Stop guessing if this is a Banking Crisis!

An evidence-based framework to determine if this is a systemic banking crisis

Mar 22, 2023

Two key questions determine the outcome of the current banking turmoil. First, do recent events qualify as a ‘systemic banking crisis?’ And second, if the answer to the first question is ‘yes,’ are policy interventions significant enough to end it?

During times of elevated uncertainty – it’s virtually impossible to have any conviction on whether First Republic Bank will make it – it’s crucial to put things into the ‘right’ perspective. Especially with so many (self-proclaimed) pundits sharing their uninformed views.

Laeven and Valencia (2012, 2018, 2020) provide the leading database of banking crises going back to 1970. They identify a total of 151 banking crises over the period 1970–2017.

Their work is consistent and empirically sound, giving an insightful overview of the (common) policy responses we have witnessed over time.

The definition of a systemic banking crisis

Laeven and Valencia define a banking crisis as systemic when two conditions are met:

Significant signs of financial distress in the banking system (as indicated by significant bank runs, losses in the banking system, and/or bank liquidations.)

Significant banking policy intervention measures in response to significant losses in the banking system.

I put a poll out on Twitter and LinkedIn asking if these two conditions are currently met. And the first results (the poll is ongoing) reveal the jury is still out. On Twitter, 58% responded that both conditions are met, against 52% on LinkedIn. This outcome is one (of many) reasons why equity markets have held up so well despite the banking sector issues spreading.

jeroen blokland @jsblokland

POLL! 1) Significant signs of financial distress in the banking system (as indicated by significant bank runs, losses in the banking system, and/or bank liquidations)? 2) Significant banking policy intervention measures in response to significant losses in the banking system?

Concerning the first, I think this condition has been met. Bank runs caused the swift collapse of Signature Bank and Silicon Valley Bank, the 2nd and 3rd biggest bank failures in US history.

This chart from the Federal Deposit Insurance Corporation (FDIC) website reveals that while there have been only two failures in 2023 – the FDIC does not count Silvergate’ Bank’ as a bank failure – the volume of the assets involved, USD 319 billion, is strikingly close to that linked to the 25 banks that failed in 2008. And it’s almost double the assets linked to the 140 banks that failed in 2009. These numbers exclude the Credit Suisse – UBS deal, which will likely go down as an ‘acquisition’ despite CS failing as well. Based on the Q4 numbers, CS’s balance sheet held CHF 233 billion (USD 252 billion) in deposits.

In addition, this now well-known chart from FDIC shows US banks have massive (paper) losses on the securities they bought with their deposits. As recently as March 6, FDIC Chairman Martin Gruenberg stated that ‘the current interest rate environment has had dramatic effects on the profitability and risk profile of banks’ funding and investment strategies.’ And he added, ‘Unrealized losses on securities have meaningfully reduced the reported equity capital of the banking industry.’ Hence, the criteria for significant banking system losses has also been met.

The second condition – significant banking policy intervention – likely divides the poll respondents. What makes policy intervention significant? Laeven and Valencia (2012) ‘consider policy interventions in the banking sector to be significant if at least three out of the following six measures have been used:’

extensive liquidity support (5 percent of deposits and liabilities to nonresidents)
bank restructuring gross costs (at least 3 percent of GDP)
significant bank nationalizations
significant guarantees put in place
significant asset purchases (at least 5 percent of GDP)
deposit freezes and/or bank holidays.

1. Extensive liquidity support

The FDIC announcing that all Signature Bank and Silicon Valley Bank depositors will be made ‘whole,’ are two more examples of what we have seen many times before. Residential deposits are guaranteed or protected during almost every banking crisis. Without this, the fractional banking system would lose all credibility and implode. Therefore, Laeven and Valencia focus on liquidity provided to institutions directly by the central bank and government, including cross-border measures like extending currency swap lines among central banks, announced on March 19.

However, little use has been of the swap extensions, as shown in the chart below. In addition, the sum of borrowing at the Fed’s Discount window and the new Bank Term Funding Program, USD 160 billion, represents less than 1% of total US Commercial Bank Deposits.

In the case of Credit Suisse and UBS, the case is less straightforward. The Swiss National Bank has stated that both banks have unrestricted access to the SNB’s existing facilities and that Credit Suisse and UBS can obtain a liquidity assistance loan with privileged creditor status in bankruptcy for a total amount of up to CHF 100 billion guarantee. With Q4 deposits of the combination totaling CHF 860 billion, liquidity support potentially tops 5% of deposits. Technically, the number of CHF 100 billion is set as a maximum, also to shield UBS from execution risks as it tries to ‘integrate’ CS. But overall, this would likely qualify as extensive liquidity support.

2. Bank restructuring gross costs

There has been no bank restructuring, with US national and regional banks swooping up part of the failed banks’ deposits. The same is the case in Switzerland, where UBS has taken over Credit Suisse, and the mess that comes with it, backed by Swiss National Bank (SNB) guarantees.

Interestingly, Laeven and Valencia note that while macroeconomic policies have been used aggressively in recent advanced economy crises, actual bank restructuring has been relatively slow. This raises questions about the pace of

recovery and the optimal policy mix in resolving financial crises in advanced economies.

3. Significant bank nationalizations

Bank nationalizations were crucial to the ‘solution’ during the Great Financial Crisis. The current environment includes no nationalizations until now.

4. Significant guarantees put in place

As described above, this is the case in Switzerland but not in the US. First, while a significant range of eligible securities may be deposited at the Fed Bank Term Lending Program at par value – theoretically omitting the issue of paper losses – more exotic instruments on the banks’ trading books have not been included. Second, with national and regional banks taking over failing competitors’ deposits, the current dynamic resembles a game of ‘Whac-A-Mole.’ No structural, over-arching solution has been put into place. One example of this would be to guarantee all regional bank deposits.

5. Significant asset purchases (at least 5 percent of GDP)

Laeven and Valencia refer to asset purchases from financial institutions implemented through either the Treasury or the central bank. With several central banks trying to reduce their balance sheets, including quantitative tightening – this condition of significant policy intervention is not met.

Even though the Federal Reserve’s balance sheet spiked by nearly USD 300 billion in the week of March 15, this was almost entirely explained by lending at the Fed’s Discount Window, the Bank Term Funding Program, and FDIC loans to make Signature Bank and Silicon Valley Bank depositors whole. Recent policy intervention by the Fed should not be considered Quantitative Easing but does qualify as liquidity.

6. Deposit freezes and/or bank holidays.

As Laeven and Valencia point out, deposit freezes, while rare, are most frequently used by emerging economies.

Not ticking the boxes

Based on the definition used by Laeven and Valencia in their Systemic banking crisis database, we have not seen ‘significant banking policy intervention.’ Moreover, even if we focus on the case of Switzerland exclusively, only two out of the six criteria for significant policy intervention are met, and not at least three, as suggested by Laeven and Valencia.

As a result, currently, only the first of the two conditions for a systemic banking crisis is being met:

Significant signs of financial distress in the banking system (as indicated by significant bank runs, losses in the banking system, and/or bank liquidations.)

And not,

Significant banking policy intervention measures in response to significant losses in the banking system.

Obviously, the caveat here may be that the banking crisis will grow systemic BECAUSE policy intervention has been relatively muted compared to previous crises. With the US banking intervention resembling a game of ‘Whac-A-Mole,’ investors should not underestimate this possibility. As the Great Financial Crisis painfully demonstrated, things can continue heading south for years before the crisis finally arrives. One final observation that may suggest we are not out of the woods yet is that most banking crises have historically started in the second half of the year.