economic downfall

And here in Australia the week’s news on consumer confidence and business activity points to a similarly abrupt and painful halt, even as the number of announced job layoffs starts to mount. We did get some good news in the form of a massive US$2 trillion stimulus package passing the US Senate and on its way to the House, the prospect of which had been steadying US financial market nerves and even prompted a three-day Wall Street rally, despite an absolutely enormous surge in US initial jobless claims.

What I’ve been following in Australia . . .

What happened:

The ANZ-Roy Morgan weekly measure of consumer confidence slumped to a 30-year low of 72.2 last week. That 27.8 per cent drop left the headline index just above the all-time lows recorded in 1990 and 17 per cent below the lowest point reached during the GFC in October 2008.

graph

The economic conditions subindices also fell sharply with ‘current economic conditions’ dropping 37 per cent and ‘future financial conditions’ dropping almost 26 per cent. In a further blow to retail conditions, ‘time to buy a major household item’ plummeted by more than 37 per cent.

(Covid-19 has required a shift in survey methodology from face-to-face to phone and online interviews. ANZ Economics noted that this might have impacted the results but judged that the consistency of responses made this unlikely.)

Why it matters:

Economic indicators are now picking up the dramatic impact of the coronavirus pandemic and the policy response that are providing further support to the view that a major hit to the real economy is now unfolding – no surprise against a backdrop of news of mounting job layoffs and shuttered businesses.

A weekly fall in consumer sentiment that took us past the depths of the GFC and back to the 1990-91 recession implies that for Australia the ‘CVC’ is likely to be worse than the GFC.

What happened:

The latest CBA Flash Composite Purchasing Managers’ Index showed (pdf) a steep decline in business activity in March. The overall index fell from 49.0 in February to 40.7 in March, where a result below 50 signals a deterioration on the previous month. (So-called ‘flash’ measures are based on about 85 per cent of the survey responses that make up the PMI.)

graph

The drop was largely driven by services activity, with the services PMI falling from 49 last month to 39.8 in March amid declining new business orders, especially from overseas.

In contrast, manufacturing was relatively resilient, only edging down from 50.2 to 50.1, although the reported decline in the manufacturing output index was the fastest drop in the survey’s history.

The survey also reported a sharp drop in confidence about the prospects for output growth over the coming year, with sentiment falling to the lowest level on record (although note that, overall, companies still expected activity to be higher than current levels in 12 months’ time).

Why it matters:

The Flash March PMI confirms what we would have predicted by watching events unfold over the past couple of weeks: that the virus and the associated policy response were starting to have a big impact on the services sector at the time of the survey, and that the scale of that impact has continued to grow, with many businesses finding activity has been brought to a juddering halt over the space of a just a few days or weeks.

">graph

And as already noted, with services a critical source of employment, the impact of that shock is already being felt in reports of soaring job losses.

graph

Exposed sectors such as retail (the source of 1.4 million jobs as of Q3 last year) and accommodation and food services (1.1 million jobs) are major employers and while there will be offsetting increases in some subsectors (supermarkets, delivery) as well as in other sectors (health, public administration and safety) employment will be vulnerable across much of the economy as the effects from the initial round of shut downs and layoffs start to build.

What happened:

Economic forecasters are rolling out their initial estimates of how badly the CVC might impact economic growth and employment in Australia.

According to the latest monthly Bloomberg survey (covering 34 economists polled between 19 and 24 March), the annual pace of GDP growth is expected to be negative in all but the first quarter of this year, and average 1.3 per cent over 2020 as a whole, while the unemployment rate is forecast to rise to 7.2 per cent by the fourth quarter. That compares to the February 2020 Bloomberg survey which saw GDP growth this year averaging 2.1 per cent and unemployment ending the year at 5.2 per cent.

