Raising bank capital requirements

I’ve been a little slow to get round to much comment on the Reserve Bank Governor’s proposals to require banks to fund a much larger share of their balance sheets with equity capital.  These aren’t small changes: from a starting point where they more or less accept that New Zealand bank capital ratios are already higher than those in most advanced countries, they want to compel banks incorporated here to adopt capital ratios that would be higher than (almost?) anywhere else in the world (including, notably, higher than in Australia).

I had a couple of initial posts (here and here) immediately after the consultative document was released, and one a couple of weeks ago on the way in which the Bank has continued to discount the results of demanding stress tests it has required banks to undertake.     Since I wrote that post the Reserve Bank has finally released (in response to OIA requests from me and others) various background papersOne of them specifically addressed the issue of the stress tests.  Suffice to say, whatever the validity of some of the specific points the authors identified, it did not materially my view of an awkwardness that the Bank still has to contend with; repeated very demanding stress tests suggests a very robust system, and yet the Bank proposes to go out on a limb with its capital requirements for banks incorporated here.  Various other people have written comments on the Reserve Bank’s proposals, and I’d commend to your attention most of a column by my former Reserve Bank colleague Geof Mortlock, and some observations from Roger Partridge of the New Zealand Initiative on the striking absence from the consultative document of any serious attempt at a cost-benefit analysis.

It is striking just how little substantive supporting analysis, and robust testing of alternatives, there is in the consultative document.  And although they have finally released various background documents, some of which are interesting, as those papers stretch back over almost three years through a couple of changes of Governor, it is impossible to know what bits the current Governor is or is not drawing from in reaching his proposals.  There has been no substantive speeches from, or searching interviews with, the Governor or his deputy since the document went out.

Of course, the Bank has a long history of not taking consultation very seriously –  they’d jump through the formal required hoops, and perhaps even amend the odd working detail here or there, but there was never much of a sense of being genuinely open to alternative perspectives.   On this particular proposal, so much money is involved that one can only hope that they are eventually forced to do better.  Perhaps there really as a compelling case for going out on a limb as they propose, but if so that case has not yet been made.  On the specific issue –  imposing more-demanding capital requirements than almost any of our peers (eg small open floating-exchange rate countries, with moderate-sized locally-focused banking systems, and government finances in good order) – nothing in the consultative document even makes the attempt.  Apparently we are just supposed to take the Governor’s word for it that it is all fine; that he knows what is best for us.

There is a range of areas where the analytical support is very weak.  I might come back to some of them. But today I’m going to start at the end and highlight one absence that I found particularly glaring.   Having reached the conclusion that a 16 per cent capital requirement would be appropriate for the large locally-incorporated banks, there is almost no discussion at all as to how banks and the financial system more generally might respond, whether in the protracted transition or in the longer-run.   Which might be okay in some quasi-academic document (of the sort they cite from overseas, trying to estimate “optimal” capital ratios), but shouldn’t be remotely acceptable when a regulator is consulting on far-reaching proposals, directly affecting private businesses (and potentially firms and households more generally), that it wishes to put the force of law behind.   We should expect to see that regulator lay out its workings, and tell us how it thinks things will unfold, in the specifics of the New Zealand economy and financial system.  Only then can we really unpick and challenge their reasoning and assumptions (implicit or otherwise).

The bottom-line appears to be that they think bank lending margins will rise to some extent (they report  –  reasonably enough –  not believing that Modigliani-Miller offset effects hold in full, at least in a potentially protracted transition) but that otherwise banks’ shareholders will happily divert most of their profits over the next few years into reinvesting in the business.  But they show no sign of having tested those assumptions, let alone of having thought more widely.

Why, for example, would the Reserve Bank not suppose that existing locally incorporated banks would respond to large increases in capital requirements in part by pulling back on their willingness to lend?  There are two ways to increase actual capital ratios: increase capital or reduce the volume (at least the growth rate) of assets relative to your prior plans.  It is not as if no banks anywhere have ever responded to increased capital ratio expectations (market or regulatory) by looking to reduce risk-weighted assets (it was a common in the EU after the last crisis).  Whether or not, in the long run, Modigliani-Miller effects really hold nearly in full, shareholders don’t tend to believe they do,  and if the New Zealand authorities insist on higher capital requirements, lending into the New Zealand market is going to look less attractive to shareholders in banks subject to those requirements.  Perhaps the Governor thinks there is already too much credit in the New Zealand economy and wouldn’t be unhappy if it becomes harder to get.  But none of that is discussed or analysed in the discussion document.

Similarly, there is no discussion at all of the fact that we are now eight or nine years into a growth phase, and that most probably there will be another recession along at some stage in the next five years (the planned transition period).  Now, of course, in truth the best time to increase capital requirements was always five years ago, but it doesn’t change the fact that  –  with capital ratios already comfortably high by international standards –  the specific proposal is to increase them over the next five.  In recessions borrowers tend to become more reluctant to borrow, and (typically quite rationally) lenders become more reluctant to lend, but if we go into a new recession in the next few years with banks still facing several years of increased capital requirements, isn’t it quite plausible that banks would become even more reluctant to lend than usual, exacerbating –  or dragging out –  the downturn.  And recall that the Reserve Bank has nowhere near as much monetary policy leeway in the next downturn as it did in previous ones.  That might reasonably make one more nervous than otherwise about ramping capital requirements over the next few years even if one could make a decent case in the abstract.  But none of this –  none –  is discussed, not even to allay concerns, in the consultative document.

There is also no discussion or analysis of the way in which the proposals will affect Australian (and other foreign) locally-incorporated banks differently than New Zealand owned banks.   Dividend imputation means that debt and equity are taxed similarly for New Zealand owned companies.  But dividend imputation doesn’t hold for foreign-owned companies, and substantial increases in capital requirements will impose a direct additional cost on the shareholders in foreign-owned banks (in this case, the Australian parents).   In other words, the Governor’s proposals skew the playing field away from the Australian banks (large and more diversified) in favour of the domestic banks (small, undiversified, and typically capital-constrained).  The Governor is known to have some animus towards the Australian banks –  and on that score he might even have political sympathisers (at least from rabble-rousers like Shane Jones) –  but none of this is identified or discussed in the consultative document.   I’d argue that we are generally better off having big banks that are part of offshore banking groups. Perhaps the Governor disagrees, but if so the onus should be on him to make his case, not to attempt to slip through regulatory measures that consciously disadvantage the foreign-owned banks incorporated here.  It is far from clear that boosting the competitive position of a 100% state-owned bank (Kiwibank) is any sort of longer-term gain to financial system soundness or efficiency.

There was also no discussion in the consultative document on the potential lenders who will not be subject to the Governor’s proposed capital requirements.  For example, those requirements will not apply to non-bank deposit-taking lenders (let alone any lenders who aren’t deposit-takers at all).  NBDT capital requirements aren’t something the Reserve Bank can set itself (same goes for LVR restrictions, which is why LVR restrictions were never applied to the NBDTs) and so it would appear that one set of regulated intermediaries will have materially lower capital requirements than another set.  All else equal, mightn’t we expect to see at least some disintermediation to these lenders?   At the other end of the spectrum, banks that aren’t locally incorporated are also not subject to Reserve Bank capital requirements.  If the Governor proceeds with his proposals, wouldn’t we expect to see more lending gravitate towards these lenders (subject to the local incorporation threshold limits), and at the “big end of town” to banks that aren’t registered in New Zealand at all.   The SME down the street might need to get credit from a local lender, but Fonterra or Air New Zealand don’t.   These effects wouldn’t arise if authorities around the world were all raising capital requirements to a similar extent, but they aren’t.   Not even APRA.  And yet there is no sign the Reserve Bank has thought hard about the potential disintermediation issues.

And even if all banks (broadly defined) globally were subject to much the same capital requirements, one might have to think about disintermediation to non-institutional providers of credit.  If capital requirements on a portfolio of mortgage loans are increased markedly and there is incipient pressure for bank lending margins to rise, it opens the door for more securitisation options.   But, again, none of this is touched on in the consultative document.

There have been suggestions, notably from at least one investment bank/broking firm, that the Reserve Bank’s capital proposals could increase New Zealand lending interest rates by as much as 100 basis points.   That, frankly, seems unlikely to me, not because banks (especially the Australian banks) won’t pull back on the supply of credit or because bank shareholders won’t look to maintain high targets ROEs for a while, but because of the potential disintermediation from the existing big-4 institutions.   Those alternatives limit the extent to which lending margins could increase (while the potential for offsetting OCR adjustments limits the extent to which retail lending rates themselves would rise).     On my story, there is a cost to the efficiency of the financial system –  a key constraint the Bank is required by law to take account of.   The core big institutions –  already very robust –  might be made a bit stronger, but in the process they would be of diminished importance in the financial system.  Any gains from stronger core institutions could be compromised by the greater proportion of credit flowing through other channels.  (A topic for another day is that entire proposal is built around the idea that the costs policy should try to mitigate relate narrowly to the failure of regulated institutions, rather than to the gross misallocation of credit –  and investment resources – in the good times, whether or not any core institutions fail when the reckoning occurs.   But there is no sign the Reserve Bank has considered whether the allocation of credit, and lending standards, will be maintained if there were to be significant disintermediation.)

I could go on, but won’t for now.  My point wasn’t to attempt to isolate every possible channel, but to note that in their consultative document –  where they are judge and jury in their own case –  the Reserve Bank has provided no analysis of any of these issues, not even to set our minds at rest.   Just nothing.  We should be able to expect better.