ALWAYS BET ON THE dEEP sTATE:

The results in Table 2 show that 2016 turnout was by far the strongest predictor of 2020 turnout. However, after controlling for 2016 turnout, the data show that states that mailed absentee ballots directly to voters had a significantly higher turnout in 2020 than other states. Similarly, states that allowed the use of drop boxes for absentee voting had significantly higher turnout than those that required voters to put their absentee ballots in the mail. Finally, early in-person voting had a small negative impact on turnout but this effect was not statistically significant.

It should be emphasized that although some of these effects on turnout are statistically significant, all of them are quite small -- no greater than 2 or 3 percentage points. The most important development regarding turnout in the 2020 election is that it increased everywhere and by a rather substantial amount. Voters were highly motivated to participate in the 2020 election, just as they were in the 2018 midterm election before the pandemic hit the United States and many states changed their voting procedures. Turnout surged in 2020 in all types of states regardless of their partisan inclination and regardless of their voting rules.

The other major question about the effects of voting rules and procedures involves their impact on party performance. Did any of these voting procedures favor one party's candidate over the other party's candidate? In order to answer this question, I conducted a second regression analysis, this time with the Democratic vote margin in the 2020 presidential election as the dependent variable and various election rules and procedures as independent variables. I included the 2016 vote margin in each state as a control variable because there has been an extremely high degree of continuity in the outcomes of presidential elections at the state level in recent elections. In fact, the correlation of .993 between the Democratic presidential margin in 2016 and the Democratic presidential margin in 2020 was the strongest for any pair of consecutive elections since at least the end of World War II. Almost 99% of the variation in Joe Biden's margin in 2020 is explained by Hillary Clinton's margin in 2016. [...]

The results displayed in Table 3 show that after controlling for the Democratic margin in the 2016 election, none of the election rules included in the regression analysis had any discernible impact on the Democratic margin at the state level in 2020.

Tweet   @brothersjudd