Wednesday, 19 October 2016

Now you see it, now you don't

Yesterday's inflation data held no great surprises. Everybody expected a pretty low number for inflation during the September quarter and for the year to September, and they got it: 0.2% for the quarter, 0.2% year on year.

Cue for hand wringing over the Reserve Bank yet again allowing inflation to stay too low.

But here's the thing.

We know the overall headline rate of 0.2% can be split into two bits, the bit that happens in the 'tradables' world of exports and imports, and the 'non-tradables' bit that happens in our own economy, the likes of the local authority rates, the doctor's bill, or the school fees. And we know that the overall 0.2% outcome was made up of tradables prices falling over the past year by 2.1%, while non-tradables prices rose by 2.1%.

The Reserve Bank has sod all influence over the tradables bit, except to the extent that it might manage to control the exchange rate, which influences how much of the world inflation rate comes through to us. But it (and any other central bank) tends not to be able to steer exchange rates terribly well, so in practice whether the Reserve Bank is on top of things comes down to whether it is steering non-tradables inflation to where it needs to be.

So let's have a look at that non-tradables inflation in more detail. Here it is.

The blue line is annual non-tradables inflation, which is running at 2.1%. So remind me again how the Reserve Bank hasn't got inflation back up to 2%, the midpoint of the 1% to 3% band it's supposed to be focussed on?

But (you'll say) that 2.1% rate of non-tradables inflation isn't all it's cracked up to be. It's not really 2.1%. It's inflated, innit, by the housing market. And you're right, it is. But if you take out the cost of new houses, you get the green line, non-tradables ex housing. It's running at 1.8%. That's not too bad, either, if you're supposed to be aiming at 2%.

Course (you'll reply), there are other housing-market-related things still being counted in that non-tradables ex housing line, in't there? Rent. The cost of keeping the house in good running order. The rates. And again, you're right. So let's purge the non-tradables inflation of every damn house-related thing - the cost of a new house, the rent, yadda yadda yadda.

That gives you the red line, non-tradables less anything to do with a house. On that basis non-tradables inflation is running at 1.25%. That's short of the Reserve Bank's 2% focus, so you could beat them about the ears if you felt like it. On the other hand, it's at least crept back into the 1% to 3% band, and it's clearly headed in the right direction. Every one of these measures has been on the rise all year.

One of the big mysteries of macroeconomics recently has been, where's the inflation gone? Why hasn't it come back like it used to when things pick up? That's a big topic, and everyone from Janet Yellen at the Fed to Philip Lowe, the new governor of the Reserve Bank of Australia have been having a crack at it, and I'll come back to it one of these days.

My thought today, though, is this: maybe it's actually come back, and we haven't noticed.


  1. Non-tradables ex all housing costs might be a measure of inflation that is worth using. But that's not what what the PTA requires the RBNZ to achieve.

  2. Thanks Lindley for the comment, and you're right on both points. I think it's got some value as a supplementary indicator for monetary policy purposes, but may also have other value. I've sometimes felt that it might be a useful sighting on the overall flexibility of goods and labour markets, and possibly of the strength of domestic competition, and might also be another insight into where the economy is relative to potential output. So not easy to interpret, but if it was running very hot or very cold you'd at least want to try and see what might be going on.

    While we're talking about it as a measure, over on his Croaking Cassandra blog Mike Reddell in this post said that

    "Non-tradables inflation typically averages well above tradables inflation (for various reasons) and so can’t meaningfully be compared with the CPI inflation target midpoint, although some people – including some who should know better – do so"

    I suspect I am some (of all) of those "who should know better"! But that's all right. Having thought about it a bit more, Mike's also right, and you'd probably want to see it running more like 3% before you could say the RBNZ was hitting its target. The current 1.25% (or 1.8% or 2.1% whichever version you like) isn't there yet.

    That said, I think that (a) even if the RBNZ is still behind the 8 ball, it's not as far behind as it was, and (b) we may have reached (or indeed gone past) a turning point in inflation that the headline CPI hasn't revealed at all. We've had false dawns before, so who knows, but the graph is quite suggestive.