Economic webcast

Our latest monthly economic webcast is now available, entitled “Near-term growth challenged by second wave.”

Overview

This week’s note covers a range of topics, including the latest COVID-19 virus statistics. However, the focus of the report is very much in the economic space, discussing how this recession differed from the norm, our latest business cycle analysis, a run through the latest economic developments and a glance at how inflation is evolving. Financial market developments and themes are acknowledged, Brexit then gets a quick nod, before what is arguably the main act: a U.S. election preview.

In terms of the flow of recent developments, they have arguably been more negative than positive. Negatives include:

• The ongoing COVID-19 surge in the developed world.
• Mounting expectations of economic damage from the second wave, particularly for Europe.
• Slightly diminished vaccine expectations, both as vaccine timing slips slightly and as a new study argues that antibody protection may not last more than about six months.
• The risk of an unclear election outcome for days or longer.

Fortunately, there are also a few positives:

• Based on current polling, it is quite possible that the election outcome will be clear by Wednesday morning.
• The latest run of economic data was actually quite positive.
• A variety of upside event risks beckon, as discussed later in the report.

Virus developments

The COVID-19 figures remain almost universally challenging. Globally, the world now records around 500K new infections per day, and around 6.5K daily deaths (see next chart). Both trends are rising significantly, and the fatality numbers look as though they could exceed the spring peak within a matter of weeks.

Global COVID-19 cases and deaths

Note: As of 11/02/2020. 7-day moving average of daily new cases and new deaths. Source: CDC, Macrobond, RBC GAM

Developed world countries continue to record the majority of the new cases (see chart below).

COVID-19 emerging markets vs. developed markets infections

Note: As of 11/02/2020. Calculated as the 7-day moving average of daily infections. Source: ECDC, Macrobond, RBC GAM

Within the emerging market space, Brazil and India continue to report declining new infections but other hot spots such as Russia and Poland are deteriorating quickly.

Testing positivity

Among developed countries, the great majority are suffering an accelerating pandemic. Providing some sense for this, 9% of U.S. COVID-19 tests now register positive, alongside 8% in the U.K., 5% in Canada and 2% in France. This is greatly in excess of the roughly 1% rate that prevailed in many of these countries over the summer. It also suggests that testing efforts are becoming insufficient to handle the growing number of infections, or at a minimum that the extent of the COVID-19 increase may be underestimated.

Very high transmission rates

The COVID-19 transmission rate is now substantially above the critical threshold of one that separates a virus in retreat from one that is advancing (see next chart). This is not a case of the transmission rate sitting at 1.05, such that the number of daily infections increases by around 25% over a month. The transmission rate in most of these countries is 1.2 or higher. This is consistent with the infection count doubling every month, indefinitely. For those with a transmission rate of 1.5 or higher, the number of daily infections are on track to increase by a startling factor of five over the span of the next month.

Transmission rate over one means COVID-19 accelerating

Note: As of 11/02/2020. 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

This is clearly bad from an infection perspective, but it is no less bad news from an economic standpoint. To bring the transmission rate back below one will require more than just tweaking the number of people allowed to congregate or adjusting the operating rules for a handful of businesses. It arguably demands more aggressive action, as demonstrated by the recent major policy pivots from the likes of France and the U.K. to lock their economies back down.

These new restrictions are not as extreme as the measures implemented last spring, but are nevertheless serious. In the U.K., bars, restaurants and non-essential businesses such as barber shops and gyms are being closed. Furthermore, people are only allowed to leave home for specific purposes. These include:

• education
• work (only if unable to work from home)
• food shopping
• health reasons
• exercise outdoors with one’s own household members or, when unaccompanied by members of one’s bubble, with a single person from another household

Conversely, schools have been prioritized and will remain open. These new rules will stay in place until early December, and could yet be extended.

Europe

France remains the most adversely affected European country, with around 40K new infections per day. However, it is now making a serious effort with a lockdown of its own that closely resembles the U.K. with the added restriction that recreational activities must occur within just one kilometre of one’s home and last no more than one hour.

In fact, even though the measures were only implemented last week, the French infection numbers tentatively appear to be peaking (see next chart). To the extent that social distancing rule changes usually only change the trajectory of the virus after a period of two weeks, this hints that the French public had already begun to modify their behavior before the implementation of the new rules. In turn, this could mean that the new rules are stricter than they need to be. Let us see whether the improvement persists and how quickly the infection figures decline before reaching that conclusion.

COVID-19 cases and deaths in France

Note: As of 11/02/2020. 7-day moving average of daily new cases and new deaths. Source: CDC, Macrobond, RBC GAM

Italy’s daily virus count is now above 25K and rising. Spain has increased to 20K per day and Germany records 15K each day. In all countries, the fatality rate is now also rising seriously, albeit well short of the spring readings.

The U.K. is now up to almost 25K new infections per day. However, the combination of its latest aggressive shutdown and an already decelerating trend make the strong argument that the country should start to enjoy an improving trajectory within the next two weeks (see next chart).

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

Note: As of 11/02/2020. 7-day moving average of daily new cases and new deaths. Source: CDC, Macrobond, RBC GAM

United States

The U.S. is now up to around 90K new infections per day – a record – and seems likely to continue its worsening trend from here. In part this is because the southern half of the country is only now starting to encounter the cooler weather that may be a factor in accelerating the spread of the virus. It is also in part because the U.S. has been actively reopening certain sectors and activities rather than shutting them down (with a few exceptions), and because the U.S. has two potential spreading events in November: the election and Thanksgiving (see next chart). Fortunately, the fatality rate has not yet moved significantly.

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

Note: As of 11/02/2020. 7-day moving average of daily new cases and new deaths. Source: CDC, Macrobond, RBC GAM

Canada

Tragically, Canada’s cumulative death toll from COVID-19 has now passed the 10,000 mark. The country now records nearly 3,000 new infections per day – a record, but this is rising at a slower rate of growth than in early October (see next chart). The fatality figures remain low, but are clearly rising.

COVID-19 cases and deaths in Canada

Note: As of 11/02/2020. 7-day moving average of daily new cases and new deaths. Source: CDC, Macrobond, RBC GAM

Provincially, Alberta is experiencing a particularly notable acceleration in cases, as is British Columbia. Ontario continues to deteriorate but at a notably decelerating rate that hints of stabilization in the not-too-distant future. The province cited Canadian Thanksgiving as a spreading event, which gives some hope that the rate of advance may slow as that holiday fades into history. Meanwhile, Quebec – the most adversely affected province in both the first and second waves – continues to make tentative progress in reducing its daily infection count. Quebec’s social distancing restrictions are among the strictest in the country, so this is not a coincidence. Provinces continue to announce incrementally tighter restrictions on a regular basis.

An unusual recession

With the benefit of several months of contemplation it is worth highlighting just how unusual this recession has been -- before the debate begins as to whether any decline in economic activity later this year constitutes a new recession or not.

The agonized discussion we detailed in an earlier #MacroMemo speaks to this, regarding whether the contours of the recession and subsequent recovery could best be defined as a “V”, “U”, Nike swoosh, “W” or even “K.” Normally, the options are merely “V” versus “U”.

The most unique aspect of this recession is that it did not come from economic or financial market excesses, but instead from a completely exogenous shock: a mere 100 nanometre-sized virus. The resulting economic decline was the result of a government-mandated shutdown, paired with the social distancing that naturally occurred when people began to understand the threat. In fact, International Monetary Fund (IMF) analysis argues that, in the developed world, the latter factor was responsible for a greater fraction of the economic decline than the former.

From a timing perspective, the recession arrived unusually abruptly. This is not to say that all recessions are anticipated with pinpoint precision, but normally there are at least a minority of pundits presaging doom due to one excess or another. China’s experience with COVID-19 did technically give a few months’ warning. However, this was not fully heeded because China was tackling the virus aggressively and there were hopes it would not spread beyond the country. In the end, half of the assessment was correct: China did manage to corral the virus, but not until after it had gone global.

One might argue that this recession and recovery have been somewhat tidier than most. We knew from the beginning what sectors would be most damaged: those that require the greatest amount of human interaction and that were shut down by the government. It hasn’t been the organic sprawling mess that recessions usually are.

Similarly, for a recession of this magnitude, there has arguably been less suffering than usual on the aggregate given the completely unprecedented level of fiscal support given to workers who lost their jobs and, to a considerably lesser extent, to businesses. The fact that household incomes are higher than normal is unprecedented during a recession. Furthermore, many sectors are operating at nearly normal levels. However, certain sectors have been nearly completely wiped out given the unequal effects of social distancing rules. To these poor sectors, this is far, far worse than a normal recession of the same magnitude.

In theory, the fact that a large fraction of the economic damage has come from government edicts means that economies can recover unusually quickly as those restrictions are lifted. This has already happened to a significant extent, though the second wave is now temporarily interfering with that process. And, of course, the virus needs to be vanquished before all restrictions can be lifted.

Finally, we wonder whether the next economic expansion might prove somewhat shorter than the last two. Providing convenient support to this musing, the last two – at around a decade each – were unusually long. Furthermore, to the extent this recovery didn’t involve the right-sizing of previously bloated industries or financial market excesses, it argues that such vulnerabilities could still lay in wait in the future. The housing market comes to mind: it has boomed rather than the usual experience of busting during traditional recessions.

But this is still a highly speculative thought. The ultimate measure of whether an economy is at risk of overheating in the coming years is the output gap. This shows room for many years of future growth.

Business cycle continues to advance

Having spent many years charting the glacial evolution of the prior business cycle from “Mid” to “Late” and then tentatively into “End of cycle,” it is amazing how quickly the U.S. business cycle has evolved over the past few quarters.

According to our scorecard approach, which combines 17 different buckets of variables into a single conclusion, the U.S. business cycle has advanced from “Recession” two quarters ago, to “Start of cycle” last quarter, to “Early cycle” today.

To be clear, the beauty of the scorecard approach is that it allows for dissenting opinions (see next table). These are useful to understand where the balance of risks lie.

U.S. has transitioned from “Recession” to “Start of cycle” to tentative “Early cycle”

Source: RBC GAM

For instance, in the present context, “Early cycle” gains the most votes but only by a smidgen over “Start of cycle.” And “Recession” continues to receive a handful of votes, mostly from such indicators as the low level of bond yields, the continuation of ultra-stimulative monetary policy and largely unchanged levels of leverage (see next chart).

U.S. business cycle score

Note: As at 10/29/2020. Calculated via scorecard technique by RBC GAM. Source: RBC GAM

To provide an example of the sorts of variables underlying the scorecard, here is a measure tracking the extent to which private investment contributes to GDP. This normally rises during recoveries and declines from late cycle through to a recession. It has been substantially reviving recently.

Good gauge of late cycle: Private investment / GDP

Note: As of Q3 2020. Shaded area represents recession. Source: BEA, Haver Analytics, RBC GAM

Economic developments

Three challenges

From an economic standpoint, three new challenges argue for a more pessimistic economic outlook.

1. The second virus wave has now prompted some countries to significantly restrict their economies, pointing to the potential for an outright economic decline over the next few months. We have been flagging this for several weeks, but now a new Financial Times survey of economists finds that most expect Eurozone GDP to decline in the fourth quarter. In fact, France’s Finance Minister indicates he expects the French economy to shrink by a big 15% during the second wave – half the decline of the first wave. This sounds too large to us. Similar restrictions in the U.K. are only expected to result in roughly flat activity in the fourth quarter there. Nevertheless, the broader message remains – the Eurozone economy is likely to underperform and probably shrink in the fourth quarter.

