There have been three New Zealand recessions since inflation targeting was introduced. That, in turn, wasn’t long after interest rates were liberalised and the exchange rate was floated in 1984/85. Of those recessions, two were severe and one (the 1997/98 episode) was more moderate. Here is how 90 day bank bill interest rates – the benchmark indicator before the OCR was introduced – fell (and were allowed to fall by monetary policy) in those episodes.
Mid-late 1990 to mid-late 1992 780 basis points
Mid-1997 to end 1998 340 basis points
Late 2007 to mid 2009 590 basis points (OCR cut 575 basis points)
Over the first of those periods, medium-term inflation expectations fell by about 2 percentage points (from about 4.3 to 2.3 per cent) – that was in the midst of the major drive to lower the inflation rate. In the other two episodes there was no material change in surveyed inflation expectations. So in the two severe recessions, short-term real interest rates fell (or were cut) by about 580 basis points, and in the less serious recession they fell by 340 basis ponts.
At present, the OCR is 1.75 per cent (and 90 day bank bill rates are about 1.9 per cent). As things stand today (current laws, rules, and central bank practices), no one is confident that the OCR could be cut further than around -0.75 per cent. Below that, it seems likely that it would become economic for large scale moves into physical cash (which earns zero less storage and insurance costs) to occur – mostly not by households, but by market participants with large holdings of New Zealand dollars. To the extent such shifts happen, market interest rates wouldn’t fall much even if the OCR was cut further. That would dramatically undermine the effectiveness of conventional monetary policy (which works mostly either through direct interest rate effects, or through the influence of interest rates on the exchange rate).
There isn’t anything very controversial about that story. At the Reserve Bank it was a conclusion we got to in a project I led back in about 2012 when the euro-crisis seemed to foreshadow risks of new externally-sourced crisis/recession. It is consistent with the revealed practices of other central banks (no one has cut below -0.75 per cent), and is just pretty standard analysis.
So if the next recession hit today, the Reserve Bank could count on having around 250 basis points of policy adjustment capacity (the OCR could be cut that much). But it has needed more than that in each of the (small sample of) recessions in recent decades.
And it isn’t that the New Zealand numbers are unusual. In the US recessions going back to the 1960s, the median cut in the (nominal) Fed funds rate has been just over 500 basis points.
If the OCR can’t be cut as much as normal, monetary policy cannot do its job. We have active discretionary monetary policy to minimise the output and employment losses in downturns (adverse economic shocks come along every so often, like it or not). And if markets, and businesses, know that monetary policy is thus hamstrung, they will factor that into their expectations and the actual downturn will probably end up even worse (the monetary policy cavalry aren’t expected to ride to the rescue).
And so, every so often since I started writing this blog, I’ve been highlighting the potential problem the next time a serious recession comes along, and lamenting the apparent (certainly in public) indifference of our Reserve Bank, Treasury, and Ministers of Finance to the issue. Other countries ran into the limits of conventional policy in the last recession. They couldn’t do much about it then, and paid the price in a very sluggish recovery (slow closing of output and employment gaps). But no country needed to find itself in this position if it prepared the ground well before the next recession.
In raising these concerns, I’ve been in good company. Since shortly after the last recession various prominent economists – of pretty impeccable orthodoxy – have been raising the possible need to think about an increase in inflation targets. Two of the most prominent were former IMF chief economists, Ken Rogoff and Olivier Blanchard. Their logic is simple. Inflation targets were set on the assumption (implict or explicit – in our case, we actually wrote it down at times) that the near-zero lower bound on nominal interest rates was only a theoretical curiosity, and of little or no practical relevance. Experience in various countries proved that assumption wrong. And in the medium to long term, the most reliable way to raise nominal interest rates – and thus leave room for substantial cuts in future recessions – is to raise actual and expected trend inflation. (One counter to this argument for some countries a few years ago was that since most central banks were having trouble meeting existing inflation targets, and had already exhausted conventional capacity, how could they hope to credibly target still higher inflation.)
Other economists – Miles Kimball, Willem Buiter, and more recently Ken Rogoff – have focused on the other side of the issue: can changes in currency laws or practices be put in place which would mean that nominal interest rates could be cut more deeply. As various observers have noted, some estimates of a conventional Taylor rule suggest that ideally the Fed funds rate would have been cut – for a short time perhaps – as low as -3 or even -5 per cent at the height of the 2008/09 recession.
