The GOP created the “rigged vote” myth.
Trump really isn’t doing anything but taking seriously what the GOP, in a racist attempt to limit Democratic voters’ participation, has been saying for years.
This is the best book to help you understand the wild 2016 campaign – Vox
“Partisan loyalties are largely built up from fundamental group identities rather than based on profound ideological commitments, and swing voters swing in large part for no good reason at all — maybe because of a recession, but maybe because of a swing in global oil prices or because the Steelers lost or almost anything else.”
How Half Of America Lost Its F**king Mind | Cracked.com
One of a growing genre of “what are Trump voters thinking?” articles. I have issues with this article, and with this trend of articles, but I still think it’s worth reading.
AskTrumpSupporters
Along similar lines, the “Ask Trump Voters” reddit is interesting reading.
The Midwest’s Racial Incarceration Problem
The South puts more black men in prison in absolute numbers – but as a percentage of prisoners, the North and Midwest are worse.
I recently moved to California from points east, just in time for the elections. (Well, more than just in time–if it had been legal for me to still vote in my previous swing state, I might have been tempted.) So this is my first time dealing with CA’s propositions system. And that means getting the official voter information guide. It has the stuff I’d expect in it (summaries of the propositions and a level-headed disinterested analysis of the effects it might have). But it also has arguments in favor and opposed (and rebuttals to those arguments), which can apparently be submitted by anyone in CA, although certain types of arguments get priority if multiple arguments are submitted. And they’re fascinating.
I’ll be the first to admit that I seek out certain kinds of spaces to have political discussions in, and they tend to be fairly low on emotion and high on principle and wonkiness. I deliberately avoid some types of political discussions, in other words, and yet here they are in the official voter guide! Here’s maybe the best one, on the question of whether grocery stores (etc) should be banned from giving out single-use plastic bags.
(I’ll note that there’s an argument against this kind of law that the thick plastic bags are so much worse for the environment that it’s not clear these laws help if lots of people buy them instead of reusing. But that argument doesn’t appear here.)
Anyway, I didn’t know this existed, or that something like a voter guide distributed in the most populous state in the nation could produce arguments with the same flavor as letters to the editor in my Midwestern hometown (both the good and the bad).
The question that’s been weighing on each of our minds, but that we have been unwilling to acknowledge—until now: If I drink a shot each time a Democrat wins a Senate race, up until I know they will control the Chamber, will I end up in the hospital?
This ends up being a tougher question then I’d bargained on. The answer depends on –
1. The current partisan split among Senators,
2. The list of states hosting senate races,
3. The party affiliation of each of these challenged incumbents,
4. The time when polls close in those states (and thus, when journalists will make forecasts),
5. The relevant time zones for each state (Many states operate in more than one time zone!),
6. The alcoholic content of the shot,
7. My weight and metabolism,
8. Etc.
Yikes! Good think I didn’t start drinking early. Here’s my take:
According to FiveThirtyEight, there are basically ten senate races remotely in contention: FL, IL, IN, KY, MO, NC, NH, NV, PA, and WI.
Unfortunately for Democrats, that party start out with only 46 Senators. So they’ll have to win five seats to gain a majority.
Fortunately for Democrats, they only have to defend one of the ten contested seats; the rest of the challenged incumbents are Republicans. By my count, Dems need gain four seats plus keep their existing seat (or win a different one elsewhere): 4+1 means the Dems need to win 5 seats – out of 10 possibilities. Even odds, huh?
More fortunately for Democrats, if they win the Vice-Presidency, that will count as a seat. So long as they can muster 50 votes, they can battle all issues to a tie—and Vice President Tim Kaine would then pass the deciding vote. (OK, smarty-pants, set aside the cloture rules for an evening, ok?) So arguably the Dems only need four seats, not five. But this complicates the analysis: When will we know if Clinton/Kaine have won?
