Long lines at polling stations, unreliable voting machines, disputed ballots, parochial administration—and, of course, hanging chads. While voting is at the core of American politics, it has been fraught with far too many issues for the world’s so-called model democracy.

Addressing these shortcomings is challenging, particularly in the current atmosphere of partisan intransigence where hard lines seem to thwart even modest advances.

But two of the nation’s most well-respected voting experts, drawn from opposite ends of the political spectrum, came together in 2013 to lead a team that took a very hard look at voting in America.

Benjamin Ginsberg, a Republican, and Robert Bauer, a Democrat, were tapped by President Obama to co-chair the Presidential Commission on Election Administration, a yearlong investigation into voting in the United States. As expressed in the president’s executive order, the goal was to “promote the efficient administration of federal elections and to improve the experience of all voters.” Their ultimate findings, “The American Voting Experience: Report and Recommendations of the Presidential Commission on Election Administration,” published in 2014, offer an overview of the current deficits and recommendations to address them.

That the president chose Bauer and Ginsberg to lead this effort is not surprising—both are considered among the best in the field, with deep experience working with candidates in their parties. In the 2012 election, Bauer was President Obama’s campaign counsel and Benjamin Ginsberg was the top lawyer for Republican opponent Mitt Romney. Bauer, who has now returned to a partnership position at Perkins Coie, served as both personal attorney and general counsel of the Obama for America presidential campaign and White House counsel to President Obama. He also served as general counsel to the Democratic National Committee. Ginsberg, now a partner at Jones Day, served as national counsel to the Bush-Cheney presidential campaigns in the 2000 and 2004 election cycles and played a central role in the 2000 Florida recount. In 2012, Ginsberg served as national counsel to the Romney for President campaign.

And not only did the two men overcome partisan strains to produce a thorough and fair set of recommendations, they became friends—their easy banter in the interview that follows far removed from the headline-grabbing infighting of Washington.

Conducting the interview is Nate Persily, JD ’98, the commission’s senior research director, who worked closely with Bauer, Ginsberg, and the membership. Persily, the James B. McClatchy Professor of Law, is a constitutional law expert and sought-after nonpartisan voice in voting rights with extensive firsthand experience—including serving as a court-appointed expert to draw legislative districting plans for Georgia, Maryland, and New York and as special master for the redistricting of Connecticut’s congressional districts. In the following, Bauer and Ginsberg discuss with Persily this year’s presidential primaries, election issues, and the commission’s report www.supportthevoter.gov.

Persily: Let’s start by talking about the 2016 election. What has surprised you thus far? Ben, do you want to begin, because there has been far more activity on your side?

Ginsberg: I think the passionate anger of the electorate and the way that has impacted a lot of preconceived notions about the way the race would turn out have been surprising. The conventional wisdom has been turned on its head. I think the role of fundraising in the election is another surprise, in terms of super PACs. I think the number of candidates on our side [Republican] has been surprising too.

Persily: What about you, Bob? What surprises you thus far?

Bauer: I think, as Ben is correct in pointing out, a lot of expectations, in different ways for the different parties, have been not satisfied. I mean, we obviously have, as of today, a competitive race for the Democratic primary nomination. The enormous success that candidates on both sides have had with raising money online but less success in translating super PAC money into influence—those are new developments that hadn’t been readily anticipated.

Persily: Ben, Are you surprised by the ineffectiveness of the super PACs?

Ginsberg: Yeah, clearly the amount of money has not correlated to showings in the polls. And that was a common assumption that I think many people had going into the election. On the other hand, super PACs have historically been the most effective in drawing contrast. And a lot of the early super PAC money—certainly not all of it—was trying to credential candidates in a positive way.

Bauer: Well, on both sides that’s true. Ben knows better than I do about the Republican side, but you’ve had super PAC credentialing exercises—certainly when Jeb Bush went in with, I think, the best-funded super PAC, and all of the advantages that that was supposed to have brought with it. But that benefit did not translate into any significant value to him. But then, of course, you also have the enormous amount of negative advertising from at least a couple of super PACs. One of the super PACs dedicated itself against Marco Rubio, but he obviously had, at least from his point of view, a successful showing in the Iowa caucuses despite that. And then again in South Carolina.

So, a lot of the assumptions made about negative advertising and the amount of money raised demonstrate once again that the effect of this money is, frankly, highly contextual. You have to take a number of factors into account to really assess how well that translates into success for the candidates.

