Chaos in Repo
Updated: May 1
Repo is short for “sale and repurchase agreement” which are effectively short-term collateralised loans, often overnight. They are the mechanism by which those in need of short-term funding, such as banks and brokers, get the cash they need from those with cash to lend, looking for a return. Normally, the rate of interest on these loans closely tracks the central bank policy rate because otherwise there would be an opportunity for primary dealers to make an arbitrage risk-free profit by borrowing at one rate and lending at the other.
On 17 September, the repo rate in the US spiked to as high as 10% intraday and 6% on a closing basis. In theory this should never happen due to the arbitrage opportunity, which begs the question why it happened? The obvious answer is that there was too much collateral (Treasury bonds) and not enough cash in the system. Those with funds were either unwilling or unable to lend as they normally would. There were lots of explanations such as a swath of new Treasury bond issuance hitting the market and large cash withdrawals for tax-related payments. However, at that moment there simply wasn’t enough money in the system to keep it running.
The Federal Reserve rode to the rescue. They immediately offered $75bn of funds into the repo market which was gobbled up immediately. Since then the Fed has offered larger repo facilities and over longer timeframes, which has seen its balance sheet reverse its decline and grow by around $300bn. There is the expectation that the Fed will have to do more, particularly over year-end when there is often a technical shortage of cash.
This wasn’t a crisis per se because the Fed can and will provide as much money as the system needs. However, it was a massively important signal on where this is heading. If the market rate for money without the Fed’s help is 6%-10% then very quickly the US yield curve would find itself in that postcode. The problem is that in our highly financialised economy, at 6%-10% everyone is dead. For example, with a $22tn debt burden to service, the interest cost alone is 40%-60% of US tax receipts. And that’s before the economy collapses. The Fed will provide ANY amount of dollars to stop this happening.
The Fed have said that this is not QE because of its temporary nature (the purchase and simultaneously agreed sale of the Treasury) rather than outright purchases. However, when one looks at arguably the main driver of the pressure in repo, which is the growing treasury issuance to fund growing US deficits, and realises that these deficits are structural and only going to get bigger, then one must also conclude that unless the Fed is genuinely ready to unleash Armageddon, their balance sheet MUST start to grow indefinitely over the long term.
The combination of US private sector and overseas demand will not be able to fund the US government requirements at the low rates required.That leaves the choice of “it doesn’t get funded” or “the Fed funds it”.The first choice means the US defaults internally to its citizens or externally to its creditors, an outcome to which we would attach a very, very low probability.The second, has major implications for asset markets and most notably the US dollar, as the Fed begins the long task of monetising future US deficits.