Moreover, that headline consensus figure conceals some dramatic differences between forecasters, likely at least in part reflecting differences in the timing of their latest forecast revisions. Goldman Sachs saw GDP growth falling six per cent this year, for example, while UBS pencilled in a 5.4 per cent fall and Capital Economics a 4.1 per cent drop.

By way of comparison, the AFR reported that the consensus forecast from ’a dozen leading economists’ was for the economy to shrink three per cent and the unemployment rate to reach 9.4 per cent in 2020.

Why it matters:

The standard cruel comment about economic forecasters is that they (ok, we) were put on this world to make astrologers look good. Which is a bit harsh (and no, not on astrologers). One impact of the CVC is that the uncertainties that always bedevil economic forecasting are even more problematic than usual. Hence just last week the Treasurer pushed back the 2020-21 Budget to October, explaining that the unprecedented nature of the current situation and the high degree of uncertainty ‘makes it extremely difficult to formulate reliable economic and fiscal estimates over the next few months.

The wide range of predictions about the damage to the economy as a result of the virus are an indication of the huge unknowns that remain around the likely severity and duration of the pandemic itself, the breadth, intensity and longevity of the associated physical distancing and lockdown measures that are being employed here and overseas to combat it, and the consequent disruption to business. Not to mention the potential for collateral damage via financial market turmoil.

Just one aspect of the current forecasting challenge, for example, is that each round of government policy designed to improve health outcomes also takes a new and heavy toll on economic activity. Westpac economics currently sees the unemployment rate rising to 11.1 per cent and GDP contracting by 3.5 per cent in the June quarter with no sustained economic recovery until the fourth quarter of this year. That would entail GDP contracting three per cent through the year and the unemployment rate ending 2020 at just below nine per cent. Just a week earlier, the bank had released a new set of forecasts predicting that the crisis would see unemployment peak at seven per cent and the economy grow at 1.5 per cent through the year. Between those two forecasts, Canberra had announced dramatic new restrictions on business and social gatherings with profound implications for business activity and employment.

The standard solution to rapidly changing conditions (other than to not forecast at all) is to ‘forecast often’ and be prepared to make frequent and sometimes significant revisions on the back of new data. That is probably the best we can do to cope with the radical uncertainty now besetting us.

What happened:

The ABS released the results of the first of its new Business Impacts of Covid-19 surveys, drawn from a sample of 3,000 businesses (with a 41 per cent response rate) and based on data collected between 16 and 23 March. The survey found that 49 per cent of the businesses surveyed had already experienced an adverse impact as a result of the virus and 86 per cent expected to be affected in future months.

By sector, the biggest hit was in accommodation and food services, where 78 per cent reported that they had experienced adverse impacts in the previous two weeks and 96 per cent anticipated negative impacts over the coming months. Businesses in Professional, scientific & technical services (21 per cent), Electricity, gas and water supply (34 per cent) and Mining (37 per cent) reported the lowest rates of adverse impacts in the previous two weeks.

graph

Source: ABS

A fall in local demand was the most common impact reported for the previous two weeks (cited by 82 per cent of firms) and was also the most common impact expected in coming months (81 per cent). Of those businesses already reporting an adverse impact, 36 per cent had experienced staff shortages and 59 per cent expected to experience staff shortages in coming months.

graph

Source: ABS

Why it matters:

Given that the government’s announcement of Stage 1 Restrictions on social gatherings came on 22 March, this ABS survey only picks up the very start of business reactions to the first big round of physical distancing measures. And at that stage already roughly half of all businesses were reporting adverse consequences. No surprise then that almost 90 per cent expect to be hurt in some way over the next six months. Note, however, that the questions ask whether and how a business was adversely impacted, but not about the intensity of that impact, which would also be expected to vary considerably by sector and would be useful to know.

What happened:

As part of an additional set of measures announced by the government on 24 March, real estate auctions and open house inspections were prohibited. Private appointments for inspection were still allowed.

Why it matters:

While the measures still allow the housing market to operate, they seem likely to slow sales and dampen sentiment against an already-deteriorating economic backdrop.

