Listen: Political Scientists Frances Lee and Patrick Egan ’92 Clarify the Stakes of the Midterm Election

Political scientists Frances Lee of the University of Maryland and Patrick Egan '92 of New York University recently participated in a campus discussion entitled “Do Congressional Elections Matter? (Um, Yeah!)”, which explored the polarizing 2018 midterm congressional elections with Claude C. Smith '14 Professor of Political Science Rick Valelly '75.

“So the question that was posed for the panel was whether midterm elections matter," says Lee. "Now we probably wouldn't be holding this panel here if the answer weren't yes, but rather than focusing on predictions about what will happen in the midterms, especially given how chastened we political scientists have been about making predictions, I'm going to instead talk about how midterm elections matter for the performance of our political institutions."

“Midterm elections should be viewed as one of the most important checks and balances in the American political system, the constitutional system. The framers of the Constitution didn't necessarily anticipate that midterm elections would work out that way, but they served that role for at least the last century in American politics.”

The event, which was sponsored by Swarthmore’s nonpartisan Get Out The Vote Steering Committee and the Department of Political Science, is part of an ongoing campaign to get students to engage meaningfully in the political process for the 2018 midterm election and beyond.

Egan is an associate professor of politics and public policy and the director of undergraduate studies for politics at New York University. He specializes in public opinion and institutions in American politics; the formation of political attitudes; the politics of climate change; LGBT issues and politics

 

Audio Transcript:

Rick Valelly: ... So welcome everybody to our panel discussion on the House and Senate elections this year and before we get started, I'd like to thank Emily Weisgrau of the Communications Office for helping me put this panel together and also to Phillip Stern of the Communications Office for making such a terrific poster. I'd also like to thank the President's Office, the Lang Center and especially Katie Price and Delores Robinson at the Lang Center and the Department of Political Science for material support for this panel.

Rick Valelly: This year's election cycle has become highly nationalized by which I mean that the great majority of voters about three in four evidently care deeply about which party will control the two chambers of Congress and the substantial, but small majority, about three in five, tell [researchers 00:09:17] that for them the President is also on the ballot. And public debate has been pitched at the level of regime [inaudible 00:09:24] that the future of the regime is in question.

Rick Valelly: In an October [inaudible 00:09:30] Op-Ed for USA Today, the President had this to say, "The centrist Democratic Party is dead. The new Democrats are radical socialists who want to model America's economy after Venezuela. If Democrats win control of Congress this November, we will come dangerously closer to Socialism in America. Every single citizen will be harmed by such a radical shift in American culture and life. Virtually everywhere it's been tried, Socialism has brought suffering, misery and decay."

Rick Valelly: Which is true, but however that may be, you can see that the stakes are set pretty high with that kind of public [talk 00:10:10] and meanwhile such liberal commentators as George Packer of The New Yorker and Paul Krugman of The New York Times have written that the future of American democracy is on the ballot. So it is a suspenseful midterm election cycle. I can remember at one time that that was actually an oxymoronic phrase, but for at least 20 years it has not been. Midterm election cycles have been big deals.

Rick Valelly: So to help put all this in perspective, we're very lucky to have two really distinguished political scientists here to talk with us about how they think about congressional elections in general and its implications for governance and this election in particular. So to my left, hang on a second while I get the bio, is immediately to my left here on this Phil Donahue set [inaudible 00:11:10].

Patrick Egan: They don't know who that is.

Rick Valelly: So true, okay. Is Pat Egan who is Associate Professor of Politics and Public Policy at NYU where he specializes in public opinion, political institutions and their relationship in American politics. His current work includes research on social media and political knowledge, how politics is leading Americans to shift their identities, Americans' understanding of the complex problem of climate change about which he has written this positive article that appeared in Nature with Megan Mullin of Duke and the political identities of LGBT Americans. He is the author of Partisan Priorities: How Issue Ownership Drives and Distorts American Politics which came out with Cambridge University Press in 2013 and his peer reviewed research has appeared in such distinguished outlets as Nature, The Journal of Politics and The British Journal of Political Science and he's also a recipient of a teaching award at NYU.

Rick Valelly: And next to Pat is Frances Lee. Pat and Frances, it turns out that your work has been assigned here. And so Frances, your APSA paper started out our Political Science II Introduction to American Politics course. Frances Lee is professor of Government and Politics at the University of Maryland. She is co-editor of Legislative Studies Quarterly. Most recently she has authored Insecure Majorities: Congress and the Perpetual Campaign, Chicago, 2016. She is also author of Beyond Ideology: Politics, Principles and Partisanship in the U.S. Senate, 2009. Was that also with Chicago?

Frances Lee: Mm-hmm (affirmative).

Rick Valelly: Okay. And co-author of Sizing up the Senate: The Unequal Consequences of Unequal Representation in 1999 and a textbook, Congress and Its Members. Her research has appeared in The American Political Science Review, The American Journal of Political Science, Journal of Politics and other outlets. And in 2002-2003, she worked on Capitol Hill as an American Political Science Association Congressional Fellow. I forgot to mention, strangely, that you are an alumnus, Class of '92, but I think everybody knows this.

Rick Valelly: So Pat is going to speak first for about 20 or 25 minutes followed by Frances and then we'll have question and answer.

Patrick Egan: Hey, everybody, can you hear me okay? Okay. It's really nice to be back and especially on this fall day. I think when I was in college in the early '90s by now the trees would be kind of brown, but a little something has taken place in the last 25 years which kind of delays the fall and winter from happening and I think that's probably why it's still pretty green here.

Patrick Egan: So again, I'm going to kind of set the stage a little bit and talk about the electoral dynamics both generally that happen with congressional politics and midterm elections specifically and then also talk about the specific dynamics at play here in 2018. Frances is then going to pick up the baton and talk about the implications of these electoral dynamics that were policy making in Congress and it's relationship with the President going forward. So hopefully, this will make a good one, two act for all of you to follow.

Patrick Egan: So let's start with what I would say are four stylized facts about congressional elections in the present era. So the first thing is that midterm elections and of course, by that I mean elections that take place off years when there is no presidential election on the ballot like this year, generally result in a loss of seats for the party holding the presidency in the House. This has been true in most midterm elections since World War II with two really big recent exceptions that are worth noting.

Patrick Egan: The first was in 1998 when the Democrats won House seats in the wake of the GOP's impeachment of Democrat President Clinton. That didn't go so well for them and 2002 when Republicans won House seats when George W. Bush, the President, in the first election since 9/11 and were riding a big wave of approval and popularity for that Republican President. So that's the first fact.

Patrick Egan: The second fact is that there are generally very, very few House seats truly up for grabs in any congressional election. So there are a total of 435 seats in the U.S. House of Representatives and this year analysts say that no more than about 70, about 20% of them, can even possibly change party control in this election. So regardless of all of the kind of hype and hoopla that you're hearing about this election, what that means is that there are really only 20% of Americans who are in congressional districts where they're actually in a race that matters. Okay?

Patrick Egan: So if you're thinking about one of the reasons why Americans feel a little disenfranchised and uninterested in the congressional elections, that's one reason is that there's just not a whole lot going on for four out of five Americans. They're just not really in a very competitive congressional district. Okay. So that's actually true here in this area by the way, okay? So you are in the Pennsylvania's 5th congressional district which was redrawn by the Pennsylvania Supreme Court after it imposed its own map on the state after finding that the Republican map that had been drawn was a partisan gerrymander and therefore, too unfair. So the map that is in place today is actually a brand new map. Elections have not been run in these districts yet.

Patrick Egan: And this 5th congressional district which is mostly centered in Delaware County is not very competitive, okay? So the Democratic nominee whose name is Mary Gay Scanlon is heavily, heavily favored to win election here in two weeks. And that is again, not an unusual occurrence that we are not expecting a very competitive election here in Delaware County or in Swarthmore.

Patrick Egan: So some of the reason why there are non-competitive elections in most parts of the country is due to something you've probably heard a fair amount about which is gerrymandering. Okay? And that is a party's long lines, district lines, that advantage their incumbents. They tend to draw lines that are packed with voters of one party or another to make incumbents safe at election time. Okay? In most states, politicians draw the lines themselves and when you let politicians draw the lines themselves, sure enough they're going to draw lines that advantage them, that is the people who are already in office.

Patrick Egan: That's completely and perfectly constitutional and in fact, has long been the way that districts are drawn here in the United States. There are some states, and more and more of them, that have adopted what are called redistricting commissions that try to establish some independence from the politicians in terms of how lines are drawn, but most states the lines are drawn by politicians. And that's one of the reasons why there are so few competitive seats throughout the country.

Patrick Egan: Gerrymandering is not the only reason why there are uncompetitive seats for the House of Representatives throughout most of the United States. It's also due to the fact that voters are what we would call more residentially segregated on the basis of partisanship than before. So what that means is that there are geographic clusters of Democratic and Republican voters around the country and in particular, the Democrats are increasingly concentrated in urban areas throughout the country that are high density and because House districts are drawn in a compact fashion and as we like districts that look kind of like compact circles rather than giant spaghetti strings or pizza slices or crab claws, right, that because compactness is often a criterion on which we judge maps both aesthetically and sometimes legally, this actually disadvantages Democrats because they are focused on these high density urban areas.

Patrick Egan: So to just give you a little bit of illustration of that, this leads to a lot of Democratic wasted votes. So I live in Manhattan and Manhattan as I think you know is a long, thin island. It's also made up of basically Democratic voters, something like 85 to 90% of the residents of Manhattan island are Democrats. They voted for Hillary Clinton. They'll vote for Democratic members of Congress this year.

Patrick Egan: So Manhattan is worth roughly three congressional districts. The population of Manhattan is about 1.6 million and that's going to translate into 2 1/2 to 3 districts depending on how the maps are drawn. And so I could make a very nice looking map of Manhattan where I just divided it maybe at like 42nd Street, 96th Street and divided the island into three districts. And that would look very, very aesthetically pleasing. Right? There's something very sort of logical about that when we think about the map, right?

Patrick Egan: But that would create three districts that are just packed with Democrats and in fact, waste a lot of Democratic votes in these uncompetitive, big, big Democratic districts. How do I create competitive districts with Manhattan island? You have to draw basically spaghetti strings from, I don't know, how many of you are from the New York area, from West Chester County, Orange County, Rockland County, all the way from the sort of more Republican suburbs into Manhattan and those districts don't look very nice. Okay?

Patrick Egan: So nice looking districts which by that I mean compact districts tend to disadvantage Democrats who are crowded into these high density urban areas. And so when you hear about somebody saying wow, I would like maps that looked kind of nice, generally that sort of nice criterion which usually means compactness is one that these days disadvantages Democrats. So Democratic votes are wasted in New York and that's also true in Philadelphia, Chicago, Los Angeles, San Francisco, Atlanta, Boston. These are all big Democratic cities where compact districts waste Democratic votes.

Patrick Egan: So gerrymandering which I mentioned a minute ago and residential concentration both contribute to a third style as fact about congressional elections which is that they're generally biased against Democrats these days, okay? So analysts generally agree that at this point, the Democrats need to win the national congressional popular vote by like five or six percentage points even to have a shot at winning a majority of Congress. Okay? So I just want to let that sink in which is the Democrats can win more of the votes nationwide and still lose control of the House because of these two courses, residential segregation and gerrymandering, okay? So there's a built-in bias in favor of Republican Party in these House elections that Democrats have to overcome if they want to have any hope of taking over the chamber. Okay.

Patrick Egan: So a final fourth fact is that very few people vote in midterm elections. So generally, as you may know, U.S. voter turnout is really abysmal. In the 2016 presidential election, about 60% of all eligible voters showed up and that puts us well behind most of the developed world if you look at statistics from around the world, but things are even worse for midterm elections. Typically just about 40% turn out in these elections.

Patrick Egan: Now who tends to vote more? Educated people, wealthy people, white people and old people. All of which usually favor Republicans. Those are all groups that tend to tilt in the Republican direction. These stats are particularly stark in midterm elections which tend to give the Republicans an even bigger boost, an advantage in addition to the institutional advantages I mentioned a minute ago.

Patrick Egan: Now this year, one thing may be different and that's the story of the first group which is educated voters. There's been a shift among highly educated voters toward the Democratic Party and that can actually benefit the Democrats in this midterm cycle and I'll talk about that more in just a sec. Okay. So some of this differentiation in terms of who turns out and who doesn't is due to things like voter suppression and things like felon disenfranchisement. So there are active efforts mostly by the Republican Party to make it harder for people to vote, voter ID laws, those kind of things, all things that I'm sure many of you have heard about.

Patrick Egan: But it's also true and really overwhelmingly so that it's due to lack of interest, motivation and time to vote among the American populous that is distrustful of government, dislikes politics and is turned off by what they see as the endless fighting and posturing among the two major political parties. So we can pin some of the blame for low turnout in certain groups on active efforts to suppress their participation, but unfortunately the fact remains that even with none of those laws and effort in place, you would still see a big differential in terms of how much people turn out and a lot of it has to do with they're just not motivated and they're often not mobilized by their parties to turn out and vote.

Patrick Egan: Okay. So keeping these four facts in mind, the out party tends to pick up House seats. Very few seats are competitive. The House map is biased against the Democrats. And turnout is generally low. What do we know about this particular election that we're going to have in less than two weeks?

Patrick Egan: So first the landscape right now is that the division of the parties in the House currently stands at 240 Republicans and 195 Democrats. This means that the Democrats need to pick up a total of 23 seats to gain a majority of 218 to 217 and that number currently appears to be a very, very strong possibility. All the indicators point in the direction of solid gains for the Democrat Party and that includes a distinctly unpopular President of the opposing party, a wave of retirements among Republican incumbents, many in swing districts. Recruitment of very, very strong candidates by the Democrats to run in competitive seats. Solid fund raising advantages for the Democrats at both the elite and the grassroots levels and probably most important at this point, two weeks to go, national polls that have consistently found the Democrats to have a seven to nine point lead in the national popular vote which as I said a minute ago, puts them over the bar they need to win House control. They think the Democrats kind of need to clear a bar five or six points. If they can turn in a performance of more than seven to nine points they should be able to gain control.

Patrick Egan: So with less than two weeks to go, history tells us that last minute shifts in these advantages are typically pretty unlikely. Typically. There's one thing that political scientists have learned since 2016 is we don't make strong predictions about anything because a lot of people were making strong predictions about what was going to happen in 2016 and they were smoked. So we are cautious about the kind of predictions we make, but at least today, as we stand before you in this room, things look very, very good for the chances of Democrats taking over control of the House.

Patrick Egan: So why are the Democrats doing so well? Well, not surprisingly much of it comes back to Donald Trump who is a distinctly unpopular President given how strong the economy is right now. So I think we sort of, a lot of the media coverage of Trump sort of paints him as a guy who can get away with anything. And he certainly has a base of supporters who don't seem to be particularly moved by any of his really unusual and norm breaking behavior.

Patrick Egan: But it's important to note that right now he stands at about 43% approval rating and previous presidents governing when the unemployment rate was this low and economic growth was this strong, had approval ratings more like 60% or 65%, okay? So if we think about kind of the gap between where Trump is and in theory where he should be given how strong the economy is, it's pretty wide and that tells us something about his relative performance to previous presidents.

Patrick Egan: So why is this happening? Well, partly is because instead of appealing to middle of the road and moderate voters, Trump has really doubled down on adopting policies and politic rhetoric aimed at mobilizing a conservative base. It's also because he blatantly disregards many of the norms and expectations we have about presidential behavior, he's turning off a lot of voters who would otherwise be inclined to grant him credit for the strong economy. Right?

Patrick Egan: So again, one thing to kind of think about Trump again, regardless of where you stand on the political spectrum, is that he's not a particularly good President. That is, he's not very good at the job just objectively in terms of out performing the expectations about what Americans would think about him given the economy. And surprisingly, perhaps not surprisingly, not very good at this particular [performance 00:29:00].

Patrick Egan: So Trump is taking what would already be a weak hand for Republicans that is trying to hold seats in the House while they control the presidency and making it worse. I think that's sort of like the tag line here. Polling indicates that this has particularly helped the Democrats in districts that have highly educated voters, that are racially diverse or both. So I'm working at NBC News for the elections. I'll be there behind the scenes on November 6th and I have a colleague at NBC whose name is Dave Wasserman and he said somewhat jokingly, but I think actually quite accurately, calling these districts the Whole Foods districts.

Patrick Egan: The districts that have these highly educated voters who are going to places like Whole Foods, I'm looking you all in the eye and these are places that typically are swing districts because they are often high income places so that tends to favor Republicans, but they're also highly educated places, hence the Whole Foods thing, and these are the kinds of voters who have been turned off by the performance of Donald Trump and are finding themselves drawn toward the Democrats just as they were toward Hillary Clinton in 2016.

Patrick Egan: So as an example, not far from here are New Jersey 2nd and 3rd districts which are largely ex-urban and suburban places that include many college educated white voters and they're also rapidly diversifying. Both currently have Republican incumbents and it's possible that both of these districts will fall into the hands of the Democrats. So similarly around the country, Democrats are competing strongly in places like the suburbs of Philadelphia, New York, Chicago, Los Angeles, Houston, Minneapolis and Miami. These are all suburban districts that are often swing districts. They're relatively well off which again sends them generally in the direction of the Republicans, but these highly educated white voters plus a rapidly diversifying set of suburbs is giving the Democrats more of an advantage than usual.

Patrick Egan: What's important to note is that Democrats are having a harder time in some of the places where Trump's appeal to white less educated voters has been particularly affected. So that includes rural parts of the Midwest and Upstate New York where Republicans appear to be kind of keeping some of their advantage that they did in 2016. So here his tactic of making appeals to these voters' economic interests and racial and ethnic fears continues to be appear to be effective in maintaining their support. And on balance, this is what we'll have to see in two weeks, there appear to be basically enough Whole Foods districts available for Democrats to win control of the House outnumbering these other rural districts.

Patrick Egan: So I've been talking about the House and I've got to wrap up in a sec, but I just want to talk a little bit about the Senate where Republicans have just a 51 to 49 seat advantage right now. Here we should note that the story is strikingly different. Unlike the House where all 435 seats are up every two years, in the Senate elections are held for only one third of the seats in any given election. And this year, it just so happens that many of the Democrats up for reelection are in seats where Trump did quite well two years ago. So these include Indiana, Missouri, Montana, North Dakota and West Virginia. All of these have Democrats who are up for reelection and they are among the five most vulnerable incumbents on the ballot in the Senate this year.

Patrick Egan: It's very unlikely that all five of them will win reelection and thus for the Democrats to have any hope of winning control of the chamber, they need to win Senate races in Republican held states like Arizona, Tennessee, Nevada and Texas. And at the beginning of the fall, it looked like perhaps some of those states were truly in play. And what I think has happened since then, however, is that we're really only looking at one of those states, Nevada, where a Republican incumbent is really vulnerable at this point, this late in the cycle.

Patrick Egan: The other three states, Arizona, Tennessee and Texas, I think it's very, very likely that the Republican incumbents are going hold up. So on balance it suggests the Senate is likely to remain in Republican hands and in fact, if Republicans get lucky, it's possible that control could swing quite decisively in the Republican direction, even a 55-45 Senate in the Republican favor is not completely out of the question.

Patrick Egan: Two states to keep an eye on where Democrats may be in trouble although things have firmed up a little bit for them over the last few weeks according to polls is Florida and New Jersey which have Democratic incumbent Senators both of whom have been pretty vulnerable this cycle.

Patrick Egan: Okay. So to wrap up. I think we're looking very likely at a shift in control of the House to the Democrats perhaps decisively while the Republicans are very likely to maintain control of the Senate. Again, perhaps quite decisively. And I'll leave it to Frances to tell us what this means for the future, particularly in terms of governments and policy [making 00:34:05]. Thanks.

Frances Lee: I want to start out by thanking Rick for inviting me to come here to speak [inaudible 00:34:27]. It's my first time ever to visit a campus so I'm excited to be here and excited to talk about midterm elections coming up in less than two weeks.

Frances Lee: So the question that was posed for the panel was whether midterm elections matter. Now we probably wouldn't be holding this panel here if the answer weren't yes, but rather than focusing on with the predictions about what will happen in the midterms especially given how chastened we political scientists have been about making predictions, I'm going to instead talk about how midterm elections matter for the performance of our political institutions.

Frances Lee: Midterm elections should be viewed as one of the most important checks and balances in the American political system, the constitutional system. The framers of the Constitution didn't necessarily anticipate that midterm elections would work out that way, but they served that role for at least the last century in American politics. Why do they work this way?

Frances Lee: Well, first, one of the category discussed is that the President's party almost always loses seats in the midterms. In 35 out of 38 midterm elections that have happened since the end of the Civil War, the President's party has lost House seats. That totals up to 92% of the time. 35 out of 38. Since 1913 when the 17th Amendment to the Constitution was adopted instituting direct election of senators, the President's party has lost seats in 19 out of 26 elections so that's 73% of the time.

Frances Lee: So the midterms loss for the President's party is simply one of the most reliable patterns in American politics. There aren't very many laws of politics, but this is about as close to a law in American politics as we have. In midterms since 1934, the President's party has lost an average of 27 seats in the House and 3.9 seats in the Senate in midterm elections. This pattern ultimately results in a divided government. In fact, roughly half of midterm elections since the Civil War have created divided government where it didn't previously exist.

Frances Lee: The news media typically treat midterm elections as a verdict on the President's performance so you see this here in the headline from 2010 which were read, the 2010 midterm elections were read as a referendum on President Obama. The same was true in 2006 when George W. Bush lost his Republican majorities in Congress. So the 2010 shellacking was like 2006 thumping. There's some good reason to view midterms as presidential referendum. In part because midterm election outcomes do bear a relationship to the public's approval of the President.