Bauer: But really, there have been fewer surprises than people seem to think. There’s a set of established expectations that are probably somewhat wobbly and a little suspect. Then, when those incorrectly constructed expectations are not satisfied, people tend to talk about surprises. Take, for example, the Iowa caucuses. If the Clinton campaign thought this was a walkaway and looked at the history of caucuses and at the electorate, with all that intelligence available to them, they wouldn’t have put together the organization on the ground that they did and they wouldn’t have expended the money that they did. I mean, the fact of the matter is they devoted substantial resources from the beginning because Democrats who run in the Democratic primaries—and I think Republicans that run in Republican primaries—realize that the best possible policy you can have is to expect the unexpected. So when we say “surprise,” it’s really only a surprise because this narrative develops that people subscribe to but is not necessarily well-founded.

Persily: What about the polling?

Ginsberg: I think one of the surprising things is precisely how inaccurate polling thus far has been. You know, in Iowa it missed on both sides. The national polling has been done with really small samples, but, nevertheless, has been talked about a lot in the media. And the polls have, at least so far, proved not to be very accurate.

Persily: Our friend Charles Stewart actually is making the argument that the polls are more accurate than people think, if you look at the averages.

Ginsberg: I think there are two problems. One is that some polls get put out by the press and there are reasons to have real doubts about their validity—whether it’s sample size or the screens that are used. For example, one of the polls in Iowa that was published maybe a month ago: While other polls were showing a relatively competitive race, it had one of the candidates ahead by about 33 points. How did that poll wind up being published? There’s a lack of discrimination here in the publishing of this polling data.

The second point, which I think may be somewhat responsive to Charles Stewart’s views, is that polls are systematically misrepresented and are relied upon to excess. That is to say, they are just not read properly. They are useful in certain respects, but they are definitely over-promoted and there are certain developments that polls cannot accurately predict or forecast—particularly in very close races. But, unfortunately, they are just sort of set out there as a depiction of what will happen. And that’s not, even for the best of polls, how they function. That’s not their value.

Persily: Let’s talk about other surprises. I must ask you a Trump question, Ben. What does Donald Trump say about the state of American politics and the state of the Republican Party?

Ginsberg: So, he has obviously been the loudest voice in the debate and has had an enormous impact on the plans, hopes, and dreams of the other Republican candidates by his personality and the strength with which he makes his arguments. And if you go down the list of Republican candidates—a brilliant, historically talented field—you can see how each and every one of them has had to change his or her tactics and strategies. Trump has dictated many of the items that have been debated through the way that he’s presented the issues. And I think this is partly due to the role of the debates in the Republican process.

The debate process, in effect, was a reaction to the 2012 issue that we had with so many debates. And the process put in place for 2016 was designed to protect the frontrunner, which, in this cycle, happens to be Trump. So the paucity of debates this time plus the use of polls to winnow out the field have really served to give Trump an awfully good platform and structure to this point.

Persily: Bob—do you want to comment at all on Trump, or should I ask you a Bernie Sanders question? What does Sanders’ ascendency tell us about the Democratic Party and the election?

Bauer: Well, Ben is correct. I mean, you have an electorate that is very distressed at the moment and doubtful of the efficacy of the political process and that believes the political system is unresponsive. Obviously, I’m speaking in very sweeping terms, but I think you can feel that restiveness. You can also see some significant demographic divides—the way that some voters think of the Sanders’ candidacy versus the way younger voters respond to the Sanders’ candidacy. So, it’s an unusual year. I don’t think, for a variety of reasons, that Trump is a case apart. People have a tendency to put Trump and Sanders in the same sentence. I tend not to do that, just because it’s very difficult for me to think of anybody who’s quite like Trump. When Trump was thundering away about what he called Cruz’s theft of the Iowa election I thought the rhetoric and communications style that he displayed, as a serious candidate with significant standing in the polls, unlike anything I have seen in many years.

Persily: Is it a one-off, do you think? Or is it a precedent for something else?

Bauer: I think that in a number of respects, it is a one-off.

Persily: You raised the debates earlier, Ben. There has been a lot of fighting over these debates and at these debates, particularly looking at what happened in the first one, the Fox debate, and then the Trump boycott. And more debates were added on the Democratic side. Talk a little about the role the debates are playing in the primary season.

Ginsberg: Debates are generally the electorate’s window into the candidates. And they play a very important function. I think where the debates —whether primary or general—get into trouble is when they are more about the setting and the format and the moderators than they are about what the candidates are saying. The Republicans have had a particularly challenging time this round because there are so many candidates on the stage and, besides, we feel sorry—for a nanosecond—for the moderators. It’s pretty tough to manage 10, 11—even seven or eight—people on a debate stage, so you tend to ask questions that don’t let the candidates talk about their positions on issues—they tend to be comparative. Sometimes, nastily so. So that’s led to the good and the bad moments in the debates so far.