2. U.S. election polling has narrowed somewhat and the result could well be unclear or disputed for a period of weeks, potentially discouraging risk-taking economic activity over the duration.

3. The flow of vaccine news has been incrementally negative over the past few weeks. This argues for a slightly slower 2021 growth trajectory to the extent a smaller fraction of the year may be unencumbered by the virus.

Five things to look forward to

But let us recall that all of these things are well appreciated by financial markets. In fact, these factors largely explain last week’s stock market weakness. As such, perhaps the smarter question is instead whether any of these problems might be resolved over the coming months. And on that front, there are several more optimistic thoughts to share:

1. Whether the election uncertainty is resolved on the evening of November 3 or not until December, this is a short-term concern that will be dealt with relatively soon.
2. The most likely election outcome is a Biden win and a Democratic Party sweep. Financial markets appear to favour this outcome based on how they have responded to polling swings over the fall. As such, there could be room for upside here.
3. It is likely that a vaccine manufacturer will announce that their Phase 3 trials were successful over the next several weeks. This would be an important positive. Furthermore, emergency usage is likely to be approved shortly thereafter, permitting actual deployment of the vaccine.
4. It is quite likely that the U.S will opt to deliver another major round of fiscal stimulus. This could arrive either during the lame-duck session this fall (if the election outcome yields the same Republican President, Republican Senate and Democratic House of Representatives) or in early 2021 if Biden wins. 
5. Recent aggressive European restrictions may hurt the economy, but they should also start to tame the virus – an important positive.

The latest U.S. economic data

On the aggregate, economic data has been surprisingly good over the past week.

In the U.S., weekly jobless claims fell to 751K from 791K – the lowest of the cycle, and resolving a two-week period when jobless claims were higher than an earlier reading. It would appear that U.S. job creation is continuing. Indeed this Friday’s monthly payrolls report is expected to yield net job creation of 600K for the month of October.

The U.S. ISM (Institute of Supply Management) Manufacturing Index for October rose impressively from 55.4 to 59.3 – an extremely high reading that makes a further claim for U.S. economic exceptionalism: Europe’s economy appears to have been sputtering during the month, whereas the U.S. was apparently barreling ahead. The subcomponents were no less impressive. This includes a jump in new orders to a spectacular 67.9 and the elevation of the employment index from 49.6 to 53.2.

U.S. third-quarter GDP was also released. While undeniably dated (it spans July through September), the figure was nevertheless above the consensus with a 33.1% annualized increase.

Canadian data

Canada’s August GDP print rose by 1.2% – a slightly better-than-expected outcome. But the real story lay in the details of the report, which presented a forecast of a 0.7% increase in September. Unsurprisingly, the deceleration in growth is expected to continue, but even this diminished increase is still pretty good when compared to a “normal” rate of growth of less than 0.2% per month. If this forecast proves accurate, Canada’s Q3 GDP will have grown by a heady 45% annualized.

To be clear, while this is stronger than the U.S. equivalent, the reason is mainly that the Canadian economy declined more substantially in the second quarter.

Providing a rare glum reading, Canada’s CFIB (Canadian Federation of Independent Business) Business Barometer – a measure of the activity and sentiment of small and medium-sized businesses – fell from 59.2 to 53.3 in October. This is the lowest reading since May, and argues that the Canadian economy continued to decelerate into October. Fortunately, the latest figure is absolutely nothing like the trough of 30.8 recorded in March.

Canada’s October job numbers will provide some further insight into the economic contours of October when released on Friday. Job creation is expected, though substantially weaker than the surprising strength of September.