Frances Lee: This slide here shows that relationship between presidential approval in October before the midterms and the seat swing that you see in the election outcome. So if 2018, based on this simple [inaudible 00:37:56] relationship, if 2018 performs as one would expect Republicans should lose around 34 House seats based on President Trump's approval rating. The Democrats only need 23 seats to take control of the House so this would be enough for Democrats to win a majority.

Frances Lee: Note though that Obama's approval rating was quite similar to Trump's in 2010 in the lead up to those midterm elections and Democrats lost 63 seats, but I would not expect Republicans to lose that many seats. Republicans have, as Pat discussed, a structural advantage in congressional elections in that Republican voters are distributed more efficiently across jurisdictions than are Democrats. You can see this quite clearly just comparing the performance of presidential candidates in the national vote with the number of states and congressional districts they carry.

Frances Lee: So for example, President Trump only won 45.9% of the national vote, but that constituted 30 states so he won 60 senators' constituencies and 240 House districts, a clear majority of House districts. So even though he got less votes than his presidential opponent, Hillary Clinton, he carried far more House and Senate jurisdictions. The same was true in 2012. So this is a long term pattern in American politics. It's not recent in that Republicans tend to win more jurisdictions in congressional elections than do Democrats based on the same overall national performance.

Frances Lee: But midterms elections are not simply referendum on the President in that even popular presidents tend to lose seats. Here I circle only those midterm elections that occurred when the President had an approval rating above 50%. So and you can see that in all but two of those cases, the President's party still lost seats. The exceptional cases were the ones that had [inaudible 00:40:21] in 2002 and 1998. In fact, there are some cases in 1974 and 1958 where the President was quite popular and yet the midterm loss was quite devastating.

Frances Lee: Now what does this mean for governance? So midterm loss is one of the most important checks on the presidents. The midterms tend to in turn exacerbate divisions within the President's party. Members of the President's party in Congress typically blame the President for the losses. Naturally, members of Congress prefer to blame the President than to blame themselves for bad outcomes. They're not going to say well, we did a bad job and that's why voters threw us out of office. They're going to say it's the President's fault. So it tends to introduce more back biting and contentiousness within the President's party in Congress.

Frances Lee: But in this sense, presidential elections tend to undermine the President's authority in Congress. In turn, tend to reduce presidential success with Congress. You can see that in looking at presidential success rates on the votes on which they take a position in Congress. President Trump's success rate in 2017 was startlingly high. All the votes on which he took a position, his position prevailed 98.7% of the time in 2017, but as you can see that tends to be presidents' high water marks and after midterm elections, their success rates in Congress tend to come down especially so in the cases that I've circled in which the midterm elections resulted in the introduction of divided government.

Frances Lee: Now what I'd like to do is to set 2018 in broader historical context. We've been talking about the possibility that Republicans would lose control of one or both chambers of Congress. Since 2016, literally just a week after the 2016 election, there began to be news stories speculating about the possibility of a change of party control in the Congress at the midterms.

Frances Lee: This is one of the most striking features of our contemporary political context, this ferocious party competitiveness. Back in the 1960s, 1970s it used to be conventional wisdom to describe American politics as being made up of a sun party and a moon party. This was a widely used metaphor and what it meant was that the normal state of affairs in American politics is that there's a dominant party and a secondary party and that the major issues of the day will be worked out among factions within the dominant party rather than in competition between the two parties.

Frances Lee: But since 1980, Democrats and Republicans have each held the presidency about half the time. Divided government has been the typical condition since 1980, 75% of time since 1980. In both House and Senate since 1980, Democrats and Republicans have held majorities in Congress roughly equal numbers of Congresses. Nine Democratic majority Congresses during that period. Ten Republican majority Congresses in that period, both House and Senate. The evenness of this partisan balance is just not normal for American political history.

Frances Lee: This figure displays a simple measure of two party competition at the national level since the Civil War. So just tracks through time starting with 1861 elections to the far left up through the most recent elections, 2016 on the far right. What I did to divide this measure of party competition is I just averaged three quantities, the Democratic party's share of the two party vote for president, the number of House seats Democrats hold in Congress in the House and the number of Democratic seats in the Senate. So I averaged those three quantities together and then I subtract 50 from that average in order to display periods with Democratic majorities above the line and periods with Republican majorities below the line. Taller lines means that the majority party is more dominant.

Frances Lee: So first thing you can see from looking at this figure is that the periods since 1980 stands out as unusually competitive for its narrow and switching majorities. The more typical pattern over this long swath of U.S. history is for one party to enjoy a substantial advantage over the other party in controlling national government. The Republicans, as you can see, were dominant in national politics for a decade after the Civil War and then again between 1896 and 1932. Democrats were similarly dominant for decades after The New Deal. The period most similar to the present competitive environment, the surprising stalemate after the Hayes-Tilden contested presidential election of 1876 until the elections of 1894.

Frances Lee: Those decades of the late 19th Century are probably the most analogous period to the present. Like our era it was a period of frequent divided government, intense partisan conflict and limited congressional productivity. I'll just flag for you the two other highly competitive periods. They're both rather brief during this history. You have the progressive era split in the Republican party that generated party competition between 1910 and the so-called return to normalcy with President Harding's election in 1920. Before the present, the last major burst of intense competition lasted for about a decade around the middle of the 20th Century from the loss of the Democratic majority at Truman's midterm in 1946 until the crushing Republican defeats in the 1958 midterms after which Democrats held nearly a two to one majorities in both House and the Senate.

Frances Lee: This next figure just sums up and simplifies the patterns. Displayed in the index is divergence from a 50-50 balance for each decade. So the Republican leaning era is shown red. Democratic leaning era shown in blue. Competitive eras shown in purple. What you can see here is that the time since 1980 stands out as the longest sustained period of near parody between the parties since the Civil War. So control of national institutions as here in 2018, but more generally in this whole post 1980 era has been in play. And when I say in play what I mean is that both parties when they're out of power see themselves as within striking distance of retaking control.

Frances Lee: You can see this in news coverage of congressional elections. After 1958 when this starts and up through 1980 which is right here, there were virtually no stories about the possibility that party control might shift in the upcoming congressional elections. So I just look at the two months preceding the elections counting news stories. So there's a count and then the line shows you the percentage of all stories about the congressional elections that dealt with the possibility that a switch of party control might happen.

Frances Lee: What you can see is that there's preoccupation with the question of which party will control Congress in news coverage of congressional elections. So in 2016, there weren't that many stories about the congressional elections because we were so focused of course on the presidential election, but a large percentage of those stories were about the possibility that a majority control might switch in 2016. But if you look at news coverage in earlier eras, the question was more about how were the parties doing in different regions of the country or it would be profiles of particular races that were interesting, but not questions about majority control. And of course, that reflected the Democratic dominance of Congress before 1994 and before 1980. 1980's the key election for afterwards when the Senate goes into play. So the Senate becomes the so-called permanent Democratic majority ended in 1980 in the Senate and in 1994 in the House.

Frances Lee: So how do these circumstances of intense party competition affect congressional politics. So the midterm elections of 2018 are part of a larger pattern and one that matters for how members of Congress behave between elections, not just during elections. So let's talk a little bit about how this affects political incentives.

Frances Lee: First it undercuts incentives to negotiate bipartisan agreements. Elections are not just important during election seasons. The prospect of future elections affects Congress all the time. So members of Congress look towards their expectations of where they'll be after the next set of elections. First it's possibility that the majority control might switch in the next elections always opens up the possibility that the party that's not in power will be in a better position to negotiate a more favorable deal after the next elections. It just leads to a focus on the near turnaround, that the time horizon gets shorter.

