SOFR Reconsidered | Seeking Alpha

LIBOR’s past

The LIBOR market was once the world’s source of valuation of low risk unsecured corporate debt. For three reasons:

  1. Bank debt was more homogeneous than the alternative source of short-term wholesale unsecured funds, non-bank commercial paper. Moreover, the market for bank debt was more liquid.
  2. “Top Tier” bank deposits traded at a common yield when LIBOR first traded. Thus, LIBOR was not tied to the fate of a single borrower.
  3. Unlike wholesale CDs, the primary source of regulated commercial bank funding early in the period of LIBOR dominance, LIBOR rates did not include a deposit insurance fee paid to the FDIC. Therefore, the extra cost of CDs made the cheaper LIBOR-based market the preferred source of bank funds.

During the heyday of LIBOR, interest rate risk was more significant than it is today. LIBOR’s doom was sealed with the Financial Crisis of 2007-2008. Two effects of the Financial Crisis spelled the end of a role for LIBOR in the interest rate risk transfer process.

  1. The dramatic decline in interest rate risk driven by zero target rate monetary policy and inflation rates below the Fed’s two percent inflation goal.
  2. Pressure from bank regulators, leading banks to reduce their interest rate risk exposure.

LIBOR’s present. The problems of the LIBOR transition

LIBOR is dying, to be replaced primarily by the secured overnight financing rate (SOFR). There are at least three problems with the SOFR-based index. The first, legal exposure, affects any rate index that might replace LIBOR. The second, lack of credit risk, affects any secured rate. The third, function-relevance, affects any overnight rate. But function-relevance is the essential reason for LIBOR’s success, and lacking function-relevance will be the reason for SOFR’s failure.

Legal exposure

Renegotiation of contracts that include LIBOR is such a thicket that some form of legislation-based remedy for legal wrangling over LIBOR replacement seems likely. But securing legislation is less problematic than it might seem. Congress has trouble changing tax and spending policy to bail out taxpayers, but Congress had no trouble changing the bankruptcy code to protect interest rate swap traders. Thus, changing the law to accommodate the needs of the large dealer banks with LIBOR-based loans is feasible.

Lack of credit risk

LIBOR reflects the market’s estimate of the probability of bank default. SOFR is based on Treasury-collateralized repurchase agreements (repo). Thus, SOFR fails to reflect changes in the market estimate of credit risk. One can adjust an overnight rate for estimated credit risk included in other market yields. But if these other yields were accurate barometers of anticipated risk, the replacement of LIBOR would be most effective if the rate used to adjust SOFR would simply replace LIBOR instead of SOFR. This is not proposed because these credit-risk adjustments come from illiquid markets.

But this is circular reasoning. If credit risky markets are too illiquid to replace LIBOR, how are they sufficiently liquid to provide an interest rate risk adjustment? Logically, the difference between the credit-risky yield and SOFR is the appropriate adjustment for credit risk. But the result of combining this risk adjustment with SOFR is the credit risky rate in the market from which the credit adjustment is taken – an illiquid market, no?

Some versions of the credit risk adjustment include further modification of the difference between liquid and illiquid interest rates (moving averages, term-to-maturity adjustments). But these further adjustments do not change the fundamental fact. One cannot use a credit risk adjustment based on an illiquid market yield to adjust a liquid market rate without losing liquidity.


LIBOR is a forecast of expected future overnight interest rates. SOFR is a summary of past overnight rates. Since yields on bank credit and other debt indexed to LIBOR are set at the beginning of each payment period and paid at the end, these yields reflect the lender’s forecast of future overnight rates. Thus, the safest source of interest rate risk protection for the lender is a rate that forecasts future overnight rates during the payment period in question. No overnight rate has this capability.

Worse, term indexes of overnight SOFR used to settle loan agreements – such as Eurodollar futures and LIBOR-based interest rate swaps – are backward-looking. They fail to forecast the expected costs of unsecured overnight money. This flaw further reduces the function-relevance of SOFR, explained in greater detail here.

SOFR does not have function-relevance. Nor would any overnight rate index, since all term estimates of overnight rates are backward-looking.

