Subscribe

Job ads and vacancies: parting company

A puzzle is emerging in the labour market data with the divergence of the number of job advertisements and job vacancies.

Click image to zoom Tap image to zoom

Until recently, job ads and vacancies tended to be correlated and tracked the unemployment rate. Historically, ANZ’s job ads series - released monthly rather than quarterly and a month ahead of the Australian Bureau of Statistics’ (ABS) vacancies series - provided a more timely and regular read on the state of the labour market.

"An assessment by Indeed found the jobs mismatch is higher in Australia than in comparable countries."

However, between May 2018 and February 2019 job ads fell by 5.6 per cent while vacancies increased by 3.0 per cent. Over the same period, the unemployment rate fell by 0.4 percentage points - a reversal of its historically inverse relationship with job ads.

The gap between ads and vacancies has been widening markedly for around two years. Although vacancies fell in May 2019 for the first time in three years, the gap with job ads continued to expand.

Click image to zoom Tap image to zoom

To try determine what was causing the recent divergence in the two series, ANZ Research looked at three possible explanations:

  • Less efficient labour market matching - Are employers having less success finding suitable applicants to fill vacant positions?
  • Increased presence of labour hire agencies - Are employers relying more on labour hire agencies to recruit workers?
  • Changing recruitment preferences - Are employers shifting away from “advertise and apply” towards a “find and engage” approach, using sites such as LinkedIn?

Mismatched

Both vacancies and job ads (as a share of the labour force) have strong inverse relationships with the unemployment rate however job ads have recently bucked this trend by falling at the same time as the unemployment rate.

Although vacancies have maintained the historical relationship, the curve has shifted outward. Increased participation is likely playing a part but the shift suggests the labour market is becoming less efficient at matching the supply of and demand for workers. This suggests the mismatch between the skills required and the skills of job seekers is increasing. An assessment by Indeed found the jobs mismatch is higher in Australia than in comparable countries.

Click image to zoom Tap image to zoom

If employers are struggling to attract and engage workers with suitable skills to fill vacant positions, they may be less willing to invest money and resources in advertising for and processing applicants. If this is the case, ANZ’s job ads series may actually be less reflective of labour demand.

A good test case for this theory is the construction industry which has long complained of shortages in skilled labour and expertise. Data from a survey from the Department of Employment, Skills, Small and Family Business’s confirms this problem.

For construction trades in 2018, only 44 per cent of vacancies were filled and, on average, there was only one suitable applicant per vacancy. That outcome is worse than at any time during the mining boom or residential construction boom.

In 2017-18, engineering occupations saw the poorest results in six years with 59 per cent of vacancies filled and an average 2.3 suitable applicants per vacancy.

ABS data show construction job vacancies relative to employment have increased by more than half over the three years to May 2019 and are now at a rate not seen outside the mining boom. The steepest growth in the number of vacancies (+33.1 per cent year-on-year) was in the year to February 2019, at the same time that construction employment contracted by 4.0 per cent. Over the same period, a proxy for construction job ads was down 9.6 per cent.

This disparity suggests less efficient labour market matching and skills mismatches could explain some of the divergence in vacancies versus ads.

Labour hire

If the use of labour hire is increasing relative to job ads, this could undermine the ANZ Job Ads series. However, this theory is not supported by the ABS or Department of Employment, Skills, Small and Family Business data.

The ABS’ Characteristics of Employment survey reports little movement in the shares of workers being paid by or registered with labour hire firms between 2014 and 2018. The Department of Employment’s Survey of Employers’ Recruitment Experiences also does not show any upward trend in the share of vacancies which employers used labour hire companies to recruit for (along with recruitment agencies/Government employment services).

Click image to zoom Tap image to zoom

Furthermore, private labour hire company data does not appear to explain the divergence between vacancies and ads, particularly over the past year. Although between mid-2017 and mid-2018 there was a ramp-up in labour hire jobs relative to job ads, this has not continued.

Find and engage

If employers are becoming more likely to “find and engage” suitable applicants to fill their vacancies, including via social media sites such as LinkedIn, rather than use the traditional “advertise and apply” approach, ANZ’s job ads may not be capturing this trend.

One test of this theory is to look at a broader measure of online recruitment, the Sunsuper Australian Job Index. This tracks “digital job ads on job boards, employer career portals and recruitment company websites from over 4,000 sources”. However, this index has been largely flat since Q1 2018 and does not show any recent correlation with the vacancies series.

Another test is to look at the use of social media in recruitment. Department of Employment, Skills, Small and Family Business data show the use of social media as a recruitment tool has soared from 1 per cent of vacancies in 2012-13 to 11 per cent in 2017-18.

However, this is still very low compared with the internet (excluding social media), which was used as a recruitment method for 58 per cent of vacancies in 2017-18.  Additionally, there has been a clear upward trend in the use of the internet in recruitment over the past five years.

On the whole, it appears the shift towards “finding and engaging” potential employees on social media, via recruitment companies, or using other digital tools may be a partial contributor to the divergence of job ads and vacancies data, as these methods would not be captured in ANZ’s series.

Click image to zoom Tap image to zoom

Missing puzzle pieces

While these trends may contribute to the divergence to varying extents, ANZ Research doesn’t think they fully explain it. Although not fully explored, other factors could also be playing a role including:

  • Changes in market share in online recruitment

Both ANZ Research’s job ads and the Internet Vacancy Index (IVI) use SEEK and Australian JobSearch to measure online ads and the IVI also uses CareerOne.

According to a 2018 report from IBISWorld, SEEK accounts for more than 85 per cent of online recruitment industry revenue in Australia. However, Indeed, the largest job search engine globally, which aggregates ads from various online and offline sources, has expanded its presence in Australia and a future launch of Google for Jobs in Australia could also impact the market.

  • Other recruitment methods

According to the Department of Employment, Skills, Small and Family Business, employers using word of mouth and/or directly approaching applicants jumped, as a share of vacancies, by 6 percentage points to 32 per cent in 2017-18. However, the Department advises not to rely on year-to-year changes.

Time will tell whether this turns into a trend. It is possible that changes in the degree of internal recruitment and upskilling might be affecting the data, but this is difficult to ascertain.

  • Skilled migration.

When employers find it hard to recruit domestically, they turn to skilled migration. However, at a high level, employer demand for skilled migrant workers appears to have weakened recently.

There were fewer first-stage applications (as an indicator of employer demand) received in the employer sponsored category for skilled visas in 2017-18 compared with 2016-17. It is possible this could mask differences by industry/occupation/region or challenges caused by changes in listed occupations, visa conditions and application processing times, rather than reflecting a change in employer demand for skilled migrant workers.

ANZ Research will continue to monitor movements in the Job Ads and Job Vacancies series and how they track against the unemployment rate and employment growth.

Job vacancies fell for the first time in three years in the May quarter, so it will be interesting to see if there is a further decline in the August quarter.

Catherine Birch is Senior Economist at ANZ

The views and opinions expressed in this communication are those of the author and may not necessarily state or reflect those of ANZ.

editor's picks

13 Sep 2018

The other side of the infrastructure boom

Daniel Gradwell | Associate Director - Property, ANZ

Australia’s $120bn in construction projects a boon for the economy but rising cost pressures a real risk.

04 Oct 2018

Construction concern

Daniel Gradwell and Joanne Masters | Senior Economists at ANZ Research

Housing work under construction will support the market in the near term but the outlook for 2020 is darker.