We hear from Ole Rogeberg on occasion:
2009: Taking Absurd Theories Seriously: Economics and the Case of Rational Addiction Theories
2011: Descriptive statistics, causal inference, and story time
2012: Scientific fraud, double standards and institutions protecting themselves
2013: Struggles over the criticism of the “cannabis users and IQ change” paper
2015: Cannabis/IQ follow-up: Same old story
It’s been a few years and it’s time for some new material!
Rogeberg writes:
I recently thought of your work [with Jonathan Auerbach and Yair Ghitza] on how the popularity of the sitting president predicts future voting pattern, with stronger “shaping” effects for individuals in certain age ranges. Would it be possible to use the already estimated model to estimate how the Trump presidency may shift future voting patterns relative to a counterfactual where Trump is given a more typical popularity score pattern across his presidency (e.g., replacing his scores with those of George W. Bush or “average scores across recent Republican presidents” or something)?
I haven’t done the calculation, but my quick guess is that Trump’s four years of unpopularity will be bad news for future Republican support among people who are currently between 15 and 30 years old. So we can expect a continuation of the recent generation gap among voters, a gap that is unprecedented in the history of polling. I’m not sure what the implications of this will be for politics. I’m not saying that the Republicans can’t win future elections, just that, based on what we know now, I expect that they won’t get much of the young-adult vote.