October 2020 DC Lunch with Michael Barone

Washington, DC Lawyers Chapter

Event Video

Listen & Download

On October 23, 2020, The Federalist Society's Washington DC Lawyers Chapter held a virtual luncheon featuring Michael Barone, Senior Political Analyst at the Washington Examiner.  Mr. Barone spoke on the 2020 Elections. 


  • Michael Barone, Washington Examiner
  • Introduction: Reginald J. Brown, WilmerHale


As always, the Federalist Society takes no position on particular legal or public policy issues; all expressions of opinion are those of the speakers.

Event Transcript



Reginald J. Brown:  Hello. Welcome to this month's D.C. Federalist Society Chapter Virtual Lunch. My name is Reg Brown, and I am pleased to serve as the President of our chapter. I trust that you've all ordered Chinese takeout and are settled in for a great presentation and discussion consistent with the finest traditions of our organization. I am awaiting my order of Kung Pao chicken and Szechuan beans, which should be arriving by drone delivery any minute.


      I've received a last-minute note from Jeffrey Toobin, who says that he's otherwise occupied and will not be able to join us today, which is too bad.


      We're pleased to be joined today by the distinguished lawyer, historian, and author Michael Barone, who literally wrote the book on American elections and political activity. That book is called The Almanac of American Politics and has been published consistently every other year from 1972 through 2020. The Almanac is chock-full of amazing data and is now nearly 2,000 pages in length. In connection with his work on The Almanac, I'm told that Mr. Barone has traveled to all 50 states and all 435 congressional districts. I'd like to know whether he's also made it to Guam and Puerto Rico as well or if they're on his list.


      Mr. Barone is a graduate of Harvard University and the Yale Law School, where he was editor of the Yale Law Review, and following law school, clerk for Judge Wade McCree on the Sixth Circuit. Mr. Barone has received the Bradley Prize from the Lynde and Harry Bradley Foundation and is also a recipient of the Barbara Olson Award and the Carey McWilliams Award from the American Political Science Association.


      Those of you who are members of our beloved society know that we do not take positions as an organization on political issues, but this being D.C., the set of issues around elections are of great interest for a whole host of reasons. There are fascinating constitutional and statutory issues tied to campaigns, campaign finance, election results, and the mechanics of the Electoral College, for instance.


      Likewise, the results of this year's election may have extraordinary implications for the courts, including possibly the size and membership of the Supreme Court in the years to come. I'm sure that all of these issues will come up today, and whether you are a conservative, a libertarian, a progressive, or something in between, I know that you will find our guest insightful and informative.


      The format that we're going to follow today will start with a presentation from Mr. Barone, after which you are welcome to put questions into the chat function, and I will try to faithfully recount those questions to Mr. Barone.


      So with that, we're all excited to hear from our guest. And I'll turn it over to you, Michael.


Michael Barone:  Well, thank you very much, Reg. It's very nice to be with you giving my quadrennial pre-election talk to The Federalist Society. As you note, among my credentials in life, I am a graduate of the Yale Law School. I like to say that it's the next best thing to no law school at all. And I have a -- I can claim to have a professional career in the law. I was a law clerk for two years to Judge Wade H. McCree, Jr. of the United States Court of Appeals for the Sixth Circuit at a time when the Sixth Circuit was a very collegial court. I also engaged, to some extent, in the private practice of law to the extent of four billable hours for which I was paid for two. So consequently, I moved on to other things.


      I like to say that I've moved—in my professional life—from one career to another, each successive one, which tends to pay less and to have a lower degree of integrity and intellectual honesty than the one before. So I started off in the law at a very high level, obviously, by those criteria. Then I took a small step down to political consulting, and then rather a steeper plunge into journalism. And the only thing now left for me is academia. I'm not at present a recipient of any offers to any colleges or universities at all.


      As I recall, when I spoke to this group back in 2008, I began by saying, "You can't win 'em all." I don't think I'd began this way four years ago in 2016, but I do begin today by acknowledging that the president, who has probably appointed more Federalist Society members to life federal judgeships than any other president in history, is trailing the noted legal scholar who was scored 16 years chairman of the Senate Judiciary Committee, Joe Biden, by a 7.9 percent margin in the RealClearPolitics average of recent polls. That's down from ten percent ten days ago, but it's still well behind.


      And if we were interpreting these poll numbers without the knowledge of the 2016 election before us, I think that we might very well just say, "Well, this election is over. With 11 days to go, Joe Biden has a lead popular vote outside the margin of error," and so forth. It is larger than the last poll, in 1948, which showed Thomas Dewey 5.0 percent of Harry Truman in the last Gallup poll conducted in 1948. That was a poll that was conducted between 9 and 19 days before Election Day. They figured it was all over, and there was no reason to go out and do what you had to do then in polling, which was conduct door-to-door interviews and then send the interview responses back to the head office by post office, at which point they would be hand-tallied by clerks who would put four vertical lines and one horizontal line to tally each five votes. It was a time-consuming thing, and they didn't figure they had enough time to do it. And of course, that resulted in one of the biggest upsets in the history of polling, which now goes back to 1935—85 years ago.


      But I see two reasons for caution about overconfident extrapolations from those poll results to who is going to be elected president in the result when people will be voting on Election Day, November 3rd, or when the election is decided when the votes are counted at some subsequent date.


      One reason for caution is the 2016 result. You go back then, and you see that Donald Trump won that election despite Hillary Clinton's consistent poll leadings. We—those of us who watched this on television—can remember seeing all those faces of people preparing to celebrate Hillary Clinton's victory at the Javits Center, on the west side of Manhattan, with its glass ceiling. They weren't going to go break the glass ceiling, literally, but the idea was perfect, where it would be a metaphorical breaking of the glass ceiling. They were pretty confident it was going to happen. And sometime between 9:00 and 10:00 Eastern Time, the mood changed and became very much darker. And I think even those of us who were not in favor of Hillary Clinton's candidacy, but who had spent some significant number of election nights feeling very unhappy about some of the results, had to sympathize with the people at the Javits Center then. They had a very unhappy surprise from their point of view. And this was something that wasn't anticipated.


      And so consequently, I think that Hillary Clinton's lead over Donald Trump in the 2016 polls was slightly less in the national polls than what we've got for Joe Biden now. But it was still -- looked to be a significant lead, and in fact, she did win the popular vote.


      Which leads me to the second reason for caution, and that's summed up in the word "Californian". One of the things I'd like to do is put things in historical perspective. And from the 1820s, when, as you may recall, John Marshall was the chief justice -- John Marshall was appointed, by the way, by President John Adams about ten days before his term in office was ending, after he was defeated by Thomas Jefferson and confirmed by a senate, which was going out of office at that same March 4th date when the Federalists would no longer have a majority in the Senate. And he lasted for 34 years in the Supreme Court, and nobody seems to have made the argument, at least successfully, that he was an illegitimate chief justice. To certain people, he was a consequential one.


      But from the 1820s until the early years of this century, the nation's largest state -- and that was New York, from the 1820s until 1963—California since 1963. The nation's largest state voted pretty much close to the national average, within five percent of the national average, in every election. New York was not an overwhelmingly Democratic—or an overwhelmingly Republican state—during the 140 years it was the largest state. California was very close to the national average for a long time. That's no longer true. California's now an outlier, the number two Democratic state in 2016, voting just about 14 points more Democratic than the national average, ceded in Democratic percentage only by Hawaii, the home of the legal scholar, Senator Mazie Hirono. And California is responsible for all of Hillary Clinton's popular vote plurality and more. If you take California out of the popular vote percentage, you see that Donald Trump is ahead in the popular vote by a 48 to 46 margin, reversing Hillary Clinton's margin that happens if you include California.


      And so California weights any poll that you see, basically—national poll. There are about six points for the Democratic nominee from California. There's only about two or three points for Donald Trump or the Republican nominee from California. And the results that are going to make a difference, beyond California's 55 electoral votes, are those coming from other states.


      So I'd like to take a look, at this point, at the poll numbers in the six top target states in this election. And they were all for Donald Trump four years ago, many of them contrary to expectations: Florida, North Carolina, Pennsylvania, Michigan, Wisconsin, Arizona.


      Currently, Joe Biden leads the collective results from those six states by about 4 percent, as Hillary Clinton did 11 days out in the 2016 election. His lead is three points or less than Florida, North Carolina, and Arizona. That's clearly within the margin of error and cannot be considered, in my judgment, safe for either Biden—the candidate who's leading in those polls—or for Trump; five points in Pennsylvania and Wisconsin; eight points in my home state of Michigan. You can cherry-pick and find some polls in each of those states that are better for Donald Trump. You could find some polls that are better for Joe Biden. But in terms of getting an approximate idea of where the race stands, I think it's better to take an aggregate form to get the average of the polls as you can get RealClearPolitics and Nate Silver's fivethirtyeight.com thing.


      And so what I would say for that result, is that, despite the national polls showing about an eight-point lead for Joe Biden, Donald Trump is within range. But in order to win, he's got to do something that looks like drawing to an inside straight in poker. The good news for Donald Trump is that that’s possible. That sort of thing happened four years ago. The bad news for him is, of course, that the odds are against it. So consequently, those odds are against it.


      So basically, polls -- and also, polls don't always -- aren't always tilted towards the same party in election after election. 2016 national polls showed Hillary Clinton leading the popular vote by three points. She led it by two points. It's very close, given limitations of polls. The polls in some of these states, like Wisconsin and Michigan, were either few and far between or far off the final result, indicating either a late move, an unusual turnout, or the fact that the polls just didn't hit target and were wrong. So the polls tilted towards the Democrats in 2016. You can go back four years, 2012, they tilted towards the Republicans. Pollsters try to learn from their errors, and sometimes they try to compensate for them. Sometimes they overcompensate—go too far in the other direction.


      In 2016, many pollsters undersampled white non-college voters—non-college graduates. They didn't look at them as a separate category or one that they ought to weight their responses for because, in the past, white non-college voters basically, overall, tended to vote like white college graduates overall. There weren't much differences. But basically, going back a few cycles, you can see this movement started, but it became a kind of crescendo in 2016. Non-college whites [inaudible 16:25], so have been trending Republican, especially in 2016. White college graduates have been trending Democratic. We saw that in 2016 and more in 2018 in the off-year elections.


      This time, many pollsters have taken care to try and adjust their sample or to weight their responses in order to reflect the percentage that white non-college voters have in the electorate as a whole. A percentage that, by the way, has been larger than what is recorded in exit polls that the consortium hired by the television networks and other news organizations has indicated.


      Other pollsters are not adjusting for white college status and are at risk, in my judgment, of missing changes here. So basically, non-college white votes were for Trump, and he did much better with that group than, basically, any Republican has done before, going back at least to Ronald Reagan in 1984. And in some cases, he's done -- like western Pennsylvania, he ran ahead of Ronald Reagan among that group. And so, he won 100 electoral votes that Mitt Romney lost in 2012 in Florida, Pennsylvania, Ohio, Michigan, Wisconsin, Iowa, and that one vote in the 2nd congressional district of Maine. He came within one, two, or three points in three states with 20 more electoral votes: New Hampshire, Minnesota, and Nevada.


      And the question now is -- overall national polls are showing him doing worse among white non-college voters, including maybe particularly among women. There's some danger when you look at overall polls to try and slice up the electorate—the sample size is so small, that the error margin is so large—that you can't really make assumptions about it even as he's running slightly better, it seems, than four years before, among black American voters, among voters classified as Hispanic, despite the charges that are constantly made—usually without any supporting evidence—that Trump is a racist. He seems to be doing somewhat better with those voters.


      The key state, in my view, in many ways, is Pennsylvania. And it's interesting that—for those of you who watched the debate last night—President Trump seemed to be very careful about informing the voters—after Vice President Biden indicated that he wanted to phase out fracking and fossil fuels and so forth—he said, "Pay attention, Pennsylvania." He was aiming at those electoral votes.


      So basically, my overall assumption is there's still a chance for Donald Trump. We've had some -- 45 million votes have already been cast by people voting early. The voting, by people voting by mail, is heavily tilted towards Democratic voters because, as we've seen in response to the Coronavirus, Democratic voters and Democratic politicians have been more risk-averse and more unwilling to participate in crowd events. You've had the teacher's unions insisting that schools not be reopened despite the overwhelming evidence that there’s no significant danger of deadly disease for either students or teachers.


      And so, you see a lot of Democratic voters that wanted to vote by mail this year. That's going to be something of a problem in counting the votes in some of these states which are not used to having a large number of votes delivered by mail. And I believe it's the case in Pennsylvania and Michigan, for example, that election group people are -- clerks are not allowed to count those votes until Election Day. Some other states, they're able to count them early.


      We've had some decisions in which the Supreme Court has actually taken a part in some cases where typically, it's Democrats who want to extend the time during which votes can be submitted or ballots. Republicans have tended to say, "No. We've got to enforce the rules that are in force there." In most of those cases, the Democrats have not prevailed, although, in Pennsylvania, with a decision of their Supreme Court and an evenly divided U.S. Supreme Court, they have.