To date, the housing market has held up reasonably well, and indeed entered the CVC in the middle of a strong upturn: back in February, combined capital dwelling values were up 1.2 per cent over the month and 7.3 per cent over the year, while housing values were at record highs in Melbourne, Brisbane, Canberra, Hobart and Adelaide. The timelier data we have on auction clearance rates do show some impact from the virus, but it’s been limited. The latest CoreLogic weekly numbers report a preliminary auction clearance rate of 61.3 per cent, which although well down on recent highs was still substantially higher than this time a year ago. This resilience seems unlikely to last, however.

graph

A big drop in market turnover would have negative implications for those states who rely heavily on stamp duty as a source of revenue. Based on ABS data as reported by CoreLogic, stamp duty revenue delivered an $21.7 billion of tax revenue to state governments last year. A return to falling prices would also hit state revenues as well as have adverse consequences for household wealth and consumer sentiment (see also the readings below for an update on Australian household wealth as of the final quarter of last year).

. . . and what I’ve been following in the global economy

What happened:

In the United States a record 3.28 million workers applied for unemployment benefits in the week ended 21 March, up from 282,000 the previous week. In Pennsylvania, the worst-hit state, the number of claims filed increased by more than 363,000, with ten states reporting increases in excess of 100,000.

graph

Why it matters:

The reported increase in US initial unemployment claims dwarfed the previous record one-week jump (695,000 claims in October 1982). It was also much worse than already-gloomy market predictions of a 1.7 million spike. The usual pattern for ‘regular’ recessions is for a gradual increase in unemployment. The economic sudden stop triggered by the coronavirus crisis (CVC) measures saw more claims filed last week than across the whole of March or April 2009 (each much less than three million), the peak months for job loss during the GFC. This is an unprecedented labour market shock.

What happened:

IMF Managing Director Kristalina Georgieva said that the outlook for global growth in 2020 was negative, predicting ‘a recession at least as bad as during the global financial crisis or worse.’ But the Fund expects a recovery in 2021.

Why it matters:

While we still have to wait before the IMF updates its quantitative forecasts for the world economy, the qualitative judgement that the growth outlook for this year is likely to be at least as bad or worse as during the GFC is a good proxy for the current baseline for expectations for the CVC.

As a quick reminder, world GDP growth measured at purchasing power parity exchange rates fell from 5.6 per cent in 2007 to three per cent in 2008 and to minus 0.1 per cent in 2009 before recovering to 5.4 per cent in 2010. For advanced economies, growth slowed from 2.7 per cent in 2007 to 0.2 per cent in 2008 to a contraction of 3.3 per cent in 2009 before recovering to 3.1 per cent in 2010. That was a swing of about six percentage points between 2007 and 2009. Using market exchange rates (which reduces the relative weight of faster-growing emerging markets and increases that of the advanced economies that were at the centre of the GFC), world GDP growth slowed from 3.8 per cent in 2007 to 1.5 per cent in 2008 before contracting by two per cent in 2009. Growth then recovered to 4.1 per cent in 2010.

While the GFC was a bad year for the world economy, in relative terms Australia came through reasonably well. Our growth did drop quite markedly, slowing from around 4.4 per cent in 2007 to about 1.9 per cent by 2009, but that swing of about 2.5 percentage points was a much better outcome than the six-percentage point drop suffered for the group of advanced economies overall. That relative resilience reflected a combination of factors including one of the more aggressive fiscal policy responses, the fact that our financial sector was not part of the meltdown that triggered the GFC, and the cushion provided by China’s economic resilience and Beijing’s own aggressive stimulus package. This time around, we look to be more directly exposed to the CVC than we were to the GFC and China is less likely to provide a cushion. We also have much less conventional monetary policy ammunition to spend, although we did start with ample fiscal space. So, for Australia, in terms of the ‘at least as bad or worse’ than the GFC prediction, the most likely outcome is that the CVC is going to be worse.

