The Financial Stability Report: falling short again

The Reserve Bank yesterday released its six-monthly Financial Stability Report.   With (fortunately) no new direct controls to announce, the latest FSR didn’t get a great deal of coverage, and I won’t be writing about it at great length either.  But there were a few points that I thought it was worth making.

First, I thought it was a little surprising that the Bank did not have a bit more discussion of euro (and EU) break-up risk.  It has been a slow-burn process (materially slower than pessimists like me had expected) but none of the underlying stresses seem to be going away.  This weekend alone we have the Italian referendum, which could see the fall of the mainstream Italian government, and the rerun of the Austrian presidential election.  In the first half of next year, there are elections in France and the Netherlands, and in both countries anti-euro parties are polling very strongly.  The Reserve Bank does note problems in the banking systems of various European countries, but these are symptoms of some of the underlying problems not the source of the main risk.  As with all tail-risks, it is impossible to get the timing right, but to me these euro/EU risks remain by far the largest visible disruptive threat to the world economy and world markets over the next decade (China, by contrast, is more closed to the world, and has a strong (grossly over-strong) central government).  Perhaps the Reserve Bank feels uncomfortable writing about these euro/EU risks –  it might make for awkward conversations for the Governor the next time he goes to BIS meetings –  but these reports are for New Zealand citizens, not for the international fraternity of central bankers.  If it is too awkward to reflect on real and substantial risks, one has to ask how much value there is in FSRs more generally.  Conventional, and anodyne, references to Brexit and/or the US election don’t really cut it.

Second, it was a bit disappointing that even in a financial system piece like the FSR, the Bank still couldn’t help itself, and insists on running the line that

“New Zealand’s economy is strong relative to other advanced economies, growing 2.8 per cent in the year to 2016”.

And our population grew by 2.1 per cent in that year.  Per capita GDP growth is weak, and has been over the last few years taken together. I illustrated a while ago that relative to other advanced countries, growth in real GDP per capita since 2013 has been around the median.  And there has been no labour productivity growth here at all.

At one level when the Bank writes misleading things like this it doesn’t matter very much.  But at another level it does. Society has delegated an enormous amount of power to the Governor, including the resources to produce major reports like the FSR. We should expect them to avoid rose-tinted political perspectives.  If they don’t on small issues like this, how can we have much confidence in their willingness to accurately represent things in conditions of much greater stress or risk?

In the FSR the Bank puts quite a bit of weight on the increased use banks have been making of offshore wholesale funding.  I was a bit puzzled by this increased use, since there has been no sign of the current account deficit widening notably (and bank offshore borrowing tends to grow most notably when the current account deficit is large –  it was the main channel through which the larger deficits were financed).  I was interested in this chart, drawing on data from the Reserve Bank website.  It shows annual growth in Private Sector Credit to resident, and a broad measure of the money supply (M3, from residents), both adjusted for repo transactions.

psc and m3.png

The deposit side of the financial system balance sheet is growing at around the same rate as the credit side, hardly suggestive of any particular problems.  If anything, it looks as though banks’ increased reliance on foreign wholesale funding markets isn’t reflecting some deteriorating domestic macroeconomic imbalances, so much as the specification of the Reserve Bank’s core funding requirement (CFR) which, somewhat arbitrarily, distinguishes between some types of domestic deposits and others.  I don’t have a strong view on how the CFR is set up, and I believe some funding restrictions are necessary –  to internalise what would otherwise be a willingness of banks to simply rely on the central bank in times of stress.  But no such rule is perfect, and it would be good if the Bank were a bit more explicit about where the pressure for more (more expensive) foreign wholesale funding seems to be coming from.

One of my longstanding criticisms of FSRs over the years (including when I sat on the editorial committee reviewing them internally) is the weak treatment of the efficiency side of the Bank’s statutory responsibilities.  The Bank is required to use its statutory bank regulation powers to “promote the maintenance of a sound and efficient financial system”.  There is an almost inevitable tension between those two strands –  something I believe that Parliament recognised in phrasing things the way it did –  and the Act also requires the Bank to write FSRs in ways that allow the conduct of policy to be evaluated against the statutory criteria.

