New payroll data from the ABS indicated job losses across all industries between 14 March and 4 April of six per cent, with losses exceeding 25 per cent in the accommodation and food sector and close to 19 per cent in arts and recreation. The CBA’s flash PMI readings for April showed activity in the service sector collapsing. The RBA governor said that the central bank now expects Australian output to fall by ten per cent in the first half of this year and be down by six per cent across 2020 while Q2 unemployment is likely to hit double figures. The global oil market has been in turmoil, with futures prices making an unprecedented trip into negative territory as a supply glut meets disappearing demand. Governments across the world have responded to the downturn with substantial fiscal stimulus which is translating into a sharp ramp up in global public debt. The IMF sees Australia’s general government deficit widening by six percentage points of GDP this year and our general government debt jumping by more than 14 percentage points.
This week’s readings look at the Australian labour market, the challenges involved in trying to buy your own barrel of oil, coronavirus policy lessons from wartime and ‘superforecasting.’
What I’ve been following in Australia . . .
The ABS released new experimental weekly estimates on the impact of COVID-19 on employee jobs and wages. The new numbers are sourced from the ATO’s Single Touch Payroll (STP) system and are part of the suite of new products the ABS has introduced to provide more timely information on the impact of the coronavirus.
This first release in the new series shows that between the week ending 14 March 2020 (the week in which Australia record its 100th confirmed COVID-19 case) and the week ending 4 April 2020 employee jobs decreased by six per cent and total wages paid fell by 6.7 per cent. The bulk of those declines took place in the last week of the three week period covered by the data – that is, between 28 March 2020 and 4 April 2020. Employee jobs fell by 5.5 per cent in that week compared to a decrease of 0.5 per cent in the previous week. Similarly, total wages paid fell by 5.1 per cent compared to a decrease of 1.3 per cent over the previous week.
By industry, the biggest proportionate declines in jobs over the whole three week period were in accomodation and food services (down more than 25 per cent) and arts and recreation services (down almost 19 per cent), followed by mining, rental, hiring and real estate services, and professional, scientific and technical services.
Unsurprsingly, the industry distribution in terms of declines in the total wage bill was similar, with the largest drops in accomodation and food services (down 30 per cent) and arts and recreation services (down almost 16 per cent).
By State and Territory, the biggest losses in employee jobs between 14 March and 4 April were in Tasmania (down 7.3 per cent) and Victoria (down 6.8 per cent) while the largest falls in the total wage bill were in the Northern Territory (down 17.6 per cent) and Victoria (down 8.8 per cent).
Why it matters:
We noted last week that because the March labour market report relied on survey data collected between 1 and 14 March, it missed the big adverse impact on jobs that had started to unfold during the second half of the month and then accelerated in the final week or so. This new ABS release confirms that the hit to the job market during that period was indeed extremely severe.
It’s worth noting there that the ABS cautions that the new data is ‘experimental’ and that these indicators only provide an indication of movements in total jobs and total wages and not estimates of levels. The ABS also notes that not all jobs in the Australian labour market are captured within these estimates: as of 6 April 2020, approximately 99 per cent of employers the ATO classifies as ‘substantial employers’ (that is, those with 20 or more employees) were reporting through STP but only about 71 per cent of small employers (those with 19 or less employees) were doing so.
Keeping those caveats in mind, it’s still useful to think about what the kind of percentage falls reported here imply for actual job numbers. One crude way to do this is to apply these declines to the distribution of Australian jobs by industry as reported by the December quarter labour account numbers.
If we do that, then this week’s numbers imply job losses approaching 300,000 in accomodation and food, exceeding 100,000 in professional and scientific services, and substantially exceeding 50,000 in administration and support services and construction. Again, these are not concrete estimates. But they do provide a feel for the kind of magnitudes involved.
We can do a similar back of the envelope calculation in terms of the implications for the unemployment rate. The March labour force survey suggested there were a bit more than 13 million Australians employed in the first half of the month, along with around 719,000 unemployed. That translated into an unemployment rate of 5.2 per cent. If we apply the six per cent job loss figure from the payroll data to the number of employed, that gives another crude estimate of total ob losses of more than 780,000. That in turn would imply an unemployment rate of almost 11 per cent by 4 April.
The good news is that this number is likely too high: we’re stitching together different data sets, there’s an imperfect match between jobs and employment, and we’ve assumed no changes in behaviour (an offsetting drop in the particiaption rate seems likely, for example). The bad news is that forecasts of double-digit rates of unemployment in the second quarter of this year now appear to be the official consensus.
RBA Governor Philip Lowe delivered his first speech since his 19 March presentation setting out the details of the central bank’s policy package in response to COVID-19. This time the topic was an economic and financial update with a focus on the immediate outlook for the economy and on the possible nature and speed of any recovery.
In line with last week’s discussion of the economic prospects, the RBA thinks that the first half of this year will see the Australian economy ‘experience the biggest contraction in national output and income that we have witnessed since the 1930s.’ After stipulating the difficulties involved in producing any kind of forecasts in the current environment, the governor said the RBA now expects:
- National output to fall by around 10 per cent over H1:2020, concentrated in the June quarter.
- Total hours worked to decline by around 20 per cent over H1:2020.
- The unemployment rate to be around 10 per cent by June.
- And it ‘is quite likely’ that annual headline inflation will turn negative in the second quarter, thanks in part to the slump in global energy prices.
What about the outlook into the second half of the year and beyond? Here Lowe noted that prospects were heavily conditional on the future status of Australia’s public health measures, but went on to suggest that one ‘plausible scenario is that the various restrictions begin to be progressively lessened as we get closer to the middle of the year, and are mostly removed by late in the year, except perhaps the restrictions on international travel.’ Based on that scenario, ‘we could expect the economy to begin its bounce-back in the September quarter and for that bounce-back to strengthen from there. If this is how things play out, the economy could be expected to grow very strongly next year, with GDP growth of perhaps six-seven per cent, after a fall of around six per cent this year.’
Even with that somewhat V-shaped recovery, however, the RBA also thinks that unemployment will be will stuck above six per cent over the next couple of years, wage growth will dip below two per cent, and underlying inflation will likewise remain below two per cent over this period too. So, not much of V really.
The governor also included a review of the RBA’s policy measures, noting that the new policy of yield curve control (YCC) introduced in the 19 March package of special measures had been successful in lowering the yield on three-year government bonds to around the target level of 25 basis points, after having been around 50 basis points immediately prior to the announcement. He noted that there had been a marked improvement in liquidity in the Australian government bond market. That combination of hitting the bank’s target yield and an improvement in market conditions had also allowed the RBA to scale back the size of its daily bond purchases.
Why it matters:
Governor Lowe’s speech provided a useful insight into the RBA’s expectations regarding the economy over the next few months. The projection for a 10 per cent unemployment rate in Q2 lined up with the Treasury forecast discussed in last week’s note. The double-digit drop in GDP expected to be concentrated in the second quarter looks consistent with the big declines in business and consumer confidence we’ve already seen, as well this week’s collapse in the service sector PMI (see below). And the overall forecast for a GDP decline of around six per cent this year is in the same ball park as last week’s IMF forecast of a 6.7 per cent fall.
Things get much more uncertain beyond that, of course, although based on the current trajectory of new COVID-19 cases, the RBA’s scenario appears plausible. If an unwinding of the lockdown from the middle of the year does prove to be the case, the projected bounce back of six to seven per cent would be welcome (and again, is roughly in line with IMF forecasts of a 6.1 per cent growth result in 2021). Much less welcome will be the prolonged period of labour market sluggishness the RBA anticipates, with an unemployment expected to be stuck well above its pre-CVC natural rate of 4.5 per cent dragging down wage growth and therefore capping domestic demand. If that’s going to be the case, then once the initial ‘snap back’ in growth has happened, the outlook for activity seems destined to look very soft again. After all, remember that real growth before the crisis, in 2019, was a meagre 1.8 per cent.
The governor awarded the RBA a deserved pat on the back for its success to date in stabilising Australia’s government bond market. That in turn has allowed it to scale back its bond buying program. The latter got underway with a $5 billion purchase on 20 May but has since been tapered to about a tenth of that. As at the time of writing, the RBA had bought a total of $48.75 billion of Australian government debt.
The ramp-up in government bond issuance over coming months will offer an interesting test of the durability of this tapering exercise.
Finally, it’s worth noting that in the Q&A session following the speech, Lowe took the opportunity to make the case for several long-standing economic reforms that would help boost the economy’s performance in the medium-term. The governor’s wish list included ‘the way we tax income generation, consumption and land’, the flexibility of the industrial relations system, the building and pricing of infrastructure, the nature of education and training, and the regulatory framework applying to innovation. A familiar bit of economic policy advice is to never waste the opportunities offered by an economic crisis or recession, and this was the RBA governor taking his chance to set out some priorities. Readers familiar with the AICD’s own DSI results will see a strong overlap with directors’ priorities in Lowe’s list.
The CBA Flash Composite PMI Index slumped from 39.4 in March to 22.4 in April. That was largely driven by a collapse in the services PMI, which fell from 38.5 to 19.6. The drop in manufacturing, while still significant, was much more modest, with the index tumbling from 49.7 to 45.6.
The PMI readings on jobs also showed sharp declines for both services and manufacturing in April, consistent with the severe hit to the labour market already underway in late March as set out in the ABS payroll data discussed above.
(So-called ‘flash’ PMIs are based on about 85 per cent of the final survey responses and are intended to provide an early reading before the final, full set of results are available – usually about a week later.)
Why it matters:
The April PMI readings indicate a dramatic collapse in service sector activity as well as a significant contraction in manufacturing.
The April results also picked up an interesting difference in the experience of services vs manufacturing in terms of input costs. In the case of the former, lower wages and falling fuel prices translated into lower input cost pressures. But manufacturers reported that the fall in the dollar and some shortages in raw materials has driven an increase in cost pressures.
Australian retail turnover rose 8.2 per cent over the month in March 2020 (seasonally adjusted) and was up 9.8 per cent over the year, according to preliminary data from the ABS. These estimates are another of the new ABS products and are based on early data provided by businesses that make-up approximately 80 per cent of total retail turnover.
The big jump in March was driven by the food retailing industry with supermarkets and grocery stores, liquor retailing, and other specialised food retailing all recording increases in demand. The food industry rose 23.5 per cent, with the supermarket and grocery store subgroup rising 22.4 per cent. ABS analysis found that monthly retail turnover for perishable groceries and all other groceries soared last month, with turnover doubling relative to February for a range of products including toilet and tissue paper, flour, rice and pasta, and with turnover for canned food, medicinal products and cleaning goods rising by more than 50 per cent. The ABS also reported strength in other non-food sub-groups, including in electrical, hardware and other retailing, where businesses reported an increase in sales of items related to the set-up of home offices, for example.
All of this activity more than offset some large falls in reported turnover in cafes, restaurants and takeaway food services, in clothing, footwear and personal accessory retailing and in department stores.
Why it matters:
This was the strongest rise recorded in the history of the data series, surpassing the previous record increase of 8.1 per cent in June 2000 when households brought forward expenditure ahead of the introduction of the GST. As such, it’s a measure of the intensity of precautionary buying (and therefore the underlying behavioural change) triggered by COVID-19.
The RBA published the minutes of the 7 April Monetary Policy Meeting of the RBA Board. The minutes note that the RBA’s policy package announced on 19 March had ‘helped to lower funding costs and stabilise financial conditions’, and ‘contributed to an easing of strains in government bond markets.’ They also confirmed that the RBA Board ‘reaffirmed the elements of the policy package announced on 19 March 2020.’
Why it matters:
With the 7 April meeting following on quite soon after the RBA had announced its package of policy measures on 19 March, the expectation had been for no change in policy settings to be announced this month, and no change was what was delivered. In that context, the minutes would normally be of interest mainly to the extent that they offered any additional insight into the RBA’s thinking around future policy settings. But the governor’s latest speech has now given us a more up to date take on the central bank’s perspective, so the April minutes are included here as much for completeness as anything else.
. . . and what I’ve been following in the global economy
On 20 April the futures price for West Texas (WTI) crude fell to minus US$37.63 a barrel. At one point, the price per barrel was trading below minus US$40. At the time of writing prices were back in positive territory, but the drop marked an historic moment: this is the first time that this has happened. Ever.
Why it matters:
The global economy has been a crazy place in recent times, and we’ve had to get used to some very peculiar things, not least negative central bank interest rates. And now we can add negative oil prices to the list of weird stuff.
If you’re interested in digging into the details of what happened on Monday, there are a couple of useful links in this week’s reading list below, but the short story is that the global oil market has been buffeted by simultaneous supply and demand shocks:
- The supply shock came in the form of an oil price war between Saudi Arabia and Russia that has seen both producers pumping huge amounts of oil onto the market. There was supposed to be a peace deal (brokered by the United States and the G20) to calm things down, but the general assessment has been that the proposed production cut of 10 per cent or a bit less than 10 million barrels per day (mb/d), may have been ‘too little, too late.’
- The demand shock has, of course, been triggered by COVID-19, which the International Energy Agency (IEA) has estimated will see global oil demand fall by a record 9.3 mb/d year-on-year in 2020. Demand in April is estimated to be an incredible 29 mb/d lower than a year ago, down to a level last seen all the way back in 1995.
The resulting oil glut has seen global oil storage facilities approaching full capacity, meaning space for storing physical oil has not just been getting scarce but also (and therefore) very expensive. Futures contracts for May delivery were due to expire on 21 April which meant that on 20 April there were a bunch of traders with oil contracts falling due. Faced in some cases with the prospect of having to receive and then find costly storage for physical oil, it turned out that the less bad option for them was to sell their contacts at a negative price. That is, to pay someone else to take them off their hands.
By 21 April prices were back in positive territory again, but the fundamental problem facing the oil market of a major demand-supply imbalance is still in play.
The IMF has estimated that current global health spending plus government tax and spending measures to combat COVID-19 comes to about US$3.3 trillion. Additional support in the form of public sector loans and equity injections (US$1.8 trillion) and govenrment guarantees and other contingent liablities (US$2.7 trillion) takes total estimated fiscal support to US$7.8 trillion. Add in the loss to budget revenue from the decline in economic activity (the Fund reckons that worldwide budget revenues will be cut by 2.5 percentage points of GDP) and the result is a substantial shift in global fiscal accounts. As a result, the IMF’s latest projections see the GDP weighted averge general government global fiscal deficit risign from 3.7 per cent of world GDP in 2019 to 9.9 per cent of GDP this year. That in turn is expected to see global general government debt rise from 83.3 per cent of GDP last year to a forecast 96.4 per cent of GDP in 2020.
Much of that fiscal activity is taking place among advanced economies, where the Fund sees the GDP weighted average general government defcit rising from three per cent of GDP last year to 10.7 per cent of GDP this year. General government debt is projeced to rise from 105.2 per cent of GDP to 122.4 per cent of GDP over the same period.
According to the Fund’s data, Australia’s so-called ‘above the line’ fiscal support (that is, general tax and spending measures that directly influence the budget bottom line) stands out for its size. At more than 10 per cent of GDP, it’s one of the most aggressive responses amongst its advanced economy peers, with only Japan deploying a similar level of budgetary largesse and well-ahead of even the big spending United States (6.9 per cent of GDP).
Other advanced economies have been relatively more reliant on the deployment of ‘below the line’ fiscal measures (loans, guarantees, contingent liabilities), with Italy (more than 30 per cent of GDP) being particularly notable here, but with other big commitments from the UK (about 15 per cent of GDP) and France (about 14 per cent).
