“The first six months expenditure for the Australian JobKeeper scheme lifted total national government expenditure by one quarter”

During the Covid-19-crisis Australia quickly introduced the “JobKeeper scheme” – a large scale and easy to implement wage subsidy program to keep firms and employees together. However, the differences to “European-style” job retention programmes are striking. In this interview, Bob Gregory, Professor Emeritus at the Australian National University provides insights on the Australian approach and sketches some possible lessons for policy makers in other countries.

What were the key design features of the Australian job retention scheme?

Australia’s first ever job retention scheme, JobKeeper, was introduced on March 30th, 2020, for an initial six months, extended and modified for two additional three month periods and then discontinued. JobKeeper encompassed around one third of Australian businesses and employees.

For the first six months the Australian Taxation Office (ATO) would provide each eligible firm with a uniform amount of 1.500 Australian Dollar (roughly 900 Euro) to be paid to every eligible employee every two weeks, irrespective of the actual salary and hours worked before.

For the first six months, benefits were paid irrespective of actual salary and hours worked.

Even for those employees who had no working time reduction due to the pandemic?

Yes, many employees worked and produced output as normal and received unchanged pay, but now $1,500 would be covered by the Taxation Office contribution. The ATO subsidy had to be passed on to the employee in an accounting sense. At the same time, there were no restrictions on how any wage saving to the firm had to be used. For one fifth of total recipients, mainly part-time employees, the $1,500 lifted their pay above their normal pay. The amount of uniform wage subsidy was very close to the level of the Australian minimum wage.

What were the eligibility criteria?

To establish eligibility the firm applied to the ATO and stated that they anticipated a 30 per cent decline in turnover during at least one month in the following six. A few private sector industries such as banks, finance companies, hospitals and universities were excluded from the scheme. For very large firms the turnover decline had to be 50 per cent or more. JobKeeper was a “one-in-all-in” system applied to Australian residents employed in the JobKeeper firm on March 1. 2020. However, casuals who had not been employed at the firm for at least twelve months were excluded. Although around 23 per cent of payments went to large firms, with a normal turnover exceeding $50 million, 41 per cent of recipient firms (12 per cent of payments) were sole traders of which 90 per cent had no employees.

Have there been any changes after the initial six-month period?

Yes, there were two important changes: actual turnover declines replaced self-assessed predicted declines and the uniform wage subsidy of 1.500 Australian Dollar was split into two tiers – a uniform rate for full-time workers and a lower uniform rate for part-time workers. Each tier was reduced in two steps – in September 2020 and in January 2021. For the last three months the full-time worker subsidy was about 60 per cent of the initial subsidy, when the scheme was introduced, and 40 per cent for part-time workers.

JobKeeper expenditure for the first six months was equivalent to 7 per cent of GDP.

The Australian scheme was very simple, relative to many European job retention schemes, and very expensive. During the first six months, JobKeeper expenditure was equivalent to 7.0 per cent of GDP. It lifted total national government expenditure by one quarter.

Why was the Australian scheme so simple and why did JobKeeper account for such a large share of government expenditure? Were there any special political goals at the time of the introduction?

The simplicity and “generosity” of JobKeeper can be explained by the special nature of the Covid Pandemic. Within a few weeks in March 2020, unemployment increased at an unprecedented rate and a worst case Treasury estimate, in the absence of a policy response, was that GDP could fall by as much as 25 per cent. A massive and rapidly introduced fiscal stimulus was needed. With this response, and the quick development of a vaccine, the speed and depth of the economic down-turn might be short-lived. In addition, a fast economic recovery from the downturn might be facilitated if firms and workers could be kept together in the economic downswing. Perhaps a new and untried program, a large and generous job retention scheme could play an important role to moderate the downswing and  increase the speed of the recovery. But to do this a speedy and large take-up rate was required. It was firmly believed that the simpler and more generous the new program the more likely this could be achieved.

Government was also advised that unless “lock-down” and “social distancing” policies were adopted Covid could result in a high death toll – perhaps 50,000 to 100,000 people. The Australian community widely supported these Covid health policies but it was possible that political support might dissipate within a short time if large income losses were quickly accumulated by firms and individuals. This “political” judgment also supported exceptional generosity and simplicity to encourage high take-up rates by firms. The generosity of the scheme also included the provision that firms would make no financial contribution and all wage subsidy expenditure would be met by government borrowing.

Were there any measures to avoid misuse and deadweight losses?

Not really. Both misuse and deadweight losses are likely to be judged as “relatively unimportant” in a very temporary job retention scheme primarily designed to generate business confidence and deliver a large macro stimulus.  However, half-way through the JobKeeper program the firm eligibility requirement was moved from a self-assessed to an actual turnover decline and this reduced take-up considerably. This change, however, was a logical step. Now that the JobKeeper take-up was successful, economic recovery appeared to be well under way and actual turnover data had become available. The subsidy reductions beginning at the six month point were also primarily a result of the schemes success but there was some feeling on equity grounds that the initial subsidy for part-time-workers was too high. The paring back of the scheme after six months therefore was not really a response to misuse or deadweight losses although both obviously existed.

JobKeeper sped up the recovery process and appears to have had no adverse effects on resource reallocation.

Did the support system in your view slow down the necessary reallocation processes.

No. JobKeeper sped up the recovery process and appears to have had no adverse effects on resource reallocation. Within 12 months, employment levels in the economy had returned to previous levels and within 18 months the employment-population ratio was the highest experienced over the last half century.

Were there incentives for firms to offer on-the-job training?

No. This issue was not relevant given the very short-term nature of the program.

Were there incentives for benefit recipients to look for new jobs?

No. There were few jobs available during the covid crisis and the government objective was to encourage individuals to stay connected to their firm even if the firm did not require their labour during the twelve month period.

Is there a debate as to whether to include elements of experience rating in the national jab retention scheme?

No. There is very little Australian discussion as to the nature of good job retention schemes. The general view among policy advisors is that job retention schemes are not appropriate for Australia, given the very flexible labour market. Employment usually grows at about 2 per cent per annum, there are few depressed economic regions and in general terms jobs are not scarce.

The general view among policy advisors is that job retention schemes are not appropriate for Australia, given the very flexible labour market.

How effective was the speed and bureaucratic support for the introduction of the program? 

There was widespread bureaucratic support for the fast introduction of JobKeeper. The Australian Taxation Office (ATO), the delivery agency, coped extremely well, primarily because of the recent adoption of a new tax collection technology – “Single Touch Payroll” – that was fully introduced by July 2019. This technology integrated the firm’s payroll payment system directly with ATO computers to facilitate the firm’s payment of worker superannuation contributions, profit and employee withholding taxes. The technology encompassed 99 per cent of firms and enabled the ATO to easily match employer and employee records. The development of this technology was a necessary pre-requisite for JobKeeper to be implemented.

How was the termination of JobKeeper justified?

The economic recovery, over the JobKeeper period, was so strong that there was little opposition to the program termination. The rapid increase in government debt, in response to the Covid fiscal expenditure, also contributed to the support for termination.

From a short term perspective, the JobKeeper scheme was exceptionally effective. 

How would you judge the general effectiveness of the job retention scheme?

From a short term perspective, the JobKeeper scheme was exceptionally effective: the speed of introduction was fast, the extent of macro stimulus to generate an economic recovery was large because of the high take-up rate, the linking of firms and workers together was extensive, and the underpinning of community support for “lockdowns” was also very successful. Another indicator of success is that unemployment rate in Australia is now at the lowest level for half a century.

But it is important to realize that any large macro stimulus introduced so quickly will necessarily include undesirable side effects. If the alternative to JobKeeper was to send cash to households, as in the global financial crisis for example, it would be expected that many households would not need the cash.

And what is your judgement from a longer-term perspective?

Here, JobKeeper evaluations may be less “glowing”. I would like to mention some contentious issues. A large proportion of recipient firms did not need the subsidy. A comparison of actual firm turnover for the first six months of JobKeeper, with a comparable period the year before Covid, indicates that about one half of the disbursements went to firms that did not experience a turnover decline of 30 per cent or more. Furthermore, one third of disbursements were paid to firms that experienced a turnover increase!!! On average, during the Covid period, the profit share of GDP reached the highest level for half a century. If the alternative to JobKeeper was to send cash to households, as in the global financial crisis, it would also be expected that many households may not need the cash.

One third of disbursements were paid to firms that experienced a turnover increase!!!

Of course, firms not needing the subsidy may, to some extent, be a result of the macro “success” of the program. Large expenditure levels must have been an important reason why GDP did not fall by the worst case scenario of 25 per cent, considered when designing the program, and why actual turnover outcomes were not so dire.

Another issue is that micro evaluations have indicated a high cost per additional job saved between 50,000 and 100,000 Australian Dollars, which does seem a lot. But these estimates should be compared to the cost per additional job saved from alternative macro stimulus packages – such as tax cuts, or substantial cash disbursements to households – which might have been adopted instead of JobKeeper. These comparisons across programs are not easy to make.

I should also add that in the Australian context the economic case for keeping firms and employees together is not that compelling. There are many complex issues involved here. In the past, Australian deep recessions have been associated with substantial increases in long term unemployment and this outcome is often thought of as evidence for a loss of human capital that can come about because of job separations. However, the long term unemployed, on average, seem to have possessed poor labour market skills before becoming unemployed, a high incidence of poor health and do not appear to have possessed a significant degree of firm specific skills. Furthermore, in normal times, as much as one fifth of all workers have been in their current job for less than a year and the separation of workers from jobs during a recession is usually concentrated on those with short tenure.

Would you like to add anything?

Three final points: First, the focus of the interview has been on JobKeeper but it should also be remembered that there were other programs that affected the macro recovery and a final assessment of Jobkeeper should not only involve a comparison with these programs but also address whether the total stimulus from all sources of fiscal and monetary policy was too great. In particular, I am thinking about the generous boosting cash flow, which delivered up to $100,000 to small businesses with no requirements as to how this should be spent. The expenditure on this program over the year was equivalent to 2 per cent of GDP.

Second, in the process of future assessments it is important to be fully conscious of how exceptional and fast moving the deterioration of the economy was during the six weeks beginning mid March 2020 and how widespread was the belief that it was urgent for policy to respond quickly. It has even been suggested, on the basis of New Zealand experience, that the introduction of the policies one or two weeks earlier would have made a significant difference, as pointed out by Jeff Borlands and Jennifer Hunts recently published paper in The Australian Economic Review.

Third, Treasury has recently announced an independent enquiry into lessons to be learnt from JobKeeper and many readers might be interested in their forthcoming assessment.

References

Australian Government/The Treasury (2020): The JobKeeper Payment: Three month review, June 2020.

Australian Government/The Treasury (2021): Insights from the first six months of JobKeeper, October 2021.

Borland, Jeff; Hunt, Jennifer (2023) JobKeeper: An Initial Assessment, The Australian Economic Review, vol. 56, no 1, pp. 109-123. 

Cassells, Rebecca; Duncan, Alan, (2020) JobKeeper: The Efficacy of Australia’s first short time wage subsidy, Australian Journal of Labour Economics, Vol. 23/2, pp.99-128. 

Kennedy, Steven (2022): The 2022 Sir Leslie Melville Lecture, 27 July, The Australian Treasury.

Watson, Timothy; Tervala, Juha; Sainsbury, Tristram (2023): The JobKeeper payment: How good are wage subsidies?, Centre for Applied Macroeconomic Analysis, Working Paper 36/2022, Australian National University, Canberra.

DOI: 10.48720/IAB.FOO.20231031.03