In yesterday’s FSR there is one paragraph (and one table) on the efficiency side of the Bank’s responsibility, under the somewhat ambiguous heading “Some indicators suggest New Zealand’s banking system is operating efficiently”.  Am I meant to take from that that other indicators, not reported, suggest otherwise?  I’m also a bit genuinely puzzled why the Reserve Bank considers that a low ratio of NPLs is an indicator of system efficiency.  In credit booms there are typically very few NPLs –  they come after the boom has bust.  In very risk averse systems there are typically very few NPLs.  In isolation, it simply doesn’t look like an efficiency indicator.      And the Bank simply does not address the inevitable adverse efficiency implictions of its increasing reliance on direct controls on specific classes of lending by specific types of institutions.   There has never been a proper cost-benefit analysis on the repeated waves of new controls.  Perhaps it is hard to do one formally, but the FSR should be an opportunity for some more considered analysis exploring some of these issues. As it is, it risks veering towards being a propaganda sheet for the Governor’s chosen policies.  Perhaps it is hard for it to be otherwise, especially under the current governance model –  the Governor signing off on a report on his own policy –  which simply highlights against the desirability of structural governance reform.

The FSR reports no reason to doubt the ability of the insurance sector to cope with the recent severe earthquake.  Perhaps the event was too recent to allow them to deal with the issue in any greater depth, but for the next FSR it might be worth posing the question as to what might threaten the ability of New Zealand insurers to keep getting reinsurance in large volumes for earthquake risk.    The NOAA database suggests that of the 33 earthquakes of 6.5 magnitude or greater to have hit New Zealand since 1840, 10 –  or almost a third –  have hit in the last 10 years (and the very destructive February 2011 aftershock was less than 6.5).    Is there a risk that a continuation of that sort of higher quake frequency could serously impair the ability to insure earthquake risk in New Zealand?  As we know that the Reserve Bank solvency standards for insurers were not set to cope with a repetition of the Christchurch quakes, what sort of tail risk concerns might this raise –   not just for the insurers, at least for the wider (currently) insured population?

The Reserve Bank has underway a review of the capital requirements for banks. on which they expect to consult next year.  As they note at present

The Reserve Bank will also look at international norms when considering calibration of New Zealand’s capital requirements. In raw international comparisons, the current capital ratio of New Zealand banks is relatively low. But a simple comparison is potentially misleading as the risk weights that New Zealand banks must apply to certain asset classes are set conservatively. This means New Zealand banks’ capital ratios appear lower than foreign banks’ ratios for the same underlying portfolios.

After adjusting for the conservative approach New Zealand takes, the Reserve Bank’s preliminary assessment is that New Zealand banks’ riskweighted capital ratios have been near or above international norms.

Looking ahead  (emphasis added)

As part of the review, the Reserve Bank will consider the appropriate amount and quality of capital New Zealand banks should be required to maintain. This will involve a survey of recent academic and central bank studies on optimal capital ratios. Care will be taken when interpreting these studies as they tend to be sensitive to assumptions about the effect of capital on banks’ funding costs and the scale of GDP losses that are directly attributable to banking crises.  It is also unclear how the inclusion of different types of capital affects the results of these studies.

I found those comments encouraging.  While I would hope that the Bank always interprets literature carefully, there are particular issues in this area.

Thus, those arguing for much higher bank capital ratios often suggest (for example) that all or most of the underperformance in economies since 2008 can be ascribed to the banking crises. If one assumes that, it is worth society spending almost any amount of money to avoid a repeat.  A more plausible story is that the banking crises themselves explain only a rather small amount of a structural global slowdown in productivity growth (that was already underway well before the crisis).  On other hand, those opposed to higher capital requirements often argue that having to fund a larger proportion of loans from equity would materially increase the cost of funding.  That approach tends to ignore both the potential for the Reserve Bank to cut the OCR, to offset any incipient rise in interest rates, and –  more importantly –  ignores the extent to which a higher reliance on equity reduces the riskiness of the institutions, and should reduce the required rate of return on equity.  Tax systems can complicate that story, but the New Zealand (and Australian) tax systems, with dividend imputation, tend to treat debt and equity more or less equally.

