Thursday, December 17, 2015

2016 Republican Delegate Allocation: VERMONT

This is part thirteen of a series of posts that will examine the Republican delegate allocation rules by state. The main goal of this exercise is to assess the rules for 2016 -- especially relative to 2012 -- in order to gauge the potential impact the changes to the rules along the winner-take-all/proportionality spectrum may have on the race for the Republican nomination. For this cycle the RNC recalibrated its rules, cutting the proportionality window in half (March 1-14), but tightening its definition of proportionality as well. While those alterations will trigger subtle changes in reaction at the state level, other rules changes -- particularly the new binding requirement placed on state parties -- will be more noticeable. 

VERMONT

Election type: primary
Date: March 1 
Number of delegates: 16 [10 at-large, 3 congressional district, 3 automatic]
Allocation method: proportional (with majority (50%) winner-take-all trigger)
Threshold to qualify for delegates: 20%
2012: proportional primary

-- 
Changes since 2012
There are only so many ways to allocate 16 delegates in a state with just one congressional district. Four years ago, Vermont Republicans met the less onerous RNC proportionality requirement by allocating its (11) at-large delegates in a proportionate manner to candidates with more than 20% of the vote statewide. The congressional district delegates (and automatic delegates -- 6 delegates total) were allocated in a winner-take-all fashion to the statewide winner as was allowed (considered proportional overall) in 2012.

The winner-take-all part of the 2012 Vermont delegate selection plan is not consistent with the changes the RNC has made to the national delegate selection rules for 2016; the stricter definition of proportionality. And that necessitated some form of change on the state level in a number of states. Rather than continue to make a distinction between the types of delegates and how they are allocated, for 2016, Green Mountain state Republicans have opted to pool all 16 delegates and allocate them proportionally based on the statewide vote.

And really, that was their only play if they were to comply with RNC rules given the March 1 presidential primary date. There are not multiple congressional districts in Vermont, and thus the single district vote is the statewide vote. As such, that synchronicity between state and district eliminates the ability to separately and proportionally allocate two different types of delegates (at-large and congressional district). For Vermont, then, there is no real meaningful distinction for the purposes of allocation. They are all just delegates.


Thresholds
The Vermont rules on 2016 delegate allocation set the bar for qualifying for delegates at 20% of the statewide vote. Candidates above that threshold are eligible for a share of the 16 delegates. Those below are left out. Importantly, though, there are triggers that would make Vermont a winner-take-all state. A candidate who receives more than 50% of the vote would be entitled to all 16 Vermont delegates. The usual field size caveats apply. The larger the field of candidates is on March 1, the less likely it is that one candidate will approach the 50% vote share mark. As the field is winnowed, however, that likelihood increases.

Additionally, there are scenarios where a crowded field might also force a winner-take-all allocation in Vermont. That happens in the situation in which only one candidate surpasses the 20% threshold. If there is only one candidate north of 20% then that candidate would -- like the candidate over 50% -- claim all of the Republican delegates from the Green Mountain state. There is nothing in the Vermont rules prohibiting such a backdoor winner-take-all allocation, and furthermore there are no detailed description of the "one candidate over 20%" contingency.

What the rules do lay out are plans for a situation in which no candidate receives more than 20% of the Vermont primary vote. In the event that no candidate is over 20%, the threshold drops to 15%. If no candidate is over 15% the threshold is lowered to 10%. Different states have dealt with this "no candidate over the minimum qualification threshold" differently. Some states, like Texas, define a specific number of (top) candidates to receive delegates if no one reaches the threshold. Others, like Tennessee (with its statewide, at-large delegates) or Minnesota, eliminate the minimum threshold altogether in the event that no candidate is above the threshold.

Vermont, though, is like Georgia (with its statewide, at-large delegates). Both have created a graduated threshold that decreases incrementally, but still maintains a minimum qualifying threshold. That moving target has implications. The cut points in the graduated threshold are somewhat arbitrary, and that has a decided impact on how delegates are ultimately allocated. If the winner finished at 15.1% and the runner-up was at 14.9%, the Vermont threshold would decrease to 15% and the winner would take all of the delegates despite winning only narrowly. And that is a different result than if the threshold dropped to 10% initially. In that case, the runner-up would claim some delegates.

FHQ understands the obvious counter to this: "Well, we're only talking about 16 delegates here." Absolutely, but a 16-0 delegate advantage is a lot better than a 9-7 edge and that is even more true when discussing states with a larger delegate cache. The point to all of this is that the moving target threshold is not without potentially significant ramifications.


Delegate allocation
The Vermont delegates -- all 16 delegates in total -- will be proportionally allocated to candidates with a vote share above the 20% mark. There are no polls that have been conducted in Vermont and the state of the Republican presidential nomination race there. Our best option then for exploring the impact of the rules in the Green Mountain state is to look at how the 2012 allocation would have been different had the 2016 rules been in place. In 2012 there were three candidates with a vote share greater than 20%. Mitt Romney claimed nine delegates and Ron Paul and Rick Santorum evenly split the remaining eight.1  Applying the 2016 rules to 2012 results, Vermont's allocation would look something like this2:
  • Romney (40%) -- 7.600 delegates
  • Paul (26%) -- 4.874 delegates
  • Santorum (24%) -- 4.526 delegates
  • Gingrich (8%) -- 0 delegates
  • Huntsman (2%) -- 0 delegates
  • Perry (1%3) -- 0 delegates
In this case, all three candidates who qualified for delegates would round up, but that would lead to an overallocation of delegates. Rounding Romney (8), Paul (5) and Santorum (5) up would allocate 18 instead of 17 delegates. Under Vermont Republican Party rules, any overallocation is squared by subtracting the necessary number -- one delegate in this case -- away from the qualified candidate with the fewest votes.4 That would yield an allocation as follows:
  • Romney -- 8 delegates
  • Paul -- 5 delegates
  • Santorum 4 delegates
Compared to 2012, the allocation under 2016 rules would shift one delegate away from Romney's total and add it to Paul's. A minor change at best. That, however, is more a function of Vermont's small number of delegates than anything else. As the delegation size increases, so, too, does the impact.


Binding
All 16 Vermont Republican delegates are bound through the first ballot at the national convention according to the provisions of Rule 11.i of the state party bylaws. The release process entails a delegate being freed if their candidate does not have their name placed in nomination at the national convention or if their candidate has suspended their campaign or withdrawn from the race.


--
State allocation rules are archived here.


--
1 Vermont Republicans had 17 delegates in 2012 and has subsequently lost a bonus delegate for 2016, dropping their total to 16.

2 Again, this is a simulation of the impact of the Vermont rules for 2016. Note also, that this simulation will be done using the 17 delegates from 2012. That will allow a better comparison of the impact of the rules.

3 The total sums to more than 100% because the percentages were rounded to the nearest whole number. The allocation is based on the full, voted-based percentage.

4 The Vermont rules also account for an under-allocation. Should, after rounding, there be an allocation of fewer delegates than are available, the unallocated delegate(s) are awarded to the top vote-getter.


