Sunday, September 27, 2015

The Chaos Theory of Republican Presidential Nominations Under an Iron-Fisted RNC

Steven Rosenfeld had an interesting delegate selection rules/process piece up at Salon/AlterNet last week. He reached out to FHQ to discuss those rules and their implications for the RNC and Donald Trump, but unfortunately we never were able to connect. While I wish that hadn't been the case, I do think that both Richard Berg-Andersson and Tony Roza from the Green Papers, nonetheless, pretty closely captured a number of items I would have raised.

Still, FHQ has some comments in reaction to Rosenfeld's piece.

First, the rigging frame -- that the RNC will engineer the state-level delegate selection rules in such a way as to prevent a Donald Trump nomination -- is not consistent with what we know about how the presidential nomination process operates. The Republican National Convention adopted the 2016 national delegate selection rules at the 2012 convention in Tampa. It tweaked them slightly in the time between the convention and August 2014; readopting a proportionality requirement affecting a smaller calendar window, revising the language empowering the national party to enforce the new binding requirement that came out of Tampa and adding a rule to limit the number of primary debates among others.

A majority of the convention passed the 2016 rules, and the changes made in the time since had to garner majority support at the RNC Rules Committee stage before passing a three-quarters supermajority threshold before the full 168 member RNC in order to be implemented. And while there were voices in dissent at every checkpoint -- be it at the convention or any of the seasonal meetings the RNC holds -- the so-called establishment position won each time. And, in the case of the changes made outside of the convention, those changes achieved near consensus-level support from the members of the Republican National Committee.1

Those are not rigged results. In actuality, they are consensus-built national party delegate selection rules. The parties on both sides of the aisle often fight this last war; drafting rules consistent with those already in place, but that troubleshoot the problems that existed in the previous cycle(s).2 Again, that is not rigging. That is crafting rules in a context where different people from state parties all over the country have different views on what the rules should be; what the goals of the party and the nomination process should be. In other words, these rules are the product of politics. That process has yielded a pretty clear signal to this point: despite the diversity of opinion among members, the 168 members of the RNC have been pretty unified on the matter of the 2016 delegate selection rules. That is not to suggest that there is not dissent, but that, to the extent it exists, it has been minimal.

But Rosenfeld is not really talking about the RNC influencing the national rules of the nomination process. Instead, he is picking up on and advancing UT-Austin law professor Sanford Levison's wild theory that the Republican National Committee will massage state-level delegate allocation rules -- whether winner-take-all, proportional or some method in between the two extremes -- to derail Donald Trump's candidacy. This notion is ridiculous for a number of inter-related reasons.

First, state parties make the decisions on how delegates to the national convention will be allocated to particular candidates based on the results of primaries and caucuses. This does not mean that those state parties are not open to influence from the national party, but that task is and would be quite difficult. In other words, it is much easier said than done. The national parties themselves are somewhat decentralized as already described, but the state party executive and state central committees that vote on and thus choose the rules that will govern the state-level delegate allocation processes are another layer of decentralization altogether. If there is a diversity of views within the national party, then there are even more opinions involved when more people -- in the form of state party executive committees and state central committees -- are invited into the process. Additionally, there is variation on this from state to state. Some states are more aligned with the positions of the RNC than others.

But keep in mind also that state parties are like the national parties in one important respect. Since, they too are decentralized to some degree and are balancing a varying multitude of views, state parties often are plagued by the same politics with which national parties can be afflicted. Given a political environment, parties -- whether state or national -- often take the path of least resistance. Like the national parties, state parties tend to carry over the same delegate selection rules from the previous cycle. That can be a function of either majority agreement that those are the best rules or that disagreement within the state party over the rules means that the status quo position (the same rules from the last cycle) prevails. The one exception is when the national party requires a rules change (but FHQ will come back to that momentarily).

The presidential nomination process is all a huge instance of party coordination in a decentralized environment. That alone speaks volumes about how mismatched the Levinson hypothesis is with how this all actually works. But there is more to this presidential nomination coordination problem than decentralization. That hovers over the entire process, but does not provide an adequate picture of how off Levinson's view is.

One other factor that ought to be included is how the Republican National Committee approaches or has historically approached the regulation of state-level delegate selection rules in the modern era of presidential nominations. They have tended to be very hands-off; allowing states to devise rules that are tailor-made for that state. In that way, it is consistent with national party's general adherence to the notion of limited (federal) government. State and local interests/parties are better able to make those decisions than the national party.

Over time, however, the RNC has become slightly more hands-on in its dealings with state party-devised delegate selection rules. Those cases have tended to be in response to state-level abuses of the national party guidelines -- the national party delegate selection rules. Again, this fits in with the idea that the national parties are often fighting the last battle, the one from four years prior. It is all cyclical: national parties set rules, states react (either following or breaking the rules), national parties react to those abuses. And on and on the cycle goes.

In both parties, there have been problems during recent cycles in dealing with the handful of states that have chosen to hold either primaries or caucuses outside of the national party-designated window. Those states have gone too early. But in those cases, the national parties have tended to deal with those problems between cycles rather than within them. The one exception is that the DNC affords its Rules and Bylaws Committee the option of increasing the (50%) penalty on states that go too early. That is intended to create some additional leverage that can be used against states before the actual voting begins. But the RNC has no similar ability. It can only make rules changes between cycles.

And even if the national party did have that ability, getting state governments to respond is another matter. This is another way in which the process is decentralized. In the majority of states, the delegate selection/allocation process is filtered through a primary election. And in most of those states it is the state government -- not the state party -- that is making the decision on when the primary will be held.3 The state party has limited say in that process. The national party has limited say in that process. Yes, the state party has the final say in whether they opt into a primary or foot the bill for a caucus themselves. But those state parties have incentive to choose the cheaper option, the primary.

The fact that Florida, Michigan and Arizona broke the RNC rules on timing during the 2012 cycles should tell us something about exactly how much power and influence the national party over the states.

The final problematic factor in the Levinson hypothesis is something he basically points out. The incentives of wading into the patchwork of state-level rules are not clear.
There are risks for an anti-Trump GOP establishment with either approach. Pushing more states into the proportional representation camp lengthens the race to accumulating 1,236 delegates. But that “would presumably assure Trump of getting some significant number of delegates, assuming he hasn’t flamed out,” Levinson said. However, pushing more winner-take-all states “obviously is a greater roll of the dice.”
If a national party is coordinating this process, even if only loosely, it is likely not inclined to "push" states to have either winner-take-all or proportional rules. The field of candidates will winnow over time, but how it winnows and importantly when it winnows are unknown factors at this point. That those are unknowns now makes it increasingly difficult for a national party -- or a state party setting their own rules for that matter -- to effectively and exactly make rules across a series of states that will definitively advantage (or disadvantage) one candidate or one type of candidate. The parties at all levels are wary of unintended consequences and this is one area -- state-level rules making -- that can produce them quite quickly.

Now, the RNC may attempt to influence state parties' delegate selection plans, but their rate of success there is likely to be mixed at best. It is not that the national parties are not powerful. They are. But they are also limited in how they can respond to states. As Tony Roza points out in Rosenfeld's article, the state parties submit plans to the RNC and the RNC green lights those plans or rejects them. [The same is true for Democrats, though the sequence of submission and approval/rejection is frontloaded.] While that potentially provides an opportunity for the RNC to massage state-level delegate allocation rules, the party does not tend to handle these matters that way. First, there is no evidence that the RNC does anything other than give a thumbs up to these plans. Faced with a Florida plan in 2011 that called for both a non-compliant primary date and non-compliant allocation formula (winner-take-all in the proportionality window), the RNC, nonetheless gave a thumbs up to the plan. And that was an instance where the party had national level rules (penalties) in its favor, backing it up. There was a violation. Additionally, when the RNC considered increasing the penalties on rogue states in the late summer of 2011, it opted not to.

Most of the state party rules are publicly available, and most of them follow closely the rules of four years ago. There are exceptions, but collectively the picture of the rules that are public sets a baseline now for us to look back on later in the year when the RNC has approved and thus finalized the state-level rules. We will know then whether the RNC has in fact massaged the rules. Chances are pretty good that the national party will leave well enough alone. If it pressed forward with such an intervention, word would get out and that would further stir up the hornet's nest of discontent that is already brewing, for the time being propping up (in the polls anyway) outsider candidacies like Trump's. The RNC does not want that kind of backlash, and besides does not operate with that kind of iron fist anyway. They might intervene on matters like who participates in primary debates (i.e.: no more kiddie table debates, an increasing threshold of support for participating in those debates or a decreasing number qualify over time American Idol-style, etc.), and while that is important, it speaks to how limited the RNC is in controlling the process. The national parties really only seek to facilitate or manage the presidential nomination process.

The second comment-worthy item in Rosenfeld's piece concerns the Haugland-driven, rules-based chaos theory of the 2016 Republican presidential nomination process. If one works hard enough, it can be tied in with the delegate allocation discussion above, but in practice, it makes for a Frankenstein's monster of an article when bringing the two parts together. The article really seems little more than an attempt to shed some light on a couple of fringe hypotheses, Levinson's and Haugland's.

By FHQ's count this is at least the third vehicle for Haugland's Rule 40-based chaos theory; following Dave Catanese's and Reid Wilson's lead. The premise is basically this:
  1. There are now 15 candidates vying for the Republican presidential nomination.
  2. RNC rules require a candidate or candidates to hold a majority of delegates in at least eight state delegations at the national convention in Cleveland.
  3. RNC rules require states with contests during the first two weeks of March to proportionally allocate their delegates to candidates based on the results of their primaries and caucuses.
  4. There are likely to be between 20 and 25 states that hold contests in those two weeks (March 1-14).
  5. Together, all of the above in combination make it less likely that any candidate gets to the level of majority control of eight delegations.
  6. Brokered convention!
Yeah, maybe. But in reality, this hypothesis is the victim of overanalyzing the rules and underanalyzing some of the regular patterns of the presidential nomination process in the post-reform era. Stated differently, it puts the cart before the horse.

2016 could be different. It could also be that it falls into the regular winnowing pattern that tends to mark these processes. That is one element that is common to each of these pieces: They emphasize the chaos because it makes for a better story than "the field will winnow and there will be a ho-hum pep rally of a convention that gives way to the general election phase of the campaign". There are reasons to believe that the 2016 Republican field of presidential candidates will winnow in ways similar to most of the rest of them since 1972.

First, as Catanese mentions, states whose contests fall in the proportionality window have the option of putting in place thresholds (of the vote) that candidates have to meet in order to win delegates at the statewide and congressional district levels. Proportionality window states can set that threshold as high as 20%. On the surface, that minimum threshold is intended to limit the number of candidates who emerge with delegates bound to them from a given state. That is even true in the scenarios where no candidates reach the threshold (the top two candidates win the delegates) or only one candidate receives 20% or whatever the threshold is set to (in most cases, either that candidate gets all of the district/statewide delegates or that candidate shares those delegates with the district/state runner-up). Most of the SEC primary states fit this category.

That is going to have the likely effect of creating at least two classes of candidates: those with delegates and those that have none or very few. That latter category will face increasing pressure in a sequential process to put up or shut; to win contests and/or delegates or drop out. Yes, that is true even with super PACs. Donations to and spending from those groups is going to becoming increasingly results-driven as the process continues. That goes for the remainder of the invisible primary phase (increase your poll numbers or get out) as well as when the process moves from Iowa to New Hampshire to South Carolina and onward (win delegates or get out). Neither super PACs nor voters are going to be drawn to candidates who have only demonstrated that they cannot move up the polls or win contests/delegates. This is a naturally occurring phenomenon typical of the sequential presidential nomination process.

But what if the winnowing only occurs to a certain point, leaving four or three or two candidates? Actually, it is likely that the field shrinks after the February contests and continues in March/April until it is down to two candidates.

But why?

It is tempting to say that that the above threshold is like the electoral process in a parliamentary system, the membership of which is dependent upon a system of proportional representation. Those systems have thresholds too. Duverger's law would tell us that that type of system would end up with multiple parties in the governing coalition even with a minimum threshold for gaining seats in the parliament. Why would the Republican presidential process with thresholds for delegates be any different? Why would it not produce a stable group of multiple (and not just two) candidates?

The answer lies in the fact that the presidential nomination processes for both major parties in the United States are actually 50(+) contests that are happening in a sequence, rather than just one vote in the parliamentary example. Each step in the sequence gives onlookers -- voters, donors, super PACs, the candidates and their campaigns -- an opportunity to revisit the tally, the score in the horse race. Those opportunities -- those chances to reflect on how the process is progressing -- means that there are constantly points where the collective process is separating the wheat from the chaff.

But some may ask, if you are a candidate who has a fair number of delegates, why would you drop out and give up any leverage at a convention that may not be a coronation for a presumptive nominee? Why would you free up delegates -- your own -- to have no choice but to gravitate toward that presumptive nominee? This is the heart of the Haugland argument: if there's chaos, don't give up your delegates/leverage. That is why his comments are often peppered with statements like his intention to educate delegates on their responsibilities.

But there are a couple of reasons why candidates might and, in fact, very often do give up that leverage and release their delegates. First, like Scott Walker, they reason that there is no utility to be gained by continuing the fight. There is nothing to be won and plenty to be lost. Now, not all of the candidates fit that category. In fact, one could easily foresee candidates like Donald Trump or, say, Ted Cruz completely ignoring that Walkerian rationale and fighting on if they have delegates.4 The rest of the field is likely more pragmatic; willing to play the Boehner role and drop out for the good of the party and its chances of winning the general election.5

And that brings us to the second reason Republican candidates may feel compelled to bow out in favor of another better positioned (in primary season) candidate: Hillary Clinton. Well, not Hillary Clinton, per se, but how the Democratic presidential nomination process is going by comparison. If the advantages Clinton has in polling, fundraising and in party endorsements collectively translate into votes throughout presidential primary season, then the Democratic process will reach its natural conclusion more quickly. That comparison is important. If Republican voters, super PACs, donors and/or the campaigns feel like that gives any advantage (real or imagined) to the Democratic candidate, then some herding toward a resolution -- a candidate -- is the likely result. There would at the very least be some pressure to hasten the ultimate resolution. That pressure is exerted formally and informally on the candidates; for them to drop out and release their delegates in order for the party to focus on the general election task at hand.

Chaos theory proponents will point toward 1976 and the contested nomination battle that cycle. Of course, that was a nomination race shaped by a much different set of conditions. It was the first Republican nomination race under the new-to-them, Democratic Party-triggered nomination system where winning primaries affected who attended the conventions and thus who won the nomination. In other words, the Republican Party was new to the process and had not adapted to it in the way that it has in the intervening 40 years. 1976 was also a year in which Republicans only ever had two options: status quo (Ford) or outsider (Reagan). A bigger field in 2016 means more instability, but it also means that there will be a winnowing process that, in combination with the delegate allocation rules, may give an advantage to one candidate that is difficult for any other candidate to overcome, if there is significant competition to become the alternative to that frontrunner. Any delay in that secondary jockeying is a win for a front-running candidate.

Out of chaos may come order, but the type of chaos that the adherents of this theory describe is not likely to produce any of the order they desire until after 2016. And when there are presidential elections only once every four years, that means putting it off until 2020. And actually, that may be the goal of this: to hope for chaos that will yield if not a candidate that is acceptable to the chaos theory proponents, then enough mayhem to force some changes to the rules of the game.

Hey, it worked for Democrats in 1968. Chaos gave way to a new system of nominating presidential candidates. FHQ is not sure how much order they got out of that, though. At best it had a delayed impact.

1 And to be clear, these members are not stooges installed by the national party to do their bidding. These are the state party chair and the Republican National Committeeman and National Committeewoman from each of the 50 states, Washington, DC, Puerto Rico and the territories. Each of the three members that form the state delegation to the Republican National Committee are elected at the state level by the members of the state party. That leaves a diversity of different voices to form the core of the national party.

2 This is less manufacturing than it is carrying over the rules from previous cycles and augmenting them in ways to prevent the problems of the past. To build consensus, the path of least resistance is to take what already exists and tweak it. That often leads to the type of contradictions that North Dakota Republican National Committeeman, Curly Haugland, mentions later in Rosenfeld's piece.

3 Party-in-state-government is not necessarily the same as the state party. The goals of the individuals involved in each are institutionally different and even absent those differences, the groups can be completely different in make up.

4 Both are in a similar space and are unlikely to continue going on together against, for example, one alternative "establishment" candidate.

5 Some of the candidates in the 2016 field have already had experience dropping out in the past. See Huckabee and Santorum.

No comments: