34

If we look at the population of the US at large, as of time of writing, 84.7% of Democrats, 43.0% of Independents, and 9.4% of Republicans support impeachment. Meanwhile, in Congress, literally 100% of Democrats support impeachment, and literally 0% of Republicans do. If we assume that the national values are representative, then the odds of all 133 Democrats who've expressed an opinion landing on the side of impeachment is extremely low, in the range of 0.000000003%, and yet here we are.

Why are the Democrats & Republicans in the House so homogeneous? Statistically it's extremely unlikely to get this 100% support/oppose, yet that's what has happened.

Things I can think of:

  • The Democrats & Republicans in the House aren't representative of the population. If so, why? It seems weird too: it means the elected leaders don't actually represent the population, which seems to run contrary to the whole point of democracy.
  • The Democrats who oppose impeachment and the Republicans who support it are in the "no response yet" column. However, it's unclear to me why this would make a big difference - presumably some of the US population polled don't respond either.
  • They are under threat by their party to either vote along the party line or be expelled from the party. This is conceivable, but seems unlikely, because when Boris Johnson expelled rebels from his Conservative party a few months ago, 1) there were public warnings that he would do so and 2) it didn't stop the rebels from voting against his proposal. I haven't seen public warnings by either the Democrats or the Republicans, but it seems nobody wants to rebel anyway.
  • They might not agree with/oppose impeachment on a personal level, but they are voting the other way because they believe that's what the people that voted for them want. It's conceivable, but there's no evidence for this in the New York Times list - the closest is Republican Michael Guest who said "I will not vote to silence the voices of over 700,000 Mississippians who voted for our president", but he didn't express a personal opinion. Update: there're now news items claiming that even vulnerable Democrats are supporting the impeachment, which seems to refute this as an explanation.
  • 5
    @Allure There is a small group (10) of Dems who have floated the idea of censure rather than impeachment - I'm guessing they are in vulnerable districts and want to avoid swatting the beehive that is impeachment. If they decline to vote for it, that makes it 95% not 100% - We shall see. – SurpriseDog Dec 13 '19 at 5:30
  • 4
    The current prediction market is guessing that's its likely that 1-8 Dems will vote No on Impeachment with a longshot bet of 15+: predictit.org/markets/detail/6197/… – SurpriseDog Dec 13 '19 at 5:33
  • @SurpriseDog are their predictions about Republicans voting yes? – Burt Dec 13 '19 at 6:29
  • 1
    @Burt 17% chance of at least 1 R voting yes. - predictit.org/markets/detail/6198/… I'm pretty doubtful of this happening myself considering R support for impeachment is only at 9% – SurpriseDog Dec 13 '19 at 6:36
  • 17
    "literally 100% of Democrats support impeachment" Not true. House Democrat says he plans to vote against all articles of impeachment: "Rep. Jeff Van Drew of New Jersey, one of two Democrats to vote against formalizing the impeachment inquiry, said he plans to vote against all the articles of impeachment "unless there's something that I haven't seen, haven't heard before." ..." – Just Me Dec 13 '19 at 10:37

10 Answers 10

42

Actually, group polarization is not statistically unlikely for correlated variables like opinions, nor is it a phenomenon unique to politics.

People radicalize in concert with like-minded others due to the mutual affirmation of a shared identity. This behavior intensifies their shared attitudes, including a negative view of outsiders. This, in turn, generates the polarization of party platforms and officials.

In support are cited a couple of papers, one (1996) with a couple of otherwise unrelated experiments, (amusingly) on [the opinion on] dentist chairs, or mate attractions, but which nonetheless has this general conclusion:

social corroboration is capable of increasing the extremity of behavior having real consequences for participants

Actually a 1999 paper suggests it is applicable in many more contexts:

For example, people who are opposed to the minimum wage are likely, after talking to each other, to be still more opposed; people who tend to support gun control are likely, after discussion, to support gun control with considerable enthusiasm; people who believe that global warming is a serious problem are likely, after discussion, to insist on severe measures to prevent global warming. This general phenomenon -- group polarization -- has many implications for economic, political, and legal institutions. It helps to explain extremism, "radicalization," cultural shifts, and the behavior of political parties and religious organizations

(This conclusion seems to be based on some experiments with pre- and post-group-deliberation polls on more political/legal topics like child support or gun control.)

And finally a 2014 paper more properly on US politics finding group polarization really does apply there too:

party cues exert powerful effects on nonpolitical judgments and behaviors. Partisans discriminate against opposing partisans, doing so to a degree that exceeds discrimination based on race [...]

As for why the elite (i.e. representatives) vs popular polarization differs a bit in its intensity on this issue of impeachment... it may be a little harder to answer non-conjecturally, but the two-party system is probably plays a large role, as it does with polarization on other issues. In fact, [despite the protestations you often hear here on politics SE that the unidimensional left-right divide is meaningless] in the US Congress one dimension explains most of the voting patterns.

voting can be increasingly accounted for by a single dimension that distinguishes the parties. This situation directly contrasts with that of the mid-twentieth century, when the parties divided internally on a variety of issues primarily related to race and region

enter image description here

[...] Using the terminology of Converse (1964), issue constraint at the congressional level has expanded dramatically.

So, yeah, supporting or opposing Trump's impeachment seems to have become one of those issue constraints on congresspersons (on both sides of the isle).

With respect to Trump in particular, the coattail effect appears substantial, even extending to some policy issues, i.e. he swings the opinion of some loyalists in some experiments. Actually, there was highly popular question here on what explains Trump's popularity within his own party.

| improve this answer | |
  • 1
    Group polarization would explain a vast majority following the party line, but does it really explain 100%? – Barmar Dec 13 '19 at 21:11
  • 9
    Imagine a world in which politicians would base their opinion on impeachment on the merits rather than on party lines... – gerrit Dec 14 '19 at 13:14
  • 1
    @Barmar FWIW Justin Amash left the Republican party to become an independent and Jeff Van Drew is expected to leave the Democratic party and join the Republican party both for reasons regarding impeachment. Another democrat, Collin Peterson, also voted against impeachment. So we aren't at 100% (although it is close). – Magnus Jørgensen Dec 15 '19 at 0:06
  • 1
    @MagnusJørgensen So it's almost ther other way around: being for/against impeachment defines whether they're a D or R. – Barmar Dec 15 '19 at 1:00
  • 1
    If this would've been truth we wouldn't see any "new" opinions in ethical topics ever pop up. Group polarization also means that a new opinion can never form and take hold, a new opinion "enslaving people is bad" "woman have same rights as man", "homosexual behaviour is against the will of god", "god doesn't exist", would be shunned away from and not given a ground to grow – paul23 Dec 15 '19 at 18:39
50

Primaries. In order to win re-election a politician must win both a party primary and a general election. In most Republican primaries 80-90% of the electorate approves of the president and thinks he should not be impeached. So if you are a Republican running for office it is virtually impossible to win if you support impeachment because the people voting in a Republican primary really like the president. The same goes for the Democrats for them it is just 80-90% that support impeachment and dislike the president.

| improve this answer | |
  • 31
    @Allure not the primary they won last time, the primary they will fight next time. – Caleth Dec 13 '19 at 9:38
  • 27
    @Allure most of them will seek re-election. Any R who observably supports impeachment, and any D who observably opposes, will be attacked for that if they stand again. – Caleth Dec 13 '19 at 9:54
  • 14
    @Allure There's a fair-sized group of House Republicans that have announced they will not be running for reelection, for one reason or another. If I were trying to find Republicans who might support impeachment, I'd start there. – Geobits Dec 13 '19 at 14:16
  • 3
    @Allure 100% of House seats are up for election every two years. There are no “already elected” Representatives. – Joe Dec 14 '19 at 16:14
  • 1
    @Allure Every seat in the house is up for re-election 11 months from now. The “next primary” for all of those seats is starting right now. Nobody who isn’t retiring is “already elected”. – Joe Dec 15 '19 at 13:34
22

You have said,

85% of democrats support impeachment and you expect, therefore, that 100% of Congress Democrats will support impeachment? I don't follow your logic. I expect 85% of Congress Democrats to support impeachment, not 100%

But your logic is flawed.

I am going to use a crummy ASCII picture to demonstrate, P is pro-impeachment, A is anti-impeachment.

Here is your visualization of just Democratic districts, with all Anti voters in one district who's representative is then Anti:

PPPPPPPPPP || PPPPPPPPPP || PPPPPPPPPP || PPPPPPPPPP || PPPPPPPPPP || PPPPPPPPPP || PPPPAAAAAA

Here is a more realistic (but still massively simplified) visualization, with the same number of voters but Anti voters spread randomly throughout--none of these representatives would be Anti because even the most Anti district is only 20% Anti, and the representative can't vote "80% pro and 20% anti":

PPPPPAPPPP || PPPAAPPPPP || PPPPPPPPPP || PPPPPAPPPP || PPPPPPPPPP || PPAPPPPPPP || PPPPPPPPPA

Of course the reality is slightly more complicated. 1) Anti voters are probably not distributed evenly; 2) Each Democratic district would contain some percentage of Republicans who are more likely to be Anti, 3) Some of the Democrats (both Anti and Pro) are actually living in Republican districts.