Wednesday, December 16, 2015

2016 Republican Delegate Allocation: TEXAS

Updated 3.1.16

This is part twelve of a series of posts that will examine the Republican delegate allocation rules by state. The main goal of this exercise is to assess the rules for 2016 -- especially relative to 2012 -- in order to gauge the potential impact the changes to the rules along the winner-take-all/proportionality spectrum may have on the race for the Republican nomination. For this cycle the RNC recalibrated its rules, cutting the proportionality window in half (March 1-14), but tightening its definition of proportionality as well. While those alterations will trigger subtle changes in reaction at the state level, other rules changes -- particularly the new binding requirement placed on state parties -- will be more noticeable. 

TEXAS

Election type: primary
Date: March 1 
Number of delegates: 155 [44 at-large, 108 congressional district, 3 automatic]
Allocation method: proportional (with majority (50%) winner-take-all trigger statewide and in congressional districts)
Threshold to qualify for delegates: 20% (both statewide and within the congressional districts)
2012: proportional primary

--
FHQ has discussed to some extent the Republican Party of Texas delegate allocation rules for 2016 already. However, there have been some tweaks made to the rules that have a fairly decided bearing on how Texas delegates will be allocated to candidates based on the results of the March 1 primary. Initially, the Texas GOP attempted to devise a primary-caucus plan roughly similar to the one Texas Democrats have utilized for years.1 That plan entailed raiding the pool of at-large delegates -- those allocated based on the statewide presidential primary results -- and granting the power of allocation to the state convention. It was a clever bid to get two bites at the apple in a wide open presidential primary season. Candidates would come in to the Lone Star state once for the primary and those still around and active in the nomination contest for the June state convention would return to woo the state convention and its delegates.

As 2015 progressed, however, that plan changed. By March 2015, the State Republican Executive Committee had added language to the rules that provided for a fall back option if the Republican National Committee rejected the primary-caucus plan. That proved a somewhat prophetic move as the RNC General Counsel's office informed the Republican Party of Texas via letter in early June 2015 that RNC Chairman Reince Priebus had interpreted the change -- the addition of the convention step -- as a violation of the Rules of the Republican Party.2

In the end, what this amounts to is a Texas delegate allocation plan that is one step less complicated without the allocation of 38 at-large delegates (one-quarter of the total number of delegates) at the state convention. But it also means that those 38 delegates will not be hanging out there as a potential June wildcard -- the intent behind the creation -- as primary season winds down.


Changes since 2012
The addition and then subtraction of the state convention portion of the allocation plan constitutes a change since 2012, but it will not ultimately be implemented for the 2016 cycle. Still, there are some changes to the current method as compared to how the state party handled its 2012 delegate allocation.

Redistricting challenges in the courts held the 2012 Texas primary date (and thus the eventual delegate allocation) hostage into March 2012. That back and forth had the Texas primary in March before considering an April date and then ending up on the last Tuesday in May. When the rules were made for the 2012 cycle, Texas Republicans were planning for a March primary and thus settled on a unique proportional allocation plan. Basically, Texas Republicans proportionally allocated all of their at-large and congressional district delegates based on the statewide results in the primary. It was all one big pool of delegates. However, only candidates who were above 20% of the vote within a congressional district were eligible for a share of the pool designated congressional district delegates. The plan was probably overly and unnecessarily complicated, but it essentially worked out to a small bonus for those candidate who did well.

For 2016 and a March primary within the proportionality window, Texas Republicans scrapped the 2012 plan and adopted a proportional plan that is more comparable to some of its SEC primary neighbors (see Alabama or Georgia, for example). As opposed to proportionally allocating a larger pool of at-large and congressional district delegates based on the statewide primary results (oversimplification, but see above), Texas Republicans will have a more unit-specific allocation. The at-large and automatic delegates will be proportionally allocated and bound based on the statewide primary results, but the congressional district delegates, unlike 2012, will be fully allocated according to the results within each of the 36 Texas congressional districts.


Thresholds
While the Texas Republican delegate allocation plan is proportional, it meets that mark, but with provisions that potentially limit the number of candidates who qualify for delegates or provide a statewide or district winner to be allocated a disproportionate share of the delegates from those units. Texas Republicans have kind of followed the letter of the law in laying out thresholds that go up to the RNC designated limits. That means candidates must hit 20% of the vote in a congressional district or statewide to be eligible for delegates, but if a candidate wins a majority in either unit, then that candidate is entitled to all the delegates either statewide or in a congressional district.

Contingent upon how many candidates clear the thresholds, there are a number of different directions in which the allocation can go. The limits of that statewide are clear. There can only be so many candidates beyond the 20% threshold. Four or fewer candidates will split those 47 at-large and automatic delegates. Unlike some of its SEC primary compatriots, Texas prohibits the backdoor winner-take-all scenario, where just one candidate breaks the 20% mark. Should only one candidate clear 20% statewide, that would trigger a top two allocation. The statewide winner and runner-up would be proportionally allocated a share of the at-large delegates (regardless of whether the runner-up is over 20%). Compared to others, then, Texas limits how much of a boost a winner can get out of the allocation.

If no one crosses the 20% threshold statewide, then the allocation is proportional and functions as if there is no threshold. That would open the allocation up to all candidates. However, as has been the case elsewhere, this becomes less and less likely as the size of the field decreases (even with a threshold as high as the one in Texas).

At the congressional district level there are still more contingencies. And, they, too, narrow the possibilities of who qualifies for those three delegates. In the situation where no candidate wins a majority in a congressional district, then the top two candidates -- provided at least one is over 20% in the district -- win the delegates. The winner is allocated two delegates and the runner-up claims the remaining delegate. If no candidate tops 20% in a congressional district, then the top three finishers each receive one delegate.

What is clear about the Texas plan, at least as compared some others -- is how limited the advantage is to the winner(s). The only way for a winner to put some real distance between him- or herself in the Texas delegate count is to win majorities or pluralities consistently across all congressional districts. That may or may not occur. This is exacerbated by the lack of backdoor winner-take-all provisions.

This set of rules tends to greatly reduce the number of candidates who qualify for delegates, but also is likely to produce a pretty tightly clustered delegate count.


Delegate allocation (at-large and automatic delegates)
The at-large and automatic delegates -- 47 delegates in total -- will be proportionally allocated to candidates with a vote share above the 20% mark. Based on the last poll conducted on the race in Texas (the late October/early November Texas Tribune poll), the statewide allocation would look something like this3:
  • Trump (27%4) -- 23.572 delegates
  • Cruz (27%) -- 23.428 delegates
  • Carson (13%) -- 0 delegates
  • Rubio (9%) -- 0 delegates
  • Bush (4%) -- 0 delegates
  • Fiorina (4%) -- 0 delegates
  • Paul (4%) -- 0 delegates
  • Huckabee (2%) -- 0 delegates
  • Christie (1%) -- 0 delegates
  • Kasich (1%) -- 0 delegates
  • Santorum (1%) -- 0 delegates
Here is where being the winner -- or more precisely the order of finish -- matters: rounding. The allocation occurs sequentially from the top down. Any fractional delegate rounds up, even if it is just a tiny fraction. How tiny? Well, Trump and Cruz basically tied in this poll. One more respondent favored Trump than Cruz. Trump could have a .01 fraction and still round up. This problem is mitigated to some extent by the fact that there is an odd number of delegates. Even in a one vote difference situation, Trump would always round up to one more delegate.

Rules matter. In this case, rounding rules matter.

