I’ve just been finalising for publication some comments I presented late last year at the IJCB-RBNZ conference marking the 25^th anniversary of inflation targeting, and reflecting again on the wider set of options from which authorities can choose.  Looking ahead, I’m not convinced that inflation targeting is clearly superior to the alternatives.  The slightly unconventional option that I would prefer to see explored in more depth is nominal wage targeting.   Wages are the stickiest of the nominal prices, and nominal wages are also the key element underpinning the servicing of nominal debt.

But that is a topic for another day.  Today I want to focus more narrowly on the question of “if we are going to run an inflation target, where should that target be set”.  In our system, the Minister of Finance has the lead in setting the target[1].

The approach to setting the level of the first inflation target was pretty simple. The Act specified that the primary goal of monetary policy was “stability in the general level of prices”, and in setting an inflation target range the idea was simply to have something as close as feasible to that.   I think most of those involved thought a 0 to 2 per cent annual inflation target was a pretty close approximation to “price stability”.

The more hard-line among us were a bit disappointed when Winston Peters and Don Brash later agreed to shift the target up to 0 to 3 per cent, and then Michael Cullen and Alan Bollard agreed to a further shift up to 1 to 3 per cent.   These changes seemed to arise from some mix of political product differentiation, and convergence towards the targets being used in other countries.  But I don’t think anyone at the Bank really thought these changes would make much macroeconomic difference, good or ill.  Nominal targets generally shouldn’t.  The tax system works a little less well with slightly higher inflation, but any downward nominal rigidities are also a little less pressing.

One factor that never got much attention was the near-zero lower bound on nominal interest rates.  We knew it was there –  if policy interest rates got much below zero people could simply shift to cash – but it never seemed likely to be much of an issue for New Zealand.

Craig Ebert, senior economist at the BNZ, put out an interesting note a month ago arguing that New Zealand should think seriously about lowering its inflation target.  The gist of the argument is that deflation isn’t particularly harmful and that the current global low inflation is, from a New Zealand perspective, a good thing, which should be accommodated rather than offset.

I’m not persuaded.   Deflation isn’t likely to be harmful if it results from positive productivity shocks, and occurs in economies where private debt has only a modest role.   In most of the older deflations that the recent BIS paper looked at, not only was private debt much less important than it is now, but the near-zero lower bound was not a binding constraint.  We have never had a period in (modern?) history when policy interest rates have been near-zero for so long, and when private debt (and private financial assets) have been so pervasively important.

Last year, a leading figure in global central banking over the last 20 years visited New Zealand.  I asked him whether, purely with the benefit of hindsight, he wished that inflation targets had been set nearer 5 per cent than 2 per cent.  Doing so would have provided additional leeway to reduce real policy interest rates, to more deeply negative levels, during the period since 2007.  I didn’t really expect him to agree with the proposition, but what really surprised me was how few arguments he could put up in defence of  inflation targets centred on 2 per cent.  For countries that have now spent years at the near-zero lower bound, the with-hindsight case for higher initial inflation targets seems pretty clear-cut.  In the old line from James Tobin “it takes a heap of Harberger triangles to fill an Okun’s gap”.   Yes, the tax system worked a bit better with lower inflation, but there are an awfully large number of people unemployed across the advanced world at present.  The limits of monetary policy are part of the story.

But where should inflation targets be set now?  Is there a case now for raising them to 4 or 5 per cent, as commentators as eminent as Olivier Blanchard and Ken Rogoff have suggested?  In particular, is there such a case for New Zealand?   Eric Rosengren, President of the Boston Fed recently called for a debate on whether the Fed’s inflation target should be raised.  The FT suggests he was the first serving policymaker to openly canvass the issue.

The issue is perhaps more relevant for New Zealand (and Australia) than for most other advanced countries.  Why?  Quite simply because advanced countries that are at the near-zero lower bound  at present have no way of credibly raising inflation, and inflation expectations, from something around current inflation targets to something around 4-5 per cent.

If they could then inflation expectations would rise and real interest rates would fall, drawing forward more demand, and absorbing the current excess capacity (getting people back into jobs).  But policy interest rates in these countries can’t (with current institutional constraints) be cut materially further, and no one really believes that QE can make that degree of difference.   And people might reasonably conclude that an announced higher target was simply a desperate measure for crisis times, and that the authorities would renege on it once economic recovery got underway.

