|
| |
The Internet and the American Political Process
From: Columbia University
| By:
David EpsteinSharyn O'Halloran |
EDITOR'S INTRODUCTION |
In December 1999, the White House ordered the National Science Foundation (NSF) to conduct a one-year study on the feasibility of online voting. Legal scholars, technical specialists and political scientists were asked to investigate the question of whether Internet ballots will increase voter turnout without creating problems with fraud and privacy. Two Columbia professors, David Epstein and Sharyn O'Halloran, were among those asked by the NSF to examine these important questions facing American society in the cyber age. Epstein and O'Halloran began their investigation by conducting an analysis of the results of the 2000 Arizona Democratic primary, in which voters were allowed to cast their ballots online. The second stage of the study, not yet
undertaken at the time of this interview, involved joining forces with scholars at UCLA and Northwestern University to look at how changing different parameters of the Internet voting process will affect election
outcomes. Epstein and O'Halloran spoke to Fathom about the current and future challenges of Internet voting. |
Fathom: What was unique about the Arizona primary? |
David Epstein: The Arizona primary was the first large-scale coming-out party, if you will, for Internet voting. It was held both on the Internet and in real voting booths so that voters had their choice of which way to vote. There were four Internet days of voting, and then on the final day of the vote you could go to the polls and either vote on the Internet there or vote the regular way. |
Since the Arizona primary was the first large-scale use of Internet voting, it had the novelty aspect to help boost turnout rates. In addition, the company that developed the software used in the election, Election.com, spent a lot of money informing voters that this was happening, and spent a lot of man-hours on the ground troubleshooting in case any technical problems arose, and so there was clearly a big push on this one. In the future there probably won't be as much attention paid to every single individual that is voting, but hopefully things will be smoother and people will be more accustomed to Internet voting to the point where you won't need that. |
 | |
| Official ballot of the Democratic Party, presidential primary election 2000, state of Arizona. | |
|
Fathom: Had this been discussed for a long time? |
Sharyn O'Halloran: Internet voting has been kicked around since the 1996 elections. There has been a lot of discussion as to whether in fact this was a feasible ... whether issues of security could be overcome and so forth and so on. So yes, this was a first flagship attempt to go ahead and use the Internet. |
Fathom: Tell me about the work you are doing in analyzing the Arizona vote. |
O'Halloran: We have access to all the returns from the primary, and what we're going to do is link this up with census data, which will give us a good profile of the folks that voted using the Internet. The point then is to look at things such as participation rates, turnout rates and actual voting outcomes from these results, and the question is, "Does the Internet improve the quality of democracy? Does it improve the participation of different segments of our society that tend not to become involved in the voting process?" And this is the first cut at that. |
Epstein: There are two sides to the coin. The first is whether or not there's more participation in general, whether or not more people go to the polls or will vote via Internet voting when they wouldn't necessarily vote otherwise. Second is whether this increased participation might be biased in some way. Will it perhaps increase participation overall, but increase it most among affluent white voters that are already voting in disproportionate numbers anyway, and therefore relatively disadvantage, say, minority voters? In fact, before the Arizona primary was allowed to happen there was a voting-rights suit by an organization that said that this was illegal because it was expected to disadvantage minority voters. The judge let the primary go ahead, but there are now suits still pending, so they may in fact throw it out. But even if this one survives, the general question about greater participation versus biased participation, I think, will live on with Internet voting. |
Perhaps our research could help inform the suit. We might show that in fact minorities did show up and vote in greater numbers in proportion to everyone else, and that it didn't disadvantage them. When people think of Internet voting they think, oh, since it's usually connected to your home or at work, therefore how are people who don't use computers all day going to vote via the Internet? That's true, that's a bit of a problem, but they did set up voting computers in public places as well such as libraries, post offices and the like. So even if you don't own a computer or use one daily, you could walk in and use a computer that way. |
Fathom: What if data show that the method of voting was biased against certain voters? |
Epstein: I would expect that if we found that if the voting adversely affected minorities that in fact, yes, that would be seized upon by a number of groups to say, "Look, this is a discriminatory mode of voting." Maybe someday it will be feasible, when everybody has access to the Internet, when you know minorities are ready to vote on the Internet at rates comparable to other segments of the population, but right now they might say it's still too early to have Internet voting. |
Fathom: What other issues are you going to be considering? |
Epstein: Our research on Internet voting is going to address, on the one hand, the issues of: Does increased participation actually get people more excited about the democratic process? Is it possible that by voting online and being able to click on, say, hyperlinks to position papers by candidates you can learn more about them? Or, if you're going to build a new school, look at how the finished school would look if it's fully funded, partially funded and so on. We'll see if that helps voters actually learn more about what they're voting on, and perhaps create a more informed electorate to improve the overall quality of democracy. |
Another aspect of our research on Internet voting deals with alternative methods of voting that are possible using the Internet that aren't possible otherwise. So, for instance, instead of having one single election and having everybody vote on that date, you can have what's called a rolling plebiscite, where people vote once, then you show them an interim tally of the votes, then you can change your vote and so on and so forth up until the final date of the election. So instead of having an election at a single point in time--everybody has to vote and that's it--you can have more of a discussion, more of an evolving process. And that's something that's possible with Internet technology that just really hasn't been technically feasible before. |
A lot of states and localities see Internet voting as a cheaper way of doing the voting. You can put up fewer polling places, you can hire fewer election employees, so it actually turns out to be a bit of a money saver. Beyond that, I think one of the reasons that the Arizona Democratic Party did what they did was just to create excitement, to create some interest, to give people an alternative method of using this appliance they use every day, namely the Internet, and connect it up with their political lives, which sometimes are a little bit disconnected from other things that we do. |
O'Halloran: It also is a fundamental question of representation in the democratic process. I mean, one of the concerns that you have is that our normal elections are biased against minorities, youth voters and so forth. And it's disproportionately skewed toward older folks that have a little bit more time to get involved or more extreme voters who have vested interests. By reducing the overall costs of actually getting to the voting booth, in many ways you may be leveling the playing field. And so, there's a real push for many people who believe that they would benefit from this by increasing the size of the electorate, to go ahead and use this new technology. |
 | |
| |
Epstein: One of the things about voting the way it is now, we may not think about it all the time, but essentially it's a very controlled process. You go into a ballot booth, everything you see has been preapproved by the Election Commission. And so you think when you see a ballot that it's more or less officially released information. The Internet makes the election officials lose control of voting to some extent, because once you're on the Internet you can click anywhere. You can click to the Sierra Club's page, from there you might go somewhere else and the question is, are we sure we're going to get back to the ballot? Might they just start doing research, click away and forget to answer a few questions? So, for instance, the way California is thinking of doing Internet voting is to send all voters a voting CD, so that you would actually reboot your computer on that CD. It would give you a little ballot right there, there might be some information but it's all very self-contained, and only once you've completed the voting form and actually send it off via the Internet to the voter commission are you done voting. You can't actually click away to the Internet while you're browsing the CD, so it's a way to try to recapture a little bit of that control. |
Fathom: Are there other models of Internet voting being considered? |
Epstein: Another model is you just give all voters a PIN number and you let them vote at any Internet terminal, so they could vote at home, they could vote at work or they could vote at one of these terminals set up in libraries--that's the freer version of this method of voting. The other would be to send voters a disk. And they put the disk in their computer, they vote on the disk and send the disk back in, so they actually have to go mail something. It's more like an absentee ballot, but it's on a computer disk rather than a piece of paper. |
Fathom: When will we see the results of your analysis of the Arizona data? |
O'Halloran: Well, I believe the first view of the Arizona data we'll have by the summer. So by fall we'll have those results. Whether it impacts the 2000 election is not clear, because I don't believe there'll be Internet voting for the presidential elections. I don't know of any state that's actually moved in that way, but 2004, that's an open question. And I think that a lot of the state and local governments will be adopting this process much more quickly than at the national level, and I think we'll see it on school boards, I think we'll see it for local bond issues, I think we'll see it just because it's cheaper to do. And then once each state comes up with their own set of election laws on the way in which the Internet is going to be allowed to be used--informational content issues and the actual means of processing the votes--then in fact you'll see it moved to the national level. But I think it'll be probably another four years. |
