Banks avoid senior bail-in
APRA has released an important new consultation paper on the regulator's approach to ensuring the major banks have sufficient "Total Loss Absorbing Capacity" (or TLAC). I have written at length about this in the AFR here (or AFR subs can click here).
You might recall that since the financial system inquiry (FSI) report was released, APRA has required the big banks to build "unquestionably strong" going concern, or core first-loss equity, capital amounting to at least 10.5% of their risk-weighted assets, which is among the highest risk-weighted equity ratios of any large banks globally. (I helped draft the original terms of reference for the FSI with David Murray and Joe Hockey in 2010 and repeatedly argued the banks needed to radically deleverage in the years that followed.)
This has resulted in the major banks deleveraging their balance-sheets from roughly 30 times their non-risk-weighted equity in 2013 to about 18 times today, which is one of the main reasons their returns on equity are converging towards their cost of equity, as we argued they would in 2015 (and called for their price/book multiples to drop from 2-3x to 1.0-1.5x where they sit today).
In this new paper, which APRA will discuss with the industry over the next few months, it takes a uniquely Australian approach to solving the TLAC problem, as we have said it should do. APRA has crucially avoided requiring the major banks' senior bonds to be bail-in-able into equity, which is now common practice in most countries overseas. It has also targeted an additional layer of TLAC summing to 4-5% of risk-weighted assets, which the banks will have until 2023 to develop. The idea is that the extra 4-5% of TLAC is "gone concern", rather than "going concern", capital that can be used to help recapitalise a failing bank that has wiped out most of its existing equity.
Rating agencies cheered this approach as "credit positive" for the banks, with S&P flagging that it would upgrade the majors' senior bonds to AA- "stable" from their current "negative" outlook. This will ensure the banks can continue to raise capital globally via their senior bonds in the most stable and low-cost manner possible. Indeed, the major banks' senior bonds will emerge as one of the few classes of senior-ranking debt securities globally that are not bail-in-able into equity and which are still eligible to be sold to central banks as part of their liquidity, or repurchase, facilities. (We have long held the contrarian position that this is what would come to pass.)
Most other countries have forced their banks to replace their traditional senior bonds with bail-in-able alternatives. As a little known technical aside, APRA can, in fact, indirectly impose losses on---as opposed to directly bail-in---the banks' senior securities through APRA's asset transfer powers under the Banking Act whereby it can take control of a failing bank and then sell its assets to a third-party at fair market value with no time constraints. This is really no different to the economic outcome that any senior creditor would receive in bankruptcy, although given the large banks' access to government guaranteed deposits and government guaranteed liquidity via the RBA's Committed Liquidity Facility, it is an extremely unlikely, if not impossible, contingency.
For what it is worth, this loss-absorption mechanism is almost identical to the one US lawmakers developed under the post-GFC Dodd Frank Act (Title II). It means that while Aussie banks might not have as much explicitly bail-in-able capital as peers overseas in the sense of securities that can be swapped into equity under contract or statute, they do have world-beating theoretical loss-absorbing capacity if you count loss absorption via asset transfers. This in turn means that APRA does not have to double-up and insist on the banks having the best official TLAC ratios in the world. (I realise this is a nerdy distinction that most are uninterested in.)
The one wrinkle in APRA's paper that will need to be addressed is how the banks raise the official TLAC. APRA has suggested they can use existing regulatory capital securities (CET1, AT1 hybrids and T2 bonds), and argued that most of the TLAC could be sourced via the "cheaper" T2 bonds. After accounting for existing T2 maturities, the new issuance required to raise 4-5% TLAC, and growth in risk-weighted assets over time, this implies up to $100 billion of issuance.
Anyone with deep fixed-income expertise will recognise that sourcing $100 billion of funding via T2 issuance is a pipe-dream and would require current global demand for T2 bonds to more than double. Put differently, this funding does not actually exist for any banks---and certainly not Aussie banks.
One counter might be that perhaps the funding might emerge at the "right price". Beyond the fact that this price would be immensely expensive, and be passed on to the banks' borrowers (lifting lending costs), we also know that when financial markets become volatile it is very difficult to raise T2 capital at any price, let alone $30 billion plus every year. (In 2015 and 2016 insurers and the major banks struggled to close relatively modest T2 issues.)
APRA is upfront in its report in revealing that it has not carried out any capacity analysis on whether this funding exists nor any analysis of the price that would be required during rosy market conditions (there will be no price during turbulent markets)---it simply assumed that the banks can raise up to $100 billion at "current spreads", which is clearly not going to be the case.
So, what is the solution? The first is that APRA could slash the TLAC requirement to something manageable, like 1-2% of RWA. But this would potentially leave the banks short of the minimum bail-in-able equity required during a resolution process.
An obvious fix is to simply allow the banks to issue the bail-in-able bonds that global investors are prepared to fund in vast volumes: these are typically called Tier 3 or non-preferred senior securities and sit between T2 and senior bonds in the capital structure. As long as APRA can bail this stuff into equity via a non-viability clause, which is found in existing AT1 and T2, it should be indifferent as to whether the banks fund the 4-5% TLAC gap via AT1, T2 or T3.
Rightly or wrongly, global investors are much more comfortable with T3. In particular, global investors tend to have much larger portfolio limits for T3 or non-preferred senior bonds than they do for T2 subordinated securities, which is why the funding exists. I personally think this is silly---to my mind, T2 and T3 are almost identical. But the fact that so much more capital exists for T3 than T2, and at a much lower price, makes this a bit of reverse regulatory arbitrage for APRA.
We have actually seen this debate unfold before. Back when the first draft TLAC standards were released in 2014, the world thought there would be a tsunami of T2 issuance. Then the banks all sat down with their regulators and agreed that they needed a product that investors were happier to fund. And so we saw an explosion in T3 or non-preferred senior issuance post 2015, which has become the main TLAC asset-class globally that banks are reliant on. Nowadays there is relatively little global T2 issuance, but massive amounts of T3 hitting markets every day. (I personally dislike T3 and prefer good ole non-bail-inable senior bonds!)
Another rejoinder might be: but won't T3 just price like T2? Not if the global precedents are any guide. T3 has proved much cheaper for banks than T2. That's probably a function of the fact that the global funding for T3 is much larger than T2. The major banks' T3 will price off global T3 curves, which will be cheaper and far more liquid than T2.
A final point is that we do not want the major banks ideally sourcing their bail-in-able AT1, T2 and T3 capital from local (domestic) investors. This means the catastrophe insurance that APRA is seeking is being funded by Australians, when the practice overseas has been to try and diversify away this risk by getting banks to issue T3 around the world. APRA does not want to be bailing-in super funds, the Future Fund, or retail investors at the same time as our banks are blowing up and their equity portfolios are getting smashed. They should ideally be bailing-in offshore investors that therefore share some of this risk with Australians.
Chris co-founded Coolabah in 2011, which today runs $7 billion with a team of 33 executives focussed on generating credit alpha from mispricings across fixed-income markets. In 2019, Chris was selected as one of FE fundinfo’s Top 10 “Alpha...