SOFR Is A Bad Idea, So Is Repo

by: Kurt Dew
Summary

LIBOR and the London dollar deposit market are thoroughly discredited.

But conveniently, repo-based lending and the repo rate have replaced London dollar deposits and LIBOR as the primary basis for financial institution borrowing globally following the Crisis.

So the Fed proposes a repo rate, the Secured Overnight Funding Rate (SOFR) to replace LIBOR.

SOFR is different, but not better, for well understood reasons.

And Treasury repo itself never made much sense.

“For every complex problem there is an answer that is clear, simple, and wrong.”

  • HL Mencken

SOFR, or the Secured Overnight Financing Rate, a Fed-introduced version of the overnight Treasury repurchase agreement rate, is the Fed’s candidate to replace LIBOR. SOFR is a mistake. Many analysts, including senior bank regulators, explain what are now well-understood reasons why SOFR won't work. Read here, here, here, for example. Briefly, SOFR is an overnight rate. Thus one-, three-, and six-month averages of SOFR look backward. They describe the past; they don’t forecast the future as a term loan, futures and swap rates must do. And SOFR is a Treasury-based rate. Thus, it does not reflect the risks inherent in lending as the instruments it will price need to do. SOFR is a flawed interest rate index.

Is the repo itself, like the London dollar deposit upon which LIBOR was based, also a mistake? Sadly, yes.

Yet, Treasury-collateralized overnight money – or repo, the ultimate security-based loan – has become the primary source of bank and nonbank financial institution funding, since the Crisis. It has largely replaced the old staple of wholesale bank funding, the term London dollar deposit, and its ubiquitous price, LIBOR. What’s more, securities lending against repo has largely replaced LIBOR-based trading activity as a source of government-protected oligopoly profit for the megabanks, [Bank of America (NYSE:BAC), Citigroup (NYSE:C), Goldman Sachs (NYSE:GS), and JP Morgan Chase (NYSE:JPM)].

Image result for securities lending

Combined with the disrepute into which LIBOR has fallen, with LIBOR’s scandalous apparent manipulative potential for purposes other than raising deposit money, repo’s new role has seduced financial market regulators into a simple logical conclusion: Since repo has replaced London dollar deposits as a source of funds, why not replace LIBOR with a repo rate such as SOFR, a new index for the cost of bank money?

But wait a minute! We know SOFR is flawed. But is any repo rates index a good idea? Does a Treasury repurchase agreement itself make sense as a credit instrument? The question should be given consideration. After all, the London dollar deposit, which has served as the cornerstone of short-term financing for almost four decades, proved to make no sense at all.

In hindsight, the London dollar deposit market never made any logical sense as the globe’s index for dollar-based lending. It's in London. And the price of getting the choice of index wrong turned out to be incredibly high. We have an entire dollar-based financing system based on a rate, LIBOR, which never existed in the form that was claimed - the cost of wholesale bank funding! LIBOR was never a price. It was an opinion.

We couldn’t be making the same mistake again, could we? Yes, we could. We are moving from one nonsensical source of indexing for the world’s over-the-counter (OTC) short-term credit-risky interest rates, to another. But this time, we have no excuse. LIBOR was brand new. Repo is over 80 years old. It was a bad idea from the start. It has shown its flaws multiple times, most prominently during the Lehman collapse. It’s a bad idea now.

How did we fall into the strange habit of using repo?

Because the Fed did it first.

In its original manifestation, a vehicle of monetary policy, repo was the “brilliant” invention of Benjamin Strong, then President of the Federal Reserve Bank of New York (back when that job was at the heart of financial power.) As an instrument of monetary policy, repo makes perfect sense on the Fed loan side. The Fed wants to increase and reduce the amount of Fed liabilities, or reserves, held by commercial banks, and do that without creating credit risk for itself. Lending cash in exchange for 100%+ Treasury collateral does that nicely. A Treasury primary dealer might object to such a steep collateral requirement, but there is only one source of Fed money. Nobody listens to dealer complaining. But the reverse transaction, where the Fed provides 100%+ collateral in exchange for cash, makes absolutely no sense whatever! What? Is the Fed not good for the cash? Nonsense. The Fed creates cash.

The “reverse repo,” which the Fed called a “matched sale-purchase” for several decades to avoid admitting that the transaction was a loan to the Fed, made sense in Strong’s time because Strong was trying to counteract the Fed’s post-WWI problem of a surplus of gold. At the time, prices were inflating because gold was in surplus and the money supply, tied to gold, in surplus as well, creating inflation. Open market sales of securities mopped up the money and broke the inflationary bond with gold. Strong wanted to break the tie connecting the Treasury’s stock of gold and the quantity of money in the hands of the public. So repo was a pretty good temporary seat-of-the-pants kind of solution to Strong's problem.

