On Friday, we had the extraordinary admission from our Federal government that they had overestimated the injection required to fund their wage subsidy JobKeeper program by some $A60 billion. When the overall program was announced the Treasury allocated $A133 billion to it. So now they are admitting to a 45 per cent forecasting error, which sort of dwarfs the worst errors that the IMF makes, and they sure make some bad mistakes in their projections. Whatever the reason for the mistake, the way the Treasurer has defended it is quite repugnant – claiming virtue out of the incompetence. And while all the Labor Party economists are talking about seeing the error from space, none of them picked it up or had the nous to realise that the figures didn’t add up when the Government originally released them. I am the only economist who wrote that the figures published by the Government didn’t make sense. I did that on April 29, 2020. I also wrote to the Treasury and the Treasurer requesting answers to questions that reflected my concern. They didn’t bother replying. Now everyone is wise after the fact. Anyway, the $A60 billion is a nice round figure. And I outline a plan in this blog post on exactly how the Treasurer can spend it and improve the well-being of more than a million Australians with a stroke of the pen.
As everyone knows now, the boffins in Treasury have made the largest official forecasting error in Australian history by admitting that their celebrated JobKeeper wage subsidy program, which they claimed would inject $A133 billion over 6 months into the economy and reduce the unemployment rate by 5 percentage points (from 15 per cent to 10 per cent), would now actually cost $A60 billion less than thought due to an ‘accounting’ or ‘form fill out’ error.
Here is their press release (May 22, 2020) – JobKeeper update.
Instead of 6.5 million workers being covered, the Treasury now claims that only 3.5 million workers will receive payments.
Strangely, the Treasury continues to claim:
Treasury’s overall view of the labour market is unaffected by this reporting error … It remains the case that in the absence of the JobKeeper program, Treasury expects the unemployment rate would have been around 5 percentage points higher. Treasury continues to expect the unemployment rate to reach around 10 per cent …
Regular readers will know that on April 29, 2020, just after the JobKeeper scheme was announced, I wrote this blog post – JobKeeper wage subsidy – some strange arithmetic is afoot (April 29. 2020).
It was clear to me that the estimates provided by the government didn’t add up or jive with reality.
I was the only economist who made that claim at the time – conservative or ‘progressive’.
I wrote to the Treasury and the Treasurer’s office at the time with a series of question and requests for more information pointing out what appeared to be errors.
In the past, academic researchers such as myself have often written to government departments to seek such information or briefings and the government typically has responded positively.
This time, I didn’t receive a reply from either the Treasurer or the Treasury. Closed shop these days.
It is a pity that the Treasury didn’t take heed of that post and my questions.
I wonder who is going to take responsibility for this scale of error.
There was no way the original estimates could have been correct.
It is all very well for the Treasury to claim there were bookkeeping errors but their economists should have had the nous to realise that it all didn’t add up. It was obvious.
The Labor Opposition Leader put out a Tweet about the error on Friday evening (20:01):
A mistake you could have seen from space.
If they can’t get this right, then how can they be trusted to get the recovery right?
But, no-one in the Labor Party picked the error up.
The shadow treasurer didn’t pick it up and praised the JobKeeper program.
Another economics spokesperson, who likes to claim MMT is nonsense, didn’t pick it up.
None of the Labor advisors – formal or informal – didn’t pick up the absurdity of the numbers at the time. They tweet a lot about themselves and what smart guys they are, but couldn’t even cope with simple arithmetic.
It is easy to say that you “could have seen … [the error] … from space” after the fact.
Worse still – trying to make good out of a disaster
At his press conference on Friday, the Treasurer tried to make virtue of the fact that his government was starving a severely recessed economy of $A60 billion when just a week before the ABS had published horrific labour market data – see below.
In an Interview on the ABC television (May 22, 2020) – Job Keeper Interview – he told the host that:
1. “these numbers are good news for the taxpayers. $60 billion less than would have otherwise been spent because the Australian economy has not deteriorated by as much as Treasury initially forecast. This is all borrowed money …”
The host ticked that lie.
2. He kept repeating the line “This is good news for the taxpayer” and “its all borrowed money”.
3. “Every dollar that we’re spending is borrowed money. It means that our children will be paying it back. It will take years to pay back the accumulated debt from the coronavirus period, there’s no doubt about that.”
4. “The JobKeeper program and the JobSeeker program are to work in complement to each other. One has got 1.6 million workers on it, the other will have 3.5 million workers on it according to Treasury. That’s nearly half the Australian labour force.”