In practice, nothing much has changed yet. No one has changed their inflation target – or adopted, say, price level of nominal GDP level targeting which some (including the head of the San Francisco Fed) believe could provide more resilence – or taken steps to deal directly with the practical lower bound. We are drifting towards the next severe recession, with the toolkit severely depleted.
In New Zealand, it has been harder than in most places to get any serious debate going at all. I suspect a variety of factors contributes to that, including:
- the fact that we didn’t get particularly close to the near-zero lower bound ourselves in the last recession,
- the persistent belief that before long interest rates will again be much higher,
- the belief that New Zealand has lots of fiscal headroom such that even if monetary policy is constrained in some future recession it won’t matter,
- given the perpetual discontent in some quarters in New Zealand around monetary policy (it has been an election issue for some or other party every election since 1990) a desire, among the orthodox, not to be seen as “giving aid and comfort to the enemy”.
There is also a bit of handwaving around the possibilities of QE (direct asset purchases), but this is mostly handwaving and attempting to play distraction as no one in other countries – that have actually used QE – believe it is an effective substitute, on feasible scales, for the conventional interest rate instruments.
I would not, myself, regard a higher inflation target as any sort of first-best option. Indeed, in an ideal world, I’d be more comfortable with a regime that delivered average inflation near-zero over time (allowing for the modest measurement biases in the CPI). Inflation has costs, although economists have struggled to find convincing estimates of large adverse effects at relatively low inflation rates. And many of the costs that do exist arise from the fact that the tax system is designed for a zero inflation rate. Inflation-indexing the tax system (mostly around the treatment of interest and depreciation) could tackle that issue quite directly (and there are official reports from decades ago, here and abroad) identifying how it could be done. (It would treat savers, and borrowers, more fairly, even with a 2 per cent inflation target: perhaps I point I might make in my submission to the Tax Working Group.)
But if there are modest – largely avoidable – costs of a slightly higher inflation target, they quite quickly pale in comparison with the output and welfare losses if monetary policy isn’t able to operate as effectively in leaning against recessions. Drifting towards imposing that cost on ordinary New Zealanders – and it won’t be the Treasury or Reserve Bank officials who face those cyclical costs – should be pretty inexcusable.
And so I was encouraged when, the other day, Radio New Zealand’s Patrick O’Meara rang up. He’d been reading my post which had noted, in passing, the IGM survey of US economists in which 86 per cent of those senior US academics agreed that
Raising the inflation target to 4% would make it possible for the Fed to lower rates by a greater amount in a future recession.
and was interested in how one might think about these issues, and do something about them (immediately and in the medium-term), in a New Zealand context. We talked at some length, and then he talked to some other people, and the result was this story.
I was quoted this way
While uneasy about higher inflation, economic commentator and former Reserve Bank official Michael Reddell said increasing the Reserve Bank’s inflation target band of one to three percent was worth discussing.
“It’s really important to start planning, having the discussions about how we’re going to cope with the next downturn.”
“The next recession could be relatively minor,” he said.
“We could just get away with needing to cut the OCR by 50 or 100 basis points [0.5 to one percent].”
“But the typical severe recession, whether it’s in the US or New Zealand needs, typically, [400-600] basis points of interest rate cuts, and we’re just not positioned for that,” Mr Reddell said.
What surprised me was the comments from others that O’Meara talked to – not so much the conclusion, but the argumentation. In fairness to the individuals, I don’t know exactly what they were asked, or how much of their response was used, so what follows is a response to the Radio New Zealand reporting.
There was Kirk Hope, head of Business New Zealand, who (frankly) seemed an odd person to ask about details of macroeconomic policy and contigency planning for future recessions.
Business would also suffer, Business New Zealand chief executive Kirk Hope said.
“If the target is too high and there’s too much inflation and interest rates are too high, then it reduces investment in the economy … and that in the end costs jobs.”
As quoted, those proposition are simply wrong. They don’t, for example, distinguish – quite importantly – between real and nominal interest rates. All else equal, higher real interest rates might reduce investment. Unchanged real interest rates – actually perhaps a bit lower in a transition period – are unlikely to affect investment or jobs adversely.
And there is no sign that Hope had even engaged with the “how do we handle the next severe recession” – when investment and jobs really will be adversely affected if monetary policy is hamstrung – question. Perhaps it wasn’t put to him?
