Is jobs growth enough for economic success?

We need good jobs and more good jobs

Australia’s economic recovery from the COVID-19 crisis has so far been impressive in its speed and magnitude. Government support has undoubtably played a major role, but the surprise factor has been the strength in consumer spending.

The extraordinary level of income support from the government has driven household disposable incomes higher, despite record low wage growth. Rising disposable household income has enabled households to build-up savings, as well as spend, mostly on goods rather than services. 

What happens next in the journey to economic recovery is all about the underlying strength of the labour market, because it needs to pick up the role of driving household income as government support winds down.

Getting Australians back to work is already happening. Despite the recessionary environment in 2020, participation in the labour market is at record levels and the unemployment rate continues to decline.

But looking at just employment misses the point: household income is what drives consumer spending, and income is not merely a question of whether you’re employed, but how much you earn per hour, and how many hours you work. What happens to income growth, and what households do with the war-chest of saving will be critical to Australia’s economic success and longer-term outlook.

To date, the calibration of an improving labour market with the withdrawal of government support looks to be well balanced. Almost 900,000 people lost their jobs in the first two months of the pandemic, but as of December, eight out of nine jobs had returned, 140,000 fewer people were on JobSeeker and the number of businesses meeting the revenue-based eligibility criteria for JobKeeper have fallen faster than expected.

But there’s a way to go as support continues to taper, and the balance is fragile, particularly in coming months as policies such as JobKeeper come to an end. Get it wrong and an income recession would likely see private consumption falter and risk the economic recovery.

Encouragingly, leading indicators suggest the economy will continue to create jobs. The number of job vacancies nationally have lifted to their highest level on record to be 12 per cent above their pre-COVID level – equivalent to there being about one job vacancy between four unemployed people. This ratio compares favourably to Australia’s last recession where there was one vacancy between 34 applicants. It now seems likely that the unemployment rate peaked at 7.5 per cent in July and will continue its downward trend. 

So will this jobs-led recovery be enough?

While the jobs recovery is undoubtably impressive, this is only part of the picture. Wages growth is at record lows and hours worked have been slower to recover than the level of employment. Hours worked remains 1.4 per cent below pre-COVID levels, compared to employment at 0.7 per cent below.

This is because much of the recovery in jobs has been in part-time roles, which are up 24,000 on pre-COVID levels, compared to there still being 112,000 fewer full-time jobs. While an increase in part time roles is expected during periods of economic uncertainty – as businesses hesitate to hire full-timers—the numbers also reflects a much bigger trend at play that was well underway prior to the pandemic: the growing casualised workforce.

Increased labour market flexibility as a result of changes to labour laws between the 1980s and 2000s have made our labour market more dynamic and contributed to a rise in part-time and casual jobs. This has been a positive for the economy, it has allowed people to work more flexibly if they prefer, which has helped Australia avoid longer term downturns and reduced the costs associated with long term unemployment—such as skill atrophy and reliance on government assistance.

This increased flexibility in the labour market can be seen when we analyse how the labour market has adjusted following previous economic downturns. EY analysis of Australia’s labour market finds that since the 1990s, more firms have looked to adjust people’s hours worked within a job rather than let workers go—resulting in lower job losses during the peak of downturns. The Reserve Bank of Australia has previously acknowledged  “the share of labour market adjustment due to changes in average hours worked has increased threefold since the late 1990s” .

But, the COVID-19 recession has so far been different. As the above chart shows, the labour market response is unique in how quickly hours worked fell and in the mix of adjustment between jobs and average hours, which has been roughly even and more akin to that seen in the 1980s and 1990s.

Over 2021, we expect that employment growth will continue to outpace the recovery in average hours worked, similar to the recovery we saw in the DotCom bust and Global Financial Crisis.

One consequence of this style of recovery is that part-time employment is likely to continue to rise at a faster rate than full-time employment, creating the risk that some people are likely to struggle to see their hours restored or their skills fully utilised.

These people are considered underemployed – employed but would prefer, and are available, to work more hours than they currently do – equivalent to 8.5 per cent of the labour force, up from an average of 6.7 per cent during the late 1990s

The underutilisation rate, which measures both unemployment and underemployment, has declined since its peak of 20.2 per cent in May to 15.1 per cent in December but remains 1.4 percentage points above its pre-COVID level.

