When liabilities move into assets – What we can tell from the SVB bank run, and what might be next
Recent events involving several banks in the USA are a clear reminder that interest rates risk and ALM management is the core of banking – the balance sheet is everything. Just following ALM, Liquidity and Accounting regulations is not enough. A lesson learned is that unrealized losses in banks liquidity portfolios is not just an “accounting issue”.
Few would have expected to see the bank run in the Silicon Valley. Until last week. The overall context needs to be carefully understood. Not at the least to manage future interest rate and liquidity risk scenarios.
The niche bank Silicon Valley Bank hit by the recent crisis was funded by deposits in the form of raised funds from primarily tech start-ups, but also from private individuals.
When the venture funding market dried up about half a year ago, it inevitably had to lead to a structural cash outflow. A fast burn rate of the tech start-ups deposits, no new deposits and unsuccessful capital raising rounds is a lethal combination.
In order to cover for their deposit outflows, SVB had to sell substantial amounts of their assets. This turn of events highlighted a significant risk within the banking systems, where the steep increase in interest rates have led to unrealized losses in the banks liquidity portfolios of unignorable scale.
With the benefit of hindsight, EVE calculations, which are not affected by different accounting principles, need to be performed and reported by all sizes of banks. This is a regulatory requirement in the EU, but it is unclear how well EU banks have implemented current interest rates regulations. It may also be concluded that the effect of a high degree of deposit funding requires duly performed liquidity risk analysis, considering also that deposits not covered by a state deposit guarantee scheme and with high rates are not sticky.
Rates and liquidity risks must be analyzed thoroughly and covered in an ICLAAP – if the outcome of the analysis is that excess capital and liquidity is needed to cover risks, it is not prudent to maximize the return on capital. In the case of SVB, this materialized as significant capital shortage when assets had to be sold at market value far below accounting values. This resulted in a need for a new issue of USD2.25bn in stock. The obvious equity crunch caused depositors to make a run for it. The rest we now know.
Only 2,7% of the Bank’s deposits were fully covered by the US deposit guarantee limit of USD250K, which likely led to very swift and large outflows. To further set the perfect scenario for a bank-run, SVB was highly over-funded with on-demand deposits, which SVB could not invest in its traditional lending business. Instead, SVB bought US treasury bonds and MBSs, corresponding to 57% of their total assets, or about USD120bn. In practice, this made SVB more of an asset manager than a traditional bank.
The majority, or roughly 80%, of the liquid assets on the balance sheet of SVB were accounted for as “hold to maturity”, meaning that changes in asset market value were not accounted for through the bank’s PnL. The purpose of the investments was not, and should not have been, to sell them. The purpose was to yield interest income. Is this why SVB did not hedge risks and the PnL effects of their positions?
To further increase the interest rate risk in SVB’s balance sheet, the average duration of these investments was, according to the annual reports as of December 2022, 6.2 years. Even if the assets never were intended to be sold, SVB seems to have opted for a maximized NII by increasing the duration of the assets without any consideration of the EVE risk they were exposed to.
The collapse of SVB demonstrates that a bank can never completely overlook the aspect of the structural interest rate risk in the balance sheet. If an analysis of the interest rates risk had been done, a simple EVE stress calculation with a 200 bp jump in rates would have shown a approx. 11 % decline in asset value for the hold-to-maturity bonds, corresponding to roughly 13.2bn to be compared to the 16bn of equity.
This development further calls for the question, what if SVB had been obliged to adhere to the increased regulatory requirement on asset and liability management and liquidity risk after the 2008 financial crisis? Now, due to the regulatory threshold implemented in 2018 in the US they were not. The same counts for many other financial institutions that are generally referred to as niche banks which may by comparison be quite large.
From an outside perspective, it seems that the SVB followed the regulatory requirements on HQLA (high quality liquid assets). However, it appears likely that SVB may have broken the regulatory requirements on LCR and NSFR if they would have been applicable. Monitoring of regulatory measurements can identify increased liquidity and interest rate risk, but regulatory compliance alone is never an assurance of proper rates and liquidity management. The SVB case demonstrates that applying liquidity management is not just about regulatory requirements, it is a case of business continuity, if not survival.
FCG is working increasingly with clients to implement new interest rates risk management models in the banking book regulation. Generally, given the market developments, banks may need to further strengthen their rates risk management, but also more actively address their ICLAAP work in rates and liquidity areas.
For more information, please contact: