Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
The GOP Nomination and the Rotten Boroughs Problem
Of late, the GOP nomination process has favored those on their second try: Reagan failed in 1976 before winning in ’80; George H. W. Bush failed in 1980 before winning in ’88; Dole failed in 1988 before winning in 1996; McCain failed in 2000 before winning in ’08; and Romney failed in 2008 before winning in ’12. (We’ll come back to George W. Bush, the exception that proves the rule, in a moment.) But we shouldn’t be too hard on the GOP nominees failing the first time around because the process depends heavily on understanding the power of the GOP’s rotten boroughs.
To win the GOP nomination, a candidate must win a majority of the 2,472 delegates, which are allocated through a complicated series of formulas. The biggest problem is that these formulas create rotten boroughs, where a relative handful of voters in liberal parts of the country control more delegates than larger numbers of conservative voters elsewhere.
Let’s use an absurd hypothetical to demonstrate the actual absurdity of the 2016 nomination process. Let’s say the ghost of Nelson Rockefeller wins every delegate in every state that Barack Obama carried in 2012, while the reanimated corpse of Barry Goldwater wins every delegate in every state that Mitt Romney carried in 2012. Rockefeller would win the Republican nomination with 1,231 delegates to Goldwater’s 1,166 (these don’t quite total 2,472 total delegates at stake in 2016 because I’m not counting the territories like Guam and Puerto Rico, but given that they tend to vote Democratic, they’ll likely wind up with Rockefeller rather than Goldwater). In other words, the blue states outweigh the red states when picking the nominee. The Ethics and Public Policy Center’s Henry Olsen likes to make the point that somewhat conservative and moderate Republicans have a dominant role to play in the nomination process, and the rotten borough effect is a significant reason why. Some actual numbers might help illustrate what’s happening.
If Rockefeller wins Vermont with 30,000 votes (it cast 60,850 votes in the 2012 primary), then he’d get 16 delegates: three for the congressional district, 10 for the state at-large, and three more for each local member of the RNC. But if Goldwater wins TX-13 (Amarillo, the most Republican district in Texas), then he might only get three delegates, even though more than 100,000 Republicans will likely vote in that primary. It’s uncertain because Texas’s delegates are doled out in proportional representation (unless someone wins a majority of the vote statewide, in which case that person takes all the votes), so just winning big in that one district isn’t enough: to get the three delegates of TX-13 requires a candidate to win a lot elsewhere in the state. In other words, Rockefeller got each of his 16 delegates for less than 2,000 votes each, while Goldwater will need to win more voters than voted in all of Vermont just to get the three delegates at stake in TX-13. Repeat that across multiple states, and it should no longer surprise anyone that old Rockefeller Republican families like the Romneys and the Bushes keep winning GOP nominations. The northeastern and midatlantic states are filled with rotten boroughs, which is why Gov. Chris Christie might be underrated. He’s probably the candidate closest to easy delegate pick ups in Delaware, New York City, and New England.
Further, how the delegates wind up with each candidate get seriously screwy. Some states are winner take all, others use proportional representation, still others use conventions, and others yet (like Texas in the example earlier) use proportional representation but become winner-take-all when the winning candidate clears a certain percentage of the vote. It’s like a chaotic videogame where you probably need to have played and lost at least once to have a chance to win. That probably explains why George W. Bush was able to win on his first outing: he was, effectively, a junior staffer on his father’s campaigns and already knew how to play.
We cannot totally eliminate some areas getting more weight than they deserve, but we can try to have it make more sense. If RNC Chair Reince Priebus (or his successor) wants a legacy, fixing the GOP nomination process so it better reflects the GOP’s voters would be a good start. Here’s my proposal to reform the process:
The first principle of my plan is that the process should give more weight to districts where the GOP is winning elections. Under the current system, it seems as though the less likely it is for Republicans to control their local government, the greater their say is when picking presidential candidates. Second, we should place less emphasis on the states and more emphasis on congressional districts, which better reflect the party’s base. Gerrymandering is less of a problem than you might think. Republicans tend to lose at gerrymanders in states where they aren’t winning much, like Illinois, Massachusetts, and Maryland, but they tend to win at them in states like Texas, Georgia, and North Carolina. The net effect of gerrymandering could be to have a presidential candidate represent the base more closely than a neutral drawing of the lines.
My proposed system is that the delegates should be determined by congressional districts and that these districts should be weighted according to elected officials, winner take all in each district. Each congressional district should get one delegate by default (hey, even Vermont and San Francisco have some Republicans). If the district is represented by a Republican in the US Congress, add four delegates, on the principle that we ought to give a boost to outposts of conservatism in otherwise liberal states, such as Staten Island (NY-11). Then, add an extra delegate for each each Republican governor or US senator representing that district.
So under this system, districts can range from being worth one to eight delegates. Here are some examples:
- CA-12, San Francisco, would be worth only one delegate, as it isn’t represented by any Republicans in the governor’s mansion or in its congressional delegation.
- MA-8, South Boston, would be worth two delegates, thanks to Republican Governor Charlie Baker.
- IL-3, Chicago, would be worth three delegates, thanks to Governor Bruce Rauner and Sen. Mark Kirk
- TX-35, Austin, would be worth four delegates, thanks to Governor Greg Abbott and Sens. Cornyn and Cruz.
- NY-11, Staten Island, would be worth five delegates thanks to Republican Rep. Daniel Donovan.
- Montana, an at-large state, would be worth six delegates thanks to Rep. Ryan Zinke and Sen. Steve Daines.
- FL-1, Chumuckla, would be worth seven delegates thanks to Rep. Jeff Miller, Sen Rubio, and Governor Scott.
- TX-1, Tyler, would be worth the maximum eight delegates.
Right now, the relative handful of Republicans in states liberal leaning states get a hugely disproportionate say in the presidential nominee. For example, California gets 172 delegates to Texas’s 155. This is crazy, since California Republicans run next to nothing in their state, while Texas Republicans utterly dominate theirs. Under my plan, California would get 109 delegates to Texas’s 244.
My proposal isn’t perfect, since some places will still get more say than they deserve, but it moves the nomination process in a more representative direction. It also means that we might finally start getting GOP nominees that reflect the base.
The rotten boroughs are probably going to pull the eventual nominee leftward this year, but maybe we can get smart about it by 2020.
Published in General
Great article. Thanks for doing this math.
It’s the same issue with the electoral college. Millions of voters in Upstate New York, Downstate Illinois and fly-over country are overshadowed by NYC, bussed in voters in Philly and dead Chicagoans.
I vote for a good old fashioned smoke-filled room.
Fascinating. On a forst read-through, one flaw might be that the increased focus on #Winning would increase the prevalence of unprincipled win-by-whatever moderates.
But the system is already screwy, and a more transparent system might be medicine enough to make the gaming of the system less rewarding, and a focus on principles more so.
Thank you for this concise summary and interesting proposal. Good stuff. I suspect that some of our podcast players not named Rick Wilson or Mike Murphy would find this engaging.
That makes a lot more sense.
As you say, any system will contain some degree of unfairness. The trick is to design it so that it’s unfair in the right way.
If your goal is to win the general and accurately represent what Republicans around the country actually want, you would apportion delegates based on raw numbers. The bonus delegate system you’re proposing would just be a different system of gerrymandering skewed towards more conservative districts. Maybe it would be better just to add up the total number of votes over the whole process and divide the delegates out proportionally with no winner take all states.
The quick way to eliminate the electoral college problem is to follow the Nebraska / Maine approach, which is highly Constitutional. One vote per congressional district, two extra votes for winning the state, which isolates the Blue area vote fraud. The Presidential candidates would have to appeal to the whole country.
We need a bottom up approach to control the county, not top down from Washington. I think Mike’s proposal is similar to NE/ME above, except his weighting +4 for congressional Republican districts IMHO is way too high, but it does reward bottom up.
One of the challenges to change is that this system is rooted inside republican party machinery. If you want to meet the walking talking definition of the establishment it is the people who came up with the existing formula.
Getting a seat at this table probably requires 6 or 7 figure donations and washing Reince’s car or something.
I am not proposing the existing formula is a conspiracy, just that it is institutionalized and will not be easy to modify.
I actually like that approach myself — and have advocated for it before — but that’s for the Electoral College, not the parties’ nomination.
So when Rob blames the primary voters for our outcomes, can I point to this and say “wait, it is rigged!”?
Is it rigged or just very poorly designed?
The way the current system works, my vote is meaningless. I live in Williamson County, Tennessee, which votes close to 75% Republican. One more vote from me isn’t going to change the outcome. (In sports, I would be accused of running up the score.) Of course, I do vote – but at this point it’s mostly theatrical. Kinda like the TSA screeners at the airport.
Well, the reason I went with +4 per district is that any less means that liberal oases in otherwise conservative states—think of Austin (TX-35)—would outweigh conservative bastions in liberal states—think Staten Island (NY-11) or Bakersfield (CA-23).
I agree. But I’m hoping that if one of the GOP outsiders (like Ted Cruz) wins the nomination, he can put someone in the RNC who’ll force through a reform.
If Ted Cruz gets the nomination we will have to put RNC headquarters on suicide watch and set up air mattresses around the building. It won’t be pretty.
I think it is designed to reflect the GOP strength of the turn of the last century. When McKinnley and TR and Coolidge were winning, the GOP’s strength was New England–including NYC–Chicago, the Mountain West, and California. This is also the time when the Primary became the method of nominating candidates. There was still the need to negotiate nominations during the convention, but these negotiations were reflected in the Primary process. So the design is poor because it reflects a reality that hasn’t existed for over 100 years.
I’ve lamented before on Ricochet that the primary “only” system (mostly since WWII) has not produced better candidates than the “smoke filled rooms.” I think Mike’s approach to giving more weight to individual congressional districts has merit, now let’s try to work out the details. For example, if we could find a primary system that rewards purple states to turn red, that would be wonderful. It’s insane that the presidential race is always chosen by Ohio, Florida, Virginia, North Carolina, and Colorado.
Weighting the delegates towards places that elect Republicans has some merit, but the reason that California has more delegates, despite Republicans there not controlling much, is that California has more Republicans. They may be liberal Republicans (though once you get off the coasts, the liberalism drops precipitously), but Republicans all the same. Even your TX-13 to VT comparison is unfair because you are comparing winning a large district to winning a small state. If you win the large state, you win a lot more delegates. The proportional representation of the South is because they set their primaries early -trading delegate numbers for the chance to winnow the field to their favorites. If Texas wishes to swamp Vermont, then they can vote later in the cycle. As it is, Texas is trading that power for the chance to knock out, say, Kasich, by denying him many delegates even if he can pull a bunch of votes in that large state.
If you want to balance the factions of the party, then you need to up the number of required delegates -to 2/3 or 3/5 maybe. Let the conventions actually serve as conventions after the first ballot, and let them choose a candidate acceptable to an overwhelming majority of the party.
Adding up the total number of votes would be like direct voting for the President. It’s too easy to game the system due to vote fraud. Also, since primaries are presently held on different days, the Main Stream Media and others could more easily manipulate the outcome by their coverage. And it doesn’t do anything to help the purple states turn red.
The reason why I want different parts of the country weighted differently is that it’ll make it easier to get a nominee. I want it skewed towards places where Republicans are actually winning elections and making policy.
After the fiasco that was 1968, the Democrats did more or less what you suggested: they made every contest proportional representation: that’s how they got George McGovern as the nominee.
Every state in the Democratic presidential primary is still proportional representation, but they’ve hedged their bets with superdelegates. The other problem with this approach is that without some winner take all aspects, you get a grinding nomination battle where one candidate is winning but cannot score a knockout punch, like the endless Clinton-Obama fight in 2008.
Pretty? No.
Beautiful? Yes.
I made that comparison to point out the rotten borough effect, but I can make a other comparisons to illustrate the weakness of the current system. Right now, California gets 172 delegates to Florida’s 99. Florida sends 17 Republicans to congress versus California’s 14, plus Florida Republicans control one Senate seat the the Governorship. Under my proposed plan, California would still get 109 delegates, but Florida would get 149. Given that Florida is actually a swing state where Republicans often win statewide, shouldn’t its residents get more of a say in the GOP nomination than California’s?
There are 2 different considerations winning and accurately representing the base of the party. Directly proportional apportionment represents what the base of the party wants. If you want to focus on winning instead of reflective representation, you would move primaries in purple states forward in the primary cycle. Followed by marginally red states taken from a geographically disparate areas representing varied concerns to solidify the nomination. Red states that are probably staying red like Texas and blue states probably staying blue like California would be moved to the back of the cycle, hopefully the nomination would be sewn up by the time it got to them.
I agree that until we get the NE/ME type Electoral College voting, the main thrust is the purple states. If the “Republicans” in the purple states knew better how to win that state, and could also change their neighbors to vote Republican, that would be the best outcome. However, as one who has voted in the Democrat primary (Rush Limbaugh “Operation Chaos”) not all “Republican” primary voters have the same motivation.
I agree that the way the GOP apportions delegates helps to select moderate candidates. The question is: Does selecting moderates help win presidential elections?
I think your plan has merit, but it would nice to figure out a way choose candidates that would better appeal to the swing voters in swing states, if that can even be done.
Our country has become so polarized that the presidential election is chosen by a small minority in a small number of states. It really is an metropolitan-rural/suburban divide more than anything else.
Great article, great commentary.
Should they? Florida has 20 million people, California has 39 million. Florida voted 49% for Romney, so with really simplified math, that’s 10 million GOP voters in Florida. California gave Romney 37%, that’s 14 million GOP voters. Should we really disenfranchise those 4 million conservative voters in the Inland Empire because they have the misfortune to share their state with San Francisco? Now, by all means 172 to 99 is probably too steep a lean towards the larger state, but the idea that California should get less votes than Florida is by no means obvious.
One practical reason to give California some delegates is Rob Long’s “skin-in-the-game.” I guess a sizable amount is raised from CA into the Republican coffers. It’s probably too much to ask of people, but since the Californians know their vote is effectively wasted, maybe those delegates can choose the best candidate, much like the old “smoke filled rooms.”
But I don’t think we’re disenfranchising them. They still get the +4 bonus for having a congressman like Paul Ruiz (CA-36) or Ken Calvert (CA-42) or Paul Cook (CA-8). The IE isn’t as conservative as it once was: Mary Bono, David Dreier, and Gary Miller are all gone. I wish my old stomping grounds were still as conservative as they once were (I grew up in Cucamonga) but they just aren’t.
California still get a ton of delegates from its size, but it really doesn’t shape the Republican party any more. It should get less of a say in picking the nominees.
One can’t automatically assume that because a district has relatively few Republicans, that those Republicans will be less conservative.
But that does seem to be the case.
To a first-order approximation, sheerly numeric, that assumption will be valid. Assume a normal distribution analyzed as two lesser normal distributions, and that’s what you’ll get.
But normalizing distributions is a business of assumptions in itself. It could be that being surrounded by lefties helps to radicalize conservatives.
It would be interesting to see where the mode is for conservatives as a function of percentage of the whole.
This suggestion is undemocratic. All states should be represented proportionally, according to population. Democracy, it’s called. Republicans living in blue states deserve fair representation on principle.
Fixing the system to empower “the base” makes it less likely that the party will ever regain strength in states presently captive to the Democrats. The goal should be the opposite. Have some candidates who can stir a populist uprising in blue states, and break the (D) stranglehold. We have several Governors in blue states now. Does this mean Massachusetts Republicans should get better treatment than Virginia ones? Ridiculous! Massachusetts and Maryland tipped because Democrats and Independents saw the light. Governors are ambitious, and some of the best ones like to think they have some shot at national office. It’s not fair to rig the game against them.
States like California are important to the party’s future. Actors, tech wizards, and billionaires can be useful in politics. If our (R) primary votes become meaningless, why vote in primaries, or contribute financial support to a party which discounts us? Jungle primary laws already leave some districts voting between the lesser of two Democrats in November. Abandoned, people will think maybe our votes will count more supporting a new fiscally conservative wing of the Democrat party. If two factions of Democrats control blue state districts, say goodbye to Republican congressional majorities.
2016 could conceivably demonstrate the value of alliances between “base” and blue state Republicans. What if Rubio and Bush wind up in a delegate race against Trump, Carson, and Cruz? Trump’s strength in some of the more expensive major media markets such as New York could defeat the Bush bankrolled forces. If the race goes to the convention floor, the choice could come down to Trump-Cruz vs. Rubio-Kasich. A lot of Rush Limbaugh listeners would think Trump-Cruz is the more conservative ticket.
Blue state conservatives aren’t naturally opposed to all conservative ideas. We Republicans in California often reject the religious right agenda, but embrace hard right positions on immigration, tax policy, gun rights, and the Dem’s efforts at redistribution which we’ve seen up close and personal.
Delegate rules are quiet, behind-the-scenes, deep in the weeds games. Try something like this, and the media spotlight would magnify the shootout, with purple state voters in states like Virginia caught in the crossfire. Not good.