The Partisan Myth: How Voting Laws Actually Affect Election Results

>> Justin Grimmer: Election laws, from voter identification and felon disenfranchisement to automatic registration and no excuse mail voting, are depicted by partisans in the media as having massive consequences on who votes and ultimately, who wins elections. Many people express dismay about laws they disagree with, arguing the laws have dire consequences for American democracy.

Some claim that laws allow voter fraud to swing elections in one way, while others claim that votes are suppressed to swing elections the other way. Listening to candidates, pundits, and partisans, it would be easy to conclude that election laws are ground zero for heated partisan battles, with the United States status as a democracy at stake.

But research I co authored with my colleague Eitan Hirsch suggests that the heated rhetoric around election laws just doesn't match the evidence. In our study, we looked at the actual effects of policies like voter identification laws, election-day registration, and early voting on election outcomes. We found these laws have negligible impacts on who wins elections.

This may seem surprising given how fiercely the two parties fight over voting rules and legislative debates and court battles. Democrats often accuse Republican backed policies like voter ID of suppressing turnout of democratic voters. Republicans claim Democratic proposals like mail in voting invite fraud that benefits the other side.

Both parties imply the stakes are high whenever voting rules change. And yet our research shows that partisan stakes are actually very low. That's because most voting laws only target a small fraction of voters, barely affect turnout rates, or tend to evenly target Democrats and Republicans. Take voter identification laws, which require voters to show approved photo identification at the polls.

Critics charge these laws deter democratic leaning groups like minorities from voting. Yet studies have consistently found that voter identification laws tend to have no effect on overall turnout and even a minuscule effect on turnout for individuals who lack identification. Because the laws have such a small effect on turnout, it just isn't possible that they could swing many votes to either party.

And as a result, voter ID laws simply can`t swing all but the very closest of elections. It's a similar story with other high profile voting policies we analyzed. Expanding mail in voting might raise turnout just slightly, but that increased turnout doesn`t tend to benefit either party. Other laws have small effects that sometimes benefit Democrats and sometimes benefit Republicans.

For example, we find that giving felons the right to vote benefits Democrats in some states and Republicans in others. We provide a Mathematical framework to explain our results, let me give you an example. In a state with an electorate split 51:49 Democrats to Republicans, let's suppose a new law is implemented that imposes additional requirements to vote.

Let's also suppose that these new requirements only affect 4% of the electorate and among this group turnout drops three percentage points. And finally, let's suppose that those who are deterred from voting are disproportionately Democratic voters. So 60% of those deterred from voting are Democrats, while 40% are Republicans.

What would happen to the outcome of the elections after implementing this law? Well, the policy would cause a 0.12 percentage point decline in overall turnout. That's barely an 8th of a percentage point, and the vote for the democratic candidate would fall by 0.011 percentage points. In other words, if the state had 1 million eligible voters, the policy would deter 720 Democratic voters and 480 Republican voters, netting the Republicans a 240 vote shift.

In reality, few voting laws even come close to depressing turnout among those targeted by even one percentage point, let alone three percentage points. And those laws rarely target just one party's base so, precisely. Factoring in these realities, our study shows most voting rules barely make a dent in actual election results.

This explains why academic studies like ours consistently find negligible partisan impacts from reforms like mail voting, voter ID, and same day registration. Now, some claim parties and campaigns countermobilize against voting restrictions, masking their true effect. But my co author and I show counter mobilization is far too limited in scope to explain the null results found across dozens of studies.

The bottom line from our research is the partisan stakes of most voting laws are far lower than the overheated political rhetoric suggests. This doesn`t mean election rules are unimportant, but it does explain why the partisan outcomes rarely match the doomsday prophecies of political parties and pundits. So next time you hear politicians or talking heads on TV insisting otherwise, remember the empirical evidence.

Voting laws can raise valid concerns about equality, ethics, and implementation, but their partisan impacts are marginal. The data are clear even when the rhetoric isn't.