Treasury revealed its huge recalculation in a joint statement with the Australian Taxation Office, which also revealed there had been a large reporting error in estimates of the number of employees likely to access the program.

The costings revision, while highly embarrassing for treasury, is extremely good news for the government, which had committed around $200 billion to support measures to get the country through the pandemic.

Treasurer Josh Frydenberg said: “It is welcome news that the impact on the public purse from the program will not be as great as initially estimated”.

The government will be able to use money saved to reduce projected deficit and/or for other spending.

The news of the revision immediately prompted calls for JobKeeper to be widened to include workers, especially many casuals, who are not covered under its present rules.

But Frydenberg told the ABC: “We’re not making wholesale changes to the JobKeeper program. We’ll have a review, as we’ve always stated, mid-way through the program, and we’ll wait for the results of that review if there are to be any changes.”

When JobKeeper was developed, Treasury anticipated about 6.5 million employees would access the program, which provides a flat $1500 a fortnight for workers who remain connected to their employer. The assistance is available to employees of businesses which have had at least 30% fall in their turnover, or 50% in the case of big businesses.

Writing in The Conversation in late April, Melbourne University economists Roger Wilkins and Jeff Borland pointed to a disparity between the dive of 2.6 million full time jobs expected by Reserve Bank Governor Philip Lowe and the 6.6 million jobs the Treasury was preparing to fund under JobKeeper.

“What is surprising is the size of gap between the predicted number of payments and the predicted number of jobs at risk,” they wrote.

Treasury now expects only some 3.5 million workers to need JobKeeper.

While the Treasury revision of the scheme’s likely cost is driven by the fact circumstances have not born out its original assumptions, a reporting error by many businesses masked what was actually happening, so treasury’s numbers for a time appeared correct.

At a Senate committee hearing on Thursday Treasury was still talking about the program covering more than six million employees.

Explaining the wrong forecast, Friday’s statement said the original cost estimate was made when COVID-19 cases were “growing significantly” in Australia and restrictions were being tightened here and abroad.

“The difference between Treasury’s estimates at the time and the number of employees now accessing the JobKeeper program partly reflects the level and impact of health restrictions not having been as severe as expected and their imposition not having been maintained for as long as expected at the time.

“This has been reflected in some improvement to the outlook for the economy since the original estimate was developed,” the statement said.

“The variation in estimates also reflects the inherent uncertainty associated with estimating the take up of a demand driven program in the current circumstances.”

The enrolment forms completed by 910,055 businesses had indicated the program would cover about 6.5 million eligible employees – in line with treasury’s thinking.

But the ATO has now found about 1,000 of these businesses had made big mistakes in estimating eligible workers.

“The most common error was that instead of reporting the number of employees they expected to be eligible, they reported the amount of assistance they expected to receive.

“For example, over 500 businesses with ‘1’ eligible employee reported a figure of ‘1,500’ (which is the amount of JobKeeper payment they would expect to receive for each fortnight for that employee).”

The reporting error does not affect the payments already made to businesses.

This is because those payments are linked to a later declaration from a business in relation to every eligible worker.

“This declaration does not involve estimates and requires an employer to provide the tax file number for each eligible employee.”

The information where the reporting error occurred was just collected to obtain an early indication of how many employees were likely to go onto JobKeeper.

The mistakes were detected when the Tax Office investigated the large gap between the expected number who would go onto JobKeeper and the much smaller number actually accessing it.

By May 20, 910,055 businesses had enrolled in the program, with 759,654 making claims for eligible employees; $8.7 billion had been paid to those businesses, covering about 2.9 million employees.

Anthony Albanese said: “This is a mistake you could have seen from space. This is a government that couldn’t run a bath, let alone be good economic managers.”

Calls for the scheme to be revamped came from both the employer and employee sides.

The Australian Industry Group said the government should address the program’s anomalies and alter “rules which leave many employees without support and mean that many employers are facing unfair competition.

“With the estimated budgetary costs reduced by $60 billion there is considerable scope for refinements to the program,” the Ai Group said.

It urged the inclusion of low-margin businesses which had not had a 30% reduction in turnover but were “under greater stress than higher-margin businesses that do qualify for Jobkeeper”.

The ACTU tweeted “We have millions of workers who were left out of #JobKeeper on the premise that there wasn’t enough money. Now we know that it’s been underspent by $60 billion. There is no excuse – @JoshFrydenberg can fix this with a stroke of his pen. Expand JobKeeper now.”

The Transport Workers’ Union said the government should immediately pay the thousands of airport workers shut out of the scheme.

The Greens said the revision meant the government had “no excuse to return the Jobseeker payment to $40 a day at the end of September”.

Treasury still expects unemployment to reach 10% and says it would have reached 15% without JobKeeper.

Michelle Grattan, Professorial Fellow, University of Canberra

This article is republished from The Conversation under a Creative Commons license. Read the original article.