Fathom: Will we abandon the polls altogether? |
Epstein: I believe there will be some elections that are going to be Internet only. Probably very localized, but just as Oregon had a mail-in-only election, we may very well see Internet-only elections. I doubt at a state level or national level, but perhaps at localities. |
By law, everybody has to be able to be allowed to vote, so you have to make it practical for everybody who wants to cast a ballot to be able to do so. So any Internet-only election would certainly have to have public terminals whereby anybody could walk up and vote. |
Fathom: What are some potential problems with Internet voting? |
Epstein: Internet voting raises some security issues. One type of issue is with the possibility of people hacking into the system and stuffing the ballot box electronically, so you have to make sure that you're counting the votes actually cast. Another is that people are going to be voting from work--then there's a big issue about whether or not companies, which usually reserve the right to monitor e-mail of employees and all electronic transmissions, will be able to observe employees' votes. This is actually a major concern holding up California at the moment: they're worried about whether or not they can get employers to sign a contract or somehow commit themselves not to monitor their employees votes, thereby destroying the secret ballot. |
Fathom: Are any of the candidates speaking to this issue? |
Epstein: I haven't seen any public official take a position yet on Internet voting. Who has taken a position, of course, is President Clinton, who said this is the way of the future. He's encouraging it and he's directed some federal agencies to make special studies of Internet voting. One of those agencies, the National Science Foundation, is going to be sponsoring our research. |
Fathom: Do you have any predictions to make about Internet voting? |
Epstein: The general expectation is that Internet voting would simply increase participation. That it would bring a new and exciting dimension to voting, to voters who are used to just having to trudge to the local school or local apartment building every few months. I think that so far the early results have certainly borne out this promise: the sheer number voters who voted in Arizona were much higher than in the '92 or '96 primaries. So, even though the race was actually over by that point--that is, it was clear that Gore was going to win--they still had a large number of voters to show up and vote via the Internet. So its early dry run has been pretty much a success. |
The Internet can also have big effects on politics other than just voting. Campaign finance is one example of this, where John McCain raised $750,000 on the Internet in the three days following the New Hampshire primary, so we've already seen a real explosion. No one remotely anticipated that fund-raising was going to be that easy over the Internet. Another thing that voters could do, for instance, is to circulate e-petitions where you have voters sign up by e-mail and then send those off to candidates. That might represent a new way to jump-start communication between voters and representatives. So there's a number of other ways that the Internet can impact politics short of actually holding elections on the Internet. |
O'Halloran: The other issue with the e-petitions is their impact on direct democracy. In California you have many referenda, much more than any other state, where voters make decisions that in other states are usually left to the legislature to choose: budget decisions, passing of environmental policy, and so forth and so on. So this may be a move to allow more direct democracy, more direct representation and broaden that not just to California but to other states as well. |
Epstein: There is also the possibility of more grassroots political organization done on the Net. In fact, there's a new company called Grassroots.com that's trying to set itself up for exactly this purpose, to facilitate group formation. I predicted four years ago that the biggest impact of the Internet was not going to be necessarily on national elections where everybody is pretty mobilized anyway--they know the candidates and know the issues. I thought the biggest impact of the Net was going to be on local elections and local politics, where you kind of hear things swirling around you but nobody really has the time, and it's not on the news all the time so you don't necessarily know what's going on. The biggest impact of the Internet could be on organizing these kinds of local groups for action on a smaller scale. So we might see more of that emerging as well. |
Fathom: You're both enthusiastic about the potential for Internet voting? |
O'Halloran: It is a democratizing force, just because it reduces the costs of actually voting, but it can be misused and it can have consequences that may be unforeseen, so you definitely need to understand what the impact is. And if in fact, if it does have skewing results--i.e., say minorities or other groups are disadvantaged by it--then you need to figure out a way to compensate for that. And so I think it's important for us to go ahead and do a systematic analysis of this to understand how each of these components play into the final outcome. |
|
| |