Anyway, here’s a list of closing times for polls, by state. Interestingly, at least three of the states in contention—FL, IN, and KY—span the Eastern Time/Central Time line, and thus have polls closing at different times within the state. But it’s my understanding that journalists have not been reluctant to forecast the electoral outcomes in these states, even while some of the polls have remained open. So for practical purposes, they might count as Eastern Time states.
To complicate matters further, the last competitive Senate race of the night—in Nevada—is also the sole Democratic Senator in contention. So if I’m going to get to bed sooner than 10pm, I’m back to needing five Republican incumbents to face defeat.
Last step: Based on FiveThirtyEIght forecasts, here’s a rough schedule of my alcohol consumption:
• 6pm: Early polls close in IN, KY, with Dems forecast to win the former. One shot.
• 7pm: Early polls close in FL. Who knows when they’d actually get any poll results—but it probably won’t matter, ‘cuz Dems won’t beat Rubio. No shot for you. But there’s some chance that journalists, seeing that Trump has lost FL, will call the election for Clinton/Kaine. So I may wet my whistle yet.
• 7:30pm: Polls close in NC. This one is a nail-biter. FiveThrityEight says no. But Hamilton says I’m not giving up my shot! Who ya gonna believe?
• 8pm: Polls close in IL, MO, NH, PA, and Democrats run the table. Four shots, and I’m down for the night! And once journalists call PA for Clinton/Kaine, you gotta suspect that they’ll be calling the election, too. (Probably it’s been an hour or two since my last shot, so I needn’t end up in the emergency room. Not like during that second debate. Who would have guessed that agreeing to take a shot every time someone says “DISASTER” would turn out to be such a, well, disaster…?)
• 9pm: If things don’t turn out as well as forecast for Dems, or if you have more stamina than I, you might stay up to drink to WI’s Democratic victory.
• 10pm: Finally, we’ll be able to place all the nametags on the Senate seats when the polls close in NV. Looking good for Dems here, too.
Anyway, here’s a list of closing times for polls, by state. Interestingly, at least three of the states in contention—FL, IN, and KY—span the Eastern Time/Central Time line, and thus have polls closing at different times within the state. But it’s my understanding that journalists have not been reluctant to forecast the electoral outcomes in these states, even while some of the polls have remained open. So for practical purposes, they might count as Eastern Time states.
Indiana is weird — only a little bit of it is in Central time, but that little bit includes Gary, which is something like 85% black. If you’re watching Indiana numbers come in live, you can pretty much always assume that there’s going to be a huge jump in the number of Democratic votes all the way at the end of the counting. (Also, that you’ll have The Music Man stuck in your head all week.)
And Nancy, thanks for the link. It’s hard to feel sorry for that racist dumbass, but as a general principle I think we’re all better off if it were a norm that people aren’t fired for what they say when they’re not at work.
It ruled that despite declaring abortion a fundamental right, that didn’t mean Medicaid had to extend that right to poor women,
Do you think that the Federal government should have to fund the exercise of a fundamental right for people who cannot afford to do so themselves? Consider that the right to keep and bear arms is a fundamental right.
It could start to develop a robust right to vote and limit gerrymandering.
In my home State of Illinois, gerrymandering gives the Democrats +2 House seats over what they would have otherwise.
Re: Gerrymandering. Yup, and that’s horrible. We need to end gerrymandering nationwide, including in Dem controlled and districted states. Which means recognizing that it is a infringement on the right to vote, and a violation of equal protection, and acting accordingly.
When House representation is 10 points off of generic congressional ballot, something is deeply wrong.
The thing is, even with using more natural political or geographic boundaries you can get House representation that varies from a generic ballot.
For example, let’s take a State that has an urban center. For simplicity’s sake, figure the urban center has enough people to form one House district, a suburban area that has the same number of people, and 2 rural districts. The urban district goes 90% D / 10% R. The suburban district goes 55% R/45% D. The two rural districts go 65% R/35% D. The State as a whole is 51% D / 49% R. So a generic Statewide ballot would give you roughly even representation, but the actual representation will go 3:1 R. You don’t need gerrymandering to get an imbalance; it’s an issue of the fact that different areas have different interests, and that we live in a Federal republic, not a pure democracy.
