Productivity measurement in the time of a pandemic

The COVID-19 pandemic generated an unprecedented shock to the Australian economy and the labour market. This article discusses how these impacts flowed through to productivity measures in the 2019-20 reference year.

Multifactor productivity (MFP) measures output per unit of labour and capital factor inputs to production. Gross Value Added (GVA) is output less intermediate consumption of goods and services used in the production process. Labour productivity is measured as the ratio of GVA to hours worked. As such changes in GVA will impact MFP and labour productivity.

GVA estimates are compiled through the Supply-Use framework, the last period is for 2018-19. The estimates for 2019-20 are derived by extrapolating the annual benchmarks, using the best available indicators at the time of compilation. A key assumption underpinning this output indicator method is that gross output, intermediate consumption, and GVA grow or decline at the same rate. The ABS reviewed this assumption and assessed that no intervention to the indicator method was required in the latest year (for further information, see Output indicators in the national accounts). Output indicators will continue to be actively monitored to assure they reflect the real economy and to maintain the quality of the latest GVA estimates.

The COVID-19 pandemic and resultant lockdown impacted GVA [1], with twelve of the sixteen market sector industries recording falls. The largest falls were in Accommodation food and services, Arts and recreation services and Transport, postal and warehousing industries. For more details on the timing and stringency of containment measures, see International comparisons.

The government’s economic response to the COVID-19 pandemic included the introduction of JobKeeper and the Boosting Cash Flow for Employers policy. The payments from these schemes are classified as ‘other subsidies on production’ (for further information on the classification of these policies, see Economic measurement during COVID-19: Selected issues in the Economic Accounts). Despite record levels of subsidy payments to business, there is no directly observable impact on GVA or labour productivity [2]. Subsidy payments allocated to employers and employees will change labour and capital income shares in the latest and following year, which may impact multifactor productivity. For further information on the allocation of these subsidies by industry see Government support for business.

Historically, labour input for productivity compilation has been sourced from the hours worked series of the Labour Force Survey (LFS). For the 2019-20 reference year, the national accounts moved to hours worked movement from the Australian Labour Account for industries and the market sector aggregates. This more accurately captured the fall in hours worked due to COVID-19 related restrictions which occurred after the LFS reference period (for further information see Assessing the impact of COVID-19 on the Labour Account).

Labour input can also be measured as Quality Adjusted Labour Inputs (QALI). QALI takes into consideration labour composition change due to educational attainment and experience by sourcing information from the latest Census data. The government response to develop short, highly discounted, online courses at higher education levels to upskill or retrain workers displaced by the pandemic [3] may result in changes in labour composition, which will be reflected in productivity measures once Census 2021 data is available.

Stringent social distancing measures introduced from late March resulted in business capital being unused or under-used for productive activities. This varied across industry, with more pronounced impacts on customer-facing industries such as Accommodation and food services and Transport, postal and warehousing.

Capital services flows are not directly observable, and are estimated using productive capital stock. Underpinning capital services measurement is the assumption that capital utilisation remains constant. The ABS reviewed this assumption as part of article Variations in the Utilisation of Productivity Inputs.

Growth in labour productivity has been recognised as the key driver for improvements in living standards. Two contributions to labour productivity growth are growth in capital services per hour worked (capital deepening) and multifactor productivity growth.

Stringent social distancing and lockdown measures introduced from late March 2020, resulted in reduced business activity and temporary business closures. These had significant impacts on drivers of labour productivity growth. In particular, the capital deepening component is likely to be overstated which may impact the accuracy of the labour productivity measure.

At the aggregate level, labour productivity of the market sector reflects not only the industry productivity performance but the changes in allocation of labour input. For 2019-20 labour productivity rose 0.6%, boosted as less productive industries saw the largest falls in hours worked due to stricter containment measures. This is a relative reallocation toward higher productive industries that either did not fall or fell less, and will have a larger share of hours worked than in previous years.

Multifactor productivity (MFP) shows how labour and capital inputs are used to generate GVA and are interpreted as an indicator of economic performance of the economy. MFP fell 0.7% in 2019-20. Government response to COVID-19 had an impact on economic efficiency as containment measures impacted business activity.

The size of the economic shocks experienced during the pandemic introduces measurement challenges, especially at the industry level. In the short run, this may cause a departure of MFP from its conceptual definition.

Productivity is measured residually and has historically been difficult to interpret, especially in the most recent periods. This is because productivity measures include a number of drivers including technical change, scale and cyclical effects which are difficult to separately identify. The COVID-19 pandemic has compounded this issue, as it has had varying impacts on productivity estimates for 2019-20. Care should be taken when interpreting year to year productivity growth for the market sector and by industry. Users are encouraged to continue viewing productivity growth through growth cycles (for more information, see Experimental Estimates of Industry value added growth cycles).

[1] While GVA is used to compile the headline productivity, the ABS also publishes gross output based productivity measures up to 2018-19.

[2] JobKeeper payments have been recorded as “other subsidies on production” from the government to employers. An increase in “other subsidies on production” will reduce GVA. When employers pass JobKeeper payments to employees they are recorded as wages and salaries (compensation of employees) which increases GVA by the same offsetting amount. Timing and other reporting issues may also mean this does not hold perfectly however. Any indirect impacts on firm behaviour and hence output and GVA are also unclear.

[3] https://ministers.dese.gov.au/tehan/short-courses-now-available-upskill-australians