Hot weekends in mid-July usually define slow news periods, even with a tight race for the U.S. presidency and the Republican convention beginning Monday, but Saturday’s assassination attempt on former President Donald Trump kicked the news cycle into overdrive. Obviously, the nation went to bed with shock and disgust at the assault and senseless loss of life, but Sunday morning broadened the focus to implications for the election and everything associated with that. Among these, of course, are the economy and capital markets.
To echo the statements of politicians, leaders of all types, and the entire world media, there is no place for violence in political discourse and it’s overwhelmingly sad that we are again faced with this reality. Frankly, we don’t have much to add to the general commentary. America is a nation suffering through a lack of consensus between large, closely balanced constituencies on binary, emotionally charged, hot button issues, and we fear that one possible consequence is horrific events such as this.
We do, though, have something to add with respect to the impact of shocks on markets. Many years ago, we read a paper that tracked short- and intermediate-term changes in the Dow Jones Industrial Average following a variety of unexpected events. These ranged from natural disasters to wars and political upheaval. We liked the format and have kept the tracking up to date, adding events that mimicked those identified as “shocks” in the original paper. Based on our own views regarding what constitutes a shock and threat to markets, over time we have generated a universe of 95 events for analysis, and we have subdivided these into a variety of different threats.
Exhibit 1 provides a summary of the impacts of these 95 individual events separated into six specific threat categories, and then separately for the thankfully small set related to threats on a president’s life (or, in the current case, a former and possibly future President), or existential challenges to their administration (ie: Watergate).
Exhibit 1: The market through crisis – median experience for the U.S. stock market
While one can quibble with the events that we added and certainly those that we didn’t, the data is interesting and, we think, quite useful. The range of outcomes is wide, but some general conclusions can be drawn. First, on median, shock periods are usually contained. The U.S. equity market tends to react to shocks negatively, falling a median of 4% over the first nine days and then typically recovering the full amount of that loss another seven days following the shock. Another interesting insight into the full set of data is that for those shocks that occurred either during or prior to a recession taking hold, damage to markets proved much more durable than for periods when the economy remained in a growing trend.
Looking only at the small dataset for shocks related to the U.S. presidency, the median decline is just 1.3%, lasting only one day with full recovery in the index a day and a half following that (Exhibit 2). Perhaps surprisingly, this dataset shows the most limited impact from shocks in terms of both duration and damage. Only one event dealt a double digit blow to the Index and in two of the eight events listed, the market’s rise was uninterrupted.
Exhibit 2: The market through crisis – presidential shocks
As with all shocks, last night’s assassination attempt is a reason to revisit our outlook for the economy and markets. The most difficult part of any assessment is understanding changes in investor confidence, and it’s possible that despite the current fractiousness having developed over many years, it could be a tipping point issue. Or, with valuations in the small set of spectacular market leaders so high, a larger than normal and perhaps overdue correction could be triggered. But for investors looking beyond the very short term, the real threat is the degree to which this shock finds pathways to altering the prior consensus for growth and inflation. The past data related to presidential events suggests that’s not the most likely course.