I wouldn’t say it’s an infringement on the right to vote. The right to vote means that if you are a citizen of voting age and have residency in your district, you get to vote. It doesn’t mean that the party you favor wins.
If the professor was wearing the “Black Lives Matter” T-Shirt during class he/she is teaching, I think the students would have a point. The professor is in the classroom to teach law, not to make a point of his or her political opinions. OTOH, if he’s just wearing it around campus, too bad for the students – just as it’s too bad for other students or faculty if the students are chalking “Trump 2016” on campus sidewalks.
There’s some natural variance to districting, sure, but until the great 2010 gerrymander (project REDMAP), the difference between the national congressional ballot and actual representation was within a much smaller range. Gerrymandering has gotten much, much stronger with the advent of computer graphing.
The right to vote means that if you are a citizen of voting age and have residency in your district, you get to vote. It doesn’t mean that the party you favor wins.
No one’s saying that “the party you favor wins.” That is just a straw man. What we’re saying is that the party that the majority of voters favor should win most of the time in the House of Representatives, which is supposed to have proportional representation. When things are so gerrymandered that one party can lose the overall popular vote by millions of votes, that’s a serious problem. When the party in that position is also the one that benefits from the lack of proportional representation in the Senate (as the Republicans currently do), it is an even bigger problem. Structurally, people in large, urban areas are underrepresented in both the House of Representatives and the Senate right now. Yet, it is conservative rural voters screaming about vote rigging.
The House is supposed to have proportional representation by a combination of population and geographic region and the Senate by sovereign political entity; neither is supposed to reflect party affiliations. The U.S. Constitution (unlike a Westminster style Parliamentary system) does not recognize the concept of political party and makes no provision for a role for them in government. I think it would be a very bad idea to start now. Congressional districts were intended to be constituted on the basis of geographic and municipal boundaries, not those of party affiliation. That’s why the creation of a convoluted district for Elbridge Gerry’s election to the First Congress was so notable.
What do you mean by “people in large urban areas are underrepresented in both the House of Representatives and the Senate”? On what basis?
Here in Illinois the urban Congressional districts are deliberately shaped in such a fashion as to include portions of the surrounding suburbs as well, so that the overwhelming Democratic majority in Chicago and East St. Louis can be used to offset the Republican voters in the suburbs. I find it difficult to believe that we are the only blue State that has this phenomenon. Note that in my example the imbalance between State-wide party voting and the actual representation is not a function of gerrymandering, it comes naturally from the political leanings of the geographic regions.
I’ve considered that there should be some way to base legislation on some formula wherein no Congressional district’s ratio between boundary length and area could exceed a given (and somewhat arbitrary) number. This would make convoluted districts such as Illinois’ Fourth District illegal.
The reason why that graph looks like it does can be accounted for by the fact that the Democratic party appeals to a large number of people but those people are concentrated into cities. So States that don’t have a high proportion of urban population are not going to go Democratic. If the Democratic party could broaden it’s appeal to non-urban voters they could take the Senate and the House.
Here in Illinois the urban Congressional districts are deliberately shaped in such a fashion as to include portions of the surrounding suburbs as well, so that the overwhelming Democratic majority in Chicago and East St. Louis can be used to offset the Republican voters in the suburbs. I find it difficult to believe that we are the only blue State that has this phenomenon.
In the purple state where I live, most of the “urban” districts include part of a city, a bunch of suburbs, and a whole lot of surrounding rural area. The city that I live in, plus the first ring of suburbs, would be about the right population to be a district, but instead, my district is half the city, half the suburbs, and about four or five rural counties. The other half of the city is in a separate district, which is composed similarly. My district has been Republican for decades; the other one swings back and forth. I honestly have no clue what sort of local interests my representative actually represents — his district is defined so weirdly that it includes urban, suburban, and rural areas. (The other district within my city also includes a couple of smaller cities nearby. My district is just my half of the city, a bit of strip-mall suburbia, and farms.)