The other person asked specifically was Arthur Grimes, these days a researcher on all manner of interesting things, but formerly a senior manager at the Reserve Bank (and, for a time, chair of the Reserve Bank Board). Arthur has long been a vocal, and articulate, defender of the status quo.
Here was what RNZ reported of his views
Victoria University School of Government professor Arthur Grimes was adamant New Zealand’s existing inflation target band was more than adequate.
Professor Grimes said raising inflation to four percent, for example, would do nothing but hit households in the wallet.
“Why would we want the cost of living to be rising any faster than that? Don’t forget the cost of living makes it harder for people to live. It’s wonderful that inflation expectations are low and that inflation is pretty low.”
His response was more surprising, given his background. Not that he didn’t favour raising the target (I don’t either) but the quoted thrust of his argumentation. Higher (expected) inflation – and inflation expectations are at the heart of the story about nominal interest rates – don’t “hit households in the wallet”; they see both wages and prices (and welfare benefits) rising a bit faster than otherwise. If there are real adverse costs of higher inflation they come from things like the tax system effects (see above). The confusion of reals and nominals in these quoted remarks seems pretty extraordinary. If wages and price are both rising at 2 per cent per annum, and then subsequently – well foreshadowed – they are both rising at 4 per cent, there simply isn’t a “cost of living” problem which “makes it harder for people to live”.
And again – and perhaps he wasn’t asked – there is simply no engagement with the case that people like Blanchard and Rogoff used in raising the option of a higher inflation target: how do we cope with the next severe recession when the OCR can’t be cut as much as we are used to?
My own position remains much as I outlined it the other day. The first priority needs to be some serious engagement on the issue, and recognition, of the likely threat – the constraint on the ability of monetary policy to do the job we’ve asked of it since the end of the Great Depression. In my view, the second priority should be serious work on removing or greatly attenuating the near-zero lower bound, by taking steps (and being open about them) to limit the scope for large scale conversions to cash. Far better to deal with the issues, and risks, now, than to attempt to grapple with them in the middle of the next serious recession (especially given recognition lags).
And for the immediate future, in the context of the PTA that has to be agreed in the next few days, and any associated letter of expectation to the new Governor, as I suggested the other day
In conducting monetary policy, and without derogating from its obligation to act to keep inflation within the target, the Reserve Bank should be required to have regard to the desirability of there being as much effective policy capacity (or at least rather more than at present) as possible to respond to the next serious recession, and
consistent with that, the Minister could indicate that he would be more comfortable if core inflation over the next few years fluctuated in, say, the 2.0 to 2.5 per cent part of the target range, than if core inflation continued to fluctuate around 1.4 per cent as it has now for a number of years.
Raising the inflation target itself should only be a fallback option. I deliberately don’t use the words “last resort” – that way nothing will happen until well into the next severe recession when it will be too late – but if, after careful and open considerations, officials come to the conclusion that whatever can feasibly be done around easing or removing the near-zero lower bound won’t produce (with certainty) the desirable degree of policy flexibility, than we should be seriously considering a higher target. It might not be ideal, but we don’t live in a first-best world. If one of the key assumptions that underpinned the current targets has been invalidated, and can’t be dealt with directly, the target would need to be revisited.
And finally, as much to anticipate commenters as anything, a couple of quick thoughts on the exchange rate and fiscal policy:
- as I’ve pointed out here more than once, in such an adverse scenario – with our OCR at the floor – the exchange rate is likely to be very much lower. Which is consoling, but unlikely of itself to be adequate. After all, in typical New Zealand recessions we have both large OCR cuts and large falls in the exchange rate,
- our net public debt is quite low, and clearly there is more room for fiscal expansion than in many countries. Nonetheless, experience suggests that that room will prove smaller than it might appear (not so much technically but politically). And since few other advanced countries will regard themselves as having much fiscal room, the advanced world as a whole will be short of offsetting stimulus. Moreover, typically monetary policy can be deployed much more quickly than fiscal policy, suggesting that at best fiscal headroom is a poor substitute for fixing the monetary policy constraints.
And for anyone interested in another analysis of the option of raising the inflation target, this recent piece from the Bruegel thinktank in Europe, emphasising the importance of robustness, covers some of the ground quite nicely.