Frances Lee: For decades in the 20th Century, the only real question about congressional elections was how big the Democratic majority was going to be. That's a very different political environment from one in which we wonder which party will have the majority. It's hard to look past the next elections to think about long term institutional questions like the balance of power between Congress and the President or how Congress could reform itself to work better as an institution. It's hard for members of Congress to enact public policy to settle major policy questions because it's hard to negotiate a deal when the party out of power expects to be in power again on short term horizon.

Frances Lee: It amplifies risk aversion. It's harder for leaders and members to take political risks given their majorities. Your hope is that you'll be able to win power at the next election or hold on to the power that you currently have. This leads to a cautiousness about risking the support of your political base. Disappointing your base is the last thing you want to do if you want to win or hold majorities in Congress. Bipartisan deals are risky if you're seeking to hold on to this political support from your base.

Frances Lee: So a minority party that works productively with a majority party on legislation, cuts a bipartisan deal, gives up arguments that it can make against the performance of the majority party. How do you work productively with the parting opposition and then turn around and say that they're doing a bad job? How do you negotiate agreements that you both vote for and then say well, you should vote them out of power anyway?

Frances Lee: So to work together cooperatively tends to validate the party that has a majority. So it just makes compromise look less rational to minority party, this expectation being able to win the power. It always seems to look to say at the next elections well, the reinforcements may be here. We may be in a better position. Why strike the deal now? So this sheds some light on the tendency in recent congressional politics to constantly kick the can down the road on big issues.

Frances Lee: Second key effect of this closed competition is it focuses Congress on messaging, on partisan messaging operations. Partisan messaging is always about amplifying the differences between the parties. A party wants to give an answer to the question why should you support us and not the other party? The parties are always trying to answer that question via messaging. The answer to that question no matter what it's content is has to be because we're different from the other party. You should support us because we are not like them. We're different. We're better. So we're constantly drawing differences using party messaging.

Frances Lee: You can see this in the institutional development of Congress in terms of the increased capacity to develop and drive party messages. There's been the creation of an elaborate infrastructure in both parties and in both chambers of Congress to create and disseminate partisan messaging. I show this to you in just staffing decisions. So this figure displays the total number of staffers working for leaders in the Senate. So that's the bar. Total staff. The number of people. And then the line gauges the percentage of those staffers who work on party communications functions.

Frances Lee: You can see that there were no Senate leadership staffers working in communications before the late 1970s. You begin to see Republicans bringing on board communications staff in the late '70s in the lead up to the critical watershed 1980 elections. And after that, it's sort of been an arms race between the two parties as they constantly increase their staff levels to improve their communications capabilities since 1980. So that by the end of this time series, about half of all the people who work for Senate leaders, work on messaging. Speech writing, red pages, their events planning, communications effort [inaudible 00:56:11].

Frances Lee: Trends in the House are a little different, but also tied to competition for control. Before the 1990s, the share of leadership staff and communications held around or below 10%. During the 1990s, of course 1994 is the critical election when the so-called permanent Democratic majority in the House was ousted, you begin to see throughout the 1990s the number of leadership staff doubled and the share working in communications nearly tripled.

Frances Lee: By the end of the series, the share of House staffers working for House leaders who do communications is about one third of all staff. So what you have today in Congress is a legislative branch that includes a workforce of hundreds of professional communicators whose job it is to drive a partisan message. Every day they work to try to enhance the perception of their own party and undercut public perceptions of the opposing party.

Frances Lee: Finally, I'll say competition for power, this [inaudible 00:57:27] competition for power fuels partisan confrontation in Congress. Parties trying to get back in power often calculate that they're better off if they use every opportunity to confront and undercut the party in power with the goal of setting up favorable issues for the next election. So this is part of the reason for the hyper partisanship that we see in congressional politics today. To show you this, I'm just going to give you a series of quotations from different leaders over time where they just explain the relationship between competition for control and the need for partisan confrontation.

Frances Lee: We'll start here with Senator, in this case Minority Leader, Robert Byrd, right after Democrats lose their Senate majority and go into the minority in 1981. This is what he told his colleagues. He said, "What we need is a solid record of opposition to Mr. Reagan built brick by brick, piece by piece. I told my colleagues that if we should go out there and offer their alternatives, even though they knew they would be voted down. What looked like defeat after defeat in 1981, would look differently a year and a half down the road." So in other words, even though you know you're going to lose, frame the issue, force the vote, define the differences between the parties and make your case to the American people for a change of party control. It's a minority party strategy out of power.

Frances Lee: Here's Dick Cheney back when he was Republican Policy Committee Chair, 1985. He said this, "The confrontation fits our strategy. Polarization often has very beneficial results. If everything is handled through compromise and conciliation, if there are no real issues dividing us from the Democrats, why should the country change and make us the majority?"

Frances Lee: I did some interviews with long, long serving staffers for my last book and these are just a couple of quotations. I got a lot of statements along these lines, but these are [inaudible 00:59:37]. A veteran of the Republican Senate Chief of Staff said, "In the minority, you don't want to fuel the success of the majority. Too much deal making can perpetuate them in the majority. It makes them look good. If you cut a deal with them, it makes them look good if you're in minority and you're working with the majority."

Frances Lee: Veteran Democratic House Leadership aide said, "It's self defeating to work in a collaborative way. It's a constant fight. This is true on the floor, but it's also true in the committees as well. You look for opportunities to differ with the other party. You want to stigmatize the opposition. The stakes are huge and so you do this all the time."

Frances Lee: Here more recently, this is from a Republican strategy meeting held the night of President Obama's inauguration in January 2009. "If you act like you're the minority, we're going to stay in the minority," said Minority Whip at the time, Kevin McCarthy. "We've got to challenge them on every bill."

Frances Lee: And here Mitch McConnell has been quoted many times, probably quite famous for saying this. He said this to Carl Holz in The New York Times, "it's absolutely critical that everybody be together," talking about Republican strategy during Obama's first two years in office. "It is absolutely critical that everybody be together because if the proponents of the bill were able to say it was bi`partisan, it tended to convey to the public that this okay. They must have figured it out. It's either bipartisan or it isn't. I think the reason my members are feeling really good is they believe that their reward for playing team ball this year was the reversal of the political environment and the possibility that we will have a bigger team next year. So hold your ranks together. Oppose, deny their party in power bipartisan legitimacy and take the issues to the voters in the upcoming elections." These are incentives in a highly competition environment.

Frances Lee: So this leads me to my concluding thought for you when we think about is there such a thing as too much electoral competition? That normally we celebrate the virtues of political competition for purposes of producing accountability and clear choices for voters, but the logic, when you see these politicians, political leaders, articulating over time, political leaders of both parties, point to some distinct downsides of competition.

Frances Lee: Politicians face trade-offs between campaigning and governing, between their electoral prospects and cutting bipartisan deals. This is true generally. It's especially true for out parties. Governing is difficult and it involves concessions that will disappoint party constituents so a party that wants to win back power may well be better off promising pie in the sky, all the great things we'll do if we get our power back and refusing to negotiate on anything they're willing to [inaudible 01:02:48] disappointing base voters.

Frances Lee: Out parties probably do improve their chances by harsh demonization of the in parties and by refusing to blur the lines by dealing. The implication is they are making less bipartisanship in contemporary American politics now and in recent decades because the electoral stakes are so high. Maybe all this competition isn't always good for governmental performance. Thank you.

Rick Valelly: Thank you. Both of you. That was terrific. Absolutely terrific. We have half an hour for questions and let's start with anybody who wants to throw out a question. Yes, sir, [how are you 01:03:52]?

Audience: I guess one of the things that come to mind was we often hear voters talking about how we want some more partisan efforts and more working across the aisle between the two parties, but you talked about how these bipartisan efforts tend to disappoint constituents and that kind of seems an oxymoron. So can you elaborate on that more?

Frances Lee: What I would say is that the broad electorate approves of bipartisan deals, but party bases are often unhappy with them. One thing you always have to be sensitive to when people are being polled on questions about wanting to see more bipartisan deals, bipartisan deals on what terms? And so strong partisans may want to see some bipartisanship meaning that the opposing party goes along with what they want to do. So it's often hard to know exactly what to infer from polls that show that. But clearly, the minority, the out party, is aware of the tendency of voters to infer that legislation that commands broad bipartisan support is good legislation. So they know that that's how people will understand it if legislation is enacted with big bipartisan majorities. So knowing that and realizing the implications that has for their future ability to make a case against the party in power, it gives them political incentive to refuse to work together.

Rick Valelly: Sam.

Sam: This question is more for Pat. You were talking about Whole Foods districts and there's kind of [inaudible 01:05:49] talk to me about these districts also being called some of them being Romney Clinton districts. Districts that voted for Romney and Clinton in 2016. I just wanted to hear what your thoughts are about those districts staying blue in the future or reverting back [inaudible 01:06:07].

Patrick Egan: Great question. So let me know if you can't hear me. The mic tends to make me even louder. So the question is basically are we seeing kind of a permanent shift right? So basically, if you think about the difference between the electorate in 2012 and how it shook out in 2016, there's one big transformation which is that white educated voters shift toward the Democrats and white voters without a college education moved toward Republicans, okay? Everything else pretty much stayed the same, but that shift was pretty much unprecedented in recent American political history. Okay? In that white voters moved in opposite directions with educated voters moving toward the Democrats, less educated whites moving to the Republicans.

Patrick Egan: And so a big question that's come up is is this a permanent shift? Because one answer would be well, that's just a function of Donald Trump. It's his rhetoric, it's in his sort of appeal to a white ethnonationalism that is very attractive to less educated voters who don't have the same core set of values in terms of diversity, of caring about equality among different groups and also feel threatened by things like globalization, international economy, etc. So it could be that Trump sort of caused this shift and once he's off the political scene, we'll kind of revert back to a place where voters who weren't divided along economic lines where more economically affluent voters, white voters, support the Republicans and less affluent voters support the Democrats. And that would mean that we'd see it flip back where the Romney Clinton districts become Republican again and the Obama Trump districts become Democratic again.

Patrick Egan: The other thing is, and I actually think this is much more likely, that like the rest of the developed world, the United States is currently in a very fundamental debate over its national destiny. The question is are we a multi-cultural democracy where the white majority is slowly becoming more comfortable with giving up power to a diverse electorate on lots of different dimensions or is that white majority going to call for what we might call a restoration of the white ethnonational power structure that has ruled in the United States since its founding. My sense is we're seeing that debate happening in places like Britain, in France, in Germany, in Italy, all over even in Canada, we see some of these provincial elections-

Rick Valelly: Oh, Canada.

Patrick Egan: ... it's even in Canada. We're seeing this kind of debate emerge. And it's a result of international globalization so people feel the pressure of a global economy that squeezes wages particularly at the bottom. And also international migration which is creating much more diverse populaces in countries that have long had white European majorities. So my sense is that these shifts which you quoted as sort of Romney Clinton districts is probably now a more permanent fixture in American politics and this is basically the debate that we are going to have as an electorate, as a polity for decades to come.

Rick Valelly: Gabriel.

Gabriel: I'm curious, you touched on this in the discussion about gerrymandering, whether you think this sort of sexiness of gerrymandering and voter suppression as issues now is going to lead to policy shifts because you see movements toward independent redistricting commissions by referendum and then the referendum in Florida and Michigan with restoration of voting rights. So I'm curious if you think that there could be a sustainable bipartisan, do you think that there's going to likely be a bipartisan sort of shift towards that becoming more of an issue or if you think that that's specific for Democratic is gerrymandering out of power.

Patrick Egan: Yeah, I think it's a maybe, I don't know if it's [inaudible 01:10:25]. So what we're seeing in this cycle is a number of efforts to shift, that's both very good points, that basically move policy away in ways that take districting out of the hands of politicians and put them into independent commissions and also to roll back some of the disenfranchisement around people with felony records and other criminal records.

Patrick Egan: What's notable is that both of those things are taking place at the ballot box. That is they are not being done by legislative majorities and especially are not being done in states where Republicans control the legislature. For better and for worse, the Republicans have figured out that these laws help them because they tend to keep voters who favor Democrats, they make it hard for those voters to vote and so what Democrats have now figured out is that they can put these things on statewide ballots then they have kind of a shot at reversing these rules.

Patrick Egan: We'll see, you're talking about a referendum in Michigan that would put the districting in the hands of an independent commission. You're talking about a referendum in Florida that would restore voting rights to something like a million people in Florida. I mean it's a really, quite substantial shift and both of them are polling very, very well right now. If those two initiatives win on November 6th, my sense is we're going to see a wave of these efforts across the country that may roll back some of these advantages of the Republican Party.

Rick Valelly: [Mallory 01:12:01].

Mallory: I have a question for Dr. Lee. Have you ever encountered anyone who said that bipartisan work is better regardless of the strategic power plays that you mentioned? And adding on to that, do you think this attitude towards bipartisan cooperation is more a reversal of responses to our sort of polarized culture or do you think it has to do with the egos of politicians?

Frances Lee: So some big questions. I mean in the interviews that I've done on Capitol Hill and the reading I've done on politician's attitudes like in memoirs, politician's own attitudes towards bipartisan agreement, they recognize that it's usually necessary in American politics in order to get any major changes implemented. That it's very rare that you can legislate how Senate get a presidential signature on narrow party lines. Almost no legislation passes like that on the Obamacare model.

Frances Lee: So most of the time, they recognize if they're going to do anything, they need to work together, but there's just too many veto players in American politics, bicameralism, the frequency of divided government, that you wind up having to work across the aisle to accomplish things. So they're cognizant of that. They bemoan the focus on politics that gets in the way of their ability to do policy making as they see it. They see a trade-off between politics and policy.

Frances Lee: It's a source of frustration for members. The expectation, I think what keeps us stuck in this moment, is that parties have hope. They believe that down the road there will be a realignment and that there'll be a new real majority party like they used to have in American politics. If we can just get there. And so they say well, once we get there, then we can begin to resolve some of these issues and so it's that looking toward the future and the expectation that we'll be able to get some real things done later even if we find ourselves unable to do it now given our political predicament.

Frances Lee: But it's an open question. I mean Pat pointed out how similar the 2016 election results were to previous elections. You know we're not seeing any big electoral change in American politics. The 2000 map looks like the 2004 map or the 2008 map. Presidential elections look the same year after year in terms of which states go which direction, in terms of which party wins, which were the battleground states. I mean there's just not a lot of new. We can describe the constituencies that support the Democratic party and the constituencies that support the Republican party and change is modest. So I haven't seen any real prospect of breaking out of this close, competitive environment anytime soon, but I think politicians, elected officials, are constantly looking towards the future in the expectation that there'll be a breakthrough that just doesn't come.

