Our latest monthly economic webcast is now available, entitled “Responding to the second wave.”
The bad and good news has been roughly evenly distributed over the past week.
Key negatives include:
- The COVID-19 numbers continue to deteriorate across most of the developed world.
- U.S. President Trump has tested positive for the virus – injecting more chaos into the U.S. election and the way forward more generally.
- Real-time mobility data continues to decline and economic data points to only sluggish growth.
Conversely, positives include:
- We highlight the prospect of upgraded economic forecasts.
- The U.S. COVID-19 numbers are not rising as badly as in many countries.
- Political pundits point to a rising chance of U.S. fiscal stimulus before the election (though we are still unconvinced).
- Brexit negotiations have recommenced, with a plausible path toward a trade deal now visible.
Versus prior pandemics
Gazing back into history, COVID-19 is now the 20th most deadline pandemic in recorded history, as measured by fatalities as a share of the global population. Thankfully, only 0.013% of the world’s people have died of COVID-19 so far. Of course, this nevertheless represents more than a million deaths. Any loss of discipline could allow the virus to spread more fully and ultimately boost the fatality rate by as much as 10 to 50 times.
Even a worst-case scenario should leave COVID-19 with fewer total deaths than the HIV/AIDS epidemic (0.57% of the global population, in ninth place). The Spanish Flu is in fifth place, with a global fatality rate of 2.7%. The two most deadly pandemics in history – The Black Death of 1347-1351 and the Plague of Justinian from 541 to 542 – killed 42.1% and 28.5% of the global population, respectively. It could be far, far worse.
The daily COVID-19 infection figures continue to inch higher, with nearly 300,000 new infections per day across the world. Over 5,000 deaths per day continue to be recorded (see next chart).
Global COVID-19 cases and deaths
Note: As of 10/05/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM
As such, the global transmission rate remains above one, but only just barely (see next chart). It remains plausible that the global infection numbers could stabilize or even improve from here, given the recent trend.
Global transmission rate hovering around key threshold of one
Note: As of 10/05/2020. 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
The deterioration over the past month is entirely a function of the developed world (see next chart).
COVID-19 infections: emerging markets vs. developed markets
Note: As of 10/05/2020. Calculated as the 7-day moving average of daily infections. Source: ECDC, Macrobond, RBC GAM
The fraction of tests coming back positive in the developed world has risen visibly over the past month, hinting that the true increase in developed-world cases is even greater than officially registered. Of course, the miscounting during the first wave was even more extreme.
Among emerging market countries, the overall COVID-19 infection trend is roughly flat. However, this masks a theme of convergence. The big countries that had previously been badly affected – such as India, Brazil and South Africa – are now seemingly improving. Russia represents an exception to this rule.
On the flip side, the very fact that the overall emerging market infection rate remains roughly flat indicates that many smaller countries must be suffering a rising number of new cases. Eastern European nations are particularly prominent in this regard.
If it defies belief that a country like Brazil is managing to reduce its virus count despite a government that has attempted to diminish the severity of the disease, one possible explanation is mask usage. Mask usage has been unusually thorough across Latin America, where 70% of survey respondents indicate they always wear a mask when going out. This is in contrast to less than 30% in the U.S. and U.K. Simple solutions like universal mask-wearing can obviate the kind of sector-level restrictions that cause so much economic pain.
The U.S. COVID-19 daily infection numbers continue to rise, but not as forcefully as elsewhere. They remain well below the levels reached during the country’s (mid-summer) second wave. Nevertheless, even after factoring in the country’s immense population, its 55,000 new cases per day are not very good (see next chart). Furthermore, hospitalizations have begun to inch higher again.
The U.S. remains a perplexing mix of jurisdictions in which social distancing restrictions are being eased (such as in Florida) and regions in which the rules are being tightened anew, including New York City.
COVID-19 cases and deaths in the U.S.
Note: As of 10/05/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM
The U.K. virus numbers continue to leap higher, with the country now processing around 7,000 new cases per day. Fortunately the U.K. is now becoming increasingly aggressive about controlling the spread of the virus. This includes asking office workers to work from home, closing bars and restaurants by 10 p.m. and banning indoor team sports. More may be needed.