I’m wondering whether “I didn’t believe what I said, I was just saying what I thought would hurt the most” is an excuse in any meaningful sense.
I don’t think it is. Someone willing to be racist for instrumental purposes is, to my way of thinking, still being racist. In a way, claiming that he knows better makes it even worse.
I went to do some digging on gerrymandering, as I remembered some recent academic work had been done on quantifying it.
First, there’s the efficiency gap. That takes election results and measures how many “wasted” votes there are–that is, votes that either didn’t go to a winner of an election, or were in excess of the 50% + 1 vote that are needed to win. Gerrymandering means winning more seats based on the same number of votes, so gerrymandered districts show a discrepancy in the fraction of wasted votes per party. The measure is simple, but is subject to some weirdnesses (as the paper itself details).
A more complex thing to do is make a bunch of fake district maps and use precinct-level results to see what the distribution of election results would be, and then to compare the actual results to that distribution. That nicely captures the point about spatially-clustered Democrats leading to some “natural” gerrymandering, since it’s included in the maps. However, it’s more complicated, which means among other things it’s more subject to errors & assumptions made by the people who made the maps; still, this looks reliable to me (though I’d love to hear criticisms if people know of them!). This group did the simulated-maps thing. What’s interesting is they find a net gain of only about 1 House seat for Republicans based on gerrymandering. That’s because of a combination of California, which they find has the largest gerrymandering magnitude in favor of the Democrats, and a pro-Democratic bias caused by VRA-mandated districting plans; those two effects combined almost outweigh the partisan gerrymander in many Republican states.
However, I’ll note that they use 2008 precinct-level results to predict post-2010-plan Congressional results, and I think 2008 was weird enough in several ways that this might fail. (In particular, 2008 was such a sweep for the Democrats that some closely-gerrymandered districts might have gone Democratic in the simulation that actually didn’t in 2012 or 2014, since the gap between D and R went from ~7% to ~4% in the presidential race.) So my guess is that this is a best case scenario in terms of how many seats Democrats lost due to gerrymandering, and the real answer is probably higher.
Anyway, I found both those things interesting reading; thought you folks might as well.
And to address that interesting chart from Ben Lehman: 2012 is clearly an outlier, but I’m not sure about 2014. That grey blob is just the convex shape that includes all the pre-2012 data points and nothing else. If you think it should be a linear trend–and it kinds of looks like that–and you imagine a grey band that follows the trend, but has some width to it, I think 2014 would actually be in that band extrapolated down. It’s below the grey blob just because it’s so rare for the Democratic share of the vote to be that low, and all the previous low points happened to favor the Democrats relative to the usual trend.
Why was 2014 less unusual than 2012? I think might get into a point mentioned in the efficiency gap paper above: the distribution of votes in districts can change quickly enough that gerrymandered maps become less gerrymandered, both because people move and because party affiliations can change as the parties do. Or maybe it’s just statistical noise, or maybe it’s that gerrymandering’s less effective when one party’s support drops unusually low. Shit’s complicated, man. :)
“I honestly have no clue what sort of local interests my representative actually represents — his district is defined so weirdly that it includes urban, suburban, and rural areas.”
My guess is that his interests are intensely local – local to his own person, which he wants to make sure gets re-elected every two years. This is why districts should be compact and based on natural geographic and municipal boundaries – so there some kind of consistent set of interests that the Representatives is responsible for.
To reiterate the point – given the disparate kinds of interests that the two parties represent and given the unevenness of the distribution of the people with those interests, there is no particular reason why the representation in the House should match some generic or average ballot.
In Illinois the Democratic party (which controls the General Assembly and thus controls re-apportionment) is careful to ensure that the number of urban voters in the mixed urban/suburban districts outweigh the suburban voters. It sounds like the opposite is happening in your State – the re-apportioning authority is making sure the rural/suburban votes outnumber the urban votes.
For the most irritatingly designed districts I’ve ever seen: zoom in on Austin. There are something like five or six districts that have little bits of the city plus a ton of surrounding area. Some of those districts spread out a few hundred miles beyond the city. None of the other Texas cities look like that. https://upload.wikimedia.org/wikipedia/commons/5/53/Pagecgd113_tx.pdf
Trump fans on reddit think that the polling is skewed and that the election will be a blowout in Trump’s favor.
Oy.
I read “The Moral of the Story” at work. Luckily, I just managed to stifle the groans before I bothered all my office mates…
(which is to say, thanks for the link)
Is anyone else a little squicked about about the “what if only $ingroup/outgroup voted” charts, in terms of their wording?
As this is an open thread–
I recently moved to California from points east, just in time for the elections. (Well, more than just in time–if it had been legal for me to still vote in my previous swing state, I might have been tempted.) So this is my first time dealing with CA’s propositions system. And that means getting the official voter information guide. It has the stuff I’d expect in it (summaries of the propositions and a level-headed disinterested analysis of the effects it might have). But it also has arguments in favor and opposed (and rebuttals to those arguments), which can apparently be submitted by anyone in CA, although certain types of arguments get priority if multiple arguments are submitted. And they’re fascinating.
I’ll be the first to admit that I seek out certain kinds of spaces to have political discussions in, and they tend to be fairly low on emotion and high on principle and wonkiness. I deliberately avoid some types of political discussions, in other words, and yet here they are in the official voter guide! Here’s maybe the best one, on the question of whether grocery stores (etc) should be banned from giving out single-use plastic bags.
(I’ll note that there’s an argument against this kind of law that the thick plastic bags are so much worse for the environment that it’s not clear these laws help if lots of people buy them instead of reusing. But that argument doesn’t appear here.)
Anyway, I didn’t know this existed, or that something like a voter guide distributed in the most populous state in the nation could produce arguments with the same flavor as letters to the editor in my Midwestern hometown (both the good and the bad).
The question that’s been weighing on each of our minds, but that we have been unwilling to acknowledge—until now: If I drink a shot each time a Democrat wins a Senate race, up until I know they will control the Chamber, will I end up in the hospital?
This ends up being a tougher question then I’d bargained on. The answer depends on –
1. The current partisan split among Senators,
2. The list of states hosting senate races,
3. The party affiliation of each of these challenged incumbents,
4. The time when polls close in those states (and thus, when journalists will make forecasts),
5. The relevant time zones for each state (Many states operate in more than one time zone!),
6. The alcoholic content of the shot,
7. My weight and metabolism,
8. Etc.
Yikes! Good think I didn’t start drinking early. Here’s my take:
According to FiveThirtyEight, there are basically ten senate races remotely in contention: FL, IL, IN, KY, MO, NC, NH, NV, PA, and WI.
Unfortunately for Democrats, that party start out with only 46 Senators. So they’ll have to win five seats to gain a majority.
Fortunately for Democrats, they only have to defend one of the ten contested seats; the rest of the challenged incumbents are Republicans. By my count, Dems need gain four seats plus keep their existing seat (or win a different one elsewhere): 4+1 means the Dems need to win 5 seats – out of 10 possibilities. Even odds, huh?
More fortunately for Democrats, if they win the Vice-Presidency, that will count as a seat. So long as they can muster 50 votes, they can battle all issues to a tie—and Vice President Tim Kaine would then pass the deciding vote. (OK, smarty-pants, set aside the cloture rules for an evening, ok?) So arguably the Dems only need four seats, not five. But this complicates the analysis: When will we know if Clinton/Kaine have won?
Anyway, here’s a list of closing times for polls, by state. Interestingly, at least three of the states in contention—FL, IN, and KY—span the Eastern Time/Central Time line, and thus have polls closing at different times within the state. But it’s my understanding that journalists have not been reluctant to forecast the electoral outcomes in these states, even while some of the polls have remained open. So for practical purposes, they might count as Eastern Time states.
To complicate matters further, the last competitive Senate race of the night—in Nevada—is also the sole Democratic Senator in contention. So if I’m going to get to bed sooner than 10pm, I’m back to needing five Republican incumbents to face defeat.
Last step: Based on FiveThirtyEIght forecasts, here’s a rough schedule of my alcohol consumption:
• 6pm: Early polls close in IN, KY, with Dems forecast to win the former. One shot.
• 7pm: Early polls close in FL. Who knows when they’d actually get any poll results—but it probably won’t matter, ‘cuz Dems won’t beat Rubio. No shot for you. But there’s some chance that journalists, seeing that Trump has lost FL, will call the election for Clinton/Kaine. So I may wet my whistle yet.
• 7:30pm: Polls close in NC. This one is a nail-biter. FiveThrityEight says no. But Hamilton says I’m not giving up my shot! Who ya gonna believe?
• 8pm: Polls close in IL, MO, NH, PA, and Democrats run the table. Four shots, and I’m down for the night! And once journalists call PA for Clinton/Kaine, you gotta suspect that they’ll be calling the election, too. (Probably it’s been an hour or two since my last shot, so I needn’t end up in the emergency room. Not like during that second debate. Who would have guessed that agreeing to take a shot every time someone says “DISASTER” would turn out to be such a, well, disaster…?)
• 9pm: If things don’t turn out as well as forecast for Dems, or if you have more stamina than I, you might stay up to drink to WI’s Democratic victory.
• 10pm: Finally, we’ll be able to place all the nametags on the Senate seats when the polls close in NV. Looking good for Dems here, too.
Good night!
Indiana is weird — only a little bit of it is in Central time, but that little bit includes Gary, which is something like 85% black. If you’re watching Indiana numbers come in live, you can pretty much always assume that there’s going to be a huge jump in the number of Democratic votes all the way at the end of the counting. (Also, that you’ll have The Music Man stuck in your head all week.)
The best political ad of 2016, says the tweet, and I’m not sure I’d disagree.
A VERY MERRY UNBIRTHDAY TO ALL!
(Void for people actually born on this date.)
Man fired for off-the-job hate speech.
http://www.phillyvoice.com/pa-man-fired-over-racist-tirade-anti-fracking-protest/
Thanks, Nobody!
And Nancy, thanks for the link. It’s hard to feel sorry for that racist dumbass, but as a general principle I think we’re all better off if it were a norm that people aren’t fired for what they say when they’re not at work.
The president of South Korea was probably in the thrall of a horrifying cult leader for most of her adult life.
Ampersand, you’re welcome.
I’m wondering whether “I didn’t believe what I said, I was just saying what I thought would hurt the most” is an excuse in any meaninful sense.
From #1:
Do you think that the Federal government should have to fund the exercise of a fundamental right for people who cannot afford to do so themselves? Consider that the right to keep and bear arms is a fundamental right.
In my home State of Illinois, gerrymandering gives the Democrats +2 House seats over what they would have otherwise.
RonF.
Re: Gerrymandering. Yup, and that’s horrible. We need to end gerrymandering nationwide, including in Dem controlled and districted states. Which means recognizing that it is a infringement on the right to vote, and a violation of equal protection, and acting accordingly.
When House representation is 10 points off of generic congressional ballot, something is deeply wrong.
The thing is, even with using more natural political or geographic boundaries you can get House representation that varies from a generic ballot.
For example, let’s take a State that has an urban center. For simplicity’s sake, figure the urban center has enough people to form one House district, a suburban area that has the same number of people, and 2 rural districts. The urban district goes 90% D / 10% R. The suburban district goes 55% R/45% D. The two rural districts go 65% R/35% D. The State as a whole is 51% D / 49% R. So a generic Statewide ballot would give you roughly even representation, but the actual representation will go 3:1 R. You don’t need gerrymandering to get an imbalance; it’s an issue of the fact that different areas have different interests, and that we live in a Federal republic, not a pure democracy.
I wouldn’t say it’s an infringement on the right to vote. The right to vote means that if you are a citizen of voting age and have residency in your district, you get to vote. It doesn’t mean that the party you favor wins.
Re: #28
If the professor was wearing the “Black Lives Matter” T-Shirt during class he/she is teaching, I think the students would have a point. The professor is in the classroom to teach law, not to make a point of his or her political opinions. OTOH, if he’s just wearing it around campus, too bad for the students – just as it’s too bad for other students or faculty if the students are chalking “Trump 2016” on campus sidewalks.
There’s some natural variance to districting, sure, but until the great 2010 gerrymander (project REDMAP), the difference between the national congressional ballot and actual representation was within a much smaller range. Gerrymandering has gotten much, much stronger with the advent of computer graphing.
Here’s a chart of generic congressional vote vs. representation since WWII . Note the post 2010 years.
No one’s saying that “the party you favor wins.” That is just a straw man. What we’re saying is that the party that the majority of voters favor should win most of the time in the House of Representatives, which is supposed to have proportional representation. When things are so gerrymandered that one party can lose the overall popular vote by millions of votes, that’s a serious problem. When the party in that position is also the one that benefits from the lack of proportional representation in the Senate (as the Republicans currently do), it is an even bigger problem. Structurally, people in large, urban areas are underrepresented in both the House of Representatives and the Senate right now. Yet, it is conservative rural voters screaming about vote rigging.
The House is supposed to have proportional representation by a combination of population and geographic region and the Senate by sovereign political entity; neither is supposed to reflect party affiliations. The U.S. Constitution (unlike a Westminster style Parliamentary system) does not recognize the concept of political party and makes no provision for a role for them in government. I think it would be a very bad idea to start now. Congressional districts were intended to be constituted on the basis of geographic and municipal boundaries, not those of party affiliation. That’s why the creation of a convoluted district for Elbridge Gerry’s election to the First Congress was so notable.
What do you mean by “people in large urban areas are underrepresented in both the House of Representatives and the Senate”? On what basis?
Here in Illinois the urban Congressional districts are deliberately shaped in such a fashion as to include portions of the surrounding suburbs as well, so that the overwhelming Democratic majority in Chicago and East St. Louis can be used to offset the Republican voters in the suburbs. I find it difficult to believe that we are the only blue State that has this phenomenon. Note that in my example the imbalance between State-wide party voting and the actual representation is not a function of gerrymandering, it comes naturally from the political leanings of the geographic regions.
I’ve considered that there should be some way to base legislation on some formula wherein no Congressional district’s ratio between boundary length and area could exceed a given (and somewhat arbitrary) number. This would make convoluted districts such as Illinois’ Fourth District illegal.
The reason why that graph looks like it does can be accounted for by the fact that the Democratic party appeals to a large number of people but those people are concentrated into cities. So States that don’t have a high proportion of urban population are not going to go Democratic. If the Democratic party could broaden it’s appeal to non-urban voters they could take the Senate and the House.
In the purple state where I live, most of the “urban” districts include part of a city, a bunch of suburbs, and a whole lot of surrounding rural area. The city that I live in, plus the first ring of suburbs, would be about the right population to be a district, but instead, my district is half the city, half the suburbs, and about four or five rural counties. The other half of the city is in a separate district, which is composed similarly. My district has been Republican for decades; the other one swings back and forth. I honestly have no clue what sort of local interests my representative actually represents — his district is defined so weirdly that it includes urban, suburban, and rural areas. (The other district within my city also includes a couple of smaller cities nearby. My district is just my half of the city, a bit of strip-mall suburbia, and farms.)
Nancy:
I don’t think it is. Someone willing to be racist for instrumental purposes is, to my way of thinking, still being racist. In a way, claiming that he knows better makes it even worse.
I went to do some digging on gerrymandering, as I remembered some recent academic work had been done on quantifying it.
First, there’s the efficiency gap. That takes election results and measures how many “wasted” votes there are–that is, votes that either didn’t go to a winner of an election, or were in excess of the 50% + 1 vote that are needed to win. Gerrymandering means winning more seats based on the same number of votes, so gerrymandered districts show a discrepancy in the fraction of wasted votes per party. The measure is simple, but is subject to some weirdnesses (as the paper itself details).
