Politics and Polls - 28 October 2020
Vizcarra faces a new corruption allegation. Biden is likely to win next week.
I cover recent polls every Wednesday. If you want to be added to the newsletter distribution list, please enter your email at https://boz.substack.com/ or email me at boz@substack.com and I will add you.
Two items today:
Peru - Vizcarra remains popular in spite of new controversy
US - Biden very likely to win the election
Peru - Vizcarra remains popular in spite of new controversy
President Martin Vizcarra faces allegations of corruption and a potential impeachment inquiry, but remains popular. Yes, that’s the same key point from a newsletter I wrote last month, but this is a new round.
This time, Vizcarra is accused of taking about a half million dollars in bribes from a construction company while he was governor of Moquegua. This allegation is far more serious than last month’s allegation of lying about a singer visiting the presidential palace.
An Ipsos poll last week shows 78% of Peruvians believe the president should finish his term in office in spite of the allegations. A significant portion of citizens who disapprove of the president do not believe he should be impeached.
Vizcarra maintains a 54% approval rating with 42% disapproval. That’s gradually tracking down but remains quite positive in a region where many presidents are far less popular.
Manuel Merino, the head of Congress who would replace Vizcarra if he were forced out of office, has 22% approval and 68% disapproval. Merino’s unpopularity likely plays a role in the support for Vizcarra amid the new impeachment threat.
The same poll shows 24% of Peruvians believe the next president should prioritize the fight against corruption. 61% of Peruvians mention corruption as one of the country’s top three problems, by far the highest ranking for any issue. That’s a shocking number given the country has among the highest number of dead per capita from Covid-19 and a severe economic downturn.
US - Biden very likely to win the election
Above: Aggregate of Superforecaster predictions from Good Judgment, Inc.
I did my usual search for “encuesta” across various Latin American media outlets and came back with page after page of results about the US elections. Well over 90% of the coverage of polls this week in the Latin American media was about the US. The US election is important for Latin America. So fine, six days away, I’ll cover it.
Joe Biden leads Donald Trump by around 7-9 points in most polling aggregators. He regularly polls above 50%, something that Hillary Clinton almost never did in the final six months of the 2016 election. Biden needs to win by about 4-5 points nationally to ensure an electoral college win.
Totally separate from my Latin America work, I’ve been tracking my US election prediction this year as part of this project.
I’m currently at 86% probability that Biden wins the US election. That’s fairly close to the current Superforecaster aggregate and 538’s current forecast. However, one big difference is my model has been relatively confident about the Democratic candidate’s chances this entire year (no matter who was nominated). I started around 70% in January and have remained above 80% since Biden clinched the Democratic primary, well above most other prediction models and betting markets in the early months of this year.
Why has my model been so confident of an opposition victory?
Multiple, high quality and regular polls in battleground states show the Democratic candidate winning by more than the margin of error.
Trump’s low approval rating and high disapproval rating. Having a net negative approval rating and a disapproval rating above 50% makes reelection less likely, particularly when facing an opponent with a net positive approval rating.
As a hemisphere, we’re in an anti-incumbent moment. Incumbents usually have an advantage when running for reelection (and that is built into prediction models), but the past 12 months have been a very different environment in which most incumbent parties have lost.
Since March, the mismanagement of the Covid response, the drop in the economy and the controversy over George Floyd’s murder have further harmed Trump’s chances. I haven’t seen much this year to make me think the incumbent is doing better than he was in January when I first put together my prediction model.
Biden’s lead over Trump has been far more certain and stable than Clinton’s was. I think many analysts who are more cautious with their predictions are overcompensating for the lessons of 2016 and missing the differences between the race this year and the one four years ago.
Thanks for reading
Have a great afternoon!