Skip to content
 

What happens to the median voter when the electoral median is at 52/48 rather than 50/50?

Here’s a political science research project for you.

Joe Biden got about 52 or 53% of the two-party vote, which was enough for him to get a pretty close win in the electoral college. As we’ve discussed, 52-48 is a close win by historical or international standards but a reasonably big win in the context of recent U.S. politics, where the Democrats have been getting close to 51% in most national elections for president and congress. I’m not sure how the congressional vote ended up, but I’m guessing it’s not far from 51/49 also.

Here’s the background. From a combination of geography and gerrymandering, Republicans currently have a structural edge in races for president and congress: Democrats need something around 52% of the two-party vote to win, while Republicans can get by with 49% or so. For example, in 2010 the Republicans took back the House of Representatives with 53.5% of the two-party vote, but they maintained control in the next three elections with 49.3%, 52.9%, and 50.6%. The Democrats regained the House in 2018 with 54.4% of the two-party vote.

And it looks like this pattern will continue, mostly because Democrats continue to pile up votes in cities and suburbs and also because redistricting is coming up and Republicans control many key state governments.

And here’s the question. Assuming this does continue, so that Republicans can aim for 49% support knowing that this will give them consistent victories at all levels of national government, while Democrats need at least 52% to have a shot . . . how does this affect politics indirectly, at the level of party positioning?

When it comes to political influence, the effect is clear: as long as the two parties’ vote shares fluctuate in the 50% range, Republicans will be in power for more of the time, which directly addresses who’s running the government but also has indirect effects: if the Republicans are confident that in a 50/50 nation they’ll mostly stay in power, this is a motivation for them to avoid compromise and go for deadlock when Democrats are in charge, on the expectation that if they wait a bit, the default is that they’ll come back and win. (A similar argument held in reverse after 2008 among Democrats who believed that they had a structural demographic advantage.)

But my question here is not about political tactics but rather about position taking. If you’re the Democrats and you know you need to regularly get 52% of the vote, you have to continually go for popular positions in order to get those swing voters. There’s a limit to how much red meat you can throw to your base without scaring the center. Conversely, if all you need is 49%, you have more room to maneuver: you can go for some base-pleasing measures and take the hit among moderates.

There’s also the question of voter turnout. It can be rational, even in the pure vote-getting sense, to push for positions that are popular with the base, because you want that base to turn out to vote. But this should affect both parties, so I don’t think it interferes with my argument above. How much should we expect electoral imbalance to affect party positioning on policy issues?

The research project

So what’s the research project? It’s to formalize the above argument, using election and polling data on specific issues to put numbers on these intuitions.

As indicated by the above title, a first guess would be that, instead of converging to the median voter, the parties would be incentivized to converge to the voter who’s at the 52nd percentile of Republican support.

The 52% point doesn’t sound much different than the 50% point, but, in a highly polarized environment, maybe it is! If 40% of voters are certain Democrats, 40% are certain Republicans, and 20% are in between, then we’re talking about a shift from the median to the 60th percentile of unaffiliated voters. And that’s not nothing.

But, again, such a calculation is a clear oversimplification, given that neither party is anything close to that median. Yes, the are particular issues where one party or the other is close to the median position of Americans, but overall the two parties are well separated ideologically, which of course is a topic of endless study in the past two decades (by myself as well as many others). The point of this post is that, even in a polarized environment, there’s some incentive to appeal to the center, and the current asymmetry of the electoral system at all levels would seem to make this motivation much stronger for Democrats than for Republicans. Which might be one reason why Joe Biden’s talking about compromise but you don’t hear so much of that from the other side.

P.S. As we discussed the other day, neither candidate seemed to make much of a play for the center during the campaign. It seemed to me (just as a casual observer, without having made a study of the candidates’ policy positions and statements) that in 2016 both candidates moved to the center on economic issues. But in 2020 it seemed that Trump and Biden were staying firmly to the right and left, respectively. I guess that’s what you do when you think the voters are polarized and it’s all about turnout.

Relatedly, a correspondent writes:

Florida heavily voted for 15 minimum wage yet went to Trump. Lincoln project tried to get repubs and didnt work. florida voted for trump because of trump, not because of bidens tax plan.

To which I reply: Yeah, sure, but positioning can still work on the margin. Maybe more moderate policy positions could’ve moved Biden from 52.5% to 53% of the two-party vote, but then again he didn’t need it.

P.P.S. Back in his Baseball Abstract days, Bill James once wrote something about the different strategies you’d want if you’re competing in an easy or a tough decision. In the A.L. East in the 1970s, it generally took 95+ wins to reach the playoffs. As an Orioles fan, I remember this! In the A.L. West, 90 wins were often enough to do the trick. Bill James conjectured that if you’re playing in an easier division, it could be rational to go for certain strategies that wouldn’t work in a tougher environment where you might need regular-season 100 wins. He didn’t come to any firm conclusions on the matter, and I’m not really clear how important the competitiveness of the division is, given that it’s not like you can really target your win total. And none of this matters much now that MLB has wild cards.

P.P.P.S. Senator Lindsey Graham is quoted as saying on TV, “If Republicans don’t challenge and change the U.S. election system, there will never be another Republican president elected again,” but it’s hard for me to believe that he really thinks this. As long as the Republican party doesn’t fall apart, I don’t see why they can’t win 48% or even 50% or more in some future presidential races.

It seems nuts for a Republican to advocate that we “challenge and change the U.S. election system,” given the edge it’s currently giving them. In the current political environment, every vote counts, and the winner-take-all aggregation of votes by states and congressional districts is a big benefit to their party.

79 Comments

  1. Jordan Ellenberg says:

    How did I never know you were a fellow Orioles fan?!

    • Andrew says:

      Jordan:

      I’m not much of an Orioles fan now, but back in the day . . .

      Did you ever read Earl Weaver’s autobiography? It has a great passage . . . I don’t remember the exact words, but it’s relating some important game where Brooks Robinson was up at bat, and Earl told him, “Swing away, Brooksie.” Weaver always maintained that it was the players who won the game, not the manager, and that the manager’s job was to get the best out of each of his players. This is a good general principle, I think, and similar to Deming’s views on management.

      Also it was good that when Weaver came back in 1985, his team did really badly. I mean, no, it wasn’t good, it was bad, it made me sad. But from a statistical perspective it was good to be reminded that even though Weaver was famous for being a genius, that wasn’t enough. Just cos you’re a genius, it doesn’t mean your teams will automatically win.

      • jonathan says:

        Huge Earl Weaver fan. He managed prima donas like Palmer, got 4 pitchers to keep their heads together for 20 wins each, understood the value of defensive play, and – hard to say this correctly – he realized the key to success was old-fashioned, that you needed to beat the bad teams, which meant the players had to understand those were the games that mattered because if you went 28-0 against the bad teams, then you only had to be competitive with the best. His teams didnt let down. Used to frustrate me to no end as a Tigers fan. And they got used to playing at a high level as a team, so one guy would go into a slump and the others would step up because they had internalized those roles.

  2. Tom says:

    If the extra 2% is in CA/NY then why is it not really 50/50 for where political capital is spent? Do you have the breakdown between safe states and swing states for where that runup in votes occurs? You mention big runups in cities and suburbs but it also matters in which states those votes lie.

    This gets back to the argument that a popular vote election would be different than an Electoral College election.

    Also, the structural advantage is a feature, not a bug. So isn’t this a case of Chesterton’s Fence? It seems this question should be flipped around to whether politics is working as designed, what incentives should be in place, and why are we deviating from those incentives (if we are).

    • Andrew says:

      Tom:

      The extra 2% is not in any particular states; it’s everywhere. To put it another way, it’s not like the Democrats could push a button and transfer their votes from one place to another. Also, for the purpose of this post we don’t really care about the counterfactual in which the election were held by simply adding up all the votes. My point is that to win the elections as now constituted, the Democrats need something like 52% of the two-party vote, and that will affect the strategies of both parties. Regarding the structural advantage, I don’t think it’s a “feature” or a “bug.” The current imbalance is very large in the context of recent American politics (see here). There’s no reason to think this was part of any design. It is what it is.

      • Joshua says:

        Regarding whether it’s a feature or a bug.

        I run into this a lot from right-wingers: the argument is that by design, we live in a republic not a democracy, where we would be subject to “mob rule.”

        We saw rhetoric along these lines recently from Senator Mike Lee (before he walked it back when it became politically inconveneient).

        As near as I can tell, the thinking is that there is no problem with a structural imbalance that allows for a distinctly minoritarian dominance in power. And objections are sour grape whinning of losers (i.e., Democrats) .

        The question I have about that view is whether proponents think there is any theoretical limitation to this “feature.” Republicans (and as an extension white, rural voters) have enjoy a clear power imbalance. There’s no reason to think that imbalance won’t continue to grow in magnitude even though this year the imbalance in Congressional race outcomes relative to the # of votes cast was less pronounced in more recent elections. So then, is there any point at which that is no longer a “feature” short of, say, white rural voters voting for demz because they don’t like the imbalance (or i suppose demz adopting policies that favor white, rural voters but presumably thar would cost them support from diverse urban communities)? Is there any point at which a tyranny of the minority becomes unacceptable? What are the defining criteria for reaching thst point?

        • Tom says:

          This doesn’t address anything substantive about the US Constitutional structure.

          For example, state appointment of Senators was to ensure states rights were defended against a centralized government. That rationale is explicit. The popular vote of Senators by the 17th amendment in 1913 weakened that incentive. An empirical question is whether federalism declined after that change. Original rationale and incentives, change in incentives, subsequent outcomes vs original rationale.

          The criticism leveled here is similar. Whatever power structure you observe is the outcome of the checks and balances in the Constitution. That framework must be understood before we consider changes to the framework for what is perceived to be “bad” (Chesterton’s Fence).

          I don’t have an answer. Why the imbalance exists, what purpose it serves, and what incentives were built into the original framework and how they’ve changed provides a positive analysis framework with an explicit objective function. Asserting that the imbalance will grow and that it a problem without recognizing these issues is an unanchored normative analysis of what the government should be.

          • Zhou Fang says:

            But isn’t that constitutional structure prior to the invention of political parties and the notion that Senators are primarily loyal to their party? It seems like any claim of this defending local rights goes right out the window when there’s two inherently federal organisations defining most of what Senators do.

          • dhogaza says:

            “For example, state appointment of Senators was to ensure states rights were defended against a centralized government. That rationale is explicit. The popular vote of Senators by the 17th amendment in 1913 weakened that incentive.”

            That is simply wrong. Giving each states two senators was a structural solution to ensure that less-populous states would be less subject to domination by more-populous states. How those Senators were chosen were left to the states (until the 17th Amendment) and did not affect the balance of power between the states.

            • confused says:

              Eh, it may not be quite that simple. It didn’t *legally* change it, but I have heard the argument that the change made Senators more willing to take the side of federal vs. state power (as they were no longer vulnerable to recall by state legislatures).

              • Tom says:

                Not really important, the Senate was just a motivating example for how to view whether a system is performing as desired. Andrew’s response is the most pertinent and is fine for me. He is just looking at the structural advantage and its effect on the exercise of power (a well defined objective).

                As to whether this is “simply wrong” and that the Constitution left the selection of Senators to the states, well that’s just wrong.

                Article 1, Section 3 – Senators chosen by the Legislature:
                “The Senate of the United States shall be composed of two Senators from each State, chosen by the Legislature thereof”

                This is in contrast to Article 1, Section 2 for the house – Reps chosen by the People:
                “The House of Representatives shall be composed of Members chosen every second Year by the People of the several States”

                Article 1, Section 4 is where the election power resides:
                “The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof;”

                But that “election” for Senators is by the state legislatures, not directly by the people. The Senate’s webpage also confirms this.

                “The election of delegates to the Constitutional Convention established the precedent for state selection. The framers believed that in electing senators, state legislatures would cement their tie with the national government, which would increase the chances for ratifying the Constitution.”
                https://www.senate.gov/artandhistory/history/common/briefing/Direct_Election_Senators.htm

                The later popular vote reforms had senators elected by non-binding primaries or referenda. Oregon required the legislature to vote according to the popular will, so they kind of gave up their power. However, in the end the legislature was always the body that elected the senator and it can be argued that a current legislature cannot bind how a future legislature votes.