What happened:

Preliminary (‘flash’) Global PMI data showed dramatic falls in service sector activity across Europe and the United States. As in the case of Australia (see above) the manufacturing sector was more resilient although there was plenty of weakness here, too.

In the Eurozone (pdf), the Flash Composite Output and Services PMI indices both hit record lows in March while the Manufacturing PMI Output Index fell to a 131-month low. The collapse in the survey’s service sector business activity index of more than 24 points took it down well below its prior February 2009 low. There was a record fall in export business as cross border trade flows seized up, while service sector jobs were cut at the steepest rate since May 2009 and the composite indicator showed the steepest decline since July 2009. The survey also reported widespread and near-record supply chain delays. And business sentiment about the year ahead dropped to the most pessimistic on record.

In the UK (pdf), the Flash Composite Output and Services Activity indices both fell to record lows, while the Manufacturing Output Index fell to a 92-month low, with the combined monthly drop across manufacturing and services again exceeding that suffered during the GFC. New orders across the private sector fell at the sharpest rate seen since December 2008 and business expectations for the year ahead fell to the lowest level recorded since the measure started in 2012. The flash PMI also signalled a fall in employment across both the manufacturing and services sectors at a scale not seen since July 2009. There were only small pockets of growth to be found across the economy, in food manufacturing, pharmaceuticals and health care.

Finally, in the United States (pdf), the Flash Composite Output and Services Business Activity Indices each fell to new series lows in March while the US Manufacturing PMI and Manufacturing Output Index both slumped to a 127-month low. The overall decline in activity was the steepest recorded since comparable survey data were available in October 2009, the fall in services was the fastest contraction in business activity in the series’ ten-and-a-half year history, and firms overall reported reducing their workforce numbers at the fastest pace seen since December 2009.

Why it matters:

The collapse in business activity across the Eurozone was far worse than that seen at the peak of the GFC , in the UK the initial hit to activity is larger than that delivered during the worst months of 2009, and in the United States, companies reported the steepest downturn suffered since 2009. Economists at data provider Markit estimate that, based on the past relationship between the survey results and GDP growth, the March PMI is roughly indicative of Eurozone growth currently dropping at a quarterly rate of around two per cent and UK GDP declining at a quarterly rate of between 1.5 and two per cent. And with the scale of lockdowns in the both the US and UK having increased since the time of the survey, the results reported here seem likely to worsen as we head into the second quarter of the year.

Overall, then, the PMI results from Europe and the United States indicate a dramatic ‘sudden stop’ in economic activity and a grim outlook for employment. That suggests that the new official view- that the current downturn will be at least as bad as the GFC and quite possibly worse - looks about right.

What happened:

The US Senate passed a huge US$2 trillion stimulus package (approaching ten per cent of US GDP), described by Senate majority leader Mitch McConnel as ‘a wartime level of investment into our nation,’ by a 96-0 vote. The package now goes to the House for a vote Friday US time. At the time of writing there was still a little confusion about the exact details, but highlights include:

  • A US$500 billion fund for loans to businesses hurt by the virus, including US$25 billion for passenger airlines, US$4 billion for cargo airlines and US$17 billion for companies ‘critical to maintaining national security.’ The balance (US$454 billion) will be available to backstop loans to businesses, state and local government.
  • Around US$301 billion in direct payments to US households worth US$1,200 per adult plus US$500 per child. The payments will start to be phased out starting at annual incomes of US$75,000 per person so that wealthier households will not benefit.
  • About US$250 billion to increase unemployment benefits by US$600 a week for four months which will be paid in addition to payments from the claimant’s state, along with an expansion of benefits to the self-employed and workers in temporary employment and an increase in the duration of benefits from the 26 weeks typical in most states to 39 weeks.
  • Roughly US$350 billion in loans and grants to small businesses with less than 500 employees, with loans that are used to cover payroll expenses, rent, interest on mortgage obligations and utilities to be forgiven (turned into grants).
  • Around US$340 billion in emergency supplementary funding for the battle against the virus including US$117 billion for hospitals and veterans care, US$11 billion for vaccines, diagnostics and related needs, and US$45 billion for the Federal Emergency Agency.
  • A US$150 billion Coronavirus Relief Fund for state and local governments.