Canada’s infection numbers also continue to climb, with a 32% increase over the past week alone. The country is now recording nearly 2,000 new cases per day (see next chart). This is officially more than the peak of the first wave in the spring. In practice, though, serious under-testing at that point means that the second wave is definitely still smaller in reality, at least for now. The fatalities figures have also jumped several-fold, though remain well below the spring peak.
COVID-19 cases and deaths in Canada
Note: As of 10/05/2020. 7-day moving average of daily new cases and new deaths. Source: ECDC, Macrobond, RBC GAM
Regionally, Quebec has once again claimed the mantle of most new cases, with a trend rate of nearly 1,000 per day and still rising. However, the province stands a better chance than most at flattening its curve. It is now aggressively locking down the most inundated parts of the province, closing bars, casinos, cinemas and sit-in restaurants. Furthermore, private gatherings are now forbidden: people cannot have any visitors from another address in their homes, and outdoor visits are also now banned.
Ontario is now logging over 500 new infections per day and rising quickly. Like Quebec, Ontario has also opted to target specific areas of the province, but it has taken a much less aggressive approach so far. Restaurants, bars and nightclubs are still open, albeit capped at 100 patrons. Gyms and banquet halls will now be limited to 50 members. We flag the high probability that these actions prove insufficient by themselves, with further tightening efforts likely necessary. Ontario’s virus control efforts are further stymied by a significant testing backlog.
Contrary to the country’s two biggest and adversely affected provinces, British Columbia’s virus numbers are now edging slightly lower again.
Europe’s second wave
Europe’s second wave continues, albeit with some signs of improvement. France may now be finally starting to improve after cresting at around 12,000 new cases per day (see next chart). In addition to prior tightening efforts, it recently announced the closure of the country’s bars for two weeks.
COVID-19 cases and deaths in France
The Spanish situation has similarly improved, though remains precarious with a still-high infection rate of over 10,000 per day.
On the other hand, less affected countries such as Italy and Germany are continuing to get incrementally worse (see Germany on next chart). It is a frustrating situation: it seems countries are unable to learn from one another, or even to recall the lessons from their own prior waves.
COVID-19 cases and deaths in Germany
Risk of openness index
We have long used Oxford University’s COVID-19 policy stringency indices to help in the evaluation of the relative severity of social distancing rules in different countries, thus gauging the probability of each nation successfully controlling the virus.
Recently, Oxford added a further measure called the “risk of openness” index. It combines the aforementioned stringency data with information about the quality of health care in a country, the rate at which the virus is spreading and the extent to which the public has been educated on preventative measures. In short, it is – in theory – a superior metric of which countries would be at greatest risk if they were to completely reopen, versus which would fare better.
It should be noted that we suspect almost every country would run into serious trouble if it were to completely abandon COVID-19 restrictions. Perhaps China and New Zealand constitute plausible exceptions. Nevertheless, the risk of openness index says something about which countries are best positioned in the context of COVID-19, and arguably even more accurately than the older stringency data.
In the developed world, the countries at greatest risk are, perhaps unsurprisingly, Spain and France (see next chart). The U.S. comes third. The U.K. is rising quickly, in fourth. There are then four other countries that sit considerably lower on the risk axis: Japan, Italy, Germany and Canada. Canada is actually ranked best of the bunch, though even it is suffering a rising infection rate presently. All are considered more vulnerable than they were in the middle of the summer.
Risk of openness in developed markets
Note: As of Sep 2020. The index shows the amount of risk a country faces if its government adopts an ‘open’ policy stance. Source: University of Oxford, OxCGRT, RBC GAM
Among emerging market countries, Latin American nations have the greatest risks, though the degree of risk is now declining in some cases (see next chart).
