Global spillovers into Asian bank credit linger

The recent spate of Western bank failures has cast a spotlight on banking risks and chilled sentiment in financial credit. Despite the promise of support to depositors by US policymakers, liquidity pressures on US regional banks have not fully abated.  The Fed’s new Bank Term Funding Program (BTFP) facility has continued to see high demand, with usage growing to a record of USD87bn as of 17 May.



A regulatory write-down of Credit Suisse’s Additional Tier 1 (AT1) bonds, while preserving value for shareholders, has also upended creditor hierarchy understanding in financial credit. Given these developments, Asian bank credit has been adversely impacted, with investors still demanding significantly higher spreads (35-70bps) across subordinated credit, though these have partially narrowed from extremes in late March.




Assessing Asian banks’ liquidity risks

Are higher spreads in Asian financial credit warranted? With liquidity strains faced by Western banks being a key driver of financial credit risks, we examine the risks of similar liquidity stresses also occurring for Asian banks. We shall (i) discuss liquidity metrics for banks, (ii) analyse the impact of monetary policy on liquidity, as well as (iii) review institutional and market features that could influence liquidity risks in Asia.


LTD ratios indicate better Asian buffers

We focus on the LTD ratio as a traditional metric of liquidity risk—the risk that banks do not have sufficient liquid assets and must accept large discounts in their assets in exchange for cash on experiencing funding stresses. Comparing banks’ main illiquid asset (loans) against their most stable source of funds (deposits) indicates how vulnerable banks are to liquidity strains. The higher is the LTD ratio, the lesser is the liquidity buffer. During periods of deposit outflows, banks with high LTD ratio would need to quickly secure non-deposit funding which is usually more costly and less available subject to economic circumstance or realize losses in selling off their illiquid assets.



For US banks, we adapt the LTD ratio to include held-to-maturity (HTM) assets on top of loans. The failure of SVB shows that HTM securities matter for US banks, given their high holdings of these assets and large unrealized losses. All three recently restructured major banks (SVB, FRC, CS) show a very high adapted LDR ratio of 95% to 115%, compared to an average 79% for the Big 4 US banks. This clearly indicates their elevated vulnerability to liquidity strains.

While Basel III reforms have introduced the more sophisticated Liquidity Coverage Ratio (LCR) and Net Stable Funding Ratio (NFSR), LTD ratio are still relevant. First, not all banks will disclose LCR or NSFR, with larger banks typically the ones that do so. Second, assumptions underpinning the net outflows component of the LCR may not always hold, as evident from CS’s liquidity stresses despite a very high LCR of 144%. In contrast, the LTD ratio does not make assumptions about the rate of outflows and is more robust to extreme stress scenarios.



Given the sharp rise in US rates and US banks’ large holdings of long-duration MBS, we also include long-term assets (> 5Y maturity) into the US LTD ratio, as these assets will need to be funded to realizing losses. With our adapted LTD ratio being above 100%, US banks do appear to be vulnerable to liquidity risks. Similarly, European banks have a quite high LTD ratio of over 90%. For Asian banks, most have LTD ratios that are below 90%, except for Korean banks. This suggests that liquidity vulnerabilities are generally lesser in Asia compared to the West.

We note that Korean banks are a special case with high LTD ratios. This is because of their higher dependence on bank debentures sold to retail investors who want to earn higher rates. Korean debentures may thus be considered as a more stable source of funding, unlike typical debentures sold to institutions. In our view, this unique funding channel is likely more stable than wholesale funding, but do not offer the same degree of stability as bona fide deposits. We observe that the Korean debenture market was impacted negatively during the 2008 Global Financial Crisis. Korea’s LTD is also slated to fall this year, as the Financial Services Commission (FSC) had previously raised the LTD ratio cap to 105% given a credit crunch last year, before an eventual normalization to 100% in end June.

Besides being lower compared to the West, Asian LTD ratios have also generally been falling (or stayed stable) compared to their 10y average level from 2010 to 2019. This indicates that Asian banks have also not been aggressive in extending loans compared to historical norms. Only Chinese banks saw a significant 12% rise in their LTD ratio from the decade average, which largely reflects liberalization after China removed a 75% cap on the LTD ratio in 2015


Lesser risks from QT & monetary policy

Recent research by Acharya, Chauhan, Rajan, and Steffen (2022) suggest that banking vulnerabilities can perhaps be heightened by liquidity dependence. This means that bank deposits and credit tend to rise alongside an increase in reserve money under QE, but there is no mechanism for a symmetric decline in deposits when money supply falls under QT. As such, QT poses risks of systemic liquidity shocks to the banking sector as reserve money shrinks, while deposit claims stay unchanged.

Of course, such a view presupposes that there is no “excess” liquidity in the banking system. Concerns about system liquidity can also be partially mitigated by mechanisms to provide reserve money to banks, such as through the discount window or the Fed’s BTFP facility. Nevertheless, such reserve money facilities are costly compared to deposits, and there are thus economic constraints for banks to tap them.  Excessively tight liquidity conditions thus remain a potential risk under QT. How should we assess these conditions? The growth of monetary aggregates, such as M2, can be a useful metric, with lower M2 growth associated with tighter liquidity. Interestingly, only the US has seen its M2 growth turned negative on a y/y basis, after the Fed began quantitative tightening in June 22. Furthermore, the US has also seen the sharpest fall in M2 growth relative to its 5y average growth from 2018-2022.



As it stands, Asian and European monetary policies are nowhere as tight as in the US, with M2 growth still positive across all countries. Asian central banks also do not have overhang from a large balance sheet that needs to be shrunk, unlike the ECB which is slated to accelerate the pace of its balance sheet reduction in July by ending APP reinvestment. Thus, from a monetary perspective, we expect Asian banks’ liquidity to be less fraught with challenges, unlike US and European banks. Hong Kong banks could face some risks given the USD/HKD peg and a diminished aggregate balance, but risks are well-mitigated due to Hong Kong banks’ lower LTD ratio.


Institutional and market features help funding

In our previous publication (see Asian banks: Better insulated from contagion risks, 24 Mar 23), we highlighted that deposit funding are more stable for Asian banks for three reasons:

• Vast majority of deposit accounts (>90%) are fully insured
• High confidence with many institutions being partially owned, or sponsored, by governments
• General orientation towards retail and corporate clientele with high deposit stability

Furthermore, Asian bond markets have less longer-tenor bonds compared to the US. Over 60% of the US Treasury and agency MBS market comprises of bonds with over 10 years of remaining maturity. The inverse is true for Asia, with under 40% of government bonds having more than 10y of remaining maturity. This implies that scope for Asian banks to face duration risk mismanagement is low, even if rates volatility is to pick up with Asian central banks being more hawkish than expected. We do not see risks of unrealized losses for portfolios held at amortized costs being a risk or triggering deposit flight for Asian banks.


Credit risks and NPL trends have improved

One fact in banking is that liquidity strains do not just happen by random chance but are usually catalysed by increased uncertainty over credit quality. Given the wide profile of credit risks across Asian banks, we choose to focus only on Chinese real estate risks given widespread Chinese developer defaults since late last year, and a possibly high impact to the rest of Asia. From a macro perspective Chinese banking risks from real estate are likely small.

First, China’s property outlook has unquestionably improved in 2023. Following China’s reopening at the end of 2022, residential property prices have now rebounded strongly, with broad-based increases across major cities. Mortgage boycotts that were prevalent in Aug-Sep have also dissipated after the authorities decreed loan funding to resume construction. As such, NPL risks for Chinese banks’ mortgages should be significantly reduced.

One remaining concern is developers’ liquidity, which is still problematic. Property sales have also softened somewhat after a surge post-reopening. Nevertheless, the rise in collateral values is still good news, with developer loans typically secured by land and property. Overall, Chinese banks’ NPL and special mention loan ratios have steadily declined back to pre-pandemic levels. Large banks’ NPL ratio has eased to 1.27%, while shareholding banks’ NPL ratio has also fallen to a record low of 1.31%.

Finally, we observe a trend of declining NPL ratios across Asia ex-Japan. Most economies NPL ratios have peaked in mid-2021, before declining amid reopening momentum. Only Indonesian and Philippine banks continue to show NPL ratios that are still higher than pre-pandemic 2019 levels. But even for them, the trend of gradual improvement of asset quality has largely continued.


Confidence-building with upcoming AT1 calls

Despite the resilient Asian banking outlook, market confidence has, to a degree, been affected by the write-down of CS AT1 by FINMA in March. This case has also made clear that national regulators’ approach towards AT1 bondholders are not all the same. 

In Asia, MAS has affirmed that it intends to abide by the hierarchy of claims for liquidation during a bank resolution (or restructuring). Holders of Singaporean banks’ AT1 bonds are entitled to claim any difference between what they receive in a resolution, and what they would otherwise receive in a liquidation. Similarly, HKMA has clarified that capital holders should expect to be treated in resolution in accordance with the priority they would enjoy on a winding up of the institution. This implies that losses are to be first absorbed by equity holders, before AT1 and Tier 2 bond holders. However, no other Asian regulators, besides MAS and HKMA, had made a commitment to prioritize AT1 bondholders over equity holders in a restructuring.

Upholding creditor hierarchy in a resolution helps keeps the restructuring mechanism strategy-proof, with regulators’ inherent bias or private information being irrelevant in affecting the strategy of AT1 investors. The is because the outcome of AT1 bonds will obviously dominate the outcome for equity under all scenarios. If regulators have some discretion deciding how to allocate losses between equity and bondholders in a resolution, then the strategy of AT1 investors will become more difficult to analyse, as it is now subject to their expectations of what a regulator’s course of action could be. This could mean that AT1 investors will simply price the worst-case scenario of a permanent write-down (or equity conversion) when banks’ CET1 ratio falls below regulatory minimums, or when there is a viability-threatening liquidity event.

Tail risks are too small at this juncture for such worst-case considerations to matter. We expect Asian subordinated credit spreads to narrow further, as investor confidence returns eventually when a slew of AT1 calls materialize, starting in late July.


To read the full report, click here to Download the PDF. 

Chang Wei Liang

FX & Credit Strategist, Global

[email protected]