Some have suggested that expansionary fiscal policy provides a way through.  In principle that sounds fine, but in most (but not all) of these countries public debt is already very high (and understated in official figures, which often don’t record public sector pension liabilities), populations are ageing and –  probably not unrelatedly –  the political appetite for materially expansionary fiscal policy is largely non-existent.  Witness, as an example, the current UK election.

New Zealand is not at the near-zero lower bound.  But neither was most of the rest of the advanced world in 2007.  Indeed, almost all those countries had policy interest rates then above New Zealand’s current 3.5 per cent.  In the last rate-cutting cycle, the OCR was cut by 575 basis points and in the previous cycle, in the 1990s, short-term interest rates were allowed to fall by a similar amount.    Future recessions will happen, and there is no obvious reason to think that they will be smaller than we’ve seen before.  But the OCR could not now be cut by 575 basis points or anything like it.  Our Minister of Finance, and his advisers, should be treating that as a pressing concern.

Nominal targets shouldn’t matter very much for medium-term economic outcomes.  And they wouldn’t without the near-zero bound, and the free option of converting to cash.   Some argue that a negative 5 per cent Fed funds rate would have been helpful in the US in the depths of the recession. But if such a rate had been implemented there would have been large scale conversions  to physical cash and effective interest rates would not have gone materially negative.  Every central bank knows this, and none has been willing to cut policy rates so far as to see those large scale conversions.  These provisions are acting as a real constraint on central banks now, and constraining the speed at which economies can rebound from recession.

New Zealand should be acting now to be better positioned when the next serious downturn comes.  We don’t know when that will be.  If things turn nasty in China, in the rest of the emerging world, or in Europe once the first country leaves the euro, it could be very soon.  But it could be years away.    Or it might never happen.  No one can forecast timings.  But contingency planning isn’t about timing, but about planning for identified vulnerabilities.

There are two broad options:

• Serious policy work on overcoming the near-zero lower bound, or
• Raise the inflation target

What could be done about the near-zero lower bound?

• In the longer-term, it might involve looking to phase out physical central bank notes and coins (which now play a very small role in day-to-day transactions)
• It might involve repealing the legislative monopoly the Reserve Bank has on issuing physical notes and coins. The scope for innovative market-led alternatives might make it easier to end physical issuance by the central bank altogether
• The stock of notes and coins could be capped at current  levels.  To the extent that physical currency has value as a retail payments medium it would still be available, but if it became scarcer the price would rise.
• Central banks could put impose a fixed fee for converting settlement account balances into physical currency.  A 10 per cent fee, for example, would be likely to provide material additional leeway to allow policy rates to be cut more deeply negative.

People like Miles Kimball, Willem Buiter and others have written on some of these issues in much more depth.  I’m not sure which of these options, or others that have been touted, would be best.  But they are issues that central banks and finance ministries should be exploring in more depth now.  In New Zealand’s case, a joint working party with Australian officials might be an efficient way to address the issues.

Dealing with the near-zero lower bound –  mostly a government intervention which has become troublesome – should be a better option.  Something like a stable value of money seems a better outcome for society than a steady targeted debasement of the value of the currency.    If that could be done, a lower medium-term inflation target, centred on (true) zero might be desirable.

But if –  for whatever reason –  the authorities are not willing to do anything active about removing the near-zero lower bound, then they need to be thinking much harder about raising the inflation target[2].  New Zealand can actually deliver higher inflation, and do it quite quickly.  The OCR is 3.5 per cent, and there is nothing to suggest that the transmission mechanism is so impaired that lower policy rates would not flow through into lower retail interest rates, a lower exchange rate, and to higher inflation.   In some ways it would be a shame to have to do it.  But the possibility of a prolonged period in which the scope for conventional monetary policy has been exhausted, and the number of unemployed people sits far above normal levels for years, is not a risk that a democratic government should contemplate with equanimity.

[1] Section 9 of the Reserve Bank Act requires that “The Minister shall, before appointing, or reappointing, any person as Governor, fix, in agreement with that person, policy targets for the carrying out by the Bank of its primary function during that person’s term of office, or next term of office, as Governor.”  Since the Minister has a blocking veto on any recommended person for appointment as Governor, the Minister should normally have the dominant influence on the content of the Policy Targets Agreement.

[2] And would then need to do something about indexation of the tax system.