Conventional Chaos, Part 2: Delegated Disloyalty
I have a lot of friends who are asking a lot of questions about brokered conventions these days. In this series, Conventional Chaos, I’ll be explaining how a Republican party convention works… and why 2016’s convention could be very different from the dull pageants we’ve seen since the 1970s. In Part 2, we'll take a close look at the delegates -- and why the official delegate count tells you less than you think. If you're just joining us, start with a basic overview of the convention process in Part 1. 1996 Delegate Credentials
So there will be 2,427 delegates at the 2016 Republican National Convention in Cleveland. But who are they? What will they do? What can they do?
Bound or Unbound?
As we discussed in our last post, most Republican delegates are bound, which means they are required to vote for a specific candidate at the convention, regardless of their personal feelings, at least up to a certain point. Some Republican delegates are unbound,* which means they can vote for whichever candidate they prefer, at any time. There will be approximately 170 officially unbound delegates at the convention (from Wyoming, North Dakota, Colorado, and several U.S. territorial possessions like Guam), plus 60-90** bound delegates who became unbound when their candidates dropped out. (Only a few states allow this.)
This means the convention will have around 230 unbound delegates and 2,242 bound delegates. (1,237 delegates is enough to form a majority.)
Who Will The Bound Delegates Be Bound To?†
The binding of the delegates is the great question of the primaries, to which the Associated Press and the Election Wizards devote nearly all their energies, and the primaries aren't over yet. As of today, 754 are bound to Donald Trump, 465 to Ted Cruz, 112 remain bound to Marco Rubio, 144 to John Kasich, 111 are unbound, and 10 remain bound to other minor candidates (Bush, Carson, Fiorina, and Paul).
By the time the convention rolls around, there are four likely scenarios. Bear these four scenarios in mind, because we'll be coming back to them in future posts. The first three scenarios are (more or less) taken from FiveThirtyEight; the last is based on Kasich dropping out.
Trump Barely Clinches - Trump wins enough delegates in the remaining states to have a majority of the delegates bound to him on the first ballot. This is known as "clinching," and it requires 1,237 bound delegates. Trump just barely gets there, but he gets there. (This is Trump's best likely scenario; it is widely considered unlikely he will clinch by a wide margin.)
Trump Barely Misses - Trump doesn't clinch, but he has enough delegates to put him in striking distance of a win, as long as he can rally just a handful of unbound delegates to him.
Trump Plurality - Trump can't get within close range of clinching, but goes into the convention with a plurality of delegates.
Kasich Drops Out - Kasich drops out (before Wisconsin) and Cruz (finally) is able to consolidate the non-Trump vote behind him. Trump still sweeps most of the Northeast, but Cruz makes up ground elsewhere (including a decisive win in California) and they enter the convention with Trump closer to Cruz than to clinching.
Here are some approximate delegate totals for each scenario, based on the map as it stands today (23 March 2016): Trump Clinches Trump Near-Miss Trump Plurality Kasich Out Trump 1240 1204 1146 1063 Cruz 693 686 739 913 Kasich 187 230 235 133 Rubio 112 112 112 112 Unbound 230 230 230 241 Other 10 10 10 10 Teaser for later: most people think that, if Trump "clinches" with 1,237 delegates or more, there is no way to prevent him from winning the nomination. As we'll discuss in later entries of this series, this is false. The delegates can still nominate someone else by amending Convention Rule 40, although doing so would no doubt carry a significant political cost. But, like I said, we'll come to that later. Let's get back to our delegates.
Who Will the Bound Delegates be Loyal To?
Delegates can be bound to vote for a candidate even if they don't personally support that candidate. It happens all the time, actually. In 2012, a number of insurgent Ron Paul supporters in various states managed to get themselves elected as state delegates in states Mitt Romney had won. These supporters were bound to vote for Romney, but were actually loyal to Paul. It doesn't have to be insurgents, either: in 2000, Michigan voted for insurgent John McCain in the primary, but its delegates were a bunch of long-standing Republican establishment officials who had backed George W. Bush. They were bound to McCain but loyal to Bush. John Yob calls these disloyal delegates "Supporters In Name Only," or SINOs.
Normally, SINOs don't matter. The vast, vast majority of bound delegates are also loyal to their candidate, and the frontrunner candidate has such a vast, vast lead that a few -- or even a few hundred! -- SINOs couldn't make even a tiny difference if they tried.
But, at a close or contested convention, it is ultimately delegate loyalties, not delegate bindings, that will determine the outcome.
Problem: it is not easy to determine the loyalties of the delegates, because of the way delegates are elected.
How Are the Delegates Elected?
In some states, bound delegates are selected directly by the presidential candidates they're bound to. This is not actually a very good system, in my opinion, but it does make analysis very simple: those delegates are essentially guaranteed to be loyal to their bound candidates. (I am told that the Democratic Party does this in every state.) The total number of delegates elected this way is 455 -- most from California and New York. (I show the details at the very bottom of this page, in the footnotes.)
In some other states, bound delegates are elected directly on primary ballots with endorsement by the candidates. This leads to some pretty crazy ballots, plus virtually-guaranteed delegate loyalty. This is a better system than direct candidate selection, if you insist on having a primary. 252 delegates are elected this way.
In most states, however, Republican national delegates are still chosen through the old system of local caucuses and state conventions. 1503 bound delegates are chosen this way, plus 121 unbound delegates. Under this system, as you'll recall from the previous post in this series, party stalwarts gather locally, work out the local platform, endorse local candidates, and elect their best people to represent them at a state (or congressional district) convention, and then those higher-level conventions elect their best people to be national delegates. The binding primary system invented in 1972 hasn't actually swept away this system or the national delegates it selects; it's just taken away the delegates' freedom to vote for the candidate of their choice, as they are nearly all bound to one candidate or another. 2016 Delegate Election Method Count Candidate Selection 455 Endorsed Election 252 Convention - Bound 1503 Convention - Unbound 121 Other 141 Total 2472 State Conventions: Where SINOs are Born
So the majority of national delegates are picked at state-level conventions. Who, specifically gets elected at those conventions?
Since national delegates haven't had any real effect on the nominating process for decades (since Reagan/Ford in 1976), state parties, over two generations, have formed the habit of electing people national delegates as a way of saying "thanks" to the most loyal Republicans. Party officers, who work hard behind the scenes to keep things running with very little public thanks, often get the nod. So do the party's elected officials (the governor, an ex-senator, and so forth). Occasionally, a smaller convention will give a spot to one of their their beloved volunteers -- you know, like that 65-year-old woman who's brought campaign lit and plates of cookies to every door-to-door in her congressional district since the battle against Mondale, and who says in her speech this has always been her dream but it's probably her last chance because she's getting up in years. Everyone in that room has enjoyed her famous peanut-butter cookies for decades, so of course they vote for her.
The delegates' first-choice candidate doesn't usually matter all that much: the winner of the primary is always someone broadly acceptable to party loyalists, even if not their first choice, so when Gretchen the Santorum supporter gets bound to vote for Romney, she doesn't mind all that much... and, even if she does, what can she do about it? It's not as if there's an organized opposition to the presumptive nominee -- at least, not one with any chance of winning. Not in a normal year.
This isn't to say that the delegate election process is without tension: there are always more people who want to be delegates than there are slots, there's always a lot of behind-the-scenes politicking to get on "the slate" (all the moreso because the candidates are all behind-the-scenes politicos themselves), and, above all, there's the great ideological tension between the two wings of the Republican Party: the Establishment and the Conservatives. Ever since the Goldwater/Rockefeller battle of 1964, Establishment and Conservative Republicans have vied for total control of the party, and their tug-of-war has gone back and forth down the decades: the Establishment won with Ford in 1976 (who beat Reagan), but the Conservatives won with Reagan in 1980 (who beat Bush). The Establishment won in both 2008 and 2012 (McCain and Romney, respectively, and Romney's supporters in particular put in place some rules designed to crack down on the grassroots), but Conservatives were optimistic about 2016 and began coordinating their efforts early this year, in hopes of a takeover.
Mr. Donald J. Trump blows all that away. In a convention system that pits Conservatives and Establishmentarians (of varying intensity) against one another, Trump is a figure despised by both wings of the Republican loyalists, because he is awful. Establishmentarians loathe him because he is unelectable and because his economic and immigration policies do not sufficiently benefit rich people. Conservatives abominate him because he's not a conservative and does not even seem likely to cooperate with conservatives the way the Establishment often does. Trump is popular with a broad swath of the Republican electorate... but, by all indications, his support is lowest -- and that is very low indeed -- among precisely the Republican loyalists who elect the actual delegates.
