Source: Golden Ten Data
Author: Wu Yu

So far, any austerity impact on the real economy caused by the Fed's balance sheet reduction may have been mitigated, but an important buffer is rapidly receding.

Whether the Fed's wording change last week acknowledged this potential tightening is an open question, but credit and liquidity observers are calling for the time to consider how long the Fed's bank reserves will remain stable in 2023.

The Fed's reverse repurchase tool is rapidly draining

The debate focused on financial liquidity management issues associated with the Federal Reserve's “quantitative tightening” (QT). Over the past 18 months, the Federal Reserve has been gradually selling off the $7.8 trillion in bonds and notes it still holds on its balance sheet.

Due to the Federal Reserve's large debt purchases during the COVID-19 pandemic, the size of its balance sheet has more than doubled, reaching a peak of nearly $9 trillion. The Federal Reserve buys bonds from banks and lends its commercial bank reserves to banks, which can then use these reserves to continue lending to businesses and households.

As inflation soared, the Federal Reserve announced the launch of QT and has been selling these bonds at a rate of $95 billion per month since mid-2022, including $60 billion in US Treasury bonds and $35 billion in mortgage securitized assets (MBS).

But the Federal Reserve has also been absorbing what Atlanta Fed Chairman Bostic once called “pure excess liquidity” through its overnight reverse repurchase tool (RRP). Usage of this tool reached a record peak of $2.554 trillion at the end of last year.

As the Fed increases the return on RRP in accordance with policy interest rates, money market funds and banks see it as an extremely attractive “destination,” but currently the funds they have stored there are being withdrawn at a very rapid pace, and are scrambling to lock in high yields on a large number of new longer-term bonds before the Fed cuts interest rates.

Currently, the scale of use of RRP tools has been reduced by about $1.6 trillion, compared to just $768 billion last week. Judging from the rate at which RRP has shrunk over the past 3 months, it will run out in March next year. Perhaps coincidentally, the interest rate futures market currently expects the probability that the Fed will cut interest rates by at least 25 basis points in March next year by more than two-thirds.

Perhaps the sharp decline in the use of RRP instruments is only a fluctuation and detour in the currency market, but this sign may also be critical: whether for Bank of America credit, which is likely to face next year's economic slowdown, or asset prices, many believe that asset prices are directly affected by changes in capital liquidity in the financial system.

Solomon Tadesse (Solomon Tadesse) of Société Générale said that the key to how QT affects overall credit is how it erodes bank reserves, yet so far, the impact seems to have been minimal due to the influx of large amounts of excess cash into the RRP buffer zone.

Tadesse pointed out that in another example of the Fed's QT action in 2017-2019, the decline in bank reserves was in sync with the reduction in the Fed's overall balance sheet. But not this time.

The consequences of not having a “QT buffer”

“Although the Federal Reserve is continuing QT, the main channel for this move to influence liquidity has not worked. That is, the Fed's bank reserves have remained unchanged, as most of the capital to finance the surge in government debt has flowed out of RRP,” Tadesse said.

In the last QT operation, the Fed's overall balance sheet was cut by more than 600 billion US dollars in less than two years until 2019, and bank reserves were reduced by more than 800 billion US dollars.

However, in the past 18 months, bank reserves are still at mid-2022 levels despite the Fed's balance sheet being cut by about $1.1 trillion.

Tadesse pointed out, “In the 2017-19 QT, the sell-off of assets on the Fed's balance sheet was almost directly proportional to the reduction in bank reserves, thereby extracting liquidity from the system more quickly... and leading to a liquidity crunch in September 2019. This is currently not the case, as the RRP buffer is absorbing the loss of liquidity expected by QT.”

His key point is that as RRP shrinks further, QT will hit bank reserves and begin to have a greater impact on already deteriorated credit growth. So far, the decline in credit growth is mainly due to falling demand for credit due to rising borrowing costs and declining investment prospects.

This will weaken asset prices at that time. Presumably the Fed could have anticipated this too. But whether to change the policy approach to overcome this problem is still a matter of debate.

Last week, New York Federal Reserve Chairman Williams claimed that the depletion of RRP was on schedule, that some time was still to go, and that a shortage of bank reserves was far from over.

At the beginning of November, three officials, including Federal Reserve Vice Chairman Jefferson, told a US senator that it is not clear how long the reduction in balance sheet size will last, but there are no fixed goals, and it will not end soon.

However, if QT is not controlled, the decline in credit demand and supply may impact the US economy more strongly next year than the tightening process so far.

This is probably just one of the long-term and variable lagging effects associated with the impact of monetary policy, and may also highlight why the Fed is preparing to cut interest rates. Changes in the speed of policy shifts may determine the consequences.

As for the impact on the market, other liquidity experts insist that more depends on global liquidity. Global nominal liquidity levels are actually picking up and are only slightly below this year's high. This does not take into account the liquidity that the European Central Bank, which may be the first to cut interest rates, is about to release. But for the US, without a “QT buffer,” next year is likely to be even more rocky.

editor/tolk