How the economy — and Philadelphia — could determine the presidential election

The race for the presidency is in full swing, and the state of Pennsylvania — and more specifically Philadelphia — may well decide who wins.

This forecast is based on a model of presidential elections that my colleagues and I at Moody’s Analytics have been using to predict who will be the next president in elections going back to Bush vs. Gore in 2000. The model predicts whether the incumbent presidential candidate, Joe Biden in this election, will win the popular vote in each state and the District of Columbia, and thus the necessary Electoral College votes to win the election.

Given that we are economists, it is no surprise that the principle underpinning our model is that the economy may not be at the top of voters’ minds in every election, but it is rarely less than a close second. In other words: “It’s the economy, stupid.”

However, while the economy’s performance and perceptions of this performance are critical to deciphering the behavior of undecided voters, making them crucial to predicting close elections, political variables are also potent in predicting state-by-state voting.

The most important political variable is how states voted in previous elections. Wyoming, for example, invariably votes for the Republican nominee, and as such is expected to vote for former President Donald Trump this go-round. Rhode Island always supports the Democrat and will go to Biden.

Voter turnout is another critical variable in determining election outcomes, and even modest shifts in turnout can flip an election. The 2020 presidential election saw the second-highest turnout since 1936, with well over two-thirds of the voting-eligible population casting votes. For the upcoming election we assume that voter turnout will be the same as in 2020.

Third-party candidates can also swing elections, and that is a real possibility this year. Well-known independent contender Robert F. Kennedy Jr. is actively trying to get on state ballots, while No Labels, a self-identified bipartisan organization, appears set to put its own candidate forward.

As for the economic factors, particularly important is the growth in real household incomes, a measure of voters’ purchasing power. To Biden’s benefit, after a bout of high inflation, after-inflation wages are growing again and by Election Day should be well above pre-pandemic levels in most states. Unemployment is low and stable from coast to coast and is supporting strong wage gains.

Gasoline prices play an outsized role in shaping voters’ perceptions of inflation and their financial well-being. Most Americans purchase gas regularly and are well aware of the price and how it is changing.

While sanctions on Russian oil and OPEC+ production cuts pushed gas prices to a historic high over $5 per gallon, they are currently back near $3.50, about where we expect them to be on Election Day. Having said this, forecasting oil prices is dicey. If prices move up over $4 again, the damage to Biden’s reelection bid will likely be insurmountable.

The fixed mortgage rate is another important factor in our election model and may play an even bigger role in this election given how quickly it has risen since Biden became president. Combined with the surge in home prices since the pandemic hit, housing affordability has been hammered. Many view the purchase of a first home in emotional terms as an anchor in a good community, a path to building wealth, and often a rite of passage into the middle class. Seeing homeownership drift out of reach can tap deeper feelings of economic insecurity and frustration.

While fixed mortgage rates remain elevated at just under 7%, they are down considerably from their peak closer to 8% late last year, and we expect them to move down a bit more by Election Day, a modest plus for Biden’s reelection chances.

Of course, the election model results hinge on our forecast for the strength of the economy between now and Election Day, and various political assumptions regarding voter turnout and the prevalence of third-party candidates. But while the election will almost certainly be a nail-biter, we feel confident in the model’s 2024 prediction for who will be the next president. That is, Biden.

More precisely, Biden is expected to win 308 electoral votes, 38 more votes than the 270 needed to win reelection. This is nearly identical to his tally in 2020, when he won 306 votes. Like 2020′s close and contentious race, the 2024 election will be determined in a few battleground states. Five states will be decided by less than one percentage point. Those states are Arizona, Georgia, Nevada, North Carolina, and, yes, Pennsylvania.

Biden’s projected 308 electoral vote tally provides some cushion. If we start flipping the results of his slimmest victories, the loss of North Carolina and Nevada would trim his vote total to 286, still enough to achieve victory. Losing Georgia, which has 16 electoral votes, would then bring Biden to the exact threshold he needs to win a second term. Pennsylvania is thus the key to the national election, and if history is a guide, Philadelphia will be key to how the state votes.

Of course, this won’t be lost on Biden and Trump, so count on lots of political ads and rallies in coming weeks.