Sophia Rodrigues gets the wobbles on QE3

There is one little-known RBA watcher who has got things more-or-less right of late: Central Bank Intel's Sophia Rodrigues. About one month before the RBA announced its open-ended, $4bn/week QE3 taper in July, Rodrigues was on this trade. She got the yips immediately before the July meeting, and advocated for $5bn/week. In her defence the AFR's John Kehoe and many others were saying the same thing. And the RBA revealed the choice between $4bn/week and $5bn/week was a line-ball call at board level.

The problem with the RBA's statement on this was that the economic situation after July deteriorated substantially with the elongation of the NSW (it's not just Sydney now) lockdown, and additional lockdowns in Victoria, Queensland and elsewhere. We know with the Delta variant the lockdowns are going to be longer and harder than last year, and more frequent, in turn implying the rebound will be a lot soggier. And of course there is not as much fiscal stimulus being doled-out, which the RBA seems to be overlooking when assuming that the recovery will be just as strong as those experienced last year.  

Instead of accepting that its July decision to taper QE was wrong, the RBA doubled-down, remarkably releasing even more optimistic economic forecasts following its August meeting to rationalise its decision one month prior. It basically ignored the lockdowns altogether, and then surprisingly argued that the bar for deferring the QE3 taper was going to be very high. 

In its defence, the RBA did very clearly attach one huge caveat: it would review the situation in again September, just before the taper commenced, and if conditions did materially worsen, it could defer then. And to its credit, it pushed this message through key journos, including Kehoe and News Ltd's Terry McCrann, who all stressed this important point. McCrann was interesting, because he had previously argued that anyone suggesting QE3 was going to be deferred in August was stupid. Yet immediately after the meeting he was at pains to emphasise it could in fact happen at the September meeting.

The idea that there was such a high bar to deferring QE3 in August was surprising given only one month earlier the RBA had claimed that starting the third open-ended QE program at $5bn/week (no taper) or $4bn/week (taper) was effectively a toss-of-the-coin proposition. And justifying its decision to maintain the taper in the face of a huge deterioration in economic conditions relied on the RBA's rubbery forecasts, which I have written about extensively in the AFR here. 

In short, given the impossibility of economic forecasting during a 1-in-100 year pandemic, the RBA had promised to be data-dependent and "nowcast". Yet fast-forward, and suddenly it was telling us all that its QE decisions would, in fact, be based on those unreliable forecasts, not the previously-advocated, and much more dependable, nowcasts. Well of course the RBA's July forecasts were blown-up within days of its decision to announce its QE3 taper. (Even during the good times, the RBA's forecasting track-record has been pretty ordinary.)

In any event, Rodrigues was one of the few journos to get the RBA's insistence to not defer its QE3 taper at its August board meeting right. She was convinced that it basically did not matter how bad the lockdowns got---the RBA would (rather irrationally) maintain its July decision. 

And following the August board meeting, Rodrigues was one of the first to roll-out the argument that there was a "very high bar" to the RBA deferring the QE3 taper in September even though the economic depression rippling across east coast Australia continued to get worse. 

To be clear, everyone knows the RBA made the wrong decision in both July and August. There was a rich range of policy choices it could have made in both months that would have had immediate positive impacts and no costs. It should have stuck to its original plan of moving to an open-ended QE3 program at $5bn/week, reviewed quarterly, as first unveiled by the AFR's Kehoe. When it decided to announce the $4bn/week taper in July, it should have deferred this in August. 

If it was worried about market capacity constraints on QE3, which many believe to be the case, these restrictions were of its own making and are very easily resolved: most obviously, it could have increased its underweighted purchases of State government bonds relative to its overweighted buying of Commonwealth government bonds and/or extended the tenor of its purchases from 5 to 10 years to, say, 5 to 15 years.

I've argued here that one big saving grace for the RBA might be the Committed Liquidity Facility: crushing this down to zero---as APRA has repeatedly argued was likely to occur---would in turn force banks to buy more than $100 billion of government bonds. 

As much as the banks would squeal about having to do so, it would be a de facto form of QE3 (allowing the RBA to continue to taper QE3), although I have no idea whether the RBA has even considered the possibility. But it is frankly the only theoretical rationale that would help make sense of the RBA's decision-making.

Anyway, Rodrigues has started to get the wobbles on her call for there being a very high bar for the RBA deferring its QE3 taper in September. Whether there is actually any signal in this move, nobody knows, because the RBA's shift to leaning on its rather meaningless forecasts makes its own decision-making hard to predict. Writing for her own Central Bank Intel subscribers, Rodrigues appears to open the door to a September taper deferral:

RBA’s Shifting Thinking As Reprieve From Lockdowns Getting Delayed
For the first time since the Covid-19 pandemic, the Reserve Bank of Australia’s forecasts surprised to the upside, and ironically in less than a week, its thinking has shifted in the direction of downside risks. 
The shift in thinking isn’t surprising given positive health outcomes was the main reason why the economy bounced back quicker than the RBA projections at each round of forecasting. This time the number of Covid cases continue to rise, and it now looks certain that the lockdowns will last longer than what the RBA forecast in the latest Statement on Monetary Policy...
Longer lockdowns are worrying, and longer-than-expected is a bigger concern because of the risk that it would lead to job losses, and the unemployment rate will take longer to return to the prelockdown level... 
Since the SOMP’s release, the RBA’s thinking is tilting towards the downside scenario because the news on lockdowns and Covid cases continue to disappoint. But because there is still optimism on the vaccination front, it is difficult to know if only the nearterm downside risks will come to pass or whether, and how much, this would affect forecasts for next year...
But if persistent lockdown news means the outlook for next year would clearly come under threat, it has implications for monetary policy...
BACKTRACK ON TAPER REMAINS AN OPTION
Lowe kept the option open by saying, “We will, however, keep the situation under review, and we are prepared to act in response to further bad news on the health front that affects the outlook for the economy over the year ahead.” 
As CB-Intel wrote earlier this month, the hurdle to backtrack on tapering decision is very high but the more the RBA tilts in the direction of downside scenario, the greater the possibility that the bar would get lower.

As noted above, the RBA has clearly signalled that it was retaining the optionality to defer QE3 in September. And the data has only continued to get a lot worse. Perhaps the "high bar" might be hurdled, although I am not holding out hope at this point given how reluctant the RBA has been to acknowledge the change in the facts it faces.