Flattening the (Funding) Curve

The impact of COVID-19 on the funding spread

Recently, amidst the unprecedented volatility and asset price shifts cause by COVID-19, the Funding Valuation Adjustment (FVA) has become a major source of accounting losses for several US banks totaling almost $2bn [1]. FVA, which measures the cost (and benefit) of funding uncollateralized trades, depends largely on two factors: the exposure on the trades and the funding spread. One cause of the sudden rise in FVA may be attributed to the lowering of interest rates by reserve banks around the world. For example, should a bank have more receiver swaps than payer swaps, then a lower interest rate will invariably increase the exposure and consequently the FVA.

The other cause would be a sudden rise in funding spreads. In the simplest case, when banks enter into an uncollateralized trade with a counterparty, they simultaneously enter into a collateralized hedge position with a central counterparty. When the trade becomes in-the-money, the pledged collateral on the (out-of-the-money) hedged position earns a risk-free rate but must be funded at the bank's unsecured rate - the spread between these two rates is the funding spread.

To investigate the latter cause, we calibrate to the recent FVA values from our Totem xVA Service at the 2-, 5- and 10-year tenors for USD and EUR. Note that, unlike exposures, funding spreads are independent of the bank's portfolio and capture the state of the lending markets amongst the major global banks. The calibrated funding spreads using the in-the-money receiver swap are shown below:

Our results show that, the short end of the funding spreads is the most sensitive to the recent market conditions. One sees a similar phenomenon in bank credit spreads - this is reasonable, as both funding spreads and credit spreads reflect the bank's riskiness as viewed by lenders (bondholders, in the case of credit spreads). As an example, a comparison of jumps in the funding spreads to JP Morgan's credit spreads is shown in Table 1.

Table1. Jumps in spreads from February 21 to March 24, 2020.

In addition, the funding spread term structure (not shown) were upward sloping pre-crisis, but flattened in March, for both USD and EUR. This is simply an artifact of the short end of the funding spread term structure jumping more than the long end - a reflection of liquidity drying up in the short-term funding market. Recall that in Basel III, the Net Stable Funding Ratio (NSFR) protocol was designed precisely to mitigate such scenarios. The calibrated funding spreads quantifies the risk(s) banks face in their derivative portfolio management when relying on cheap short-term funding sources to cover longer dated assets.

Finally, like credit spreads, it appears that funding rates have peaked in mid-March and are now trending down - the actual peak in the funding spread, judging from daily quoted credit spreads, was likely on March 20 when the Federal Reserve of New York announced a plan to supply $1 trillion a day to the repo market.

Funding spreads contains vital credit information of the derivatives market, particularly the liquidity of the interbank lending market. While the initial shock from the COVID-19 crisis has appeared to have subsided, whether the New York Fed's recent intervention is sufficient beyond April remains to be seen - trends in the future funding spreads will confirm this. For the latest calibrated funding spreads with comparisons to simulated credit curves, see IHS Markit's Financial Risk Analytics Credit Forecasting Utility.

[1] FVA losses back in spotlight after coronavirus stress. Risk.net, April 16, 2020

This article was written by Ryo Takei who is on the Financial Risk Analytics team at IHS Markit

S&P Global provides industry-leading data, software and technology platforms and managed services to tackle some of the most difficult challenges in financial markets. We help our customers better understand complicated markets, reduce risk, operate more efficiently and comply with financial regulation.

This article was published by S&P Global Market Intelligence and not by S&P Global Ratings, which is a separately managed division of S&P Global.