But I hope this helps you understand the basic flaw in your assumption.

| improve this answer | |
  • 7
    Absolutely correct - you don't have 15% of Democrats that represent 100% anti-impeachment districts, you have 100% of Democrats representing 15% anti-impeachment districts (somewhat simplified). None of the democratic districts are majority anti-impeachment. – Nuclear Wang Dec 13 '19 at 15:37
  • 2
    How would all the Democrats know whether their districts are anti-impeachment? Are they conducting constant polls on all the topics they vote on? Also, what about bullet point #4 in the question? – Allure Dec 13 '19 at 20:25
  • 3
    @Allure It's statistics: If a majority of democrats are pro-impeachment, probably a majority of democrats in an individual district are also pro-impeachment. Also, representatives (or specifically their secretaries) likely receive a lot of letters and calls and such from people in their districts that can easily inform their decisions. – Klaycon Dec 13 '19 at 21:50
  • Am I correct in interpreting this as referring to the primaries? Otherwise I assume there are some democrats representing electorates that are up to 49% republican? – craq Dec 14 '19 at 9:20
  • @craq That's the "reality is more complicated than this" disclaimer. Too complicated to speculate about distributions, are Democrats in Republican districts more or less likely to be anti-impeachment?... etc. I'm just demonstrating the basic idea of how 15% anti-impeachment Democrats doesn't equal 15% anti-impeachment representatives. – user3067860 Dec 16 '19 at 15:50
10

US politics has become increasingly polarized since, perhaps, the 80's, but definitely since the 90's.

Partisan antipathy rose dramatically compared with 1994, when only 21 percent of Republicans and 17 percent of Democrats had highly unfavorable views of the other. By 2016, those figures had risen to 58 percent and 55 percent, respectively. [1]

Cross party-line voting is less likely than in the past. By extension, endorsing positions of another party is less likely than in the past. So the answer to your question is that due to increased polarization the population tends to closely adhere to party line positions.

I wrote an answer about polarization previously, you can find sources for my claims there.


[1] United States: Racial Resentment, Negative Partisanship, and Polarization in Trump’s America, Alan Abramowitz and Jennifer McCoy, The ANNALS of the American Academy of Political and Social Science, Vol. 681 Iss. 1, https://doi.org/10.1177/0002716218811309

| improve this answer | |
  • 4
    Anyone who is at least 50 years old sees the increased polarization. The bigger problem is the way liberties are being overrun by zeal for centralized control. Smart people talk openly about the need to curtail free speech. The biggest problem developing during this time frame is the abandonment of fact-reason-based problem solving, in favor or winging it. – DanAllen Dec 13 '19 at 22:26
  • 2
    @DanAllen careful there. I am 50+, white, male myself and it seems to me that 50+, white, male US voters themselves aggregate largely into 1 of the 2 parties. While 50+ voters may remember more amicable times, they are hardly exempt from polarization themselves. Oddly enough, being French, and having had to put up with true-dyed French Communist Party voters (an extremely intolerant lot) has made me a lot more accepting of people with different viewpoints. – Italian Philosophers 4 Monica Dec 14 '19 at 5:04
  • It's certainly been increasingly polarized since the 60s, not just the 80s and 90s, and I imagine if you go further back you can continue to find lynchpin politicos that served to exacerbate the issue of polarization. Us vs Them has always been a quick and easy sell. – TylerH Dec 17 '19 at 20:58
8

We do not have direct democracy, we have majority-rule representative democracy.

Let me use an example to illustrate. Assuming the 15% of Democrats who disapprove aren't crammed into only a few districts, let's say each district with 51+% impeachment approval produces a representative who is pro-impeachment. That representative isn't 85% for impeachment, s/he's 100% for impeachment. While you and I would hope that our representative is considering the opinions of that 15% who were anti-impeachment, s/he would be a fool to ignore the will of 85%. So that's how a 85% majority turns into a 100% impeachment approval in the House.

| improve this answer | |
  • How would all the Democrats know whether their districts are anti-impeachment? Are they conducting constant polls on all the topics they vote on? Also, what about bullet point #4 in the question? – Allure Dec 13 '19 at 20:32
  • 1
    @Allure They know their districts support them because people keep voting them in. It doesn't really matter why they vote them in, only that no cause has yet been so far over the line that people are willing to vote for "the other side". It doesn't even matter if their constituency supports impeachment (or any single issue), the alternative is voting for a wholly separate platform that isn't to their political taste. While the intention of the system is to vote for people, in reality votes count for parties, because the parties tightly control all the candidates. – Knetic Dec 14 '19 at 23:54
  • As an example, Republicans in the 90s did exactly what Democrats are doing now - they tried to impeach a president after a two-year investigation, on fairly technical legal grounds that most people agreed were too petty to warrant removal from office. The outcome was that Republicans lost seats when they were expected to gain them, and the leadership (Gingrich) never held office again - voters had had enough of them. It didn't matter which issue or clip or person pushed it over, people stopped voting for them. – Knetic Dec 14 '19 at 23:58
  • @Knetic well when the people voted them in, impeachment hadn't even started, so one doesn't know if the people actually are in favor of impeachment or not. They are voting based on their opinions on other issues. The parties at the time don't have an opinion either. The representative would have to come to his/her own conclusion on impeachment, and that would indicate that ~15% of Democrats would oppose it. The last impeachment attempt failed, but circumstances for this are different. That's how I interpret it. – Allure Dec 15 '19 at 5:23
7

OP: Why are the Democrats & Republicans in the House so homogeneous? Statistically it's extremely unlikely to get this 100% support/oppose, yet that's what has happened.

This can be answered partly by the rest of my answer below, but it can also be explained by the fact that it is all a power struggle for control of the government. We've become so polarized as a nation that progress as taken a back seat to "winning".

In his essay, Fate Of Empires, Sir John Glubb (b.1897 - d.1986) says:

Another remarkable and unexpected symptom of national decline is the intensification of internal
political hatreds. One would have expected that, when the survival of the nation became precarious, political factions would drop their rivalry and stand shoulder-to-shoulder to save their country. ... True to the normal course followed by nations in decline, internal differences are not reconciled in an attempt to save the nation. On the contrary, internal rivalries become more acute, as the nation becomes weaker.

This passage was prefaced by the notion that great empires throughout history have a predictable lifecycle (The Age of Pioneers (outburst), The Age of Conquests, The Age of Commerce, The Age of Affluence, The Age of Intellect, The Age of Decadence) which leads to an eventual decline and collapse due to internal factors. (I highly recommend this read, it is an entirely non-partisan perspective on the eventual fate of empires)

OP: The Democrats & Republicans in the House aren't representative of the population. If so, why? It seems weird too: it means the elected leaders don't actually represent the population, which seems to run contrary to the whole point of democracy.

The best explanation I've ever heard as to why politicians aren't representative of the population can be found in the video Why Government Fails.

In it Prof. Antony Davies explains, if you have two politicians: Politician-A and Politician-B.

Politician-A's primary motivation is to seek the common-good. Politician-B's primary motivate is to get elected.

On average, Politician-B, the guy who seeks to get elected is the guy who is going to win. And in doing so, he must say what the greatest number of people want to hear, while not perfectly representing any one group's best interest.

| improve this answer | |
  • Wow. It's really sobering to realize we're following the pattern so predictably. – Foo Bar Dec 14 '19 at 18:02
  • @Foo Bar , likewise, its also reassuring reassuring, in the sense that even the most dystopian global empire will weaken, crumble and fall on its own accord – holaymolay Dec 16 '19 at 16:22
4

As with most things related to politics, there are a lot of factors that would encourage Congress members of the same party to be all say they are in favor or in opposition on a particular issue. The key two (which have been discussed in other answers) are:

  • Distribution of Voter Support - It's unlikely than anyone in Congress represents a district that is leans heavily against the party line.
  • Preparing for Primaries - Acting against the preferences of the vast majority of one's party risks being challenged and losing in primaries.

There are a lot of other factors, though, that have some influence. These include:

  • "Inside" Knowledge - Those who serve in Congress are, on average, much better informed than the general population about the political issues. It's their job to know this stuff! Sometimes what may seem like the obvious action to take to someone in Congress may not always be apparent to voters with limited information.
  • Disconnect Between Representatives and Voters - When a vast majority of voters in a Congress person's district support their position, they may never hear voters who disagree and if they do, they may incorrectly assume that disagreement is coming from members of the opposite party.
  • Party Loyalty - Party leaders have only so much tolerance for dissension. If a Congress person undermines the party's message on a crucial issue, the offender will likely have less or no support for re-election from their party.
  • Talk is Cheap; Silence is Golden - If you are a politician whose position goes against the grain, it makes sense to either stay quiet or even lie instead of announcing your actual thoughts. The media is far more likely to spread a contradictory statement than a contradictory vote. It may make more sense to wait until the last possible moment before "defecting."
  • Powerful Incentives - Members of Congress are still people. They are fallible and may not ensure that the preferences of voters are accurately reflected in their decisions. Instead, they may act in a fashion that strengthens the political power of themselves or their party.