Why it matters:

The size of the US stimulus package (assuming it passes the House) will be the largest congressional bailout in recent US history, dwarfing the US$800 billion Obama stimulus package in response to the GFC. The program will provide important support to a US economy that is starting to reel under the impact of the virus. But there are already concerns that it will not be enough to head off a recession.

The prospect of this fiscal stimulus – along with intervention by the US Fed – has also acted as a stabilising force for financial markets, delivering two consecutive days of large gains on Wall Street for the first time in about month. But there’s a risk that this relief proves short-lived: at the time of writing global markets were again looking unsettled as coronavirus case numbers continued to climb.

What I’ve been reading

Again, lots of Covid-19 stuff, I’m afraid, but inevitable under the circumstances.

Michael Keating worries that the government’s wage subsidy through the ‘Boosting cash flow for Employers payments’ measures – which he estimates as equivalent to about a 20 per cent subsidy on average – won’t be enough to maintain employment in many industries.

Steve Grenville reviews the benefits of and limitations to the RBAs actions: they will help on the margin and there isn’t much downside, he reckons.

NAB economics analyses a 2006 paper on the economics of pandemics co-authored by now-Treasury Secretary Steven Kennedy. That paper finds that a collapse in confidence drives GDP growth lower through falls in consumption (especially services), business investment and residential investment, with plummeting household income and profits then delivering a further squeeze on spending and investment. Employment also collapses, but the paper assumes that much of the net impact on the unemployment rate is offset by a parallel fall in labour supply. The drop in activity alone is estimated to weaken the budget balance by 3.5 per cent of GDP in the absence of any fiscal stimulus (an assumption that the paper acknowledges is unrealistic).

Three links from the ABS this week. First, the ABS provided a guide to using detailed labour force data on hours worked to assess the impact of Covid-19 on the labour market. While analysis of February’s results find no identifiable impact from the virus, the ABS points out that the reference weeks for the monthly labour force survey fall in the first half of the month, and during the first two weeks of February there was only a relatively low number of confirmed cases in Australia and the WHO had yet to declare a global pandemic.

The ABS has also released new data on regional population growth in Australia. It finds that the population living in Australia’s capital cities grew by 303,100 during 2018-19, bringing the total to almost 17.2 million people, with capital city growth accounting for 79 per cent of total population growth over the year. The largest population increases were in Melbourne (up 113,500 or a 2.3 per cent increase), Sydney (87,100, 1.7 per cent), and Brisbane (52,600, 2.1 per cent). All capital cities except for Darwin (down 1,200 people or -0.8 per cent) experienced a rise in population. Net overseas migration was the big driver of population growth in Melbourne and Sydney and was also the major contributor for Adelaide and Hobart. Perth and Canberra saw natural increase make the largest contribution, while Brisbane had a three-way tie between natural increase, net overseas migration and net internal migration. The geographic centre of Australia’s population is now found 50km east of the small town of Ivanhoe in western NSW and has moved 2.4km southeast over the past year, reflecting rapid population growth in Melbourne.

And finally, the ABS reported that household net worth (wealth) rose $360.9 billion or 3.3 per cent in the December quarter. This was driven by a $382.8 billion increase in assets that was only partially offset by a $21.8 billion rise in liabilities. That rise in total assets reflected the housing market recovery at the end of last year, where the value of residential land and dwellings rose by $282 billion, due to real (inflation-adjusted) holding gains of $233.4 billion – the strongest such gain since the final quarter of 2009. Household wealth per capita rose $12,809 to $442,705, which again was the largest increase since 2009.

