Having taken their long summer break – not heard from since 11 November – the Reserve Bank’s Monetary Policy Committee will be out with their Monetary Policy Statement on Wednesday. Much has changed in the economic data and indicators, here and abroad, since then, and it will be interesting to see what the Governor and has committee have made of it all. There are some genuine surprises and puzzles that the Committee should have been grappling with – and most other macro economists and commentators too, but the rest of us don’t get to set monetary policy. And if the strength of the economic rebound is a surprise – and it would appear to have been to the Bank too – how resilient is that rebound likely to prove, and under what conditions?
I’m somewhat sceptical of the idea of a resilient rebound this year – and with more than a few questions/puzzles about quite which data we can really count on at present – but without a compelling explanation for last year, one has to be even more hesitant than usual about backing a view about the future (macro forecasting is mostly a mug’s game anyway).
My own approach to monetary policy would probably be the one – “least regrets” – the Bank has repeatedly articulated over the last couple of years, if rarely followed in practice. That is especially so because the last year has made me more sceptical than I was about attempting to use fiscal policy for macro stabilisation (as distinct, say, from income relief amid a lockdown). Interest rates are the prices that balance savings and investment intentions, and monetary policy is about allowing interest rates to do that job.
And so even if the level of economic activity – even in per capita terms – is now back to something like it was at the start of last year, we still have
- a central bank that has done nothing to reduce the (true) effective floor on the nominal OCR (even if they have very belatedly ensured that banks can cope with modestly negative rate),
- core inflation that is still (a little) below the midpoint of the target range, not having been at or above that midpoint for the best part of a decade),
- inflation expectations (surveys and market prices) that are still typically below the target midpoint, often by quite a long way (and this is so even though there has been quite a – welcome – lift in recent months),
- the unemployment rate (inevitably measured less precisely than usual) is still non-trivially above reasonable guesses at where a NAIRU might be,
- most other countries’ economies are doing less well cyclically than New Zealand’s and if vaccination programmes are well underway in a few of them, anything like normality still seems quite a way away,
- there is a great deal of uncertainty (inescapable, unavoidable) about the environment in which firms and households will be operating, and uncertainty tends not to encourage either consumption or investment spending, and
- if the US is having another fiscal splurge, more generally across advanced countries the pressure in the next year or two is likely to be towards fiscal consolidation – not necessarily dramatically so, but certainly in contrast to last year. There isn’t much sign New Zealand will be any exception to that (nor, in my view, should it).
And then there is the wider backdrop. Even if we recover from this unusual Covid recession more readily than many had expected, the issue that has increasingly dogged monetary policy over the last decade has not gone away: nominal policy interest rates in more and more countries (now including former high interest rate countries New Zealand and Australia) are now near zero leaving rather limited monetary policy capacity when the next serious recession – grounded in economic developments not infection ones – comes along. That might be 10 years away, but it might be only a handful. The best thing monetary policy can do to help ensure there is some policy leeway next time is to err strongly on the easing side at present, generating inflation (and inflation expectation) outcomes that are – for a change – in the upper part of the target range. The Bank could articulate something like the Fed’s average inflation targeting approach – or, since the Minister is the one supposed to set the target, the Minister could tell them to – but a decent start would be to start acting as if they would be totally comfortable if, by chance, core inflation averaged say 2.3-2.5 per cent over the next five years. That doesn’t require actively targeting such numbers, but it does require recognising that central banks (including our own) have consistently over-forecast inflation over the last decade, and still don’t adequately understand why they’ve made that mistake. So by being actively willing to embrace higher inflation outcomes, perhaps the Bank and the MPC might just give themselves a better chance of delivering outcomes around 2 per cent – what successive ministers of finance have asked them to do.
If it were me, then, I would still be cutting the OCR, perhaps to zero this time. It would add a bit more macroeconomic stimulus, and would also be more realistic – since we don’t know the future – than idle pledges to keep the OCR where it is for some arbitrary length of time (recall that their last, hawkish as it turned out, arbitrary commitment only expires next month). And I would continue to express a willingness to take the OCR negative – and not a grudging willingness, but a genuine “do what it takes” approach to getting the economy back to full employment and inflation back to target.
And what of the Large Asset Purchase programme? If it were me, I would discontinue it now. That isn’t inconsistent with my macro stance (see above) because as regular readers know I’ve long been of the view that the LSAP was not making much macro difference at all (even if it may, at the margin, have helped a little in stabilising bond markets in the couple of weeks of global flurry last March), while it continues to (a) act as distraction (enabling the Bank to look and sound as if it is doing more than it is, and (b) has led some people to believe that somehow monetary policy, notably the LSAP programme, is greatly exacerbating that unnatural disaster of the rigged New Zealand housing market. Scrap the LSAP and nothing of substance will change around the housing market – access to finance, access to (use) land, supply of finance, demand, or even the shorter-term interest rates that are relevant to most mortgage borrowers. (And, of course, more generally the unnatural disaster has almost nothing to do with monetary policy – and even for those who want to “blame” interest rates, bear in mind that very long-term market rates, that central banks have little direct hold over most of the time, have been falling for decades.)