My own, provisional, view is that for banks operating in New Zealand somewhat higher capital requirements would probably be beneficial, and that there would be few or no welfare costs involved in imposing such a standard.  My focus is not on avoiding the possible wider economic costs of banking crises (which I think are typically modest –  if there are major issues, they are about the misallocation of capital in booms), but on minimising the expected fiscal cost of government bailouts.  As I’ve explained previously, I do not think the OBR tool is a credible or time-consistent policy.

Many of these have been relatively small points, at least for now.

I am more uneasy about two aspects of the report, one of which is missing completely.

There is simply no discussion at all –  and has been none in past reports –  of what the implications of drifting ever closer to the near-zero lower bound on nominal interest rates might be for financial system resilience in a further severe recession.  One of the great merits of a floating exchange rate system has been the flexibility it provides to national central banks to adjust policy interest rates, more or less without limit, as domestic conditions require. But the limits are now becoming increasingly visible.  It is remiss of the Bank – both in its MPS, its FSR, and its more corporate documents such as its Statement of Intent, or governors’ speeches – not to even begin to address these issues openly.  Simply ignoring them doesn’t make them go away, and there will no acceptable excuse if we find ourselves in the same position many other advanced countries found themselves in after 2008, having been given a decade’s sharp-focus notice of the potential problem.

And my final source of unease is around the Reserve Bank’s analysis of the housing market.  The Bank puts a great deal of importance on developments in the housing market, and the market for housing credit.  It is has imposed successive rafts of direct controls, of the sort never seen before in post-liberalisation New Zealand, all on the basis of little or no research.  And in the latest FSR, there is no fresh analysis, just the looming threat of debt to income restrictions –  if the market takes off again –  so long as the Governor can persuade the Minister of Finance to give him political cover (and of course he will, or otherwise the Opposition will attack the Minister for standing in the way of the Reserve Bank).

I’ve noted in the past that we’ve seen no analysis from the Bank on, for examples, countries that have had large increases in house prices and where there has been no subsequent financial crisis, or even collapse in house prices.  New Zealand, Australia, the United Kingdom and Canada in the 2000s are just some of the examples that spring to mind.  And although the Bank keeps talking about the problems in housing supply, they show no sign of having really thought deeply, or researched, the role of land market restrictions.  The New Zealand housing market is unlikely to be sorted out by simply building enough houses to keep the occupancy rate stable – or even a few more.  There is a much more structural issue around restrictions on land use (accentuated by the heavy regulatory costs imposed on building new houses).  I’m not aware of countries with floating exchange rates and tight land use restrictions where urban house and land prices have sustainably come down again.  Perhaps the Bank is, but if so surely the onus is on them to disseminate the research, rather than continuing to hand-wave and treat all house price increases as more or less equally risky.  When house prices are bid up on the back of congressional and executive pressure on lenders to lower lending standards, on pain of potentially losing favour with the regulators, that is a very different issue than when the combination of population pressures and land use restrictions drive prices up. But the Reserve Bank has never articulated that sort of distinction.

It is more than three years now since the Reserve Bank set off down the path of increasing direct controls (with no end in sight, let alone the prospect of removing the controls).  I went through the list of their Analytical Notes, Discussion Papers,  and  Bulletins over the last three years or so, and was struck by how little housing-market research they have published.  For an organisation with so much discretionary power, and a substantial publicly-funded research capability, it really isn’t good enough.

I could repeat a variety of points from previous commentaries –  including questions about what other risks banks are assuming if they are prevented from taking on as many high LVR housing loans as they would like –  but will leave it at that for now.