Incumbent US presidents tend to win elections except during recessions

Published on30 NOV 2023

As the race for the 2024 US presidential election heats up, history suggests economic metrics can help predict the most likely outcome, according to Goldman Sachs Research.

“This far in advance of an election, we have previously found that ‘fundamentals’ outweigh polling in predicting outcomes,” Goldman Sachs Research Chief US Political Economist Alec Phillips and economist Tim Krupa write. Polls a year before an election are not particularly reliable, with an average absolute error of nearly 10 percentage points since 1948.

First-term incumbency typically provides an advantage — unless there’s a recession during or just before the election. When there is no recession, the incumbent has always won in the post-World War II era. Goldman Sachs Research estimates a 15% probability of a recession over the next 12 months (equal to the average historical probability).

Since 1951, when the constitutional amendment was ratified to limit presidents to two terms, the incumbent has lost when the election took place soon after a recession (in 1976, 1980, 1992, and 2020). The party in the White House also lost after a recession in two instances when the incumbent candidate was not on the ballot (1960 and 2008).

What the economy tells us about elections

The real economy tends to offer better signals than the financial markets about how elections will pan out. Broad economic indicators including income, employment, GDP growth, and consumption matter more than market measures such as equity prices, Phillips and Krupa write.

The absolute level of an economic variable usually doesn’t reveal as much as its change, and economic data late in the year prior to the election and early in the election year have the strongest relationship with the ultimate vote. Our economists find that some economic variables, including real consumption and real disposable income, are more predictive over a longer horizon. But the strongest statistical relationship with election outcomes is often with variables measured in the second quarter of an election year. 

Inflation appears to be less predictive of election results than indicators related to growth or the state of the labor market. Headline inflation, in a sample going back to 1952, is weakly statistically significant as a predictor of election results, Phillips and Krupa write. Elections during periods of high inflation show a stronger relationship, but even in these elections, the signals provided by growth and labor variables are stronger.

Core inflation has an even weaker relationship with election results than headline inflation. “This is intuitive, as core inflation excludes items with the greatest salience for consumers in forming inflation expectations,” such as gasoline and groceries.

What elections tell us about the economy

Election outcomes can also influence economic policy.

Whoever wins in 2024 will face an unusually large budget deficit. And, in general, a larger deficit at the start of a presidential term tends to lead to greater fiscal restraint. This has been true on average over the last 60 years, though fiscal policy will depend at least as much on economic circumstances as election outcomes, with Congress apt to approve some degree of fiscal support in a recession regardless of party control.

On average, fiscal policy grows more expansive ahead of elections. And winning incumbents tend to tighten policy somewhat following the results. This pattern has partly been formed by a few episodes in which the government pursued a large fiscal stimulus in response to an economic downturn. The median change shows little fiscal expansion going into an election, though it still shows a fiscal contraction when an incumbent starts a second term.

When a first-term president’s party also controls Congress, a loosening of policy is more likely early in the term. In divided governments, fiscal policy has generally tightened after an election. “Control of Congress impacts fiscal policy at least as much as control of the White House,” the authors write.

This article is being provided for educational purposes only. The information contained in this article does not constitute a recommendation from any Goldman Sachs entity to the recipient, and Goldman Sachs is not providing any financial, economic, legal, investment, accounting, or tax advice through this article or to its recipient. Neither Goldman Sachs nor any of its affiliates makes any representation or warranty, express or implied, as to the accuracy or completeness of the statements or any information contained in this article and any liability therefore (including in respect of direct, indirect, or consequential loss or damage) is expressly disclaimed.

Explore More Insights