As our FT colleagues wrote last week, people in the US are worried right that we are going to see another spike in the repo rate — that is the rate that US banks pay to borrow from one another, with collateral, overnight — as the year draws to a close.
We have already seen a substantial spike in repo rates this year, such as back in September when the overnight rate jumped to just shy of 10 per cent. Bear in mind that at the time, the upper limit of the fed funds target range, which the repo rate tends to track, was 2.25 per cent.
The reasons driving those fears are numerous. But we want to focus on one in particular here: the impact of the Fed’s attempts to squeeze some of its post-crisis liquidity out of the system.
When the crisis struck and the Fed’s balance sheet ballooned, there was so much liquidity in the system that it was more a question about keeping the repo rate within and not below the range targeted by the Fed than above it:
Last year, as the chart above shows, we saw the Fed begin to shrink its balance sheet as it sold off some of the Treasuries it has bought under its quantitative easing programme. It appears that markets are not handling that withdrawal of liquidity very well.
At the same time John Williams, the president of the New York Federal Reserve, the branch of the central bank that doles out — and removes — liquidity from the banking system, has little sympathy for participants’ pleas to provide more support to the repo market.
This is what he and Lorie Logan, who heads the markets group at the Fed, told the FT after the late September spike:
The Federal Reserve Bank of New York is examining why banks with excess cash failed to lend to the overnight money market, following a week that revealed cracks in the US’s financial plumbing.
John Williams, president of the New York Fed, on Friday questioned the hesitance of the banks in an interview with the FT. “The thing we need to be focused on today is not so much the level of reserves [held at the Fed],” he said. “It’s how does the market function.”
Mr Williams and Lorie Logan, senior vice-president in the markets group at the New York Fed, said officials were looking at why cash failed to move from banks’ accounts at the Fed into the repo market, where banks and investors borrow money in exchange for Treasuries to cover short-term funding needs.
Ms Logan pointed to the concentration of excess cash at a small number of banks as one potential issue. “Reserves are concentrated, the excess reserves relative to the minimum level each bank is demanding is concentrated,” she said. “And the key question is how those reserves, as the level was coming down, would get redistributed, and how smooth that redistribution process would be.”
You might say Williams has a point in placing the onus on market participants to lend, rather than on the Fed to dole out more liquidity. The Fed’s balance sheet is still way above its pre-crisis level so there should in theory still be enough reserves out there to leave the repo rate close to the federal funds target range.
Yet we think the way in which they are viewing the “excess reserves” issue requires a closer look. The prime reason being that the common definition of excess reserves — that is, reserves that banks have over and above those which need to be held at the central bank — might no longer apply.
We spoke to money markets guru Lou Crandall of Wrightson ICAP and his thinking was that, rather than two levels of reserves — required and excess, it made more sense to think about four.
First come the absolute must-have reserves: those needed to meet banks’ enhanced post-crisis liquidity requirements.
Second, there are the nice-to-have reserves that aren’t strictly necessary, but give banks a greater level of comfort in dealing with cash flow uncertainty. These are the reserve holdings that may become stickier around quarter-end.
On top of that, lies a layer of Fed balances that banks didn’t necessarily plan on holding, but were quite willing to have as a more liquid substitute for Treasuries.
Last was an “excess” layer that was truly surplus to requirements.
It’s important to stress how much of a big deal money market participants also think regulatory constraints are. On calendar dates at the end of each month, various regulatory snapshots limit the amount of capital banks can deploy into funding markets. The issue of adequate reserves also comes into play on large collateral settlement and tax payment dates, as in mid-September. In turn, rules around daylight overdrafts mean the US banking system needs an abundance — we are talking two times above requirements here — of reserves to ensure intraday payment liquidity is adequate. A lot of people in the market want the Fed to allow access to its discount window to meet daylight overdrafts, but the central bank is not keen.
Let’s not forget too that we have a whole generation of people in financial markets who have never experienced anything but a very cheap and very plentiful supply of central bank cash, and who are likely to panic at the first signs of stress.
Why does all of this matter? Well because it suggests the Fed is vastly underestimating the level of liquidity market participants need in order to keep on lending to one another.
Fed chair Jerome Powell has signalled a change in stance on daylight overdrafts. But unless the Fed actually follows through in the coming weeks, expect more spikes at the close of 2019.