Why it matters:
As we discussed in the context of Australia a couple of weeks ago, the Coronavirus Crisis (CVC) is set to have a substantial impact on debt and deficits. The numbers in the IMF’s Fiscal Monitor show that the Fund thinks Australia’s general government budgetary position will deteriorate by a hefty six percentage points of GDP this year. But that will still be less than many other advanced economies and comfortably below the advanced economy average of a 7.7 per cent of GDP swing.
Those Fund projections imply that Australia will run a total general government deficit of around 9.7 per cent of GDP in 2020. Again, that will still be below the advanced economy average of 10.7 per cent of GDP but in this case, note that the latter has been boosted by a massive US budget shortfall which the IMF puts at more than 15 per cent of GDP.
These deficit numbers also imply correspondingly large increases in government debt. In Australia’s case, the IMF thinks that general government debt will rise by more than 14 per cent of GDP this year. That compares to a projected increase across all advanced economies of around 17 per cent of GDP.
Finally, however, those big forecast increases in our debt burden still leave our net stock of government debt looking relatively comfortable by international standards. The IMF’s numbers have our net government debt to GDP ratio at 35 per cent this year. That compares to an advanced country average that is expected to be more than 94 per cent of GDP.
Last Friday, China’s National Bureau of Statistics (NBS) reported that GDP growth in the first quarter of this year fell by 9.8 per cent over the quarter and was down 6.8 per cent in annual terms.
Why it matters:
This data release is from last week but is still worth highlighting given the role that China has been playing as a kind of leading economic indicator for the rest of the world economy during the CVC. The key point here is that this is the first contraction in quarterly GDP since China began reporting the series back in 1992. And according to some analysts, it represents the worst growth outcome (officially) recorded since the 1970s.
Even the powerhouse that is the Chinese economy has been unable to sustain its growth momentum in the face of COVID-19 (although note that last week’s IMF forecasts do assume that China will still be one of the few countries to manage to grow its real GDP over the whole of 2020).
What I’ve been reading
The Grattan Institute has a go at estimating the size of the labour shock the Coronavirus Crisis (CVC) will impose on the Australian economy. They reckon that the unemployment rate is set to rise to between ten and 15 per cent.
ANZ Bluenotes on COVID-19 and Australian agriculture.
Heather Smith and Allan Gyngell think about the post-CVC world order in the AFR.
This is just a transcript of a government press conference, but it did include a useful update on the status of some key policies. For example, the ATO has now approved 456,000 applications allowing Australians early access to their superannuation, totalling $3.8 billion with an average withdrawal of around $8,000. The Treasurer also reported that the ATO has now paid out $3 billion to 177,000 businesses employing 2.1 million workers as part of the government’s cashflow support for business. And more than 900,000 businesses have registered their interest in accessing the JobKeeper payment, with 275,000 businesses now having completed formal applications, of which around half are sole traders.
Econbrowser explains negative oil prices. See also this from Visualcapitalist.
Via FT Alphaville (and from back in 2015), what happened when a Bloomberg journalist attempted to buy an actual barrel of oil.
And here FT Alphaville casts a sceptical eye over an MIT proposal for opening up the economy as part of its long-running ‘axes of evil’ series.
This is a handy link to BIS COVID-19 related research.
Blanchard and Pisani-Ferry think that we don’t need to worry about the monetisation of all those ballooning fiscal deficits discussed above.
Coronavirus lessons from World War Two.
Related, an argument that economies (well, the US economy in this case) will grow out of our debt burdens.
The Economist magazine’s Free Exchange discusses the possibility of a return to inflation.
A WSJ piece
on COVID-19 and inequality. Past pandemics have had significant distributional impacts, shifting the balance of power towards labour (on the rather brutal grounds that large death rates imply a negative labour supply shock). Successful public health interventions mean that the coronavirus is unlikely to repeat this (hi)story. But will there be other impacts instead?
This is a great episode of the Macro Musings podcast with Nicholas Bloom. A bunch of interesting ideas and arguments to be found here.
And another good listen and also the transcript if you prefer to read: Tyler Cowen in conversation with Philip ‘superforecasting’ Tetlock.