In a recent seminar, I had the chance to listen to a talk by Raghuram Rajan discussing his paper titled Liquidity Dependence and the Waxing and Waning of Central Bank Balance Sheets. The paper examines the new situation of central bank policies, where the cycle between QE and QT creates financial instabilities instead of mitigating them. 

Key Points from the Paper

The paper puts forward the idea of "liquidity dependence," where the supply of reserves during quantitative easing (QE) leads to an increase in bank liquidity claims. In turn, an even larger central bank balance sheet is needed. This buildup of liquidity claims can make quantitative tightening (QT) difficult, while also dulling the impact of QE on the broader economy. 

Some key findings from the paper include:

• During QE, commercial banks financed the increase in reserves by issuing more demand deposits and reducing time deposits. They also issue more credit lines.
• When the Fed halted QE and started QT, there was no corresponding decrease in bank demandable claims. This left the system more sensitive to liquidity shocks.
• Weakly capitalized banks were found to engage more in maturity shortening of deposits during QE, taking on more liquidity risk for modest returns. They were also more exposed during the COVID liquidity shock.
• The authors suggest that more liquidity regulation and supervisory policy that accounts for changing conditions may be needed to mitigate these issues.

Visualizing the Change in Demandable and Time Deposits 

Simply put, during QE, time deposits are replaced by demandable ones:

And during QT, banks experience a shrink in asset value as a result of interest rate risk:

This can explain the SVB bank case, where its long-term securities lose value. On its liability side, there remains a lot of demandable, instead of time deposits, which increases its liquidity risk. 

Here is a simple visual analysis, where dimmed regions represent QEs. A similar figure can be found in the paper. Note that the QT period starting in 2015 does not change either the time or demandable deposits. 

Notes from the Talk

• A timely mention of Goodhart's Law and Murphy's Law in the case of monetary policy. Continued drastic policy swings bring unexpected changes. 
• The asymmetry between QE and QT. QE's impact is more focused on pension funds and QT on commercial banks.
• The increase in demandable deposits and decrease in time deposits during QE made the banking system more fragile. QT does not fix this. 

During the question section, topics such as holding assets to maturity, deposit betas, duration risks, and CBDCs were discussed.

Ending Notes

Finally, some excerpts from Keeping At It by Paul Volcker, his autobiography:

A lesson from my career is that such success can carry the seeds of its own destruction. I’ve watched country after country, faced with damaging inflation, fight to restore stability. Then, with victory in sight, the authorities relax and accept a “little inflation” in the hope of stimulating further growth, only to see the process resume all over again. The sad history of economic policy over much of Latin America offers too many examples.

...

The late Bill Martin, as I recalled earlier, is famous for his remark that the job of the central bank is to take away the punch bowl just when the party gets going. The hard fact of life is that few hosts want to end the party prematurely. They wait too long and when the risks are evident the real damage is done.