How LIBOR's death could affect investors
Investing Journey has a portfolio update, at the end of this post
The London Interbank Offered Rate (LIBOR) was widely used for decades to set interest rates on trillions of dollars worth of bonds, mortgages, interest-rate swaps and many other securities. An example is the bond with a yield of 2% plus LIBOR.
But LIBOR was set by a rather small group of global banks - and it was shown in 2012 that they were manipulating the rate to their financial advantage. The whistleblower was awarded US$200 million (informants are entitled to 10% to 30% of the fines levied).
The scandal has led the U.K. Financial Conduct Authority (FCA) to ban LIBOR outright. Most of the 35 variants (spread across 5 currencies and 7 maturities up to a year) will cease publication by the end of this month.
The exceptions are the heavily-used U.S. LIBORs, i.e. those with terms other than 1 week and 2 months. They will cease publication after June 30, 2023. A synthetic LIBOR may be temporily used on some complicated U.K. and Swiss securitiess.
In the U.S., LIBORs will be replaced by the Secured Overnight Financing Rate (SOFR). It is the rate set in the deep and transparent U.S. repurchase (repo) market, where U.S. Treasuries are sold and repurchased the next day (thus, repos are short-term collateralized loans). Other countries are publishing their own replacement rates.
The transition could get messy: LIBOR-indexed securities maturing after the switchover dates need to have their documents amended to say they are tied to the SOFR instead of LIBOR. “Updating legacy contracts can prove a complicated process, raising the risk of a chaotic transition that has been likened to Y2K,” noted William Shaw, Bloomberg’s LIBOR expert. “Asset managers face a … risk of fines, litigation and reputational damage if they poorly manage the transition,” he adds.
Even if the transition goes smoothly, one wonders if the change to new rates will spawn a new set of problems or unintended consequences. For example:
1. SOFR is based on secured loans while LIBOR is based on unsecured loans. So, returns linked to the new reference rates are not that sensitive to credit risk and may not rise to compensate investors for risk like LIBOR did
2. SOFR trades only on an overnight basis whereas LIBOR had maturities ranging from overnight to a year. Regulators have simulated this term structure for SOFR using historical data and SOFR futures but this estimation may not always reflect reality.
3. Countries have developed their own replacement rates with methodological differences (for example, Japan’s TONA is an unsecured rate whereas SOFR is a secured rate); this may give rise to new discrepancies across reference rates globally.
4. Whereas the switch away from LIBOR may be relatively easy for assets like bonds, loans are more challenging because this market has a much wider range of lenders and borrowers – and is largely private, thus more opaque.
5. Some financial instruments are quite complex and are exceedingly difficult to switch over to new rates, for example non-linear derivatives like swaptions (options on swaps).
Portfolio update
I trimmed some of my economically sensitive positions. Of note was the sale at a 10% gain of the SOXX semiconductor ETF bought in late October, as discussed in an Oct. 29 post. It was not meant to be a long-term hold anyway - just a play on the late innings of the chip cycle. As for cyclical stocks in general, sentiment is becoming noticably less bullish in the market with the advent of Omicron and the prospect of higher interest rates to fight out-of-control inflation. Much of the proceeds from the trimming were moved into arbitrage plays, which can deliver positive returns in bull and bear markets. I hope to post about these investments in a forthcoming post