Now I don’t for a moment suppose that the Bank will do anything of that sort, on any of what I’m suggesting about monetary policy. But I hope they do give us some sort of serious framework outlining the sorts of specific factors that might eventually lead them to discontinue the LSAP. It is, for example, hard to see how they could justify continuing it if (a) they now believe banks can adequately cope with negative interest rates, and (b) if they get to a point where they think the risks are no longer skewed to the downside.
On such things, I’ve been reading over the last week a new book by the British economist Jonathan Ashworth on the experience this century with central bank asset purchase programmes (it is 20 years next months since the Bank of Japan first launched its quantitative easing). Quantitative Easing: The Great Central Bank Experiment was published last year and clearly was completed on the very eve of Covid – a couple of 2020 references, but no mention of the Covid recessions/interventions at all. It is a really nice summary treatment and documentary record of the activities in this area of the Fed, the Bank of England, the ECB, and the Bank of Japan, up to and including the Fed’s partial withdrawal from QE, as it finally raised interest rates after 2015 and wound back the size of its balance sheet. Although the publisher – launching this new series of books on aspects of the global financial system – describes the approach of the series as “resolutely heterodox”, in fact the book is strikingly orthodox. It is, therefore, quite a nice summary of the likely way the Reserve Bank and The Treasury were seeing the possibilities, and limitations, of quantitative easing when they were advising the government at the start of last year. It is also a good single point of reference if, like me, memories of some of these programmes grow somewhat hazy over time. And for anyone wanting a good introduction it is a fairly accessible read.
The orthodox view tends to be that asset purchase programmes have had some, perhaps significant, macroeconomic benefits. The case is probably strongest in the midst of the 2008/09 crisis when both the UK and US launched such programmes (although with important differences between those programmes) although Ashworth seems to favour interpretations in which later programmes have also had useful effects. I’m more sceptical, for a variety of reasons. Much of the work in this area rests of event studies around the announcement of programmes, and so it is a shame that Ashworth does not engage with (for example) the published work of former senior St Louis Fed researcher Dan Thornton who has critically reviewed claims in that are (see, for example, this journal article, and this policy piece). Ashworth rightly highlights how wrong were the people who claimed a decade ago that the asset purchase programmes would lead to a huge upsurge of inflation (much the same claims are made in some quarters on the right about the latest asset purchase programmes) but doesn’t really probe deeply questions as to whether a large scale asset swap can really make very much sustained macro difference. He doesn’t, for example, engage with the idea that things might be different if a central bank was buying bonds yielding, say, 10 per cent, and paying zero interest on settlement cash balances (as would once have been the norm) than if the central bank is purchasing assets yielding under 1 per cent (sometimes under zero) and paying the full policy rate on the resulting settlement cash balances. And although he usefully looks at the Fed’s balance sheet wind-down pre-Covid, his conclusion that that policy choice had little or no macro impact doesn’t seem to lead him to reflect afresh on whether the earlier policy interventions really had as much sustained effect as many central bankers prefer to believe. (One of my own sceptical arguments over they ears has been that there was little sign that bond yields had fallen further relative to policy rates in countries that used the LSAP tool – say the US or UK – than they had in countries that did not – say New Zealand or Australia.)
One point Ashworth does usefully highlight – and which I hope the RB will touch on on Wednesday – is the stock vs flow distinction. If QE has an effect, it is from the transactions in the market at the time (the flow) or from the accumulated withdrawal of bonds from the market (a stock effect). He notes that the literature tends to favour the stock story. If that is correct – and if QE has much effect at all – then, for example, the Reserve Bank could discontinue the LSAP now and continue to assert that the stock of bonds they had purchased was continuing to have a material stimulatory effect.
And just in case you think that LSAP-scepticism might just be some Reddell idiosyncrasy, I can leave you with a couple of quotes, The first is from the body of the book, from Paul Krugman, quoted in 2015 observing of the unconventional monetary policy tools “the bad stuff [presumably inflation risks] unpersuasive, the good stuff maybe, but not really compelling, this has just not turned out to be the game changing policy too that people had expected”. The other quote is from the Foreword to the book by the eminent academic and former central banker, Charles Goodhart (also a former colleague of Ashworth’s). Goodhart clearly likes the book, and commends it to readers, but notes that his own view that beyond intense crisis periods – in which bond purchases can respond to liquidity and market dysfunction stresses – the direct effect on the real economy via interest rates [ and recall that Orr claims the LSAP works by affecting interest rates], either actual or expected, and on the portfolio balance, was of second-order importance. QE2, QE3 and QE Infinity are relatively toothless”.
As I’ve noted previously when you have a tool that largely involves swapping one lots of (longer-term) government liabilities for another lot of (shorter-term) government liabilities – both paying low but market interest rates – and when your swap doesn’t even displace many existing holders of the long-term assets, it is inherently unlikely that you could use such a tool to generate large or sustained macro effects. My best read of the experience to date – abroad, nicely described in the book, or at home – is that we’ve seen just what we should expect, but with lots of central bank handwaving (the need to be seen to be doing something) that has distracted people into thinking that the tool is much more powerful – for good or ill – than it actually is.