The RBA’s language change this week revealed a change in thinking driven by shifts in the data. But some data surprises might just be head fakes.
Something happened between 28 November and 9 December to shift the RBA’s thinking. The first of these dates marked a speech by Governor Bullock that can only be described as hawkish. It highlighted the RBA’s post-Review framework of treating instances of inflation above target as a sign of a positive output gap. The labour market was characterised as being too tight to be consistent with full employment. The slowdown in wages growth was not even mentioned.
Roll forward less than two weeks, and the language in the Board statement following the December meeting was rather different. The post-meeting statement highlighted that ‘some of the upside risks to inflation appear to have eased’. The possibility of a rate hike was no longer canvassed. And it dropped the ‘not ruling anything in or out’ and ‘vigilant’ language that had been in the post-meeting statement for most of the year.
Post-meeting comments by both the Governor and Deputy Governor have highlighted that both wages growth and the national accounts were weaker than expected. The Wage Price Index data were already available in time for the Governor’s speech, so it must have been the combination of the two that forced the shift in thinking. (The national accounts also include important measures of growth in wages and labour costs, which slowed noticeably in the quarter.)
The change in language certainly shifted market pricing of future rate moves. And to be fair, the probability of a rate cut earlier than our current base case of May 2025 has lifted, both because of the data flow and the RBA’s evident response to it. But a lot can happen between now and May.
One example of the ‘lot that can happen’ is that, since then, we have seen the surprisingly strong labour market data for November. But, Westpac Economist Ryan Wells cautions that this could, like last year, be a bit of a ‘head fake’ related to shifting seasonal patterns rather than an indication of a genuinely stronger labour market. Black Friday sales are becoming a bigger thing than they used to be, which is shifting the seasonal pattern of both employment and consumer spending. (Normal seasonal adjustment processes can’t fully offset seasonality that is shifting.)
Given ongoing cost-of-living pressures, it is no surprise that households are responding by going hard in the sales and holding back when things aren’t on sale. This also means that the next couple of months of household spending data could also be a bit of a ‘head fake’. And because Black Friday came late in the month, the December figures will be affected as well, as we saw in the Westpac Card Tracker this week. The RBA, and other observers of the economy, will need to be careful that they don’t jump at economic shadows in the next few months.
Along with seasonality, the labour market data might be providing a bit of a ‘head fake’ because of the skewed nature of current employment growth. As we have previously highlighted, most of the growth in employment over the past year or so has been in the non-market sector. Governor Bullock correctly pointed out in her post-meeting media conference that the expansion of the care economy represents important work in its own right – and work that frees up family members from unpaid care work.
But, as we have noted previously, the current skew towards care work has two implications that can be misleading if not treated carefully.
Firstly, care work tends to be lower-paid and less capital-intensive than many other jobs. So these jobs account for less GDP per hour worked than most other jobs. The result is apparent weakness in productivity growth driven by compositional change, even though individual workers and firms are not becoming less productive. If you wondered at Deputy Governor Hauser’s comment this week – that the narrative about slow productivity did not resonate with the RBA Board members involved in businesses – this is why. They are involved in market-sector industries, where labour productivity growth has been reasonable in recent quarters.
The problem here is that the models the RBA uses to estimate the output gap rely on estimates of trends in potential output that in turn rely on estimates of trend productivity growth. These estimates of trend are based on statistical techniques that impose considerable smoothness on those estimated trends. They are not designed to capture the kinds of compositional effects currently occurring.
Secondly, the current run-up in non-market employment will not last forever. The rapid run-up in spending in these areas has already spurred a reaction. Some news reports suggest that growth in NDIS spending is slowing a bit earlier than originally planned. As spending on these programs slows, so will the expansion of care-related employment. When that occurs, what will keep employment growth swift enough to keep pace with still-rapid population growth?
One thing working in the other direction is that labour force participation might not stay where it is either. Participation rates in Australia have been on a multi-decade upward trend, reflecting rising female participation in the workforce as well as increased participation by older people of both sexes as life expectancy and health outcomes improve. Around that upward trend, participation can fluctuate in line with job opportunities. At least some of the recent increase in participation was likely to have been a response to cost of living pressures, as we have previously noted. As those pressures ease, so too could the participation rate. As was the case this month, this could hold down unemployment due to people exiting the labour market. It would also contribute to other measures of labour market slack and could even mislead some observers about the true state of demand and the labour market.
Given all this, it will be worth taking care to avoid getting too hung up on every data point in the next few months, and instead focus on how they fit together.