                So yes, it *legally* changed it but it may not have *practically* changed it. Which is the whole point of the example.

              • Tom says:

                And just to round out this discussion (I can’t reply to myself), the popular vote compact that comes into effect if enough states sign on to get 270 electoral college votes doesn’t change the fact that the electoral college elects the President. It changes nothing *legally* about the Constitution, it just is a loophole around that prescription that has the *practical* effect of neutralizing it.

        • Joshua says:

          Tom –

          > Asserting that the imbalance will grow and that it a problem without recognizing these issues is an unanchored normative analysis…

          Well, I said there’s no reason to think that it won’t grow. Perhaps you can offer one?

          I didn’t actually assert that it is a problem (although I think it is at some levels, not the least is that it could be unstable).

          So are you saying that you don’t think there is a problem no matter how wide the imbalance might be? So as an example if 20% of the national population were to elect the majority of senators, the president, have a disproportionately lare representation in the House, and dominate SCOTUS as a result, there’s no reason to think that would be inherently problematic irrespective of the intention of the founders? That’s the question I was hoping to get answered.

          • confused says:

            Not really to defend the current system, but I wouldn’t really expect the imbalance to continue to grow indefinitely.

            There’s nothing inherently “Republican favorable” about the system — that’s an artifact of the current post-1960-or-so realignment of the parties, and the increasing economic and cultural urban-rural divide.

            But the party platforms aren’t fixed. Nor are demographic trends in voting necessarily as solid as often thought; Florida and South Texas in this election seem to show that.

            So I would expect one of three things to happen…
            – The Republican Party’s reassessment between now and 2024 leads to a different platform with a broader appeal, possibly by taking more of a “working class” focus & dropping the more hostile rhetoric on immigration.

            – The cultural after-effects of COVID lead to more movement away from dense cities, so the demographics themselves become less unbalanced.

            – If neither happens, after the 2030 Census redistricting they probably won’t be able to win the Presidency even *with* structural advantage, so the platform will shift after *that* (possibly following an era of one-party national-level control, as the Republicans had at the end of the 19th/beginning of the 20th century).

            • Joshua says:

              confused –

              > There’s nothing inherently “Republican favorable” about the system

              Of course not. But with the level of gerrymandering that we’ve seen and the consequent structural imbalance already in play there is a momentum in play for it to keep going in the same direction in the House.

              And there is a structural imbalance that favors rural citizens in less populated statesn which, as it happens favors Republicans (and white people) and again, that has created a power imbalance that in itself generates a momentum towards an increasing imbalance.

              Doesn’t necessarily have to keep going in thst direction but I don’t see an compelling reason to argue the it won’t. Do you?

              > – The Republican Party’s reassessment between now and 2024 leads to a different platform with a broader appeal, possibly by taking more of a “working class” focus & dropping the more hostile rhetoric on immigration.

              Remember “The Party Decides.” That’s what we thought would happen before Trump and instead he consolidated support around a more exteme, rightwing center of gravity. The idea that he has move pubz more towards the working class is only true if you exclude minorities from the” working class” even though they comprise thst class disproportionately.

              • confused says:

                >>Remember “The Party Decides.” That’s what we thought would happen before Trump and instead he consolidated support around a more exteme, rightwing center of gravity.

                Sure — I’m talking about the reaction against Trump’s loss, between now and 2024.

                >>The idea that he has move pubz more towards the working class is only true if you exclude minorities from the” working class” even though they comprise thst class disproportionately.

                Sure. I was talking about post-Trump shifts in the party, not Trump’s jobs rhetoric.

                I am sure some strategists in the Republican Party are seeing the unexpected performance with Hispanic voters in Florida and South Texas right now.

                If they went to less hostile rhetoric on immigration, the demographics might look much more favorable to Republicans…

                There is actually a lot of cultural common ground there – high religiosity and family values/social conservatism for example…

              • Joshua says:

                confused –

                > I am sure some strategists in the Republican Party are seeing the unexpected performance with Hispanic voters in Florida and South Texas right now.

                If they went to less hostile rhetoric on immigration, the demographics might look much more favorable to Republicans…

                Maybe. First if all, some of that might be overblown to the extent it is based on exit polling which is always problematic but may be particularly so given the ideological skew in in-person voting this year.

                Also, at time point it becomes pretty useless to characterize the “Hispanic” vote as a distinct category. Many who are considered as falling into that group basically consider themselves as white. And cetainly treating them as a coherent group is problematic. Are Cuban-Americans more like Mexican-Americans than they are like Caucasian Americans? And Latinx vote seemed to play a different role in outcomes in Arizona and Nevada than in Miami-Dade

                > There is actually a lot of cultural common ground there – high religiosity and family values/social conservatism for example…

                That’s somwehat problematic as well. For example, blacks are pretty “religious” and tend towards “social conservatism” and Jews are big on “family values” but both groups lean heavily democratic. I think there’s some circular reasoning in your logic. But sure, it’s theoretical that pubz might pivot and go in a different direction – my guess is that isn’t going to happen, though. From what I’ve seen they generally think that Trumpism, which heavily focuses on animating a fairly constrained base, is the way to go.

              • confused says:

                >>some of that might be overblown to the extent it is based on exit polling which is always problematic but may be particularly so given the ideological skew in in-person voting this year.

                It’s not just exit polling, though. Some of those South Texas places are over 90% Hispanic, so a 25-point (or whatever) swing in the county’s vote is I think more than can be explained by that sort of thing.

                >>Also, at time point it becomes pretty useless to characterize the “Hispanic” vote as a distinct category.

                Absolutely!!!

                My point is just that I think there is an assumption common on *both sides* that “minorities always skew Democratic”, and that this is probably actually more easily changed by strategy than currently thought. Looking at the results in Florida and South Texas might break that assumption.

                >>I think there’s some circular reasoning in your logic.

                I don’t think so? What I’m getting at is economic vs. cultural/social motivations for supporting one party. There are a *lot* of economic “liberals” who are also social “conservatives” in the US.

                >>From what I’ve seen they generally think that Trumpism, which heavily focuses on animating a fairly constrained base, is the way to go.

                Eh, I don’t know how much we can judge about where the post-Trump reassessment will end up a week after Election Day, when Trump is still actually in office. I think it will take years, and the future face of the party will likely be determined in the 2024 primaries.

          • confused says:

            As for my own views, I think that making things more balanced would be good, but I think it’s critical to do it in a way that maintains the US’s identity as “an union of states” rather than a monolithic blob. IMO we have already gone a bit far in the direction of centralization of power.

            Perhaps something like expanding the House of Representatives so there are something like 200,000 citizens per Representative, and expanding the electoral college along with that?

            So Wyoming (~600k) would have 3 Representatives and 5 electoral votes, whereas California (~40M) would have something like 200 Representatives and 202 electoral votes.

            I think that could be done without a Constitutional Amendment, so it could actually happen! (Not this cycle, probably, unless the Democrats win both Senate runoffs, but still…)

            • Joshua says:

              As for personal views, I’d like to see additional viable parties – where coalition building would develop and where there’s less, lesser of two evils, scorched earth, zero sum tradeoffs.

              Not very realistic I’m afraid – but how realistic are any structural reforms given the stranglehold that pubz have on maintaining the status quo?

              • confused says:

                I think the only kinds of reforms likely to happen in the “near future” are those which can be passed with a narrow margin. It’s possible that the Democrats could gain the Senate in 2022*, and things that are purely legislative (as opposed to Constitutional amendments) would become possible.

                *COVID vaccination should be done by then & the economy recovering because of that, which could make Biden’s first two years look very good; and a R Senate would likely prevent Biden from doing anything drastic that would be easily painted as “socialism”.

      • bxg says:

        By my calculation (not sure I’m using the very latest totals) the presidential vote is split astonishingly close to 50/50 if California is excluded. (In fact to three decimal places based on the numbers I found!)

        If I were a Republican strategist, and assuming I count California as a completely certain loss, shouldn’t I be thinking of having to do rather better than 50% in the rest of the country (to outweigh California). So rather than aiming (A) slightly more-Republican positions relative to the population median, I should actually aim (B) for slightly more-Democratic positions relative to the population-minus-CA median?
        These *could* be the same thing, but that that would need to be proven. To the extent Californian is genuinely atypical these are more likely to be distinct. And as that hypothetical strategist, I don’t really care whether (A) and (B)
        are the same: (B) is a more relevant goal in either case, that’s what I should be thinking in terms of.

      • Tom says:

        CA is the only state where a candidate won by more than 1 million votes. Biden won 4.7 million. The total margin so far is 4.4 million. So yes, it is all CA. The rest of the country is relatively evenly split.

    • Jonathan (another one) says:

      +1. Outside of California, the Republicans had slightly over 50% of the two-party vote, last I looked… it was close. So you might well think of this as a country in which the Democrats start with California and then the Republicans have to try to make that up with more than 50% of what’s left.

      • Jonathan (another one) says:

        Or, the Dems start with NY and California and the Republicans have to get well over 50% of what’s left. You say above, Andrew, that the 2% is not in any particular state, it’s everywhere. It’s true that Republicans start with a bunch of small states as well, with well over 50% state margins there, so I’m not claiming anything unique here. But the extra 2 percent for Democrats doesn’t seem like it’s even close to evenly distributed across the country. And the Republican’s California (Texas) is getting swingier every year, though ain’t quite there yet.

        The other way to think about this is in the old baseball saying, that every team wins 40 and loses 40, and the good teams are the ones that win 80 percent or more of the 40 in the middle.

        • Phil says:

          The fact that “Republicans start with a bunch of small states” brings up another issue, which could perhaps be discussed on another thread: the fact that every state gets the same number of senators. Montana, Wyoming, North Dakota, South Dakota, Nebraska, and Idaho have a combined population of about 6.5 million, fewer (by about a million) than live in the San Francisco Bay Area. But those six states combine to have 12 U.S. senators, whereas California has only 2. Low-population agricultural states have disproportionate power in the Senate — wildly disproportionate to their population — and I think this causes all kinds of distortions in political decision-making. I’m not even talking about Democrats / Republicans, really, although of course that gets tangled up in it, but just things like farm policy and trade policy that can cross party lines.

          I’m all for preventing the “tyranny of the majority” but it’s not like tyranny of the minority is any better.

          • Jonathan (another one) says:

            There are several aspects to the Electoral College that upset people, and they often aren’t entirely clear about which ones they dislike. The “minimum of three” rule indeed gives extra power (relatively) to very small states. But people are often more exercised about giving states any recognition at all. The third aspect (which has been addressed partly by Nebraska and Maine) is winner-take-all. While the last one is entirely resolvable intrastate (though not if you insist on national uniformity) the other two really require a Constitutional Amendment. The rules are pretty much what everyone signed up for. And before you cite the National Compact workaround, I really don’t believe it is stable. It will work fine until one year when 90 percent of Californians vote for candidate X and they are forced by the compact to give all their electoral vote to candidate Y.

  3. Ben says:

    > As we discussed the other day, neither candidate seemed to make much of a play for the center during the campaign.

    Well, it wasn’t a conclusion of the discussion or anything.

    I liked this comment: https://statmodeling.stat.columbia.edu/2020/11/04/how-the-election-might-have-looked-in-a-world-without-polls/#comment-1573795

    > To which I reply: Yeah, sure, but positioning can still work on the margin. Maybe more moderate policy positions could’ve moved Biden from 52.5% to 53% of the two-party vote, but then again he didn’t need it.

    Could you explain more explicitly what you mean here? Biden didn’t win 52.5% of the vote in Florida. Is the margin here nationally? The comment from your correspondent was re: Florida, a swing state.

    • Andrew says:

      Ben:

      Yes, above I’m talking about the national vote. My impression (for example here is that there are electoral benefits to moderation but they are small. If Trump had staked out more moderate policy positions, maybe that would’ve given him another 0.5% of the national vote, and if Biden had staked out more moderate policy positions, maybe that would’ve given him another 0.5% of the national vote. I wouldn’t think that would’ve been enough to change the winner of Florida this year. Pennsylvania maybe.

      • Joshua says:

        > If Trump had staked out more moderate policy positions, maybe that would’ve given him another 0.5% of the national vote,

        And maybe it would have cost him 1% of the national vote. It certainly seems to be his belief that moderation would not have worked to his electoral advantage.

        >and if Biden had staked out more moderate policy positions,

        Such as?

        >maybe that would’ve given him another 0.5% of the national vote.

        And maybe it would have been a tipping point that would have cost him an equivalent amount of “progressive” support.

        It’s early to judge, but I’d say at this point that Biden became the nominee (and perhaps won in the general) precisely because he ran more towards the middle of the party than the left. None the least because by running to the middle he appealed more to black democrats who in general are in the more moderate wing of the dem party.

  4. Joshua says:

    > Yes, the are particular issues where one party or the other is close to the median position of Americans, but overall the two parties are well separated ideologically, which of course is a topic of endless study in the past two decades (by myself as well as many others). The point of this post is that, even in a polarized environment, there’s some incentive to appeal to the center, and the current asymmetry of the electoral system at all levels would seem to make this motivation much stronger for Democrats than for Republicans. Which might be one reason why Joe Biden’s talking about compromise but you don’t hear so much of that from the other side.

    This seems to assume that voting behavior is more a function of stances on issues than it is ideological identity orientation. People’s views on issues fluctuate as their identity orientation dictates. I don’t think that the flow is uni-directional.

    >. P.S. As we discussed the other day, neither candidate seemed to make much of a play for the center during the campaign. It seemed to me (just as a casual observer, without having made a study of the candidates’ policy positions and statements) that in 2016 both candidates moved to the center on economic issues. But in 2020 it seemed that Trump and Biden were staying firmly to the right and left, respectively. I guess that’s what you do when you think the voters are polarized and it’s all about turnout.

    Here, also… This seems to be assuming that the party constituencies are static. But that is questionable. For example, I think that Biden lobbied at the center of his party even as the Overton Window moved – meaning that his positions were to the left of Clinton’s.

    • Joshua says:

      But I do think that yes, it is pretty clear that there is a structural inequality thst plays out as the pubz being able to have a viable candidate who apples to a more extreme cohort of the electorate (and where many more moderate pubz have left the party) where as the demz likely need a pretty mediocre candidate (ideologically if not necessarily personally) who can hold the middle – not repelling moderates and independents and not repelling the “progressive” left.

      Seems fairly axiomatic at this point to me.

      • Joshua says:

        (for Martha’s benefit – apples = appeals, all puns are welcome.)

        • Martha (Smith) says:

          I did realize that “apples” was a typo (or probably a !@#$% auto-correct?) and “appeals” was intended, but I hadn’t thought of looking for a pun (Too tired, I guess). But since you suggested it: Maybe there’s something that one could do involving eating the forbidden fruit and being thrown out of the Garden of Eden?

  5. Marc Intrater says:

    Andrew, I do see your point that with only ~ 20% of the electorate up for grabs, a 2% difference may be significant. However I don’t think focusing on the overall, national difference in median voter would be a good research project.

    a) The structural advantage which causes this difference is due to districting, both at the level of states and districts within states. So any attempt to determine how this advantage affects politics should look mostly at these local level effects, not the aggregate. It is not a matter of R’s needing only the 49th percentile (from the right of the national scale) voter while the D’s need the 52nd percentile. It is that both need the 50th percentile in the swing states, which are not the same as the 50th percentile nationwide. Maybe the research project would be to look at issues that sway these voters, and how they differ from those that sway the national 50th percentile. As for Congressional control, many representatives have more reasons to be concerned about a primary challenger from their flanks than a general election contender from the other side of the median.

    b) The last two presidential elections were so much more about personality, group identity and signalling than policy issues. I don’t know that a play for the center would have had much impact. Biden was the most centrist of the major candidates in the Democratic primary, and Trump had several positions (e.g., drug prices) to the center of most Republicans, but I don’t know that this made much of a difference in either election.

    b)

  6. Zhou Fang says:

    > It seems nuts for a Republican to advocate that we “challenge and change the U.S. election system,” given the edge it’s currently giving them. In the current political environment, every vote counts, and the winner-take-all aggregation of votes by states and congressional districts is a big benefit to their party.

    Republicans won’t be happy until it’s impossible for them to lose. That’s what this means.

  7. John Williams says:

    Do you win elections based mainly on the policy positions you take, or on what kind of campaign you run? This weekend the NY Times had interviews with AOC and Conner Lamb, who took opposing positions on this question:

    https://www.nytimes.com/2020/11/07/us/politics/aoc-biden-progressives.html
    https://www.nytimes.com/2020/11/08/us/politics/conor-lamb-democrats-biden.html

    • Andrew says:

      Jonathan:

      I think both factors make a difference. This year the president was in so much trouble that maybe it didn’t matter too much what positions and campaigning was done by the Democratic challenger. Congress is another story, so the debate between the two young Democratic congressmembers is more relevant there, I guess.

  8. Kaiser says:

    I’m reminded of the Hotelling law which according to Wikipedia is the Comparative Midpoints Model in political science. Is there a multi-dimensional version of this? I think the collapse of dimensions into a Conservative-Liberal line may be creating blind spots. The swing voters are all presumed to be in the middle of that line… but there are also swing voters on either extreme (not voting/3rd party/write-in). A multi-dimensional view would be even richer – clusters of swing voters scattered in that space.

  9. jonathan says:

    I just want to question the correct strategy. The way you present it, time flows from the near term elections out toward longer term efforts to gerrymander your way, meaning control over states. That is a good way to fit to whatever near term process you are repeating, because the other side matches you. Or rather, I assume scenarios in which the other side cancels you, is annihilated by you, or annihilates you. So, you can lay out how you might achieve a percentage adjustment in results, but if you presume the other side also models, then these would cancel to the point where my guess is you see a relatively random breakdown in each election that offers a new-ish pattern but which is really an effect of models colliding, and which thus is never going to be reliably predictable unless you can determine a persistent modeling error on one side (that they cant see or fix even though it’s open and obvious). My guess then is that appositive point is being assumed: that there is an error on the other side that can be outmodeled. I don’t see that because that side has been doing pretty well.

    I think a lot of the issue people have is buried in the idea that popular vote total is somehow purer and better, though it largely seems to mean that CA is not the same as the rest of the country. If that one state were competitive in the same range as the rest of the country, most of the popular vote difference disappears. In a lower turnout election, it could entirely disappear, with most of that disappearing into the LA region. think it becomes a ‘moral’ issue not seeing that CA shapes the game: if CA were competitive, it would be much less necessary to gerrymander. If CA were competitive, we wouldnt even talk much about gerrymandering, outside of racial issues. But if you think ‘an election that counts only the popular vote is fairest’, then you have a problem thinking that it isnt. Imagine if CA were Texas, meaning it voted Red, with 5 million extra GOP voters. But of course the point is that you should by careful imposing a moral judgement prior on the shape of a game.

    As not really an aside, one of the weirdnesses of recent years has been the Democrats refusal to make the kinds of policy statements the strategy proposed requires. Example: they’re going to win the cities and areas where immigrants cluster, which means they were able to move to the center without losing cities, but they didnt. They couldnt say anything about immigration other than to fault the GOP. So, if you are projecting a model, you have to examine or assume a capacity and willingness. Let’s say you pick 5 issues, from overseas defense to immigration, and you look at what you can do, but you fail to examine the constraints on your actual ability to do a lot of what you can do. This is a common issue in corporations; they dont match their goals to their capabilities, and then they put goals ahead of developing the capabilities, and end up blaming culture and information problems.

    I see little capacity in the Democratic Party to make the statements the model would require to move their vote share. We are seeing that in the press now, with a spate of articles from Representatives saying that ‘Defund the Police’ and Progressive policies cost them seats. In CA, every single Progressive initiative was rejected. So why did the Party hew this line? The answer is because the Progressives attack their own. So if you have a model, then if you divide the Party into wings, the Democratic Left will attack its own Party vociferously. You’d need to model the GOP split differently: outside of specific states on specific issues (mostly abortion), they dont eat each other, and they dont eat at the Party itself in the same way. I’m not sure how to value ‘you’re not a true Republican’ versus ‘if you’re not Progressive, you’re evil’, but it seems they’re different.

    To be honest, I was afraid to say what I meant, which is the Democrats are far more afraid of being labeled racist. Go back to my comment about needing to be careful when imposing moral priors that shape the game. But since I try to be honest, at the party and candidate and campaign level, the Democrats value the label ‘racist’ differently than the GOP. It’s easy to summarize: the Democrats view the use of the label as carrying a horror, while the GOP views the use of the label as the use of a label by people who want to label them bad things. Any Jew should know what I mean because we’re exposed to labels of so many kinds, many of them applied in offensive ways. We can draw fine lines between labels which have some truth and labels which are outright lying disgusting trash. So, yeah, Jews will say to each other we can be pushy and we can be loud but we’re just trying to shine our lights in the world, but if you believe we’re monsters that run secret conspiracies or we control all the wealth, then you are worthless and may even be beyond redemption. I went through that to reach the end: to the Democrats at the party level, the racism label is like saying ‘beyond redemption’. This tilts their game board.

    So, the ‘best’ strategy is not at the aggregate level of popular vote totals but is more complicated. It goes back to Hillary in Detroit where low black turnout meant she lost. I put ‘best’ in quotes because I mean without moral values: you have to trade turnout in Detroit for wider appeal. Using Detroit, which I do reflexively because it’s where I grew up, highlights the racial dimension to city votes, but of course other cities are less black so the same thing could be said: you have to reduce appeal to cities, because now they are in a turnout game in which they need to produce all their votes to win versus the state as a whole. In some states, you could say the same thing about the GOP. Note that in CA, the way the government works has made much of the state into a ‘city’ in this sense, meaning the scale of government is larger in key ways. (And note the GOP realizes that and uses that as marketing.)

    I’m not advocating anything. But honesty in modeling is honesty in modeling.

    So, my summary is that the models should collide, should vary enough from cycle to cycle so apparent patterns appear and then shift, and that the state of the game reflects the internal constraints the sides have which prevent or enable them from appealing to the other side’s voters. These are apparent on both sides. Since the posting comes largely from a Democratic Party perspective, examining the obvious differences without imposing moral judgement is revealing, but not surprising since this is a failure to play a game as well as it could be played by the rules in force. In that regard, while I’m not a fan of Mitch McConnell, he sometimes says very astute things, like when asked why they were confirming a new Justice, he said the Democrats would do the same, and they should if the places were reverse. What he meant was: we’re playing this game for our side, not for yours. We happen to believe our side is best for the country. You keep insisting that your side is best, and you want us to agree to do what you want, but we’re elected by our voters under legal rules and we’re not only entitled to act for our side but that’s our job. I sometimes think a problem with Congress is fewer lawyers these days because lawyers internalize that.

    • Joshua says:

      > CA is not the same as the rest of the country. If that one state were competitive in the same range as the rest of the country, most of the popular vote difference disappears

      Yes, if the electorate were different than it is, then the election outcomes would be different.

      For examole, if white, rural voters moved to big cities then they would no longer have disproportionate influence in determining national-level policies.

      • TBW says:

        The Democrats should just pay hipsters to move to places like Wyoming and the Dakotas, in some of those states 200k people would dramatically change the shape of the electorate.

        • confused says:

          It’s possible — Montana and Wyoming had like a 120,000 vote margin for Trump in this election. But I don’t know that you’d find that many people who are both committed Democrats and willing to move. (IE, people who would be happy living in Wyoming are likely not committed Democrats!)

    • anon e mouse says:

      I will admit I started scanning after your second paragraph because your post is way too long, but the argument in your second paragraph has an obvious flaw. If the GOP had to compete in California and New York to win nationally, the national GOP would have to change its positions, or it would lose over and over and over again like it does in statewide CA races now. You seem to be assuming this would be a bad thing, but… would it? If a party can repeatedly control veto points and sometimes the entire trifecta despite consistently running on and enacting measures that are deeply unpopular nationally (tax cuts for the rich, privatizing/cutting Medicare and Social Security, repealing the ACA without anything lined up to replace it, etc.) that seems like a bad thing for the legitimacy of government.

      • Jonathan (another one) says:

        To be fair, they enacted tax cuts which had higher relative cuts (in income taxes…. the highly regressive payroll tax is another matter) for the part of the lower incomes that actually pay income taxes. (I realize that statement is somewhat convoluted, but so were the tax cuts. The rejiggering of the standard deduction hugely favored those who were taken the standard deduction, which wasn’t the rich, and the capping od State and local tax deduction entirely took from the those itemizing in high tax states, ie skewing mostly rich. The lowering of general rates, as always, favors those who pay those rates, so it’s not quite as clear as you make it.)

        They have run on privatizing/cutting Medicare and SS, but have passed exactly nothing actually doing it. So either people who vote for them on that basis will eventually realize they’re being hoodwinked, or already realized that they weren’t really serious about it.

        They only repealed the unenforced ACA mandate and have never mustered the votes to repeal it. I admit they seem to have no replacement for it, but they didn’t repeal it even when they had complete control.

        So it looks like the legitimacy of the government is pretty much intact… ignoring Trump, which now seems pretty safe.

        • anon e mouse says:

          Republican politicians and their affiliated groups continue to pursue lawsuits that would require the rest of the ACA to go away. If they didn’t actually want the rest of it to go away, they could… stop doing that. And they clearly have absolutely no plan if that happens, which it very well might with the now-very-conservative SCOTUS.

          Also I don’t know how you interpret the results on Medicare/Medicaid/SS as “they aren’t really serious about it.” It seems very clear to me that the party is very serious about it; Paul Ryan considered it his top priority and W. expended significant political capital on his failed SS proposal. It just doesn’t happen because it’s so unpopular that they can’t quite rally enough swing state Rs to run headfirst into the meat grinder. But given the views of an increasing number of recent SCOTUS nominees I think the strategy now is probably just to let SCOTUS gut the state so the R reps can claim it wasn’t their doing.

          • Jonathan (another one) says:

            We’re getting a bit far from both the statistical and median-voter-theorem aspects here, so I’ll keep it short.
            (a) Paul Ryan indeed had those views, and he left DC when he realized that even in a Republican administration, he had no chance of seeing any of it instantiated, among other reasons. He’s confirmatory evidence for *my* hypothesis that R’s aren’t serious about entitlement reform.
            (b) The severability argument about ACA promulgated by Trump’s Justice Dep’t (and certainly not by most R’s) is hot garbage that will go nowhere. it made a great talking point for D’s in the last election and represents one of dozens of Trump own goals, but expecting the courts to do with a spurious argument about severability that which you couldn’t do when you could have done so if only you could have mustered enough votes in your own party is again, to me, not a serious argument. I’ll give you 10-1 odds the the mandate is severed cleanly, and probably 9-0, though I may be wrong about Thomas, so no worse than 8-1.

        • Joshua says:

          > They have run on privatizing/cutting Medicare and SS, but have passed exactly nothing actually doing it

          Are you sure? They have run on cutting the debt but my sense is usually they run away from actually advocating “cutting” Medicare and SS (as distinguished from. Medicaid).

          Maybe a little bit years ago, but mostly it was promising structural reform that would magically avoid actual cuts to their constituents’ benefits, no?

          • Jonathan (another one) says:

            Good point. I was a little imprecise. First exactly who “they” is unclear. Paul Ryan (see above) for sure, and a few others, but obviously not enough people to actually get big changes made. Second, I fell into the standard Democratic trap of describing future changes to SS as a cut, rather than a reduction in the expected growth rate. Almost every SS change grandfathers existing benefits, sorta, but does things like increase the retirement age, which clearly cuts SS benefits but doesn’t change the benefits to anyone receiving them.

            • confused says:

              I honestly don’t see why we don’t raise the age for benefits by quite a bit, given how much life expectancy has increased since Social Security was instituted.

              Especially since birthrates were much higher then, so more workers per retiree. Part of the issue with the program IMO is that its structure assumes pre-1970 demographics.

              • anon e mouse says:

                These programs are popular in part because they are (basically) universal, so there’s a reticence to change that. The workforce has bifurcated, though. Those of us who sit at a desk can mostly work until we die, assuming we don’t develop anything neurological, although whether we should want that for people is questionable imo. People whose work involves physical activity, though, still are pretty limited by biology as far as how long they can safely and effectively work.

  10. David Marcus says:

    We actually have three parties: Two Democratic (corporate wing, progressive wing) and one Republican. The corporate wing is going to have trouble winning elections:

    https://cepr.net/donald-trump-and-being-deplorable/
    https://theintercept.com/2020/11/06/election-biden-democrats-progressives/

    The corporate wing tries to win by using a lot of money instead of by doing things that will be popular. Of course, the Republicans also do that. The result is that neither does things that are popular, so both are vulnerable to being voted out.

    • confused says:

      I’m not really sure how popular, nationwide, the progressive wing is either! Progressive House members may win by huge margins in their own districts, but those aren’t representative of the nation.

      I think the fact that Trump lost but Republicans in Congress did OK this election may warrant a re-evaluation of 2018 as more a response to Trump’s presidency specifically than evidence of the nation moving left.

      The 2020 vote patterns in South Texas and Florida may also suggest that a Democratic-favoring shift due to changing demographics is less inevitable than usually supposed.

      IMO, while party identity is strong, positions on specific issues tend to be more moderate.

    • Joshua says:

      > The corporate wing is going to have trouble winning elections.

      The corporate wing just netted the most votes for president in history, and a fairly large % of the presidential vote.

      The results in Congressional races are hard to interpret as there were losses (and wins) across the ideological spectrum of the dem party.

      • David Marcus says:

        The progressives supported Biden because the alternative was awful. But, a corporate President at the top of the ticket is a drag on the progressive candidates on the rest of the ticket. Polls showed that Bernie’s positions were more popular with the voters in the primaries, but many thought Biden was more electable. I think the results of the election show how misguided that was. Bernie also polled better against Trump than did Hillary. The corporate Democrats just don’t realize how much their policies have failed a large segment of the country. See https://cepr.net/donald-trump-and-being-deplorable/. They just keep throwing money at the problem instead of doing what people want. You would have thought the Democrats would have learned this after Hillary lost, but they didn’t. Of course, one reason to be a corporate Democrat is the corporations provide lots of money for your campaign. There is too much money in our election system. A system like the Seattle Election Vouchers would help: https://cepr.net/combating-the-political-power-of-the-rich-wealth-taxes-and-seattle-election-vouchers/.

        • confused says:

          >> Polls showed that Bernie’s positions were more popular with the voters in the primaries,

          Sure, but the voters in the primaries are largely Democrats! You have to appeal to independents to win.

