Inflation targeting three decades on

It is almost three decades since the Reserve Bank of New Zealand became the first central bank to formally structure its monetary policy around a specific and publicly announced target for inflation.  Since that time, the great majority of central banks have also adopted an explicit inflation target.

Targeting inflation rather than some real economic variable reflects the academic wisdom of the time that the best contribution that monetary policy can make to economic growth and employment is to ensure that the price system works effectively. And having that target publicly announced, with the governor of the central bank personally accountable for delivering that target, was seen as a way of minimising the real costs of reducing inflation from the high levels of preceding years by changing inflation expectations.

Did it work? Inflation as measured by the CPI certainly fell quite quickly – indeed, slightly more quickly than required by the target.  Unemployment rose quite strongly at the same time, reaching 11% in 1991 as inflation fell sharply, and some observers have suggested that that shows the policy failed to reduce the real costs of disinflation. Perhaps, though because a substantial range of other policies were changed at the same time – quantitative import controls removed, export subsidies scrapped, over-staffed public sector enterprises privatised, etc – it is impossible to know for certain whether having an explicit and publicly announced inflation target did reduce those costs.

Though there is now widespread acceptance that central banks should be transparent about what they are trying to achieve, and that keeping prices, however defined, stable or rising in a slow and predictable way should be at the heart of what monetary policy is focused on, there is no agreement on who should be involved in setting that inflation target.

In New Zealand, the inflation target was – and still is – set in a formal written agreement between the minister of finance on behalf of the government and the governor of the central bank. That agreement is made when the governor is appointed, and must be made public. 

The board of the central bank was made responsible for monitoring the governor’s performance in achieving the agreed target, but was to have no part in decision-making with regard to monetary policy. It was hoped that making the governor personally accountable for achieving the agreed inflation target – with an explicit understanding that he could be dismissed for failing to deliver the target in the absence of some clearly identifiable exogenous shock – would help drive inflation expectations in the right direction.

I am aware of only six other central banks that combine full instrument independence and an inflation target determined with explicit political input … but I strongly favour such input provided that it is indeed explicit and public

At this stage, I am aware of only six other central banks that combine full instrument independence and an inflation target determined with explicit political input – Australia, Canada, Israel, Norway, Sweden and the UK – but I strongly favour such input provided that it is indeed explicit and public. It very largely protects the central bank from criticism from the government, provided the bank is delivering the agreed inflation rate, and helps government decision-makers understand that the decisions they make concerning fiscal policy are very likely to have monetary policy implications.  Even more important, it rightly involves the elected government in a decision that is inherently highly political – the rate at which the country’s currency loses value.

But what of the target itself?  In 1988, at the de facto commencement of the inflation targeting era (the law providing for the new framework was not changed until the following year), our target was 0–2%.  To the extent we had a rationale for that target – as distinct from having a general commitment to very low inflation – we assumed some upwards bias in the measurement of consumer inflation – judged to be 1% per annum by the Boskin Committee in the US – so the target was “genuine price stability plus or minus 1%”.

The New Zealand inflation target has gradually been amended over the years, and now sits at 1–3% over the medium term, with a mid-point of 2% – the mid-point of most other central banks’ inflation targets.

Is 2% too high a target?  After all, abstracting from measurement bias, a 2% target doubles prices roughly every 35 years – meaning that prices increase nearly eight-fold over the course of a century.  Hardly price stability.

But perhaps 2% is too low a target?  Given the downward rigidity of nominal wages and the zero bound for interest rates – though of course that assumption has been called into question lately – perhaps the target inflation rate should be higher? I suspect that there will in time be more flexibility in nominal wages – with a larger part of total remuneration being subject to circumstances – and perhaps we have to accept that quantitative easing could become a routine part of monetary policy unless other aspects of economic policy are able to play a more substantial role. I would certainly be reluctant to see central banks deliberately target a faster erosion of the purchasing power of the money for which they are responsible.