The underemployment problem

Joey Moloney


October 28th, 2013

Joey Moloney looks into a worrying trend in the Australian economy.

It is unambiguous that the Australian economy is in an enviable state. All the key parameters—inflation, unemployment, growth—are more or less where we want them. However, where Australia falls away from being a world leader is in the often forgotten measurement of underemployment.

The Australian Bureau of Statistics (ABS) classifies anyone who is currently employed, but wants more work and cannot find it, as underemployed. It may be preferable to unemployment, but it is far from ideal, and it is trending in the wrong direction. A 2010 report from the OECD noted that Australia’s underemployment rate has been consistently high by international comparison. The graph below illustrates the underemployment rate in Australia from 1978 to the present.

Figure 1: The Australian underemployment rate (%). Source: ABS, Labour force, Cat. No. 6202

The high underemployment rate deserves attention for a range of compelling social and economic reasons.

On the individual level, research has found that underemployed workers are more likely to exhibit lower job satisfaction, higher job turnover, poorer mental and physical health and persistently lower income.

Furthermore, underemployed workers are overwhelmingly employed as casuals, with little job security and negligible superannuation accumulation. Banks are severely averse to extending credit to individuals lacking secure and sufficient income, which means underemployed individuals often find themselves trapped in rental accommodation with little chance of breaking into the housing market.

On the macroeconomic level, the underemployment rate has important implications for the level of underutilisation in the labour market. The ABS combines the unemployment rate and the underemployment rate to define the labour force underutilisation rate. Therefore, by focusing solely on unemployment, we cannot accurately assess whether or not we are efficiently utilising our available labour force.

Even with a low unemployment rate, we may be letting a large volume of hours, which workers are willing to put to use, go to waste. The following graph illustrates that the average employed person in Australia is working fewer and fewer hours.

Figure 2: Average monthly hours worked per employed person. Source: ABS, Labour force, Cat. No. 6202

Underemployment is not evenly dispersed across different demographics. For example, women are over-represented in the underemployment figures. The ABS data for August 2013 identified 964,300 underemployed workers, of which 566,800 (or 58.7%) were females. Overall, the rate for women was 9.8%, compared to 5.9% for men.

Another group hit hard by the growing trend is young people. In August 2013, the underemployment rate for people aged 15–24 was 14.8%, compared to an overall rate of 7.8%. To put that in historical context, in 1978 the youth underemployment rate was 3.0%, with an overall rate of 2.6%.

Aside from women and young people, other groups found to be at risk are immigrants, people with a lower educational attainment and individuals living in rural areas.

Furthermore, a highly significant factor was found to be labour market history. Research has established that workers who experienced underemployment in a previous period are considerably more likely to experience underemployment again. This path-dependency was equally true for males and females.

Since underemployment is best understood as people wanting to work more hours than they can currently attain, it represents a mismatch between the demand for, and supply of, hours of work. More specifically, it means there could either be an increasing trend in the amount of hours people want to work (demand), or a decreasing trend in the amount of hours available (supply).

The most plausible explanation is on the supply-side. The Australian economy has been going through a gradual shift in the composition of industries. Our manufacturing sector is shrinking while our services sector grows.

The services sector lends itself to casualisation much easier than the industrial sector, hence we have seen an effective doubling of the proportion of employees classified as part-time (<35 hours/week). The increased supply of part-time work relative to full-time work assists in explaining the reduction in the hours available to the typical worker represented above in Figure 2. The graph below illustrates the shift towards part-time employment.

Figure 3: Employees classified as part-time (% of total employed). Source: ABS, Labour force, Cat. No. 6202

An unsettling implication from the proliferation of part-time work and the demise of full-time work is in poverty statistics. A recent inquiry from the National Centre for Social and Economic Modelling posited that a person in a family in which the principal ‘bread-winner’ was employed part-time had a 17.0% chance of living in poverty, compared to a figure for full-time employment of 3.0%. Even more concerning is the fact that the former probability has grown from 11.4% in 2001.

The greater availability of part-time work also helps explain why women and young people are affected the most by underemployment. Both groups are much more likely to work part-time in labour-fluid service sectors such as retail and hospitality.

Moreover, an additional explanation for the startlingly high youth figure is that there is an increasing growth rate of tertiary educated young people, yet only a steady growth rate of full-time graduate jobs. Research has suggested this has led to many skilled young people drifting hopefully through casual job opportunities, rather than settle for a stable, lesser-skilled job with more hours.

This inference is complementary to a 2008 report from the Productivity Commission, which found that 36% of underemployed persons cited a lack of suitable job vacancies as the main barrier to sufficient employment.

As with a plethora of economic problems, any government intervention carries a high probability of unforeseen consequences. For example, incentives for firms to hire employees full-time could tempt higher unemployment. Unless there is growth in the aggregate supply of hours, increasing the supply available to a typical employee will simply muscle some out of employment.

The best avenue would therefore be to have the aggregate supply of hours grow at a rate superseding that of the total labour force. Such a dynamic would primarily be a product of population growth and the composition of economic growth, and is outside the scope of this article. Exactly how it would be accomplished is another matter entirely.

An important transition would be for discourse to revolve around how much people are working, and under what conditions, rather than to simply analyse the unemployment rate. Regardless of whether or not a silver-bullet solution for growing underemployment exists, an increasing number of people essentially expressing decreasing life satisfaction is an issue that deserves recognition and debate.

The views expressed within this article are those of the author and do not represent the views of the ESSA Committee or the Society's sponsors. Use of any content from this article should clearly attribute the work to the author and not to ESSA or its sponsors.

  • Sharon Lai

    Fantastic article Joey – I really enjoyed reading this. It certainly draws attention to an issue that is often overlooked in public discourse on Australia’s economic performance. What are our underemployment rates relative to other advanced economies? How about across industries (presumably industries dominated by women, youth, and part-time workers would have higher underemployment rates)? I’m curious about the manufacturing sector specifically, given that it has been declining for some time.

    I also wonder what your thoughts are on the two notable spikes in the underemployment rate shown in Figure 1. The first coincides with the early 1990s recession and the second coincides with the GFC. Surely a sign of firms reducing hours rather than letting go of employees?

    Interestingly, the 1990s recession was also accompanied by a significant spike in the unemployment rate, which rose by a larger proportion than the underemployment rate. Although the unemployment rate subsequently fell quite substantially, we have not seen an analogous recovery in the underemployment rate. Conversely, the GFC saw a larger rise in the underemployment rate than the unemployment rate. Do you have any thoughts on reasons behind the different dynamics between the unemployment rate and the underemployment rates during those two time periods?

    Finally, any plans to continue with the theme of underemployment in future articles? I think the issues raised in your final three paragraphs would be worth exploring further!

  • Joey Moloney

    Thanks for the feedback and good vibes Sharon!

    It’s definitely an interesting topic, particuarly because it seems, as the OECD concludes, that our underemployment rate is well above the average for developed economies.

    A lot of the research I came across had a sentence in the introduction stating some variation of ‘there is a shortage of research into underemployment and more is needed’. But I feel it is a safe-ish assumption to think that the shift toward services and away from manufacturing, combined with the higher rates for women and most of all young people, implies the problem must be most concentrated in retail and hospitality. I certainly observed many young people wanting more hours than they could get when I worked in a pub for a few years.

    I’m glad you noticed the massive climb and non-recovery from the 1990-91 recession, I found it most interesting. It seems the recession we had to have left us with underemployment we had to keep! I wanted to discuss it but couldn’t find an explanation

    My thoughts are that maybe during the 1990-91 recession firms stopped hiring (raising unemployment), and cut down the hours of the remaining workers (raising underemployment). Then when things got better, possibly they decided against raising the hours of the existing workers, and instead hired more workers at lower hours (consistent with the casualisation they had been embracing for ~15 years). Thus unemployment recovers but underemployment does not.

    Then maybe by the time the GFC came around, business was locked into the path of labour casualisation to the point where rather than necessarily completely halt hiring or lay off workers (leaving unemployment relatively undisturbed), firms just reduced their total supply of hours and casualised a bit more (meaning underemployment is hit harder).

    That’s my theory anyway, and it should probably come with a disclaimer that it is pure speculation, and I have nothing to back it up!

    I think I may have raised more questions than I’ve answered. The dynamics of our last two slumps, why Australia’s rate is particularly high (there are other economies transitioning from manufacturing to services that have lower rates), and what can be done to reverse the trend, all seem unclear. Maybe I’ll look into it a bit deeper next year.





    • Joey Moloney

      Haha I have no idea why it has double cheers’ed there! Please ignore.

      Also, I might just add that it would be my suspicion that underemployment would not be a hugely prevalent problem in manufacturing. It tends to lend itself to low-turnover, full-time workforces better. Factory work is completed most efficiently by a stable workforce showing up 9-5, whereas services have to accommodate unconventional hours and changing circumstances.

  • Sharon Lai

    Thanks for providing further insights Joey! This made me chuckle: “It seems the recession we had to have left us with underemployment we had to keep!”

    I would certainly be interested in reading further articles from you on this issue. In the meantime, well done on a fantastic year of writing and all the best with upcoming exams!

    • Joey Moloney

      Thanks mate! Well done to you on a fantastic year of making it all happen, and best of luck for exams as well.

  • Kim Liu

    Great article Joey — I definitely agree that we are in a most enviable condition, but at the same time, the strict, narrow ABS definition of unemployment leaves much to think about the wider labour market. Just to pipe in with Sharon, I would really like to see an exploration of “hidden unemployment” and the underemployment of skilled workers in these tight conditions; university graduates stuck in entry-jobs or undertaking undesired further education I think is a worrying trend, especially if they are pushing out the less skilled and/or reducing skill premiums.


    • Joey Moloney

      Cheers for the comment Kim!
      The skilled underemployment is an curious case. Whenever the discourse around the demise of manufacturing pops up, we hear that if we are to be a high wage economy, our manufacturing sector must be high-skill. Yet, here we are, with more skilled people than we seem to know what to do with.
      Then on the other hand, you do often hear that not enough people are studying science these days. So maybe, there is a high-skill manufacturing sector ready to take off, but just too many arts and the like graduates and not enough science graduates. Maybe, the underemployment of graduates is not so much a product of a general surplus of graduates, but a surplus of graduates with unspecific qualifications (I hold a BA so I honestly mean no offence to arts students!).
      Either way, this is a lot of speculation. Definitely another sign I should go deeper next year!

Founding sponsors




Gold sponsors



Silver sponsors