Risk of openness in emerging markets
Note: As of Sep 2020. The index shows the amount of risk a country faces if its government adopts an ‘open’ policy stance. Source: University of Oxford, OxCGRT, RBC GAM
Risk by activity
We continue to revisit (some might say obsessively!) the subject of what activities and circumstances contribute to the greatest amount of COVID-19 spread. The honest answer is that there is no single, definitive answer. Tracking and analytic work yield only broad, tentative empirical conclusions as outlined in our prior two weekly notes.
Of course, governments have access to much more data than we or academics do. They are the ones that possess all of the tracing data and even have access to hospital admissions notes that might provide hints as to the provenance of each infection. While we await the release of such details, it seems reasonable to assume that the government’s own classification of what activities are a low risk, medium risk and high risk should align fairly closely with this unpublished data.
Leaning heavily on the Government of Canada’s assessment (though ultimately informed by a variety of jurisdictions – the conclusions are remarkably similar for all), the following table lays out the relative ease of transmitting the virus by activity type:
Activities by risk (and what might need to be halted to control virus)
Note: Bolded activities are ones with direct economic consequences. Source: RBC GAM, Government of Canada
We have bolded activities with a significant economic component so as to delineate between activities whose halt might damage the economy versus those with a lesser impact (though conceivably affecting the quality of life of the participants).
Now, as much of the developed world struggles with a rising virus count anew, the obvious solution is to limit many, if not all, of the high-risk activities. Helpfully, most are not economic necessities in the way that a medical appointment (medium risk) or a trip to the grocery store (low risk) are. Despite this, any shutdown is devastating to the affected economic sectors. ence the hesitant response by policymakers so far.
A further policy consideration is that this list only indicates the extent to which an individual would be at risk of contracting COVID-19 by engaging in a specific activity. From a society-wide perspective, it also matters how many people are engaging in that activity. For instance, there are far more people who normally work in an office or who attend school than who go to casinos. As such, while the casino is the more dangerous pursuit, the offices and schools could well be responsible for many more cases. Policymakers must consider this factor as well, and also the relative necessity of each activity.
Of course, one can alternatively break the risk of transmission into its elemental parts. What those lists actually capture is the extent to which people are in close proximity to one another, how long that proximity lasts, the extent to which masks are worn, hands are washed and so on.
A new research report from a Princeton University researcher looked at the transmission of COVID-19 in India, resulting in the largest study yet conducted on the subject. The key takeaway is that 71% of patients were not responsible for any further spread of the virus. Another 21% accounted for around 40% of the virus’ spread, while a mere 8% of the people who had been infected were responsible for 60% of the spread.
It isn’t so much that the innate biology of these people make them more likely to spread the virus, but instead that their behavior enabled it: transmission was disproportionately a function of their proximity to others and the length of time they were close to others. Unfortunately, this means that super-spreaders cannot necessarily be identified in advance, at least not in terms of their basic physiology. Instead, the key is to avoid high-risk situations. It is furthermore heartening that 71% of patients managed to avoid transmitting it to anyone else. Most people are behaving responsibly.
A recent study found no link between the spread of COVID-19 and the temperature, the amount of rainfall or the velocity of the wind. But there was a huge link between the spread of COVID-19 and the humidity, with each 1 percentage point drop in humidity responsible for a big 7% to 8% increase in new infections. As such, while the winter might not translate into more infections due to colder weather, it likely will due to the dry air that comes with that colder weather.
A recent Japanese study concluded that face shields don’t do much to stop the spread of COVID-19 when the wearer is having a conversation. Of course, face shields have a substantial opening at the bottom. More conventional masks still appear to work.
The coming months contain a social calendar filled with transmission risks, including:
- Canadian and U.S. Thanksgiving
- The U.S. election
- The holiday season
The risk is that even if government restrictions prove sufficient to quell the second wave in the near term, these potential super-spreading events could keep the virus circulating at an elevated rate into early 2021.
There are a number of important developments to share with regard to the science surrounding COVID-19.
Another crack at the true virus numbers
We wrote last week that the true number of people who have been infected by COVID-19 in the U.S. could plausibly be in the range of three to six times greater than officially tabulated. This is due mainly to information gleaned from serological blood testing. In other words, whereas 2.1% of the U.S. population has officially been infected by COVID-19, the true figure might be as high as around 12%.
