Hidden findings in banks' stress test

Christopher Joye

In the AFR today I reveal that the regulator's latest stress-tests of Australia's 13 largest banks show a vast improvement over the results of the 2014 stress-test when some banks' common equity tier one (CET1) capital ratios fell below 5%, triggering the conversion of their hybrid securities into equity. In the 2017 stress-test released this week APRA found that the big banks' CET1 ratios on average only fell from 10.5 per cent to slightly above 7 per cent, which means that hybrid securities would not have been converted into equity (given the capital trigger sits at 5.125%). This is despite APRA assuming house prices fell by 35%, GDP contracted 4%, and the unemployment rate jumped to 11% (AFR subs can click here). Brief excerpt:

The boss of Australia's banking regulator, Wayne Byres, delivered an important speech during the week that carried several hidden messages that are relevant to investors. The first was to crush UBS's housing "credit crunch" thesis by confirming this column's argument that the vast bulk of the tightening in lending standards has already occurred. "The heavy lifting on lending standards has largely been done," Byres explained, highlighting that the Australian Prudential Regulation Authority (APRA) had been compelling banks to improve credit quality since December 2014. For the sake of clarity, Byres predicted that "any tightening from here on is expected to be at the margin as banks seek to get a better handle on borrower expenses, and better visibility of borrower debt commitments". This slap in the face for doomsayers prophesying an acute credit crunch was reinforced on Wednesday by the official housing finance data, which showed that both the number and value of new loans rose in May, trumping economist expectations of a sharp fall. This reconciles with the Reserve Bank of Australia's data on the stock (as opposed to flow) of outstanding home loans, which likewise expanded in May and (as Byres also noted) climbed by a healthy 5.8 per cent over the past 12 months. Our central case has long been that there will be a welcome deceleration in housing credit creation, but that it should remain positive and will certainly look nothing like the deeply negative contraction projected by UBS's Jon Mott and his acolytes. APRA should be congratulated for engineering a striking improvement in financial system stability through marked reductions in: speculative investment lending (annual growth has more than halved since 2015); the share of riskier interest-only loans (down from over 50 per cent three years ago to below 20 per cent today); and the proportion of loans worth more than 90 per cent of the value of a property which for owner-occupiers (investors) has fallen from almost 15 per cent (8 per cent) in 2014 to less than 9 per cent (1 per cent). Since 2014 APRA has also required banks to assess borrowers' repayment capacity using a minimum 7 per cent interest rate, which is hundreds of basis points above prevailing rates. The most interesting part of Byres's speech was its discussion of the results of APRA's stress tests of Australia's biggest banks, which had not been disclosed for years. (It would be better if APRA published a consistent annual summary of this exercise.) APRA's stress test does a very good job of approximating the worst imaginable downturn with economic growth contracting 4 per cent, unemployment doubling to the 11 per cent level last seen in the 1991 recession, house prices falling by an unprecedented 35 per cent, wholesale funding markets closing coupled with rating downgrades, and credit spreads exploding. APRA also assumes that bank executives do nothing to mitigate the downturn (eg, by raising extra equity, cutting costs, and/or repricing loans). In 2017 APRA introduced a new headwind, which was "an operational risk loss event involving misconduct and mis‑selling in the origination of residential mortgages…adding a further shock to bank balance sheets". A key question was how the banks fared relative to the 2014 stress test, which Byres said " the conversion of additional tier 1 instruments…as losses mounted" for some institutions. In the 2017 exercise Byres reported that "the common equity tier 1 (CET1) ratio of the industry fell from around 10.5 per cent at the start of the scenario to a little over 7 per cent by year three", which, crucially, is miles above the 5.125 per cent CET1 level that triggers the conversion of bank hybrids into equity. It is also substantially better than the 6.3 per cent CET1 trough banks dropped down to in the 2014 analysis (some went below 5 per cent on an individual basis). Even when APRA added the new operational risk crisis into the mix, CET1 levels stayed around 6 per cent. In a footnote Byres remarked that "no bank reported a CET1 ratio at its lowest point that was below 5 per cent". This is the benefit of APRA adopting its new "unquestionably strong" capital benchmarks, which has forced the big four to raise more than $50 billion in equity that on a globally harmonised basis clearly ranks them among the best capitalised banks in the world. More equity and less leverage reduces the probability of default, albeit with the cost of lower returns on equity. But APRA's mission is to protect depositors, not shareholders.


Comments

Please sign in to comment on this wire.