Is wages growth really as weak as we think?

In the run up to Australia’s federal election, the issue of low wages growth received considerable media attention, describing it as the lowest growth on record.

But claims like this are actually misleading, as they focus too much only on growth in nominal wages.

What should matter for workers and their families isn’t how much wages are increasing in money, but how much people can purchase with those money gains.

And despite the campaign assertions of former Labor leader Bill Shorten, the reality is that Australian wages have been rising by more than the prices of most goods and services.

It is true that until recently there has been barely any growth in the quarterly Wage Price Index (WPI) – the Australian Bureau of Statistics data source that is most often used (at least in the media) to measure growth in real wages; that is, the difference between growth in nominal wages and price inflation.

Over the five years until March 2018, real-wages growth averaged just under 0.3 per cent a year, before an unexpected drop in the rate of consumer price inflation caused growth to rebound to a very respectable 1.0 per cent in the latest year.

However, many Australians may be surprised to learn that the WPI doesn’t actually measure how much people earn. Instead, it is a measure of the wage attached to a specific job title and classification.

So, it doesn’t include, and was never intended to include, wage increases gained by promotions and job changes.

This is entirely sensible if the aim is to measure changes in the underlying cost of labour, but less so for measuring changes in actual worker earnings.

But what about the subdued real wage growth we see in the ABS’s average weekly earnings figures?

These figures are intended to measure what workers earn, and they too show a rate of growth in real average weekly earnings among full-time adult employees of just 0.4 per cent a year on average over the 5 years ending in November 2018.

But ABS average weekly earnings data come from a cross-section survey and, as a result, is affected by compositional changes in the workforce. If employment is growing (as it has been) and is concentrated, for example, on young inexperienced workers, the measured average earnings will be pushed downwards.

A very different approach to measuring wages growth is to use longitudinal data tracking the same workers over time.

One source of data like this is the Household, Income and Labour Dynamics in Australia (HILDA) Survey, though this data isn’t very timely. The most recent available data comes from 2017.

In the table below, we looked at the median annual percentage change in real hourly wages for those who are employed for two consecutive years. The sample is restricted to adult employees of working age, defined here as people aged 21 to 64 years.

TABLE: Median annual (per cent) change in real hourly wages (employees aged 21-64). Source: HILDA Survey Data General Release 17

The table shows that for this group of workers, real earnings growth has declined, from about two per cent a year in the period 2001 to 2013, to just 1.3 per cent a year in the period 2013 to 2017.

These rates of real wages growth are far higher (about a full percentage point higher) than those suggested by the WPI.

The table also splits workers according to whether they work full-time (35 hours per week or more) or part-time hours in their main job.

For full-time workers, real hourly earnings growth has been much stronger than for part-time workers, where growth has been close to zero.

Since we have longitudinal data we can also see that changes in employment status have a marked bearing on earnings, at least in the short term.

Full-time workers who switch to part-time typically experience marked increases in their hourly wage (in large part, because many of these new part-time jobs are casual and attract a casual wage premium).

Conversely those that switch from part-time to full-time experience a marked decline.

Many people might be surprised by the relatively high rates of earnings growth reported here, at least for the full-time employees.

The HILDA Survey data come from households and so provide earnings data that are inherently noisier than data sourced from employers, such as the WPI.

Nevertheless, when we treat the HILDA Survey as a series of cross-sections, we get estimates of average percentage changes in hourly earnings that overlap very closely with what is obtained from the WPI.

But once we focus on workers who maintain employment from one year to the next, we find the typical Australian adult employee in full-time employment has been doing much better than suggested by the WPI.

And in hindsight this shouldn’t be unexpected given two key avenues for securing wage increases – promotions and changing employers – aren’t captured by the WPI.

It means that contrary to all the hyperbole we hear about low wages growth, Australians in full-time work are doing pretty well – provided they remain in employment.

The people who are missing out are part-time employees, and it’s perhaps this sub-group of workers that needs more of our attention.

Banner: Getty Images