An Observation on Geography-Based Representation
A thought inspired by the PA Senate race.
In the comment section of my post on the possibility of a spurious majority in the next Congress, a discussion broke out about the nature of representation and its linkage to geography. I have, as regular readers might expect, a lot of thoughts on this subject. I am not sure I am in the mood for a full exploration of this issue this afternoon, but I will say the following. I think that in the rank-ordered preferences that should guide the evaluation/construction of democratic systems, the interests of people are more important than the interests of places. Indeed, and moreover, I would note that the only reason that places have interests in some abstract sense is because the people who live there have interests.
So, yes, I fully understand that localized interests (such as different industries, specific resource issues, etc) are quite relevant to legislative representation, I think that we still over-focus on the importance of location, whether it is in terms of things like “Red States” and “Blue States” which suggests monolithic ideological and political preferences that are radically simplified (after all, all states are actually some shade of purple, even California and Mississippi). Likewise, our reliance on single-seat districts is an example of over-emphasis on location despite all the known pathologies of that structure. Indeed, there simply is no such thing as a fair system of drawing those lines, yet reformers are often obsessed with finding it, certain that because we have done it this way for so long means it can be fixed (which ignores the question of whether or memories of the past functionality of US politics are warranted, among other issues).
That is already more than I intended to say on this general issue. What brings me to the keyboard this afternoon is Mehmet Oz (and to a lesser degree Hillary Clinton and Herschel Walker). And before I continue, let me note that this is not supposed to be some slam-dunk example of what I am discussing either above or in the comment thread that inspired these thoughts. This post is more an example of something that I think illustrates that despite all the talk about geography and place, the revealed preferences of voters via elections are that they care more about interests/ideology/party than they do about the alleged specialness of place.
I was listening to a news story run down the Senate races this morning and it was noted that Oz is really from New Jersey, not Pennsylvania and yet he has a very legitimate chance of becoming the next U.S. Senator from Pennsylvania (a 29% chance according to FiveThirtyEight). Even if he loses, he will win a substantial percentage of the vote (the polling average puts him at 47.1%).
So, here’s the question: if the Senate is supposed to represent the interests of the state why would it be so easy for a person not from a given state to have a real chance at representing that state? Honestly, you would think that not being from a place would be a bigger deal in a Senate election than any other kind of election. And yet, Hillary Clinton strolled into one of the Senate seats in New York a couple of decades ago, as did Tommy Tuberville in Alabama a couple of years ago. Sure, he coached at Auburn, but as best I recall he was living in Florida when he decided to run for the US Senate.
Herschel Walker is certainly from Georgia, but he has lived his entire adult life outside the state.
I do recognize that these are all celebrity candidates and that is a factor that needs to be taken into consideration, but if the interests of place were in fact as important as it often claimed (especially by people who want to return to appointed Senators or who argue that the Senate’s representational skew is worth because of the importance of state-level interests) then why do voters often tell us, by voting for newbies in these race, that maybe other factors are more important?
It is incontrovertibly true that John Fetterman knows more about Pennsylvania and its needs than does Mehmet Oz. And maybe that alone will sway some voters. But as an act of a mass public, millions will vote for Oz despite his non-PA credentials. Those voters will be saying that being from and of the state is not as important as the policy/philosophical/ideological signifiers of party affiliation.
Maybe regionality (NJ is right next door, after all) or maybe the assumption is that it is more likely that the carpetbagging Republican can learn to understand Pennsylvania’s needs than it is likely for the Democrat to vote in the Senate like a Republican. Put another way: the interests of most Pennsylvanian Republicans are defined not by simply being Pennsylvanians but rather by being Republicans.
I realize to some degree all of this is obvious, but I think it also is relevant to an understanding of representation and what is means to represent a specific set of voters.
This also connects, yet again, to the issue of primary elections as nominating mechanisms as being distinct from the general election. What might help a given candidate to win the nomination, such as celebrity status despite other flaws, becomes overpowered in rank-ordered preferences of voters in the general election when partisan choices become the central issue.
Regardless, the point that I am trying to make is the specifics of place are not as unique as the general discourse often suggests that it is. For example, practically any Republican from any state would represent the overall interests of Alabama in the Senate better than the most local of Democrats because the needs of place can be learned and assessed while partisan affiliation and ideological points of view are more fixed.
Quite frankly, when it comes to voting behavior, we can even see plenty of examples wherein politicians who can handily win re-election often vote against the objective interests of their states/districts because when it comes to actual voter behavior, partisanship is more important (based, at least, on how people act in the aggregate).
