Chapter 1. Skill imbalances in Australia

International evidence

The 2016/2017 Manpower Talent Shortage Survey found that 40% of employers globally report having difficulty filling positions, an increase from the low of 31% during the global financial crisis in 2008. At 38%, the share of Australian employers reporting difficulties filling jobs is close to the global average, but is at the lowest percentage since 2007 when 61% of employers reported difficulties. The end of the commodity boom and the associated decline in demand for skilled labour is no doubt behind the reduction in hiring difficulties. The hardest jobs to fill are in skilled trades, and these have topped the list for the last ten years. Engineers, management executives, sales representatives, and IT staff are also identified as hard-to-fill positions. According to Manpower, employers complain that positions are hard to fill because candidates have a lack of experience (23% of employers report), there is a lack of available applicants (21%), or candidates lack technical skills (20%).

The Hays Global Skills Index provides another international comparison of skill shortages in national labour markets based on a range of quantitative data. The index combines data on wage pressures in the total economy and in high-skill industries and occupations, with indicators of labour market flexibility and the degree to which the qualifications of the unemployed align with employers’ needs. As with the Manpower Survey, the Hays Global Skills Index shows an average level of skill shortage pressure for Australia compared with other countries. The accompanying analysis flagged that there is evidence of significant wage pressure in high-skill industries in Australia, and rising mismatch between the skills unemployed workers possess versus those that employers are looking for.

Figure 1.1. Hays Global Skills Index, selected countries, 2017

Index values

Source: The Hays Global Skills Index 2017 (www.hays-index.com).

Evidence from the Department of Jobs and Small Business’ skill shortage research

According to Australia’s official assessment (Department of Employment, 2017[8]), skill shortages were limited in 2017, although many employers reported recruitment difficulties (i.e. cases where there is an overall adequate supply of skilled workers but employers are unable to recruit suitable workers due to lack of experience, specialist skills, or mobility barriers). There were 11.1 applicants per vacancy (Figure 1.2), but 37% of vacancies went unfilled due to difficulty recruiting suitable candidates. In terms of number of occupations in shortage, the Department of Jobs and Small Business (formerly the Department of Employment) found that 46% of consistently-assessed skilled occupations1 were in national shortage in 2017, compared with 90% in 2007.2

Figure 1.2. The proportion of vacancies filled has been in decline

Source: Department of Jobs and Small Business (2018), “The Skilled Labour Market: A pictorial overview of trends and shortages.” Survey of Employers who have Recently Advertised.

In 2017, shortages were concentrated in technician and trades worker occupations. A number of skilled trade occupations have been in continuous national shortage over the last decade, despite the end of the commodity boom and relatively weak employment growth in these occupations relative to total employment growth. Jobs and Small Business attributes continued national shortages in the skilled trades to several factors, including low participation in VET, and relatively poor pay and working conditions in some trades occupations (Department of Employment, 2018[9]). Only a small share of applicants for skilled trade vacancies are considered suitable due to lack of experience, qualifications or skills.

Despite stronger growth in employment in professional occupations (34%) than in trades (10%) over the decade since 2007, just 13% of professional occupations were in shortage in 2017, compared with 80% in 2007. Widespread national shortages were apparent for only five professional occupations in 2017: audiologist, sonographer, veterinarian, surveyor and architect. Rising enrolment in higher education has in part contributed to softer graduate outcomes and some surplus labour activity within professional occupations.

In addition to assessing skill shortages at the national level, Jobs and Small Business also investigates differences in employer recruitment experiences for skilled occupations between states and territories, and between metropolitan and non-metropolitan regions. There are marked differences between states and territories in the share of vacancies which go unfilled (Figure 1.3, Panel A). Labour demand strengthened across the country in 2017, but the strongest demand for labour was seen in the eastern states, with elevated levels of construction activity contributing to this. Western Australia and Queensland have experienced looser conditions since the end of the mining boom. There is also variation in recruitment difficulties within states and territories. Employers in the rural and remote regions of the country tend to attract fewer applicants and fill a lower proportion of vacancies than employers in urban areas. This gap has narrowed in recent years (Figure 1.3, Panel B), likely owing at least partially to declines in labour demand in regional areas due to the end of the mining boom. In 2017, six occupations were deemed to be in shortage in rural and remote regions of the country: medical diagnostic radiographer, optometrist, hospital and retail pharmacist, physiotherapist, midwife, and radio communication technician – almost all of which are in health professions.