This sets us up for a 2016 convention with an unprecedented number of supporters-in-name-only. Establishment and Conservative Republicans across the country will continue to have their battles for control of the party, of course, and it's hard to say (especially now) what will happen in those battles... but, whoever wins, Trump loses. Conservatives are far more inclined to support Cruz than Trump; Establishment types prefer Kasich to Trump. For Trump to win a lot of loyal delegates, either the state parties would have to roll over and hand over their coveted bound delegate slots to Trump supporters who have not "earned" their place through years of volunteerism -- and state parties never surrender their delegate slots unless forced to by a Ron Paul-style floor fight‡ -- or Trump would need to discover a whole lot more support among party activists than either the evidence or the candidate's own attitude toward activists would suggest.
South Carolina is a perfect example. South Carolina binds delegates based on a winner-take-all-by-district allocation model. Because of a deeply divided anti-Trump vote, Trump won all 50 of South Carolina's delegates on February 20 with just 33% of the popular vote. However, according to Sasha Issenberg (I have not been able to verify this in state party rules), the only people who are eligible to become or vote for South Carolina national delegates are delegates to the 2015 state convention -- a convention that was held weeks before Trump even entered the presidential race. This virtually guarantees that the South Carolina delegation to Cleveland, while bound to Trump on the first ballot, will be either conservatives of the Nikki Haley-style or establishmentarians of the Lindsey Graham-style -- both of whom are fiercely opposed to Donald Trump. That's 50 Trump supporters-in-name-only... in just one state.
Multiply this across the country and, even without pulling out a spreadsheet and doing lots of counting, it's very easy to imagine Donald Trump coming into the convention with 1,200 bound delegates (just shy of a majority) but only 800 or so loyal delegates (far short of a majority). The other 400 Trump delegates would be SINOs, bound to Trump but loyal to others.
(Literally as I put this article through final proofreading, Politico published this piece on the "shadow primary" for delegates, with a special focus on South Dakota, where the fight for SINOs is already in the open.)
But What Can a SINO Do?
"Well, who cares?" you might ask. As long as a Trump SINO casts the required vote for Trump, as required by oath, it makes no difference to Trump whether his delegate is loyal or not: all Trump needs is the right number of votes, loyal or not, and he wins the nomination. Heck, even if a Trump SINO breaks his or her oath and votes for another candidate while still bound to Trump, there is a rule in the Republican Party that the faithless vote must be ignored and counted for the original bound candidate instead. (That's RNC Rule 16(a)(2) for you rules nerds out there.) So there is literally nothing a Trump SINO can do to change his or her bound vote: it will be counted for Trump whether the delegate cooperates or not.
Thing is, in most states, binding doesn't last forever. If, after the first formal vote for a presidential nominee at the national convention (known as a "ballot"), there is no immediate winner, most states release all their delegates to vote their conscience. Nearly all of the rest release their delegates by the third ballot. (Eight states, representing 301 delegates, bind their delegates beyond the third ballot, though with some caveats.) So if Trump doesn't "clinch" his bound majority and win on the first ballot (and, again, we'll discuss later how even a bound majority might not win him the first ballot), the second ballot will see 1,503 delegates suddenly voting their loyalty, instead of their binding. If all those delegates are loyal to their bound candidate, that has no impact. But if the convention has a few hundred Supporters-in-Name-Only, their change from binding-based to loyalty-based voting could swing the convention from a Trump-led convention to a Cruz-led convention, or even a complete free-for-all, without skipping a beat.
There are simply no precedents for a situation like this: the last time a Republican convention went to a second ballot was in 1948, more than twenty years before bound delegates (and, thus, SINO delegates) were even invented. But it seems that the odds of second-ballot SINO-driven chaos breaking out this year are now almost even money.
Beyond the obvious effects on a second or third ballot, about which the entire commentariat is breathlessly speculating, SINOs could have another huge impact on a convention: while bound delegates are required to vote for their candidates in nominating ballots, that is all they're required to do. In nearly§ all states, delegates are free to vote their conscience on any other measure, including rules changes that could dramatically change the terrain of the convention floor fight. This is a major point we'll be coming back to, because the few who are talking about it aren't doing a great job explaining it to a lay audience.
So, Seriously: Who Will the Delegates be Loyal To?
The truth is, we have no idea, and no way of determining, how many SINOs there are -- and which candidates they're truly loyal to -- until the possible second ballot at this year's national convention actually happens.
So far, this article has discussed the idea of Trump SINOs (bound to Trump but loyal to others), both because Trump SINOs would be the most consequential at a close convention where Trump enters with a lead in the bound delegate count, and because it seems likely Trump will have a lot more SINOs than the other candidates. But there will no doubt be at least some Cruz SINOs (bound to Cruz but loyal to others) and Rubio SINOs and probably even some Kasich SINOs. Indeed, there is evidence that Cruz SINOs who are actually loyal to Kasich are already organizing. The Cruz campaign openly advertises that it is working hard to win the loyalties of delegates bound to both Trump and Rubio. It's even possible that Trump himself has a more organized delegate-selection operation than he lets on (I have seen and heard evidence for this on the ground in Minnesota, but only anecdotally), and Trump will end up controlling the loyalties of a good bunch of delegates bound to other candidates. Nearly two-thirds of national delegates are elected by a caucus/convention system, and every active campaign is going to be fighting for the loyalties of every single one, bound or not.
Unfortunately, there isn't going to be a big ledger at the end of the primary season where all the delegates faithfully mark down their loyalties, the way delegates must mark down their bindings. We're going to hear a lot of rumors and a lot of guesses... but we aren't going to know until the convention which campaigns were most successful getting SINOs loyal to them elected to other candidates' bound delegate slots. We are fumbling in the dark.
If I may offer my guesses, though: Cruz is the candidate of the conservative branch of the party, and has been for quite some time now. He is also, by far, the best organized, most data-driven candidate in terms of get-out-the-vote efforts and delegate accumulation. In any normal establishment-vs.-conservatives year, the relative disarray of the establishment field compared to previous years plus the conservatives' clear organizational edge would, in my opinion, give the edge to conservatives generally and to Cruz specifically. Conservatives have way less money, which means they tend to lose primaries, but they make up a lot of ground in conventions, where their numbers and high levels of activism give them power. (Indeed, that was precisely the fight Cruz was gearing up for pre-Trump.) With Trump disrupting the race and the establishment now slowly rallying to Cruz as a last-ditch effort against Trump, and Cruz now fighting on the friendly terrain of state conventions, I would be pretty surprised if Cruz didn't clean up in conventions across the country, controlling not only the loyalties of most of his bound delegates, but a large chunk of Trump's and plenty of Rubio's and Kasich's as well.
Cruz will not hold on to the loyalties of all his bound delegates, but I suspect he will be losing them more to Kasich than to Trump. Trump may win plenty of delegate loyalties in the Southern states, where he did well and many party appartchiks warmed to him, but it's difficult to see how he wins his share of delegate loyalties in the Northeast, where the blue-collar base (which is expected to propel him to sweeping wins) has little representation in the upper echelons of the state parties. Kasich is much more suitable to the bigwigs of the Acela Corridor. And it is nearly impossible to see Trump getting any loyal delegates out of, say, my home state of Minnesota, even though 8 of our delegation will be bound to him, because this part of the country is simply very hostile to Trump's brand of politics -- and very open to the Cruz brand of conservatism. (I mean, hey, Santorum won our state's popular vote last time.)
It is, of course, possible that Trump will surprise me and sweep up a whole lot of state delegations with an insurgent guerrilla delegate-accumulation campaign, along the lines of what Ron Paul's insurgency did in several states in 2012. Many in the Minnesota state party infrastructure were caught totally off-guard by the number of Paulites elected by precincts to the BPOU conventions (a Minnesota thing; don't ask), and Paul supporters were able to seize majority control of enough BPOU conventions to ensure that the state convention was also majority Ron Paul. Despite fitting neither the establishment nor conservative mold, and despite finishing third in the Minnesota caucus straw poll, Paul left the state convention with 33 loyal delegates out of 40 awarded statewide -- including all of the delegates elected at the state convention.
However, Paul's campaign actively and publicly pursued a delegate-accumulation campaign from the get-go, spending heavily in caucus states where he could maximize his advantage, training his supporters (me included) throughout the country to win specific convention contests based on their specific local rules, and depending on the passionate, sustained engagement of his mostly-youthful base. I see some evidence of that in the Trump campaign -- organized speeches, precinct captains, specific instructions to supporters to get elected as delegates -- but it's only enough evidence at this point to make me nervous, not enough to persuade me that Trump actually has a functional delegate insurgency planned. (And, worth remembering: despite running the most effective insurgency in history, Paul still only "won" a handful of state delegations. Five, as I recall.)
But, like I said, I'm just guessing. The campaigns, with millions of dollars to spend and the phone numbers and home addresses of every single state-level delegate in the country, may be able to construct a reasonably complete picture of who's being elected to each delegate slot in Cleveland and where their loyalties lie before the national convention begins. We won't.
Now that you know all about the delegates who will carry out the convention (and also know what we don't know about them), we can get down into the nitty-gritty of convention procedure... and how the chaos of a closely-fought convention might play out. Stay tuned for Conventional Chaos, Part III: The Republican Rulebook!
FOOTNOTES
*Democrats often compare Republican unbound delegates to Democratic "superdelegates," but the analogy is a bad one. While they have similar powers, "superdelegates" are a deliberate addition to the process by the national Democratic party, designed to block undesirable candidates. Republicans unbound delegates exist for no special purpose: they are simply a historical artifact from a handful of states that never updated to primaries from their old caucus systems. There are way fewer of them, they are geographically concentrated in just a few places, and their numbers are unpredictable.
**By my count, there are already 60 delegates who have gone from bound to unbound. (55 from Rubio, 2 from Carson, 3 from Bush, from NH, AR, GA, MN, DC, and WY). If Kasich drops out (which would make stopping Trump much, much easier), we can expect another 20 or so delegates to join the unbound pool (9 from the District of Columbia, 2 from Maine, plus several more in upcoming contests).
†I mean, "To Whom Will the Delegates be Bound?" but I'm pretty sure there's a web-readability rule against using "whom" in headers.
‡Gosh darn it, I wish Jeff Johnson were Minnesota's governor today. I was at this speech. Heck of a speech.
§There are a handful of states -- namely Arizona and, arguably, Alaska and Georgia -- where delegates selected by convention must nevertheless pledge to make their "best effort," or equivalent, to support the candidate to which they are bound in all matters, not just the nomination vote. In theory, this prevents delegates from these states acting against their candidate's interests during rules fights. True, these pledges are nearly impossible to enforce as a practical matter, but many delegates are honorable people who will not (and should not) betray their oaths. Most states, however, merely bind their delegates to a single vote, not to a broad "best effort" standard.
Presidential candidates either directly select or get effective sign-off on their delegates in CA (172), CT (28), HA (19), ID (32), NH (23), OH (66), WI (42), and in the congressional districts of NY (81). They also select half the at-large delegates in TN (16). Total 479.
However, in each of these states (except, obviously, New York's congressional districts), the National Committeeman, National Committeewoman, and State Party Chairman take up three delegate slots, bypassing the candidate selection process with three consummate insiders. This amounts to 24 delegates, and reduces the total to 455.
Delegates are directly elected by voters, with endorsements by the candidates, in AL (50), RI (19), WV (34), Puerto Rico (23), and the congressional districts of MD (24), LA (18), IL (54), and TN (27). TN also elects half its at-large delegates the same way (15). Total 264. Again, in each of the 4 states that directly elect delegates at-large, 3 slots are occupied by party officials, who bypass the direct election. That peels away 12 of the 264 to the "Other" category. Total 252.
In states where delegates are selected by convention, party officials are chosen at the same time by the same people as the national delegates themselves, often based partly on their support for the appropriate candidates, so party officials aren't really "bypassing" the delegate election process the way they are in the other states. Delegates are chosen, unbound, in conventions in American Samoa (9), Guam (9), the Virgin Islands (surprise: 9), CO (37), ND (28), and WY (29). They are chosen, bound, in all other states and territories (1503), with two exceptions:
New Jersey (51, minus 3 RNC officials) directly elects its delegates, but they are not endorsed by the different candidates. Rather, the state party is entitled to endorse a single slate, which then typically wins. In most states, I'd say this party control over the process favors Cruz or perhaps Kasich. In New Jersey, however, with Gov. Christie's endorsement of Donald Trump and the general shambles of a party they have ("sad!" as Trump would say), I expect it will work to the benefit of Mr. Trump, and he will get all these delegates.
Pennsylvania's congressional districts (54) directly elect their delegates, but they are not endorsed by anyone. All delegates officially run uncommitted. This means that candidates will need to be highly organized to communicate the "loyal" delegate slates to their various supporters. The state is naturally Trump in the rural areas, naturally Kasich in the cities, and yet Cruz's deep organization gives him a decided advantage in this race. I've no idea what will happen there, and would love some insight from native Pennsylvanians.
In Florida, I should note, candidates get to submit recommended delegate slates to the state party, but the state party has final approval. It is unclear how willing the state party would be to overturn candidate instructions, though -- often, in these matters, custom matters as much as rules.
UPDATE 24 March 2016: Accounted for RNC "bypass" in delegate selection process.