Thursday, 22 September 2016

Leave well enough alone

No surprises from the Reserve Bank in this morning's review of the Official Cash Rate - it was held at 2.0%, as widely expected (in this survey, for example).

It's been steady as she goes over the ditch, too, not just in the sense that the Reserve Bank of Australia also stood pat at its latest decision, and left the Aussie equivalent at 1.5%, but also in the sense that, earlier this week, the Aussies re-committed to their inflation targetting regime, which is broadly similar to ours. They're not identical - the Aussie target is "keep consumer price inflation between 2 and 3 per cent, on average, over time" and ours is "keep future CPI inflation outcomes between 1 per cent and 3 per cent on average over the medium term, with a focus on keeping future average inflation near the 2 per cent target midpoint", for example - but they're clearly close cousins. The occasion for the Aussies staying with what they've got, by the way, was the change at the helm of the RBA, with incoming governor Philip Lowe taking over from the outgoing Glenn Stevens.

It would be nice to think that our local politicians will see the Aussies signing on for more of the same and draw the right conclusion - inflation targetting is still the mainstream serviceable model for monetary policy - but of course they won't. As sure as eggs is eggs, come to election time we'll have some of the political parties (and on past form, nearly all of them) promising to 'do' something about our monetary policy regime.

It's not an entirely discreditable exercise: some folks are beginning to wonder if there mightn't be something better. The Economist, for example, ran an editorial piece in its August 24 edition, 'When 2% is not enough: The rich world’s central banks need a new target', canvassing two ideas: raising the current target (typically 2% or so in many developed economies) to 4%, or moving on from inflation targetting to targetting of nominal GDP. The Economist argued that "A 2% inflation target is ill-suited to the rich world today. Doubling it would be an improvement, but targeting nominal GDP would be better still. Time for a new era".

It wasn't the most convincing thing the Economist has ever run. It argued, for example, that "credibly enacted", a 4% inflation target could be a goer, but "credibly enacted" magicks away the whole of today's biggest challenge: if central banks can't hit 2% today, despite throwing unprecedented firepower at it, what makes anyone think they can hit 4% tomorrow? Putting that aside, however, it's nonetheless plausible that there are at least some potential candidates for the Next Big Thing in monetary policy.

But as the magazine also said, "Changing targets is not something policymakers should do lightly; their credibility depends on stability". Exactly right: our current system may not be perfect, and as everyone knows we and other countries have been struggling to get inflation to where it should be, but unless there's a very clear benefit to change, net of the considerable credibility costs involved, we ought to do what the Aussies have just done, and stick with the programme.

Not that I expect this unasked-for advice will make a blind bit of difference to the next sets of election manifestos, but as I said last time I looked at some research on these issues, plea to the pollies is this.
Back off. We've got a working system that's done what it said on the label. It takes forever for new monetary systems to get bedded in and for people to get their heads around them: a central bank's credibility takes decades to lock down. We've got there: let's stay there.


  1. I would just like to know why anyone would think more inflation is better than less inflation? The problems that inflation causes are greater the higher is the rate of inflation so why do we want more inflation?

  2. I agree, 4% in the normal run of events is too high, and as I said the whole 4% argument didn't look very convincing. The argument is (probably) that something needs to be done to (temporarily) shock inflation expectations higher and get us away from zero lower bounds and all that. A better approach would be to recognise that persistently sub-2% inflation across many countries must mean that there is a global negative output gap, and we'd be better off addressing that than going for off-the-wall monetary experiments

  3. Here are two of the various arguments why moderate inflation is better than very low inflation, and much better than deflation, in the present context.

    Inflation erodes the effective burden of debt. I doubt any readers here need this process to be explained. Some may ask "why is that a good thing?'. Given our current monetary system nominal debt (and I mean total debt: public + private) must grow if we are to have an expanding money supply which, unless velocity of money increases, is necessary to support economic growth. (If anyone out there is still in denial about bank lending creating the great majority of our money check out 'Money Creation in the Modern Economy' on the Bank of England website from their Quarterly Bulletin 2014 Q1) If we have little or no inflation, yet nominal debt continues to grow, the cost of servicing an ever greater debt burden will depress demand even with historically low interest rates. Of course there are ways other than inflation to reduce the debt burden. If banks had had to face the same market discipline as workers and productive businesses do there would have been no bailouts or comprehensive emergency govt guarantees in 2008 and no QE to prop up their balance sheets since. The resulting bloodbath of bankruptcies would have substantially cleared the decks of debt but at the cost of enormous disruption the consequences of which no one could accurately predict. Still, if you really believe in free markets and that profit is the reward for risk then this was the only defensible course of action (or inaction).
    Another way to clear debts is by some political action: default, negotiated write-downs or write-offs, revolution, jubilee, etc. Plenty of historical examples of all these.

    Second reason to have moderate inflation is that it will stimulate demand. Economic actors will spend and invest more if they fear loss of value of their money holdings. Extremely low inflation (or worse, deflation) will have the opposite effect, slowing the circulation of money and further depressing demand in a feedback loop. Is there anyone out there who does not want to see aggregate demand rise?

    Lastly, there is much talk of monetary levers having been exhausted and that too much is expected of central banks in the role of ending this stagnation. There is a tool in the box ideal for this situation. Direct Monetary Finance. Lord Adair Turner does an excellent job of presenting the case for this and I would love to read some decent objections to it. The Central Bank would create debt-free money (or interest-free loans with explicitly no expectation of repayment, which amounts to the same thing) and either give it directly to consumers (a version of Milton Freidman's helicopter money') possibly with a stipulation that those who have debts use it first to repay them (Steve Keen's idea), or alternatively to finance govt spending. Some hold up their hands in horror at this although the same people don't seem to be aghast at unprecedented money creation to make whole the banks who got us into this mess. Yes, there are political risks. If it were up to the govt of the day to decide how much money to inject then the temptation to overdo it might well be irresistible. Therefore leave it to the (nominally) independent central bank to decide how much and how often. Govt gets to decide how to spend it. Personally I would like to see it finance infrastructure, esp renewable energy and transport - also housing. But that is a political question. The relevant point here is that it would add to aggregate demand without adding to the debt burden on future generations. Yes there would be a redistributative side effect. Does anyone still really think this would be a bad thing?

    1. Thanks for another thoughtful comment, and apologies for taking a while to respond.

      My own arguments for moderate inflation would be a bit different. The classic one is that moderate inflation makes it easier for relative prices to adjust without outright price cuts. This is traditionally mentioned in the case of pay rates - it's a good deal easier, socially, to hold pay steady when inflation is 2% and let real wages erode by 2%, if that is a mechanism that's needed (eg to improve international competitiveness), than it would be to try and cut pay by 2% when inflation is zero - but it also applies more generally.

      Another conventional argument for targetting moderate inflation that I think is right is that measured inflation may not account properly for quality changes in what's being bought. Inflation may be measured as 2%, but that could well be a misleading artefact that has missed out on the fact that the price in the shops may have gone up 2% but you're actually getting (say) 2% more functionality from what you buy. If you were to target a zero inflation rate as officially measured, you'd actually risk targetting a true inflation rate of -2%.

      I'm not enamoured of the argument for inflation to erode the value of debt (except perhaps in extremis). It seems to me an argument for policy made in bad faith. A government issues bonds and then, deliberately and later, runs an inflation rate to damage their value to investors. Nah. It's not good government. And as I mentioned before, it's going to be self-defeating, as you cannot fool all of the investors all of the time.

      In general I don't believe in the ability of monetary policy to influence long term living standards: you can't print your way (or legislate your way) to prosperity. It can, of course, have strong shorter-term effects. But its main longer-term effect is to provide a stable macroeconomic environment in which the real drivers of long-term living standards (productivity and institutions) can work properly.


Hi - sorry about the Captcha step for real people like yourself commenting, it's to baffle the bots