Australia: Is the RBA heading to QE?

The latest rate cut by the RBA on 1 October took the cash rate to just 0.75% and comes after consecutive rate cuts in June and July of this year. The Australian economy has been hit by a combination of stagnation in some of the extraction industries following years of excess investment and capacity increases, and more recently by the impact of the trade war on global demand, notably for commodities, and the slowdown in the housing market.

Decent employment growth has not generated much of an upturn in wages, but has underpinned consumer spending; the economy has held together reasonably well given the headwinds it has faced. More recently, RBA Governor Philip Lowe hinted at a gentle (upward) turning point – an idea supported by what appears to be some bottoming in the housing market. But like many other central banks, the RBA is concerned about missing its inflation target. This is higher than in most OECD countries, with a mid-point of 2.5% for CPI. Currently, inflation stands at only 1.6%YoY (2Q19).

Recent research by the RBA also seems to suggest that the unemployment rate needs to fall to 4.5% for wages and prices to start rising. We believe it is this, and small increases in the unemployment rate (August print was at 5.3% from 5.2% in July) that led to the RBA easing aggressively during the middle of the year, and again more recently. Today's release shows unemployment was at 5.2% in September, still far from the RBA target, but improving.

Leaving aside the rights or wrongs of this approach (and we have serious misgivings) some observers are beginning to speculate that the RBA might eventually embark on a quantitative easing programme of its own, like other central banks including the US Federal Reserve, European Central Bank, Bank of England, Bank of Japan and Sweden's Riksbank.

Some are beginning to speculate that the RBA might eventually embark on a QE programme

Certainly, if the RBA is intent on seeing the unemployment rate fall to 4.5 %, then the economy isn’t growing nearly fast enough to achieve that, and so this could be one argument for embarking on unorthodox policy measures. We would argue that getting the unemployment rate this low with monetary measures alone, or even with the help of fiscal measures too, would require unrealistic amounts of stimulus. The negative consequences of doing this would likely far outweigh any positive benefits.

But worryingly, central bankers these days seem to be far more sanguine about the effects of such policies, as a recent BIS paper from a committee co-chaired by the RBA’s Governor Lowe indicates.

All in all, we suspect that to push the RBA to embark in quantitative easing we would need to see inflation still around 1.5% after two more rate cuts (cash rate at 0.25%) and some uptick in the unemployment rate (around 5.5%).

The most obvious approach for the RBA would be to purchase Australian government debt in much the same way as the Federal Reserve was a net buyer of Treasuries from 2009 to 2015. By April 2010, 12 months after commencing outright purchases the year before, the Federal Reserve had increased its holdings of Treasuries by around $1.1tr, at an average pace of $95bn per month.

Purchasing occurred in waves, though successive rounds of QE2 did not match the initial pace of purchase either in terms of the absolute change in outright ownership, or as a proportion of the monetary base. Bear in mind of course that the monetary base was itself swollen rapidly by these procedures.

For Australia, conditions could not be more different. The Australian economy is not imploding, asset prices, real and financial, are steady or rising, and there is frankly no crisis that needs such emergency monetary policy tools.

Bearing these very different backdrops in mind, then If the Fed effectively doubled the monetary base in the 2009/10 period for QE1, there is no plausible excuse for the RBA doing more than a fraction of that, say a 30% increase of the monetary base. Base money already grows by about 5% per year to keep track with the nominal growth of the economy, so an additional 25% increase would currently entail an annual increase of AUD28bn per year, from today’s monetary base of AUD112.6bn. That would be about AUD2-3bn of additional outright asset purchases per month.

Historical evidence suggests that, on a general basis, quantitative easing and the consequent expansion of a central bank balance sheet negatively affect rates (especially on longer maturities) and weakens the domestic currency, other things equal.

In a notable example, the asset purchase programme implemented by the ECB starting from early 2015 is widely recognised as a main driver of the euro weakening. In an ECB research paper, it is estimated that every expansion in the relative ECB-Fed balance sheet by 1.9% caused a 1.1% depreciation of the EUR/USD. According to such estimates, the ECB APP prompted a total 20% EUR/USD depreciation between the announcement of the programme in September 2014 and the end of 2016.

Should the RBA decide to explore quantitative easing in the next quarters, the reasons, size, domestic and international backdrops would only be few of the differences with the post-crisis QE experiences in the US, Eurozone, Japan and United Kingdom. However, it seems reasonable to assume a negative impact on the Australian dollar, given QE’s inherent power to dampen interest rates.

At this point, the relative size of the RBA vs the Fed's balance sheets does not show a meaningful correlation with the AUD/USD cross. However, this was also the case for EUR/USD and the ECB/Fed balance sheet before 2015.

For EUR/USD, we find that the negative correlations between the assets held by the central bank and the currency became evident only once the asset purchasing started. A similar case was seen in EUR/SEK, with the cross following the dynamics of the ECB/Riksbank relative balance sheets only after QE was introduced in Sweden. We would expect the same trend to occur for AUD/USD, and the correlation between the cross and the balance sheet to pick up only once the RBA starts engaging with or signals asset purchases.

Should (a) the RBA start the QE programme now and deliver AUD 28bn of QE (as discussed above) or (b) the Fed increases its balance sheet to curb money market volatility (note that we partly discount the effect of the Fed balance sheet increase in our calculations as it does not impact the UST curve but rather target money market dislocations), this would translate into an increase in the RBA/Fed balance sheet ratio by approximately 8%. If we use the “currency multiplier” provided by the ECB paper, the result would be a 5% depreciation in AUD/USD, all things being equal.