          I’ve read that Sanders polled worse than Biden against Trump, possibly enough to lose given the Republican electoral college advantage – though obviously how much one believes these sorts of polls may vary…

          Moderate/independent voters may theoretically not like corporate influence, but they *also* tend to not want radical change. And this election I think was a specific “back to normalcy” one (whereas 2016 was a reaction against the establishment – as was 2008 for the other party).

        • Zhou Fang says:

          > Polls showed that Bernie’s positions were more popular with the voters in the primaries, but many thought Biden was more electable.

          Ignoring the primary voter aspect, if you mean stuff like healthcare, polls did not in fact show that. Polls showed whilst most voters liked both, in a head to head 1v1, voters preferred Biden’s public option to medicare4all.

          https://www.kff.org/health-reform/press-release/poll-democrats-like-public-option-medicare-for-all-but-overall-more-people-support-public-option-including-significant-share-of-republicans/

          • Zhou Fang says:

            Also, it’s fallacious to make the claim that if you just pick the most popular side of every policy decision, you end up with a *platform* that is actually popular.

            The issue is first that you might lose apparent coherence if you have too many policies, with people starting to think you’re just making stuff up instead of focusing on any particular issue, and that it’s impossible for you to actually execute on your promises, and second that popularity cannot be simplified to a binary. Suppose for instance each “popular” policy has a small minority of people for whom they are absolute dealbreakers. Then have enough of those policies, then even though each are mildly liked, you might well accumulate enough voters who have decided they would never vote for you because of some one individual issue.

        • Joshua says:

          David –

          > But, a corporate President at the top of the ticket is a drag on the progressive candidates on the rest of the ticket.

          I think that is too broad a statement. Everyone has been trying to figure this out for years. Which is more politically expedient – Go for independents/moderates by doing Republican-lite (where you effectively get two for the price of one), or focus on animating a “base” of young voters and minorities with a populist economic platform.

          I have feeling strongly that Republican-lite was a losing strategy to not being sure any more. I think that maybe Biden did as well as any candidate might, even as he was totally mediocre in all respects. Maybe he got independent/moderate votes while not completely alienating progressives. His platform was as far “progressive” as we’ve seen with any presidential candidate but he’s still towards the moderate end of the political spectrum. He was the only Dem candidate who got any black support at all, basically. That’s why he was the nominee. Would a more progressive candidate have gotten as much support from the black community? Well, as much as I might want to believe otherwise the primaries certainly indicated not. Maybe in the general another candidate might not have been as vulnerable as he to Trump’s lobbying for the African American vote (“super predator” and all that jazz).

          At any rate – I think that it’s really tough to apply some general principle in a way that’s particularly useful. Seems to me that the political implications of corporate vs. progressive really rely to a great deal on specifics of the candidates, their opponents, current events, and the specifics of the district involved.

  11. Rui Viana says:

    Applying the median voter theorem to the share of the popular vote in a Electoral College system seems incorrect to me. A strict application would instead predict that the two parties should be fighting for the median voter of the median state in the EC. I had multiple friends ask me last week: in a country of 350+ million people what are the odds that the election would come down to a few thousand votes in so many states? But that’s precisely what the median voter theorem predicts! The parties are fighting for the median voter in swing states, not for the 52th percentile voter of the US voting population.

    • Andrew says:

      Rui:

      Yes, but the point is that swings are mostly national, so to win that electoral college majority the Democrats have to win something like 52% of the popular vote. If they could just get a swing in the swing states, that would be fine for them (at least for the presidential race), but it doesn’t really work that way.

      • Rui Viana says:

        I appreciate the point Democrats needing 52% of popular vote and national swings. But I don’t see that it follows that the Republicans have more room to maneuver than the Democrats. Both parties are in a tight fight for Pennsylvania and Wisconsin (and others), and I don’t see either as been able to take positions that are unpopular in those states.

      • Chebyshev says:

        Is it a bit like saying the only important points that count in a tennis match at the Open are in the 5th set tie-break?
        One will have to get to the fifth set, let alone play a tie-break?

        • Andrew says:

          Chebyshev:

          In theory there are all sorts of ways for the Democrats to win the presidency and congress. In practice, there’s approximate uniform partisan swing, and the way for them to win is to get more than 52% of the national vote. Similarly, in theory there are all sorts of ways for the Republicans to win, but in practice what happens is that they do well or poorly in the usual areas and get a national vote of at least 49%.

          • Anonymous says:

            “In practice…”

            Democrats are unwilling to compromise on policy, so they can’t get the votes they want to take power.

            Andrew, this insinuation you keep making that there’s some sort of physical law that dictates people’s party vote is ridiculous. The simple fact is that both parties have become entrenched in their positions. They’re not moving, so why should the voters move? They can move. They just wont.

  12. Anonymous says:

    “while Republicans can get by with 49% or so. For example….they maintained control in the next three elections with 49.3%, 52.9%, and 50.6%”

    jeez:

    (49.3 + 52.9 + 50.6)/3 = 50.93%

    not ‘49%’ right? Did I miss something, new math or what?

    This constant blame on gerrymandering is a joke. Democrats have used it just as frequently. It just so happens that, at the moment, Democrats general platform is bad enough that they can’t garner the votes necessary to overcome it.

    • Andrew says:

      Anon:

      As I wrote, the Republicans can “get by” with 49% or so. More than 49% is even better. I didn’t say the Republicans get 49% in every election. Sometimes they get less than 49%, sometimes they get more. What I said is that 49% is all they need.

      Regarding gerrymandering, sorry to burst your bubble but it’s real. It’s not the only thing, nor is it the biggest thing, but, yes, it’s part of the story. It’s one of the reasons why Democrats need something like 52% of the vote to win the congressional elections. Again, gerrymandering is not the biggest thing, and of course it does not affect the electoral college or the Senate, but it does have an effect on U.S. congress and state legislatures.

      • Martha (Smith) says:

        Yes, gerrymandering is real! I have been moved from congressional district to congressional district, even though I have lived at the same address for more than 40 years. Sometimes my district has stretched from Austin south to San Antonio; sometimes it has gone out to West Texas, and currently it stretches up almost to Fort Worth.

Leave a Reply