Wages growth has peaked and started declining. At current rates, even quite low productivity growth would be compatible with inflation being sustained at target. So why is the RBA so worried?
The Wage Price Index increased 0.8% in the September quarter and 3.5% over the year. This was in line with our expectations but – as Westpac Economics Senior Economist Justin Smirk pointed out – slightly below consensus expectations. The extent of the step down in the year-ended growth rate was well anticipated, because it reflected the dropping out of the outsized 2023 National Wage Case and related decisions from the calculation.
The RBA does not publish a full quarterly wages forecast profile, only the forecasts for year-ended growth as at June and December quarters. So, we do not know exactly what they expected for the September quarter. However, it would now need to see a bounce back in quarterly growth to around 1% for the December quarter for its end-2024 forecast to come true. Even allowing for some recent health-care agreements, we consider such a bounce to be beyond the bounds of plausibility given how smooth this series tends to be. There are no strong reasons for a change of direction of this kind, either. Surveys, data on awards and enterprise agreements and feedback from our own customers would all suggest that a sudden bounce back in wages growth is not happening.
We therefore expect that the RBA will have to revise down its near-term wages growth forecasts again in February, having already done so in November.
Forecasting is hard, so some revisions and near-term misses are par for the course. Even so, is there something going on with the way some observers think about domestic labour costs, that could be affecting their interpretation of the economic outlook? And in the case of the RBA, could this be affecting its monetary policy decision-making?
Some insights can be gleaned from the following passage from the latest Statement on Monetary Policy:
At current rates of productivity growth, WPI growth remains somewhat above rates that can be sustained in the long term without putting upward pressure on inflation. All else equal, when productivity growth is positive, WPI growth is able to outpace inflation while still being consistent with inflation at the midpoint of the target range. As trend growth in labour productivity is likely below its rate in previous decades, the sustainable WPI growth rate is probably lower than in the past and below the current rate of growth. That suggests it would be difficult to sustain wages growth at its current pace in the longer term without a higher pace of trend productivity growth.
There are a few things worth noting about this passage.
First, this reasoning comes from the markup model for forecasting inflation. As explained in a previous note, this model starts from the presumption that prices are a (roughly stable) markup over costs, including labour costs. A bit of algebra later leads to a relationship that states that wages growth minus productivity growth is approximately equal to inflation (prices growth). As discussed in that previous note, there are a lot of assumptions underlying the use of this relationship for forecasting. But more fundamentally, the WPI is not the measure of labour cost growth that maps most closely to the one implied in that relationship. Rather, the more volatile average earnings measures from the national accounts are more relevant.
Presumably the RBA has used the smoother WPI measure for ease of exposition. In that case, though, one should be even more circumspect about how tightly the relationship should hold.
Second, there are some interesting implied choices of time period used in that paragraph. For example, it is stated that future trend productivity growth is expected to be slower than the average of previous decades. This is not controversial: the late 1990s was a period of strong productivity growth globally, largely because of the adoption of personal computers and other new technologies. More recent productivity growth was slower, but not zero. The real question is whether future productivity growth will be slower than the average of more recent times, such as the years leading up to the pandemic. Perhaps this is true, but the reasons for a further slowdown have not been elucidated. While any boost from AI and other technology will indeed take time to show up in the productivity figures, just as PCs did, a further decline in global trend productivity growth is not the base case for the profession more broadly.
Third, even granting the reduced noise from using the WPI, and assuming a further slowdown in global productivity, there is the question of why the RBA repeatedly referenced the sustainability of the current rate of growth. At the time of publication, this was the year to the June quarter figure of 4.1%, not the year to September quarter figure of 3.5% just reported. Yet the RBA surely anticipated the step down in growth that was already baked in to awards and many enterprise agreements. Why the focus on the sustainability of a growth rate that everyone knew was not going to be sustained? The question also arises of how we reconcile wages growth having already rolled over, to annualised rates in the low 3s, with the RBA’s view that the labour market is still tight.
Later in the document, the step down in unit labour cost growth from 7% annualised to 3½% annualised in just six months was noted (as we had previously expected and written about). So why the implication that growth in labour costs was much stickier than that?
The deeper question is: with wages growth tracking in the low 3s and productivity growth not being zero, why has the RBA focused so much on the risk that wages growth is unsustainable?
I can’t help thinking that this partly reflects deep-seated narratives about the Australian economy not being competitive. These narratives stemmed from the so-called ‘real wage overhang’ that emerged in the 1970s following the policy-induced wages breakout then. Another bout of this belief system emerged after the mining boom and attendant strong income growth. Since then, restoring competitiveness by crimping wages growth has been a common go-to in the policy discourse in Australia, far more than elsewhere in my observation. It is as if people forget that exchange rates tend to move much faster than domestic labour costs.
In any case, even if productivity growth averages a touch lower than 1% (worse than recent history), then by the RBA’s own figuring, WPI growth averaging 3.2% (the annualised rate of the past three quarters) is well and truly consistent with inflation averaging 2½% or below. Perhaps we need to let go of the pandemic-era hangover.