Finally, the Bank of Canada rendered its latest verdict this week. The Bank kept its policy rate unchanged and pledged to leave the measure unchanged until no sooner than 2023, when the economy is forecast to be approaching normality again. A neutral policy rate is now assumed to be just 1.75% to 2.75%, 50 bps less than the prior assumed range. The amount of bond buying is slated to decrease somewhat because the composition of purchases is being shifted further out the yield curve where they can deliver an outsized benefit for the many term-based loans that exist in Canada.

The Bank largely aligns with our own view that the economy has outperformed expectations in recent months, even as it now slows. Indeed, the Bank of Canada’s new GDP forecasts are quite similar to our own, with a prediction of -5.7% in 2020 and +4.2% in 2021.

International economic data

The Eurozone also published its third-quarter GDP number, announcing a 12.7% increase – substantially better than the +9.6% consensus. Lest any confusion set in, this is a non-annualized figure, in contrast to the standard practice in North America. In European terms, the U.S. third-quarter increase was just 7.4%. Thus, Eurozone growth outpaced the U.S. and much of the world in the third quarter, but only because it so badly undershot other countries in the second quarter. And there is already evidence it has slowed in October.

Japanese industrial production for September rose by a big 4% relative to August – a strong outcome.

China’s two Manufacturing PMIs (Purchasing Manager Indexes) came in roughly flat in October, while the first of two Non-manufacturing PMIs rose slightly. All remain consistent with moderate economic growth of the sort China has already been delivering.

China also held its Fifth Plenum meeting – a confab for top party leaders that sets the strategic agenda for the country over the coming years. Among the highlights:

• China will emphasize strengthening its economy, increasing its technological independence and becoming especially strong in the digital economy.
• China continues to play up the interconnectedness of the economy and the state, as per the plan to “promote a better combination of efficient markets and effective governments.”
• There was nothing to challenge the expectation that President Xi will remain the country’s leader for a third term starting in 2022, despite a long-standing tradition of stepping down after two terms.

Inflation revival

There are two distinct patterns of interest with regard to developed-world inflation.

1. North American inflation readings are reviving noticeably. Having bottomed at 0.2% year-over-year (YoY), U.S. Consumer Price Index (CPI) has now rebounded to 1.4%. Canada bottomed at -0.3% and has now increased to 0.6%. In both cases, the inflation readings remain below normal, but not nearly so much as during the initial phase of the shock.

2. Europe has not experienced the same revival. Eurozone CPI is still very low at -0.3% YoY. U.K. inflation has increased only tentatively from 0.5% to 0.7% YoY. It is a similar story with inflation expectations, which have rebounded notably for the U.S. but not to the same extent for the Eurozone.

It may be tempting to explain this as a demographic phenomenon given Europe’s relatively older population (and for that matter, Japan’s inflation remains in the vicinity of recent lows as well). Yet this argument doesn’t make sense. An aging population helps to explain why one country might have lower inflation than another, not why inflation would fall and then not rebound after a shock.

A more plausible set of explanations is that North America’s economies have had a less challenging second virus wave (so far). Furthermore, these economies are more dynamic than those in Europe. This squares with the observation that Australian and New Zealand inflation have both behaved more like North America than Europe. They are also relatively dynamic economies and have been much less affected by COVID-19.

In the end, inflation has largely aligned with the usual experience after a major economic shock, and should manage to edge higher over the next few years as the economic recovery continues. But this process appears set to be slower for Europe. Over the medium run, one might even imagine a bit more inflation than normal due to such factors as supply chain onshoring and distended central bank balance sheets.

Another Brexit update

The Brexit transition period ends at the stroke of midnight on New Year’s Eve, leaving very little time left for the U.K. to negotiate a proper free-trade agreement with the European Union. The Financial Times reports that the functional deadline for a new deal is mid-November given the need to finalize the text and achieve parliamentary ascent. While such deadlines have proven surprisingly malleable in the past, time is genuinely running out.