The immediate future

There are a few coming events that signal the beginning of the LIBOR transition.

The Little Bang

The near-term future begins now. Market participants call it “The Big Bang.” Beginning on October 26, 2020, OTC derivatives indexed to the Fed Funds rate will be changed. The Fed Funds rate will be replaced by SOFR in some 80 trillion dollars notional amount of swap agreements. The prognosis is good. Market participants claim they are well prepared for the change and have already simulated the effects on their portfolios. This event is covered in greater detail by Bloomberg here.

The Fed Funds rate, like SOFR, is an overnight rate, unsecured by Treasuries but perhaps the least risky of the unsecured overnight rates. Thus, Fed Funds replacement is the easiest of the coming LIBOR replacement events. Moreover, none of the problems of LIBOR replacements are shared by Fed Funds replacement. This replacement is more like a LIBOR replacement with training wheels.

The Regulators have clarified the optional status of SOFR

Another positive development in the LIBOR transition process is the letter from senior representatives of US government regulatory agencies, stating:

“The transition away from LIBOR is a significant and complex undertaking. Financial market participants should not delay their preparations for transitioning away from LIBOR and should replace LIBOR with suitable alternative reference rates, which include but are not limited to SOFR…. Supervisors will not criticize firms solely for using a reference rate (or rates) other than SOFR.” [emphasis mine]

This has the much-to-be-desired effect of permitting the market to resolve the problem of LIBOR’s death more effectively than with a SOFR-based LIBOR replacement.

The Credit Sensitivity Group (CSG) was formed by the Federal Reserve Bank of New York in February of 2020 to address the adequacy of any credit risk adjustment to SOFR. The letter above and other documents related to the deliberations of this group may be found here.

The more distant future of LIBOR replacement

To consider the more distant future, compare the original qualities of LIBOR to the index requirements of the International Organization of Securities Commissions (IOSCO), to be found here. In other words, consider what Adam Smith’s invisible hand required to establish LIBOR compared to what a blue-ribbon committee of financial market regulators expects.

A summary of the IOSCO expectations:

  1. A responsible accountable administrator for index determination.
  2. An administrator without self-interest in index value.
  3. Administrator authority to control the process of index determination.
  4. Index transparency.
  5. Periodic review of the index methodology.
  6. Index quality: An index that reflects conditions of market supply and demand in a liquid market.

The first five requirements, quite appropriately, are expectations for an agency or organization that decides the validity of the index. Only requirement six is a requirement for the index itself. Moreover, this requirement boils down to two Adam Smith-like expectations.

  1. The index should reflect conditions of demand and supply.
  2. The market source should be liquid.

But SOFR does not satisfy condition one. The value of SOFR, an overnight rate, is manifestly entirely the result of the Fed’s monetary policy objective – a near-zero Fed Funds rate.

What market conditions are consistent with a market-created LIBOR replacement?

For the short-term credit markets to produce a LIBOR replacement, two market conditions that made LIBOR desirable 50 years ago are necessary.

  1. Volatile interest rates inspired by the presence of inflation rates exceeding two percent.
  2. An index-creating entity aware of the commercial value of supporting a liquid market for short-term credit.

What requirements would a LIBOR replacement need to meet to be as effective as LIBOR was? It would provide a market yield that is:

  1. not dependent on a single (or ideally even a few) commercial firms’ credit risks,
  2. not easily manipulated by market participants,
  3. not determined by decisions of government bank regulators, market regulators, or monetary policymakers,
  4. easily incorporated into financial derivatives’ settlements, and
  5. reflective of the market’s forecast of future credit conditions.

These combined market and instrument conditions for the production of a viable LIBOR replacement are a tall order. Certainly, the conditions will not be met for another three years – the length of the Fed’s commitment to keeping overnight rates near zero. None of the currently proposed replacements fit the five requirements above. On the other hand, it would take all three years of the Fed intervention period to generate the institutional structure needed to accomplish this task. One approach to index creation is described here.

Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.

Be the first to comment

Leave a Reply

Your email address will not be published.