Figure 1.3. Sub-national variation in proportion of unfilled vacancies

Source: Department of Jobs and Small Business (2018), “The Skilled Labour Market: A pictorial overview of trends and shortages.” Survey of Employers who have Recently Advertised.

Evidence from the OECD Skills for Jobs database

According to the OECD Skills for Jobs database (Box 1.2), many skills are in shortage in Australia, while only a few are in surplus – including knowledge of manufacturing and production, as well as physical abilities like fine manipulative abilities, control movement, and physical strength (Figure 1.5). But while several types of skills are in shortage, the intensity of the shortages of most skills is relatively low compared with other countries in the database. The most acute shortage pressure is evident in knowledge of education and training, health services and mathematics and sciences in addition to more transversal skills3 like verbal and reasoning abilities, basic process skills, systems skills and complex-problem solving skills.

Box 1.2. OECD’s Skills for Jobs database

Employer surveys often poll employers about their difficulty filling vacancies due to skill shortages (e.g. Manpower, Eurofound), which is a useful though imperfect measure of the degree to which skill shortages are widespread. While the employers’ perspective on skills shortages is important, it has limitations. Employers do not all interpret skill shortages in the same way, leading to inconsistency in employers’ self-reporting. Analysis shows that while many employers equate skills shortages with hard-to-fill vacancies, a substantial number do not (Green, Machin and Wilkinson, 1998[10]; Shah and Burke, 2005[3]). Some employers confuse skill gaps, i.e. the skill deficiencies of their internal workforce, with skill shortages. As a result, findings of employer surveys are often inconsistent with one another (OECD, 2017[11]).

The Skills for Jobs database, released in 2017, is an internationally-comparable index of skills in shortage and in surplus across countries. Unlike employer surveys, which can be subjective, the Skills for Jobs database relies upon quantitative information. The indicators measuring shortage and surplus are constructed using a multidimensional set of quantitative signals of skills pressure, including changes in wages, employment, unemployment, hours worked, and under-qualification (Figure 1.4).

Figure 1.4. Diagram of Skills for Jobs database method

In addition to relying on objective evidence, the Skills for Jobs database offers a portrait of which specific skills (e.g. complex problem solving, critical thinking, finger dexterity) are in demand or in surplus across countries. After first assessing shortage pressure at the occupational level (ISCO, 2-digit), a mapping from occupation to skill (using the O*NET classification) generates an assessment of shortage and surplus pressure for specific knowledge, abilities and skills. This is a departure from other international comparisons of skill shortages, which generally compare shortages across occupations, qualifications, fields of study, or by sector, but not by specific skills.

Figure 1.5. Skills shortage and surplus, Australia, 2016

Note: Positive values indicate shortages while negative values indicate surpluses. A Skill Needs indicator value of +1 represents the maximum value across countries in the database, while a value of -1 represents the lowest value. Basic Skills (Process) refer to those skills that contribute to the more rapid acquisition of knowledge and skill across a variety of domains (e.g. critical thinking, active learning, etc.). Basic Skills (Content) refer to foundational structures needed to work with and acquire more specific skills in a variety of domains (e.g. reading comprehension, listening, writing, speaking, basic math and science).

Source: OECD Skills for Jobs database 2017.

This profile of skills in shortage and surplus in Australia largely reflects global trends, like the rapid development of new technologies4 which has raised demand for skills that complement the use of these new technologies. Looking at a higher level of disaggregation than depicted in Figure 1.5, several types of technology-related knowledge are in shortage in the Skills for Jobs database across OECD countries: knowledge of computers and electronics, mathematics, physics, and engineering and technology. While these knowledge areas are also in shortage in Australia, the intensity of these shortages are smaller than in other OECD countries (see Figure 1.6, Panel A).

Shortages in technology-related knowledge are likely low in Australia due to a combination of demand- and supply-related factors. On the supply side, the OECD’s 2017 Science, Technology and Industry Scoreboard suggests that Australia has one of the lowest shares of tertiary students entering science, engineering and internet and communications technology (ICT) in the OECD area5. However, according to the PIAAC Survey of Adult Skills, Australian adults have above-average proficiency in problem-solving in technology-rich environments, with nearly 40% of Australian adults scoring in one of the top two levels. Therefore, while supply of specialised technical skills may be low, Australian adults have strong transversal skills related to the use of technology. On the demand side, the OECD’s 2017 Science, Technology and Industry scoreboard suggests that Australia performs well on some measures of technology adoption (e.g. Australia had the third highest rate of mobile penetration in the OECD in 2016), but lags behind on other indicators, including number of robots per worker, and ICT intensity of manufacturing and services jobs. Furthermore, the decline in mining activity has likely reduced the demand for engineers. Therefore, shortages in technology-related knowledge may be low due to a combination of a good supply of technology-related transversal skills, changes in industry composition, and relatively slow adoption of new technologies.

At the same time that technological progress raises demand for technology-related skills and knowledge, machines and computers change the nature of production by lowering the cost of performing routine tasks. This reduces the demand for human labour to perform these tasks. Knowledge related to manufacturing and production, an industry historically characterised by routine tasks, is in surplus in the Skills for Jobs database for all countries including Australia, which has seen several large automotive manufacturing companies shut down in recent years. But while most OECD countries also have large surpluses in knowledge related to building and construction, this is not the case in Australia (Figure 1.6, Panel B), where the construction sector is currently growing, and Jobs and Small Business expects it to be one of the largest contributors to employment growth over the next five years. Recall that Jobs and Small Business in its skill shortage research finds persistent shortages in several trades occupations, including a number of those related to construction.

Research to understand the impact that technological change could have on jobs and the tasks that humans perform at work has come to the conclusion that manual, non-cognitive, physical tasks will decline in demand, as these tasks are more cheaply performed by machines, while more cognitive and social tasks (not yet mastered by machines) will rise in demand (Autor, Levy and Murnane, 2003[12]). These trends are already playing out in Australia, where cognitive skills (e.g. reasoning abilities and complex problem-solving skills) and social skills are in shortage, and manual and physical skills, like fine manipulative abilities and physical strength, are in surplus (Figure 1.5). But relative to other OECD countries, shortages in cognitive skills are low in Australia (e.g. reasoning abilities, see Figure 1.7, Panel A), likely owing to an already highly-skilled and educated workforce. Surpluses in physical and manual skills are also low (e.g. finger dexterity; see Figure 1.7, Panel B), owing to ongoing demand for these skills in the skilled trades and construction sector.

Figure 1.6. Knowledge of Computers and Electronics and Building and Construction are in shortage

Note: Positive values indicate shortages while negative values indicate surpluses. A Skill Needs indicator value of +1 represents the maximum value across countries in the database, while a value of -1 represents the lowest value. Latest year of data available for Australia is 2016. All OECD countries in the database were included.

Source: OECD Skills for Jobs database 2017 (http://www.oecd.org/els/emp/skills-for-jobs-dataviz.htm).

Figure 1.7. Skills related to non-routine, cognitive tasks in shortage (e.g. Reasoning abilities); skills related to routine, manual tasks in surplus (e.g. Finger Dexterity)

Note: Positive values indicate shortages while negative values indicate surpluses. A Skill Needs indicator value of +1 represents the maximum value across countries in the database, while a value of -1 represents the lowest value. All OECD countries in the database were included.

Source: OECD Skills for Jobs database 2017. Latest year of data available for Australia is 2016.

Qualification mismatch

According to the 2017 Skills for Jobs database, about 35% of workers across OECD countries are mismatched by qualification. In Australia, 39% of workers are mismatched by qualification, putting the country just above the OECD average (Figure 1.9)6.

Figure 1.9. Incidence of qualification mismatch and field-of-study mismatch in Australia, and other OECD countries

Percentage of workers who are mismatched

Note: Most recent year available for each country. Data for Australia are for 2016. Field-of-study mismatch is calculated for all countries at the 2-digit ISCO level. See OECD (2017), Getting Skills Right: Skills for Jobs Indicators for details on methodology.

Source: OECD Skills for Jobs database 2017. Data for Australia come from the Australian Bureau of Statistics and are based on the two-digit ANZSCO classification.

About 20% of Australian workers are over-qualified for their occupations, while about 19% are under-qualified. The rate of under-qualification is on par with the OECD average, while the rate of over-qualification is slightly above the OECD average (20% vs. 17%). Over-qualification can arise if individuals are unable to find work in their field of study, and must downgrade to a job with a lower qualification requirement in order to find work.

Research using the Household Income and Labour Dynamics in Australia (HILDA) survey data studies how qualification mismatch varies by education pathway. Mavromaras (2009) differentiates between higher education graduates, VET graduates, Year 12 school graduates and those qualified below Year 12, and finds that VET graduates are the most likely to be well-matched by qualification in their jobs.

Apparent qualification mismatch

Over-qualification can be due to employers selecting candidates with higher qualifications than necessary to do a job, under the assumption that these candidates sport a set of unobservable qualities or skills that are likely to make them better workers (e.g. more disciplined, or productive). Empirical evidence suggests that firms do indeed employ such selection techniques (Bulmahn and Krakel, 2002[17]).

While employers often use qualifications as a proxy for a set of skills, and individuals acquire qualifications as a signal to employers that they have certain skills, qualifications are an imperfect indication of skills. Indeed, there can be considerable heterogeneity in the quality of skills acquired by individuals with the same qualification. When individuals are over-qualified or under-qualified for their jobs, but well-matched in terms of the actual skills needed to do their jobs, this is known as “apparent qualification mismatch.”

Evidence from the 2012 OECD Survey of Adult Skills indicates that among those Australians who were over-qualified for their jobs, 83.4% had the right level of literacy skills for the job, while only 13.4% had a higher level of literacy skills than was required (Table 1.1). The degree of apparent mismatch varies by field of study. Apparent qualification mismatch is highest in services, health and welfare, social sciences and engineering, manufacturing and construction, where qualifications seem to be inflated relative to the underlying literacy skills of the workers.

Apparent qualification mismatch of this type points to the low capacity of qualifications to accurately signal skill quality. Of course, qualifications may more successfully signal skills other than literacy skills. For instance, a degree in social work may do a good job of signalling human empathy skills, but a poor job of signalling literacy skills.

Field-of-study mismatch

Evidence from the OECD Skills for Jobs database suggests that field-of-study mismatch is quite common across OECD countries, with 32% of workers, on average, employed in a field that is different from the one they have studied and specialised in (Figure 1.9).

While Australia’s level of field-of-study mismatch is on par with the average (33% relative to 32%), the overlap between field-of-study mismatch and over-qualification is somewhat high in Australia. About 40% of workers who are mismatched by field-of-study are also overqualified for their job, compared to an average of 33% among countries surveyed. The OECD estimates that such workers face 20% lower wages than workers who are well matched. Analysis using the HILDA Australian panel survey finds that the wage penalties for workers who are mismatched both by field of study and by qualification level may be even higher for Australian workers: 30% for men and 25% for women (Mahuteau et al., 2015[18]).

The incidence of field-of-study mismatch varies by field of study in Australia, with 80% of graduates from Arts and humanities working in a field unrelated to their studies, compared to only about a quarter of graduates from Social sciences and Services; and less than a third of graduates from programs in Engineering, manufacturing and construction; Health and welfare; and Education (Figure 1.10). Field-of-study mismatch in Arts and humanities is higher in Australia than the average (69%), which could indicate lower labour demand in these fields relatively to elsewhere, or perhaps students with these skill sets face better job prospects or higher pay in other fields.

Figure 1.10. Field-of-study mismatch by field of study, Australia and OECD countries, 2016

Share of workers employed in an occupation unrelated to their field of study

Source: OECD Skills for Jobs database. The OECD average includes 24 OECD countries in the database.

Both field-of-study mismatch and qualification mismatch also vary by immigration status in Australia. About 30% of foreign-born tertiary-educated adults are over-qualified, compared with just over 20% of the native-born who are tertiary-educated (PIAAC, 2012). Foreign-born adults are also more likely to be mismatched by field of study: 60% work in a field other than the one in which they studied, compared with only 35% for the total population (ABS, 2015[19]). However, those adult migrants who obtained a non-school qualification after arrival in Australia had a similar level of field-of-study mismatch to the total population, suggesting that foreign credentials may not be well-recognised by employers.