Milton Friedman once said never to underestimate the importance of luck of nations. I believe that is very true and I think the same goes for central banks. Some nations came through the shock in 2008-9 much better than other nations and obviously better policy and particularly better monetary policy played a key role. However, luck certainly also played a role.

I think a decisive factor was the level of key policy interest rate at the start of the crisis. If interest rates already were low at the start of the crisis central banks were – mentally – unable to ease monetary policy enough to counteract the shock as most central banks did operationally conduct monetary policy within an interest rate targeting regime where a short-term interest rate was the key policy instrument. Obviously there is no limits to the amount of monetary easing a central bank can do – the money base after all can be expanded as much as you would like – but if the central bank is only using interest rates then they will have a problem as interest rates get close to zero. Furthermore, it played a key role whether demand for a country’s currency increased or decreased in response to the crisis. For example the demand for US dollars exploded in 2008 leading to a “passive tightening” of monetary policy in the US, while the demand for for example Turkish lira, Swedish krona or Polish zloty collapsed.

As said, for the US we got monetary tightening, but for Turkey, Sweden and Poland the drop in money was automatic monetary easing. That was luck and nothing else. The three mentioned countries in fact should give reason to be careful about cheering too much about the “good” central banks – The Turkish central bank has done a miserable job on communication, the Polish central bank might have engineered a recession by hiking interest rates earlier this year and the Swedish central bank now seems to be preoccupied with “financial stability” and household debt rather than focusing on it’s own stated inflation target.

In a recent post our friend and prolific writer Lorenzo wrote an interesting piece on Australia and how it has been possible for the country to avoid recession for 21 years. Lorenzo put a lot of emphasis on monetary policy. I agree with that – as recessions are always and everywhere a monetary phenomena – the key reason has to be monetary policy. However, I don’t want to give the Reserve Bank of Australia (RBA) too much credit. After all you could point to a number of monetary policy blunders in Australia over the last two decades that potentially could have ended in disaster (see below for an example).

I think fundamentally two things have saved the Australian economy from recession for the last 21 years.

First of all luck. Australia is a commodity exporter and commodity prices have been going up for more than a decade and when the crisis hit in 2008 the demand for Aussie dollars dropped rather than increased and Australia’s key policy rate was relatively high so the RBA could ease monetary policy aggressively without thinking about using other instruments than interest rates. The RBA was no more prepared for conducting monetary policy at the lower zero bound than the fed, the ECB or the Bank of England, but it didn’t need to be as prepared as interest rates were much higher in Australia to begin with – and the sharp weakening of the Aussie dollar obviously also did the RBA’s job easier. In fact I think the RBA is still completely unprepared for conducting monetary policy in a zero interest rate environment. I am not saying that the RBA is a bad central bank – far from it – but it is not necessarily the example of a “super central bank”. It is a central bank, which has done something right, but certainly also has been more lucky than for example the fed or the Bank of England.

Second – and this is here the RBA deserves a lot of credit – the RBA has been conducting it’s inflation targeting regime in a rather flexible fashion so it has allowed occasional overshooting and undershooting of the inflation target by being forward looking and that was certainly the case in 2008-9 where it did not panic as inflation was running too high compared to the inflation target.

One of the reasons why I think the RBA has been relatively successful is that it effectively has shadowed a policy of what Jeff Frankel calls PEP (Peg the currency to the Export Price) and what I (now) think should be called an “Export Price Norm” (EPN). EPN is basically the open economy version of NGDP level targeting.

If the primary factor in nominal demand changes in the economy is exports – as it tend to be in small open economies and in commodity exporting economies – then if the central bank pegs the price of the currency to the price of the primary exports then that effectively could stabilize aggregate demand or NGDP growth. This is in fact what I believe the RBA – probably unknowingly – has done over the last couple of decades and particularly since 2008. As a result the RBA has stabilized NGDP growth and therefore avoided monetary shocks to the economy.

Under a pure EPN regime the central bank would peg the exchange rate to the export price. This is obviously not what the RBA has done. However, by it’s communication it has signalled that it would not mind the Aussie dollar to weaken and strengthen in response to swings in commodity prices – and hence in swings in Australian export prices. Hence, if one looks at commodity prices measured by the so-called CRB index and the Australian dollar against the US dollar over the last couple of decades one would see that there basically has been a 1-1 relationship between the two as if the Aussie dollar had been pegged to the CRB index. That in my view is the key reason for the stability of NGDP growth over the past two decade. The period from 2004/5 until 2008 is an exception. In this period the Aussie dollar strengthened “too little” compared to the increase in commodity prices – effectively leading to an excessive easing of monetary conditions – and if you want to look for a reason for the Australian property market boom (bubble?) then that is it.

This is how close the relationship is between the CRB index and the Aussie dollar (indexed at 100 in 2008):

However, when the Great Recession hit and global commodity prices plummet the RBA got it nearly perfectly right. The RBA could have panicked and hike interest rates to curb the rise in headline consumer price inflation (CPI inflation rose to around 5% y/y) caused by the weakening of the Aussie dollar. It did not do so, but rather allowed the Aussie dollar to weaken significantly. In fact the drop in commodity prices and in the Aussie dollar in 2008-9 was more or less the same. This is in my view is the key reason why Australia avoided recession – measured as two consecutive quarters of negative growth – in 2008-9.

But the RBA could have done a lot better

So yes, there is reason to praise the RBA, but I think Lorenzo goes too far in his praise. A reason why I am sceptical is that the RBA is much too focused on consumer price inflation (CPI) and as I have argued so often before if a central bank really wants to focus on inflation then at least the central bank should be focusing on the GDP deflator rather on CPI.

In my view Australia saw what Hayekian economists call “relative inflation” in the years prior to 2008. Yes, inflation measured by CPI was relatively well-behaved, but looking at the GDP deflator inflationary pressures were clearly building and because the RBA was overly focused on CPI – rather than aggregate demand/NGDP growth or the GDP deflator – monetary policy became excessively easy and the had the RBA not had the luck (and skills?) it had in 2008-9 then the monetary induced boom could have turned into a nasty bust. The same story is visible from studying nominal GDP growth – while NGDP grew pretty steadily around 6% y/y from 1992 to 2002, but from 2002 to 2008 NGDP growth escalated year-by-year and NGDP grew more than 10% in 2008. That in my view was a sign that monetary policy was becoming excessive easy in Australia. In that regard it should be noted that despite the negative shock in 2008-9 and a recent fairly marked slowdown in NGDP growth the actual level of NGDP is still somewhat above the 1992-2002 trend level.

George Selgin has forcefully argued that there is good and bad deflation. Bad deflation is driven by negative demand shocks and good deflation is driven by positive supply shocks. George as consequence of this has argued in favour of what he has called a “productivity norm” – effectively an NGDP target.

I believe that we can make a similar argument for commodity exporters. However, here it is not a productivity shock, but a “wealth shock”. Higher global commodity prices is a positive “wealth shock” for commodity exporters (Friedman would say higher permanent income). This is similar to a positive productivity shock. The way to ensure such “wealth shock” is transferred to the consumers in the economy is through benign consumer price deflation (disinflation) and you get that through a stronger currency, which reduces import prices. However, a drop in global commodity prices is a negative demand shock for a commodity exporting country and that you want to avoid. The way to do that is to allow the currency to weaken as commodity prices drop. This is why the Export Price Norm makes so much sense for commodity exporters.

The RBA effective acted as if it had an (variation of the) Export Price Norm in 2008-9, but certainly failed to do so in the boom years prior to the crisis. In those pre-crisis years the RBA should have tightened monetary policy conditions much more than it did and effectively allowed the Aussie dollar to strengthen more than it did. That would likely have pushed CPI inflation well-below the RBA’s official inflation (CPI) target of 2-3%. That, however, would have been just fine – there is no harm done in consumer price deflation generated by positive productivity shocks or positive wealth shocks. When you become wealthier it should show up in low consumer prices – or at least a slower growth of consumer price inflation.

So what should the RBA do now?

The RBA managed the crisis well, but as I have argued above the RBA was also fairly lucky and there is certainly no reason to be overly confident that the next shock will be handled equally well. I therefore think there are two main areas where the RBA could improve on it operational framework – other than the obvious one of introducing an NGDP level targeting regime.

First, the RBA should make it completely clear to investors and other agents in the economy what operational framework the RBA will be using if the key policy rate where to hit zero.

Second, the RBA should be more clear in it communication about the link between changes in commodity prices (measured in Aussie dollars) and aggregate demand/NGDP and that it consider the commodity-currency link as key element in the Australian monetary transmission mechanism – explicitly acknowledging the importance of the Export Price Norm.

The two points above could of course easily be combined. The RBA could simply announce that it will continue it’s present operational framework, but if interest rates where to drop below for example 1% it would automatically peg the Aussie dollar to the CRB index and then thereafter announce monetary policy changes in terms of the changes to the Aussie dollar-CRB “parity”.

Australian NGDP still remains somewhat above the old trend and as such monetary policy is too loose. However, given the fact that we have been off-trend for a decade it probably would make very little sense to force NGDP back down to the old trend. Rather the RBA should announce that monetary policy is now “neutral” and that it in the future will keep NGDP growth around a 5% or 6% trend (level targeting). Using the trend level starting in for example 2007 in my view would be a useful benchmark.

It is pretty clear that Australian monetary conditions are tightening at the moment, which is visible in both weak NGDP growth and the fact that commodity prices measured in Australian dollars are declining. Furthermore, it should be noted that GDP deflator growth (y/y) turned negative earlier in the year – also indicating sharply tighter monetary conditions. Furthermore, NGDP has now dropped below the – somewhat arbitrary – 2007-12 NGDP trend level. All that could seem to indicate that moderate monetary easing is warranted.

Concluding, the RBA did a fairly good job over the past two decades, but luck certainly played a major role in why Australia has avoided recession and if the RBA wants to preserve it’s good reputation in the future then it needs to look at a few details (some major) in the how it conducts its monetary policy.

PS I could obviously tell the same story for other commodity exporters such as Norway, Canada, Russia, Brazil or Angola for that matter and these countries actually needs the lesson a lot more than the RBA (maybe with the exception of Canada).

PPS Sometimes Market Monetarist bloggers – including myself – probably sound like “if we where only running things then everything would be better”. I would stress that I don’t think so. I am fully aware of the institutional and political constrains that every central banker in the world faces. Furthermore, one could easily argue that central banks by construction will never be able to do a good job and will always be doomed to fail (just ask Pete Boettke or Larry White). As everybody knows I have a lot of sympathy for that view. However, we need to have a debate about monetary policy and how we can improve it – at least as long as we maintain central banks. And I don’t think the answer is better central bankers, but rather I want better institutions. It is correct it makes a difference who runs the central banks, but the institutional framework is much more important and a discussion about past and present failures of central banks will hopefully help shape the ideas to secure more sound monetary systems in the future.

PPPS I should say this post was inspired not only by Lorenzo’s post and my long time thinking the that the RBA had been lucky, but also by Saturos’ comments to my earlier post on Malaysia. Saturos pointed out the difference between the GDP deflator and CPI in Australia to me. That was an important import to this post.