To: State legislators, editorial board writers, political reporters, and interested parties
From: Marc Stier, Director, Pennsylvania Budget and Policy Center
Re: Elimination of Straight-Ticket Voting and Related Matters
Voting is the fundamental political right of all Americans. And so, election reform must mean making it easier, rather than harder, for Americans to cast their votes for every office. Sadly, there is a long legacy in this country of doing the opposite. Whether put forward by racists determined to protect white supremacy or good government “reformers” determined to weaken the political power of immigrants and the working class, American states have adopted a series of proposals—such as onerous barriers to registration, limited hours and times for voting, too-frequent purges of the voter rolls, and Voter ID laws—that, inadvertently or in many cases deliberately, have made it more difficult to vote.
A package of election reforms is now under consideration in the General Assembly. This introduction summarizes our preliminary analysis of the bill. We stress the word preliminary because there is much that is uncertain about the impact of the legislation—and unfortunately remain uncertain until we see its effects. Our evaluation of the legislation is made even more complicated because different elements of the bill will have opposite effects.
One portion of the legislation, which we review at length in this policy brief, is the elimination of straight ticket voting Our preliminary analysis leaves us concerned that the elimination of straight-ticket voting will lead to fewer votes in down-ballot, and especially state legislative elections in the future. And that effect is likely to be more dramatic in very high turnout elections, which we are expecting in 2020, and which may come to characterize American politics in the foreseeable future.
Our analysis shows that the elimination of straight-ticket voting will in:
- lead to an average increase in undervotes of 5,781 in highly competitive state Senate elections and 13,968 in highly competitive state House elections;
- lead in presidential election years to an average increase in undervotes of 17,903 in highly competitive state Senate elections and 18,568 in highly competitive state House elections;
- lead to an average reduction per district of 1241 votes in state Senate elections and 845 voters in state House elections, numbers that are greater than the margin of victory in one Senate and four House districts per year;
- lead to an average reduction per district of 3378 votes in state Senate elections and 1197 votes in state House elections, numbers that are greater than the margin of victory in two Senate and seven House districts per year;
- possibly lead to a disproportionate reduction in votes from Black and Hispanic people and people with low incomes.
Note that we define highly competitive in legislative races we analyzed in Pennsylvania and four other states as those in which the Democratic share of the two-party vote was between 45% and 55%.
The reduction in the projected number of votes cast in state legislative elections, if we did not have straight-ticket voting in Pennsylvania, may not seem great in absolute terms. But in a state with a number of competitive elections—elections that we hope will become more competitive thanks to bipartisan efforts to reduce gerrymandering—the votes lost due to the elimination of straight-ticket voting could conceivably shift party control in one or both houses of the General Assembly.
Voters with low incomes face structural barriers to participation in politics. It is much more difficult to take off from work to vote or to attend candidate forums when you are paid an hourly rate rather than a salary—and even more so if you have to work two or three jobs to make ends meet. It’s also more difficult to find time to vote if your income is low and you have child or senior care responsibilities that middle-class families can pay others to meet. In addition, because political candidates and advocacy groups believe that people with low incomes are less likely to vote, they get less attention from their campaigns.
Structural barriers to voting disproportionately affect Black and Hispanic voters because their average income is below that of white voters, because they receive less attention from political campaigns and, in the case of Hispanic voters, because of language barriers. (These barriers also affect other groups of Americans as well.)
While we are concerned about the elimination of straight ticket voting reducing votes, the other changes in the legislation before the General Assembly might lead to more people voting and thus more votes in legislative races than are lost due to the elimination of straight ticket voting. Two reforms proposed in the legislation are likely to be particularly beneficial.
The first is moving the deadline for registration closer to Election Day. The deadline is currently 30 days before Election Day and the proposed legislation moves the deadline to 15 days before the election. While this is not as substantial a change as we would prefer—the ideal is to have same-day registration—and there are serious questions about how effectively county officials will process late registrations especially the first time the provision goes into effect, this is an improvement over the status quo.
The second reform in the legislation is mail in voting.1 Currently, Pennsylvanians may only request an absentee ballot if they are ill, disabled, or will be out of state on Election Day. The legislation enables any Pennsylvanian to vote by mail.
We have called for the implementation of both of these reform in our We The People policy paper on “Making Voting Easier.” The question we cannot answer at this point is how great the benefit will be of implementing the versions of these reforms in the legislation before the General Assembly now and in the future.
There is evidence that same-day registration can increase turnout by 5% to 7%2. But the effect of by moving the registration deadline 15 days closer to Election Day is likely to be substantially smaller.
There is also evidence that all mail-in voting increases voting turnout, although the estimate range quite widely from 2% to 7%.3 But the effect on turnout is likely to be greater if vote by mail is the sole method of voting. The legislation before the General Assembly only creates as vote by mail option. We are also not sure how soon any effect will be seen. Especially when mail-in voting is an option not a requirement, it takes time for voters to understand that is available and for political campaigns and advocacy groups to take advantage of it to encourage people to vote. It’s quite possible that it could take two or three election cycles for the full benefit of the mail-in option to be realized. A strong effort by the state to advertise the availability of vote by mail would bring the benefits of it more quickly.
Our best judgment is that the net effect of all these changes will be small with minor changes in one direction or another in turnout and in undervotes in down ballot races. We are uncertain about the direction of change.
One final preliminary point: Later in the report we discuss the possible partisan impact of eliminating straight ticket voting. It appears that Republicans have demanded it as part of the legislative package. And while that, as well as our own research, might be worrisome to Democrats, it is, as we point out below, hard to estimate the partisan effects of this part of the proposal. Much the same is true for the proposals to change the registration deadline or to create no-fault absentee voting.
At any rate, given the fundamental importance of voting, and our long history of limiting the franchise and suppressing voters and votes, our primary concern should be not the partisan impact of changes in election rules but whether they encourage or discourage voting for all.
Finally, there are two other elements of the legislation that are unalloyed goods—$90 million to reimburse counties for the cost of new voting machines and $4 million for census outreach. The latter is especially important in that it will help reduce the census undercount of Pennsylvanians. Doing so will help protect hundreds of millions of dollars that flow from the federal government to our citizens.
At this point, PBPC has not reached any conclusion about the overall merits of the legislation. We are uncertain about the critical question of whether the electorate will expand or contract as a result of the legislation and, most importantly, whether any changes in either direction will protect the right of people of color to vote. We also don’t know what amendments may be accepted before the legislation is ready for a final vote. As we learn more about the proposals in the current or any revised legislation, we may say more.
In the next section of this policy brief, we present our new data on undervotes in state legislative races in the last four Pennsylvania general elections. Then we present some reflections on how straight-ticket voting makes voting easier, especially for low-income, Black, and Hispanic voters. In the final section of the memo we address some philosophical issues about straight ticket voting in the context of an overview of the economic, racial, and ethnic barriers to voting equality.
READ THE FULL MEMO BELOW (Download a PDF version here):