No other candidate surpasses 20%, and they end up with nothing to show for it in the hypothetical at-large and automatic delegate count.


Delegate allocation (congressional district delegates)
If, as we have done with other states, we extend the above statewide "results" to the congressional district level, that small advantage Trump has -- one vote/respondent -- yields a sizable delegate advantage. When two candidates are over 20% at the congressional district level, the top finisher receives two delegates and the runner-up one. That is true even if that win is by just one vote. As FHQ has stated before, it is difficult to proportionally allocate three congressional district delegates.



Binding
The Texas Republican delegates to the national convention in Cleveland are bound to candidates based on the results of the primary on at least the first ballot at that convention unless they have been released by the candidate. For those who have not been released prior to or at the convention (but before the roll call vote), they are bound through two ballots. The exception is for delegates bound to candidates who fail to meet a 20% viability threshold on the first ballot. They are released for the second ballot vote. After a second inconclusive ballot all delegates are free from their binds.

Most states examined thus far have a one ballot hold on delegates bound to candidate still in the race. Texas, however, keeps delegates bound to those with 20% or more support through two inconclusive ballots.


--
State allocation rules are archived here.


--
1 Long allowed by the DNC to be grandfathered into the future, the Texas Democratic primary-caucus (two step) process was rejected by the national party for the 2016 cycle. All of the Texas Democratic Party delegates will be allocated based on the results of the presidential preference primary and that contest alone.

2 That letter is appended to the current rules of the Republican Party of Texas.

3 This poll is being used as an example of how delegates could be allocated and not as a forecast of the outcome in the Lone Star state presidential primary.

4 Though Trump and Cruz both garnered 27% support in this poll, 163 respondents favored the New York businessman to 162 respondents for the junior Texas senator.


Tuesday, December 15, 2015

2016 Republican Delegate Allocation: TENNESSEE

Updated 3.1.16

This is part eleven of a series of posts that will examine the Republican delegate allocation rules by state. The main goal of this exercise is to assess the rules for 2016 -- especially relative to 2012 -- in order to gauge the potential impact the changes to the rules along the winner-take-all/proportionality spectrum may have on the race for the Republican nomination. For this cycle the RNC recalibrated its rules, cutting the proportionality window in half (March 1-14), but tightening its definition of proportionality as well. While those alterations will trigger subtle changes in reaction at the state level, other rules changes -- particularly the new binding requirement placed on state parties -- will be more noticeable. 

TENNESSEE

Election type: primary
Date: March 1 
Number of delegates: 58 [28 at-large, 27 congressional district, 3 automatic]
Allocation method: proportional (with supermajority (67%) winner-take-all trigger statewide and in congressional districts)
Threshold to qualify for delegates: 20% (both statewide and within the congressional districts)
2012: proportional primary

--
The Tennessee Republican Party method of delegate allocation echoes that of Oklahoma in many respects. In several others, it does not.

Changes since 2012
The biggest similarity between Tennessee and Oklahoma in terms of their respective delegate selection plans is that Volunteer state Republicans were similarly overly proportional given the 2012 RNC rules. Again, four years ago, states could achieve proportionality by simply proportionally allocating their at-large delegates. State parties were free to adopt plans for 2012 that would accomplish that while still allocating congressional district delegates in a winner-take-all fashion. Both Tennessee and Oklahoma awarded both at-large and congressional district delegates in a proportionate manner in 2012. With nearly the same sets of state-level rules carrying over from 2012 to 2016, both states were already in line with the new, tighter definition of proportionality the RNC has for 2016.

Thus, there are no real changes to those rules in either Oklahoma or Tennessee.


Thresholds
Tennessee Republicans have the highest allowable threshold under RNC rules to qualify for delegates statewide and at the congressional district level. In the vast majority of scenarios, to be awarded delegates, a candidate must win at least 20% of the vote. The only exception is if no one finishes over 20%.

If no candidate clears the 20% hurdle statewide, then the delegates are allocated in proportion to the candidates share of the statewide vote. In other words, if no one hits 20%, the Tennessee primary will basically operate as if there is no threshold.

If no candidate receives 20% of the vote in a congressional district, then the top 3 finishers each receive one delegate.

But as FHQ has stated before, March 1 -- the date on which the Tennessee presidential primary will be held -- the race will have wended its way through the carve-out states and some likely winnowing of the field of candidates. As the field decreases in size, the likelihood of no candidate getting to 20% of the vote in Tennessee (or anywhere else for that matter) decreases as well.

The rules change, however, if more than one candidate exceeds 20% of the vote. If multiple candidates are over 20% statewide, then the delegates would be allocated to those candidates in proportion to their share of the over 20% vote (the total share of just those over 20%). Should that happen at the congressional district level, the top finisher would be allocated two delegates and the district runner-up would take the remaining one.

Finally, there are a couple of winner-take-all situations. But it should be noted that it is a unit-specific winner-take-all, not a truly winner-take-all allocation.1 If only one candidate crests over 20% either statewide or at the congressional district level, then that candidate would win all of the at-large and automatic delegates or congressional district delegates. As in Oklahoma and several other states, there is a backdoor to a modified winner-take-all allocation and with a much lower threshold.

There is also a supermajority threshold for winning all of the delegates as well. If there is more than one candidate over 20% -- again, either statewide or in a congressional district -- and the winner has more than two-thirds of the vote, then that candidate would also lay claim to all of the at-large and automatic delegates or congressional district delegates. Obviously, though, that is a much higher winner-take-all trigger (but lower than the similar threshold in Minnesota).

Needless to say, there are a number of contingencies packaged around these various thresholds. The supermajority trigger seems unlikely to be tripped if the field is large, but even as it -- the field of candidates -- shrinks, the other options, including the backdoor winner-take-all route all would be probable.


Delegate allocation (at-large and automatic delegates)
The statewide results in the March 1 Tennessee presidential preference primary will dictate how many of the 31 at-large and automatic delegates are allocated to which candidates. If multiple candidates are over the 20% threshold, those candidates will win a proportional share of those delegates. Based on the last poll conducted on the race in Tennessee (a November Vanderbilt poll), the statewide allocation would look something like this2:
  • Trump (29%) -- 16.648 delegates
  • Carson (25%) -- 14.352 delegates
  • Cruz (14%) -- 0 delegates
  • Rubio (12%) -- 0 delegates
  • Bush (6%) -- 0 delegates
  • Fiorina (2%) -- 0 delegates
First off, no one is over 67%, so there is no winner-take-all allocation. There is also more than one candidate over 20%, and that means that there is no backdoor winner-take-all allocation. Out of the 6 candidates, only two cleared the barrier and nearly evenly split the 31 at-large and automatic delegates. Trump would be allocated 17 delegates in this scenario and Carson would take 14.


Delegate allocation (congressional district delegates)
If we extend the hypothetical statewide numbers above to the congressional district level, it would trigger a top two allocation. If multiple candidates are over 20%, then the district winner -- hypothetically Trump here -- is allocated two delegates while the runner-up wins the other of the three congressional district delegates.

In both the statewide and congressional district allocation simulations, the results would have been no different if the threshold was lowered to, say, 15%. If it was lower still -- set at 10%, for instance -- then Cruz and Rubio would qualify for at-large delegates. Neither would win any congressional district delegates. The 1-1-1 allocation of the three congressional district delegates is only triggered if no one is above the 20% threshold.


Binding
Like Oklahoma, it is not entirely clear how long or how many ballots the bind lasts for Tennessee delegates. That has a lot to do with how the delegates are selected. The at-large delegates are selected in a couple of different ways. Half of them (14 delegates) are elected directly, listed with candidate affiliation on the primary ballot. The other 14 at-large delegates are selected by the Tennessee Republican Party Executive Committee and with input from the candidates' campaigns. In both cases, those delegates are loyal to their candidate. If that candidate has withdrawn, then those delegates presumably become unbound (or can opt out of attending the convention, in which case the Executive Committee fills the vacancy). The district delegates also appear on the ballot affiliated with (and bond to) the candidate to whom they have pledged. The same rationale applies to them as is the case with the elected at-large delegates.

Update: The three automatic delegates (see above) are bound on the first ballot at the convention, according to Brent Leatherwood, the Executive Director of the Tennessee Republican Party (citing RNC rules). Mr. Leatherwood later tweaked this, indicating a change in TNGOP rules meant the three automatic delegates as well as the rest of the delegation would be bound through two ballots.


--
State allocation rules are archived here.


--
1 By unit specific FHQ means the winner-take-all allocation is confined to either just the at-large delegates based on the statewide results or just the congressional district delegates based on the results in the several congressional districts. A candidate would have to claim victory by a wide margin in Tennessee to win all 58 delegates.

2 This poll is being used as an example of how delegates could be allocated and not as a forecast of the outcome in the Volunteer state presidential primary.



Monday, December 14, 2015

The Real Import of Rule 40 in 2016

One of the more controversial rules changes that came out of the 2012 Republican National Convention in Tampa was the change to Rule 40. This is basically the guide to getting a candidate's name placed in nomination; to be able to take part in the roll call vote. Over the years the thresholds have gradually increased. In 2012, a candidate had to control a plurality of a delegation from at least five states in order to be nominated. But that is not a bar that had been in place forever. And again, it was certainly a barrier that increased to that level over time.

In Tampa in 2012, however, the thought process was a bit different than it is today. There and then, the rules changes were designed with one thing in mind: protecting President Romney in a 2016 bid at re-nomination. As I told Matt Viser a couple of weeks ago, few were sitting around in the Rules Committee meeting the week before the convention or later in the convention room itself thinking about the possibility of 15 candidates vying for the 2016 nomination. The rules that Republican lawyer, Ben Ginsberg pushed through the Rules Committee and were ultimately adopted at the convention protected hypothetical future President Romney in several ways, but one of them was increasing the thresholds a candidate had to meet to have their name placed in nomination at the convention.

The changes?

Instead of the five states a candidate had to win in 2012, a candidate had to win at least eight in 2016. And adding to that, a candidate could no longer skirt by with just plurality control of the the delegation. The rules adopted in Tampa for 2016 upped that to a majority. The raising of the number of states is not really of much consequence. But bumping that control threshold up from just one more delegate than your next nearest competitor to having to control half of the delegates in a state's delegation plus one is a potentially significant change. The impact of the change is minimized when one considers a scenario in which a President Romney is facing off against just one likely fringe foe. Yet, when one removes a hypothetical sitting president and inserts more than 10 candidates with no frontrunner/a nominal frontrunner/an unproven outsider frontrunner, then that change to Rule 40 really seems to start to matter.

This is exactly the type of unintended consequences a party gets when it makes plans for a nomination and convention four years ahead of time. Things change and in ways that are difficult to plan for so far out.

--
Well, what do these changes portend for 2016? It is pretty ominous, right? And that is seemingly even more true given a wild and unpredictable field of candidates.

Not really.

As David Byler explains at Real Clear Politics, Rule 40 is basically a placeholder. It was among the series of convention rules that the Republican National Convention (via the adopted Rules Committee package) extended to the next cycle with some tweaks as it routinely does without much notice every four years. And by "convention rules", FHQ means the rules that are intended to govern the next convention (Rules 26-42). Those are distinct from what one might call "primary rules" -- Rules 13-25 -- in one important respect: The convention rules cannot be altered between conventions (see Rule 12), but the primary rules can.

That is why we have seen the cleaning up of Rule 16(a)(2) and the reinsertion of shall for may resurrecting the proportionality requirement in Rule 16(c)(2) among other changes.

But Rule 40 and the other convention rules have been off limits, static since Tampa. The routine is that these convention rules stay the same. One convention passes them on to another and that next convention adopts them as the procedures for governing itself with no problems. Ultimately though, the convention of delegates has the final say so in that instance. While it is customary for one convention to adopt the rules passed on to it by a previous convention it does not have to. Basically, a convention is charged with adopting the rules that will govern itself. The RNC has streamlined this process, as described above, but the convention can take it off that track if necessary.

That is where the discussion of Rule 40 stands now for the 2016 cycle. Again, with a wide open race and a robust field of candidates, the requirements to place a candidates name in nomination at the Cleveland convention appear rather steep. Can a candidate reach majority control of delegations from multiple states? Can multiple candidates get there? Will none get there?

We do not have answers to those questions at this point and may not until well after voting has begun in February. But all the while, the RNC -- and ultimately the convention itself -- could have some say in this matter.

...if it has to.

One of the shortcomings of Byler's otherwise excellent piece at RCP is that it does not consider the possibility that Rule 40 works as intended; that only one candidate gets majority control of delegations from at least eight states and receives a majority of the 2470+ delegates available in the Republican presidential nomination race.

And there is no defined point in time in which it will be clear that Rule 40 is working (or not). Until then, Rule 40 and the thresholds for nomination contained therein matter. It is a goal for which all the viable campaigns are shooting. There a reason Ted Cruz's campaign (among others) has been involved in Guam. It counts as one of those eight states. The campaigns, then, are operating as if the provisions of Rule 40 are a thing; that it matters. And indeed it does. Getting to majority control of eight states is still a goal for the campaigns until it is not.

So, if Rule 40 operates as envisioned, then there is no reason for the convention to change it. In the meantime, though, between now and the convention, the rule puts in place a marker that the campaigns must hit and are in fact striving for. That is different than saying Rule 40 "won't affect the primary outcome".

It won't affect the outcome if it ends up standing in the way of the Cleveland convention completing part of its task: nominating a presidential candidate. If Rule 40 does serve as an impediment to that task, then it will be altered. But until it is clear that Rule 40 is a problem -- that the primary process has produced an inconclusive result in part because of the requirements in the rule -- it matters.

--
That does leave us with a couple of important secondary questions.

First, does the RNC signaling that Rule 40 is a placeholder now -- in December 2015 -- change how the campaigns approach the process? We will not begin getting an answer to this question until votes start coming in and delegates are allocated. It is at that point that the campaigns will start attempting to frame what they have won and what that means. In other words, do the campaigns keep us updated on the progress they are making toward meeting the thresholds laid out in Rule 40 -- the process of being placed in nomination?

Second, if -- and it is still an IF, folks -- Rule 40 is determined to be problematic during the course of primary season (if an inconclusive result is seemingly almost assured), then how does the Convention Rules Committee alter it? Byler addresses at the end of his piece and FHQ disagrees with him. As I said above and as I told Ben Jacobs and Walter Shapiro via Twitter last Thursday when this whole brokered contested convention discussion broke open anew, the number of states required under Rule 40 is not the issue. The biggest, most problematic change was the raising of the control threshold from a simple plurality to a majority. The convention may attempt to game things some with the number of states, but reverting to the former plurality requirement is more likely to solve a whole lot of the potential problems with the Rule 40 requirements.

But first thing's first: Let's see if, in fact, Rule 40 is going to be a problem once votes are being cast.

Throwing Out Convention Votes?

The genre can be a bit of a burr under the saddle to FHQ at times, but Philip Bump at least had a nice and pretty thorough brokered contested convention explainer up at The Fix over the weekend.1 I'll resist the urge to dig in too deeply at this point. However, there was one glaring inaccuracy in the write up that should be corrected.

It involves the small blurb about the impact of Rule 16 at the convention. Here is Bump's interpretation:
Rule 16: If delegates vote for someone besides the candidate to whom they are bound, those votes are thrown out.
This is wrong. These votes -- those cast against the delegate bind -- are absolutely not thrown out. If one reads the rule in full it ends up sounding that way by the end of the paragraph. Yet, the key is at the very beginning.

Let's look at the text of Rule 16(a)(2):
The Secretary of the Convention shall faithfully announce and record each delegate’s vote in accordance with the delegate’s obligation under these rules, state law or state party rule. If any delegate bound by these rules, state party rule or state law to vote for a presidential candidate at the national convention demonstrates support under Rule 40 for any person other than the candidate to whom he or she is bound, such support shall not be recognized. Except as provided for by state law or state party rule, no presidential candidate shall have the power to remove a delegate.
[Emphasis added by FHQ]

Bump is picking up on that vote recognition part at the end. But not recognizing a delegate's vote is different than throwing out that vote. In fact, it is impossible to discard a vote that has already been "announce[d] and record[ed]" in "accordance with the delegate's obligation". The order of this is of the utmost importance here. It sets up a sequence: record the bind and ignore the violation of it.

That is at least part of the reason why the RNC amended the language of the rule as it existed coming out of Tampa at the spring RNC meeting in Los Angeles in April 2013 without really changing the intent of the rule itself. Here is the version of Rule 16(a)(2) that was adopted in Tampa and amended eight months later:
For any manner of binding or allocating delegates under these rules, if a delegate (i) casts a vote for a presidential candidate at the national convention inconsistent with the delegate’s obligation under state law or state party rule, (ii) nominates or demonstrates support under Rule No. 40 for a presidential candidate other than the one to whom the delegate is bound or allocated under state law or state party rule, or (iii) fails in some other way to carry out the delegate’s affirmative duty under state law or state party rule to cast a vote at the national convention for a particular presidential candidate, the delegate shall be deemed to have concurrently resigned as a delegate and the delegate’s improper vote or nomination shall be null and void. Thereafter the secretary of the convention shall record the delegate’s vote or nomination in accordance with the delegate’s obligation under state law or state party rule. This subsection does not apply to delegates who are bound to a candidate who has withdrawn his or her candidacy, suspended or terminated his or her campaign, or publicly released his or her delegates.
[Emphasis added by FHQ]

This is the kitchen sink version of the same rule discussed above; the current rule. But notice that even under this rule, the vote is not thrown out, the delegate is. Then the vote is counted as if the delegate had cast it properly and in accordance with the instruction of the binding. Note also that the actions of the rule are flipped. In the early version of the rule the secretary of the convention awaits some violation before correctly recording the vote. The altered, current version completely removes the ability and motivation of a delegate to vote in violation of the binding by reversing the order. But the changed rule to eliminate the resignation of a delegate.

But the bottom line here is that in neither case would the vote have been thrown out. That would make an already messy situation -- a contested convention -- even messier.


--
1 It is not that FHQ does not like the convention discussion. It usually entails a headlong dive into the rules, and I am rarely averse to that. However, we have all been through this before. We are entering into brokered contested convention chatter season, but it ends up being a quadrennial exercise in putting the cart before the horse. There are delegate allocation rules that are going to matter a lot more in the near term than trying to calculate the conditional odds that they -- the rules -- are going produce an inconclusive outcome in June.


Follow FHQ on TwitterGoogle+ and Facebook or subscribe by Email.

Sunday, December 13, 2015

2016 Republican Delegate Allocation: OKLAHOMA

Updated 3.1.16

This is part ten of a series of posts that will examine the Republican delegate allocation rules by state. The main goal of this exercise is to assess the rules for 2016 -- especially relative to 2012 -- in order to gauge the potential impact the changes to the rules along the winner-take-all/proportionality spectrum may have on the race for the Republican nomination. For this cycle the RNC recalibrated its rules, cutting the proportionality window in half (March 1-14), but tightening its definition of proportionality as well. While those alterations will trigger subtle changes in reaction at the state level, other rules changes -- particularly the new binding requirement placed on state parties -- will be more noticeable. 

OKLAHOMA

Election type: primary
Date: March 1 
Number of delegates: 43 [25 at-large, 15 congressional district, 3 automatic]
Allocation method: proportional (with majority (50%) winner-take-all trigger statewide and in congressional districts)
Threshold to qualify for delegates: 15% (both statewide and within the congressional districts)
2012: proportional primary

--
Changes from 2012
The Oklahoma Republican Party made bigger changes to their method of delegate allocation from 2008-2012 than they did from 2012 until now. The state legislature in the Sooner state shifted back the presidential primary in the state by a month from February 2008 to March 2012. Squarely within the month-long -- all of March -- proportionality window for 2012, the reaction of the Republican Party in Oklahoma was to switch from winner-take-most (winner-take-all by congressional district) plan to an across-the-board proportional plan to comply with the new RNC rules four years ago.

But rather than proportionally allocating all of the delegates -- at-large and congressional district -- based on the statewide results, Oklahoma Republicans opted to award delegates in a proportionate manner based on the results statewide and in the several congressional districts. The at-large delegates were allocated and bound based on the statewide results and the congressional district delegates were awarded based on the results in each of the congressional districts.

That change ended up being more proportional than was necessary under the RNC rules in 2012. The only switch that was required to comply was the at-large delegate allocation. In 2012, a state could be considered compliant with the proportionality requirement if it proportionally allocated those at-large delegates. A state could continue to allocated congressional district delegates in a winner-take-all fashion (based on the congressional district results). Ohio made that incremental change, but states like Oklahoma and Georgia went the extra step and proportionalized the allocation of all of their delegates.

A total proportionalization for 2012 meant that states like Oklahoma and Georgia were ahead of the curve under the stricter definition of proportional the RNC has rolled out for the 2016 cycle; the one eliminating the winner-take-all allocation of congressional district delegates.

The historical rundown is intended to show that Oklahoma had no impetus to make any changes to its delegate allocation rules for 2016. And it has not really made any alterations.

So what does the process look like?


Thresholds
To win any delegates under the Oklahoma Republican Party delegate allocation rules, a candidate must pull in an at least 15% share of the vote in the March 1 presidential primary. That is true at both statewide and at the congressional district level. Should any candidate win a majority of the votes in the primary statewide or in a congressional district that candidate would be allocated all of the statewide, at-large delegates (25) or all of the congressional district delegates (3 in each district).

