







































AUSTRALIAN POPULATION STUDIES  2021 | Volume 5 | Issue 2 | pages 29–42 

 

© Fry et al. 2021. Published under the Creative Commons Attribution-NonCommercial licence 3.0 Australia  
(CC BY-NC 3.0 AU). Journal website: www.australianpopulationstudies.org 

COVID-19 and the Australian labour 
market: how did older Australians    
fare during 2020? 

Jane Fry  The University of Melbourne 

Jeromey B. Temple*  The University of Melbourne 

Peter McDonald  The University of Melbourne 

Alysia Blackham  The University of Melbourne 

*Corresponding author. Email: Jeromey.Temple@unimelb.edu.au. Address: Melbourne School of 

Population and Global Health, The University of Melbourne, Melbourne, VIC, 3010, Australia 

Paper received 29 June 2021; accepted 25 October 2021; published 29 November 2021 

Abstract 

Background 

In analysing the effects of the COVID-19 pandemic on the labour market, attention has focussed on 

younger people, leaving a research gap when it comes to outcomes for older Australians aged 50 

years or over, in terms of employment, unemployment, underemployment and hours worked. 

Aims 

To describe levels of labour force participation, unemployment, underemployment, and hours 

worked by older workers and job seekers during 2020. 

Data and methods 

Using Australian Bureau of Statistics data, we perform descriptive analyses of variations in labour 

market outcomes by geographic areas, public and private sector employment, industry of 

employment and demographic characteristics. 

Results 

Older employment fell in April but recovered by December. As the full-time share initially increased, 

average hours worked decreased due to reductions in hours offered to workers, increasing the 

underemployment rate. There was little recovery of employment in metropolitan Melbourne due to 

prolonged lockdown conditions. Of the largest industries, retail trade and manufacturing were worst 

affected. 

Conclusions 

By December 2020, employment levels for older workers in some sectors had recovered from the 

initial downturn caused by the pandemic. However, for older workers in some industries, there is a 

major concern about their potential for future employment. 

Key words 

Older workers; COVID-19; labour market; retirement. 

 

  

   

http://www.australianpopulationstudies.org/
mailto:Jeromey.Temple@unimelb.edu.au


30 Fry et al. Australian Population Studies 5 (2) 2021 

 

 

1. Introduction 

As of March 2021, approximately one third of all COVID-19 cases in Australia were among people 

aged 50 years and over, with this same group accounting for 99% of COVID-19 deaths, the majority of 

which were aged over 70 years (Department of Health 2021). In terms of individual health and 

mortality, COVID-19 has heavily impacted older Australians (typically aged 50 years or over), while 

younger Australians (defined here as those aged 15–24 years) have faced particularly deleterious 

effects on their mental health and general wellbeing (Varma et al. 2021). Although younger 

Australians have faced considerable increases in unemployment and underemployment (Churchill 

2020), little attention has been given to how older workers and job seekers have fared in 2020 during 

the COVID-19 pandemic. 

Ordinarily, we would expect older workers to fare worse, as they do in other recessions, with 

reduced employment opportunities (Chomik 2020), increased age discrimination in recruitment, and 

steep wage losses (Bui et al. 2020), especially given they are more vulnerable to the effects of COVID-

19. That said, this may not be the case, because younger workers are more casualised and therefore 

more easily laid off (Borland & Charlton 2020). There remains uncertainty as to the long-term effects 

on particular groups of workers, particularly on the basis of gender, age, and in particular sectors 

(Birch & Preston 2020) as shutdowns have led to job loss, unemployment and inactivity (Chomik 

2020). 

In this paper we address this research gap through a detailed examination of labour force 

participation, unemployment, underemployment, and hours worked by older workers (aged 50 years 

or over) and job seekers during the COVID-19 pandemic in 2020. At the urging of unions and some 

business groups, the Australian Government introduced the JobKeeper wage subsidy scheme to help 

keep more people employed in private businesses and not-for-profit organisations. A number of 

OECD countries have introduced similar job retention (wage subsidy) schemes, such as New Zealand, 

Canada, Estonia, Poland and the Netherlands (OECD 2020). Under JobKeeper, eligible employers 

were required to pay a minimum of $1,500 per fortnight to the employee (Forsyth 2020). JobKeeper 

objectives were to support business and job survival, preserve employment relationships, and 

provide needed income support (Neil et al. 2021). It continued throughout the period of observation 

used in this paper, March to December 2020. The effects of the JobKeeper payments on employment 

are an important backdrop to the paper given that JobKeeper essentially disguises unemployment 

(Coates et al. 2020). 

This research complements the detailed evidence base on older labour force participation, including, 

but not limited to, the determinants, barriers and enablers of older employment, the role of 

workplaces and the efficiency of government policy in enabling participation and futures of labour 

supply (OECD 1998; Adair & Temple 2012; Yeatts et al. 2000). Apart from providing a complementary 

view to previous studies, our analysis underscores the impacts of COVID-19 on older workers. 

  



Australian Population Studies 5 (2) 2021 Fry et al. 31 

 

2. Data and methods 

In this paper, we use population-weighted Australian Bureau of Statistics (ABS) data tables based on 

the Labour Force Survey. This survey of approximately 50,000 individuals in 26,000 dwellings is 

conducted monthly, with households interviewed for eight consecutive months. The nationally 

representative sample rotates with one eighth of the households replaced each month. In scope 

individuals are civilians aged 15 years and over. In addition to socio-demographic information, 

questions relate to labour market activities in a ‘reference week’ just prior to the survey. More 

information on the sample and survey methodology are available elsewhere (ABS 2021c), as are the 

underlying data tables (ABS 2021b).  