There's plenty of motivation for every single Democrat in Congress to support impeachment. Even if the President is completely blameless, as long as their is sufficient reason to believe he is not, Democrats will go forward with it to please their base and undermine the President's re-election bid. Likewise, even if the President is intentionally abusing his power in ways far greater than he's been accused of, Republicans are not going to openly support impeachment unless their base is receptive to it or the President is no longer supporting the party's agenda.

In short, it's partisan politics as usual...

| improve this answer | |
3

All politics are local.

In the U.S. the first job of elected officials is to get elected again. Any politician that says otherwise is either lying to you or retiring (whether they realize it or not). National Public opinion on an issue does not matter because those people don't vote for you... your local constituents do (if the Senate, it's the people of the state you're representing. If the House, a small subset of the people of the state your representing.). That 9-15% is not all in your constituency... they're spread out. And do not forget that there are Democrats in Republican stronghold states, and vice versa. Furthermore, Independents aren't reliable to either party for elective support, but to underestimate the independent vote has doomed many an American politicians (right now independents are the largest "party" affiliation of the common voter in the U.S., with a slightly larger than 1/3rd of the population registered so). Both parties are banking on the independent support. If one half of one third supports your position, and ~10% of your 1/3rd doesn't support your position... then why should you be concerned with the 10% who are likely to vote for you anyway on other more important issues.

| improve this answer | |
3

Republicans, at this point, spent three years claiming that Trump's action are very much within reason, and even those who were opposed to him pre-election, are now staunch supporters for whatever reason.

If they decide that this is too much, they are saying that they were somehow complicit in his previous misdeeds, including the ones he's not being impeached for.

At the same time, the Democrats have the opposite problem. They spent the last three years wanting to impeach him and have him removed from office (although that seems very unlikely). They can wait for the elections next year, sure, but if they don't do anything about him, their voting base will see them as enabling, if not complicit of, Trump's actions.

Partisan politics, as the US is based, is based on the fact that each party has a large number of voters which will vote for the party "do or die" (to quote a certain soon-to-not-be-European leader). These are usually split about halfway through the population. The real difference is with the swinging votes, the people on the fence, as well as the young voters who are joining in for the first time.

If nothing else, not impeaching Trump sends a message to the young voters that politics, at the very least, is useless, and they shouldn't bother joining in, since "both options are the same". Whereas Republicans hope to convince the young voters and the swing voters to enable them to give more power to the president in November, which they hope is Trump, because they believe him to be righteous and doing well for the country.

| improve this answer | |
  • Since my comment was deleted for being "political" I wrote an answer. While I do not like Trump, nor his effect on global politics, I tried to keep this answer as little biased as possible. – Ink blot Dec 14 '19 at 22:54
  • This doesn't seem like a good explanation though - if it's true, then it predicts that 100% of Republicans/Democrats will think what they do on impeaching Trump, and that doesn't mesh with the national poll results. – Allure Dec 14 '19 at 23:19
  • @Allure Not quite. As Inkblot says, it's a "large number of voters". But it does predict for Republicans that the more they've excused his behaviour in the past, the more sunk cost they have. When you've already excused sexism, racism, sexual assault, voyeurism of underage girls, infidelity, incestuous sexual comments about his own daughter, lies almost beyond counting, and simply being a failure as a businessman - all simple facts known even before election day - then what would make you think he's crossed a line of acceptability? – Graham Dec 16 '19 at 11:54
  • @Allure ... And conversely for Democrats, the list of reasons is already very much long enough. The issue for them isn't whether they should, because they consider they have enough evidence of his moral unfitness for office. The issues are simply that (a) they didn't have enough votes in the House, and (b) they didn't have an issue which they considered to be unambiguously true. – Graham Dec 16 '19 at 11:56
  • 1
    @Allure Some Republicans have drawn the line at Trump's behaviour (John McCain before his death, for example). There is also disagreement amongst Democrats about whether impeachment is a good idea tactically, because the Senate is controlled by Republicans so it is unlikely to result in any action against Trump. – Graham Dec 16 '19 at 15:47
2

From what I have followed in the politics, I would honestly say a huge chunk of it comes down to personalty and personal score keeping. Politicians as of late have not been playing nice with each other across party lines and at the end of the day people are kinda prone to hold grudges when say...mudslinging gets a little to personal and detached from the facts.

Yes there is many other very valid reasons why the parties are polarized but it is doubtful that anyone will be eager to extend a hand over that dumpster-fire of personal grievances.

| improve this answer | |

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .