Jobkeeper – Australia Today
In the midst of an economic crisis, we do not know what the future will hold, but we must make decisions anyway, former Bank of England governor Mervyn King and economist John Kay observe in their acclaimed book, Radical Uncertainty: Decision-Making Beyond the Numbers.
A radically uncertain crisis was what Treasurer Josh Frydenberg faced in March last year, when Treasury cobbled together in about a week an estimated $130 billion wage subsidy.
The speed of the crisis was so quick that the historic JobKeeper package unveiled on March 30, 2020, dwarfed the initial $17.6 billion in economic stimulus on March 12 and the $66 billion announced on March 22, 2020.
At the time, preliminary epidemiological modelling for the government suggested 50,000 to 150,000 Australians could die with COVID-19.
Consequently, Treasury forecast the unemployment rate to hit 10 per cent and enacted a six-month wage subsidy to retain a link between employers and staff.
With the benefit of hindsight, the health and economic warnings turned out to be a bad dream. The health and economic crises were not as bad as feared, at least in Australia.
Now, Frydenberg is facing criticisms from Labor, sections of the media and some analysts, such as corporate governance expert Dean Paatsch, that the government wasted tens of billions of dollars on JobKeeper.
Extraordinarily, tax commissioner Chris Jordan is facing a possible contempt finding from the Senate for his refusal to disclose the names of JobKeeper businesses turning over more than $10 million.
What is not disputed by the vast majority of observers is JobKeeper was necessary when the economy was shut down. It was an economic and social lifeline for millions of Australian businesses and workers, rejuvenating confidence in workers, consumers, businesses and investors.
The program ultimately cost $40 billion less than originally projected. But at a total cost of $89 billion over 12 months, it is easily the most expensive spending program in history.
In that spirit and with some hindsight, this writer offers three suggestions how JobKeeper could have been improved and delivered better value for money for taxpayers.
First, re-testing the eligibility of recipient firms after three or four months, not paying the subsidies for at least six months to all enrolled firms such as those posting rising revenues above pre-pandemic levels. New Zealand re-tested eligibility for its wage subsidy after 12 weeks and spent a significantly lower $NZ13 billion.
Third, curtailing the generosity of the separate and little-talked-about $35 billion business cash flow payments to small and medium enterprises, or merging it with JobKeeper.
Treasury, the chief designer of JobKeeper, has analysed microdata for the first six months of the program to late September 2020. It is important to understand the numbers and the context of the findings.
To qualify for JobKeeper, a business had to have a reasonable expectation that its revenue would fall by at least 30 per cent (15 per cent for not-for-profits, 50 per cent for large companies turning over more than $1 billion) in any one month or quarter between March 30 and September 27, 2020, compared to the corresponding period in 2019.
The forecast turnover test allowed eligibility to be determined quickly and to inject cash into locked-down businesses.
Some 925,000 businesses and not-for profits received $1500 a fortnight for 3.6 million employees over the program’s first six months.
Treasury’s sample data was only $47 billion of the initial six-month $70 billion program, due to data limits on some not-for-profits such as private schools and charities, and subsidiaries within big conglomerates.
So the pure dollar figures in the analysis understate the larger overall figures of the program.
About 40 per cent of the subsidies was paid to firms that recorded revenue declines of at least 30 per cent (50 per cent for large firms) across the June and September quarters of 2020.
About 30 per cent was paid to firms that recorded revenue declines less than the forecast thresholds.
The remaining 30 per cent was paid to businesses which ultimately increased their revenue.
It is the latter category, firms that increased their turnover and received JobKeeper, that is the biggest point of political contention.
About $13.8 billion of the analysed $47 billion in JobKeeper payments went to firms that recorded an increase in revenue.
This appears extremely wasteful.
But Treasury puts forward several caveats to suggest any overpayments are not as large as they seem.
Some $4.9 billion was paid to firms that did not record a fall in revenue in the June quarter 2020 when compared to the June quarter 2019 – the baseline for the original JobKeeper test. However, these firms did suffer a decline in revenue between the March quarter 2020 and June quarter 2020.
Treasury’s explanation is that many of these businesses had been fast-growing before the pandemic, due to their start-up nature, a merger, acquisition, restructuring or recovery from the earlier drought.
“These businesses were allowed to use a different test to determine their eligibility for JobKeeper to more accurately reflect the size of the business at the onset of the pandemic,” Treasury notes.
Hence, it could be argued any overpayment to firms that recorded an increase in revenue was about $8.9 billion in the scheme’s first six months.
(Again, this is only out of the $47 billion sample size, so the real figure paid to firms with rising revenue in the total JobKeeper program would be higher, probably in the order of $15 billion in six months and $20 billion over 12 months.)
A further $13.2 billion of the analysed $47 billion was paid to businesses that recorded falls in revenue of less than the 30 per cent threshold (or 50 per cent for large businesses).
A fundamental question is if we should care too much about businesses that posted declines in revenue that were smaller than anticipated and still received JobKeeper?
Arguably, it is a positive symptom of a successful economic and health response.
Treasury suggests the JobKeeper payments were reasonably well targeted.
The actual average decline in revenue for JobKeeper businesses was 21.7 per cent and the median fall was 27.9 per cent in the June quarter.
For non-JobKeeper businesses, revenues were broadly flat.
Treasury argues JobKeeper was effective in maintaining a connection between employers and employees during the early months of the national lockdown in 2020.
JobKeeper firms recording revenue falls that were not as bad as feared quickly rehired about 200,000 workers who had been temporarily laid off in March 2020 before JobKeeper was announced.
Nevertheless, Labor MP and economist Andrew Leigh says the government has failed to apply a comprehensive cost-benefit analysis.
“The fact that jobs were saved doesn’t mean every dollar spent was worthwhile,” Leigh says.
He points to preliminary estimates by Australian National University economists that the taxpayer cost of each full-time annual job saved was between $140,000 and $204,000.
Research by the Reserve Bank of Australia estimates JobKeeper saved between 700,000 and one million jobs in the first four months of the scheme, out of about 3.6 million employees enrolled.
Treasury puts forward a further argument that the scheme was reasonably well targeted; small business were big winners.
Some 99 per cent of the JobKeeper businesses that did not experience the 30 per cent revenue decline were small firms with less than $50 million in turnover and an average of four employees.
Moreover, small firms also accounted for 88 per cent of JobKeeper payments to firms that recorded rises in quarterly revenue.
Small businesses had much less capacity than large companies to tap capital markets for funding or to borrow from banks during the early months of the pandemic.
More broadly, while many firms did not experience a quarterly turnover decline, many did experience a sharp fall in revenue in the month of April 2020 – the first full month of the national lockdown.
According to Treasury, 40 per cent of businesses that report their turnover monthly did experience at least a 30 per cent (or 50 per cent) decline in revenue in April, even if their overall quarterly revenue didn’t decline as much.
Labor has focused its criticism on the quarterly numbers from a Parliamentary Budget Office analysis.
JobKeeper eligibility was based on forecasting one single month of the threshold revenue declines. No business had to assume revenue for the whole quarter would be materially lower.
“For these businesses, this outcome likely reflects the earlier-than-expected easing of restrictions and the surprisingly strong economic recovery,” Treasury notes.
That these businesses suffered a sharp downturn in revenue in April 2020 but were already recovering strongly by June raises the question of whether the payments through to the end of September went on too long?
The government knew by June 2020 that 15 per cent of JobKeeper payments were going to firms recording rising revenue and there were also private warnings from the Australian Taxation Office that some companies were gaming the system.
The government gave some tentative consideration to switching off the payments to businesses recording revenue growth or strongly recovering.
At the time in June, the economic uncertainty about the recovery was very high and there were also time lags in the economic data. The government was not sure the economy was out of the woods and a V-shaped recovery only became evident some months later.
Ultimately, Frydenberg accepted Treasury advice to err on the side of being a bit more generous to lock in the economic recovery.
A decision was taken that the fiscal cost of continuing JobKeeper in its initial form to the end of September was less than the potential cost of withdrawing support too early.
“Guaranteed support for six months provided certainty in a highly unpredictable environment and encouraged businesses to operate, adapt and innovate,” Treasury says.
“It was understood that this risked making payments to businesses that recovered quickly and may not need support by the end of this period.”
Treasury secretary Steven Kennedy drew on his experience as a top economic adviser to prime minister Kevin Rudd during the 2008 global financial crisis.
Kennedy’s experience rolling out the school hall and home insulation stimulus programs would have made him well aware of the inevitability of alleged waste in implementing a rapid-fire and huge stimulus.
But Kennedy had also witnessed Europe and America unwind their stimulus too quickly during the financial crisis and suffer sluggish economic recoveries.
By late September, a strong recovery was well established and the government felt confident to dial back JobKeeper, but continue lower payments for the worst-hit businesses based on actual – not forecast – turnover declines.
“JobKeeper was designed to support households and businesses during lockdowns so that the economy was in the best position to adapt to lockdown conditions and recover quickly once restrictions eased,” Treasury says.
“This broad fiscal support came through two channels: by directly supporting businesses and households and by supporting confidence and reducing uncertainty across the economy.”
Treasury counters calls for a “clawback” on JobKeeper overpayments or an earlier re-test of revenue eligibility before the initial six-month expiry.
Businesses may have deliberately suppressed revenue and slowed the recovery, it says.
“Claw-back mechanisms and re-testing for eligibility can operate like anti-production subsidies and perversely encourage businesses to reduce activity to qualify for support,” Treasury says.
“The decision not to include a claw-back mechanism in the design of the scheme or to retest eligibility through the first phase also reflected concerns about the potential for significant balance sheet and labour market scarring to impede the recovery.”
The share of businesses with turnover declines of slightly more than the 30 per cent (50,000 businesses with sales declines between 30 and 35 per cent) was much larger than the share with declines slightly smaller than 30 per cent (20,000 businesses with declined between 25 and 30 per cent).
An earlier re-test would have also reduced the overall amount of fiscal support – a factor Treasury and the RBA were conscious of.
JobKeeper was pumping $11 billion a month into the economy in the first six months, before declining to about $4 billion a month in the December quarter 2020 and $2 billion a month in the March quarter 2021.
The argument from policymakers is that the macroeconomic outcomes in the September quarter and beyond would not have been as strong if the JobKeeper tap had been turned off earlier.
“Both businesses and households benefit from having strong balance sheet positions that provide a financial buffer to absorb economic risk and economic shocks and allow spending to increase as conditions improve,” Treasury notes.
“The spending undertaken by one business or household, as a result of extra income, benefits other businesses and their workers, which flows on to broader spending elsewhere in the economy.
“All of these channels contributed to supporting confidence about the outlook, reducing precautionary savings behaviour and encouraging investment in new activities.”
It is almost certainly true that the overt positive employment and confidence outcomes would have been somewhat lower if JobKeeper was unwound faster for rebounding businesses.
Nevertheless, the gushing tap had to be tightened up at some point.
There was a range of other extraordinary supports in place as part of a $300 billion stimulus, including a $35 billion business cash flow program, higher JobSeeker payments, bank loan repayment deferrals, rent relief for commercial and residential tenants, insolvency relief and infrastructure investment.
The multi-pronged response from federal and state governments and banks, makes it very challenging to disentangle the contribution of each part of the support package.
Despite the second virus wave in Victoria and its four-month lockdown in mid-2020, there was also much better than expected health outcomes, and businesses and workers adapted better as the pandemic wore on.
One arguable policy shortcoming of the government is that it did not build enough “upside” positive scenarios into their stimulus responses.
Treasury and the RBA – like most private sector – forecasters repeatedly underestimated the speed and strength of the recovery.
JobKeeper’s design was largely built on a “downside” scenario, or if things got really bad.
Ditto, the $35 billion business cash flow payments to virtually all employing small businesses.
About half of the business cash flow went to firms and non-profits in the second half of 2020 when the crisis outside Victoria had passed and the economy was recovering strongly.
The non-means tested cash flow payments to most SMEs — including double-dipping by JobKeeper recipients — were worth between $20,000 and $100,000 per firm following the March, June and September quarters.
Underlining the overcompensation of business, non-financial corporation profits rose 10 per cent in the 2020 calendar year – during a huge recession.
Of course, with the benefit of hindsight there are many more experts than during the fog of war.
One investor privately reflects: “While I welcome the scrutiny of JobKeeper, most journalists, economists and politicians should re-read what they were saying 18 months ago when it was launched.”
“Most were in fervent support, and said just get it out, worry about the details later.
“Scrutiny is warranted, but I just do believe there is a lot of hindsight going and there was universal support for it.
“There should always have been safeguards. My issue is only a few pundits raised that at the time, and it was certainly possible to see it could be abused back then.”
Treasury itself notes that as things turned out, “the economy recovered more strongly than expected, illustrating the importance of flexibility in policy design, including having appropriate review points to ensure policy can be adapted as circumstances change.”
Greater inbuilt flexibility or willingness to adjust policies as the situation unfolded could have proven useful.
King and Kay note in their Radical Uncertainty book, “In most critical decisions there can be no forecasts or probability distributions on which we might sensibly rely.
“We should adopt business, political, and personal strategies that will be robust to alternative futures and resilient to unpredictable events.”