      Let me address one other set of elections that I think is particularly important to Federalist Society members or to many of you, and that is the Senate races. It's a subject of more than minimal interest because we currently have a 53-47 Republican senate, which seems clearly will confirm its third successive Supreme Court justice in the last four years in Amy Coney Barrett on Monday. The Republicans seem sure to pick up one seat by beating the Democrat Jones in Alabama—heavily Republican state—where he was able to win against a really whacko candidate. And in Michigan, candidate John James, the Republican nominee, is not far behind Gary Peters, who, for some period of time, has been the U.S. Senator with the lowest substantive name I.D. in his home state -- but concerned danger of losing many other seats. Cory Gardner and Martha McSally seem to be far behind in Colorado and Arizona. Other Republican incumbents trailing within the margin of error include Joni Ernst in Iowa; Thom Tillis, North Carolina; Susan Collins in Maine. Republicans have scarily narrow leads and seats in South Carolina, Georgia, Kansas, and Montana, all states that Donald Trump carried four years before, and at least three of which he's expected to carry easily, Georgia being the exception.


      This time, I think there's a possible upset. If I had to identify an upset at this point, I might pick John James in Michigan or maybe Jason Lewis, the Republican challenger to Senator Tina Smith in Minnesota. There was one recent poll showed that a close race. We're waiting to see if it's going to be -- there are going to be others.


      In the House, it seems pretty clear that Nancy Pelosi's Democrats will retain and marginally increase their majority, which is currently at 232-203. That means retaining most of their 2018 gains, which mostly have been in high-education districts. I'm going to start calling them high-credential voters—the kind of places where most Federalist Society members tend to live, I suspect. I always say, "If you want to see areas that have turned against Donald Trump, go get a directory of Harvard Law School alumni," to take one of a number of examples—for Yale Law School alumni, Columbia, Chicago, whatever school you want, highly selective law school. The places where those people tend to live are the places that have been historically, in many cases, Republican with high-income voters tilting now towards the Democrats and repelled by the persona or perhaps by the policies of Donald Trump. He seems to have been a too heavy load for them even to carry. And this is part of a trend, which Trump did not begin, but which has been emphasized under Trump, of the Republicans becoming a more down-scale party and, at the same time, one with more racial diversity, as Trump seems to be attracting somewhat more black and Hispanic voters but has been losing white college graduates.


      Let me conclude with a note on redistricting, which is something I've been following for a period of time. As Reg indicated, I've been writing. I started The Almanac of American Politics. I'm a founding and long-time coauthor since the first edition appeared some 49 years ago. And I've been -- as a consequence of that, I have been following congressional redistricting because the book covers each state and each of the 435 congressional districts, all of which I've been in. I've followed the redistricting cycles going up before and including the Supreme Court's equal population decisions in 1963 and '64 and in the multiple cycles following. And over the past several years -- we've been hearing much talk today among people that consider themselves to be kind of "good government advocates" about how computers have enabled partisan redistricting to totally control the majorities that led Congress—led the House of Representatives and legislatures -- how this is a great threat to democracy, how it's undermining the ability of people to be represented, how redistrictors can carry all before them.


      It's interesting to me, however, that in the cycles following the 1960, 1970, and 1980 censuses, which were cycles in which the Democrats tended to have the advantage in redistricting -- Democrats then controlled most of the state legislatures. They had good electoral cycles going in 1970 and '80, which continued their controls. I can remember staying up late at night with Congressman Phil Burton of San Francisco. Many of you may not remember, but Phil used to drink. His drink of choice was vodka without ice in tumblers, and he'd go in for multiple of this getting more loosed about the redistricting situation, not only in California—where he had a splendid district gerrymander in the '70s and '80s—but all over the country. And I don't remember all these great and good reformers being very upset about redistricting when it was the Democrats who were controlling redistricting and which were using it in many parts to maintain their long-standing congressional majority in the House of Representatives. That started to change a little bit in the 1990 census cycle when, politically, redistricting was a kind of a wash between the two political parties and when the Republicans in 1994, for the first time in 40 years, won a majority in the House of Representatives. The Democrats had a 40-period of control. The second longest period of control in the history of the Congress is only 16 years. Forty years was a real long time.


      The lamentations about the evils of redistricting have been crescendoed in the cycles following the 2000 and 2010 cycles, and I think, not coincidentally, that's because redistricting tended to favor the Republicans at that point. They had majorities in many legislatures, which they had never had before. The part of that was being able to win in more down-scale districts, as well as up-scale districts. They tended then to draw plans that were favorable to them.


      But I have thought as an aficionado of redistricting -- and I've even done a little partisan redistricting myself. When I was coauthoring this Almanac book, of course, every ten years or so, I would look and try and see what the party in control of redistricting was going to do to maximize their effects. So I've gone through them—my brain functions—of imagining myself to be a Republican districtor, imagining myself to be a Democratic districtor to try and figure out what they were going to do.


