Posted by The August RBA meeting was more hawkish than expected. By placing focus on their medium term forecasts, the Board was able to duck the pressure from the current lockdowns, and hold monetary policy stable. In particular, the tapering of the bond purchase program to 4bn/wk from September was retained — against widespread expectations that the 6 July decision to taper would be reversed (purchases to remain at 5bn/wk until November) or that the pace of buying would be boosted (to 6bn/wk).
Easing was discussed, for sure. But the board eschewed the ‘signal-value’ of changing their policy settings. My guess is that there are costs to easing that they’d like to avoid. Sticking with the taper delays the looming concentration problems that will eventually limit the bond purchase program. They may also be uncomfortable with housing.
1/ Easing is being discussed
There was clearly a discussion about easing and an active decision to retain current policy settings at the August meeting. You can see this from the formatting of the post-meeting statement, which has the bulleted layout that we’ve only seen at the Mar’20, Nov’20 and July’21 meetings. Those were all meetings at which decisions were made.
The final paragraph supports this interpretation. It says that the bond purchase program ‘will continue to be reviewed’ — which strongly suggests it was reviewed at the August meeting, and may be reviewed again at subsequent meetings. The health situation is one of the review-triggers, so problems with the virus or vaccination program remain relevant. But I think you’d need a (bigger) disaster to get action.
The Board will maintain its flexible approach to the rate of bond purchases. The program will continue to be reviewed in light of economic conditions and the health situation, and their implications for the expected progress towards full employment and the inflation target.
We will find out more when Lowe speaks to the House Economics committee on Friday 6 August. I expect Lowe to emphasize the rosy forecasts and grounds for optimism about the bounce-back.
2/ The RBA is back to forecast-targeting
The RBA has resumed forecast-targeting. If there’s going to be easing, the 2023 forecasts must be downgraded.
The updated forecast looks straight through the current problem. They have an unemployment rate of 4% in Q4’23 and trimmed mean inflation at 2.25%. So in an academic sense monetary policy settings are just-right (never-mind the HUGE error-bands and historical bias to over-predict inflation).
This is a return-to-form for the RBA. After a decade of failing to hit their forecasts, many central banks, including the RBA, changed their monetary policy frameworks to put more emphasis on the current situation. The August statement’s emphasis on the rosy (but uncertain) medium term forecasts, over the (almost certain) large contraction in Q3’21 and uncertain recovery, confirms that the RBA is back to forecast-targeting.
Given the uncertainties of the virus, the vaccination program, and the rebound, a ‘COVID-risk-management’ approach would have favored deferring the taper. But that’s not what they chose.
Worries about over-stimulating the economy, due to the lags in policy transmission, ring hollow. An extra 10bn of bond-buying wouldn’t have moved the economic-forecast needle much. It would, however, have established that monetary policy was ready, willing and able to help. The downside of that signal is that it may have pumped-up asset markets (particularly Housing).
The Board didn’t make the (low cost) ‘COVID-risk-management’ decision: that suggests there may be some other non-COVID risk that has them uncomfortable.
3/ Bond market concentration might be a problem
Coming into this, I pushed back on the idea that the RBA would accelerate the pace of QE (to 6bn/wk), as I judged that the RBA’s foot-print in the bond market was becoming uncomfortably large. Holding the taper to 4bn/wk in September defers the looming concentration problem.
The most pressing issue is the mid-curve. The RBA buys significantly more mid-curve bonds than the AOFM issues. As a result, under most QE scenarios, the RBA ends up owning a very high share of mid curve bonds.
In an ideal world, the RBA would keep their share below 50%. The fix is to tweak the ratios, such that they have increased flexibility to buy across the curve. The most immediate fix would be to eliminate the distinction between the Monday (mid curve) and Thursday (long end) buybacks (there is more capacity in the long end, as that’s what the AOFM issues).
Adding Ultras to the pool of eligible assets adds ~50bn of bonds (assuming they keep ownership below 50%). Boosting the share of Semis in purchases from the current 20% would further increase the potential capacity of the program (I’ll write on how I would fix RBA-QE in another post).
If the RBA is uncomfortable with their footprint in the ACGB market, it follows that accelerating bond purchases is a remote possibility. Sustaining purchases at 4bn per week for longer, and stepping down more gradually, are more likely policy responses.
4/ Housing *is* under scrutiny
There was a subtle change to the housing market paragraph. From April to July, the RBA had said that they will be monitoring trends in housing borrowing. In August, this was changed to say that the RBA *is* monitoring trends.
Housing markets have continued to strengthen, with prices rising in all major markets. Housing credit growth has picked up, with strong demand from owner-occupiers, including first-home buyers. There has also been increased borrowing by investors. Given the environment of rising housing prices and low interest rates, the Bank will be is monitoring trends in housing borrowing carefully and it is important that lending standards are maintained.
Conclusion
Moving up the stock of bonds held by the RBA by 10bn or 20bn would have had a very small impact on the economy. The main value of would have been the signal that monetary policy was ready, willing, and able to help.
That the RBA eschewed this path despite a certain contraction of GDP in Q3 suggests there’s some cost to easing which has made them uncomfortable. It follows that accelerating QE in response to the current outbreaks is unlikely.
If the economy slows and the RBA wishes to respond, a slower taper — either buying at 4bn/wk for longer, or stepping down from 4bn /wk more gradually — seems more likely. This would give the AOFM more time to issue.
A big shock that boosts bond supply would eliminate this restraint on policy.
ricardian ambivalence 
Great note thank you