodate the often subtle, yet very real differences.

How can internalizing systems and process help banks manage legacy contracts?

In the institutional space, I think market participants on all sides have adopted a collaborative approach – transitioning legacy contracts is a formidable task, it requires a lot of parties to work together, there are a range of implications, and in many cases there is no “one size fits all” solution.  It takes time and requires close collaboration.

In the end, it may be less about internalizing systems and processes and more about recognizing that both systems and processes need to evolve and ensuring that evolution actually takes place.  Technology and process are “table stakes” – there is no question that the better your systems and processes are, the easier the transition will be, but you simply aren’t going to be able to find some technologic magic bullet that accelerates your legacy transition between now and June 30th.

How can SOFR simplify how funds transfer pricing is carried out?

I’m not sure it “simplifies” it, but the RFR transition is fundamental enough that it provides both an opportunity and an imperative to review all of your processes and assumptions to ensure they make sense in an RFR world.

It’s easy to forget that LIBOR is fundamentally a proxy for where banks borrow short term USD in the market. So to the extent that banks borrow money in the market, we may choose no longer call it “LIBOR,” but not much else changes.

Before, when both assets and liabilities were indexed to the same baseline (LIBOR), FTP didn’t need to consider things such as systemic credit and liquidity spread, because those costs were bundled into LIBOR and they were dynamic (reset every day). Now that assets (loans) are indexed to SOFR, with perhaps a static CSA, we need to ensure that the unbundled and dynamic costs on the liability side are accurately reflected.  Similarly, to the extent product offerings contain optionality, we need to ensure this optionality is properly costed in a SOFR-based world.

Over and above the incremental complexity, stakeholders have to adjust their lexicons and thought processes to incorporate things like CSAs that they never had to consider in the past.

What are the key benefits to leveraging technology & tools to recalibrate pricing?

You want to ensure you are not “reinventing the wheel” on every transaction, and that all of the complexities and implications and relevant inputs from across the organization and across the markets are consistently considered when making a decision. Technology and tools can provide a lot of leverage in achieving that goal.

But anyone who has spent any time in the financial markets knows that leverage is always a two-edged sword. When you are thinking about technology, you need to ensure you correctly understand the problem before you get too ambitious in trying to industrialize the solution.  And ultimately, realistic expectations may be a critical success factor when doing a technology implementation in this or any other space.