Traditionally, the end of the second convention marked the start of the presidential election season. Candidates did not campaign full stop, and therefore the public did not pay close attention until after both parties had introduced their nominees to the country. In recent decades, the convention demarcation line has been blurred by multi-year campaigns, 24-hour cable news channels, and social media. Decades have passed since candidates were selected during the convention. The party platform has evolved from an open debate to a pre-canned list of the nominee’s stances. Conventions have morphed into a four-day coronation of the candidate. Perhaps as a result, conventions have started later. In the 16 conventions from 1900 to 1960, the second convention ended a median of 17 weeks before the election, or in early July. Since 1964, conventions have moved six weeks later, into mid-August.

Historically, the market has rallied between the end of the second convention and the day before the election. Since 1900, the DJIA has gained 52% of the time by a median of 0.8% (table, above). There is a noticeable spread between when the incumbent party has eventually won or lost the election. The DJIA has risen 72% of the time by a median of 5.8% when the incumbent party has retained the presidency versus 23% and -1.5% when the incumbent party has lost it.

Changing Landscape?

In recent election cycles, the post-convention bounce has been non-existent, regardless of whether the incumbent party has retained residency of the White House. The DJIA has fallen each of the last five election years between the second convention and Election Day, including in 2004 and 2012, when George W. Bush and Barack Obama won reelection (table, below). Extending the study to the previous 12 elections (1976), the DJIA has risen only three times and posted a median loss of 1.4%. Median returns were slightly worse when the incumbent party lost (-1.8%) than when it won (-0.3%).

A partial explanation is that conventions have started later, removing the seasonally strong early-July to mid-August window while keeping the seasonally weak September-October period. The fact that presidential elections have dominated the news cycle for the entire election year has likely played a role as well. This year has a unique situation with the Democrats replacing their candidate so late in the cycle. Joe Biden dropped out on July 21, about when the second convention took place before 1964. Ironically, the incumbent party nominee is less known than the challenger. Voters could still blame Harris for any economic weakness before the election and vice versa. Perhaps the second convention-to-Election Day study is relevant again.