6

The US senate is frequently controlled by the Republican party because they have the support of many smaller population states, so that even if the Democratic party has more supporters overall, they are concentrated in higher population states. It has been proposed several times that large Democratic-leaning states could be divided in order to rebalance the senate in their favour. (This Wall Street Journal opinion piece is one example of a proponent for this plan.) Making new states out of existing states requires the approval of the existing state's legislature and the US Congress.

But would this really be a viable long-term strategy for the Democratic party? Couldn't the Republican party also attempt to split states the next time it got control of Congress, either the higher population southern states, or even the low population northern states like Wyoming and Montana? Splitting states seems like a strategy that is likely to invite tit-for-tat responses just as much as adding Supreme Court justices would. Is there anything that could prevent Republican states from negating any advantages the Democrats gain through state splitting?

6
  • 1
    Most red states struggle to have much of an economy - I'm not sure anyone is super excited about living in half of Montana. (I say this as someone with a lot of family and who loves Montana.) Plus - the larger the state, the more space for multiple identities to form. Nov 5 '20 at 4:10
  • 1
    @AzorAhai--hehim Splitting Montana would be pretty silly, but splitting Texas has been seriously considered. Florida as well, which could turn a swing state into two safe states (or would it be one safe and one swing?) Nov 5 '20 at 4:32
  • I was responding to that point in your Q (hence why I didn't answer). Nov 5 '20 at 5:34
  • Follow up question from a non-American: How open would Americans even be to have their state split? As far as I can see there is a certain amount of state-pride, how easily would that be given up? Nov 5 '20 at 6:59
  • 1
    Many other states might want to split of their major cities (Chicago, Atlanta, Minneapolis come to mind), but that would be unprecedented and less likely to be successful, IMO Nov 5 '20 at 16:39
7

It's hard to say.

For the sake of argument, let's ignore the immediate political barriers to this strategy (namely, the Democrats may not end up with a majority in the Senate at all, and if they do pull it off, it will likely be too narrow for extreme actions such as state-splitting). Instead, assume the Democrats have a political trifecta at both the national and state level (at least in the relevant blue states), and can do whatever they want.

If the Republicans try to retaliate with more state-splitting, the immediate problem they will run into is money. Red states, for the most part, tend to be less populous than blue states, and they also tend to have lower per-capita income and a greater reliance on federal funding of services like Medicaid. Texas is a notable exception to this pattern, but it has been trending towards the purple column for several election cycles now, and in the future, it may not be a Republican stronghold at all.

In order to split a state, you need to establish a new state government, which needs to duplicate all of the state laws and services operated by the original state government. This increase in overhead is expensive. But it's much less expensive, per-capita, if you start with a high-population state, because many parts of the state bureaucracy scale in a sublinear fashion with respect to population. In other words, larger states can be more efficient per capita, because some components of the state government scale well to larger populations. For example, there's always one governor, one state supreme court, and one (usually bicameral) legislature, regardless of how big the population is. So the Republicans, trying to split states up, will run into budget problems to a greater extent than the Democrats would.

On the other hand, this is a form of gerrymandering, and most gerrymanders don't last forever. Following the 2010 census, the Republican party heavily gerrymandered House districts throughout the United States, but they still lost the House in 2018, because of a combination of changing demographics, larger-than-expected partisan shifts, and electoral conditions that, overall, favored the Democrats.

When you make a gerrymander, your goal is, generally speaking, to reduce your number of wasted votes while maximizing your opponents' wasted votes. A vote is wasted if it is cast for the loser (in a given district), or if it is cast for the winner in excess of the number of votes cast for the loser (i.e. it's an "extra vote" that the winner didn't need in order to win). But there's an obvious problem here. To minimize your own wasted votes, you want your candidates to win as narrowly as possible in as many districts as possible (and, if you can't win in every district, you want your opponents to win by massive margins in the districts where you can't win, so their votes are wasted more than your own). However, you need to win by a large enough margin that your gerrymander can stand up to shifting political winds. If you give yourself only a +2 advantage in each district, then the moment the overall popular vote shifts -3 against you, you end up losing more seats than you would have under an equitable set of districts.

In practice, this means that, for each district, you need to figure out how far it is likely to shift, under the assumption that the national environment is strongly against your party, and then leave enough of a margin that you still barely eke out a win in that case. This is a delicate balance, since a too-small margin will doom you to the minority once enough voters decide to vote against you, but a too-large margin will make your gerrymander less effective than it otherwise would be. Since demographics change over time in unpredictable ways, it's basically impossible to consistently get this right in the long run.

How does this apply to state splitting? The problem with state splitting is that you don't get to try again every 10 years, unlike with Congressional districts. When you draw a new state boundary, it's there for good, unless you merge the states and re-split them (or split and then merge). Merging and splitting states repeatedly would generate a lot of bureaucratic churn and would not endear residents of those states to the party doing it, but if you don't do that, then sooner or later, one of those split-off states is going to become purple, and you'll lose a bunch of previously-safe electoral votes.

8
  • Ah, thanks for showing how state splitting is akin to gerrymandering. You're right that there's no guarantee the new states would keep their political makeup, even in relatively short terms. And it makes me think of another strategy Democrats could try: move large government agencies to red-leaning states. Large political-inclined companies could do the same with their office campuses. Nov 5 '20 at 4:47
  • I am not convinced that discussing gerrymandering really helps clarify things. This is not what the proposal is about. Assuming the state is homogeneous and you're not trying to optimize borders to bring about a particular result, splitting would still double the number of senators. Would that be viable legally and politically?
    – Relaxed
    Nov 5 '20 at 13:01
  • @Relaxed: No state is homogeneous, except perhaps for some of the extremely rural Republican strongholds. But the Democrats are almost always clustered around cities.
    – Kevin
    Nov 5 '20 at 17:53
  • @Kevin I think North vs. South California come pretty close for example, they are not homogeneous at the county level but the mix of cities and rural areas would be similar enough. That doesn't really matter though, my point is that if we assume they are, that still doubles the number of senators for a given population. The proposal doesn't fundamentally rely on your ability to gerrymander the states, you could even deliberately draw two halves to be similar.
    – Relaxed
    Nov 5 '20 at 20:48
  • 1
    Thinking about it, I think the discussion helped me get a better understanding of what was bothering me about the focus on gerrymandering. When you (re)draw district, you usually assume a fixed number of districts (and representatives) and try to optimize the outcome in some way. That's where the wasted vote analysis is relevant. But if you split a state, there are twice as many senators, no matter what. It works even for some hypothetical homogeneous state, which is not the case for gerrymandering. That's what makes it interesting in my view.
    – Relaxed
    Nov 5 '20 at 21:43

You must log in to answer this question.

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