Bauer: Ben and I worked together with the Annenberg School on a report about debates, and one of the conclusions of the report was that we need to look seriously at the traditional model. Candidates were becoming increasingly restive about the degree to which so many points about the structure of the debates are withdrawn from their control. This is particularly so in general elections. And they’re also very exercised about the very strong presence of the moderators in debates. That’s all been borne out recently—very much borne out. Obviously, the fabled Trump conflict with Megyn Kelly, that’s one example, and his withdrawal from the last debate in Iowa because of that conflict. And then debates were added to the Democratic calendar by the Democrats—the candidates themselves wanted more debates, added those debates, agreed on most of the locations for those debates, and agreed on the timing of those debates. And I wouldn’t be surprised, particularly based on the experience we’ve had this year, if we see a sort of reconsideration of the model for how these debates get set up with much, much more assertive candidate input and demands to have their interest in the debates reflected in the final structure.

Ginsberg: Yeah, in the current primary season, the parties have tried to play a role they’ve not played before, in terms of managing the debates. I think there have been massive frustrations on the Republican side brought up after the third debate, which was the CNBC one. And now you’ve seen it manifested on the Democratic side. You mentioned the declining role of the political parties, and I think their mishandling of the debates this time is reflective of that discussion.

Bauer: Which involves, if I could just add, surrendering a significant amount of control to media organizations, making selections on the basis of, frankly, moveable polling criteria. And I thought this second-tier debate business on the Republican side was just ludicrous—absolutely ludicrous.

Ginsberg: Just to be clear, what the Republican National Committee (RNC) ceded to the media was the ability to winnow a field of candidates based on small-sized national samples with a margin of error of plus or minus four points, at least. And the candidates, in the average of polls, were like .1 percentage points apart to .3, .5. Yet some were relegated to the undercard and some to the main debate.

Persily: So what does the ceding of control to the media say about the strength of the parties? Didn’t they willingly do this?

Ginsberg: Well, what the parties did was give the networks a set schedule of debates. The problem with that was the networks were then not as responsive to the candidates as the candidates would have liked. The RNC didn’t want to make the decisions, but what could’ve been done—and probably should’ve been done at the first debate—was to say the following: There are a lot of people who are close together; we’ve got 14 candidates up there. We’re going to have two seven-person debates and we’re going to use the polls to get the first, third, fifth, and seventh place candidates in one and the second, fourth, sixth, eighth in the other, or they could’ve drawn out of a hat to decide who got into which debate and not relied at all on those polls as a meaningful tool.

Persily: We worked together on the presidential commission. What are the reflections on the work that we did in following the president’s call to establish the Presidential Commission on Election Administration? Give us the post-mortem on that.

Bauer: It was encouraging to have a diverse range of experts who were fundamentally substantive—to have a process that was fundamentally bipartisan in approach, and bipartisan in tone, where we reached agreement and where agreement could be reached meaningfully. That was gratifying. Then the second most gratifying, or actually the most gratifying thing, was that the report when it was done didn’t simply gather dust on the shelf: There was an implementation phase. And through the good work of people like the Bipartisan Policy Center and the support of other organizations, there is going to be an extended implementation phase, and there has been very meaningful progress in bringing these reforms to pass in various states. I can’t say we could’ve predicted that, but we’ve been very glad.

Ginsberg: I would add that through the Bipartisan Policy Center’s efforts we’ve seen great implementation of the suggestions from the report. But we were also reminded of the challenge in this area of the number of jurisdictions that actually have control and authority over the voting process and the challenge of going state-by-state, and jurisdiction-by-jurisdiction, and the issues with consistency baked into our election administration system.

Persily: Both of you have been representing these parties for such a long time. Do you see fundamental changes in terms of the strengths of the political parties, versus the strengths of, say, outside groups, and what does that mean for elections?

Ginsberg: I think this presidential cycle shows that the campaign finance system, however it is designed, is upside down and no one on any side of the issue thinks the current system is working right; that the candidates are not the loudest voices in the debates; they really don’t have control over the issues that their own campaigns end up being defined by and what their own messages are. Overall, they have challenges controlling the perceptions of their candidates. The current system pretty much makes the political parties captive entities and second-class citizens once there are nominees. Then, voices that are not those of the candidates are going to be the loudest, which I think is bad.

Bauer: I share some of that exceptionally gloomy perspective, but I would say one thing that is really quite striking. We can’t predict how fundraising practices and the intersection of the politics of the country with technological developments are going to work themselves out.

But now, take a look at the Bernie Sanders campaign. That is a campaign that can finance itself easily all the way to the end of the primary season. It raised—in one month, the month of January—$20 million. It’s sometimes on the basis of a single news event, as happened recently, raising $2 to $3 million, all in small contributions, all of it fueled online. That’s good.