President Donald Trump looks set for re-election in 2020 if the economy doesn't take a huge knock and if Democrats turn out in low numbers to vote, contends Moody's Analytics in a new report.

Moody's based its prediction on purely economic grounds, that is, how consumers feel about their personal financial situations, the still resilient stock market and historically low unemployment. It argues Trump will win re-election if these three variables continue into 2020.

Trump's re-elections odds spike "if Democrats aren’t enthusiastic and don’t get out to vote,” said Mark Zandi, chief economist at Moody’s Analytics and co-author of the paper. “It’s about turnout.”

“Democrats need to be on high alert. If history is any guide and we get typical turnout, they are going to lose,” said Zandi, as reported by Politico. “They need to be at DEFCON 1 or it’s likely Trump will be reelected.”

Zandi said the race might well be determined by only a few key counties in Pennsylvania, especially Luzerne County in the northeast, which was a historically Democrat stronghold before 2016.

He described Luzerne County as "the single-most important county, no kidding, in the entire election.” Luzerne voted for Trump, 51.8% to 46.8% in 2016, after voting Democrat in the previous five presidential elections.

Zandi said Trump's doesn't even have to win Luzerne to get re-elected. He only has to make a strong showing in the county.

Moody's used three models for their predictions: the pocketbook model, the stockmarket model and the unemployment model. All three show Trump winning re-election.

Based on an average turnout, these models all show Trump winning at least 289 electoral votes. On the other hand, Trump's re-election chances plunge with maximum turnout on the Democratic side. The reverse is also true.

“Record turnout is vital to a Democratic victory,” said the report.

Moody’s found the Democratic nominee will win under the stock market and employment models if turnout among non-incumbent Democrat and independent voters were to match historical highs. The Democratic candidate will also win as an average of the three models.

Moody's said Trump does best under the pocketbook model that measured how people feel about their finances. Assuming average nonincumbent turnout, Trump gets 351 electoral votes to the generic Democrat’s 187.

The stock market model gives Trump a 289-249 win. The unemployment model favors Trump, 332 to 206. Trump wins 324-214 across all three models.

Moody's explained that its pocketbook model is the most economically driven of the three. It emphasized that if Americans voted mainly on the basis of their pocketbooks, "the president would steamroll the competition. This shows the importance that prevailing economic sentiment at the household level could hold in the next election.”

Co-authors of this paper are Dan White, Moody's director of government consulting and fiscal policy research, and Bernard Yaros, an assistant director and economist.

Moody’s modeling has been quite accurate in calling the presidential election winners since the 1980 election. It missed only once in 2016 when it predicted a narrow win by Hillary Clinton.