Within the ABS Labour Force Framework there are seven key concepts for analysing labour market 

performance:  

1. Population is identified as the number of usually resident civilians in the relevant age/sex group. 

2. Employment is defined as engaged in full-time or part-time work. 

3. Hours worked (in our context) is hours actually worked in all jobs. 

4. Underemployment is defined as either: 

a) employed full-time but worked less than 35 hours in the reference week due to economic 

reasons and available to work more hours, or  

b) employed part-time and would prefer more hours and available to work more hours. 

5. Unemployment is defined as not employed but waiting to start a new job or actively looking for 

work.  

6. The labour force comprises individuals who are employed or unemployed.  

7. Not in the labour force (NILF) comprises individuals who are neither employed nor unemployed. 

Further details are available elsewhere (ABS 2018). 

These concepts are then used to calculate six summary measures:  

1. Employment to population ratios divide the number employed (full-time or part-time) by the 

population.  

2. Labour force participation rates divide the labour force by population. 

3. Unemployment rates divide the number unemployed by the labour force. 

4. Underemployment rates divide the number underemployed by the labour force. 

5. Hours per worker divides total hours of actual work by the number employed. 

6. Full-time share divides the number of full-time employed by the total number employed. 

We refine our aggregate analysis to consider geographic areas, public and private sector 

employment, industry of employment and demographic characteristics. 

Summary measures are provided for March 2020 (or the February quarter), the initial period, which 

compares March and April (or the May quarter) to gauge the initial effect; and the longer period, 

which compares March with December (or the November quarter) to gauge the 2020 effect. These 

effects are percentage changes for selected concepts (such as the number employed) or percentage 

point changes for summary measures (except hours per worker, where we use percentages). 

Ethics approval for this project was granted by The University of Melbourne Human Ethics 

Committee LNR 2A — Ethics ID 2021-21421-15592-2. 



32 Fry et al. Australian Population Studies 5 (2) 2021 

 

 

3. Results 

3.1. Aggregate effects 

Initially, young people fared the worst: their employment rate fell by more than six points (Table 1). 

Most who lost their jobs withdrew from the labour force, cushioning the effect on the 

unemployment rate. Job losses occurred despite the JobKeeper program, for which there were 

900,000 applicants per month up to September, then 500,000 applicants per month through the 

remainder of the year. A recovery by December left the employment rate at March levels, although 

an increase in participation raised the unemployment rate as more individuals searched for work. 

Initially, hours worked dropped due to the combined effects of a loss of part-time jobs and reduced 

hours for remaining workers. This served to increase underemployment among the youngest 

workers. In the longer period, hours worked more than recovered, and desired and actual hours 

worked were in balance. 

Table 1: Effect of COVID-19 on employment and participation by age group, Australia, 2019 and 2020 

 March 2020 March-April 

2020 change 

March-April 

2019 

change 

March-

December 

2020 change 

March-

December 

2019 change 

Age 15-24 years      

Participation rate 68.9% -6.4 points -0.3 points +1.8 points +2 points 

Employment rate 60.2% -6.5 points +0.3 points +0.6 points +2.6 points 

Unemployment rate 12.7% +1.5 points -0.9 points +1.3 points -1.2 points 

Full-time share 43.7% +1.5 points -1.3 points -0.2 points +0.2 points 

Hours per worker 24.9 hours -16.4% -2.0% +4.6% +1.7% 

Underemployment rate 18.1% +4.5 points +0.8 points +0.4 points +3.5 points 

Age 25-49 years      

Participation rate 85.7% -2 points +0.3 points +0.4 points +0.6 points 

Employment rate 82.1% -2.8 points +0.3 points -0.2 points +1 point 

Unemployment rate 4.3% +1.1 points 0 points +0.6 points -0.5 points 

Full-time share 75.5% +0.6 points -0.2 points +0.1 points +0.2 points 

Hours per worker 34.3 hours -11.4% -1.8% +3.4% +2.2% 

Underemployment rate 6.7% +5.5 points +0.3 points +0.7 points -0.1 points 

Age 50+ years      

Participation rate 45.0% -1.5 points 0 points +0.4 points 0 points 

Employment rate 43.1% -1.6 points 0 points +0.1 points +0.1 points 

Unemployment rate 4.1% +0.4 points +0.1 points +0.6 points -0.2 points 

Full-time share 66.2% +1.3 points -0.3 points +0.8 points -0.2 points 

Hours per worker 32.5 hours -9.7% -0.3% +2.7% +2.2% 

Underemployment rate 

(aged 55+ years) 

6.9% +4.3 points +0.2 points -0.4 points +0.5 points 

Source: ABS Labour Force Survey. Note: The age group used for underemployment differ from those of the other measures 
because of the availability of data. Points refers to percentage points. 

Among individuals aged 25–49 years there was also a marked reduction in participation and 

employment in the initial period. Hours per worker fell substantially (11%) and underemployment 

rose. However, by December labour market conditions were almost the same as in March, although 



Australian Population Studies 5 (2) 2021 Fry et al. 33 

 

hours worked were higher. At first, the fall in the employment rate among older workers was 

matched almost exactly by a fall in participation. Initially, therefore, workers losing their jobs defined 

themselves as being out of the labour force. This meant the rise in the unemployment rate was much 

smaller than may have been expected. In the longer period, the fall in employment for older workers 

was fully compensated, but the participation rate rose more than the employment rate. Thus, in the 

longer period, the unemployment effect was worse than it had been in the initial period. 

Participation was higher in December than in March but does not appear to be seasonal based on a 

comparison with the March–December 2019 change. 

A reduction in part-time jobs (many held on a casual basis) led to a small increase in the full-time/ 

part-time job share. However, average hours per worker declined as workers initially faced reduced 

hours. In particular, fewer casual workers in food, hospitality and other service industries were 

required due to sector-specific closures and limits on social gatherings. Some employers reduced the 

hours offered to employees rather than laying them off. Indeed, for employees aged 55 years and 

over, a comparison between August 2019 and 2020 shows a large increase in employees citing not 

enough work as a reason for working fewer hours than usual: a three-fold increase for employees 

with paid leave entitlements and a 50% increase for employees without paid leave (casual) (ABS 

2021a). 

3.2. Distributional results: geographical areas 

The initial effect of COVID-19 on the participation of older workers was similar across all the 

geographies shown in Table 2, but was a little higher in Western Australia where the prevalence of 

COVID-19 was lower. This suggests that people who lost jobs initially reacted to the pandemic 

announcement by leaving the labour force. The same effects are evident for employment at the 

state level and for regional Australia, with the fall in employment roughly matching the fall in 

participation — implying minimal impact overall on unemployment. However, regional Victoria 

differs from other areas in that, while participation also fell, employment fell much less than 

elsewhere. The implied reduction in unemployment is consistent with the reduction in the 

participation rate.  

Table 2: Effect of COVID-19 on the employment and participation of workers aged 50 and over, selected 

states, metropolitan Melbourne and regional Victoria, 2020 

 Participation rate Employment rate 

March 

2020 

March-April 

change 

March-

December 

change 

March 

2020 

March-April 

change 

March-

December 

change 

New South Wales 43.4% -1.3 points +1.1 points 41.6% -1.5 points +0.5 points 

Western Australia 49.0% -1.6 points +0.2 points 46.4% -1.5 points +0.1 points 

Victoria 45.7% -1.3 points 0 points 44.0% -1.1 points -0.3 points 

Metropolitan 

Melbourne 

46.2% -1.2 points -0.5 points 44.6% -1.3 points -1.0 points 

Regional Victoria 44.3% -1.4 points +1.1 points 42.7% -0.5 points +1.4 points 

Regional Australia 42.8% -1.3 points +0.4 points 41.1% -1.2 points +0.6 points 

Source: ABS Labour Force Survey. Note: Points refers to percentage points. 

  



34 Fry et al. Australian Population Studies 5 (2) 2021 

 

 

In the longer period, as might have been expected given the duration of the second lockdown in 

metropolitan Melbourne, there was little recovery of employment in metropolitan Melbourne 

while there was strong recovery everywhere else, especially in regional Victoria. While population 

levels remained stable between metropolitan Melbourne and regional Victoria, metropolitan 

Melbourne’s share of employment declined, suggesting jobs may have shifted to regional Victoria . 

3.3. Employment by age group and sector 

Considering individuals aged 45 years or over, initially the numbers employed fell for all of the 

older age groups in both the public and the private sectors, but the largest percentage falls tended 

to be in the older ages 60-64 and 65+ years, especially in the public sector (Table 3). In the longer 

period, in the public sector, recovery was very strong for all age groups, especially among 50–59 

year olds. In the longer period, in the private sector, there was recovery for the two oldest age 

groups. This may mean that, unlike in previous recessions, 2020 may not lead to a growth in early 

retirement (Rutledge & Coe 2012). 

Table 3: Numbers employed and percentage changes in numbers employed by age group and sector, 

Australia, 2020 

 Public sector Private sector 

 No. employed % change No. employed % change 

Age group March 2020 

(‘000) 

March-

April 

March-

December 

March 2020 

(‘000) 

March-

April 

March-

December 

45-49 255.9 -2.9 +2.2 1137.0 -3.9 -2.8 

50-54 220.5 -5.9 +5.8 1029.9 -3.1 -0.5 

55-59 204.4 -3.1 +8.4 923.9 -3.8 -2.8 

60-64 150.6 -8.1 +1.7 645.2 -1.7 +2.6 

65+ 80.6 -14.3 +1.4 517.8 -9.3 +3.1 

Source: ABS Labour Force Survey. 

3.4. Employment by industry 

Between February and May, employment of older workers fell by between 4 and 9 per cent for the 

six largest industries of employment (Table 4). However, in the longer period, the outcomes for 

these six industries varied considerably, rising by 10.4 per cent for education and training but 

falling by 11.2 per cent for manufacturing. Retail also showed a heavy fall in the longer period. For 

the other three largest industries, employment returned to its February level by November. 

For the remaining industries, there were some for which the February to May effect was positive 

(construction (perhaps boosted by the ‘HomeBuilder’ program); agriculture, forestry and fishing; 

wholesale trade; financial and insurance services; rental, hiring and real estate services; electricity, 

gas, water and waste services). COVID-19 gave rise to issues of distribution which may account for 

the strong positive result for wholesale trade. There were strong negative initial outcomes for arts 

and recreation services and for accommodation and food services, due in part to spatial distancing 

measures (Coates et al. 2020).  

Between February and November, sectors outside the top six that had positive initial outcomes 

had — not unexpectedly — positive longer-term outcomes as well. That said, there were some 



Australian Population Studies 5 (2) 2021 Fry et al. 35 

 

strong reversals for a few industries that had negative initial outcomes. For example, employment 

in accommodation and food services swung from an initial fall to a longer period gain. 

Administrative and support services, accommodation and food services and mining also showed a 

large turnaround from negative to positive from the initial period to the longer period 

Table 4: Numbers employed and percentage change in numbers employed by industry, workers aged 50+ 

years, Australia, 2020 

Industry (in rank order of employment) Number employed 

February 2020 

(‘000s) 

Percentage change in number 

employed 

February to 

May 2020 

February to 

November 2020 

Health care and social assistance 597.8 -4.0 -2.2 

Education and training 345.9 -3.6 +10.4 

Professional, scientific and technical services 305.1 -5.3 -0.4 

Retail trade 300.3 -8.9 -6.7 

Manufacturing 295.4 -6.9 -11.2 

Public administration and safety 292.1 -4.6 +0.1 

Construction 288.3 +1.1 +5.5 

Transport, postal and warehousing 260.6 -15.9 -6.3 

Agriculture, forestry and fishing 174.9 +6.3 0 

Other services 139.5 -16.5 -3.2 

Wholesale trade 126.8 +10.9 +7.8 

Administrative and support services 125.3 -3.5 +12.6 

Accommodation and food services 123.5 -11.4 +12.7 

Financial and insurance services 115.0 +12.5 +7.5 

Rental, hiring and real estate services 69.7 +2.1 +8.2 

Arts and recreation services 62.6 -20.2 -3.6 

Mining 60.6 -9.0 +12.7 

Information, media and telecommunications 53.4 -23.5 -6.7 

Electricity, gas, water and waste services 36.1 +36.0 +41.2 

Source: ABS Labour Force Survey. Note: The periods used in this table differ from those of other tables because of the 
availability of data. 

3.5. Employment by demographic characteristics 

Initially, individuals aged 50–64 years from North-East Asia were worst hit by declining 

employment rates, although they recovered well (Table 5). Individuals aged 50–64 years from the 

Americas did not fare well, with an initial reduction of 3 percentage points followed by further 

decline in the longer period. Among those aged 65 years or over, highest employment rates were 

for individuals from Oceania/Antarctica and Sub-Saharan Africa. Alongside North Africa/Middle 

East and the Americas, they have the largest initial reductions in employment. However, rates for 

Oceania and the Americas recovered well, increasing by 2 and 6 percentage points respectively 

over the longer period. 

  



36 Fry et al. Australian Population Studies 5 (2) 2021 

 

 

Table 5: Percentage employed and changes in percentage employed by age group by country of birth, sex, 

partner status, relationship status and education, Australia 2020 

Characteristic 50-64 65+ 

Mar 2020 

% empl’d 

Mar-Apr 

change 

Mar-Dec 

change 

Mar 2020 

% empl’d 

Mar-Apr 

change 

Mar-Dec 

change 

Country of birth       

Australia 71.3 -1.5 pts -0.1 pts 15.9 -1.1 pts 0.3 pts 

Oceania and Antarctica 72.6 -2.9 pts 3.1 pts 21.9 -3 pts 2.1 pts 

North-West Europe 74.3 -2.9 pts -0.1 pts 13.3 -1.6 pts -0.1 pts 

Southern and Eastern Europe 64.7 -2.2 pts -2.8 pts 8.2 -1.2 pts 0.2 pts 

North Africa and the Middle East 41.2 -1.4 pts 2.4 pts 9.8 -3.5 pts -0.2 pts 

South-East Asia 64.4 -2.7 pts -0.6 pts 16.4 0.9 pts -0.1 pts 

North-East Asia 54.0 -6.8 pts 3.3 pts 8.3 -1.6 pts 3.9 pts 

Southern and Central Asia 72.9 -1.1 pts -2.7 pts 17.8 1.1 pts 2 pts 

Americas 71.1 -3 pts -4.1 pts 15.8 -3.1 pts 6 pts 

Sub-Saharan Africa 79.1 -1.7 pts -5.4 pts 19.3 -2.6 pts 0.3 pts 

Sex       

Males 74.9 -1.7 pts -0.4 pts 17.7 -1.3 pts 1.3 pts 

Females 64.8 -2.2 pts 0.3 pts 10.9 -0.9 pts 0.1 pts 

Partner status       

Partnered 73.3 -2.3 pts 0.1 pts 16.5 -1.3 pts +1.9 pts 

Not partnered 60.8 -1.2 pts -0.5 pts 10.4 -0.9 pts -1.5 pts 

Relationship status       

Has spouse only 69.1 -2.5 pts -0.4 pts 16.1 -1.6 pts 1.4 pts 

Has dependent children 76.7 -2.5 pts 0.3 pts 18.0 -0.5 pts 2.2 pts 

Lone person 61.2 -0.5 pts 1 pts 12.1 -0.9 pts -2.5 pts 

Other 51.3 -0.6 pts -2.7 pts 6.0 -0.3 pts 0.6 pts 

Education       

Bachelor and above 78.3 -2.5pts 0.5 pts 24.6 -3.1 pts +0.8 pts 

Diplomas, Advanced Diplomas 76.8 -3.3pts -2.4 pts 19.2 +1.1 pts -0.5 pts 

Certificate III/IV 75.9 -2.4pts -0.1 pts 17.7 -2.5 pts -0.7 pts 

Year 12 68.2 -4.3pts -1.3 pts 14.2 -2.1 pts +1.3 pts 

Below Year 12 56.2 -2.1pts -0.3 pts 10.2 -1.2 pts -1.1 pts 

Source: ABS Labour Force Survey. Note: Other relationships include non-dependent child, other related individual, non-
family members and visitors. Pts refers to percentage points. 

The initial reduction in the employment rate fell disproportionately for females and partnered 

individuals aged 50–64 years (Birch & Preston 2021), although these older individuals were not 

disadvantaged in the longer period (however, Wood et al. (2021) find women’s overall 

employment remains below pre-COVID-19 levels). Hidden behind the overall recovery in 

employment rates in the longer term was a gain for males and partnered individuals aged 65 years 

or over. Hardest hit by the initial reduction in employment were individuals aged 50–64 years with 

a spouse and/or dependent children. However, their prospects improved over the longer period. 

By December, the largest losses of employment were among lone persons aged 65 years or over   

(-2.5 points), but older individuals with dependent children increased their employment by 2.2 

points. 



Australian Population Studies 5 (2) 2021 Fry et al. 37 

 

Among those with identified education levels and aged 50–64 years, all experienced an initial 

decline in employment, particularly individuals with Year 12 (which concords with ISCED 3, upper 

secondary) or Diplomas and Advanced Diplomas. Individuals with Diplomas and Advanced 

Diplomas were particularly disadvantaged as their employment rates did not recover by December. 

This is an occupational effect associated with the decline in roles in the health care and social 

assistance sector. Among those aged 65 years or over, the largest initial reduction in employment 

is for holders of Bachelor’s Degrees and above, associated with sectoral contractions in health care 

and social assistance, education and training, and professional, scientific and technical services. 

However, driven by growth in education and training, this group recovers over the longer period. 

4. Discussion 

The impacts of COVID-19 on Australian employment in 2020 exacerbated and exposed inequalities in 

the labour market (Birch & Preston 2021). Initially, employment rates fell between March and April 

2020 for almost all groups, regions and industries included in the study. Young people were hardest 

hit due to their overrepresentation in casual employment and in service industries hardest hit by 

lockdown restrictions. However, older workers also suffered. Overall, there were initial period 

negative effects as older workers exited the labour force, but the economic recovery and policy 

responses such as JobKeeper (with uptake of 3.8% in December) and working from home appear to 

have ameliorated the longer period effects. The pandemic was initially broadly perceived to be an 

event that would have a strong negative impact on business and, hence, on employment. Later, 

when businesses and organisations obtained access to JobKeeper, and the pandemic’s impact in 

some geographic areas was not large, employment recovered steadily. Thus, the employment 

impacts of COVID in Australia were buffered in regions where case numbers were low. The 

introduction of government policies like JobKeeper (which favoured those in non-casualised 

employment) reduced the impact for some groups, such as older workers. At a national level, then, 

the long-term impact of 2020 on the employment outcomes for older workers is likely to be different 

to other recessions because the economic costs are likely to persist (Coates et al. 2020).  

In terms of geography, we concentrated primarily upon Victoria, but comparison was made with New 

South Wales, a similar state in economic terms (Gross State Product per capita); and with Western 

Australia, where COVID-19 was least prevalent among the states. Initially, COVID-19 was slightly 

more prevalent in New South Wales than in Victoria. However, Victoria had a very large second wave 

not experienced by New South Wales, and that experience was confined largely to metropolitan 

Melbourne. As the prevalence of COVID-19 was heavily concentrated in cities, we compare with 

regional Australia and regional Victoria. These results reflect a natural experiment: Victoria was 

subject to the same employment support policies as the rest of Australia because those policies were 

mainly Federal (the exception being paid pandemic leave for COVID-19 testing and associated self-

isolation). The different effects between Victoria, New South Wales and Western Australia were 

therefore due (almost) purely to the different experience of COVID-19 (besides any changes that 

would have taken place in the absence of COVID-19, which can be expected to be very small in the 

short term). The same argument can be made about metropolitan Melbourne and regional Victoria. 

The reason Victoria (particularly regional Victoria) had a smaller contraction in employment is that it 

is less intensive in mining, and, to a lesser extent, accommodation and food services and other 

services sectors, which all suffered large reductions in employment. 



38 Fry et al. Australian Population Studies 5 (2) 2021 

 

 

By December 2020, there were few instances where employment loss remained considerable. The 

main exceptions where employment effects were still negative at the end of 2020 were people living 

in metropolitan Melbourne, private sector workers aged 50–59 years, people living alone aged 65 

years and over, and those with the lowest education qualification. In the private sector, for those 

aged 45–60 years, the initial negative impact on employment numbers was largely sustained in the 

longer period. This may indicate that private sector employers took the opportunity to retrench 

some older, lower skilled workers. Data not shown here reveal that outcomes were worse for 

occupations at the lower end of the skills spectrum, although this was not always the case. For 

example, strongly negative outcomes in the longer period were evident for food preparation 

assistants, farm forestry and garden workers, sales support workers, and clerical and office support 

workers. On the other hand, there was a private sector recovery for those aged 60 years or over and 

this could be because these people tend to be self-employed or employed as managers/professionals 

and had more control over their own employment. In terms of occupations, for storepersons, a low 

skilled job, there were strong positive results in the longer period, probably because COVID-19 

stimulated jobs in this occupation. 

Overall, while the national picture for older workers appeared positive by the end of 2020, lower-

skilled workers in some sectors experienced significant deleterious effects. Long-term job loss for 

older workers was concentrated in the following industries: manufacturing; retail; health care and 

social administration; transport, postal, warehouse; arts and recreation; and information, media and 

telecommunications. The immediate effects of COVID-19 on these industries were obvious as 

entertainment venues, cafes and restaurants were closed, and hotels were affected by travel 

restrictions. Lockdowns probably also prevented information, media and telecommunications 

industries operating as normal. The large initial negative impact for transport, postal and 

warehousing was probably mainly in transport, as lockdowns affected airlines and public transport. 

Some industries recovered well. For example, distribution probably accounts for the large increase in 

employment in accommodation and food services as, beyond the initial period, there was a massive 

expansion in home delivery of food. Gains in administrative and support services may be due to the 

time taken to develop alternative approaches to on-site working. The turnaround in mining was 

related to increases in demand from outside of Australia. In the case of iron ore, where exports and 

prices have expanded greatly, this was the result of the disruption that COVID-19 caused in other 

countries, especially Brazil (Ker 2020).  

JobKeeper was clearly successful at maintaining attachment to the workplace, especially for higher 

level workers. However, in some industries, the data imply that employers were either ineligible for 

payments (with an insufficient reduction in turnover, or being an excluded organisation), chose not 

to enrol staff for JobKeeper, or employees were not eligible for payments. This particularly affected 

sectors with high casualisation or many migrant workers: JobKeeper was structured in favour of 

permanent employees and long-term casuals, and those with Australian residency.  

Although there was little impact on the unemployment rate, job search duration increased over time, 

suggesting no churn in the unemployed and an increase in long-term unemployment. Initially, males 

were more affected by unemployment (particularly in arts and recreation services and 

accommodation and food services). 



Australian Population Studies 5 (2) 2021 Fry et al. 39 

 

There were industries where the pandemic itself stimulated employment, even in the initial period. 

This includes financial and insurance services, wholesale trade and construction. Businesses wishing 

to take advantage of various government schemes to support employment (especially JobKeeper) 

and people making personal decisions about their financial circumstances probably led to the early 

surge in employment in the financial and insurance services sector. Growth in construction was likely 

also related to the later ‘HomeBuilder’ stimulus program, which provided grants for homeowners to 

build, renovate and buy new-build homes.  

The longer-term impacts by industry were very diverse, with some sectors benefitting substantially 

while some others suffered considerably. There remain major concerns about the employment of 

older workers in the industries where employment has remained low in the longer term, especially 

those with lower skill levels. Some sectoral shifts effected by COVID-19 — such as the growth of 

online shopping rather than retail, and the growth in food delivery — will particularly affect older 

workers’ employment going forward. From past experience in the manufacturing industry, older 

workers who lose their jobs likely have great difficulty ever gaining work again and retire completely 

from the labour force (Callan & Bowman 2015; Chomik 2020; OECD 2016). While this has negative 

consequences for income and superannuation, it can also affect mental health (which can also affect 

physical health), thereby further worsening financial circumstances through increased medical 

expenditure (Mandal & Roe 2007). There may also be negative effects on the wellbeing of others in 

the household (Siegel et al. 2004). Those who do find work can have lower earnings due to shorter 

hours or lower hourly pay rates and often find themselves in precarious forms of employment. 

Our analysis has used cross-sectional data tabulations from the ABS Labour Force Survey. This means 

our employment measures do not necessarily capture the same individuals. Rather, there could be 

churn with other labour market states hidden in these figures. Moreover, take-up of JobKeeper may 

have influenced employment results, with individuals classified as employed although they did not 

work. Future research could explore casualisation of the workforce more fully, potentially using 

longitudinal data from the Household, Income and Labour Dynamics in Australia (HILDA) survey or 

the ABS Longitudinal Labour Force dataset. Linked census data may also help to flesh out the 

longitudinal aspects. Space considerations meant we were unable to explore redundancy; the 

relationship between employment outcomes and working arrangements would also be of interest, as 

would exploring the gendered aspects of jobs that have been created post-COVID-19. It would also 

be useful to examine job search duration post-retrenchment. Future research could also consider the 

wider array of government stimulus measures, and how they have impacted job prospects in specific 

sectors. 

5. Conclusions 

Noting these limitations and extensions, our study provides evidence of the impact of the COVID-19 

pandemic on labour force behaviour and outcomes for older Australians. Usually in recessions older 

workers fare worse than younger workers. However, in the recession induced by COVID-19 this did 

not always occur, as younger workers are more casualised and concentrated in affected industries.  

The longer-term impacts by industry were very diverse, with some sectors benefitting substantially, 

while some others suffered considerably. This is the most important finding of the paper. By 



40 Fry et al. Australian Population Studies 5 (2) 2021 

 

 

December 2020, employment levels for older workers in some sectors had recovered from the initial 

downturn caused by the pandemic. However, for older workers in some industries hard hit by the 

pandemic, there is a major concern about their potential for future employment. There remain major 

concerns about the employment of older workers in the industries where employment has remained 

low in the longer term, namely manufacturing; retail trade and information; media and 

telecommunications; and transport, postal and warehousing. Future government stimulus measures 

could be particularly aimed at supporting workers from these industries.  

With the phase out of JobKeeper in 2021, and continued shutdowns in Victoria and New South 

Wales, it is important to track the continued impact of the COVID-19 pandemic on older workers and 

job seekers. With COVID now affecting employment in Australia’s two most populous states, it is 

essential that research continues to consider the ongoing impacts of COVID on particular groups in 

the labour market in the medium and longer-term. 

Key messages 

• Overall employment rates for older workers fell in April 2020 but recovered by December. 

• The longer-term impacts by industry were very diverse, with some sectors benefitting 

substantially, while others suffered considerably. 

• There remain major concerns about the future employment prospects of older workers 

(particularly those with lower skill levels) in sectors where employment levels had not recovered 

by December 2020. 

Acknowledgments 

Funding was provided by COVID19 Impacts on Society Scheme, University of Melbourne. Fry, Temple 

and McDonald are funded by the Australian Research Council’s Centre of Excellence in Population 

Ageing Research (CE1101029). 

References 

ABS (2018) Labour statistics: Concepts, sources and methods, Feb 2018. Canberra: ABS. 

https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/6102.0.55.001~Feb%202018~

Main%20Features~Overview%20of%20Labour%20Statistics~1. Accessed on 11 March 2021. 

ABS (2021a) Dataset: Characteristics of employment, 2014–2020, Canberra: ABS. 

https://www.abs.gov.au/statistics/labour/earnings-and-work-hours/characteristics-employment-

australia/latest-release. Accessed on 11 March 2021. 

ABS (2021b) Dataset: Labour force Australia, detailed. Canberra: ABS. 

https://www.abs.gov.au/statistics/labour/employment-and-unemployment/labour-force-

australia-detailed/latest-release. Accessed on 11 March 2021. 

ABS (2021c) Labour Force, Australia methodology. Canberra: ABS. 

https://www.abs.gov.au/methodologies/labour-force-australia-methodology/jan-2021. Accessed 

on 11 March 2021. 

Adair T & Temple J (2012) Barriers to mature age labour force engagement in Australia: Canberra: 

Department of Education, Employment and Workplace Relations. 

https://nationalseniors.com.au/uploads/201208_PACReport_Research_BarriersMatureAgeEmplo

yment_Full_1.pdf  

https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/6102.0.55.001~Feb%202018~Main%20Features~Overview%20of%20Labour%20Statistics~1
https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/6102.0.55.001~Feb%202018~Main%20Features~Overview%20of%20Labour%20Statistics~1
https://www.abs.gov.au/statistics/labour/earnings-and-work-hours/characteristics-employment-australia/latest-release
https://www.abs.gov.au/statistics/labour/earnings-and-work-hours/characteristics-employment-australia/latest-release
https://www.abs.gov.au/statistics/labour/employment-and-unemployment/labour-force-australia-detailed/latest-release
https://www.abs.gov.au/statistics/labour/employment-and-unemployment/labour-force-australia-detailed/latest-release
https://www.abs.gov.au/methodologies/labour-force-australia-methodology/jan-2021
https://nationalseniors.com.au/uploads/201208_PACReport_Research_BarriersMatureAgeEmployment_Full_1.pdf
https://nationalseniors.com.au/uploads/201208_PACReport_Research_BarriersMatureAgeEmployment_Full_1.pdf


Australian Population Studies 5 (2) 2021 Fry et al. 41 

 

Birch E & Preston A (2021) The Australian labour market in 2020. Journal of Industrial Relations 63(3): 

303-320. https://doi.org/10.1177/00221856211002224  

Borland J & Charlton A (2020) The Australian labour market and the early impact of COVID-19: an 

assessment. Australian Economic Review 53(3): 297-324. https://doi.org/10.1111/1467-

8462.12386  

Bui T T M, Button P, & Picciotti E G (2020) Early evidence on the impact of COVID-19 and the recession on 

older workers. NBER Working Paper no. w27448. Cambridge, MA: National Bureau of Economic 

Research. https://doi.org/10.3386/w27448  

Callan V J & Bowman K (2015) Industry restructuring and job loss: Helping older workers get back into 

employment. Adelaide: National Centre for Vocational Education Research. 

https://www.ncver.edu.au/research-and-statistics/publications/all-publications/industry-

restructuring-and-job-loss-helping-older-workers-get-back-into-employment  

Chomik R (2020) COVID-19 and the demographic distribution of health and economic risks. ARC Centre of 

Excellence in Population Ageing Research Factsheet. Sydney: CEPAR. 

https://cepar.edu.au/sites/default/files/COVID-19-and-populations-at-risk.pdf  

Churchill B (2020) COVID‐19 and the immediate impact on young people and employment in Australia: A 

gendered analysis. Gender, Work & Organization 28(2): 783-794. 

https://doi.org/10.1111/gwao.12563  

Coates B, Cowgill M, Chen T, & Mackey W (2020) Shutdown: estimating the COVID-19 employment shock. 

Melbourne: Grattan Institute. https://grattan.edu.au/report/shutdown-estimating-the-covid-19-

employment-shock/  

Department of Health (2021) Coronavirus (COVID-19) current situation and case numbers. 

https://www.health.gov.au/news/health-alerts/novel-coronavirus-2019-ncov-health-

alert/coronavirus-covid-19-current-situation-and-case-numbers#cases-and-deaths-by-age-and-

sex. Accessed on 11 March 2021. 

Forsyth A (2020) COVID-19 and labour law: Australia. Italian Labour Law e-Journal 13(1S). 

https://doi.org/10.6092/issn.1561-8048/10812  

Ker P (2020) Brazil’s Virus Pain to Extend Australia’s Iron Ore Gain. Australian Financial Review, 28 

September 2020. 

Mandal B & Roe B E (2007) Job loss, retirement and the mental health of older Americans. SSRN. 

http://dx.doi.org/10.2139/ssrn.991134  

Neil I, Chin D, & Parkin C (2021) The JobKeeper Scheme. Australian Journal of Labour Law 34: 24-42. 

https://advance.lexis.com/api/permalink/aebe5bcb-6da7-4e76-928c-

fc1fae2fcc93/?context=1201008&federationidp=F2D4KJ52509.  

OECD (1998) Work force ageing: consequences and policy responses, Working Paper AWP 4.1. Paris: 

OECD Publishing. https://www.oecd.org/els/public-pensions/2429096.pdf  

OECD (2016) Back to work: Australia. Improving the re-employment prospects of displaced workers. Paris: 

OECD Publishing. https://www.oecd.org/employment/emp/Back-to-Work-Australia-AR.pdf  

OECD (2020) Job retention schemes during the COVID-19 lockdown and beyond. Paris: OECD Publishing.  

https://www.oecd.org/coronavirus/policy-responses/job-retention-schemes-during-the-covid-

19-lockdown-and-beyond-0853ba1d/  

Rutledge M S & Coe N B (2012) Great Recession-Induced Early Claimers: Who Are They? How Much Do 

They Lose? Center for Retirement Research Working Paper 2012-12. Boston MA: Boston College. 

https://crr.bc.edu/wp-content/uploads/2012/04/wp_2012-12-508.pdf  

Siegel M J, Bradley E H, Gallo W T, & Kasl S V (2004) The effect of spousal mental and physical health on 

husbands’ and wives’ depressive symptoms, among older adults: Longitudinal evidence from the 

health and retirement survey. Journal of Aging and Health 16: 398-425. 

https://doi.org/10.1177/0898264304264208  

https://doi.org/10.1177/00221856211002224
https://doi.org/10.1111/1467-8462.12386
https://doi.org/10.1111/1467-8462.12386
https://doi.org/10.3386/w27448
https://www.ncver.edu.au/research-and-statistics/publications/all-publications/industry-restructuring-and-job-loss-helping-older-workers-get-back-into-employment
https://www.ncver.edu.au/research-and-statistics/publications/all-publications/industry-restructuring-and-job-loss-helping-older-workers-get-back-into-employment
https://cepar.edu.au/sites/default/files/COVID-19-and-populations-at-risk.pdf
https://doi.org/10.1111/gwao.12563
https://grattan.edu.au/report/shutdown-estimating-the-covid-19-employment-shock/
https://grattan.edu.au/report/shutdown-estimating-the-covid-19-employment-shock/
https://www.health.gov.au/news/health-alerts/novel-coronavirus-2019-ncov-health-alert/coronavirus-covid-19-current-situation-and-case-numbers#cases-and-deaths-by-age-and-sex
https://www.health.gov.au/news/health-alerts/novel-coronavirus-2019-ncov-health-alert/coronavirus-covid-19-current-situation-and-case-numbers#cases-and-deaths-by-age-and-sex
https://www.health.gov.au/news/health-alerts/novel-coronavirus-2019-ncov-health-alert/coronavirus-covid-19-current-situation-and-case-numbers#cases-and-deaths-by-age-and-sex
https://doi.org/10.6092/issn.1561-8048/10812
http://dx.doi.org/10.2139/ssrn.991134
https://advance.lexis.com/api/permalink/aebe5bcb-6da7-4e76-928c-fc1fae2fcc93/?context=1201008&federationidp=F2D4KJ52509
https://advance.lexis.com/api/permalink/aebe5bcb-6da7-4e76-928c-fc1fae2fcc93/?context=1201008&federationidp=F2D4KJ52509
https://www.oecd.org/els/public-pensions/2429096.pdf
https://www.oecd.org/employment/emp/Back-to-Work-Australia-AR.pdf
https://www.oecd.org/coronavirus/policy-responses/job-retention-schemes-during-the-covid-19-lockdown-and-beyond-0853ba1d/
https://www.oecd.org/coronavirus/policy-responses/job-retention-schemes-during-the-covid-19-lockdown-and-beyond-0853ba1d/
https://crr.bc.edu/wp-content/uploads/2012/04/wp_2012-12-508.pdf
https://doi.org/10.1177/0898264304264208


42 Fry et al. Australian Population Studies 5 (2) 2021 

 

 

Varma P, Junge M, Meaklim H, & Jackson M (2021) Younger people are more vulnerable to stress, anxiety 

and depression during COVID-19 pandemic: a global cross-sectional survey. Progress in Neuro-

Psychopharmacology and Biological Psychiatry 109: 110236. 

https://doi.org/10.1016/j.pnpbp.2020.110236  

Wood D, Griffiths K, & Crowley T (2021) Women’s work: The impact of the COVID crisis on Australian 

women. Melbourne: Grattan Institute. https://grattan.edu.au/report/womens-work/  

Yeatts D E, Folts W E, & Knapp J (2000) Older workers adaptation to a changing workplace: Employment 

issues for the 21st century. Educational Gerontology 26(6): 565-582. 

https://doi.org/10.1080/03601270050133900  

 

https://doi.org/10.1016/j.pnpbp.2020.110236
https://grattan.edu.au/report/womens-work/
https://doi.org/10.1080/03601270050133900

