According to BofA's derivatives strategists, if 90 years of history of US elections is to be believed, there is scant evidence that 2020 should be more volatile than 2019, as the S&P tends to be similarly volatile in election years as in the year prior. In fact, excluding 2008 when the spike in volatility had little to do with the election cycle, volatility tends to be slightly lower during election years. And yet, in light of last week's political developments, 2020 could well be an exception to the rule.
As shown in the right-hand chart above, the recent increase in President Trump's impeachment odds coincided with the simultaneous rise and fall of Senator Warren and former Vice President Biden in the market-implied probabilities for the 2020 Democratic Presidential nomination. With Biden widely regarded as a more moderate liberal/market-friendly candidate than Warren, his potential exclusion from the Presidential race could make 2020 one of the most polarized election years in modern history should Warren and Trump emerge as the two front-runners, according to BofA.
As such, the bank warns that "investors can't ignore election risk any longer."
While the investment implications of such a polarized scenario are manifold, the most straightforward of which is higher volatility, the uniqueness of this scenario also implies there is not a readily available trading playbook investors can rely upon. Below, BofA explore some of its preferred investment ideas to trade 2020 election risk.
But first things first: How much election-related risk is priced in?
Before jumping to the bank's election trading playbook, it is key to first understand how much election risk is already priced into markets. Anecdotally, the bank's strategists note that they are starting to see interest from clients in hedging both the election and its run-up, with a particular focus on the early rounds of the Democratic Party's primaries.
Iowa and New Hampshire will be the first states to vote, on February 3rd and 11th, followed by Nevada (22-Feb), South Carolina (29-Feb), and finally Super Tuesday (3-Mar, when 14 states will cast their ballots). Historically strong predictors of the eventual outcome, these early caucuses and primaries may cause the strongest market reaction if there remains significant uncertainty about who will win the Democratic nomination. Hence, any evidence of equity protection buying will be most clearly observed in Feb-Mar (S&P) option expiries. Purely from a political perspective, the risks to potentially higher vol could include the resurgence of a more moderate candidate like Biden or dominant polling numbers by Trump against the leading Democratic candidates.
The logical first place to check for election-related risks is the VIX futures market. To match the timing of Warren's recent move higher in the polls (chart below, left), BofA compared the level and shape of the VIX futures curve on 12-Sep and 30-Sep. While the entire curve shifted higher in these last 2.5 weeks, there is so far no evidence of additional premium in Jan or Feb futures (which capture S&P implied vol in Feb & Mar, peak primaries season - chart below, left). The chart below, right confirms this numerically: the cost of a "futures condor" which sells the Dec-19 and Mar-20 futures and buys Jan-20 and Feb-20 has not increased with Warren's odds. Selling the Jan-20 and Mar-20 futures to buy 2x Feb-20 futures has not gotten more expensive either.
Yet if equity markets are not yet concerned about the Democratic primaries, is this also true of the actual election? As BofA responds, unlike VIX futures (which are only listed through Jun-20 today), "S&P-based measures of equity vol allow us to look beyond the primaries to the election on November 3rd. We find that an election risk premium has emerged in the last 3 months, evident by the kink in the Dec-20 point on the S&P variance swap term structure (chart below, left)." The chart on the right plots the hypothetical cost of selling Sep-20 and Mar-21 vs. buying 2x Dec-20 S&P variance to help extract an election premium. While the size of the dislocation varies over time, we appear to have entered a new regime in which owning Dec-20 S&P vol requires a larger premium compared to owning Sep-20 and Mar-21.
Getting to the big point, should S&P 500 options markets grow more concerned about the implications of a strong Elizabeth Warren showing in the Feb/Mar-20 caucuses and primaries, strategies that sell SPX straddles expiring before the "catalyst" to fund same-strike straddles expiring afterwards stand to profit. This was the case heading into Brexit, the 2016 US presidential election, and the 2017 French elections, when such long-short straddle pairs benefited from an expansion in "event risk premium". For example, the first chart below shows the hypothetical P&L of a short SPX 4-Nov-16 2100 straddle vs. a long SPX 11-Nov-16 2100 straddle (the US election was on 8-Nov), which achieved a gross payout ratio of ~4.5-to-1. BofA would look to deploy similar strategies once S&P weekly options spanning the caucuses/primaries become listed. Owning the VIX futures fly/condor shown in the next chart below is an alternative way to benefit from a potential rise in primaries event risk premium with limited risk, though likely with less asymmetry as the VIX futures curve tends to dislocate less than the term structure of S&P weekly options in our experience.
Focus on the casualties
According to BofA, financials and health care companies will likely be the biggest "president Warren" casualties. While Warren is broadly regarded as being a less market friendly candidate than President Trump has been, Financials and Health Care are perceived to be the two sectors that stand to lose the most if Warren were to be elected, at least based on her track record and rhetoric. In particular, Financials are likely to weaken on the risk of an increase in regulation ("The real cause of the crash was not some inevitable cycle; this crash was the direct consequence of years of deliberate deregulation...", Apr 2014 - Elizabeth Warren). Arguably, Financials would also be hurt by falling yields if markets were spooked and a textbook flight-to-safety type of sell-off played out.
The Health Care sector is also at risk given Warren's push for Medicare-for-All ("I spent a big chunk of my life studying why families go broke. One of the number-one reasons is the cost of health care, medical bills. [..] Medicare for all solves that problem", Jun 2019 - Elizabeth Warren).
In terms of what is priced-in for the Democratic primaries, it is clear at the sector level that Health Care and Financials are alert to the risks, with for instance Mar-20 expiry vol showing little-to-no change since August month-end vs. a general decrease in all other expiries' vols (see below charts). Indeed, the Mar-20 minus Jan-20 ATMf implied vol spread has been on a firm uptrend over the past month for both Financials (XLF) and Health Care (XLV).
That said, since the volatility of both sectors is historically elevated, rather than recommending buying outright puts, BofA prefers financing downside protection by selling SPY puts. Here, relative value opportunities are particularly attractive between XLF and SPY, given the implied vol spread is depressed both historically and vs. trailing realized vol (the 3m realized vol ratio of XLF vs. SPY is 1.22 vs. the 6m ATM implied vol ratio of 1.13).
As an example, buying $0.85 notional of the XLF Mar-20 put (ref. 28.02, 47d) and selling $1 notional of the SPY Mar-20 put (ref. 296.98, 45d) is roughly zero cost upfront. The theoretical expiry P&L shows a highly asymmetric and favorable historical risk reward profile. For instance, the P&L was positive one-quarter of the time since Jun 09, and negative only 3% of the time. In addition, positive returns were on average 4.1%, 3.6x as large as negative returns. Importantly, the max gain of 11.9% was 4x as large as the max loss (-2.9%). Finally, while the P&L suffered some losses at the start of the 2011 and 2015 sell-offs, the trade ultimately delivered positive returns that outstripped the initial losses.