Soft activity data highlight that risks around the path for the RBA cash rate are skewed down. This has less to do with changes to the RBA’s models of the neutral interest rate and more about the risk that it will again be surprised by the domestic economy.
Soft GDP data for Q1 has again raised questions of whether the RBA is behind the curve and has left policy too tight for too long. As we have been highlighting for some time, underlying growth in Australia remains weak and sensitive to pauses in the expansion in the care economy. With inflation in the 2–3% target range already, and likely to stay there, why wouldn’t the RBA cut further and faster than previously believed?
Certainly, this is the question we keep asking ourselves and are hearing from clients. The risks to our current policy view are clearly to the downside, as we have previously highlighted. Yet a headlong switch into a more dovish near-term path for policy sits awkwardly against the Board’s own words in the minutes. The Board expressed ‘a preference to move cautiously and predictably when withdrawing some of the current policy restriction.’ And just before that sentence: ‘They also judged that it was not yet time to move monetary policy to an expansionary stance, taking account of the range of estimates involved, given that inflation was yet to return sustainably to the midpoint of the target range and the staff’s assessment that the labour market was still tight.’
Recall also the Governor’s response to Michael Pascoe’s question at the media conference, ‘We get quarterly inflation rates in this. We don’t get monthly. We get the monthly indicator, which is very volatile. We get four readings on inflation a year. Other countries get them 12 times a year.’ We read this as saying that the Board (or at least the Governor) wants to wait for the June quarter CPI, ahead of the August meeting, not the July meeting. The irony is that by late this year, Australia will have a full monthly CPI.
Predicting the movement of the stars
Part of the art of predicting what central banks will do with their policy rates goes beyond forming a view about where the economy is headed. There is also an element of forecasting future shifts in the central bank’s beliefs about how the economy works. This includes their beliefs about where the ‘star variables’ such as the neutral real interest rate (r*) and full-employment rate of unemployment (NAIRU) are, and the weight they put on these assessments in their policy deliberations. Both the location of and the reliance on these estimates can change, sometimes quite abruptly.
Ideally, the policymaker would communicate these evolving beliefs. Some of the detail is only for the aficionados, though, and therefore is sometimes – understandably – left on the cutting-room floor when putting together flagship documents intended for a broader audience. It is therefore left to those aficionados to interpret, and indeed forecast, where policymakers’ views are headed.
For example, at the beginning of this year, our views about the path of Fed policy was partly shaped by our assessment that the Fed staff view of the neutral rate was still too low and that they would progressively revise it up. There was therefore some chance that the Fed would end up lowering the Fed funds rate below neutral and have to backtrack as this view was revised. In the event, the Fed view caught up with our expectations soon enough that this backtrack did not occur. But the underlying (and out-of-consensus) view that the Fed would not end up cutting as far as had previously been priced in was sound.
Of course this was not just about judgements about where neutral is. As that earlier note discussed, there is also the question of whether there are other factors working against a ‘glide to neutral’ path for policy. Differing fiscal stances across countries is an obvious example, particularly when the parlous state of the US fiscal position is such an outlier. Working in the other direction, policy uncertainty around tariffs and other chaotic policy decisions of the Trump administration is clearly weighing on US growth prospects more than they do on countries such as Australia.
In Australia, these assessments and their likely direction of travel are also key inputs into any view of future RBA policy. For example, it is hard to shake the impression that, whatever more nuanced language about full employment being multidimensional, the RBA’s view of the upside risks to inflation are coloured by their model-based assessments (released under FOI) that the unemployment rate is below the NAIRU. The revision of the RBA’s forecasts for trimmed mean inflation from not-good-enough-dead-flat-at-2.7% to good-enough-now-dead-flat-at-2.6% is absolutely a judgement. And while it was not directly driven by models, that judgement to write down 2.6% and not 2.5% anywhere in the forecast horizon was likely shaped by the RBA’s view that the labour market is still tight and productivity growth weak.
Against that, though, we have been getting questions about an apparent downward shift in the RBA’s view of where neutral is, evident through a comparison of graphs showing estimates of the nominal neutral in a sequence of documents. A year ago, the RBA’s view was coloured by surprisingly convergent estimates of r* well above 3%, implying that monetary policy was restrictive, but perhaps not that restrictive. This, along with its analysis of productivity trends, contributed to the hawkishness of RBA decisions and communications in the latter half of last year and at the February meeting.
However, new models were introduced between November last year and February this year that were at the lower end of the existing range of models. (We understand that these were based on recent work at the New York Fed.) And unlike all the other models, these new ones did not show an upturn in estimates post-pandemic. The average of all these estimates is now in the high 2s rather than the low-to-mid 3s range we have favoured for both the US and Australia. This means that the neutral real cash rate is estimated to be barely above zero. You don’t need to look at 700 years of data to worry that such an estimate is on the low side.
At this stage, though, we assess that, while the weight of those r* estimates is indeed lower than a year ago, the weight that the policymakers put on the central estimate has also declined. There are a couple of straws in the wind supporting this conclusion, including comments by the Deputy Governor following the February Board meeting. We will be watching closely for signs that the RBA is taking these models more seriously than we currently think. More pertinently, though, we will also be watching for signs that the labour market and inflation are softer than the RBA currently expects. If the downside risks around the policy path are to come true, it will require the RBA to change its mind about how the economy is traveling.