Rick Valelly: Yes, sir.

Audience: So with that idea, sorry I'm sick. Republican confidence plummeting in Congress and the abilities of Congress to be effective and the increasing harshness of partisanship in Congress, do you think that we'll see an expansion of executive privilege or power or even resort to executive orders to get anything done in the American political arena?

Frances Lee: I would argue that it has had that effect, that a Congress that is gridlocked and mired in partisan conflict can't respond when the President interprets his authority broadly and lays down expansive executive orders. Congress can't marshal itself to act in response and so the President in effect is able to get away with it.

Frances Lee: You're seeing this right now with immigration policy under Trump. You saw it also under President Obama. That the Congress, there is no majority for anything on immigration policy either in the House or the Senate. They don't know what direction to go and so in that power vacuum created by party conflict, presidents are able to step in.

Frances Lee: The same also has been true of the courts. Part of the reason why the battles over the competition in the Supreme Court have become so fierce is that stakes are so high in a system where Congress is unable to act. The Supreme Court finds itself having to settle so many major issues in American politics. Supreme Court's word is final even when the Supreme Court is just interpreting statutes. If the Supreme Court interprets a statute like the Voting Rights Act in a way that undoes the status quo, changes precedent and Congress cannot act, cannot respond, then the decision stands even though that may not be a Constitutional decision. So Congress is overridden fewer statutory interpretations of the Supreme Court in this gridlocked, polarized era than had been the norm in earlier periods. So I'd say it's already happening, what you're describing. We see it under Trump just as we've seen it under Obama and under George W. Bush.

Rick Valelly: [inaudible 01:18:10]. Yes, in the back.

Audience: What do you think it will take for [inaudible 01:18:18] Democrats for the midterms to be competitive?

Frances Lee: I'm sorry, I couldn't hear.

Audience: What do you think it will take for parties that are outside of like for helping the Democrats become competitive?

Rick Valelly: This is a third party question.

Audience: Yeah.

Patrick Egan: Well, one thing is that you could think about the U.S. system as particularly set up to kind of channel us or funnel us toward a two party system and that's because most of our elections are what's called first-past-the-post which is that as opposed to say a proportional representation system, the candidate or the party that comes in first in any election whether it's governor, senator, member of the House of Representatives, mayor, city council, you name it. All of our elections are just the first, generally, the first candidate wins.

Patrick Egan: And so what that means is that if you vote for a candidate who is say running third, most likely and you can see this from pretty way out, your vote is going to be wasted. That is you're not going to have any real influence on that outcome. What that tends to do then is to lead for voters to vote what's called strategically. That is I'm going to vote for the second choice that's not as bad as my worst choice even though at heart I'm a green party person, but certainly the Democrats are much better than the Republicans. I know the green party can't win in this first-past-the-post system so I support the Democrats. Lots of voters make that kind of calculation and hence, you end up with very, very little support for third party candidates even when they're well funded and have a lot of momentum when somebody like Ross Perot in the '90s.

Patrick Egan: So what I would say is that in order to have real third party competition, you'd probably need to change the rules of how members of Congress and other elected officials are actually elected.

Rick Valelly: Yes, Angie.

Angie: My question is for Mr. Egan because I know you specialize in public opinion, how do you feel about the effectiveness and/or the accuracy of polls in gauging public opinion on candidates and through polling, can polling be [inaudible 01:20:31]?

Patrick Egan: Good question. So in 2016 polling came in for a lot of criticism and some of it was justified and some of it wasn't. And so it's probably helpful just to talk about that story to set straight. So in 2016, we had a whole bunch of national polls that were polling about the national popular vote which as we learned in 2016 doesn't matter, right? Because what matters is who wins in the electoral college. As Frances showed the data earlier, Hillary Clinton won the national popular vote by actually a pretty sizable margin, more than two percentage points-

Rick Valelly: Which was predicted by the polls.

Patrick Egan: ... which was predicted by the polls. Exactly. So in terms of the national popular vote, the polls were actually more accurate in 2016 than they were four years beforehand. So that's a story that people don't usually hear, but that's important to know which is that on the national level, we actually had better polling in 2016 than we did in 2012.

Patrick Egan: Where polling fell short in 2016 was in a critical key set of states that showed Clinton ahead in places like Michigan, Wisconsin and Pennsylvania where she ultimately lost. In all those cases, she lost by pretty narrow margins. I think maybe at most two percentage points in any of those places. And pollsters have spent a lot of time trying to figure out exactly what went wrong with the state polling that predicted an electoral college victory for Hillary Clinton, but actually resulted in a victory for Trump.

Patrick Egan: And there the sense is, one of the senses is that there was a last minute shift toward the Republicans in a lot of these places through the upper Midwest and other rural areas toward the Republican ticket that these state polls just didn't pick up and that's often a problem in elections in that you're not running polls in the final days of the campaign.

Patrick Egan: There's some also concern in 2016 that pollsters were not picking up and not accounting for the preferences of less educated white voters. And as I said a few minutes ago, that turned out to be critical because these less educated white voters moved so decisively to the Republican side and that also led to an underestimate of the Republican strength particularly in some of these key states.

Patrick Egan: So what we want to think now, and this gets to your question, is that polling is always trying to improve and in many ways pollsters have corrected for some of those mistakes that they learned in 2016. Polling is never going to pick up last minute shifts in opinion which was part of my hesitancy earlier to say hey, it's two weeks out. Things look pretty good for the Democrats, but you just never know what's going to happen and we're actually not going to give really, from this point out, not going to be able to pick up any last minute shifts in opinion. I think it would be very, very unlikely to kind of detect that with any real precision.

Patrick Egan: But I, as a public opinion guy, I guess this is sort of my party line, but I actually still think that polling is both generally pretty accurate, and that's true both here and around the world, and it's also a very valuable way to understand elections and to make sense of what the electorate wants.

Rick Valelly: Ellen.

Ellen: Hi, I have a question about [inaudible 01:23:50] which is a little bit tangential, but I think relevant to the fact that this matters. So what gets people out to vote? The college student aged population.

Patrick Egan: So you know there's been a lot of political science about this recently and one thing that's really kind of revolutionized how political scientists study political behavior is experiments. And just like a science experiment in where you've got a control group that gets a sugar pill and a treatment group that gets the drug and you see if the treatment group gets better compared to the control group. With political science experiments, you give treated individuals some kind of treatment that in this case might lead them to be more likely to vote compared to the control group which typically gets nothing.

Patrick Egan: Here we're finding a number of regularities about what leads people to vote and some of it is just really boring. So it's like things like, it turns out when people show up at your door and knock and do it kind of persistently, people tend to turn out to vote a little bit more frequently than the people who didn't get anybody knocking on their door.

Frances Lee: That's the best finding in that whole [crosstalk 01:25:04].

Patrick Egan: I know exactly. It's a very consistent finding. Yes, exactly. It's now been shown over and over again, right? And it's super boring. It's super obvious and it's super boring, but there it is.

Frances Lee: But in person works?

Patrick Egan: In person works. That's right. So there's one more interesting, I think, especially for all of you who are kind of more plugged into developments in social media and networking which is there is now starting to be some evidence, again from experiments, that knowing that people in your immediate social network are voting leads you to vote. So there's a very cool experiment that was done on Facebook. Now I know none of you are on Facebook anymore, right? That's like my generation.