      And my overall conclusion is that redistricting helps you with the margins. It helps a party. It does not help you win everything. If you look at the 1990s cycle, the 2000s cycle, the 2010 cycle, what you see is that neither party was able to win all five congressional House of Representative elections when a majority threw off five of those cycles. The Republicans lost in '92, in 2006, and in 2008, and 2018, and they look, still, like they're going to lose in 2020.


      The Republicans -- this cycle, I think, we're going to see the advantage swing away from the Republicans. But I think also we will not be overwhelmingly advantageous to the Democrats. We've now got some -- some states have imposed, by referendum or legislatures, supposedly, non-partisan redistricting commissions. My observation is that these always get successfully gamed by Democrats. They say, "Well, we're going to have this neutral political scientist take part in this." Political scientists across the country are about 98 percent Democrats. Your neutral political scientist is going to tend to favor a redistricting plan that looks a lot like what Democrats are going to want most of the time.


      And you've got limits by referendum imposed. For example, in Florida -- a large state, which is set to gain a couple congressional seats, which the Republicans have had the governorship of both houses of Congress. You've also got fewer states where either party has, going into it, fewer states with large numbers of House districts that -- where one party has the governorship and the two houses of the legislature. Or in North Carolina, the governor does not have the legislative veto on the redistricting bill.


      But any case, I think that looking ahead, that means that the districting for state legislatures and for congressional districts will be less favorable to Republicans than the past, but I think that that is modulated by the fact that the Democrats have a disadvantage in any equal population districting plan because they tend to have their voters concentrated in relatively few places. In central cities, some sympathetic suburbs, university towns, you find lots of places. You find areas as large as congressional districts voting 80 to 90 percent Democratic. You don't find 80/90 percent Republican congressional districts anywhere. Republican voters are spread more evenly around the rest of the country and around the rest of large states that elect large congressional delegations. So there's a -- it's a little harder for Democrats to eek partisan advantage out of redistricting than it is for Republicans.


      And so, let me just make one final point on redistricting. I can recall from the 1950s that Republicans had a theory. It was put forward in the Michigan constitution, adopted by the voters in 1963, supported by Governor George Romney, that people in rural popularities ought to be overrepresented—ought to get more representation than they'd be entitled to on population. In the 2020s, maybe we'll see some liberals and Democrats roll out theories that people in the densely populated areas ought to be overrepresented. You know they got more problems with government. They're more disadvantaged. They're more likely to be subject to racial discrimination or whatever.


      The Supreme Court in the 1960s rejected the arguments for overrepresenting lightly populated areas. I expect the Supreme Court in the next decade to reject the arguments for overrepresentation of densely populated areas and perhaps to discourage people from making such arguments as people at all.


      But if the last three Supreme Court justices had been appointed by—not by President Donald Trump—but by President Hillary Clinton, maybe that argument would get some traction. Elections have consequences. So I will just leave this bit of speculation about possible legal issues for whatever Federalist Society members might like to take from it, and perhaps, depending if the election turns out as most people expect, with some consolation for the likely results of the election next Tuesday, if you find those disappointing.


      So let me conclude at that point, and I can go to questions and answers.


Reginald J. Brown:  That sounded terrific, Michael. Thank you. Tour de force as usual.


      The chat room is open, so if you do have questions, please go ahead and get them in the queue. I will start with those, but let me just get a summary answer from you just in terms of the odds. What do you think the odds are that we're going to know the winner by 11:00 on Tuesday night? This is very important to those of us who need our sleep.


Michael Barone:  Well, to need your sleep -- for those of you who need your sleep, my advice is, take a long nap during the day --


Reginald J. Brown:  Take a nap.


Michael Barone: -- on November 3rd.


      Look, Florida tends to be tabulated pretty early, and one of the things that alerted many of us to what was happening in 2016 was the report from me that was relayed and got relayed over the internet and widely circulated from a Democratic consultant in Tallahassee, Florida, who said, "Hey. We got problems here. We're carrying the large"—we, the Democrats -- "carrying the large counties in Florida by about what we need to carry them district-wide" if you extrapolate from the results in 2004, 2008, 2012, in all of which Florida was a very close election. And of course, 2000, the official count was 537. I think the Democrats stole another 500 votes, but that's kind of a moot point now 20 years later. But the fact is he said, "Look at the smaller counties, the places where"—he didn't say this, but I will say it—where you had a higher number of percentage of non-college graduate white voters, "we're getting clocked by worse margins than we've ever seen before."


      And that was the difference that enabled Donald Trump to carry Florida by one percent and was a symbol that, to many of us also who know that, much of Florida—many of those smaller counties are filled with people who come from different parts of the Northeast, the Midwest, the South—there was going to be problems for the Democrats as well in places like out-state Michigan; northern Wisconsin; Iowa, generally; non-metropolitan Ohio; western Pennsylvania because that's where these same people from Florida were coming from.


      So I'd say this: if it's a determined -- if Donald Trump is losing Florida, I don't think we'll have to wait long for a result. I think it's going to be hard for him to win with dropping the 29 electoral votes from what is now the third-largest state in the nation and so forth. It'll be possible, but it really gets to be unlikely. And also because those Florida voters have something in common with other people.


      If Florida looks like it's going to Donald Trump, then I think you've got a race that may go on for some time because of the greater propensity of Democratic voters to vote by mail—the results of which may not come in immediately on election night—and of Republican voters to vote in person on Election Day because they are less risk-averse about being in public places.


Reginald J. Brown: Thanks very much. Got a couple of questions --


Michael Barone:  My thought is, you’re going to have Republicans -- you're going to have Trump leads in many states that may or may not be eroded is how it’s coming [inaudible 38:31].


Reginald J. Brown:  That's the so-called red haze phenomenon that Trump will look like he's leading on Election Day, and that's a haze, and it will eventually blow off as a blue wave emerges. Do you put much credibility in that theory?


Michael Barone:  Well, I think something like that could happen if -- in many of these states because it will take some time to count those votes. And I can imagine that some of these counts are going to be pretty seriously contested. I mean, we had lawyers going to the 67 counties of Florida in 2000. We may see lawyers flooding into a lot of places.


      I look back to the Washington State elections in 2004, I think it was—Dino Rossi versus Christine Gregoire. It seemed like the Democrats had a back closet in King County in Seattle where they had a whole bunch of extra votes they kept finding till they got enough to win the election. I've always been suspicious about that result, although, essentially -- because Ms. Gregoire was declared the winner and served the four years of the term. But I think you may find some real controversy about that.


      Yeah. I can't hear you, Reg.


Dean Reuter:  This is Dean Reuter. While Reg is working out his technical difficulties, Michael --


Michael Barone:  Yeah.


Dean Reuter:  -- let me go to the queue. A question from Ken Masugi, "The Trafalgar Group's polling was accurate," he says, "uniquely so in 2016. If that's the case, why not emphasize their polling this year? Do you find their innovations in measuring polling regarding shy Trump voters as problematic?" And this is echoed in a question from another audience member about, not people being dishonest, but being reluctant to identify themselves as Trump voters in polls.


Michael Barone:  Well, I worked for Peter Hart, the Democratic pollster, in 1974 and '81. And my training there, regardless of party affiliation, was to be somewhat suspicious of new techniques. Robert Cahaly, the pollster for the Trafalgar Group, says that he has a series of measurements, including asking people what their neighbors are going to vote for, that enables him to identify shy Trump voters or people that are actually for Trump but don't want to tell pollsters that they're for Trump. Other pollsters will tell you that there are no significant numbers of those people.


      I note that Trafalgar got figures for Michigan and Pennsylvania in 2016 that turned out to be pretty close to the final results. I would want to know more about those polls before I'd state publicly that I either think that the techniques he uses are justifiable or that they are not. I look at them with interest. I do not want to emphasize them over all other polls that are taking place. But I note that polling techniques that work in one cycle and tend to come close to final results don't necessarily do the same thing in the next cycle. That's been the history of polling throughout, so I'll be cautious on that.


      On the other hand, there's been some recent polling about -- particularly since the emergence of the Black Lives Matter group and people like this fraudster, Ibram X. Kendi, claiming that everybody is a racist. If you say you're not a racist, you're a racist. And if you say you're a racist, you're a racist, and so all white people are racist. They've got numbers that show 56 percent of some subgroups say that people are less willing to say, in life generally, that they hold certain kinds of beliefs, and that that's much more true of people that identify as conservatives than of people that identify as liberals. So maybe that's a fact. Maybe there's even more shy Trump voters, as it were, than they think so in the past.


      The shy thing, by the way, comes from shy -- because Tory voters in England -- particularly in the 1992 election, when the conservative party won after almost everybody expected the Labor Party to win. But in the next election, in 1997, the Labor Party won by a huge margin, so the "shy Tories" may or may not have still been shy, but there weren't enough of them.


      So I'm undecided about that. I'm watching with interest.


Dean Reuter:  Do we have Reg Brown back? Reg, are you back with us? He might be trying to reconnect. Reg, are you with us? We still can't hear Reg Brown.


      So another question. This one from Mario Loyola, Michael. He has the impression that Trump's base of support is much more committed now even than it was in 2016. How do the polls account for intensity of feeling and the impact of turnout, and how is that turnout race looking just now?


Michael Barone:  Well, we try to test intensity of turnout. And you've got a number of signs that Republican turnout may be great. You have party registrations. Republicans have been narrowing Democratic advantages and party registrations in states like Pennsylvania and Florida by significant amounts. Partly that reflects a willingness of Republican Party workers and party leaders to go out in the field and actually see people and the risk averseness of Democrats who feel that the risk from disease is too great to do that. That's an advantage that may be conferred.


      I think you have that feeling -- the kind of people that Mario and I talk to and listen among the conservative movements, you find a number of conservative voices being more pro-Trump than four years ago. I think of the radio broadcaster and blogger -- Ben Shapiro, for example, has a very wide audience. Ben has been making the point that he was neutral four years ago. He was skeptical about Trump. He's a solid Trump supporter this time, although criticizes him on some points.


      The overall polling, however, suggests that Trump is not getting the extraordinarily large percentage of non-college white voters that he was getting last time. Look at those very good maps that the New York Times has put out, the interactive maps, showing the percentage -- who carries which cities and townships. Look at Pennsylvania, Ohio, Michigan, Wisconsin, Iowa in 2016. Hillary Clinton is carrying just the central cities and the university towns. All these other areas—which you can see in previous elections, there were many areas that voted: rural areas, small towns, suburbs that voted Democratic—were all Donald Trump. Those were unusually high numbers.


      The political analyst, Ron Brownstein, for example, said, "The increasing number of Hispanics and blacks are going to make this country Democratic forever because whites can't vote any more heavily for Republicans than they did for Mitt Romney or for John McCain or for George W. Bush." Well, the whites, non-college whites, did vote more heavily for Donald Trump than they voted for those three previous Republican nominees. But that doesn't mean they necessarily will this time, and there's polling evidence that suggests they aren't prepared to do so yet. Will Donald Trump’s performance in the debate more positively for him? Eh, maybe. So you know, we'll see. But I see conflicting evidence here.


Dean Reuter:  Reg, are you back with us?


Reginald J. Brown:  Michael, this is Reg. I think I may be back --


Dean Reuter:  We can hear you, Reg.


Reginald J. Brown:  -- at this point.


Michael Barone:  Excellent.


Reginald J. Brown:  Great. I apologize for that.


      If you didn't cover it already, can you speak to some of the effects that you think the mass mail-in ballot situation will have?


Michael Barone:  Well, the mail-in ballot situation -- one thing, I think it's going to be problematic for Democrats because when you mail in a ballot, often—and it's different in different states—you have two envelopes, basically. One of them -- the outside envelope, you're supposed to sign your signature so that the counting people can verify your signature against the signature that's on file so that they certify that this is your ballot.


      And then the inner—then they take that envelope off and slough it off when they're counting. And then there's a second one that preserves the -- has no symbol of who you are and preserves the secrecy of the ballot. Some people—who are doing this for the first time—some percentage of them are not going to get this quite right. Will those ballots be counted?


      You will get a certain number of people that may fill in the ballot wrong. We've seen that there are some situations where mail ballots have not gone in with the -- the way that -- through the mail the way they should. Our postal service has a much better record of reliability and things than postal services in many countries, but it's not quite perfect, and it's perhaps somewhat less perfect than it has been historically. So there's some problematic things.


      That's one area where, if I was a Democratic Party official, I would be kind of concerned that the risk averseness of Democratic voters may cost them a few votes [inaudible 49:49].


Reginald J. Brown:  So Michael, you have an extraordinary reputation for being non-partisan or calling it like you see it. Not everyone in the media is perceived that way. And I wondered if you could comment on the impact of the rise of the partisan media and whether you think that that's something that's here to stay or perhaps will change in our lifetime. And if you could also comment on the role of social media in this campaign cycle.


Michael Barone:  Well, I don't know. I would not claim to be totally neutral in political things by any means, but I have been a supporter of Democratic candidates years ago and more supportive these days of Republican candidates. So perhaps that helps me understand a little better than I would otherwise the motivations of people who vote for one side or the other. And I try to put myself in the positions. In The Almanac book, I've always tried to write it as in a neutral way. I think I got better at that as time goes on, and feel that making a positive case for people who have been elected to elective office—whether I'd vote for them or not—is something that I felt a responsibility to do in that venue.


      Are we more partisan than we used to be? Yes, we are. Go back to about 1950. The leading political scientist then, E.E. Schattschneider, and some others said, "You know what we need? We need to really have -- we've got this muddled politics with conservative Democrats from the South, with the liberal Republicans from the North. We need to have one clearly liberal party and one clearly conservative party. That's our dream. That's our prayer."


      Well, their prayers were answered about 40 years later. And in the 1990s and the years following, we find that there aren't many liberal Republicans anymore. There aren't many liberal Democrats. And in a book that came out, I think in 2018, How America’s Political Parties Change (and How They Don’t)—a little book—I kind of explained where the conservative Democrats went and why they became Republicans, and where the liberal Republicans went and why they became Democrats, and how our parties rationalize.


      But they also rationalized, in large part, around cultural issues, around non-economic issues, on issues that relate to the moral basis on which you live or try to live your personal life. It means people feel strongly about those things. And when you bring them into politics, you find people feel strongly. The abortion issue has been one example of that. It's not unique in American history. What was the Civil War about? That was about moral issues and about how people lived their lives and things. And they felt very personal about it, and people went to war over it, literally. So these things inspire strong feelings.


      We haven't had congressmen come over and thrash a senator from the other party within an inch of his life as Preston Brooks did to Charles Sumner in 1857 or 1858. But we clearly have strong feelings. I think that's going to continue. I mean, electoral politics is -- the elections are a zero-sum game. Electoral politics is competitive. It's an adversary process. It's not like elections to be president of the second grade where you're expected to vote for your opponent and all be nice to each other and things. Serious things are at stake in terms of public policy and in terms of values and issues that people feel very strongly about. So I think that's going to continue to be the case.


      We have seen, in these last five years, accelerated by the peculiarities, eccentricities, or uniqueness of Donald Trump—but not solely because of him—some changes in party positions, some changes of what issues are emphasized because the world has changed in some ways. And I don't feel this is the end of democracy. I don't feel that we are an equivalent of Weimar Germany, or that we're dealing with an Adolf Hitler, or that it's appropriate for one side of our political argument to call themselves "The Resistance" as if they're resisting the German occupation of France from 1940 to '44. But some other people disagree.


Reginald J. Brown:  So Michael, given the common notion that Trump has a narrower base of support than maybe your standard Republican candidate or is unattractive to many voters who would otherwise be inclined to support Republican candidates, why do you think he's running one to three points ahead of the Republican Senate candidates in battleground states like Iowa and North Carolina and Arizona if that is, in fact, the case?


Michael Barone:  Well, I think I'm a little puzzled why he's running ahead of Republican Senate candidates because four years ago, in fact, Republican Senate candidates ran a little ahead of Donald Trump: Pat Toomey in Pennsylvania; Ron Johnson in Wisconsin; and so forth—Rob Portman, certainly, in Ohio, and so I think that was different at that time.


      One hypothesis is that it is difficult for a -- it has been difficult for a Republican senator in these past four years to establish a personal profile, an identification with particular issues, some of which may appeal to voters beyond the Republican half of the electorate. To do that -- because Donald Trump is sucking up—and his opponents -- are sucking up all the oxygen in the air. And that's a working hypothesis. We'll see how that works out.


      One of the things I've noticed with senate races is they can surprise. Remember, the Republicans captured a majority of the Senate in 2014. Go back 11 days before that election and see how many people were predicting that that was going to happen or see how closely the polls were indicating. It looked like Mitch McConnell was in a very tight race in Kentucky six years ago in 2014. He ended up winning by a wide margin.


Reginald J. Brown:  Thanks, Michael. I think we'll end on that note that there could be a big surprise. And we will take a nap and, hopefully, be watching you. I don't know what network to plug for you. Will you be on air?


Michael Barone:  I will be [inaudible 57:34].


Reginald J. Brown:  All right. Great. Well, thank you so much for giving us your time today. It was really thoughtful commentary—incisive, as always.


      For those of you in The Society, I wanted to remind you that we are coming up on the first-ever virtual Federalist Society convention. It will be free of charge. Registration is apparently now open on the FedSoc website. We've got very exciting speakers lined up this year for the national convention. I think the media will be very disappointed because they'll be able to watch the whole thing, and if they're not careful, they may learn something. So I tell you about that. Hope that you all will go ahead and register.


      And again, Michael, on behalf of all of us, thank you very much. We owe you a good Chinese lunch, which hopefully we'll be able to make good on in the coming year.


      Thanks, everybody, for your time.