A more complex thing to do is make a bunch of fake district maps and use precinct-level results to see what the distribution of election results would be, and then to compare the actual results to that distribution. That nicely captures the point about spatially-clustered Democrats leading to some “natural” gerrymandering, since it’s included in the maps. However, it’s more complicated, which means among other things it’s more subject to errors & assumptions made by the people who made the maps; still, this looks reliable to me (though I’d love to hear criticisms if people know of them!). This group did the simulated-maps thing. What’s interesting is they find a net gain of only about 1 House seat for Republicans based on gerrymandering. That’s because of a combination of California, which they find has the largest gerrymandering magnitude in favor of the Democrats, and a pro-Democratic bias caused by VRA-mandated districting plans; those two effects combined almost outweigh the partisan gerrymander in many Republican states.
However, I’ll note that they use 2008 precinct-level results to predict post-2010-plan Congressional results, and I think 2008 was weird enough in several ways that this might fail. (In particular, 2008 was such a sweep for the Democrats that some closely-gerrymandered districts might have gone Democratic in the simulation that actually didn’t in 2012 or 2014, since the gap between D and R went from ~7% to ~4% in the presidential race.) So my guess is that this is a best case scenario in terms of how many seats Democrats lost due to gerrymandering, and the real answer is probably higher.
Anyway, I found both those things interesting reading; thought you folks might as well.
And to address that interesting chart from Ben Lehman: 2012 is clearly an outlier, but I’m not sure about 2014. That grey blob is just the convex shape that includes all the pre-2012 data points and nothing else. If you think it should be a linear trend–and it kinds of looks like that–and you imagine a grey band that follows the trend, but has some width to it, I think 2014 would actually be in that band extrapolated down. It’s below the grey blob just because it’s so rare for the Democratic share of the vote to be that low, and all the previous low points happened to favor the Democrats relative to the usual trend.
Why was 2014 less unusual than 2012? I think might get into a point mentioned in the efficiency gap paper above: the distribution of votes in districts can change quickly enough that gerrymandered maps become less gerrymandered, both because people move and because party affiliations can change as the parties do. Or maybe it’s just statistical noise, or maybe it’s that gerrymandering’s less effective when one party’s support drops unusually low. Shit’s complicated, man. :)
ruchama:
“I honestly have no clue what sort of local interests my representative actually represents — his district is defined so weirdly that it includes urban, suburban, and rural areas.”
My guess is that his interests are intensely local – local to his own person, which he wants to make sure gets re-elected every two years. This is why districts should be compact and based on natural geographic and municipal boundaries – so there some kind of consistent set of interests that the Representatives is responsible for.
To reiterate the point – given the disparate kinds of interests that the two parties represent and given the unevenness of the distribution of the people with those interests, there is no particular reason why the representation in the House should match some generic or average ballot.
In Illinois the Democratic party (which controls the General Assembly and thus controls re-apportionment) is careful to ensure that the number of urban voters in the mixed urban/suburban districts outweigh the suburban voters. It sounds like the opposite is happening in your State – the re-apportioning authority is making sure the rural/suburban votes outnumber the urban votes.
(Decided to stop being vague about which mid-sized Midwestern city I currently live in.) This is what the local Democratic party is dealing with. They’re advocating a write-in candidate. No one has any clue how this guy keeps winning primaries. http://www.cincinnati.com/story/news/politics/elections/2016/06/19/zombie-campaign-congress-giving-democrats-fits/82892598/
For the most irritatingly designed districts I’ve ever seen: zoom in on Austin. There are something like five or six districts that have little bits of the city plus a ton of surrounding area. Some of those districts spread out a few hundred miles beyond the city. None of the other Texas cities look like that. https://upload.wikimedia.org/wikipedia/commons/5/53/Pagecgd113_tx.pdf
By the way, I’ll be in Houston this weekend. Mentioning just in case anyone here is in that area. :-)