Time to cut

Tomorrow the Governor will announce his latest OCR decision and publish the Reserve Bank’s June Monetary Policy Statement.    Having been inside the Reserve Bank for so long, I don’t claim any great expertise in reading the tea leaves as to what the Governor will actually have chosen to do when 10 days or more ago, he sat down, took advice, and made his decision.  I think he probably will not cut the OCR, but if so that will, most probably, be a mistake, compounding the series of mistaken monetary policy judgements he has made since the start of last year.  I shocked some colleagues in the middle of last year when I opined that the Reserve Bank would cut the OCR before the Federal Reserve raised US policy rates.   It was a rather speculative call at the time.    I’m still sticking to it, but I’ll be surprised if confirmation comes quite yet.

One of the things about monetary policy (at least, active discretionary forecast-based monetary policy) is that there will always be considerable uncertainty as to just what the right OCR will be.  As I’ve observed before, with that level of uncertainty there is no particular shame in being wrong, provided one learns quickly from the mistake.  The Reserve Bank doesn’t seem to have been very good at that over the last 18 months.  It should have been apparent pretty quickly that there was little or no basis for the tightening cycle that the Governor initiated last year, when he boldly foreshadowed 200 basis points of OCR increases.  But any recognition of that point has been pretty slow and grudging. Mea culpa, mea culpa, mea máxima culpa it most certainly is not.   Too bad about the people who have lingered on unemployed for longer than was necessary.  Perhaps tomorrow’s MPS will contain some retrospective self-examination of how those mistakes came to be made, and what lessons can be learned from them.  If such self-examination is not strictly required by the section of the Act governing MPSs,  it is pretty strongly encouraged.

A journalist asked me the other day if it was just embarrassment that made the Governor reluctant to reverse course.  I don’t think so.  I think is about views of the world.  All of us have them, and need to have them to make sense of things at all.   And views of what is going on in the world –  paradigms, models – don’t change quickly.  There is a tendency (again, in all of us) to discount observations that might undermine our story, and focus on data, insights, or arguments that tend to support our story.    That makes it hard to get material changes of view, and all the more so when a single decision-maker, who does not set out to foster the contest of ideas and the open-minded examination of alternative views, is in charge.    A downside of the single decision-maker model is the tendency to assign too much weight to one person’s “model”.

In the Governor’s defence, of course, outside observers are still split on what he should do tomorrow.  Many of the market economists in 2013 and 2014 were even more hawkish than the Governor, and each of them faces the same challenge of revising their view of what is going on.

I’m just going to make three other quick points:

  • I noticed the BNZ economics team out a little while ago picking that the OCR would not be cut tomorrow, but arguing that if the OCR was cut the Reserve Bank would signal a series of OCR cuts of perhaps 100 basis points or more.  I think that is very unlikely to be right.  Practically, there just is not much time between the previous OCR review and finalising the forecasts for this Monetary Policy Statement, and big changes in the whole direction of the forecast don’t seem very likely in that time.  But perhaps more important is the point about paradigms or views of the world.  It seems much more likely that the Governor would back reluctantly into any OCR cuts, treating them as little and precautionary.  He could do that within a world view in which he thought the future OCR would still have to be at least as high as it is now.  That seems much more likely than that he would –  with no foreshadowing – embrace a full-scale reversal of last year’s OCR increases.  It might happen, perhaps it even should happen, but it seems very unlikely at this stage.
  • The data.  I know some market economists have changed their view, and their OCR calls, since the last OCR review at the end of April.  But to me, the data since late April don’t seem enough for a major change of view.  For those, like me, who think the OCR should have been lower there has been nothing to contradict that view.  There is no sign of inflation pressures building, and perhaps a little more sign of the building boom tailing off, a little more sign of excess capacity in the labour market, and a little more weakness in dairy prices.   It has been enough to shift a few people across the line, but if one had been a confident believer in the sort of story the Reserve Bank was telling in its last MPS, it would have been nowhere near enough to have prompted a radical revision of one’s view.    The same goes for the world economy.  To me the data flow, especially from the emerging world, looks weak and quite worrying, but that is the read of someone inclined to a pessimistic take.  The Governor, by contrast, has repeatedly articulated a story about the strength of global growth.
  • I like the NZIER’s Shadow Board process, where a group of outsiders offer their views not on what the Reserve Bank will do, but on what it should do.  At times it has been interesting to watch divergences between what bank economists are telling us the Bank will do, and what they think it should do.    Those involved in the Shadow Board are asked to assign probabilities that a range of different possible OCRs will be appropriate, recognising the uncertainty most of us face.    I’m not sure if they are still doing it, but the NZIER innovation prompted the Reserve Bank to introduce such a system for the group advising the Governor on what he should do with the OCR.  We provided written advice (one page each) but also assigned probabilities.  It was designed to help foster debate, and to help recognise (and help the Governor recognise) that when one adviser was advocating one rate and others another rate, both recommendations were typically medians or means of a distribution of views.    The latest Shadow Board results, for tomorrow’s OCR decision, came out today.  Here is the chart of the individual respondents’ views.

shadow board

Two things strike me about the chart.  The first is how tightly bunched the distributions actually are.  It is as if all of them have a difficulty with grappling with just how uncertain we should be about what the right level of the OCR will turn out to have been.  I wonder if they are answering “what would you do if you were Governor” rather than “what OCR will prove to have been right for New Zealand given the inflation target”?  But the second observation is how the two academics in the group –  Viv Hall and Prasanna Gai –  have the tightest distributions of them all.  Viv is 85 per cent confident that 3.5 per cent is right, and Prasanna is 100 per cent confident.    Academics usually remind practitiioners of the huge margins of uncertainty in any of this stuff.  Fan charts have not been popular in New Zealand , but they do help to illustrate the historical range of uncertainties.  Presumably, given the way the PTA is written, the appropriate OCR today depends on the things that will determine the trend inflation outcomes a couple of years hence.  Who can say that they know that with any confidence?