Here’s a second CEPR e-book on Covid-19 from VoxEu.

See also these two columns from VoxEU on lessons from the 1918-1920 Great Influenza Pandemic, the so-called Spanish Flu. Robert Barro and colleagues estimate that the 1918-20 pandemic reduced real GDP per capita and consumption of the typical country by six per cent and eight per cent, respectively. By way of comparison, WWI was associated with declines in GDP and consumption of eight per cent and ten per cent respectively. The authors suggest that the Spanish Flu might have been the fourth most severe macro shock to hit the modern world economy, behind only the two World Wars and the Great Depression (note the second WSJ link below cites different research which finds a much more modest economic cost). And this column by Arnstein Aassve and colleagues finds that the social disruption during the period had permanent consequences in terms of lower social trust, which in turn acted as a headwind for economic growth.

Meanwhile, Carmen Reinhart warns that in her view there are no good insights to had from historical episodes as to the unfolding consequences of the coronavirus crisis, stressing that this time truly is different. She notes that the big difference from the Spanish Flu in the case of the United States for example is the backdrop of World War I and surging wartime production efforts.

An FT Big Read on how the UK government’s response to the coronavirus is creating a different kind of wartime economy, one based around demobilisation instead of mobilisation.

A general recommendation for Marginal Revolution, which is doing a great job of pulling together a bunch of interesting links on the coronavirus, including one linking to the views of ‘superforecasters’ on four questions relating to Covid-19. It will be interesting to check their forecasts against the data when we hit 31 March. And here is MR’s Tyler Cowen in conversation with Russ Roberts of the EconTalk podcast.

The IMF has launched a policy tracker covering the ‘the key economic responses governments are taking to limit the human and economic impact of the COVID-19 pandemic’. For now, it includes G-20 members and the EU, but the Fund promises to add more economies over time.

The WSJ asks whether the West has been drawing the wrong lessons from China’s battle against Covid-19. It wasn’t the lockdown itself that got the virus under control in Wuhan (although it did help slow the transmission to the rest of China), this piece argues, rather it was an ‘aggressive and systematic quarantine regime’ which required significant testing capacity and ‘draconian’ measures of social control. Also from the Journal, lessons from six other disasters: there is frequently a short-term trade-off between economic stability and public health and safety; enormous uncertainty at the outset leads to early responses that are often timid or off-target; there are often permanent changes to habits; and the most affected sectors and regions can take years to recover, although for society as whole the healing can be ‘remarkably quick’;

This Greg Mankiw proposal appears quite attractive. It combines the benefits (simplicity, timeliness, no need to forecast who will be most in need) of sending everyone a cheque with a mechanism for ensuring subsequent clawback from those who didn’t need the support, which is a common objection to the otherwise strong case for universal measures like this.

Adam Tooze, the author of Crashed, a history of the GFC, has started what hopefully will become a series of posts on Crashed to Corona. The first one is here. And here is Tooze comparing the coronavirus market crash with the 2008 financial crisis for Foreign Policy.

Nouriel Roubini worries that while the best we can hope for is recession deeper than that following the GFC, the policy response to date means that worse outcomes are increasingly likely.

Tim Wu reviews Shoshana Zuboff’s book on Surveillance Capitalism. (There’s little doubt that the tech sector had a major image problem before the onset of the CVC. And early in the current pandemic the WHO warned about the dangers of an ‘infodemic’. But now that its tech that’s helping keep many of us connected and at work, I wonder if the backlash might have peaked?)

John Authers looks at the ‘epic rebound’ in financial markets over the past two days, which for US stocks ended a streak without two positive days lasting for 32 trading sessions. But markets appear to be wobbling again as I write this…

Robin Wright in the New Yorker thinks we need to worry not just about an economic recession but also a social recession as Covid-19 has hit the world at a time when more people are living alone than at any time in human history.

Tom Holland on Plague and the Comforts of History (found via FT Alphaville).