Over the past week, The Economist magazine came out with its own estimates for the true number of people who have been infected. Globally, it figures the actual count may be a remarkable 21 times higher than officially captured. Instead of the 30 million confirmed infections, it estimates that around 630 million people may have been infected.
While The Economist’s numbers don’t precisely line up with our own, they aren’t quite as different as they first seem. The great bulk of the uncounted COVID-19 infections in The Economist originate from emerging markets. These are known to massively under-test and so to under-tally their infections. For instance, the report figures that India, Africa and the rest of Asia (non-India, non-China) were each responsible for more than a million new infections per day during the worst of the spring outbreak – well over an order of magnitude above the official numbers. For that matter, the overall estimate that 8% of the world has been infected isn’t so different than our U.S. estimate range of 6% to 12%.
It should be noted that The Economist generates these estimates by modelling the number of reported infections and fatalities for each country (a metric that is less likely to be drastically undercounted), further adjusting for the level of income in each country as a proxy for the extent of likely undercounting. As such, it is a highly theoretical exercise.
If true, The Economist numbers argue that the first wave wasn’t just worse than the second wave, but was radically worse, such that there hasn’t even been a second wave – the numbers continue to trend steadily downward since early May. We find this aspect of the report hard to reconcile with the fact that the global fatality numbers (refer back to our first chart in this report) are now clearly higher than they were in the late spring, despite the fact that medical enhancements should be reducing the fatality rate over time.
Additionally, the rate of asymptomatic infection, while significant, arguably isn’t high enough to yield undercounting of this magnitude. As such, we suspect The Economist study exaggerates the true number of infected. It nevertheless makes important points about the significance of undercounting, especially in the developing world.
Answering a question we had been wondering for a while, The Helsinki airport in Finland is now using dogs to sniff out cases of COVID-19. Canines are famous for their sensitive noses and ability to detect drugs, contraband and even certain health disorders. It is unclear how accurate the dogs are or whether this practice will be extended elsewhere. At a minimum, it would seem a tricky process to train the dogs while using the live virus.
It has been widely reported that flu deaths were sharply lower in Australia over the country’s winter season – by a factor of 10 –relative to the year before. This is presumably because of less social interaction diminishing the spread of all infectious diseases. It is welcome not just because the flu itself is unpleasant and potentially deadly but because this phenomenon should diminish the degree of confusion as to whether someone with a fever and cough has the flu or COVID-19.
That said, and representing something of an aside, we remain astonished that the positive rate on COVID-19 tests, even at their current slightly elevated levels relative to the summer, are only around 2% in Canada and 5% in the U.S. This is to say, 95% to 98% of people who go for testing are ultimately found not to have COVID-19. It is frankly mysterious what actual infections these people have if the incidence of flu is so low. In fairness, many are getting tests because they have theoretically been exposed to someone with COVID-19, and so this may explain the many negatives.
The drugs remdesivir and dexamethasone remain two of the most useful medicinal treatments in combatting COVID-19. However, recent studies appear to suggest that the incidence of death does not quite fall by the 30% to 50% proportion initially estimated. For instance, the European Medicines Agency has recently concluded that dexamethasone reduces the risk of death by 20% to 35%. Nevertheless, this is still quite useful.
While some studies have argued that Vitamin D radically reduces the likelihood of contracting a severe case of COVID-19, a thorough meta-analysis has concluded that the benefit is actually fairly small, though seemingly real.
While vaccines encourage one’s body to create antibodies that can defeat COVID-19, there are also efforts underway to produce synthetic antibodies that themselves defeat COVID-19. Initial trials have been promising. Incidentally, these recently came to widespread prominence after an experimental version was administered to the ailing President Trump.
While the distinction between these synthetic antibodies and a vaccine might seem trifling, in practice there are three key differences, two positive and one negative. The positives are that the body doesn’t have to fight off a (fake) illness to develop the antibodies and the treatment can even be administered when someone has already been infected by COVID-19 (as in the case of President Trump). That isn’t possible with a vaccine. The big negative, beyond their relative novelty, is that synthetic antibody treatments have historically cost tens of thousands of dollars per person – a huge bill when compared to a few dollars for a vaccine inoculation.
