Markets: Remember, Remember the Third of November

National Review Online, September 4, 2020

Writing in the Financial Times on September 1, Robin Wrigglesworth reported that markets are signaling unease about what may lie ahead in the first week of November. It is not so much the election that’s causing agita as the fear that Election Night might not resolve the result. Investors do not appreciate uncertainty, and if everything is still unresolved by, say, late the next day, the only certainty will be uncertainty.

On Bloomberg, Ryan Teague Beckwith has invited readers to “choose your own election adventure”:

So . . . how badly could this election go? In our choose-your-own-election game, see if you can avoid landing in a Supreme Court showdown or constitutional crisis and keep democracy safe (for now).

To be blunt, given the way that both the Trump and Biden camps (and parts of the media) are already throwing shade on the election process (not to speak of the sour aftermath of the 2016 election, or, for those with longer memories, the Diebold paranoia of yesteryear, or, for that matter the halcyon days when an election hung on hanging chads), the nervousness, if anything, may be underdone, especially as there are some signs that the race may be tightening, thereby reducing still further the chances of a clear-cut outcome not too long after polls close.

The president’s dark ruminations about the election — many of them centered on the iniquities of mail-in voting — are too well known to need repetition, as is the use he makes of conspiracy theories. But comments from the Biden camp also seem aimed at preparing a claim that they were robbed.

To take one example, CNN recently quoted Hillary Clinton as saying this:

“Joe Biden should not concede under any circumstances because I think this is going to drag out, and eventually I do believe he will win if we don’t give an inch and if we are as focused and relentless as the other side is. . . .”

Then there was Biden in June:

“My greatest concern, my single greatest concern [is that] this president is going to try to steal this election.”

And the Washington Post cannot be said to have been calming things down by printing a piece by Rosa Brooks describing how things might play out under various simulations prepared by the “Transition Integrity Project,” an exercise that might not have been entirely even-handed:

In each scenario, Team Trump — the players assigned to simulate the Trump campaign and its elected and appointed allies — was ruthless and unconstrained right out of the gate, and Team Biden struggled to get out of reaction mode. In one exercise, for instance, Team Trump’s repeated allegations of fraudulent mail-in ballots led National Guard troops to destroy thousands of ballots in Democratic-leaning ZIP codes, to applause on social media from Trump supporters. Over and over, Team Biden urged calm, national unity and a fair vote count, while Team Trump issued barely disguised calls for violence and intimidation against ballot-counting officials and Biden electors.

But don’t worry, counseled Brooks, steps (some of them certainly sensible) can be taken to defuse the atmosphere ahead of the vote. “The media also has an important role” to play.

Oh.

In a rather more measured article for the New York Times, Trip Daniel put, so to speak, flesh on the specters, focusing first on the likelihood that many more Biden than Trump voters will vote by mail:

An NBC/Wall Street Journal national poll last month found that nearly half — 47 percent — of supporters of Mr. Biden planned to mail in their votes, compared with two-thirds of Trump supporters — 66 percent — who planned to vote in person on Election Day.

In some states, the discrepancy is even more stark. A recent Marquette Law School poll of Wisconsin, another swing state, found that among voters planning to cast a mail ballot, Mr. Biden was favored by 67 percentage points. Among those who planned to vote on Election Day, Mr. Trump led by 41 points.

And so:

“We are sounding an alarm and saying that this is a very real possibility, that the data is going to show on election night an incredible victory for Donald Trump,” Josh Mendelsohn, the chief executive of the group, Hawkfish [a data-analysis company founded by Michael Bloomberg “to inform progressive initiatives”], told Axios on HBO. The company’s survey of registered voters concluded that twice as many planned to cast a ballot by mail as ever before, and that they were mostly Biden supporters.

A spokeswoman for the Trump campaign, Thea McDonald, called Democrats’ concerns about the president prematurely declaring victory “an unsubstantiated conspiracy theory,” adding, “President Trump and his campaign are fighting for a free, fair, transparent election in which every valid ballot counts — once.”

“Once.”

Trump did not help himself by later saying in North Carolina that those who had voted by post should “go and make sure it counted, and if doesn’t tabulate, you vote, you just vote. And then if they tabulate it very late, which they shouldn’t be doing, they’ll see you voted, and so it won’t count.” There has since been some clarification from the White House, but intentionally voting twice, it should be stressed, is illegal. What’s more, according to FiveThirtyEight’s summary of the voting rules in each state, in North Carolina, absentee ballots must be postmarked (or dropped off in person) by November 3 and received by November 6. If that’s the case and if someone had mailed their absentee ballot too late to arrive on the 3rd, it would be impossible for it to be “tabulated” in the way the president would have liked. At the very least, Trump has added yet more potential confusion to a day that will need the opposite.

But back to the Times:

A claim of victory on election night by Mr. Trump, before results are certified by officials, would have no legal effect, Dr. Hasen [an election specialist from the University of California] said. “That said,” he added, “it could have a great political effect and convince his most ardent supporters that he has won the election and any changes in the counts are due to fraud. That’s really a huge concern.”

Adding to the chance that the result may well “change” after any initial declaration is the fact that currently twelve states do not allow mail-in ballots to be processed (or, thus, I presume, be “tabulated”) before Election Day, including Michigan, New Hampshire, Pennsylvania and Wisconsin, four states that may be crucial in November.

And that is not the end of it. In some states the cut-off period by which mail-in ballots must be received is not until quite a while after Election Day. The deadline for receipt is generally November 3, but not always. As we have seen, there’s North Carolina’s November 6, but, according to FiveThirtyEight, there are other, even longer, permissible delays. In Alaska, for example, absentee ballots must be postmarked (or dropped off in person) by November 3 and received by November 13. In California, ballots must be postmarked (or dropped off in person) by November 3 and received by November 20. Minnesota ballots must be postmarked (or dropped off in person) by November 3 and received by November 10.

Even when they are received, a strikingly high percentage of mail-in ballots end up being disqualified. About a fifth of all mail-in ballots were ruled out in New York’s July primary.

The Economist recently looked at the prospects for mail-in ballots in November (my emphasis added):

For the likelihood of voting by mail, we rely on a Pew Research Centre poll of voters between July 27th and August 2nd. According to this survey, 39% of voters are planning to cast postal votes this year. A 2018 canvass of state data from the United States Election Assistance Commission, which helps states with administrative issues related to voting, revealed that the average rejection rate for postal votes was about 4% in states with the least vote-by-mail infrastructure and experience, and 1% in the best-prepared—though rates in extreme cases can exceed 15%. . . .

Although such a scenario is unlikely, even small glitches with postal voting cause a sizeable increase in the probability of a recount in one of the decisive states. Failure of election workers to attach postmarks to ballots, voters assigning signatures that do not match the state’s rolls (typically because elderly voters have trouble with the pen, or young voters have no record against which to match), and failure to deposit a ballot before the deadline are all common errors that can lead postal votes to be rejected—and they only become more common when more people vote by mail. According to our modelling, the chance that a marginal state ends up in recount territory (typically when the margin between the candidates is less than half a percentage point) is around 5%, before factoring in trouble with vote-by-mail. It nearly doubles to 9% when these postal simulations are taken into account.

It thus seems reasonable to assume that many mail-in ballots will not pass muster, raising the prospect of sustained legal challenges across the country, challenges that will take time to resolve.

Now throw in the current civil unrest, and it becomes all too easy to paint a bleak picture of what November (and its aftermath) might look like if the election is at all close, or even, once all the ballots have been tallied and litigated, it goes the “wrong” way, particularly, I would guess, if Trump emerges (or looks like he is emerging) as the victor.

Ronald Brownstein, in The Atlantic in July:

If Trump wins a second term—especially if that victory relies on another rural surge to overcome massive opposition across the big metros—the chaos in Portland might look like only the preliminary skirmish for an even more incendiary collision to come.

Over-excited talk? Maybe.

Meanwhile, Adbusters, a (Canadian!) organization that can claim some credit — if that’s the word — for launching Occupy Wall Street, appears to be planning a “siege” of the White House:

It’s been nine years since we set off the political earthquake of #OccupyWallStreet, laying siege to NYC’s Zuccotti Park and inspiring thousands of similar protests around the world. . . .

It’s time again for dramatic, decisive action. Which is why, on September 17th, in the original and enduring spirit of Occupy, we and tens of thousands of our fellow citizens will stream into Lafayette Square, in Washington, D.C.

We will lay siege to the White House. And we will sustain it for exactly fifty days. This is the #WhiteHouseSiege. … Fifty days — September 17th to November 3rd.

Over-excited talk? Maybe.

Under the circumstances, it was not a surprise to read this in Bloomberg last week(my emphasis added):

One measure of hedging in the stock market is higher than at any point in the past three presidential elections. In the interest-rates market, implied volatility is well above levels reached in 2016 or 2012. And three-month implied volatility in the dollar-yen pair — a classic haven trade — has risen above the two-month tenor by the most in two decades, signaling demand for protection from turbulence near Election Day. . . .

One area that highlights how much traders are bracing for election drama is swaptions, or options on interest rate swaps, which typically mirror the overall direction of Treasury yields. Wells Fargo analysts found that Election Day is priced for four times the volatility of a “typical” trading day in this market, based on the difference between at-the-money volatility on two-month and three-month swaptions on 10-year interest rate swap contracts. That compares to roughly two times at a similar point in the 2016 election cycle and about 1.7 times during the 2012 campaign, they wrote in a note.

The Financial Times’ Wrigglesworth has more:

Futures contracts on the Vix volatility index — which uses options to derive an estimate of the future choppiness of the S&P 500 stocks benchmark — show what some traders call a “kink” around October and November, in a pick-up that exceeds previous contests for the White House.

Monthly futures contracts on the Vix index tend to slope gently upwards, as investors normally have to pay more to insure themselves against longer-term volatility. Yet the contracts maturing at the end of October and November are unusually expensive, trading at about 33 and 32 respectively.

Those premiums are considerably higher than they were at the same time in the run-up to the 2016 contest between Donald Trump and Hillary Clinton.

Four years ago, the spot Vix index was just under 14, while the October-November contracts were trading at 17 to 18 points. In the 2012 election the contracts traded at a 5 to 6 point premium, and in 2008 there was just a 3 point premium, despite the looming financial crisis.

Wrigglesworth also cited a portfolio manager at Capstone, a hedge fund, who noted that the implied volatility for the end of the year was still higher than normal, indicating that traders were positioning for the risk of turmoil after the poll:

“What is interesting isn’t the election peak but the fact that it doesn’t come down quickly after the election,” he said. “The market is saying that we might not get a clean election.”

In 2016, MarketWatch took a look at what happened in 2002:

The Dow fell more than 5% over the first two weeks following Election Day; this came as a legal battle escalated in Florida. . . . In late November . . . Florida’s election board declared Bush the victor. And though Gore immediately contested that ruling, the stock market did begin to rally. Even so, on Dec. 12, when the legal battles were finally ended by the Supreme Court’s decision, the Dow was 3.1% lower than where it stood on Election Day.

2020 is not 2000. Those six weeks in late 2000 might look like an era of tranquility compared with what may lie ahead.

The above is an expanded version of commentary published on National Review’s Capital Matters’ Capital Note on August 31.