Despite concessions to some of the criticism its proposals had drawn, the Reserve Bank's decisions announced yesterday on the seismic strengthening of banks' balance sheets still look likely to do more harm than good.
The good is inherently uncertain. The harm, easier to predict, will be mainly to riskier borrowers, particularly small and medium-sized enterprises and farmers, who are liable to find credit scarcer and more expensive.
The final version of the policy will still require banks to find an additional $20 billion of capital, albeit over seven years instead of five.
And by allowing redeemable perpetual preference shares that meet certain conditions to count towards the 16 per cent of risk-weighted assets in "Tier 1" capital, which will still be required of the big four banks, the Reserve Bank reckons it has shaved about 8 basis points off the likely upward impact on lending interest rates.
The redeemable perpetual preference shares resemble equity in that they can be used to absorb losses, but until that point they look to investors like an issue of interest-bearing debt.
Although welcome, these are not radical changes from the original plan.
The cynical view would be: of course the big four Australian-owned banks will squeal like stuck pigs if forced to have more skin in the game, but they have been making out like bandits for years, secure in the knowledge that they are too big to fail and, regardless of official policy on "open bank resolution", will be bailed out by taxpayers if they are in danger of going bust.
But reading their submissions ahead of yesterday's decisions, and those of experts with no obvious vested interest, challenges that prejudice.
A consistent theme in the submissions is that the Reserve Bank has failed to make a convincing case for increasing banks' capital requirements from the internationally robust levels they already are.
More capital, in short, looks like a nice-to-have rather than a need-to-have.
The problem with any attempt to weigh benefits against costs is that the benefit side of the exercise is highly uncertain.
In a complex and fast-changing world, attempts, based on the past, to quantify the probability of events that would sideswipe the economy hard enough to trigger a banking crisis are fundamentally dubious. To be fair, the Reserve Bank has acknowledged that setting minimum capital levels is inevitably a judgment call, as the "nature, timing, scale and impact of future shocks [are] unknown".
Consultant and former Reserve Bank official Geof Mortlock points out that the Reserve Bank's own stress tests indicate the banks come nowhere near to the point of failure under very severe stress scenarios.
And Michael Reddell, another of the former senior Reserve Bank officials critical of the proposals, argues that recent financial crises in economies anything like ours — like the United States, Ireland, Iceland or the BNZ bailout in the early 1990s — did not arise from external shocks but from years of degraded lending standards.
That suggests the need for more rigorous banking supervision during boom times, and macro-prudential safeguards like loan-to-value ratio restrictions, rather than an across-the-board requirement for all banks to fund more — much more — of their lending from equity.
More equity is essentially an insurance premium, the cost of which has to be split between borrowers, depositors and bank shareholders.
The attitude of the latter is probably pretty accurately summed up by a succinct early submission the Reserve Bank received from one Mark Lyons: "Hi. I will make this brief ... Australian shareholders in banks are in no mood for this s**t."
The Reserve Bank's assumption that the Australian parents of our big four banks will be willing not only to put $20b more capital into their New Zealand subsidiaries but also to accept a significantly lower rate of return on equity, on the grounds that they are invested in a safer bank, has been greeted with scepticism bordering on derision.
In any case, Westpac makes the argument that when things get ugly, it is not only the level of a bank's capital but its ability to replenish it that are vital. "Banks with low returns tend to remain weak for long periods (many banks in Europe are good examples) and this creates a drag on their economies."
The Reserve Bank's lonely confidence that bank shareholders will accept a lower return on equity underpinned its previous expectation that the inevitable widening of interest margins — the spread between the rates banks pay depositors and other creditors on the one hand, and the rates they charge borrowers on the other — would only be "in the vicinity of 20 to 40 basis points".
It now reckons 20 basis points.
Other estimates have been for a much greater impact. ANZ thought 80 basis points would be more like it. Their initial estimate now is more like 30 to 60 basis points. UBS reckoned mortgage rates could rise by between 80 and 125 basis points and Macquarie estimated 70 to 110 basis points more on mortgage and small business loans.
In addition to upward pressure on lending rates, the new capital regime is expected to reduce the availability of credit. Massey University professor of banking David Tripe thinks the contraction in credit could run to tens of billions of dollars.
Banks can limit the amount of extra capital they need by restricting the denominator of the capital ratio — risk-weighted assets (loans).
And another potential constraint, in particular for the largest bank, ANZ, is that the Australian banking regulator APRA limits the extent to which an Australian bank is exposed to any one offshore subsidiary.
More generally, APRA would surely be unwilling to see capital drain from the parent to the subsidiary's balance sheet, to meet more stringent requirements than it considers necessary for Australian banks, especially when the expectation that the parent would stand behind the subsidiary underpins the latter's credit rating.
The Reserve Bank's decision to allow the use of preference shares in meeting capital requirements (up to 2.5 percentage points) and thereby allow New Zealand banks to attract capital from non-parental sources should reduce the risk of that becoming a binding constraint.
Already, the most recent Reserve Bank credit conditions survey of the banks in September found them expecting to reduce the availability of credit over the next six months, especially for agriculture, commercial property and to a lesser degree SME business loans.
The main factors the banks cited were regulatory changes, balance sheet constraints and risk tolerance.
Will yesterday's modest concessions change that? We will see, but probably not much.