For what it is worth, my own distribution of probabilities for the appropriate level of the OCR tomorrow is roughly as follows:

OCR                                       Probability

1.5                                          5

2                                              5

2.25                                        10

2.5                                          10

2.75                                        12.5

3.0                                          15

3.25                                        12.5

3.5                                          10

3.75                                        10

4                                              5

4.25                                        5

My distribution is wider, and much flatter, than any of those in the Shadow Board grouping, and yet it still feels too narrow to really grapple with the huge uncertainty we – and the Reserve Bank –  all face.

There is a great deal of complex, and pseudo-sophisticated, debate around monetary policy.  But when:

  • Core inflation has been below target midpoint for years
  • Unemployment is still above any estimate of NAIRU
  • Major commodity prices are falling
  • The building cycle –  often a key element in business cycles, and more so than usually this time given the salience of Christchurch –  looks to be turning,
  • And there is little or no sign of material demand or inflation pressures globally

the decision around the OCR really should be straightforward.  It should be cut.

4 thoughts on “Time to cut

  1. One of the things about monetary policy (at least, active discretionary forecast-based monetary policy) is that there will always be considerable uncertainty as to just what the right OCR will be. As I’ve observed before, with that level of uncertainty there is no particular shame in being wrong, provided one learns quickly from the mistake.

    When monetary policy implementation of an inflation target is described by a Taylor rule, as is typical, only current or expected inflation matter for the current change in policy. The implication is that the central bank should not respond to past deviations of inflation from its target but should continue to behave in a way that is ex ante consistent with the rule regardless of ex post failures.

    If the monetary authority chooses to correct its past mistakes, then the price level has a deterministic trend instead of a stochastic trend and the central bank can be thought of as implementing some type of average inflation targeting rule or (in the limit) price level targeting rule.

    Do you therefore advocate implementing average inflation targeting or price level targeting in New Zealand?


  2. Michael

    I’m not sure I buy that characterisation of the Taylor rule, After all, one also has to have a view on the neutral interest rate and the output gap. In a US context, pre-crisis, one might have been willing to use a fairly agreed number for the neutral rate, but output gaps are subject to huge uncertainty(and substantial revisions). Add to that that no inflation targeter owns up to operating a Taylor rule – they want to focus on forecast inflation, not actual. I think past errors in inflation are not important as something to try to reverse (the price level point) but as a pointer to the “model” being wrong.

    But on your more substantive point, personally I have been erring towards nominal wage targeting rather than inflation targeting. I’ve not been convinced by the literature or arguments around the case for levels targets (de facto, or as formal specifications). In part, it is politically infeasible – “we are tightening policy to drive inflation negative to meet our price level target” – and in part there are very few long-term fixed nominal contracts.


  3. Certainly. In my models the Taylor rule would include adjustments for the natural rate of interest, interest rate smoothing incentive, the output gap, and of course the central bank targets expected rather than actual inflation. This setup still implies that the policy rate will not respond to past mistakes, except to the extent that those mistakes have influenced inflation expectations. In this case it is still not entirely possible to judge the effectiveness of a central bank’s policies by the ex post outcomes, particularly over short periods of time; undershooting a target for a few years can be explained lots of things: unanticipated exchange rate appreciation, unexpectedly low global inflation, increased labour force participation, etc. It may be the case that the RBNZ has made policy mistakes, but it takes more than simply comparing the deviation of inflation from its target to make this case in a serious way.

    I agree that level policy targets have a number of practical problems, including the extreme assumptions about rationality and credibility required to make the policies welfare-improving. If you would be willing to share your thoughts on nominal wage targeting I would be interested in what you have to say.


    • Yes, I agree that simply looking at past inflation relative to target is not itself sufficient, altho core inflation measures help (even with all their limitations). As it happens, I wouldn’t argue that the RB has made serious mistakes at least until the start of last year. Lags are such that it is probably too soon to be conclusive about that, but I think there is now a pretty strong consensus that the OCr increases were unnecessary. Of course, if the Board or the minister were to scrutinise the Governor’s performance, he could mount a reasonable defence that most market economists were as hawkish (if not more so) than he was. Situation happened in reverse last decade – the Bank was demonstrably too slow to tighten but the bank economists were mostly more dovish.

      I haven’t developed the wage targeting idea at any length, but it has at superficial appeal based on (a) the idea of trying stabilise the least flexible prices (people typically apply it to non-tradables in the CPI, but why not wages) and (b) that financial stability in part depends on nominal incomes, and mortgage servicing in particular might be supported by predictable rates of nominal wage growth. Of course, it has its own political problems…..


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s