In actual fact, it is only 37 per cent of the Labour Force as at April 2020 – nowhere near nearly half.
But in this era of $A60 billion errors anything is close enough!
The reality is this.
What an Modern Monetary Theory (MMT) understanding furnishes one with is the insight that if there is mass unemployment then you know at least one other thing – that the fiscal deficit is too small.
In extraordinary situations, this statement might be recast as the fiscal surplus is too large.
But in recessionary times, when there is rising unemployment, the government has all the financial capacity it needs to stop those workers losing jobs.
As I note below, this government has overseen a massive rise in labour underutilisation over the last several weeks.
Now it can be said that a significant proportion of the job losses (and labour force exits) have arisen as a result of the lockdown rules being imposed by state and federal governments.
But whatever the cause the workers lose their incomes and a proportion go on unemployment benefits (JobKeeper) and those that exit the labour force as a result of a lack of job opportunities miss out altogether.
So for the Treasurer to be happy that he is no longer spending the $A60 billion is tantamount to saying he has chosen to allow around 1,480 thousand workers to be without jobs.
He is idly sitting by while the lives of these workers is trashed.
Further, as I will demonstrate tomorrow when I analyse all the debt and central bank purchases of government debt that his statements about ‘borrowed’ money are very misleading.
The bottom line is that before this mistake was realised, the fiscal stimulus was grossly inadequate.
Trying to make virtue of the fact that you are spending $60b less than you thought when you were not spending enough in the first place while the unemployment rate skyrockets should disqualify one from office.
Anyway, here is a surefire way to spend the $A60 billion and dramatically improve the well-being of the most disadvantaged.
I have a plan for the $A60 billion, Josh!
CofFEE has updated our modelling of the investment required to reduce the unemployment rate from 10 per cent to 4 per cent.
The latest updated reflect the deterioration of the labour market as captured in the April Labour Force data, which I analysed in this blog post – Australian labour force data – underutilisation rate rises to 23.4 per cent (May 14, 2020).
In April, the Australian Bureau of Statistics told us that:
1. Employment fell by 594.300.
2. Official unemployment rose by 104,500, which took the unemployment rate to 6.2 per cent.
3. The participation rate fell by 2.5 points and was 2.7 points below the August 2019 peak of 66.2 per cent.
4. That difference amounts to 554.4 thousand workers who lost their jobs and dropped out of the labour force, thus, not being counted among the official unemployed.
5. Underemployment rose by 4.9 points to 13.7 per cent and when added to the official unemployment generated a total labour underutilisation rate of (a staggering) 19.9 per cent or 2,639.4 thousand workers.
6. But if you add the workers who dropped out of the labour force during April 2020 as a result of the collapsing employment opportunities (in other words, they told the ABS they were not actively searching) then the total unemployment is closer to 9.7 per cent rather than 6.2 per cent.
And the total underutilisation of available and willing labour is closer to 23.4 per cent rather than 19.9 per cent.
In other words, a disaster, however you want to spin it.
I provided some estimates of the investment required to introduce a Job Guarantee in Australia based on the March 2020 labour force data in this blog post – A Job Guarantee would require $A26.5 billion net to reduce the unemployment rate by 6 percentage points (April 30, 2020).
But with the rapid and substantial deterioration of the labour market in April, we decided to rerun our model to take into account the increased scale of the problem.
The exercise we constructed was not to get the economy back to full employment, which we believe is on that has an official unemployment rate of 2 per cent or less, zero underemployment and participation rates at their peak (that is, no hidden unemployment).
Our estimating framework allows us to model that scenario but this time we chose something less than that objective for reasons we explained in that earlier cited blog post.
Our view is that the fiscal intervention proposed by the Federal government is grossly inadequate in size and poorly targetted in scope.
The $A60 billion admission reinforces that view.
As part of the stimulus package, the Australian Treasury estimated that the unemployment rate would reach 10 per cent by June 2020. They claimed the wage subsidy would keep the unemployment rate from rising to 15 per cent.
Well given the level of incompetence displayed by that agency that was revealed last Friday, anything could happen.
But in our modelling exercise:
1. We took the June 2020 estimate of 10 per cent unemployment as a starting point.
2. The pre-GFC low-point unemployment rate (February 2008) was 4 per cent. So, we entered this crisis in a worse position than just before the GFC.
3. We also recognised that the potential labour force at an unemployment rate of 4 per cent would be much larger than it was estimated to be by the ABS in April 2020. We thus used the August 2019 participation rate as the benchmark and added the hidden unemployed to our estimates of jobs required.
4. This what we found based upon the updated data – if the unemployment rate was 10 per cent and we wanted it to be 4 per cent, the government would have to create:
(a) 1,212.4 thousand jobs in total (assuming the participation rate doesn’t change any further). Of those, 705.8 thousand would go to those who were in the labour force in April 2020, and 506.6 thousand would go to those who had dropped out of the labour force in April 2020.
(b) 1002.9 thousand full-time jobs (assuming the full-time/part-time mix doesn’t change).
(c) 209.5 thousand part-time jobs.
5. Based upon the current national minimum wage in Australia ($A740.62 per week), an average full-time working week of 38 hours, 30 per cent on-costs (hiring costs), plus additional capital costs per worker (plant, equipment) of 35 per cent, the total annual full-time Job Guarantee job outlay would be $A77,024.48 and the annual wage received by the worker would be $A38,512.24
The capital costs etc were estimated from a survey my research centre – Centre of Full Employment and Equity (CofFEE) – undertook of Local Governments to assess supervision, plant and other costs that would be required to supplement each Job Guarantee job.
The 35 per cent assumption is based on a labour intensive suite of jobs.
The outlays for part-time workers are proportional to the ratio of average part-time to full-time working hours.
You can read about that survey and all the details of our research work in this report – Creating effective local labour markets: a new framework for regional employment policy.
6. We took into account:
(a) Labour productivity – and created a ‘composite’ job.
(b) Multiplier effects of each new Job Guarantee job – so how much does GDP and consumption expenditure increase.
(c) In turn, how this impacts on corporate profits and non-government sector employment.
(d) The extra tax revenue the government receives as a result of the higher employment levels both in the public and private sector.
(e) We also differentiated between the workers who would be receiving unemployment benefits and those outside the labour force who would not in the calculations of the increased wage income.
This distinction also conditioned our estimates of the reduced outlays on unemployment benefits under the assumption that only a proportion of the workers who were employed in the Job Guarantee were receiving income support payments (those in the labour force), which would be terminated.
This clawback was also calibrated according to estimates of different income support payments by family structure and marital (now called partnership) status.
It is important to note that our modelling did not seek to add the jobs required to eliminate underemployment. We will include that when we release the full report in a week or so.
This is the current underemployment situation (as at April 2020):
1. Total Underemployment (000s) 1816.1 thousand.
2. Underemployed preferred extra hours (hrs/week) 15.0
3. Total FTE jobs required to eliminate underemployment – 638.1 thousand.
The revised results – and the $A60 billion
To reduce the unemployment rate by 6 percentage points (say from 10 per cent to 4 per cent – in fact, it doesn’t matter where you begin the simulation), then the following results are relevant.
|Total Rise in Employment (000s)||Job Guarantee||Private Sector|
Why does the private sector also expand?
Because the income stimulus to unemployed workers at the start (which is the difference between their Job Guarantee wage and the unemployment benefit) provides stimulus for increased consumption spending, which induces further income growth and via the multiplier effects a much larger final increase in national income and that stimulates the growth in private sector employment and corporate profits.
So introducing the Job Guarantee dramatically improves the material circumstances of the most disadvantaged.
But it also improves the situation for employers and profit recipients (which, of course, is not the driving motivation but a positive derivative impact).
To accomplish that increase in employment, the Federal government would outlay:
|Investment||$A billion||Per cent|
|Gross investment required over 12 months||80.6||Net investment required over 12 months||53.9||Clawback over 12 months||26.6|
|Change in government share of GDP||3.6|
What explains the clawback?
The Government receives extra personal tax revenue ($A6.99 billion), extra corporate tax revenue ($A3.8 billion), reduced unemployment benefit outlays ($A8.7 billion) and increased indirect taxes ($A9.4 billion) at current policy parameters.
And that $A60 billion?
To implement a Job Guarantee and reduce the unemployment rate by 6 percentage points:
1. The Federal government would have to outlay just $53.9 billion in net terms over the next 12 months.
2. The wage subsidy Job Keeper program involved gross outlays of $A133 billion minus the mistake of $A60 billion to reduce the unemployment rate by 5 percentage points over a 6 month period.
3. Do the sums!
It makes no sense to oversee a labour market with drastic rises in unemployment and underemployment, when all that has to happen is for the government to announce a Job Guarantee.
That is enough for today!
(c) Copyright 2020 William Mitchell. All Rights Reserved.