The model by Michael Lewis-Beck and Charles Tien is a classic political economy model, which predicts the two-party popular vote from two variables that measure political popularity and economic growth.
For the 2016 election, the model predicts a clear victory for Joe Biden with 56.7% of the major party vote (compared to 43.3% for Donald Trump).
The regression model’s vote equation, estimated on data back to 1948, reads as:
V = 37.5 + 0.26
popularity + 1.18
|Table 1: Overview of variables used in the political economy model|
||Gallup Presidential Approval measure in July of the election year|
||GNP growth in the first two quarters of the election year|
|V||Incumbent share of the two-party presidential vote|
Lewis-Beck, M. S. & Tien, C. (2016). The political economy model: 2016 US election forecasts. PS: Political Science & Politics, 49(4), 661-663.