A survey by the Australian Bureau of Statistics (ABS) found that just under half (49.7 per cent) of underemployed part-time workers are taking active steps to look for additional hours. This means there are a higher number of applicants for employers to consider, which lessens the need to pay higher wages to fill job vacancies.

EY analysis of ABS data shows there is about 44 million hours each week of spare capacity in the labour market from people considered unemployed or underemployed. When we exclude those that are considered long-term unemployed and between jobs, there are still 35 million hours a week being lost. Based on a minimum wage, that equates to $694 million a week of income not hitting household bank accounts.

It’s not enough just to look at “jobs and more jobs”

The impact of these shifts in the labour market mean there’s no longer a straight line between unemployment and wages growth. This is an issue because wages growth is a traditional marker policymaker look for in determining the extent households can be relied on to contribute to economic growth and government taxes. 

The underutilisation rate is a more effective predictor of wage growth as demonstrated by a simple economic model, predicting around 70 per cent of the variation in private sector wages growth since the 1990s.

The correlation between lower underutilisation and faster wage growth is shown by the blue shading below. This makes sense, as firms will likely respond to a lift in demand by first offering more hours to their existing (underemployed) workforce before bidding up wages to attract new employees.

Conversely, the unemployment rate leads to only a 15 per cent effective prediction of wages growth, as seen in the lack of a clear pattern in the purple shading below. 

Low wages growth is a threat to Australia’s economic success

Forecasts for wage growth are a key determinant of monetary and fiscal policy settings so there is a lot at stake in getting them right. Private wages growth was at a record low of 1.2 per cent in the year to September 2020. Moreover, Commonwealth Treasury is forecasting growth of just 1¼ per cent to June 2022. Given that wages growth of 2.1 per cent, the average growth seen in the five years prior to COVID, was considered weak and well below the long run average of 3.2 per cent, the forecast numbers are very weak.

Moreover, if we take into account forecast inflation, real wages growth is set to go backwards over the next two financial years.

Only using traditional models that link unemployment to wage growth risks actual wage growth falling short of forecasts. In today’s context, the risk is negative real wages growth.

Wage forecasts are typically based on the gap between the unemployment rate, and what is considered full employment, currently estimated to be 4.5 per cent.

If we were to forecast wage growth based on underutilisation instead, that would require a commensurate marker of full capacity. Based on EY modelling, we estimate that figure to be around 12 per cent.

At the end of 2020, the gap between under-utilisation and full capacity was 3.5 percentage points, indicating more spare capacity than the 2.5 percentage point gap between unemployment and estimated full employment.

Using the RBA’s forecasting models*, EY analysis of this disparity suggests wages growth could be up to ½ per cent lower on average over the next four years than compared to a model which uses just the unemployment rate. Such a scenario would weigh on Australia’s largest sector – our households’ ability to consume – and limit our economy’s potential to grow over the coming years.

Our modelling suggests this would translate into a 0.6 per cent hit to GDP over four years, resulting in GDP being $12 billion lower in 2024.

While policy makers and economists were already beginning to consider the flow on effect of using underutilisation and hours worked rather than traditional singular markers of unemployment, the COVI-19 recession has sharpened the urgency around adopting these measures more broadly and applying them as a central benchmark in policy making, particularly when we consider the prospect that wages could fall in real terms if we’re not careful.

What can be done to lift wage growth, and will anything be done?

Driving wage growth requires policies designed to generate jobs but also boost aggregate demand to increase hours worked and lower underemployment.

This can be achieved by direct government spending (infrastructure spending and service delivery) or by pushing the economy faster and more productively, or both. Given the amount of fiscal stimulus already in the system, and with monetary policy already at ultra-accommodative settings, productivity enhancing reform is ever more important.

It is critical that the focus of fiscal policy shifts from respond and recover, to recover and reform. However, given the political cycle (and the prospect of a Federal election later this year) and the lack conversation about reform, it does not feel like we are gearing up for a hard-hitting reform budget this year. Reform may have to wait for 2022, and we may just face a period of very stagnant income growth as a result.

*Two scenarios for underemployment were considered in a modified RBA model based on its MARTIN wages and macroeconomic model. These included the underutilisation rate and EY’s estimate of the underutilisation NAIRU. First was a widening of the gap between unemployment and underemployment – similar to what occurred following the end of mining investment boom. The second was one where underemployment did not decline from its current level, while the unemployment rate continued its downward trend as expected by the RBA. Both scenarios produced similar results for weaker wages growth relative to the RBA’s baseline wages model.