The case of New Zealand: Inflation targeting, relative prices, monetary policy and industrial policy

According to this piece on Rates Blog, in a recent presentation Dr Geoff Bertram discussed the inflation targeting framework, and the idea of the NZ$ being overvalued.

Back in the day I had him as a lecturer, he is an extremely smart man and he brings up a number of important points.  However, I have to disagree with his view regarding inflation targeting and relative prices in the economy.  There are two main themes where I feel our views differ, fundamentally these are:

  1. The driver of the change in the relative price of non-tradable and tradable goods
  2. The relevance of industrial policy in relation to monetary policy.

Let me flesh these out.

Relative prices, inflation, the traded sector

For me, the very purpose of inflation targeting is to give relative prices “bite” – so the idea that targeting inflation has lead to inappropriate relative prices cuts to the heart of my very justification for using it.

The argument that inflation targeting has caused this is as follows:

  1. non-tradable inflaton is above target
  2. non-tradable inflation is relatively unresponsive to interest rates
  3. increasing interest rates to beat it down increases the exchange rate – reducing tradable prices
  4. these lower tradable prices are used to say “we are at target” when we aren’t.

There are several direct issues with this argument.

For one, a one off increase in the exchange rate of this kind is a “price level” shift – not a change in underlying inflation.  So as a one-off factor the Bank tends to look past this.  We would need constant growth in the exchange rate to provide the disinflationary impact Dr Bertram is after here – not a high exchange rate but an ever rising one.

Now he is completely right the price price of tradable goods has been declining relative to non-tradables.  There are two reasons for this:

  1. High non-tradable price growth due to competition issues, government (local and central) creep.
  2. LOW tradable good price growth – due to productivity growth in developing economies.

Hold up a second here – where was that last factor in the analysis provided?  In my opinion, the exceptionally low tradable good price growth (look at the data man, it has been weak) has also been very important.

Inflation targeting didn’t CAUSE either of these relative price changes – instead overseas growth, and domestic industrial policy were the drivers of the “relative price change”.  Monetary policy has merely tried to steady the ship.  As a result, when thinking about the relative price changes we have to ask “what do they mean, and is there anything policy relevant here” … it does not imply we have to do ANYTHING to monetary policy.

Causes, industrial policy, terms of trade

Now could it be that there are competition issues that have seen the cost of non-tradables (building costs, local council rates, owner occupied rents) rise more quickly – potentially.  And these are issues for the commerce commision, and also to keep in mind when we look at social policy.  Again, this does not imply that monetary authorities have failed – if anything it merely implies that government officials have failed here.

Also, don’t forget that we haven’t only experienced low growth in imported good prices … our export prices have been pretty good.  This sharp increase in our TERMS OF TRADE would be expected to push up the relative price of non-tradables to tradable goods.

Why?  New Zealand sell tradable goods oversesas, sure – but our supply curve is relatively inelastic.  This means that as prices rise, volumes don’t go up by much – this is the decision of the exporters involved.  However, as prices go up they get more income – and they demand more goods and services.  The increase in demand for goods and services leads to a corresponding rise in the production of these things pushing up the price.  It is entirely conceivable that, in a case where our terms of trade is rising (pushing up incomes) but imported good prices are very weak that the increase in non-tradable good prices could be larger.

Update:  Sorry that example was terrible – I was half asleep, and although it wasn’t wrong it was hard to understand.  Lets do this again:  Say world price of manufactured goods fall, in of itself this would push up the ratio of non-tradable prices to tradables.  However, it also pushes up the terms of trade – which leads to a boost to domestic incomes!  The increase in domestic incomes pushes up demand for non-tradables and so pushes up their price … thereby pushing the ratio of non-tradables to tradables even higher!  As a result, when we have lots of productivity improvements in manufacturing overseas we would expect to see this ratio change – if it wasn’t something funky would be going on 😉

Also, when we are talking about relative prices, lets just think about something here – do we agree that manufactured goods have become relatively cheaper over the last two decades compared to say haircuts?  I definitely feel like that, and it makes sense as that is where there is scope for technological improvement.  What do you know, this change in relative prices suggests exactly the same thing – that is where the change has occurred so the actual relative value HAS changed … if this relative price change had not occurred, then there would be a problem!

Currency mismatch

The primary focus of Dr Bertram’s piece is not on these issues – but on currency mismatch and how that could have driven the currency above fair value.  That is a fair point, although I have it under advisement that this mismatch is not as major an issue anymore.  However, I am not here to argue that – and as a result, that reason for a higher exchange rate can stay unchanged.

The point I wanted to make was that the terms of trade, productivity improvements overseas, and domestic competition issues explain the very issues he was blaming monetary policy for – I don’t agree with his attack on inflation targeting, inflation targeting did not drive this relative price change (as they would have looked past a one off shift in the price level) these other factors did.  Not mentioning them in his piece and attacking the mandate of central banks was inappropriate.

  • andrew coleman

    Matt, let us take your underlying logic further.

    Suppose you are right and the central bank has relatively little influence over relative prices in th emedium or long run. (I have not seen evidence countering this proposition .) Then a central bank charged with maintaining low inflation (or, to be precise, to ensure the general level of prices is stable) should have little interest in relative prices except to the extent that relative price movements destabilise price expectations. In a world where non-tradeable prices persistently increase relative to tradeable prices because of faster productivity growth in the latter sectors, the central bank should try to make sure both sets of prices in terms of money are increasing at a lower rate. To make this concrete, and to quote an example from a 2007 paper in the Reserve Bank Bulletin (which showed that relative price movements in Australia and New Zealand across disaggregated sectors are highly correlated in the medium term, with sectors with prices that increased relatively rapidly in NZ also increasing relatively rapidly in Australia and vice versa) , between 1990 and 2005 non tradeable prices in NZ increased by 63% (3.3% pa)while tradeable prices increased by 16% (0.8%). Since relative price movements in Australia were very similar, it should be concluded that the Reserve Bank should not be trying to reduce the rate of price increase in the non-tradeable sector compared to the tradeable sector. Rather it should be trying to ensure that both sets of prices in terms of money increased less rapidly: say by +20% and – 27% respectively over the period.

    I would contend that a problem in NZ is that the central bank allows the rate of increase of both tradeable and non-tradeable prices to be too high, particularly because of the way the tax system penalises some classes of assets in an inflationary environment more than others. If the tax system is not indexed for inflation (which it isn’t) allowing both tradeable and non-tradeable prices to persistently increase in terms of money is to create distortions that favour excessive borrowing and that penalise people who prefer to lend than purchase assets. Having a central bank deliver outcomes that persistently favour borrowing over lending is as curious public policy as would be having a bank that worries excessively over prices in the non-tradeable sectors.

    andrew

  • @andrew coleman

    Hi Andrew, thanks for another thoughtful comment.

    If I was to summarise what I took for your comment it would be as follows: Essentially you are saying that there has been a persistent trend in relative prices and that if we look past that trend the true underlying rate of inflation has been too high, especially given the effective “inflation taxes” inherent in our tax system.

    I see exactly where you are coming from, and there are two points I would like to raise in line with this:

    1) In essence we should measure the comovement in prices that is independent of prior relative price shocks and treat that as the inflation measure – with that in hand we have a clearer idea of whether the Bank is indeed meeting its mandate.
    2) The inherent bias in taxation is more an institutional issue than an issue with the implementation of monetary policy. In essence since the tax system is set up how it is, and since the inflation target is set where it is, we face this cost. To solve it, we need to change either the inflation target or the tax system – rather than anything with the implementation of monetary policy by the Bank.