Is the Fed taking baby steps back toward QE?
On September 17, the U.S. secured overnight financing rate (SFOR) spiked to nearly 10 per cent from its usual range of 2.00 per cent to 2.25 per cent as a sudden and acute funding shortage hit the overnight repurchase agreement (repo) market.
For the first time since the GFC, the Federal Reserve Bank of New York was forced to step into the overnight market to help ease the funding stress by offering up to US$75 billion of overnight liquidity. The market blamed the unexpected spike on a confluence of liquidity-draining events including the deadline for corporate tax payments and a large U.S. Treasury auction, but subsequent repo operations by the New York Fed suggest it was more than unfortunate timing that roiled the overnight market.
The overnight market is the murky plumbing of the financial system, where banks with excess reserves can lend to banks that are short on reserves in exchange for risk-free collateral and an agreement to repurchase said collateral later (hence “repo”). Every night, more than US$1 trillion is exchanged in this market and the rate at which banks lend to each other closely tracks the Fed Funds rate set by the Fed, and thus forms the transition mechanism for the Fed’s monetary policy. This obscure corner of the market typically only makes headlines when something is wrong, and so far, the overnight liquidity shortage appears to be accelerating.
Overnight, the New York Fed’s first US$30 billion term repo operation (with a term of 14 days rather than overnight) ended with over twice the demand for short-term funding than was on offer. The term repo, first of its kind, came in addition to the regular US$75 billion overnight repo operation that was also US$5 billion oversubscribed. Taken together, the demand indicates a funding shortfall of over US$140 billion. The New York Fed is expected to conduct two further term repo operations this week, bringing total temporary liquidity up to US$165 billion.
There are several plausible structural explanations for what appears to be a sudden shortage of bank reserves in the system. As a refresher, “reserves” are funds that banks, the Treasury and some other institutions keep with the central bank. When the Fed conducted QE following the GFC, it bought Treasuries from dealers (typically banks) and credited their accounts at the central bank with newly created reserves. Thus, as the Fed conducted QE, it injected excess reserves into the system, to the tune of US$2.7 trillion when the Fed’s balance sheet peaked at US$4.5 trillion in late 2014. Since the end of QE, excess reserves have been drained by Treasury borrowing, and this draining of reserves accelerated when the Fed started to shrink its balance sheet at the start of 2018.
The second structural contributor to the funding shortage is Treasury borrowing. Back when QE was running, the Treasury could borrow without primary dealers running into liquidity problems because the Fed was creating excess reserves to absorb the Treasury supply. Now that the Fed has stopped printing money and partially unwound its balance sheet, Treasury borrowing effectively reduces bank reserves dollar for dollar (until government spending finds its way back into the banking system). The yield curve inversion reduces demand for longer-dated Treasury bonds by carry traders, so primary dealer inventory is likely already inflated. The Treasury’s plan to issue more than US$800 billion of new debt between July and December this year will likely put even further stress on the overnight market heading into year-end.
Considering the overnight market is the transmission mechanism for monetary policy, sustained upward pressure on the SOFR is effectively a rate hike which the Fed may need to counter with more aggressive rate cuts. The alternatives are either a standing repo facility (i.e. QE) or a government budget surplus, both of which are politically tough sells. The upshot for investors is that rates are likely to stay (even) lower for longer.