Nevertheless, the fact that hundreds of millions of dollars are being spent pursuing synthetic antibody treatments argues that they may play a very real if underappreciated role in eventually quelling the virus, and perhaps eventually in treating a broader array of diseases.
The following table summarizes our main views on the pursuit of a COVID-19 vaccine. In short, all seven key obstacles appear to be surmountable and, as such, a vaccine solution is likely if not necessarily capable of delivering full herd immunity.
Key COVID-19 vaccine considerations
As of October 2020. Source: RBC GAM
However, the timing of a vaccine has arguably slipped slightly, as per betting markets that now point to a 70% probability of widespread distribution of a vaccine in the U.S. by the end of May in 2021, as opposed to a similar percentage expecting that achievement by the end of March just a few weeks ago. The timing has arguably slipped by two months.
Consistent with this, the most optimistic predictions for emergency availability of a vaccine is now November 2020 (notwithstanding minimally tested vaccines already in use in China and Russia). Previously October and even September had been cited. Among other implications, this means it is now unlikely that there will be a vaccine available before the U.S. election.
But let us not give the impression that progress has stalled or reversed. More vaccines continue to enter Phase 3 trials and pharmaceutical companies continue to line up deals with manufacturers to produce billions of doses and with countries to purchase those doses.
In an effort to speed up the approval process for some vaccine candidates, there is now even an initiative in the U.K. to secure volunteers who would be deliberately infected by the virus to accelerate the process of vetting if the vaccines work. Brave people!
Real-time economic and mobility data remain suspect as seasonal distortions get in the way of properly evaluating the trajectory for the economy.
As argued in the past, we are now quite skeptical of the Apple Mobility data as it flags a sharp drop in car and foot traffic even as the economy appears to continue growing according to other metrics. Seasonal distortions are likely the problem.
Now, Google Mobility data has been temporarily discontinued as the company tries to work through its own problems with consistency and how the data is reported.
We note that the Homebase U.S. hours worked data also shows a gradual decline in September, but recognize that hourly work may naturally dry up in the fall as summer jobs come to an end. It will be interesting to see whether the demand for hourly workers surges into the holiday season.
Percentage change of hours worked by hourly workers in the U.S.
Note: As of 09/28/2020. Impact compares hours worked in a day vs. median for the same day of the week in Jan 2020. 7-day moving average used. Source: Homebase, RBC GAM
Fortunately, we have enough recent data about the U.S. workforce that we are not convinced that hours worked are actually in decline. In fact, it still looks to us that the U.S. and developed-world economies grew through September.
The key insight comes from U.S. weekly jobless claims, which fell yet again from 873K to 837K. The improvement is less reliable than in the past and certainly less profound, but some labour force healing continues. Continuing claims also improved.
The U.S. September payrolls data recorded a gain of 661K jobs, and – despite some fears – the household survey employment number did not give back any of the prior month’s outsized gains. Even the leisure and hospitality sector was up, with a robust 318K new jobs in September alone.
The U.S. has now recovered more than half of the jobs that were lost in March and April. In turn, the U.S. unemployment rate fell from 8.4% to 7.9%.
It is perhaps unsurprising, then, that U.S. households are feeling fairly good. Consumer confidence soared in September according to the Conference Board, though the gain from the equally regarded University of Michigan survey was much more muted.
While U.S. personal income was recently reported to have fallen by a steep 2.7% in August, this isn’t as bad as it first seems, for a few reasons:
- The decline was entirely the function of less generous unemployment insurance benefits that were implemented at the start of that month.
- The level of personal income remains a whopping 2% higher than normal even after the drop.
- The U.S. personal savings rate remains elevated at 14.1% (though is radically lower than the unheard of 33.7% high in April).
- By virtue of this savings buffer, personal spending in August still managed to increase by a respectable 1.0%. Households are not overly crimped despite the hit to income.
