Ending Repo Madness (Wall Street Journal oped)
A Swedish friend, employed in New York by a famously prudent Swedish bank, kindly invited me to lunch last month. We discussed the recent convulsions in ‘repos’, as one does over quinoa salad in Manhattan.
That conversation prompted me to propose a radical simplification of repo markets, just published in the WSJ.
My proposal may trouble moral hazard purists, while many outside finance just won’t care.
But there’s a reason for concern: The repo market is a trillion-dollar pawnshop and the last time it went haywire was before and during the 2008 crisis.
Experts have of course suggested complicated tweaks: glitches in complex systems invariably invite quick fixes that increase complexity — and the power of insiders who pretend to know what’s what.
But it’s spiraling complexity that crashed 737s, destroyed Detroit’s manufacturing preeminence, and wrecked finance.
I’m not predicting imminent disaster. But why take a chance – why not make Toyota’s “lean” principle the default before a real geopolitical or financial accident happens?
To Make Banks Stable, End, Don’t Mend, the Repo Market
Simplifying how financial institutions borrow would make the system more resilient.
‘I don’t understand how your ‘repo’ market works,” a veteran Scandinavian banker tells me. “There’s nothing like it at home; our world is much simpler.”
Repo markets are like pawnbrokers for banks. They provide no-questions-asked cash against pledged collateral, typically Treasury bonds, and serve as crucial conduits for the credit that banks need. Many financial institutions routinely use repos to secure overnight loans from money-market funds and other institutions with surplus cash looking for a safe return. Trouble in repo markets can spill into the “fed-funds” market, another source of overnight cash for large banks. Credit problems can cripple the economy, as in 2008 after repo markets imploded.
Analysts described last month’s spike in repo rates—from about 2% to 10% on Sept. 15—as “wild” and “seismic,” sowing “shock and confusion.” Some blamed high cash demand to pay estimated taxes and settle Treasury bond auctions. More-esoteric explanations included a holiday in Japan, the “glut” of safe assets overwhelming scarce cash reserves, the Federal Reserve’s foreign-repo program for non-U.S. lenders and central banks, and even the recent attack on Saudi oil facilities.
The risks of another credit collapse that surfaced last month spurred forceful intervention by the Fed. For the first time since 2008, the central bank directly and repeatedly made repo loans. In response to criticism of the Fed’s failure to foresee the problem, Chairman Jay Powell asserted that the Fed understood repos as well as anyone and observed that market participants had also been surprised. New York Fed Chief John Williams declared that the Fed was “consistently and constructively supporting stability” in repo markets.
Critics who assert that the repo snafu marked an “unequivocal” breakdown in the Fed’s control of monetary policy have demanded forceful and permanent changes in repo policy. Some bankers have argued that the Fed should intervene in the repo market routinely, not only during emergencies, and particularly at the end of financial quarters, when banks’ cash requirements surge.
But does the repo market really warrant “constructive support” from regulators? Markets in physical commodities provide valuable price signals: A rise in the price of tin encourages miners to increase production and users to reduce consumption of the metal. But the Fed produces cash as if out of thin air and requires banks to hold this cash through its reserve requirements. Routine Fed examinations also encourage a ritualistic charade that increases the volatility of repo borrowing: Every three months banks scramble for cash to pass end-of-quarter checks of their reserves. The Fed’s monetary policies affect the availability of Treasury bonds used as repo collateral. And regulators set the rules for the repo market: Intraday borrowing, for instance, is prohibited.
Fiat money, macromonetary policies and bank reserve requirements interact with factors that regulators cannot control—such as tax payments, Japanese holidays and drone-and-missile strikes—to make the repo market complex and unstable. Complexity tends to beget more complexity. Temporizing fixes would only increase the repo market’s complexity and expand the power of insiders who claim to know what’s going on. As we know from 2008, byzantine complexity can cause financial systems to collapse.
Simplifying how banks borrow would make the banking system more resilient. The Fed could routinely offer unsecured short-term loans to banks at its target fed-funds rate. Now, the Fed grudgingly lends at its “discount rate” against eligible collateral pledged by banks. Regulators could also decouple overnight borrowing from collateral pledges by explicitly guaranteeing all the short-term liabilities banks hold. To discourage overborrowing, regulators could charge a guarantee fee, as the Federal Deposit Insurance Corp. does.
Free-market purists might regard such guarantees as an abhorrent elimination of market discipline. But there isn’t much market discipline to eliminate. Besides, we already have a parallel system that shows banks can successfully access credit without providing collateral or detailed financial information to lenders—the fed-funds market. Banks lending to each other in that market trust that regulatory examination, the Fed’s safety net, and whatever discipline stock- and bond-holders might provide are sufficient to make the system function properly except in times of acute stress.
Simple, explicit guarantees would reduce the complexity of short-term borrowing by banks. Banks would gain more-stable funding, and all would benefit from a sounder banking system. The way to prevent future repo crises, as in Scandinavia, is to make the market unnecessary.