I would note that part of the reason we see members of the legislature vote against their state/district’s interests is that it is more important to appear ideologically pure for the primary than it is to appear putting their constituents first when it comes to the general election campaign.
Having said all of that (and as is often the case the post is longer than expected, and yet clearly does not cover the topics raised comprehensively) I think that it is noteworthy that despite our national mythology about the uniqueness of place when it comes to representation, especially the importance of state’s interests (or dare I say their rights) that voters are frequently willing to entrust their vote to persons who really have no deep connection (if any true connection at all) to a given state.
I think that fact undercuts the notion that representation of place is as important to voters as is representation of specific interests (some of which may overlap with place, to be sure, but still governing philosophy is the key).
Again states and districts only have interests insofar as the people residing in them have those interests. This means, I would argue from a democratic theory point of view, that we should be thinking more in terms of how to maximize the representation of those interests over too focus on geography.
Let me stress again, this post is not a theoretical presentation on the broader question of how to discuss representation. And, moreover, I am not arguing for an utter divorce of geography from representation.* But I think the examples (Clinton, Oz, Tuberville, and Walker) all illustrate that connection to and understanding of place is less important to a lot of voters than is often assumed in American political discourse. Indeed, I don’t think these examples prove much of anything, but I do think that these examples should make us pause and think about what are often thought to be foundational elements of our politics.
(BTW, I picked the photo initially because it was a public domain shot of Oz, but the more I look at it, the more it is symbolic of how poorly prepared Oz is for the job he now seeks).
*I would note that most representative democracies link representation to place to some degree. Two prominent counter-examples would be the Israeli Knesset which is elected in one national district and the Colombian Senate, which is also elected in a single national district. I am not arguing for such a system.
Carpetbagger candidates like Clinton and Oz would certainly be anathema to the Framers. Part of it is clearly the nationalization of politics and, especially, of party politics. There are likely plenty of places where either could have won their party primary, even with no ties. And Nick Kristoff might very well have won in Oregon had he been ruled eligible.
I’m not sure I have any objection to Georgia deciding they want a native son who hasn’t lived there in years representing them. For example, lots of military officers run for Congress in states they haven’t lived in for 20-40 years. But it does tend to give lie to the notion that there’s anything especially “local” to the representation.
I also would never argue for a single nationwide district electoral system (although I would argue that the system Israel has is quite a good fit for that country). However, the Colombian Senate is quite interesting from the perspective of this post. It really allows voters to decide whether they care more about locality or interests that transcend localities. And as you, Steven, showed in your book on Colombia, quite a lot of voters opted for candidates without a strong base in a locality. (This is probably changed a bit since the further reform, but the basic point surely still applies.)
Steven, I think it is the only factor that needs to be taken into consideration.
@James Joyner: @James Joyner: Yes, an issue being implicitly raised here is what does it mean to be “local” (or “from the state”)? Is having been born there sufficient? Or must one had lived there as an adult? Have worked for local/state voters or business or other interests in some capacity?
I always like to point out how strangely relevant mere birth in a place seems to be. John Kerry ran ads in Colorado for his presidential primary campaign stating that he had been born in a military hospital in the state, even though he never actually lived there. Apparently his campaign thought there were voters who would be impressed by that.
@James Joyner:
I think you are eliding a whole lot of difference between candidates when you talk about “carpetbaggers”, especially since the original meaning (person from the North coming down to take commercial advantage of the chaos during Southern Reconstruction) doesn’t even apply. The Clintons legitimately moved to New York. Bill wanted to set up his foundation in NYC, and did so, in Harlem. They bought a place upstate and stayed there even when Bill was in the White House and moved there immediately after he left. I lived in NY state a bit west and south of where they live but still run into friends of friends who live in and around the small Hudson Valley town they reside in, and as recently as last week they report that both Bill and Hillary show up in town to shop, go the the bookstore and eat at the restaurants. And she was my Senator while she lived there and I can personally vouch for the fact that she knew our local issues better than 99% of the NY state politicians, including any but the most local of our reps.
Oz, on the other hand, picked PA out on a political map and has no meaningful connection to the state.
@James Joyner:
And the nationalization of the nation.
Being from Virginia used to mean something, back in an era with no mass media, poor transportation, and what-not.
Now? What is a Virginian, and how does it differ from a North Carolinian, Kentuckian or even an upstate New Yorker (if you match based on type of neighborhood — lot size, income, and metro size)?
States don’t really matter in most people’s day-to-day lives. Not the way they used to.
Also, the founders would find Clinton an anathema because she is a woman and Oz an anathema because he’s a Turk — they might find Dr. Oz’s exotic, oriental medical advice a bit more intriguing than leeches though.
@MarkedMan:
Totally unfair. Oz can see Philadelphia from the upper floors of his mansion.
@MarkedMan: Hillary became an excellent Senator and is just naturally the sort of person to do her homework. But that doesn’t change the fact she moved there in November 1999 in order to set up residency for the Senate campaign.
As to the “carpetbagger” label . . . I mean, the Clintons came from Arkansas, where Bill had been governor for 12 of 14 years and before that attorney general. The move to New York was strategic. First, it was a winnable and open Senate seat. Second, Hillary was more popular there than in Arkansas. Third, it’s a hell of a lot better place than Arkansas to live if you’re (about to become) rich and (already) famous.
At the founding there were big regional differences. The biggie, whether or not you could own people, but also whether we would follow a New England Anglophile foreign policy or southern Francophile, protective tariffs for manufacturing, etc. The expectation of sectional conflict seems to be one reason the Founders missed on anticipating parties.
I’ve lived in eight states. They compete for highway funds and such but there’s been no particular sectional issue. Yes, ND and IA had large agricultural interests, but so do IL and FL. The geographical difference now is cultural, rural v urban. But that’s an intrastate thing, as true in NY as in Pennsatucky. And Republicans have thoroughly nationalized their culture war.
@James Joyner:
Although Arkansas never accepted her as one of their own even after all those years — she was just that woman from Chicago that Bill hooked up with while he was off gallivanting at school Back East.
@James Joyner: I just want to note that your explanation of why the Clintons count as carpetbaggers remains lacking given that his complaint was
I think it is more tribalism than nationalism. Pick any random Republican from anywhere and place them on the ballot and they will receive pretty much all of the GOP votes. The Dems may eventually also do this but they haven’t tested the system by running someone as awful as Oz ro Walker.
Steve
@James Joyner: Sure, but “carpetbagger” implies that she had no interest or connection to the state. True of Oz bout most certainly not true of the Clintons. Hell, they are continuing to live there and be part of the community in their old age. Carpetbagger? Really?
@Just nutha ignint cracker: That the term was coined during one era doesn’t mean it doesn’t evolve. When you Google the term, “a political candidate who seeks election in an area where they have no local connections” is the first thing that comes up.
@MarkedMan: She’s as much a New Yorker as anything else in 2020. She was a carpetbagger in 1999, when she moved there to run for the open Senate seat. Hell, for all we know, Oz will become a die-hard Pennsylvanian in 20 years if he wins the Senate seat.
Up until WWII, location was, in my view, a package for a set of beliefs. In, Oregon, my home state, the “Oregon way” has been significantly diffused first by transplants coming for Federal government dam building projects, Californians moving north, and, now, high tech jobs that allow people to work remotely. Probably other migrations as well, but these are the most significant in my lifetime.
Your post resonates because I believe the next Oregon governor will be a conservative who represents her political tribe more than the moderate Oregon way in place during my lifetime regardless of the party in power.
BTW, our current Governor Brown is Jerry Brown’s sister. In my admittedly biased view, I believe she tried to represent rural Oregon as well as urban, Oregon, but she drew the Covid card and has been vilified for vaccination, shut-downs, masking, and ennsuring minorities received equitable access to care.
@MarkedMan:
I actually think that that is too reductive, but setting aside any objections I might have to your conclusion I would note that it still makes my case: voters don’t care about localness as much as they do other factors.
@Matthew Shugart:
The reforms would have made that behavior a lot harder to replicate, but I suspect there is still some of it to be seen. Maybe I need to go back and take a look!
@James Joyner:
To be clear, I have no objection either. Indeed, I have no objection if the people of NY want to elect someone who just moved to town. My point is that all of this undercuts notions about the unique representational nature of place.
@MarkedMan: @Just nutha ignint cracker: @James Joyner: Yeah, the meaning of “carpetbagger” has evolved. James is right about its application here. I used it only once in the OP because while it does describe the phenomenon under discussion its connotation is still more negative than I wanted.
But people are still doing the voting, not places. But then you also state that geography is important:
“To some degree.” So the real debate here, then is to what extent representation should be subject to subsidiarity and how should representation be devolved or divided into smaller political units (which will very likely end up being based on geography)?
If we scale that, on one extreme end, we have everyone as their own representative – direct democracy. At the other end, you have a single level of government representation with no geographic representation at all.
Both of these extremes are bad – I think we probably agree on that much.
But it’s not really clear to me where you actually stand. If the goal is to get near-perfect partisan representation in the national legislature, then ending geography entirely makes sense.
But you say you’re not arguing for that? You frequently argue against geographic representation, but you don’t want to get rid of it entirely. So there is a limiting principle to your people>places argument, but I don’t where you are drawing the line, or the justifications for why it should be drawn there. What are the objective criteria here?
More fundamentally, it’s unclear to me why national-level representation should be considered more important or better than state, local, or district representation, particularly in a huge country with a lot of diversity, which seems to be the thrust of your arguments on this.
My own view is that I think the balance in the US is about right for our country. I would maybe accept small reforms that could add some number of “at large” seats to the House and/or Senate, but much depends on the details. But that is mostly academic because such changes would essentially blow up the existing Constitution, which I think would be dangerous.
First of all, I don’t think that’s true. You don’t see most people clamoring to get rid of the current system so they can vote purely based on ideology. That many people do vote based on interests/ideology/party doesn’t prove that they want a system that would give them no other choice. Even people who vote based on national politics, partisanship, or ideology still want a representative that is accountable to them and their community.
Secondly, so what? The nationalization of politics is a recent trend that may not last, so why should we change our system to make national politics even more important and more zero-sum, simply based on aggregate voting patterns?
Nationalized offices and elections also have some serious downsides to consider. Making these elections national would promote exactly the kinds of candidates that are now carpetbaggers – ie. famous people and professional politicians who seek political power. Is “representation” really better when, due to the size of our country, the only competitive candidates will be famous people who have the $$$ necessary for a national campaign? What’s the evidence that will result in better legislative governance?
I also think it’s a mistake to look at representation only in partisan terms. Your concern that aggregate vote totals don’t match partisan representation in the House and/or Senate isn’t a problem in my view because I don’t think partisan representation is that important. Things would be different if our parties were coherent entities that could control their brand, but they are not. An AOC Democrat is much different from a Manchin Democrat for instance. Also, many people vote based on the partisan brand, but not everyone does that. So focusing on representation in terms of partisanship is a mistake IMO.
@Andy:
I think this gets to a lot of why you often disagree with things I write about–I don’t think you understand the role that parties need to play in representation in mass elections and why a system that forces AOC and the non-AOC Democrat into the same party or the Marjorie Taylor Green and Ben Sasse into the same party is a problem.
I also don’t think you understand the institutional options and how they might, or might not, foster an AOC party, a moderate Dem party, a moderate Rep party, and a Q party (or what ever the configuration might be).
Parties are just subdivisions of the population that hopefully approximate groupings of interests.
I think one of the core problems of American politics is that we have a population that needs more than two parties but the structure of power and the incentives that drive it give us only two viable ones. I am constantly trying to explain this fact one way or another.
And it is less that I am fixed on parties as much as I am on mechanisms of representation.
@Steven L. Taylor:
I think I understand those things quite well. Indeed, I broadly agree with you (and have said so many times) regarding the goal of getting more than two parties in the US and many other problems. And I certainly do understand that achieving that goal requires some serious reforms.
The problem that you specifically identified on multiple occasions is that either Republicans or Democrats don’t receive a proportional share of seats if one looks at aggregate nationwide votes for the legislature. Even assuming that aggregate vote totals in hundreds of distinct and separate contests tell us anything about the actual political preferences of the population, this framing leads one to believe the goal is to ensure that each party receives its “fair share” so that the partisan “minority” doesn’t have too much power. Please correct me if that isn’t what you’ve consistently argued.
What isn’t clear to me is how making the current two-party system more representative of aggregate vote totals achieves the goal of a multi-party system, or is even a step on the road to such an end.
Said another way, I agree that it is a flawed system. But it’s a flawed system because it forces a binary choice – it’s not a flawed system because the binary choice doesn’t meet some proportional standard when looking at aggregate vote totals. So it’s not clear to me how focusing on this very niche problem (which I don’t think is much of a problem, considering it’s existed forever) illuminates the bigger problem of uncompetitive binary choice contests, which aren’t just confined to the House and Senate.
And I am still confused when you argue that geography is problematic, but you don’t want to get rid of it entirely. What is it, exactly, that you are proposing here that doesn’t ditch geography, yet makes it far less relevant?
@Andy:
Why not both?
Well, I would note that it is hardly the only thing I have focused on. I point it out because it shows that even on its own terms it doesn’t work very well. It seems reasonable to note the panoply of issues the system has.
In simple terms, a multi-seat district still uses geography.
I have also argued for MMP, which has single-seat districts but also uses a national vote to create proportional outcomes. So that would use some geographically delimited features as well.
I agree that you and I agree on a number of things. I just also fear that there are some clear areas in which I not connecting with you the way I would like to in terms of making sure you understand what I am trying to say regardless of whether you agree with it or not.