Economy vs. Mood

Douglas Hibbs · Economy line

Bread and Peace gives Republicans 47.9% of the 2028 popular vote

That is a 33% chance of winning it. In the backtest, this model’s typical miss was ±4.7 points, so 80% of outcomes would land between 41.7% and 54.1%.

Missing 2027–28 data:Election todayFed projections
Republican share: 47.9%, with 50% and 80% ranges
Daily tracking began October 14, 2026. The bands narrow as 2028 data arrive.

Two-party popular vote, not the Electoral College. If the election were today: the latest published quarters and today’s approval stand in for 2028. Source: FRED/ALFRED (BEA, BLS, Census), FOMC projections; model fits by this site. As of October 17, 2026.

Show the numbers
Bread and Peace: Republican two-party share and chance of winning the popular vote
DateShare80% rangeChance
October 14, 202647.9%41.7%–54.1%33.1%
October 15, 202647.9%41.7%–54.1%33.1%
October 16, 202647.9%41.7%–54.1%33.1%
October 17, 202647.9%41.7%–54.1%33.1%
Embed this chart

What pushes it up or down

Real income growth per person moves this forecast most (−4.3 pts)

Two-party popular vote, not the Electoral College. Source: FRED/ALFRED (BEA, BLS, Census), FOMC projections; model fits by this site. As of October 17, 2026.

Show the numbers
Push on the Republican two-party share, percentage points
InputPoints
Typical party-in-power edge+1.6
Real income growth per person−4.3
US war deaths+0.6

Today’s inputs against 1948–2024

Where today sits in the historical range
Grey dots are the 19 or 20 past elections; the black dot is today.

If the election were today: the latest published quarters and today’s approval stand in for 2028. Source: FRED/ALFRED (BEA, BLS, Census), FOMC projections; model fits by this site. As of October 17, 2026.

Show the numbers
YearWeighted real income growth
Today−0.8
19483.6
19523.0
19562.0
19600.5
19645.7
19681.3
19723.7
19761.3
1980−2.1
19844.1
19882.5
1992−0.7
19961.3
20000.8
20040.5
2008−2.7
20120.5
20161.4
20202.0
20241.6

How it works

Douglas Hibbs’s model uses just two things: how fast real disposable income per person grew over the term, and US military deaths in wars the party in power started.

Recent quarters count most. How fast older quarters fade is estimated from the data (λ below).

Income is deflated by the consumer price index, as in Hibbs’s papers. The election quarter itself isn’t published before election day, so the latest quarter used is July–September.

Bread and Peace uses income growth through 2026 Q2, the latest published quarter weighted most.

The fitted equation

Version 1, fitted October 13, 2026 on 1948–2024 (ordinary least squares)
TermCoefficient
Constant49.378
Weighted real income growth1.903
War deaths (thousands)−0.221
Lag weight λ0.580

Its record

Predicting each election with a version fitted without it, the model’s typical error is ±4.7 points and it called the popular-vote winner in 15 of 20 elections. It missed 1952, 1960, 1976, 2020, 2024. See every election.