Overview

COVID-19 remains the dominant issue facing the global economy, and the news remains more negative than positive.

Positives include:

• The global economy continues to edge forward.
• Several new economic headwinds have so far been less problematic than initially feared.
• The COVID-19 fatality numbers remain fairly low when compared to the first wave.
• Fresh data argues that the percentage of the population with antibodies protecting them from COVID-19 is significantly higher than the official numbers.

However, the negatives arguably dominate despite being fewer in number:

• The daily COVID-19 infection numbers continue to rise across most of the developed world. This is the main problem.
• Additional social distancing and economic restrictions are likely necessary to limit further spread.

Virus developments

The global number of new infections per day continues to set new records, albeit at a slowing pace. The fatality figures remain shy of record levels, but are no longer falling and remain above 5,000 per day (see next chart).

Global COVID-19 cases and deaths

Note: As of 09/28/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM

Accordingly, the global transmission rate has slipped back down to just above the critical threshold of one. It needs to fall slightly further to put the virus onto the defensive (see next chart).

Global transmission rate hovering around key threshold of one

Note: As of 09/28/2020. Transmission rate calculated as 7-day change of underlying 5-day moving average of new daily cases, smoothed with 7-day moving average. Source: ECDC, Macrobond, RBC GAM

Among emerging market countries, the most intense source of new infections – on a per capita basis – is Latin America, with Argentina, Peru, Columbia, Brazil and Chile leading all nations in that order (see next chart).

COVID-19 transmission analysis in emerging markets countries

Note: As of 09/28/2020. Transmission rate calculated as 7-day change (presented as a ratio) of 5-day moving average of daily new cases. Source: ECDC, Macrobond, RBC GAM

Other prominent emerging market developments include:

• Poland has a low absolute number of daily infections, but is rising quickly with a transmission rate of 1.6.
• Russia has a fairly high virus count, and is continuing to suffer a rising trend.
• Indonesia has previously not attracted a great deal of attention. It now has a significant number of daily infections even when compared to its enormous population. It also sports a worryingly high transmission rate of 1.7.
• South Africa’s numbers continue to improve.
• India still has the world’s greatest number of infections at 84,000 per day, but the country has actually steadily improved for several weeks (see next chart).

COVID-19 cases and deaths in India

Note: As of 09/28/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM

Europe remains in the midst of its second pandemic wave, with Spain and France both in the vicinity of 11,000 new infections per day. Fortunately, both may now be starting to stabilize, with Spain even tentatively improving (see next chart).

COVID-19 cases and deaths in Spain

Note: As of 09/28/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM

Despite these incremental improvements, Europe remains seriously challenged. The European Centre for Disease Control recently indicated that the social distancing adjustments made so far have not been enough to control the virus. The Centre further identifies Spain and a number of Eastern European countries (Bulgaria, Croatia, Czech Republic, Hungary, Malta and Romania) as being at an extremely high risk. The U.K. and France are then described as being at a very high risk. Conversely, the risk in Germany is somewhat less.

The U.K. second wave continues to rise. The country is now reporting 6,000 new cases per day on a trend basis, up significantly from the week before (see next chart). Indeed, the country’s transmission rate is an elevated 1.5. The U.K. fatality numbers are again edging higher although, as with most developed countries, these remain substantially lower than in the spring.

COVID-19 cases and deaths in the U.K.

Note: As of 09/28/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM

Canada’s daily caseload is now up to nearly 1,400 on a trend basis, and continues to rise aggressively (see next chart). The country’s transmission rate is a considerable 1.4 – well above the ideal sub-one level. The three most populated provinces of Ontario, Quebec and British Columbia are all suffering significant increases. Quebec is now in the worst position of the three, replicating its experience last spring.

COVID-19 cases and deaths in Canada

Note: As of 09/28/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM

As we flagged last week, the U.S. virus trend is now clearly rising again as well, recording daily infections of well over 40,000 (see next chart). EEpidemiologists can debate whether this constitutes a continuation of the second wave, a new third wave, or some other configuration. It looks like a third wave to us.

COVID-19 cases and deaths in the U.S.

Note: As of 09/28/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM

Over half of U.S. states are now recording transmission rates that are greater than one, led by Texas and a number of western states. Curiously, Florida’s infection numbers continue to fall for the moment. However, the state’s recent decision to significantly re-open restaurants and bars and to diminish other social distancing precautions argues that it could soon experience rising infection numbers as well.

On a cumulative basis, the U.S. recently encountered a sombre milestone: more than 200,000 deaths due to COVID-19.

New social distancing efforts

It is extremely difficult to quantify how much social distancing measures first tightened, then eased, and are now again tightening in many jurisdictions. This is for two main reasons:

1. It is unclear how to quantify disparate policies: how much more important is it to require the wearing of masks in public spaces relative to obliging bars to close by midnight? And yet without quantifying these things, there is no way to combine the various efforts into a coherent single measure of social distancing.

2. Most efforts to measure social distancing are occurring at the national level, and yet many regulations are set at the sub-sovereign level.

As such, while we look to Oxford University’s stringency index for a gauge of the extent to which social distancing measures have tightened or eased, the measure confuses almost as much as it informs (see next chart).

Stringency of lockdown by country

Note: As of 09/14/2020. Stringency index over time and the extent to which countries have shut down in relation to the U.S.  Source: Google, University of Oxford, Apple, ECDC, UN, Macrobond, RBC GAM

For instance, any effort to look to the U.S. for guidance as to how much additional social distancing was necessary to arrest the country’s second viral wave (notwithstanding the new third wave) is foiled by the fact that the Oxford index fails to identify any tightening as having occurred in late June or July. And yet we know there were many actions taken by different states and municipalities at that time to control the spread of the virus.

Spain’s data (refer again to chart, above) proves somewhat more helpful. Additional social distancing measures were visibly applied in July, presumably as the second wave began to form. However, it is also confusing in that the country has since continued to tighten its social distancing rules into September by all accounts. Yet there is little visual evidence of that in the stringency data.

As such, we really don’t have anything quantitative to work with. From a necessarily qualitative standpoint then, we continue to gravitate back to the view that the U.S. began to run into trouble when it significantly opened indoor bars and restaurants in the early summer – indoor, highly social activities that preclude the wearing of masks. It later diminished the virus count when such actions were later reversed. It could well be that restaurants and bars are not so much the prime offenders as a canary in the coalmine for when governments and people feel sufficiently confident that they recommence social activities with an enthusiasm that eventually proves problematic.

Nevertheless, the main conclusion is that there is clearly too much social interaction happening right now in countries with a rising virus count. So far, governments are focusing on reducing the amount of interaction that occurs in a non-economic capacity – at personal barbeques and house parties, for instance. But it seems increasingly likely that further restrictions will need to be applied to economic activities with a strong social component, especially if schools are to remain open and as colder, drier weather approaches.

The U.K. provides some sense for how this is playing out. It first tightened its rules for the number of people allowed to socially interact with one another and limited certain economic activities in the hardest hit pockets of the country. It has now instituted a nation-wide 10 p.m. closing time for restaurants and bars. Furthermore, office workers are now encouraged to again work from home if possible, potentially for the next six months.

Canada is also now pivoting from solely targeting non-economic social interactions to limiting some economic sectors. Some time ago, British Columbia closed its night clubs and banquet halls. Ontario is now closing all strip clubs and is limiting the hours at bars and restaurants.

In Europe, countries have again begun to more tightly control border crossings. Italy, for instance, is now insisting that visitors from Spain, Greece and parts of France must get a COVID-19 test before they can move freely about the country.

Lockdown compliance

As many countries begin to tighten their social distancing rules once more, a key question is the extent to which the new restrictions will be tolerated. One imagines that the public will prove less compliant than the first time. This is for several reasons.

1. The virus seems less scary than it did in the spring. Thanks to additional testing and improved medical care, COVID-19 no longer carries an alarming case fatality rate of 10%. Furthermore, it was initially unclear just how hyper-virulent COVID-19 was: people were afraid to leave their homes in March and April, let alone venture into crowded places. That fear has abated, perhaps especially and most worryingly among people who have behaved irresponsibly but had the good fortune not to get infected yet. Put bluntly, everyone who hasn’t yet been infected thinks their behavior has been sufficient to keep themselves safe. And yet, mathematically, that cannot be true of everyone since the virus is accelerating. It is hard to fathom people being willing to return to their springtime restrictions even if the situation calls for it.

2. From an economic standpoint, it will be very hard on businesses if they are required to shut a second time, having struggled through a period of subpar revenue over the intervening period. The damage is cumulative.

3. The original lockdown arrived like a thunderbolt, with little warning or precedent. Lobbyists and business interests have since had time to organize. It will not be so easy for governments to shut down economic sectors again, even if the situation calls for it.

4. It is a similar situation for workers as for businesses. It is bad enough to lose a job once, but considerably worse to lose it a second time over a short period of time. Even with government support, the situation is not ideal.

5. Governments will be reluctant to seriously constrain their economies again for a mix of budgetary and political considerations.

It is not to say that these obstacles cannot be overcome: the social distancing restrictions nevertheless continue to mount and are beginning to impact the economy, but this helps to explain why the policy response has been so ginger and so lagged thus far. As we have noted in the past, we do not expect a second round of tightening to push economies into a second recession. Instead, we expect that the growth rate will temporarily sputter over the next few months.

Protests

Reflecting the unpopularity of additional tightening measures, there have been serious protests in Spain over the severity of the country’s new rules. Similarly, Israel has now imposed a fierce multi-week lockdown on its citizens, resulting in significant protests. Other countries may be discouraged from taking the necessary actions based on these adverse outcomes.

The bottom line is it will be harder to put the genie back in the bottle a second time. The will of governments to enforce tighter rules is likely weaker than in the spring and compliance will likely also be worse. In turn, it will probably take longer than it should to bring this next wave under control.

How is COVID-19 being transmitted?

We continue to return obsessively to the question of how COVID-19 is being transmitted. Until it is properly answered, efforts to control the virus will be highly imprecise.

Contradictions continue to abound. A prominent study from the U.S. Centre for Disease Control and our earlier qualitative analysis suggest that social sectors like restaurants and bars are an obvious vector for transmission. However, the province of Quebec and Ontario’s heavily-affected Peel Region indicate that restaurants and bars are not actually a primary source of transmission.

Fortunately, we recently discovered a new source of information from Canada. It is far from perfect: frustratingly, it doesn’t actually track transmission in restaurants, bars or retail stores. Furthermore, it only picks up outbreaks – transmission to multiple people – rather than individual infections.

With these caveats in my mind, the trends are still interesting:

• Nearly without fail, long-term care and retirement residences have been the source of the greatest outbreaks. They were the source of the vast majority of outbreaks in March and April, with nearly 500 each month. While the figures have mercifully shrunk since then, such facilities have remained the largest source of outbreaks every month since, with the exception of August.
• The “other” category – which, crucially, includes social gatherings – experienced a steady increase in outbreaks in every month from March through August, culminating in the category being responsible for the most outbreaks in August – more than 100. One imagines the increase was a function of people becoming incrementally less cautious and the weather becoming incrementally better. This is almost certainly why there has been such a focus on limiting social gatherings in recent government missives.
• More recently, the school and childcare centre category has increased from nearly zero to a third place showing in September-to-date. While still responsible for well under 100 outbreaks in September, it is nevertheless a new transmission channel.
• Again, frustratingly, there is no information about most economic sectors – including those seemingly at greatest risk of facilitating transmission. Exceptions are healthcare, which was fairly high in April and May, but has since been quite low; and “industrial” (which includes agriculture), a sector that suffered several serious outbreaks among foreign agricultural workers living in dormitories, but not so much recently.

The data also indicates the average size of an outbreak for each category. The industrial and healthcare sectors suffer the greatest number of infections when an outbreak occurs, at more than 10 each. Social gatherings are responsible for nearly 10 infections per outbreak. In contrast, retirement residences and schools only experience an average of around 4 new infections per outbreak. Transmission appears to be considerably more limited.

While this data doesn’t explicitly apply to countries other than Canada, it seems reasonable to imagine that the contours would be similar.

In search of herd immunity

Although less egregiously mis-measured and less discussed in recent months, it remains the case that the true number of people who have contracted COVID-19 is likely underestimated by standard infection testing. This is in part because of inadequate testing, but also because many people lack symptoms.

Fortunately, serology blood testing can provide some sense for the true fraction of the population that has developed antibodies to the virus – meaning that they were previously infected and are probably now immune.

U.S. research conducted in July yielded some eyebrow-raising figures. New York City was the most extreme location, with 18% of the population already possessing antibodies. This was around six times higher than the official estimate of people who had been infected. Other U.S. jurisdictions were less extreme, with no more than 6% of the population testing positive, but the extent of the underestimation was occasionally even greater than for New York, ranging from a true count that was two times higher than the official numbers to as many as seven times higher.

Additional tests were then conducted in September, albeit for different parts of the U.S. Whereas Florida, Texas and Georgia have officially tallied infections equal to 3% of their population, the serological tests point to a true underlying infection rate of 8% to 9% for each of the states, or three times the official numbers.

To the extent the COVID-19 experience has varied significantly by state, and furthermore that New York City was likely a considerable outlier, we cannot arrive at a precise conclusion about the degree of underestimation at the national level. But it makes sense that the extent of the underestimation was slightly smaller in September than in July. Testing has been less constrained over the intervening period. Furthermore, it is clear that New York City is an extreme outlier to the extent that it suffered by far the worst initial hit from the virus.

As a result, we are comfortable with the assumption that nationwide serological testing would reveal an immune population that is around three times higher than the official numbers. In other words, whereas 2.1% of the U.S. population has been officially infected, a better estimate would be around 6% of the population.

But this is not the final contortion we must make. A raft of research is now arguing that some people who have been infected either do not produce enough antibodies to be detected in a serology test or perhaps even relied entirely upon their T-cells to fend off COVID-19, never resorting to the production of antibodies. Testing for T-cells is a much harder proposition, and has been done only to an extremely limited extent. It is conceivable that up to twice as many Americans are ultimately protected from COVID-19 due to their T-cells, or perhaps as much as 12% of the population. To be safe, we should conclude that somewhere between 6% and 12% of Americans are likely now immune. You’ll note that we don’t say that up to 12% of the population was previously infected. This is because, while many with T-cell protection would have contracted the virus and developed their T-cells by vanquishing it, others may already have protection from a previous similar coronavirus that their T-cells are already attuned to.

Even at the most optimistic end of the spectrum, the U.S. is still quite far away from herd immunity, which likely requires around 60% of the population to be immune. The U.S. is only between one-fifth and one-tenth of the way there. This helps to illustrate why a vaccine remains the most obvious and desirable path toward herd immunity.

Nevertheless, the incremental herd protection achieved so far is still useful. As we discussed in an earlier MacroMemo, as the fraction of the population with immunity goes up, the natural transmission rate for the virus goes down. For instance, if a sick person would under normal circumstances infect three other people – that number goes down to 2.7 people when 10% of the population has become immune. Some of the people who would have been infected can no longer be infected since they are immune. More practically, the amount of social distancing required to keep the transmission rate below the key threshold of one is thus incrementally less than before.

Schools update

Unlike in the U.S., most Canadian school boards have opened fully for physical classes this fall. As such, the situation will continue to provide important insights.

Unsurprisingly, there have been no shortage of cases reported within schools. In fact, my own local elementary school now has a case. As we calculated several weeks ago, nearly 100 children per day were being infected by COVID-19 across Canada before schools had even opened, such that many dozens of schools per day should expect to discover that someone with the virus was recently within their walls, even if the schools themselves were not the location of transmission.

In practice, it now appears that some transmission is happening within schools, as per the numbers cited earlier in this report. Of course, children were also presumably transmitting the virus amongst themselves over the summer at summer camps, day cares, on sports teams and wherever else they spent their days before schools opened.

To this end, approximately 4% of Ontario’s schools have recorded a case over the first two weeks of the school year. It seems likely that many if not most schools will have to deal with this problem at some point over the long school year ahead, though the fact that students are largely cohorting within the schools means that only a small fraction of each school must be converted to temporary virtual instruction.

It is undeniably unlucky that the schoolyear has arrived at a time when the virus numbers were already starting to rise significantly. Whether physical schools remain viable – with all of the knock-on economic considerations revolving around the acquisition of human capital and the ability of parents to work – will largely be a function of several factors:

• the accelerating virus numbers can be tamed over the coming months
• the extent to which governments choose to prioritize education (most European countries have kept their schools open even as their second waves rage, for instance)
• the extent to which the virus is found to seriously incubate within schools or not.

For the moment, there are real examples of outbreaks in Canadian schools, but the outbreaks themselves have proven unusually small in size.

Although most major U.S. school boards are closed, New York City has just started physical schooling after several weeks of delays. There are already 100 cases reported in New York City schools, but keep in mind that the city educates more than 1.1 million students and so the number is not nearly as large as it first sounds.

More broadly, the New York Times reports that across U.S. colleges and universities, there have been at least 130,000 COVID-19 infections since the pandemic began. For context, there are around 20 million college students in the U.S. and perhaps another 5 million people who work at post-secondary institutions. As such, only 0.5% of such people have been infected (though many studied virtually). Framed differently, only 1.8% of U.S. COVID-19 transmission has occurred on college campuses.

The Centre for Disease Control in the U.S. has best practices for when schools should be physically open versus virtual. One key consideration is the number of new cases on a per capita basis. Canada lands below the relevant threshold – arguing to keep schools open. The U.S. is above. But this isn’t the only criteria. The CDC also indicates that the proper use of masks, distancing, hand hygiene, cleaning and contact tracing are necessary to have physical schooling. For many countries, including the U.S. and Canada, most of these criteria are met but the distancing aspect is debatable.

As such, much of Canada is considered well equipped to handle physical classrooms, whereas much of the U.S. is not. Of course, as Canada’s numbers actively deteriorate, the conclusion could well change in a matter of weeks to months. Furthermore, Canada does not fully meet the CDC’s secondary criteria. While hospital utilization is under capacity, the rising virus trend itself is considered problematic.

More broadly, the CDC indicates that best practices for schools include:

• masks
• checking temperature and other symptoms
• social distancing
• good hygiene
• cleaning and disinfecting
• contact tracing
• cohorting
• staying home when appropriate
• adequate safety and cleaning supplies
• staggered scheduling
• good ventilation
• strict control of visitors
• minimizing shared objects
• erecting physical barriers
• clear distancing instructions
• halting food services on school premises
• closing any communal space

Economic developments

U.S. Federal Reserve decision

We begin with the (now somewhat dated) Fed decision from mid-September. It progressed as expected, with a commitment not to raise rates for several more years and little dissent. Not only will rates remain low while the economy is weak, but the Fed now indicates that it expects to maintain an “accommodative stance of monetary policy” until it has achieved its inflation target of averaging 2% over time. In other words, the Fed won’t tighten until inflation is already normal, rather than as it begins to normalize. As such, and entirely in keeping with our prior expectations, bond yields look set to remain quite low.

PMI data

Purchasing Manager Index data was mostly flat in September:

• The U.S. was roughly unchanged in a position of moderate growth.
• The Eurozone was similarly roughly unchanged, in a position of modest to moderate growth.
• The U.K. PMI fell from a (strangely) strong to moderate growth signal.
• The Canadian Federation of Independent Business barometer was unchanged in a position of moderate growth.
• Japan’s PMI rose slightly, but remains below the 50 threshold normally delineating decline and growth. We believe the Japanese economy is growing, but its PMIs are lagging somewhat.

It is reassuring that most of these indicators point to ongoing steady economic growth given concerns about second viral waves. However, the indicators are no longer actively rising from month to month. This signals that the easy part of the recovery has come to a close.

Data medley

As real-time economic indicators become less useful due to mounting seasonal distortions, the U.S. weekly jobless claims measure is becoming even more important as a reliable but high-frequency indicator. The latest week showed a miniscule retreat, from 866K to 877K new unemployment claimants. Continuing claims improved slightly, from 12.7M to 12.6M, but this series is lagged by a week and so it is unclear whether the labour market has actually ceased to improve. The payroll numbers released later this week should nevertheless convey healthy growth given the strong gains made in continuing claims over the reference period.

U.S. durable goods orders for August have been released, and are up 0.4% overall, and by 1.8% for the less volatile core capital goods version. Accordingly, it appears that capital expenditures did not wither in the late summer, even as the fiscal cliff hit.

Germany’s closely-watched Ifo (Institute for Economic Research) business climate index was just released for September. The index managed a further incremental gain, from 92.5 to 93.4. This is well above the April low, but still not back to the pre-COVID-19 reading of 95.9 in February. Germany remains a success story from a virus perspective.

Japan’s supermarket sales are now up 3.3% on a year-over-year basis, consistent with the trend that people are buying more food than before the pandemic. Conversely, convenience store sales are down 5.5% YoY and department sales are a big 22% lower than the year before. Each of these is consistent with the experience in other countries, particularly the underperformance of department stores.

Sector-by-sector GDP

Canada is lucky to have a monthly series of industry-level economic output (see next chart). This data can tell us something about the sectoral trend for much of the developed world through June.

Pace of recovery varies across industries

Note: As of Jun 2020. Trough since Feb 2020. Source: Macrobond, RBC GAM

Several findings leap out. The worst initial declines – with more than 60% drops – were in the accommodation and food services sector and the art, entertainment and recreation sector. But the story since then has differed significantly, as accommodation and food services have since rebounded substantially (if far from completely), while the arts, entertainment and recreation sector remains extremely depressed.

Sectors including transportation and warehousing, retail, and manufacturing were initially down by around 30%. However, their recovery has differed significantly. The first category has lagged badly, manufacturing has recovered more than half of what it initially lost, while retail activity now sits outright higher than it was before the pandemic.

Finally, several prominent sectors had unusually mild initial declines, including public administration, utilities, agriculture, finance, and real estate. The first three of these are intuitive, the fourth seems plausible, while the fifth is something of a surprise.

Trade data

We’ve felt increasingly guilty over the past several months as our far-ranging economic investigations have failed to properly tackle international trade. We resolve that oversight now, through the lens of U.S. exports.

Unsurprisingly, trade suffered a sharp drop during the pandemic. Consistent with the historical experience, the initial decline was considerably sharper on a percentage basis than was the decline in economic output. Whereas U.S. GDP fell by around 15% at its worst, U.S. exports declined by more than 30%. Furthermore, the recovery has been less substantial so far. While we estimate that the U.S. economy had recovered more than half of its initial decline by July, well under half of the export drop had been recovered.

Of course, other complications on the trade file include:

• an ongoing trade war with China
• a broader isolationist trend
• governments prioritizing the recovery of other sectors (such as employment and consumer spending)
• some sector-specific restrictions that have not yet been resolved (see next chart).

COVID-19 impact on U.S. exports

Note: As of Jul 2020. Source: BEA, U.S. Census Bureau, Macrobond, RBC GAM

Fascinating sector-level developments include:

• Travel exports fell by more than 70% and have not yet recovered at all. For those wondering just what a travel export is, this would be foreigners coming to visit a country. Given largely closed borders, there really isn’t a way to get this back in the near term.
• Transportation exports suffered nearly as large a decline, and only a muted recovery since.
• While consumer goods and industrial supplies experienced a similar initial decline in exports, the consumer rebound has been considerably greater – consistent with enthusiastic consumers around the world.
• The demand for computers understandably rebounded after an initial dip – a logical outcome given the many workers and students scrambling for laptops in a newly virtual world.
• Service exports such as telecom and financial services held up fairly well throughout.
• Ultimately, the service sector had both the best performing and the worst performing sectors.

She-cession / Young-cession

Male workers usually fare worse than women in recessions as the construction and manufacturing sectors tend to be the most affected.

As has been widely reported, this pandemic has been different. Indeed, it has been a veritable “she-cession”, because women lost a disproportionate share of the jobs during the recession (see next chart). This is because the sectors that were least able to function under social distancing, such as restaurants and retail, disproportionately employ women. Women suffered a 17% decline in employment, versus men with a 14% decline. Arguably even more profoundly, the recovery in employment for female workers has been much slower than for men.

Young-cession / She-cession

Note: As of Aug 2020. Calculated as the percent decline in Canadian employment by gender and age. Source: StatCan LFS, Haver, RBC GAM

But it wasn’t just a she-cession. It was even more notably a young-cession. In other words, younger workers also did much worse. As the chart above conveys, 22% of Canadian workers aged 15 to 34 years lost their jobs during the pandemic, versus just 12% of those aged 35 and older. Fortunately, and continuing with these goofy terms, it has also been a young-covery in that the recovery has seen younger workers regain their lost jobs far more quickly. 94% of the jobs lost by young people have already been reclaimed, whereas only 67% of the jobs of older people have been.

Of course, there are any number of ways to slice the current situation. Turning from the economic data to the fatality data, it was a he-demic (a pandemic disproportionately killing men) globally, and beyond a shadow of a doubt an old-demic (a pandemic disproportionately killing older people).

There are estimated to have been 1.4 male pandemic deaths for every female death globally, with more men than women dying in each of the U.S., Eurozone and U.K. Representing a curious exception to this, Canada reports slightly more female deaths than male deaths.

And, of course, COVID-19 has been an old-demic on the grandest of scales, disproportionately killing older people. Someone aged 75 or more has literally a hundred times greater vulnerability than someone under the age of 50.

Alas, then, we can say that nearly every group has suffered in some regard: young people and women from an economic standpoint, old people and men from a fatality perspective.

Tracking headwinds

The second viral wave is proving concerning, spurred on by a combination of fewer social distancing rules, social distancing fatigue, reopening schools and autumn weather. In turn, we figure the rate of economic growth in affected countries should sputter over the next few months before rising more forcefully afterwards.

Fortunately, other prominent economic headwinds are proving somewhat less problematic than feared – at least so far.

Fiscal drag

The U.S. fiscal drag that comes after the fiscal boom is already underway. Way back in May there was already the beginning of a reduction in government spending as the prior month’s one-time cheque to households was not continued in subsequent months. Late July brought another inflection, as the generosity of unemployment cheques diminished. Altogether, the U.S. federal government is now spending around $200 billion less per month than it was in April.

U.S. state-level spending is also in the process of shrinking. Most U.S. states are subject to balanced budget rules. With state government revenue expected to be around 7% lower than normal in 2020 and about 15% below normal in 2021, that means an equivalent amount of spending must be curtailed. That represents around 1% of GDP potentially lost in each year.

Collectively, these changes represent significant fiscal drags. And yet, promisingly, it appears that the U.S. economy continued to expand through August.

To the extent other countries have been more cautious in withdrawing stimulus, the risk is smaller elsewhere – at least for 2020. Canada’s latest Throne Speech, for instance, appears to envision even greater government largesse including in such areas as pharmacare and daycares, rather than austerity ahead.

Lagged credit problems

Classically, credit problems accrue with a lag after a recession. Defaults don’t peak immediately, but instead over the subsequent year as financial problems crystalize.

Fortunately, the problems don’t appear to be quite as big as initially feared. While credit defaults are significant and not yet complete, the expectation is that they will peak at well below financial crisis levels.

There is still considerable ambiguity in the household credit and bank loan space. But banks seem increasingly comfortable with the credit losses they have provisioned for, and actual losses remain quite small. Of course, various loan deferral programs mean that the full extent of such losses have not yet been realized. But mortgage deferrals are improving significantly. For example:

• The fraction of Canadian mortgage holders deferring their mortgage is down from around 16% at its peak to 6% today.
• The U.K. equivalent has improved from 14% to 6%.
• In the U.S., the improvement is a more muted 8% to 7%, though the end point is similar.

Furthermore, experts estimate that only a tiny fraction of mortgages will eventually be foreclosed upon. Home prices are rising such that there are easier escapes for financially encumbered households.

Housing risks

Housing markets have so far been quite strong. At some point, it would make sense if they soften given high unemployment and low immigration. But there is no evidence of that so far, and some of the aforementioned household credit problems are already becoming less severe. We still imagine the housing market should be somewhat weaker in the future. But it could be that any housing weakness arrives with a sufficient lag that the economy is in a much better position to handle it, having already deftly navigated around many fiscal and credit challenges.

Mini financial market update

It merits mention that equity markets suffered a moderate retreat in September. This is most substantial in the U.S. stock market, which is down around 7%. The decline elsewhere is more muted. In all cases, equity markets remain massively higher than their late-March lows.

The story is disproportionately a U.S. one, driven by such considerations as:

• A tech boom centred in the U.S. that has now partially reversed.
• U.S. political concerns, revolving around such matters as fading fiscal support, a Supreme Court nomination battle and the upcoming election. At a minimum, higher-than-normal U.S. volatility seems a reasonable expectation.
• Additional Chinese sanctions have also been cited as a factor, though they seem fairly small.

For some time we have been positioned somewhat underweight U.S. equities within the broader equity basket.

Of course, as this week’s note flags, the situation is hardly perfect on a global basis. The COVID-19 numbers are not looking especially good. The economic recovery is not advancing as quickly as it once was.

That said, and despite a variety of evident challenges, we continue to anticipate higher equity markets over the coming year, grounded in rising economic activity, recovering earnings and not unreasonable valuations in an environment of very low interest rates. Additional details regarding our financial market outlook are available in the latest Global Investment Outlook.

-With contributions from Vivien Lee and Kiki Oyerinde