Despite the ticking clock, we have actually become slightly more optimistic over the span of the last quarter, reducing our assigned probability of a No Deal Brexit from 55% to 43% (see next graphic). In its place, a shallow free-trade agreement may now be most likely, with a 56% probability. In contrast, an extension is now extremely unlikely, as is a deep free trade agreement, and it is utterly impossible to find a path toward a proper Customs Union.

Brexit progress

Final deadline supposedly mid-November; shallow FTA now slightly more likely than No Deal given reports of significant progress in drafting the text of a deal (though timing is short, disagreements remain over fishing rights, level playing field):

Note: As of 10/30/2020.  GDP impact over 15-year span – not all immediate and much from diminished immigration. Source: U.K. Treasury, RBC GAM

Why are we slightly more optimistic than before?

• The U.K. has demonstrated the capacity to negotiate a modern trade agreement given its recently announced deal with Japan. The EU had already demonstrated this ability via deals with Canada, Mexico and Japan.
• These new trade deals provide a helpful, modern template for both parties in their efforts to secure a trade deal.
• While significant disagreements remain, such as over fishing rights and the specific contours of what constitutes a level playing field, it seems easy enough to reach a bare-bones agreement to minimize tariffs between the two parties. Furthermore, it seems ludicrous to kill an economy-wide trade deal on the basis of such niche concerns.
• Both parties are now negotiating daily, and significant progress has reportedly been made in drafting the text of an agreement. Previously, the EU was refusing to negotiate until the fisheries matter had been resolved.

Of course, either way, some economic damage will occur, as per the table above.

U.S. election preview

The U.S. election is now upon us. Each candidate has several factors that appear to favour them. The Republican candidate Trump is the incumbent, and incumbents usually win. He also outperformed expectations in 2016, has stoked a strong stock market performance, and benefits from the steady aging of the U.S. population given that voters generally become more conservative with age.

Conversely, the Democrat candidate Biden also has several factors that favour him. The U.S. population is becoming more racially diverse over time – theoretically favouring the Democrats. Incumbents frequently lose when their popularity is low, as is the case with Trump. Furthermore, incumbent parties have reliably lost the White House over the past 70 years when a recession or 20%-plus stock market correction has occurred in the election year, as was the case this year.

But we don’t need to operate in a vacuum, weighing theoretical pros and cons. There are a multitude of polls, betting markets and models that can help us predict the election outcome.

Polls continue to put Biden well ahead of Trump, by a weighted average of 8.6 percentage points according to Fivethirtyeight. Real Clear Politics puts this at a slightly smaller 7.8 points. This is a big gap: historically, no presidential election since the 1940s has generated a sufficiently large surprise to elevate Trump to the presidency from current polling levels. However, the differential has been narrowing somewhat in recent weeks, down from a peak of 10.5 points.

Of course, presidents aren’t elected at the national level, but rather on an all-or-nothing state-by-state basis. Florida and Pennsylvania are arguably the two most critical swing states. Biden’s advantage in Florida has shrunk from 5 points to 2 points, and his advantage in Pennsylvania has fallen from 8 to 5 points. It just so happens that Trump outperformed expectations in Pennsylvania by nearly 5 points in 2016. As such, the election isn’t quite done yet.

Betting markets acknowledge the possibility of a different outcome. PredictIt gives a 62% chance of a Biden victory – roughly steady over the past week (see next chart), while Real Clear Politics assigns a 64% chance and the Good Judgment Open assigns a materially higher 87% chance.

Biden leads Trump, but gap has narrowed in recent weeks

Note: As of 10/31/2020. Based on prediction markets data and RBC GAM calculations. Source: PredictIt, RBC GAM

Election models – which scrutinize regional polls and factor in past polling errors – give Biden an even higher likelihood of victory of 90% according to Fivethirtyeight and 96% according to The Economist magazine. It should be noted that Fivethirtyeight gave Hillary Clinton a 65% chance of victory before the 2016 election. Whether the focus should be on the fact that the 2016 outcome differed from the most likely outcome or instead that Trump’s odds of winning in 2020 are 3.5 times lower than in 2016 is debatable.

For our part we continue to believe Biden has something like a 75% chance of victory, meaning Trump has a 25% chance.

Election Day complications

The very fact that betting markets believe Trump has more than a slight chance argues that markets are profoundly skeptical of the polls. To some extent, they have every right to be. Who answers their phone these days when a stranger calls, let alone to reveal personal information? The errors of 2016 are particularly fresh in the minds of many. Expectations abound for a deluge of “shy” Trump voters who have concealed their voting intentions from pollsters.

Some context is useful here:

1. A review of the 2016 election outcome reveals that a big part of the 2016 polling error was that voting intentions actually changed as people went to the polls. That could yet happen again, but there are no recent revelations that would change a large number of minds. Most Americans have already voted, and Trump is no longer the unknown novelty candidate.

2. It is not clear that there are actually many shy Trump voters today. Anonymous online polls yield very similar results to phone-based polls.

3. Now flipping in the opposite direction, there does appear to be a repeating bias in U.S. polls that underweight Republican voting intentions. This is not universally the case and pollsters have made every effort to recalibrate their polls so as to offset this inclination. But it is notable that the handful of elections for Governors and Representatives held this year have underestimated the Republican candidates by an average of 3-plus percentage points. This is quite similar to the bias that existed in 2015—2016.

4. The person who captures the popular vote does not necessarily win the election. The vast majority of U.S. states put the entire weight of their population behind whichever candidate garners more votes. Thus, a 51% advantage becomes a 100% advantage within a state. While this was long merely a historical curiosity with only one president between 1824 and 1996 winning the presidency without capturing the popular vote, that bizarre outcome has now happened twice in the last five elections. It appears that Biden will need to receive nearly 51% of the popular vote to win the election, whereas Trump barely needs more than 49%. One election model assigns a 97% chance that Biden wins the popular vote, in comparison to a 90% chance that he wins the election.

5. The turnout represents something of an X-factor. It is on track to exceed even the turnout of Kennedy versus Nixon in 1960, with 65% or more of eligible voters set to make their opinions known. Already, more Texans have cast their ballot than voted in the entire 2016 election, and Election Day is not until tomorrow as these words are being written. With many Americans voting for the first time, the outcome is harder than usual to predict.

6. Exit polls and interim estimates of the result are likely to further diminish confidence in the polls and cause confusion as to who is actually winning. A CBS News poll anticipates that 66% of early votes went to Biden (mostly mail-in ballots), whereas 69% of the votes on Election Day are likely to go to Trump. Thus, in any attempt to determine who is actually leading, it matters enormously whether the tallied votes are disproportionately those that were made early versus on Election Day.

7. There will be considerable variation by state as to which type of result is released first, and for that matter which states release their results first. Florida, for instance, is thought likely to be fairly fast in reporting comprehensive results, whereas Pennsylvania is expected to take many days, potentially drifting into the weekend. Depending on this pattern, Trump or Biden could appear to possess false leads that later evaporate upon a fuller audit of the results.

That said, we can nevertheless learn a lot from those states that report their results early. Even the most lop-sided state will impart information to the extent that the outcome incrementally exceeds or falls short of the pre-election polls. This “miss” can then be mapped onto the polling for other states in an effort to glean how the election might be evolving differently than expected.

Furthermore, some states are critical to the election prospects of the candidates (given the interconnectedness of results across states and what this would mean for a number of other states). The loss of a single such state can nearly guarantee a win for the other candidate. For instance, if Florida goes to Biden, he has a 99% chance of winning the election. It is a similar story with Pennsylvania. Conversely, if Trump wins Pennsylvania, his likelihood of winning the overall election surges from 10% to 63% given what it likely says about voter outcomes in other close states.

Post-election complications

It is possible we will know the result of the election in the early hours of the morning after the election. If Biden’s polling holds up, this could prove a one-sided affair and sufficient information will exist to know who has won. But if the results prove to be somewhat closer than this, the outcome could remain unclear for days or even weeks to come.

A PredictIt betting market indicates that there is a 60% chance that one candidate will concede within two weeks of Election Day. Of course, that means there is a 40% chance that the election result drags on for even longer than this. Recall that the 2000 election was not resolved until mid-December of that year – just under the wire to allow the Electoral College to formalize the result. This year, the Electoral College needs to be finalized by December 8.

Election officials have suggested tabulating mail-in ballots should take no more than a few days, even in those states that are not allowed to begin counting until after the final vote has been cast. However, there is always the risk of unexpected complications, as per the Iowa Democratic Party caucus that took many days to be resolved. It should be noted that the Democratic Primaries held after the pandemic lockdown took around four days on average to be tabulated, and 10 days in Georgia.

A recount is triggered if the margin of victory in a state is less than 0.5 percentage points. Fivethirtyeight estimates there is a 4% chance that the election will hinge on a recount in a particularly close state.

There is also the possibility of more insidious complications:

The Republican Party recently attempted and failed to invalidate 120,000 drive-through votes in Texas. The U.S. Supreme Court has allowed, for now, the inclusion of Pennsylvania mail-in votes received up to three days after the election despite a Republican challenge. However, the conservative wing of the court indicated that there is a “strong likelihood” that this is unconstitutional and so those votes may later be invalidated. This kind of legal maneuvering could yet add or subtract tens of thousands of votes that might prove decisive in determining who won a particularly close state. Furthermore, such legal efforts are rarely resolved in a matter of days – they take weeks and longer to sort out.

There are even (admittedly quite unlikely) scenarios in which a state legislature opts to appoint a different slate of Electoral College voters than the one put forward by the winning candidate, potentially changing the election outcome.

There is also the possibility that President Trump declares victory prematurely, before the full process of counting, recounting and potentially litigating is complete. Conversely, in the event that he is trailing, he could refuse to concede. Either of these would complicate the interpretation of the result, but not likely alter the final outcome.

Clearly, there are quite a number of ways the election could drag on. The best case outcome is that the election result is sufficiently clear that most of this wrangling proves unnecessary or is at least irrelevant.

Implications

As we have written in the past, a Biden presidency is expected to yield a modestly stronger economic performance, higher bond yields and – possibly, though more debated – a positive stock market outcome (see details in next table). The extent of any economic boost under a Biden presidency depends in significant part on whether the Democrats capture the Senate – an outcome that has a 59% chance according to PredictIt.

Biden platform relative to Trump and implications

Note: As at 10/30/2020. Source: RBC GAM

Attempts to anticipate the stock market implications based solely on the historical performance under different governing permutations are fascinating, but arguably of limited value. When considering the 2x2x2=8 different permutations of the White House, House of Representatives and Senate, some combinations have literally never happened in modern U.S. history, while others have only happened once or twice. It seems unreasonable to suggest that the stock market’s performance over a particular two-year stretch was purely the result of the politicians in office as opposed to many other swirling matters.

That said, for what it is worth, the most likely outcome of a Democrat sweep is consistent with an average annual stock market return of 8.6% – OK, but not great versus other combinations.  The second most likely outcome, a Republican Senate flanked by Democrats in the other two chambers, has literally never happened before. Finally, a Republican President, a Republican Senate and a Democrat House – the third most likely outcome – is historically consistent with a 13.4% return – quite positive (see next table).

More generally, the stock market has historically performed more strongly with Democratic presidents than Republican ones, though this is debated fiercely given the relative good/bad luck of being in office during a booming decade such as the 1990s versus during the global financial crisis (or a pandemic).

Market outcomes based on U.S. political party control

Note: R=Republican, D=Democrat
Note: Table shows average changes in metrics for calendar years from 1968 to 2019. Source: Ned David Research, RBC GAM

-With contributions from Vivien Lee and Kiki Oyerinde