Australian wages run out of puff just as the labour market starts to soften: Justin Smirk
EXPERT OBSERVATION
The October Labour Force Survey surprised with a 19.0k decline in employment; the market median was for a 15k gain while the lowest forecast in the range was +7k.
And just a day earlier there was a disappointing update on wages.
While the rise in the September quarter Wage Price Index (WPI) was on market expectations at 0.5% it was below our forecast as we had been hoping that a combination of stronger wages growth in Victoria and the 3%yr rise in the minimum wage would have fed through to broader wage increases. That it has not, and now with signs the labour market is weakening, our forecast for a trend rising in unemployment suggests this is about as good as it will get for wages.
Wednesday saw the release of the wages update and while there were some signs of (fading) wage pressure, particularly in Victoria, we were disappointed to see that these wage pressures had not spread further. In the last year the average quarterly rise in the WPI was 0.6% but this is a soft 0.6% with rounding as the quarterly prints were 0.5% (Q4 2018), 0.6% (Q1 2019), then 0.5% in Q2 and Q3. In September the annual rate eased a touch to 2.2% from 2.3%, a pace it had held for previous four quarters and was, at best, a very modest acceleration from the low of 1.9% in 2017.
It is true that public sector wages are running a bit faster; in the quarter the annual pace eased to 2.5%yr from 2.6%yr in the June quarter, which was the fastest pace since December 2015. However, the weight for public sector wages is not large enough to change the overall picture, with private sector wages matching total wages - the annual pace for private wages easing to 2.2%yr from 2.3%yr in the June. The 2.4%yr pace in March 2019 was the fastest pace of private sector wage inflation since December 2014.
The state data did reveal some insights into wage dynamics but it also highlighted some significant questions. Much has been made of higher wage inflation in Victoria and how that it is associated with larger pay rises in the public sector, particularly for health workers. While it is true that health workers wage inflation is running faster in Victoria than the national average (4.3%yr vs 2.1% nationally and 3.0%yr NSW), this does not explain why construction, wholesale, accommodation & food services, transport, media, and professional & scientific services are also running at a faster pace than those industries' national averages.
In the end, Victorian private sector wages are running above the national average at 2.5%yr. Compare this to the 2.1%yr pace in NSW, 2.0% in Qld and 1.8% in WA. The only state with a faster pace of private sector wage inflation was SA at 2.6%yr.
As such, Westpac argues that the recent lift in wage inflation in Victoria was more to do with a tightening of labour market conditions there than outsized public sector wage gains. Victorian underutilisation fell almost 3ppt from early 2017 to early 2019. With a lag of around 6 months, this tightening in the labour market should lead a lift in wage inflation and, as expected, it did so from June 2017 with wage inflation lifting from 1.9%yr to a peak of 2.6%yr in June 2019.
The real question in our mind is why NSW wages have not responded in the same way they have in Victoria. Underutilisation was never as high in NSW as it was in Victoria (14% vs. 16%) and it has fallen to an even lower level currently (12% vs. 13%) and yet the best wages could do in NSW was lift from a low of 1.9%yr in 2016 to a peak of 2.3%yr in March 2019.
With wages in NSW responding to earlier tight labour market conditions, and growth turning, employment growth has started to soften, and with robust participation, unemployment is expected to drift higher both nationally and, in particular, in NSW.
This puts at risk Westpac’s current forecast of wage inflation drifting modestly higher from here - peaking around 2.4%yr in 2020/2021, and given how well contained wage inflation is across the nation, and between sectors, this modest uplift now looks optimistic. Note that the RBA recently lowered their forecast for the WPI in the November SMP (released prior to this week's WPI print). The RBA sees annual wages growth holding at just 2.3% through the forecast horizon.
The wage update was followed by the October Labour Force survey which highlighted the risk of softer employment growth, and rising unemployment (and underemployment). The 19.0k decline in employment was a significant surprise (market median +15k) as was that most of this decline was accounted for in NSW (–10.3k).
However, the more meaningful data point was the 0.1ppt rise in the unemployment rate to 5.3%. While participation eased as you would expect with a weak employment print, 66.0% from 66.1%, the unemployment rate still bumped up to 5.3% (5.32% at two decimal places) from 5.2%.
Our expectation has been for the unemployment rate to rise. That is despite an easing in the female participation rate, but that easing will be insufficient to offset the softer growth in employment.
Further pressure on the unemployment rate will come from a still robust participation rate amongst males.
Overall the interaction between our employment growth forecasts and participation outlook points to a lift to 5.6% in the unemployment rate by mid 2020.
Of further significance in the October Labour Force Survey was the jump in the underemployment rate to 8.5% (8.52% at two decimal places) from 8.3%. It was 8.54% back in August and the October update strengthens the trend drift higher in underemployment. This month there was a contraction in both full-time and part-time employment (–10.3k full-time and –8.7k respectively) driving a 0.2% decline in hours worked. This highlights a fundamental weakness in the report that is driving underemployment higher.
This trend drift higher in underemployment, and the slower pace of gains in hours worked compared to total employment (1.4%yr vs. 2.0%yr), reinforces the observation of the broader slack in the labour market slack beyond the 5.3% unemployment rate.
We do, however, acknowledge the month to month volatility in the labour force survey and while the October update was weaker than anticipated, it was not enough to change our view that the labour market is softening gradually which will drive a modest rise in unemployment and underemployment. Our leading indicators of employment are suggesting that the annual pace should be somewhere around 2.0%yr. The current pace is 2.0%yr and it is our forecast for it to slow to around 1%yr in mid-2020.
The Australian economy is a long way away from the RBA’s full employment aim of 4.5% and if the RBA wants to use monetary policy to drive the economy in that direction, it has a lot more work to do. However, the gradual pace of the deterioration will allow the RBA time to monitor the economy before having to act again.
In our view, while rate cuts along with the Government’s tax offset payments should help to support demand, we think it will be insufficient to get economic growth moving back to trend or above in 2020. As such we see employment growth continuing to moderate in early 2020 and expect that the resulting decline in participation will provide only a partial offset. We hold to our view that unemployment is likely to hit 5.6% by mid-2020 and see clear downside risks to our forecast for a modest lift in wage inflation.
JUSTIN SMIRK is a senior economist for Westpac
