Assessing the Election Odds – Part II
By Adam Thurgood on October 15, 2024
The end is near! No, not the end of the U.S. as we know it, but the end of the election cycle. We are just three weeks away from the end of attack-adds dominating television, robo calls interrupting dinner, and having to respond STOP to countless political text messages. While readers will no doubt have their own opinions on what may transpire if either Harris or Trump win the White House, at least we have bi-partisan agreement that election seasons are tiring, frustrating, and best when complete.
A lot has happened since I wrote the first piece on election odds back in June. Given the high level of political anxiety and general tension prevalent across the country, my goal here is not to stoke those fires by weighing in on which candidate is the better choice. Instead, this post is meant to provide a broad view of the election odds using polling data, betting market data, and market-based election indicators so each reader can arrive at their own conclusion of who will win in November. Let’s start with national polling.
The national polling data tells a story of a very close race. The latest NBC News poll, released October 13th, shows a deadlocked race with both Trump and Harris polling at 48% with Trump gaining momentum over the past two months. Other national polls, such as Real Clear Polling, have Harris with a slight edge, while FiveThirtyEight gives Harris a 70% chance of winning the popular vote.
While national polls are informative, the White House is not decided on the popular vote. The electoral college is all that really matters in the end. On this front, the race is much less certain due to the capricious voting nature of a few swing states. The results in Arizona, Georgia, Nevada, Michigan, Wisconsin, and Pennsylvania will determine who leads our country for the next four years. As you can see in the chart below, Trump held a commanding lead in the swing states until Biden dropped out. The polls then swung significantly in Harris’ favor, but Trump has since clawed his way back to even across the board, with small leads in Arizona, Georgia, Michigan, and Pennsylvania.
Real Clear Polling (RCP) has a great interactive electoral college mapping tool that you can use to see how these swing states might impact the outcome of the election. RCP believes 104 electoral college votes are up for grabs in toss-up swing states, with Harris a lock for 215 votes in blue states and Trump a lock for 219 votes in red states. RCP’s no toss-up model sees Trump winning the election with 302 votes, which would be a fairly dominant win. However, FiveThirtyEight’s electoral college simulation results show Harris winning 54 times out of 100 to Trump’s 46.
Moving away from polling data and to the betting markets, the results skew in Trump’s favor. Trump holds a commanding lead of ten points in the Polymarket betting market, while his lead in the PredictIt market is four points. The odds generated from the betting markets are interesting because people are putting their hard-earned funds on the line to the tune of hundreds of millions of dollars. Only time will tell if the bettors get it right, but at this point they collectively believe Trump will come out on top.
In addition to the betting markets, we also follow several financial market indicators of the election that have historically done a good job predicting election results. One of the easiest and most accurate election indicators is simply the return of the S&P 500 from August 1st-October 31st. When the market is positive during this stretch in an election year, the odds favor the incumbent party to keep the White House, and vice versa. Since 1972, the indicator has correctly predicted 12 out of 13 elections (92% accuracy).
Source: Bloomberg as of 10/14/2024, average data since 1972
To this point, the S&P 500 indicator is closely tracking the average performance of years when the incumbent party has won. The indicator is not fool proof, as the incumbent party has lost three elections since 1932 when the S&P 500 indicator was positive: 1932, 1968, and 1980. That being said, this is certainly positive for the Harris campaign.
Another great indicator comes in the form of the Misery index, which is the combination of the unemployment rate and inflation rate. Generally speaking, a rising Misery index is bad news for politicians. At the end of October in an election year, the incumbent party tends to lose when the Misery index is up year-over-year, and vice-versa. In fact, the indicator has predicted 14 of the last 15 Presidential election winners and has predicted every election correctly since 1980. Currently the Misery index stands at 6.5%, which is a full percentage point below the breakeven level from October of 2023. As such, the Misery index will almost certainly point to Kamala Harris winning the presidency.
With all these mixed signals, it is difficult to say with any level of certainty who will come out victorious next month. The margins are just too narrow. Trump supporters can lean on the betting markets and certain electoral college maps for comfort, while the Harris camp can take solace in the market-based indicators and national polling data. However, I would be shocked if either campaign was sleeping soundly.
While the uncertainty of which party will control the White House is high, the odds of either a Democratic or Republican sweep are fairly low. The odds of a Republican sweep currently stand at 38%, while the Democrats odds of a sweep are just 16%. With split power in Washington, major policy initiatives that skew too far left or right are likely dead-on arrival. This should provide some comfort for moderates on both sides of the aisle.
Now for the not so good news. As part of our research budget, we have access to two different DC policy research firms that help us understand the policy objectives of each party and their impact on markets and the economy. Unfortunately, both firms believe it could be 2-3 weeks after the election before we have final clarity on the winner, so while we will get relief from the nuisances that come with election season, we will likely have to deal with the electoral anxiety and political tension longer than normal this time around.