But the Fed never broke the strange custom of providing 100% collateral on funds it borrows from government securities dealers; a tribute to the supremacy of habit over thought. The repo today is, more than anything else, a bad reason for stuffing an undue amount of Treasury securities onto financial institution balance sheets.

Banks fake it.

But why do commercial banks require 100% collateral when they trade repo with each other, or with other top-quality borrowers? That’s also overdoing collateral protection.

They don’t, of course. This has become apparent repeatedly, beginning with the collapse of the dealer, Drysdale Government Securities in 1982. Chase Manhattan Bank lost $285 million in that fiasco, a lot of money at the time. Drysdale taught the world about the joys of rehypothecation.

Investopedia explains, “Rehypothecation is the practice by banks and brokers of using, for their own purposes, assets that have been posted as collateral by their clients.” So, if you provide Treasuries as repo collateral, perhaps your lender swaps it out to borrow money itself; to another lender, who also swaps it out; and so it goes… This process forms a collateral conga line. It creates a daisy chain of loans, all depending on the same collateral, which might break in the event of a bankruptcy, and make a terrible mess, as it did when the door to overnight credit closed for Lehman.

The SEC has been aware of the prospect and has been worried about it, for some time. Out of their concern, the SEC promulgated the 1972 SEC Rule 15c3-3. The rule seeks to assure that clients can withdraw most of their holdings on demand, even during a bank’s insolvency. In other words, the SEC’s rule limits the degree to which broker-dealers can swap out their client’s collateral.

Shuffling off to London… Again.

It is useful in our tale of the silliness of repo to recall how the silliness of the London dollar deposit market began. Prior to the early-1970’s collapse of the Bretton Woods fixed exchange rate agreement, the US banking system was a sleepy place. After the disastrous banking crisis of the Great Depression, and World War II, one of the sillier regulations promulgated to prevent banks from taking risk, Regulation Q, placed interest rate ceilings on bank deposit rates, both retail and business. As the inflation rate exploded after 1970, various escape valves – ways for depositors to earn interest income sufficient to keep their heads above water – were devised. The two enduring adjustments were money market funds, a substitute for Fed-regulated retail deposits, and wholesale London dollar deposits, a substitute for large Fed-regulated corporate deposits. Both fell into disrepute during the Financial Crisis.

Shuffling wholesale dollar deposits off to London appealed for reasons that explain London’s success as a financial center. If a financial business line migrates from America to Europe, there are two possible explanations: 1. the business is escaping bad American regulations; 2. the business is escaping good American regulations.

How do you spot bad behavior among financial institutions?

Securities lending to finance repo takes place mostly in Europe because it is escaping both good and bad American regulations. Once any trade or financial instrument arrives in Europe, it goes through a multi-year process of acclimatization.

First, everything needs an acronym over there. A trade is not a simply a trade in Europe, for example. It’s a securities transaction (ST). In due time, European regulators are contacted by American regulators through some international organization (the G20 or some such) and informed that stuff is going on in Europe that’s deemed not OK in the US. Europe knows what to do about that. They form a commission. The commission deliberates for several years. The post-Crisis Uber-commission, the Financial Stability Board (FSB), was formed by the G20 in April 2009.

Following the Financial Crisis in the US, there were many not OK financial instruments and trades, and it was deemed necessary to form a really important commission: the FSB, to pass regulations nearly as stiff as the American ones; but not quite, lest the business that fled American regulations returns to the States where it belongs.

The rule from which American securities lenders seek to escape is the 1972 SEC Rule 15c3-3, which limits the degree to which broker-dealers can swap out their clients' collateral. This reduces systemic risk, by preventing daisy chains of loans, all depending on the same collateral. The daisy chain might break in the event of a bankruptcy in one link of the collateral conga line.

But of course, the safer SEC rule dramatically reduces the profitability of securities lending. Indeed, while the SEC’s concern for systemic risk resulting from collateral lenders playing fast and loose with customer collateral seems well-placed; it also seems wasteful to use 100% collateral to protect a loan against failure. Hence, the repo exodus to the right-minded regulators of London.

The London alternative went badly when dollar deposits moved to London in the 1970’s chasing profit. Now that we are dropping LIBOR for SOFR, a repurchase agreement – a loan against Treasury collateral – and collateral lending for this business is located, again, in London, chasing profit. Should we find that a little disturbing?

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.