On the business side of the ledger, the U.S. ISM (Institute for Supply Management) Manufacturing Index fell a hair in September, from 56.0 to 55.4. This is still consistent with good business sector growth. The employment index actually improved on the month, though new orders became incrementally less impressive.
Internationally, the Eurozone final composite PMI (Purchasing Managers’ Index) for September squeaked out a slight gain, from 50.1 to 50.4. While higher, this is not a level associated with much growth. Arguably this makes sense given that the continent is awash in new virus cases.
China’s official PMI rose from 54.5 to 55.1 in September – a welcome increase and a healthy level (particularly in the historical context of this measure, which has often appeared depressed even as China recorded good growth). China’s economy remains a beacon of light.
Canada’s monthly GDP print for July is now out, and rose by 3% – precisely as had been expected. Of greater interest, Statistics Canada signaled a 1% gain was likely for August. In an absolute sense, that represents another robust reading – it is consistent with a 13% annualized rate of growth, after all. However, it also confirms the ongoing deceleration of growth, from 6% in June to 3% in July to 1% in August.
If the economy were actively decelerating in September and was simultaneously struck by a second viral wave, that suggests at best meagre growth in the month. We suspect Canada will find a way to continue growing in September and October, if gingerly, and note that – through August – the economy has now recovered 69% of its initial economic decline. Canadian employment will be released later this week.
We now turn to three quick nibbles on other Canadian economic matters. First, Canadian immigration remains sharply lower than normal. There have been around 25,000 fewer immigrants coming into the country than in the past – a roughly 60% decline (see next chart). It remains surprising that this hasn’t dampened demand for homes, but we have every confidence that the immigration levels will eventually normalize once the pandemic has been handled.
Immigrants admitted to Canada tumbled after border closure
Note: As of Jul 2020. Source: IRCC, Open Canada, RBC GAM
Second, Canadian hotel occupancy had increased sharply from the spring through late summer, but is now in retreat (see next chart). This may be in part as the second viral wave heats up. But we suspect the main reason is that the summer holiday season has ended and yet there is little business travel to supplant it.
Canadian hotel occupancy rate dwindling
Note: For the week ending 09/26/2020. Source: STR, Wall Street Journal, RBC GAM
Third, Canadian auto sales have staged a remarkable but not yet full recovery (see next chart). This is quite similar to the U.S. experience. On the one hand, one would have thought that with extra government stimulus in consumer pockets and a temporary distaste for public transportation, car sales would be thriving. Consumers have not been shy about buying smaller ticket consumer goods or snapping up homes.
On the other hand, this is a time of high uncertainty. Many workers don’t actually need to travel to work. The need for a second family car has thus probably rarely been lower. In the end, auto sales remain slightly depressed but we suspect they will continue to normalize over the coming year.
Canadian auto sales plunged at the beginning of lockdown
Note: As of Jul 2020. New motor vehicle sales seasonally adjusted. Shaded area represents recession. Source: Statistics Canada, Macrobond, RBC GAM
Revisiting economic forecasts
Several events have transpired over the past few months that interact with the economic outlook. Among the key considerations, the second virus wave should slow growth over a period of a few months in the affected countries. However, arguing in the opposite direction:
- The official economic data continues to outperform our initial rough-hewn real-time estimates.
- The U.S. economy seemingly avoided shrinking in July and August despite the country’s own second wave. In each month, the U.S. growth rate appears to have been about half that of Canada, but no worse. Even now, with the Eurozone dealing with its own second wave since August, the continent’s September data remains mostly consistent with economic growth.
- Various feared headwinds have so far proven less problematic than initially feared (discussed in last week’s MacroMemo. These include the fiscal drag, lagged credit problems and lagged housing problems.
Altogether, these changes collectively improve the economic trajectory relative to our existing forecasts. It now appears that U.S. GDP growth for 2020 could be as strong as -3.5% rather than our prior estimate of -6.0%. This is primarily a function of a better-than-expected outcome in the just-completed third quarter. For Canada, 2020 GDP growth now appears to be in the realm of -5.2% versus our prior estimate of -7.0%. These both now constitute above-consensus rather than below-consensus outcomes. Conversely, the outlook hasn’t changed very much for the Eurozone, U.K. or Japan.
While these are still only working estimates, the new sprightlier U.S. growth profile argues that the U.S. economy could now return to its prior peak as early as the end of 2021 instead of the prior assumption of mid-2022. Canada might return to normal by early 2022 – slightly ahead of its prior schedule as well.
Of course, in this analysis, much depends on the timing of a vaccine – a still uncertain proposition. Furthermore, this latest wave could well prove more challenging than the U.S. summer wave to the extent it coincides with colder, drier weather and is sprinkled with socially interactive holidays.
U.S. political update
U.S. political junkies have been drinking from the fire hose over the past few weeks. Not only has President Trump been infected by COVID-19 but the effort to appoint a new Supreme Court justice is becoming more complicated, the first (and conceivably last) presidential debate occurred, and the polls are very much in motion. We review these developments and more.
Trump has COVID-19
President Trump has now contracted COVID-19. Reports on his medical status run the gamut. On the pessimistic side, he spent two nights in the hospital, has received oxygen, was administered an experimental antibody cocktail and received remdesivir and dexamethasone – two therapeutic drugs normally used only on the most vulnerable COVID-19 patients.
However, he was also seen walking to his helicopter, has given several statements, has started tweeting again, and opted for an impromptu car ride. He is scheduled to be released from hospital today.
What does this mean for the election outcome? It certainly introduces a wildcard. While one might argue Trump could garner a sympathy vote, the development is probably more negative than positive for the Republican candidate. It not only ties the President more tightly to the unpopular COVID-19 issue but – even in the best possible outcome – it will leave him off the campaign trail for a significant fraction of the final month of the campaign. In the event that his health were significantly diminished afterward, this could interfere even more with his push for re-election. The President’s campaign manager has also now tested positive, further limiting the campaign’s efforts. If the President were leading in the polls, all of this might be manageable. But instead it is Biden with a significant lead and the President is now running out of time to catch up. The polls and betting markets appear to agree with our assessment that it is a net negative for Trump.
Interestingly, the political fortunes of British Prime Minister Johnson and Brazilian President Bolsonaro did not materially change despite their own encounters with COVID-19. The same goes for three U.S. state governors who contracted the virus.
On an optimistic note, one might hope that Trump’s infection will improve U.S. mask compliance, perhaps helping to limit the virus.
What would happen if Trump died from the virus? Despite his age, Trump is unlikely to die. Not only do the significant majority of people his age survive their bout with COVID-19, but he has access to arguably the best medical care in the world. He has already received an antibody treatment that virtually no other human on the planet can access.
If Trump were to become incapacitated for a period of time before the election, Vice President Pence would serve as acting President, likely in a caretaker capacity.
In the unlikely event that Trump were to die before the election, Pence would then become the President. It would be up to the Republican National Committee to decide who would be the new nominee for the election, but this would very likely be Pence. For what it is worth, Pence’s policy views are arguably less conservative than Trump on economic matters but more conservative on social matters. It is unclear how voters would respond to a suddenly new Republican candidate – and for that matter some have already voted by mail – but it would be difficult for Pence to match Trump’s expected results with such little notice.
Finally, it is unlikely that Trump would die after the election given that the virus would normally run its course before then, but once again Pence would likely end up as President in that scenario.
Supreme Court justice
President Trump has now nominated Amy Coney Barrett for the U.S. Supreme Court seat vacated by the late Ruth Bader Ginsburg. Despite Democratic Party complaints about the proximity to the election and an effort to let the next President make the selection, it initially appeared that there was little the Democrats could do to halt the process. With the White House and sufficient Republican votes in the Senate, Barrett was quite likely to be appointed before any handover of power at the start of 2021.
However, the recent COVID-19 outbreak among Republican Party luminaries has managed to infect not just the President but three Republican Senators. It just so happens that one other Republican Senator had previously indicated they would not support the nomination effort before the election. As such, the Republicans are – temporarily – one vote shy of the minimum needed to approve Barrett. But this probably doesn’t change the final result. The Senate is now recessed until October 16, at which point there will still be two and a half months to get the confirmation across the finish line. In turn, the Supreme Court will probably pivot in an even more conservative direction.
While the first presidential debate (and possibly the last one given the President’s new health challenge) occurred a mere week ago, it now feels like a distant memory given all that has since happened.
With regard to the question of “who won?” the most popular quip was that the American people had clearly lost. It was a chaotic debate with little content and plagued by constant interruptions, disproportionately by the President.
While the exit polls had Biden faring moderately better than Trump, the results were broadly in line with pre-existing polls and so not a signal of a sudden change in preference.
That said, and despite the fact that debates have rarely significantly changed the election outlook, Biden’s market-assessed odds of victory have just increased substantially. Betting markets argue that the probability that he will win has improved from just over of 55% to around 65% (see next chart).
Biden leads Trump, gap widened after first debate
Note: As of 10/01/2020. Based on prediction markets and RBC GAM calculations. Source: PredictIt, RBC GAM
Election models continue to be even bolder in their assessment, with the FiveThirtyEight model now identifying an 81% chance of victory for Biden and The Economist magazine calculating an 89% chance. Our sense is that the truth may lie in the middle. Biden is probably more likely to win than the market thinks, but perhaps not quite as certain as these models claim.
It isn’t hard to understand how the models think his prospects are so good. Of the eight most fiercely contested states in the U.S., seven now favour Biden (see next chart). If it persists through the election, this would yield a large margin of victory. In turn, the most likely Congressional orientation is a Democratic Party sweep of both the House of Representatives and the Senate, albeit the latter should be fairly close.
U.S. presidential election Trump vs. Biden
Note: As of Oct 2020. Political poll gap for the U.S. presidential election, Trump vs. Biden. Source: Real Clear Politics, Macrobond, RBC GAM
We have already laid out the differences in the Biden versus Trump platforms in the past. The following table provides a concise visual means of conveying the main economic and financial market implications of their differences. We remain of the opinion that a Biden presidency might be an economic positive (though also another reason for the public debt to rise). At the same time, it might be a stock market negative and positive for bond yields.
Biden platform and implications
Source: RBC GAM
Election result timing
While not impossible, it probably isn’t realistic to expect the full, final election results on November 3. Indeed a recent poll shows that just 14% of respondents expect to know the result the first night. Only around half expect to know the result within a week of the election. Fortunately, very few think the result will remain unknown into 2021.
The prospect of delay is largely a function of the complexities and potential delays associated with mail-in balloting, though also due to the elevated risk of a contested election.
Given the contentiousness of U.S. politics in general, specific threats made by President Trump, the unusual circumstances of mail-in voting and the general chaos of a pandemic-time election, a contested election is a distinct possibility. Contested elections are not merely a relic of the distant past: the Bush/Gore election in 2000 was ultimately litigated in the courts and took over a month to be resolved. In the extremely unlikely event that the courts cannot settle the matter, the House of Representatives gets to make the decision, which would likely yield a Biden victory.
But while this election result could well be contested in a legal sense, it is less likely to be in genuine doubt from an economic or financial market perspective. The polls and betting markets currently predict a fairly large Biden victory. If this holds, it would be hard for a few closely fought states to matter enough to do anything other than delay the inevitable.
Furthermore, in the event of an unexpectedly close outcome, that would probably mean that Congress remains divided, with the Senate in Republican hands and the House of Representatives in Democrat hands. Thus, the next president – whoever they are – would in this scenario be very limited in the legislation they could introduce. No radical policy changes would be forthcoming.
A last item
One last, small, observation: Japan now has a new Prime Minister. Abe has been replaced by Suga. Because both are members of the same party, the transition was amiable and Suga long worked for Abe as his chief cabinet secretary, no sudden policy shifts are expected from here. Abenomics of a sort should continue.
-With contributions from Vivien Lee and Kiki Oyerinde