Economics Trumps polling

One of the big news stories today is how the national polls got the results of the presidential election so wrong.

Part of their problem was the demonization of Trump by the media. When media bias turns a policy or person into the instantiation of evil as they did with Trump, those who favor that policy or candidate will not want to side with that “evil” when answering poll questions. So they tell pollsters what they think people want to hear.

One older lady told me she planned to vote for Trump but asked me not to tell anyone. I read about a rabid Clinton supporter who wrote that his mother had promised him she wouldn’t vote in this election because she didn’t like either candidate. But when he wasn’t looking she slipped out and voted for Trump. The media shot themselves in the foot on this one.

Back in July I wrote a post on economic models that were Trumpeting a win for the Republican candidate or modestly admitting the outcome was too close to call. The accuracy of those models showed that people consider the economy their most important issue.

Trump wins election, not that it matters

For someone who despises politicians as much as I do it’s really hard to avoid political news lately. But there is a nexus between the presidential election and economics: most voters think the president controls the economy. The evidence for that is the fact that economic models predicting the winners of presidential elections are the best forecasters, far superior to the myriad of polls, except the exit polls, and vastly superior to the legions of political pundits in the media. On the superiority of simple regression models over that of experts for forecasting anything, even the price of wine, read Super Crunchers.

One economic model predicting presidential elections using just GDP growth shows that the economy must grow at an annual rate of 2.5% in the second quarter of the election year in order for the incumbent party to win. Based on the latest release showing Q2 growth at an annual rate of 1.2%, Trump has virtually won.

The model created by Ray Fair at Yale University agrees, according to a story on NPR: "It's based on economic growth per capita in the four years before the election. According to Fair, because of the sluggish growth in this recovery, his model now predicts the Republican candidate will win. Fair's model has picked the winner in all but two elections since 1916."