The usual caveats apply here. The more candidates who are alive by SEC primary day on March 1, the less likely it is that that majority winner-take-all trigger will be tripped. However, as the number of candidates drops, the likelihood of a majority winner either statewide or in a congressional district increases.

It should be noted also that the Oklahoma Republican Party delegate allocation rules do not prohibit the possibility of a backdoor winner-take-all allocation. That possibility is unit-specific and limited though. A candidate can be the only one to clear the 15% threshold either statewide or in a congressional district and claim all of the either at-large or congressional district delegates, respectively. It is not possible for a candidate to win all 43 delegates from Oklahoma unless that candidate is above the 50% threshold statewide and in each of the five congressional districts or unless that candidate is the only one above 15% statewide and in each of the five congressional districts. Both outcomes are possible but not probable. Call that a limited backdoor winner-take-all allocation or a backdoor winner-take-most plan.


Delegate allocation (at-large and automatic delegates)
This is a simulation of the allocation; not a projection. The numbers are less important than how the rules operate in this exercise.

The at-large and automatic delegates -- 28 delegates in total -- will be proportionally allocated to candidates with a vote share above the 15% mark. Based on the last poll conducted on the race in Oklahoma (the mid-November Sooner poll), the statewide allocation would look something like this1:
  • Trump (27%) -- 7.560 delegates
  • Carson (18%) -- 5.040 delegates
  • Cruz (18%) -- 5.040 delegates
  • Rubio (16%) -- 4.48 delegates
  • Huckabee (4%) -- 0 delegates
  • Bush (2%) -- 0 delegates
  • Fiorina (2%) -- 0 delegates
  • Paul (2%) -- 0 delegates
  • Kasich (1%) -- 0 delegates

  • Uncommitted -- 6 delegates
The first observation is that, as is the case in other states with a minimum qualifying threshold, Oklahoma's rules would limit the number of candidates who are actually allocated delegates. If a top tier emerges (and/or is maintained), then there will be a class of have candidates and a class of have not candidates. Of course, it should be noted that the carve-out contests have winnowed some of those have not candidates from the race.

The other thing that stands out about the Oklahoma delegate allocation formula for at-large and automatic delegates is the calculation itself. Unlike most other states, the language in the Oklahoma Republican Party rules uses the total vote -- and not the qualifying vote -- as the denominator in the equation. This is similar to how New Hampshire allocated delegates. Such a plan tamps down on the number of delegates the qualifying candidates -- those over the 15% threshold -- receive, but it also leaves a cache of delegates in limbo. In New Hampshire, the rule has always been to allocated those leftovers to the statewide winner in the Granite state primary.

But such a contingency is not a part of the Oklahoma rules. In the simulated allocation above, six delegates would not be allocated.

...to anyone. They would remain uncommitted. This is similar to the state of affairs in the Louisiana rules. But here's the thing: there is a range of possibilities here. If three candidates -- say, Trump, Cruz and Rubio -- just barely clear the 15% threshold. They end up with around 45% of the total vote. That leaves 55% of the vote under the threshold. All three candidates would claim four delegates and the remaining 16 at-large/automatic delegates would be uncommitted.

If, however, those same three candidates all receive 30% of the vote -- no, that's not likely -- then they combine for 90% of the vote and leave only 10% unaccounted for. Collectively, Trump, Cruz and Rubio would receive 24 delegates and four would be uncommitted.

That is a big range, but the results are likely to be somewhere in between. However, that does mean that there will be a small group of uncommitted delegates coming out of the Oklahoma primary.


Delegate allocation (congressional district delegates)
If the above statewide numbers are extended to the congressional district level, there would be four candidates over the 15% threshold. However, there would only be three delegates to go around in a given district. Trump, Carson and Cruz would qualify for one delegate each and Rubio would be left out of the allocation.

That same sort of allocation -- one delegate each -- would also hold if three candidates cleared the 15% barrier. If only two candidates draw 15% or more of the vote in a congressional district, the district winner would win two delegates and the runner-up would be allocated the remaining delegate.   That is consistent with the baseline allocation of congressional district delegates in Alabama and Georgia.

And again, should only one candidate clear the 15% barrier (or if a candidate wins a majority of the vote) in a district, then that candidate would take all three delegates from that district.


Things left out/unclear
There are a number of matters left unclear in addition to the automatic delegate question above.
  1. What are the rounding rules? There is little guidance in the Oklahoma rules here other than to "round to the nearest whole number". It appears as if there is no process for dealing with over- and under-allocated delegates. Yet, given the structure of the rules, it would seem as if most of this is taken care of in the uncommitted delegate procedure described above.
  2. How long does the bind on delegates last? Again, this is something that is left unsaid in the Oklahoma Republican Party delegate allocation rules. 1st ballot? Two ballots? Infinitely. It is not clear. 
  3. What if no one reaches 15%? Oklahoma Republicans do describe a number of scenarios for the allocation of delegates based on the numbers of candidates clearing the qualifying threshold. However, that list does not include contingencies for the case where no candidate clears the 15% threshold statewide or in a congressional district. This seems unlikely and would require something like a seven-way logjam, but again, this possibility is not covered. Given Oklahoma's position on the calendar -- after the carve-out states -- and the pretty clear top tier of candidates in polling, the chances of a "no one above 15%" scenario seems limited.

--
State allocation rules are archived here.


--
1 This poll is being used as an example of how delegates could be allocated and not as a forecast of the outcome in the Sooner state presidential primary.



Follow FHQ on TwitterGoogle+ and Facebook or subscribe by Email.

Saturday, December 12, 2015

2016 Republican Delegate Allocation: MINNESOTA

Updated 2.29.16

This is part nine of a series of posts that will examine the Republican delegate allocation rules by state. The main goal of this exercise is to assess the rules for 2016 -- especially relative to 2012 -- in order to gauge the potential impact the changes to the rules along the winner-take-all/proportionality spectrum may have on the race for the Republican nomination. For this cycle the RNC recalibrated its rules, cutting the proportionality window in half (March 1-14), but tightening its definition of proportionality as well. While those alterations will trigger subtle changes in reaction at the state level, other rules changes -- particularly the new binding requirement placed on state parties -- will be more noticeable. 

MINNESOTA

Election type: caucuses
Date: March 1 
Number of delegates: 38 [11 at-large, 24 congressional district, 3 automatic]
Allocation method: proportional (with supermajority (85% statewide) winner-take-all trigger)
Threshold to qualify for delegates: 10% (both statewide and within the congressional districts)1
2012: non-binding caucuses

--
In 2012, the Minnesota Republican Party exploited a loophole in the RNC delegate selection rules and held a non-binding caucus a month before non-carve-out states were supposed do so. Since the preference vote at those February precinct caucuses had no direct, rules-based bearing on the delegate allocation process -- the results did not bind delegates to candidates -- the party was able to skirt sanctions from the national party. Not only was the initial step in the caucus/convention process earlier than the RNC would have preferred, but because the delegates were unbound, it meant that there was ultimately a divergence between the preference vote winner (Rick Santorum) and the candidate who controlled the Minnesota delegates at the national convention in Tampa (Ron Paul).

The RNC made some rules changes in Tampa and at subsequent meetings in the time between then and late summer 2014. Many of those rules changes were intended to target the states just like Minnesota in 2012.2 Not only will Minnesota Republicans caucus about a month later in 2016 than the party did in 2012, but the delegates will be allocated and bound to candidates based on the preference vote at the March 1 precinct caucuses.

That means no fantasy delegates from Minnesota in 2016. But it does mean digging into a new delegate allocation formula.


Thresholds
As the Minnesota caucuses are scheduled for March 1, the party's initial step in its delegate allocation process will fall within the RNC's proportionality window. MNGOP will separately allocate statewide, at-large delegates and the congressional district delegates in a proportionate manner based on the results either statewide or within each of the eight congressional districts. Candidates qualify for delegates in those respective units -- again, statewide or within the congressional districts -- if they receive 10% or more of the vote. That is a hard 10%. A candidate with 9.5% of the vote cannot round up to 10%, for example, and be awarded delegates.

The exception to this is if a candidate wins 85% or more of the vote across the precinct caucuses statewide. Under the rules of delegate allocation in Minnesota, a candidate meeting that threshold statewide would be awarded all 38 delegates. Even in a field less crowded than the 2016 group of Republican candidates that 85% barrier is a high bar to surpass. Unless, something wild happens, then no candidate is going to trigger that winner-take-all allocation. Again, the bar is higher in Minnesota than in those states in which a simple majority triggers the allocation of all at-large and/or congressional district delegates to one candidate.

On the other end of that threshold spectrum, however, is that 10% bar to qualify for delegates. In terms of what is allowed under RNC rules -- a qualifying threshold up to 20% -- the Minnesota bar is pretty low. That, in turn, reduces that likelihood that just one candidate clears that hurdle. Obviously, with a large field of candidates, that outcome is more likely, but as the field of candidates winnows over the course of the February contests in the four carve-out states, just one candidate surpassing 10% of the vote statewide or at the congressional district level decreases.

Yet, it should be noted that there is nothing in the Minnesota Republican Party delegate allocation rules prohibiting a winner-take-all allocation either statewide or at the congressional district level should only one candidate crest above 10%. That differs from, say, Arkansas in the allocation of statewide delegates and Alabama, for example, with respect to congressional district delegates. Still, those structural differences across those states would tend to balance out in terms of their effects (which is to say, it would only tend to have an impact at the margins).

Delegate allocation (at-large/automatic delegates)
Both at-large and automatic delegates -- 14 delegates in total -- will be proportionally allocated to candidates with a vote share above the 10% mark. Based on the last poll conducted on the race in Minnesota (PPP's July poll), the statewide allocation would look something like this3:
  • Walker (19%) -- 4.22 delegates
  • Trump (18%) -- 4.0 delegates
  • Bush (15%) -- 3.33 delegates
  • Carson (11%) -- 2.44 delegates
  • Cruz (7%) -- 0 delegates
  • Huckabee (6%) -- 0 delegates 
  • Rubio (5%) -- 0 delegates 
  • Paul (5%) -- 0 delegates 
  • Christie (4%) -- 0 delegates
  • Fiorina (3%) -- 0 delegates 
  • Kasich (3%) -- 0 delegates
  • Jindal (1%) -- 0 delegates 
This actually ends up nicely capturing how rounding would work under the Minnesota rules. First, the allocation is done in descending order according to how the candidates finish. Fractional delegates of .5 or greater would be rounded up. The combination of a sequential allocation and rounding is one of the potential hidden advantages in these rules across states.

In the above example, no one has a remainder greater than .5, so no one rounds up. That means Walker ends up with 4 delegates, Trump 4, Bush 3 and Carson 2. That is a total of 13 delegates, leaving one delegate out of the original 14 unallocated. In some states, that unallocated delegate is awarded to the top finisher. In still others, the candidate with the largest remainder is given that last delegate. Minnesota falls in latter category. Carson, with a remainder of .44, would gain that final delegate, pushing his total up to 3.

Carson gains in that instance, but if those in front of him had had larger remainders or fractional delegates above .5, Carson would have been at a disadvantage by virtue of being the last over the threshold (and thus last in the sequence to be allocated delegates). Those toward the end of the sequence have the potential of being squeezed out of delegates dependent upon how qualified candidates with higher vote shares statewide round.

Delegate allocation (congressional district delegates)
Allocating the congressional district delegates is easier or harder. If there are multiple candidates above the 10% threshold and they are clustered together, then the top three in the vote count within the district will be allocated one delegate each. If we extend the above statewide results to the congressional district level, Walker, Trump and Bush would be allocated one delegate and Carson, despite being over 10%, would be on the outside looking in. That is the easy way of thinking about this.

It gets more complicated when considering the possibility of one candidate receiving enough support in a congressional district so as to win two delegates. It should be noted that this is less likely too. To accomplish this, a candidate would have to win more than half of the vote among the candidates over 10% (as opposed to using the votes cast for all candidates as the denominator). That is something that would be conditioned by how many candidates break the 10% threshold. If only two candidates are above 10%, then it would only take the top finisher one more vote than the candidate in second place in the district to be allocated two delegates. More candidates above 10% pushes the margin necessary to be allocated two of the three delegates upward.

Binding
Delegates from Minnesota under these rules are bound to their candidates through the first ballot at the national convention. If a candidate with delegates from Minnesota withdraws from the race, then those delegates would attend the national convention unbound. Interestingly, should a withdrawn candidate return to the race, those delegates would return to that candidate. That return contingency is unique to Minnesota so far as FHQ can tell.



--
One fun side note is that congressional district delegates under these rules can be allocated based on the statewide results rather than congressional district results. This is a little like the idea behind the National Popular Vote plan, but more likely operates like a blanket proportional allocation of all delegates, regardless of distinction, based on the statewide vote. What's "fun" is that the rules do allow for some variation. There could be some districts that would tether the allocation of their delegates to the statewide result while others would allocate theirs based on the results within the congressional district.

But that -- a decision to allocate congressional district delegates based on the statewide preference vote results -- can only happen if the party Executive Committee on the congressional district level makes that decision before August 31 of the year prior to the presidential election. That option was meaningless for the 2016 cycle, though. The allocation rules, including this option, were not adopted until September 17, 2015, after that August 31 deadline.



--
State allocation rules are archived here.


--
1 If no candidate reaches the 10% threshold either statewide or at the congressional district level, then the threshold is lowered by rule to 0%. In practice, though, to qualify for delegates a candidate would have to be closer to the top votegetter's vote share than zero. That is particularly clear at the congressional district level where only three delegates are available in each district. Statewide, if the winner is below 10%, then there would presumably be a great number of candidates around 9%. If Minnesota was the first contest on the calendar that would be one thing, but being positioned a month after Iowa kicks off the process means that the winners -- statewide and within the districts -- will be above the 10% barrier.

2 To some extent Iowa was also similarly hit by the rules changes. As Iowa is a carve-out state, though, the Hawkeye state was only affected by the addition of the binding requirements the RNC put in place for the 2016 cycle.

3 This poll is being used as an example of how delegates could be allocated and not as a forecast of the outcome in the Land of 10,000 Lakes. Obviously, Scott Walker is unlikely to receive 19% of the vote at the Minnesota caucuses.


Follow FHQ on TwitterGoogle+ and Facebook or subscribe by Email.

Thursday, December 10, 2015

2016 Republican Delegate Allocation: GEORGIA

Updated 4.18.16

This is part eight of a series of posts that will examine the Republican delegate allocation rules by state. The main goal of this exercise is to assess the rules for 2016 -- especially relative to 2012 -- in order to gauge the potential impact the changes to the rules along the winner-take-all/proportionality spectrum may have on the race for the Republican nomination. For this cycle the RNC recalibrated its rules, cutting the proportionality window in half (March 1-14), but tightening its definition of proportionality as well. While those alterations will trigger subtle changes in reaction at the state level, other rules changes -- particularly the new binding requirement placed on state parties -- will be more noticeable. 

GEORGIA

Election type: primary
Date: March 1 
Number of delegates: 76 [31 at-large, 42 congressional district, 3 automatic]
Allocation method: proportional (with winner-take-most trigger statewide and congressional district)
Threshold to qualify for delegates: 20% (to win statewide, at-large delegates)1
2012: proportional primary

--
For the most part Georgia Republicans have retained the same method for allocating national convention delegates as the state party utilized in 2012. Unlike states such as Ohio, though, the Georgia interpretation of proportional in 2012 is still consistent with the changed definition the national party is using in 2016. But in 2012, the allocation of delegates in the Peach state was overly proportional.2 In 2016, Georgia is still compliant with the RNC proportionality requirement with much the same rules.

At-large delegates
Georgia Republicans will proportionally allocate the 31 at-large delegates apportioned to the state by the RNC to candidates who clear the 20% threshold in the statewide vote in the presidential primary election. If no candidate receives 20% of the vote, the threshold is lowered to 15%. Should no candidate reach 15%, then that threshold is decreased to just 10% of the statewide vote.

It is worth noting that despite the varying thresholds, if only one candidate clears whatever barrier is established, then that candidate would be entitled to all 31 of the statewide, at-large delegates. If, for instance, Ted Cruz is the top votegetter statewide at 10.1% and Marco Rubio is the runner-up at 9.9%, then Cruz would win all 31 at-large delegates, even with such a narrow advantage. The fact that the Georgia threshold for a candidate to qualify for delegates is a bit of a moving target adds some intrigue to the process, but it does make it more difficult to game out. It is a potentially low mark to trigger a possible backdoor winner-take-all scenario for the at-large delegates.

One new aspect in Georgia for 2016 is that if a candidate wins 50% of the statewide vote, then that candidate wins all of the at-large delegates. If the field remains even somewhat crowded though, this seems more unlikely than a backdoor winner-take-all scenario.

Unlike the RNC summary, the Georgia rules specify that winning a majority statewide entitles a candidate to all at-large delegates, not all delegates in the state.3

***
UPDATE
"Rounding" of at-large delegates
The Georgia rules clearly state that there is no allocation of fractional delegates. But there is no defined method of rounding delegates. In lieu of rounding, then, what happens is a series of repeated proportional allocations of remainders. Let's look at this using the 2016 results that are now available to us.

The first thing is the allocation equation. Georgia uses a candidate's share of the statewide vote as the numerator and the total statewide vote -- not the qualifying vote of just those above the threshold -- as the denominator. Given Tuesday's results, that looks something like this:

Round 1
[34 at-large delegates]
(Allocated delegates in parentheses)

Trump: 34 X .388 = 13.196 delegates (13)
Rubio: 34 X .244 = 8.312 delegates (8)
Cruz: 34 X .236 = 8.024 delegates (8)

That allocates 29 of the 34 at-large delegates, leaving five leftover. Those five are then allocated using the same equation from above.

Round 2
[5 leftover at-large delegates]
(Allocated delegates in parentheses)

Trump: 5 X .388 =  delegates (2)
Rubio: 5 X .244 =  delegates (1)
Cruz: 5 X .236 =  delegates (1)

That leaves just two delegates to be split among three qualifying candidates. All that is left are fractional delegates as a result. Since the remaining delegates cannot be proportionally allocated without ending in fractional delegates, the remaining two delegates go to the statewide winner.

Final allocation
Trump: 13 + 2 + 1 = 16 delegates
Rubio: 8 + 1 = 9 delegates
Cruz: 8 + 1 = 9 delegates

According to the Georgia Republican Party general counsel's office, the three automatic delegates are included in the at-large allocation and are the first three delegates to fill in the first three allocated slots the statewide winner has been awarded.

***


Congressional district delegates
The bulk of the Georgia delegation -- as is the case the larger a state's population gets -- will be allocated at the congressional district level. Each of the 14 congressional districts in the Peach state is apportioned three delegates by the RNC. The state party has opted again in 2016 to allocate those three delegates in a Top Two manner. The district winner is allocated two delegates while each district runner-up is awarded the remaining one delegate. There is no threshold to qualify. Candidates simply have to get into the top two in the district vote count.

This roughly simulates the proportional allocation of three delegates, but does present the potential to hurt a third place finisher who is tightly clustered with the top two.  That third place candidate would likely be deprived of a delegate in that scenario; a delegate that goes to the winner of the district count. On the other hand, the top two method does eliminate the possibility of a candidate winning all the delegates in a district by clearing a low threshold (as is the case with at-large delegates).

Again, there is no threshold to qualify for congressional district delegates, but there is a threshold to qualify for all three of a district's delegates. Should a candidate win a majority of the vote in a district, then that candidate would be allocated all three delegates. Newt Gingrich was able to exercise this option in a handful of districts in his 2012 win in the Georgia primary.

The method is the same at the congressional district level in Georgia as it was in 2012.

Automatic delegates
The three party delegates are functionally at-large delegates in the Georgia delegate allocation plan. The party chairperson, the national committeeman and the national committeewoman are all allocated and bound to the statewide winner of the Georgia primary. As was the case with the congressional district delegates, the automatic delegates are allocated and bound in the same manner in which they were during the 2012 cycle.

--
All Georgia delegates are bound through the first ballot of the national convention. Additionally, there is no explicit guidance in the bylaws concerning the release of delegates upon the suspension of a presidential campaign or candidate withdrawal from race. Those delegates would theoretically continue to be bound to the candidates throughout, assuming the contest is unresolved throughout. If it is not competitive, the fact that the Georgia delegation voted unanimously for Romney at the convention in Tampa in 2012 speaks to the ability of delegates to be released from those bindings.



--
State allocation rules are archived here.


--
1 If no candidate reaches the 20% threshold, the barrier is dropped to 15%. If no candidate reaches the 15% mark, the threshold is lowered to 10%.

2 Though the Ohio presidential primary is situated outside of the proportionality window on March 15 and is truly winner-take-all for 2016, Buckeye state Republicans used a modified winner-take-all by congressional district method in 2012. The statewide delegates were allocated proportionally, but the congressional district delegates were allocated in a winner-take-all fashion to the winner of each congressional district. That qualified as proportional in 2012, but does not in 2016. Georgia was ahead of the curve in 2012 and thus did not have to make many changes to its rules for 2016.

3 Now, the Rule 16F filing the Georgia Republican Party made with the RNC made indicate a totally winner-take-all scenario, but that is not something that is described in the state party rules.



Follow FHQ on TwitterGoogle+ and Facebook or subscribe by Email.