Patrick Egan: But it was on Facebook and so what happened was that people, one of the things you can do on Facebook is say I voted, right? Like you can kind of put up a sticker or whatever. So the experiment was that for some people who did that, it was broadcast to their wider social network more frequently than the control group which was not really broadcast as much. And sure enough, by tracking down voter registration records, they were able to figure out that people who got this message that their friends had voted were more likely to turn out at the polls as a result.

Patrick Egan: So I think the more interesting work, the door knocking stuff is important, I think it's a little obvious and I'm kind of over it, sort of bored by it, but the interesting [inaudible 01:26:22] is now the social networking stuff which is to sort of say is there ways to use these now very popular and obviously prevalent social networking devices and networks to turn out the vote and I think that has a fair amount of promise.

Frances Lee: Just one other point on this [inaudible 01:26:38] is that none of this works very well.

Patrick Egan: Right. That's exactly right. We're not trying to [inaudible 01:26:42] major [inaudible 01:26:41].

Frances Lee: We don't really know how to get people to turn out to vote. These interventions make very modest differences.

Rick Valelly: [Joey 01:26:53].

Joey: I just wanted to ask if you foresaw any routes towards more bipartisan agreement in the future especially considering the unpopularity of the President at the time? I mean it seems that the Supreme Court decisions and things like that. Maybe we won't see [deviation and pride 01:27:09] regardless of the outcome of the midterm elections, but I just wanted to get both your thoughts on that.

Frances Lee: So one place where I see more possibility of bipartisan agreement in the Trump era is on foreign policies and that's one of the most striking examples of bipartisanship from 2017 which was a very partisan year, was the passage of the Russia Sanctions bill on big bipartisan majorities. That I think that there's still a bipartisan consensus on foreign policy and Trump is mostly outside of it and so I see more possibility there.

Frances Lee: On domestic policy, I don't see a lot of paths forward in the near term. Possibly if Democrats retake control of the House, they might cooperate with the President on infrastructure, but that would require the President to move quite a long way from what the administration's initial proposals have looked like on. I mean nothing may have never gotten so far as an actual bill drafted, but the present position on infrastructure is very far from what Democrats would like to see, but I could see some possibility there.

Frances Lee: I guess there's an open question as to what would happen on the trade agreement that's been negotiated. It's not a major shift from NAFTA, but it may be if the President were able to get bipartisan majorities on that though that would be a hard [inaudible 01:28:43], but I could see as one possibility there, but beyond those areas it's hard for me to see where bipartisanship is likely to occur in the near term.

Patrick Egan: And I would just echo what Frances said in terms of competition which is look, November 7th the whole country shifts toward thinking about 2020. Okay? We're immediately in a presidential cycle. If Nancy Pelosi is the Speaker of the House, she doesn't want Trump to win reelection and that's her number one goal. Her number one goal is not to pass legislation. Her number one goal is to keep this guy from getting reelected. And so we are immediately in the environment that Frances has documented and researched and I agree with you that there are very, very few opportunities for bipartisan agreement in this era.

Rick Valelly: There's actually a long article by Bob Kuttner in the American Prospect in which he lays out the whole point of the next Congress is to get Elizabeth Warren elected.

Patrick Egan: Interesting. Yeah, interesting.

Rick Valelly: Yes.

Audience: Can you think of any specific structural changes that you could make potentially with a constitutional amendment that would reduce partisan competition and increase potential for bipartisan cooperation?

Frances Lee: There are a lot of proposals actively being floated about moving towards proportional representation for Congress. That wouldn't even require a constitutional amendment. The Constitution specifies the process by which states get their apportionment, it's not been done in a highly technical way, but says that they have to be based on a census every 10 years. You know, how this is apportioned based on the population, but the technical process by which a reapportionment happens, that is also statutory.

Frances Lee: So within that there's room for reform. We could have a Congress elected with proportional representation. There's nothing that stands in the way of electing our legislators in the way that most members of representative bodies around the world are represented the forms of proportional representation. That is feasible without even a constitutional amendment, but with the prior public support.

Frances Lee: I think Americans just tend to believe, I see this when I teach Introduction to American Government, Americans just tend to believe that the way we do things is basically right. And that structural change is not needed. Constitutional change is not needed. Change to the way we do our elections not needed, but we just need to get better people in there. That tends to be the generality so you have a hard battle to try to convince people that these reforms would be a good idea.

Rick Valelly: It turns out that, a couple of times I've given a public presentation and there's been Australians in the audience who raised his or her hand and announces right away, you actually don't have a democracy here in the United States because you don't have proportional representation. [Kelly 01:31:40], you have a question.

Kelly: Sure. I'll follow up on that thought at dinner, but how would you as political scientists sort out this problem. Frances, you said that you very rightly point out that this closeness of the conversation in the last roughly 35 years or so, the changeover in office, the closeness of elections has consequences for behavior. Alex Azar said that's the reason we have this wave of disenfranchisement that it's only when the elections are that close that it even begins to pay off to try to do that kind of stuff, but your argument is different from one that would say that a lot of what we're seeing is because the parties have, over this same period of time, become increasingly homogeneous that is Republicans losing the Rust Belt and the Northeast became more homogeneously conservative and got some of those Southern former Democrats.

Kelly: The Democrats also lost the South, became more homogeneous and you have a situation in Congress where the least liberal Democrat is still to the left of the most liberal Republican and that's been completely separated now for, I don't know, eight years or so. So how do you determine whether it's the closeness of the races, these highly competitive races or polarization and greater homogenization of the parties that accounts for some of the behavior in Congress?

Frances Lee: It's a great question and it's one that I cannot satisfactorily answer in that I cannot apportion the variance, but I can say this share of our polarization is a result of the homogenization of party constituencies and the fact that Republicans are consistently more conservative than are Democrats today versus what share of our polarization, our party conflict in Congress is a function of the close competition. I can't say. What I want to argue for is a [vote 01:34:08] and perspective which is I think we haven't taken enough account of how close competition affects political incentives, but that I think needs to be something we engage with more and enlist political scientists to understand our politics.

Frances Lee: I'm going to also say that I think we overrate the extent to which the parties in Congress are homogenous especially the majority party. When a party gets a majority, it tends to entail some additional diversity inside the party. That's how they got their majority. The minority parties are more cohesive. Democrats are basically reduced now down to their urban districts, but if they gained the majority, they will gain it by electing a lot of Democrats who won't support Nancy Pelosi for speaker and who will represent constituencies that are not part of the party's base.

Frances Lee: And when you can see this is the difficulty majority parties have in the cases where we have seen a unified government over the past 35 years, actually delivering on their agenda items. So they all found some big agenda items. They campaigned on, that they said they were going to do, but when they got the power, they couldn't agree. So we saw this in 2017 on repeal and replace Obamacare. You saw that with Social Security privatization under George W. Bush when he had unified government. You saw a flat cap in trade under Democrats that was their big failure in [inaudible 01:35:40]. Clinton Health Care Reform. So all the cases of unified government we can point to in recent years see a big, high profile implosion of the majority party.

Frances Lee: That tells me that they're not as unified as they look when we see in the roll call voting where it looks like Democrats are all uniformly liberal and Republicans uniformly conservative. On the critical test of a party's strength, they haven't been able to get it together on key things.

Rick Valelly: Why don't we stop because it's dinner time? Please give a big hand to [inaudible 01:36:18].