After a quarter-century of successful inflation targeting by the Reserve Bank, you would expect inflation expectations to be well "anchored" to the bank's target of 2 per cent.

But an extensive survey of people running businesses - those who set prices and wages - has found nothing of the sort. Not even close.

The survey also found a remarkable level of ignorance about the Reserve Bank's mandate, its current inflation target and even who the governor is.

The survey's findings are reported in a paper by Auckland University of Technology economics lecturer Saten Kumar and others to a conference this month at the Brookings Institution in Washington.


Four waves of the survey of (initially) more than 3000 chief executives, managing directors, general managers and directors of New Zealand businesses between late 2013 and late last year found their average expectation for inflation a year ahead clustered around 5 per cent.

That compares with forecasts from the central bank itself, which ranged from 1.1 to 1.9 per cent, and between 1.7 and 2 per cent from professional forecasters. The actual outcomes so far have been below 1 per cent.

More troubling is that during the third wave of the survey, undertaken a year ago, when the managers were asked for their pick for the average annual inflation rate over the next five to 10 years, around 40 per cent of them said they expect inflation to run at 5 per cent or higher - twice as many as opted for the 2 per cent mid-point of the bank's target band.

Other findings were also at odds with any concept of well anchored inflation expectations. The managers reported a high degree of uncertainty around their forecasts. They were liable to change them substantially over time.

And in particular, their long-run expectations swung around with fluctuations in their short-run expectations - exactly the opposite of what you would see if respondents were confident the central bank would deliver inflation rates in line with its mandate, which is keeping annual consumers price index rises between 1 and 3 per cent on average over the medium term, with a focus on the mid-point of 2 per cent.

Only 31 per cent of respondents identified keeping inflation low and stable as the bank's main objective.

When asked what the bank's inflation target is, only 12 per cent said 2 per cent, though a further 25 per cent said either 1 or 3 per cent, the bottom and top of the target band.

But 15 per cent thought the target was 5 per cent and a further 36 per cent thought it was more than 5 per cent.


And only 30 per cent identified Graeme Wheeler as governor; 17 per cent thought it was Bill English.

The reason all this is disturbing is that, among inflation-targeting central banks, it is essential for inflation expectations to remain anchored to their inflation targets.

Credibility on that point is their stock in trade. Ultimately, it is about confidence in the currency as a store of value.

What matters for monetary policy is not just current nominal interest rates, but expected real interest rates, which are a function of inflation expectations.

During the Great Inflation of the 1970s and 1980s, people had to be constantly mindful that the dollar in which they priced their goods or services, or in which their wages were paid, or their bank deposits were denominated, was liable to be worth 5 or 10 or 15 per cent less - who could say? - in a year's time.

Getting inflation under control and instilling confidence that low and stable inflation was here to stay was a costly but essential process - one we would not want to have to repeat.

So the importance of maintaining confidence among wage- and price-setters that we live in a world of low (but not too low) and stable inflation, in which near-term deviations from a 2 per cent target will be short-lived, has become an article of faith among central bankers.

The details of the survey are more encouraging. Those who said they relied on news media for information on prices tended to be more accurate than those who relied on just their own experience as consumers, and the 20 per cent of firms which relied on professional forecasts, from the likes of their bank's economists, were the best informed.

Understandably, respondents were often wrong about the relative weighting of different components of the CPI, exaggerating the importance of petrol prices, for example, and underestimating food.

The more competitors a firm faced, the less likely it was to be wrong about the Reserve Bank's inflation target and about monetary policy more generally, suggesting greater competition encourages firms to pay more attention to economic and monetary conditions generally.

Dr Kumar said firms tended to be much more accurate about their own input costs.

So the optimistic interpretation of the survey might be that if inflation really was a problem, the people running businesses would devote more attention to it and be less loose in their inflation expectations.

You might say that inflation targeting has been the victim of its own success, in that sense.

And it is fair to point out that the more regular, if smaller, surveys the Reserve Bank monitors, like the ANZ Business Outlook survey, present a more reassuring picture of inflation expectations: "Survey measures of medium-term inflation expectations have stabilised or increased slightly following declines during 2014 and the beginning of 2015," it says in the September monetary policy statement.

"These measures are at a level that is consistent with CPI inflation returning to around 2 per cent in the medium term."

But the results Dr Kumar and his colleagues report challenge that sanguine view.

See the AUT research here: