Saturday, February 27, 2016

2016 Republican Delegate Allocation: MASSACHUSETTS

This is part twenty-three 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. 

MASSACHUSETTS

Election type: primary
Date: March 1
Number of delegates: 42 [12 at-large, 27 congressional district, 3 automatic]
Allocation method: proportional
Threshold to qualify for delegates: 5% (statewide)
2012: proportional primary

--
Changes since 2012
The Massachusetts Republican Party did very little to alter for 2016 the delegate allocation plan the party used in 2012. As was the case four years ago, the party will pool its delegates and proportionally allocate them to candidates based on the results of the March 1 primary in the Bay state. However, unlike 2012, that pool of delegates will include not only the at-large and congressional district delegates, but the three automatic/party delegates as well. Additionally, the party has lowered its qualifying threshold from 15 percent to five percent. That likely yields delegates for all of the candidates still involved.


Thresholds
Dropping the qualifying threshold by ten percent is not a trivial change. A 15 percent threshold is much more likely to keep candidates with only very narrow paths (or no path at all) to the nomination out of the delegate count in Massachusetts. But at just five percent, the bar is considerably lower. No, that is not a boon to any candidates on the lower end of the order in the vote totals. Those candidates will only get a very small number of delegates.

The real importance of the change -- lowering the threshold -- lies in the fact that those are delegates that would have gone to candidates at the top of the order with a higher threshold. The lower hurdle means fewer delegates for the most viable candidates in the race.


Delegate allocation (at-large, congressional district and automatic delegates)
The allocation in Massachusetts is pretty simple; the simplest thing this side of a truly winner-take-all allocation. Candidates who clear the five percent threshold will qualify for a proportional share of the 42 delegates Massachusetts has to offer. That calculation will utilize the qualifying total -- the votes of just the candidates who clear the five percent barrier -- as the denominator and the candidates' shares of the vote as the numerator.

The Massachusetts GOP will then use simple rounding rules to determine the final count. Those candidates with fractional delegates more than .5 will round up to the nearest whole number. Any candidate below .5 will round down. In the event that the rounding leads to an overallocation of delegates -- more than the 42 Massachusetts has -- then those delegates will be taken from the candidate(s) at the bottom of the vote order. Should there be an under-allocation -- fewer than 42 delegates allocated after rounding -- any unallocated delegate(s) will be awarded to the top votegetters. 

Binding
The 42 members of the Massachusetts Republican delegation will be bound to the candidates to whom they have been allocated through the first ballot of the Republican National Convention in Cleveland. 


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

Friday, February 26, 2016

SEC Primary Delegate Number Crunch

Let's do a quick simulated delegate allocation.

Bloomberg News released a Purple Strategies survey of voters in the seven southern states that will comprise the SEC primary on March 1. These numbers may prove to be off the mark when voters head to the polls on Super Tuesday. Nonetheless, if we assume that those numbers are the vote percentages in each of the seven states -- Alabama, Arkansas, Georgia, Oklahoma, Tennessee, Texas and Virginia -- and their congressional districts, we can glean more than a little something from the hypothetical delegate allocation. The polling data are even more interesting in light of just how close second and third place are. Rubio and Cruz are tied at 20% each. That has some interesting implication for a simulated delegate distribution.

Before digging in, here are the assumptions of this exercise:
  1. The polling data are being treated as the vote shares of the candidates in the seven SEC primary states.
  2. That same polling data will also be treated as the vote shares of the candidates in all 75 congressional districts in the SEC primary states. [Yes, this is perhaps the assumption that asks for the largest leap of faith.]
  3. An exact tie between Cruz and Rubio makes this simulation a touch more difficult. That is particularly true when attempting to determine how to allocate the congressional district delegates. There are only three delegates in each of the 75 district. In the majority of these states the winner in a district gets two delegates and the runner-up receives the remaining one delegate. If Cruz and Rubio are tied, then the allocation of that runner-up delegate -- everywhere -- gets tough. For the purposes of this exercise, we will assume that Rubio received one more vote than Cruz on the congressional district level. Rubio was given the nod because he received more second choice support in the Bloomberg poll. 
  4. This obviously also has Cruz finishing third in his home state. That, too, may be something of a leap of faith. 

 Here's what happens:
SEC Primary Delegate Allocation Simulation
CandidateBloomberg PollAlabamaArkansasGeorgiaOklahomaTennesseeTexasVirginiaTotal
Trump37%28224619339520263
Rubio20%15112212174811136
Cruz20%778128121165
Carson8%------------44
Kasich8%------------33

Observations:
  • With 37% across all seven states and all 75 districts, Trump wins more than 55% of the delegates. This is another example of just how much the Republican delegate rules favor winners/frontrunners. 
  • Remember, we are assuming that Rubio received just one more vote than Cruz. They are both at roughly 20%, but Rubio has one more vote to push him into second place for the purposes of this exercise. That one vote make a huge difference in the delegate count. Everyone wants to win, but if you cannot win in these SEC primary states, then you definitely don't want to be in third (or lower). Why? In every state but Oklahoma and Virginia, third place means getting shut out of the congressional district haul in most cases in the other five states. This simulated allocation really drives that point home. Again, just one vote separated Rubio and Cruz, but Rubio ends up with more than twice as many delegates. This is a big deal for anyone who is consistently in third on the congressional district level. It means falling further behind in the delegate count. Third place is a bad place to be.
  • Statewide, only Trump, Rubio and Cruz clear the qualifying thresholds -- 20% in Alabama, Georgia, Tennessee and Texas; 15% in Arkansas and Oklahoma. While that would mean a lot at-large delegates for Trump and that Rubio and Cruz would be allocated a similar share of at-large delegates, Cruz is getting the vast majority of his delegates from the at-large pool. Rubio's advantage is totally within the congressional districts in this exercise.
  • Overall, Trump basically carries a nearly 2:1 advantage in vote share over to an almost 2:1 advantage over Rubio in the delegate count. The New York businessman already has a more than 60 delegate advantage in the real delegate count. Not even counting the other four states allocating delegates on Super Tuesday, this hypothetical allocation of SEC primary delegates would tack another roughly 125 delegates onto that lead. Together, that would be more than the 165 delegates available on March 15 in winner-take-all Florida and Ohio. There are a number of comeback scenarios out there that are predicated on those 165 delegates.
  • Imagine if Rubio and/or Cruz slip below 20%. That would mean not qualifying delegates in the four most delegate-rich states. That would definitely be true statewide with respect to the at-large delegates, but would also apply to some of the congressional district delegates in some of those states. 
This is not a prediction. It is an exercise; a simulation. The utility here is in illustrating how the delegate allocation rules will operate in these specific circumstances (under the above assumptions). The two things demonstrated quite clearly are 1) the consistent winner gets a decided advantage in the delegate count under the rules in these states [This is not and should not be a news flash to anyone.] and 2) no candidate wants to (consistently) be in third place. The latter is a death sentence in the race for delegates.


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

Monday, February 22, 2016

A South Carolina Delegate Post

South Carolina was different in 2016 than it has been in the recent past.

The first and most obvious difference is that Donald Trump won all 50 of the delegates in the Palmetto state Republican primary. Four years ago, for example, Newt Gingrich was victorious by a larger margin (statewide), but ran behind Mitt Romney in the first district, losing those delegates. Even George W. Bush won by a similar margin, but lost the first district to John McCain in 2000. The 50 delegate sweep Trump pulled off in the first in the South primary Saturday night, then, is not something that is often witnessed out of South Carolina.

However, that is a pretty minor point in the grand scheme of things. After all, even including a case like 2008, in which Mike Huckabee won two districts, his delegate grab was still much smaller than the share the winner, John McCain, left the state with. It was near winner-take-all as a result.

Yet, that is just it. Near winner-take-all or completely so, the delegate take from South Carolina was different in 2016 than it has been in the last two cycles. In both 2008 and 2012, the South Carolina Republican Party incurred a 50% penalty from the Republican National Committee for trying to stay first in the South (or ahead of Florida in those two years). A less messy calendar (formation) and an underlying rules change helped. Yes, a stronger penalty kept potential rogue states in line for 2016, but even if there had been repeat rogue activity from Florida or North Carolina, another RNC rules change even further insulated the four carve-out states. Iowa, New Hampshire, Nevada and South Carolina all had at least a month before the next earliest contest to schedule their contests if other states jumped into February. In the past, the carve-outs moving to anything ahead of February meant they ended up with a penalty like any other rogue state.

Those changes have had implications for 2016 already. New Hampshire yielded Trump delegate surplus 50% greater than the one Mitt Romney emerged with in 2012. Most of that is directly attributable to the fact that Granite state Republicans also had a penalty-reduced delegation four years ago. That is even clearer in light of the fact that Romney received a higher share of the vote in New Hampshire in 2012 than Trump did in 2016.

Further south, the story is similar. Gingrich won by more, but left South Carolina +21 in the delegate count at best. And that was a delegation cut in half. Trump, on the other hand, added all 50 (unpenalized) delegates from South Carolina to his total. That is a more than 100% increase in the delegate cushion the winner got from the Palmetto state primary, cycle over cycle.

If this process is about delegates -- and as March approaches, it is -- then that +50 is a big deal. A 21 delegate surplus is not nothing, but all that did for Gingrich was help the former speaker close the overall delegate gap he faced against Romney. The former Massachusetts governor's total was fueled by the endorsements of then-unbound party delegates before the Florida primary at the end of January 2012.

FHQ raises that non-compliant 2012 winner-take-all primary in Florida for a reason. It was winning those 50 delegates that got Romney off to a delegate lead that the other vying for the nomination could not catch as the primary calendar wended its way through more and more events.

South Carolina is that early calendar delegate boost for Trump in 2016 that Florida was in 2012 for Mitt Romney. There are not a lot of +50s out there, but two are on March 15 and both -- Florida and Ohio -- will say a lot about whether Trump ends up in same position Romney ultimately was in 2012.


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

Thursday, February 11, 2016

Bill Eliminating Presidential Primary After 2016 Clears Arizona House

The Arizona House on Wednesday, February 10, passed HB 2567 by a vote of 37-22. The legislation would appropriate state funds to fully fund the 2016 presidential preference election in the Grand Canyon state but also eliminate the election in future cycles.

The impetus behind the move appears to be budgetary, but there may also be some secondary implications. Arizona Secretary of State Michele Reagan made clear in a state House Appropriations Committee hearing that, first, the statewide election carries a cost of nearly $10 million and, second, is a tab that should be picked up by the state parties in Arizona. The secretary called the possibilities for state parties absent state funding "limitless". However, that ignores the fact that state parties in a state as large as Arizona do not typically caucus or spend upwards of $10 million for a party-run primary. Rather than limitless, then, the likely outcome is a shift to a caucus/convention system to select delegates in the presidential nomination process.

A similar bill was proposed in 2012, but went nowhere. That effort, like the current one, had the support of county elections administrators in the states, who have long supported ceding state control of the nominating contest to the parties.

--
While this bill is still early in the legislative process, it could lead to a potentially noteworthy change for the 2020 cycle. If the bill passes, is signed into law and repeals the presidential preference election, that effectively makes Arizona a caucus state.

It was not that long ago that Arizona Democrats pushed for caucuses as a means of joining the carve-out states at the beginning of the primary calendar. The Arizona pitch to the DNC Rules and Bylaws Committee in 2006 included plans to shift to caucuses. At the time the national party, in an attempt to diversify the primary electorate on the early calendar, wanted to add a southern state and a western state as well as a primary state and a caucus state (that at the time could be put in a slot in front of New Hampshire on the calendar). South Carolina gained the southern primary spot and Nevada was slotted into the western caucus spot.

Nevada has had its share of issues on both sides of the political aisle in the last two presidential nomination cycles. That, in turn, has led to some discussion about whether the Silver state is on the chopping block as an early state. If the national parties -- both of them, not just the DNC -- want a western caucus state up front and want to replace Nevada, an Arizona caucus might provide an alternative.

File that one away for 2017-2018, though.

--
Thanks to Richard Winger at Ballot Access News for passing news of the Arizona bill along.


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

Wednesday, February 10, 2016

The New Hampshire Delegate Count and Beyond -- Automatic Delegates

There was some confusion throughout New Hampshire primary day about just how many of the Republican delegates were on the line in the Granite state. If one looks at New Hampshire state law, it sets the basic terms of delegate allocation in the state. The shorthand of this is that the secretary of state allocates delegates to the national convention to candidates who receive more than 10 percent of the statewide vote in the New Hampshire primary.

That is just "delegates", not at-large delegates or district delegates. Just delegates.

However, the New Hampshire Republican Party has traditionally pooled its at-large and district delegates and allocated them proportionally based on the statewide result. According to party bylaws (see Article II, Section 1 and Article III), though, the three party or automatic delegates have traditionally been unbound. The state party chair, the national committeeman and national committeewoman are prohibited from supporting any candidates. The party rules keep them "neutral".

That would seem to indicate that only 20 of the 23 New Hampshire delegates were at stake in Tuesday's primary. That was how FHQ interpreted the rules and we were not alone. That has been the way that delegates have been allocated in New Hampshire in the Republican contest.1

But this interpretation ignores changes made to the national party delegate selection rules for 2016. New to the Rules of the Republican Party for this cycle is a requirement binding delegates to candidates based on the results of statewide contests like the New Hampshire primary and the Iowa caucuses a week ago.

That rule, Rule 16(a)(1), states:
Any statewide presidential preference vote that permits a choice among candidates for the Republican nomination for President of the United States in a primary, caucuses, or a state convention must be used to allocate and bind the state’s delegation to the national convention in either a proportional or winner-take-all manner, except for delegates and alternate delegates who appear on a ballot in a statewide election and are elected directly by primary voters. [Emphasis FHQ's]
Like the New Hampshire state law, the terms of the Republican Party rule are ambiguous. It just says that a state's delegates must be allocated proportionally or winner-take-all based on primary or caucus results. There is no distinction between types of delegates. It is just delegates or in this case a state's delegation.

That fairly inclusive-seeming definition leaves open to interpretation just how state parties like New Hampshire's or Virginia's deal with the potential ambiguity in the national party rules as compared to state party bylaws that explicitly keep the party/automatic delegates neutral or unbound. Typically, the RNC has left much of the minutiae of delegate selection up to the states to decide.

Yet, in December, FHQ was told that there were no unbound delegates at the outset of the Republican nomination process. There can be unbound delegates, but only if they are released by candidates who have withdrawn from the race. That meant that in the 40 percent of states where state party rules left the automatic delegates unbound there was something of a conflict. That is what prompted us to begin adding riders like the following to the FHQ explainers on delegate allocation at the state level. Here's an example from the Virginia post:
The automatic delegates -- the state party chair, national committeeman and national committeewoman -- from Virginia are explicitly unbound according to the September resolutions adopted by the state party. That has been the case in the past, but FHQ was informed in recent conversations with the Republican National Committee that Rule 16(a)(1) binds all delegates from a delegation. The only exception is for delegates elected directly (on the ballot). That does not include party/automatic delegates. How those delegates are allocated/bound when the state rules are not clear on their allocation is a bit of an unknown and something of a wildcard.
The Republican National Committee has a more rigid interpretation of Rule 16(a)(1) and its effect on the binding of the three party delegates when such a binding process is not specified on the state level. In a January 29 memo to RNC members, the RNC general counsel's office, citing both Rule 16(a)(1) and the November call to the 2016 convention, detailed the binding of party delegates in those states. In a state where the allocation of party delegates is not specified, those delegates are to be treated as at-large delegates and allocated in a manner consistent with the allocation of that subset of delegates.

In New Hampshire, then, the party delegates are lumped in with the full allotment of delegates and allocated proportionally. The same would be true in Virginia (rendering Morton Blackwell's endorsement of Ted Cruz somewhat moot). A state like Tennessee, where only the at-large delegates are proportionally allocated based on the statewide results (and the district level delegates based on the congressional district results), those party delegates would treated as another three at-large delegates.

This is bigger than New Hampshire, then. Other state are affected as well, and FHQ's state-level allocation primers will be updated to reflect the clarified interpretation of the rules.


--
1 Of course, it has been since 2000 that that traditional method had been used. Penalties imposed by the national party for going too early in the last two competitive cycles -- 2008 and 2012 -- meant that New Hampshire lost its automatic delegates. New Hampshire is rules compliant for 2016 and they have their automatic delegates.


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

Monday, February 8, 2016

Two Things to Watch in the New Hampshire Delegate Race on the Republican Side

UPDATED 2/9/16

Mainly the candidates for the Republican presidential nomination will be fighting for a win and/or jockeying for position in Tuesday's primary. But there are couple of factors to watch as the returns come in that will affect the resulting delegate count coming out of the Granite state.

While New Hampshire does not offer a windfall of delegates, depending on how the voting shakes out the winner could end up with a surplus of delegates -- under the New Hampshire law that guides the allocation process -- that far outstrips the one delegate that separated first from third place in Iowa's caucuses. That surplus hinges on two things:

First, how many candidates finish over the 10 percent threshold necessary to claim any delegates? The current polling in the Granite state on the Republican race has Donald Trump well out in front. And while those numbers may not translate to votes, that fact does underline something perhaps more important for the candidates immediately behind the real estate mogul: the fluidity of polling in New Hampshire. Accurate or not, the polling has a number of candidates hovering around that 10  percent  mark; something that has not gotten enough attention amid discussions of debate performances and impending voting.

Unlike in some other states, candidates in the New Hampshire primary receive a share of delegates in proportion to their vote. If a candidate wins 11 percent of the vote, that candidate is allocated 11 percent of the delegates. Other states proportionally allocate delegates in a manner that considers a candidate's share of the vote among just the qualifying candidates; those who meet or exceed the threshold. In that scenario, a candidate who wins 11 percent of the vote would be awarded slightly more than 11 percent of the delegates.

This is an important distinction because it leaves some number of unallocated delegates because of the threshold.

The second factor to keep an eye on as the votes are being counted tomorrow night is how much of a percentage of the total vote are the candidates not qualifying for delegates taking. A larger under the threshold percentage means a greater number of unallocated delegates. If, for instance, Donald Trump wins 31 percent of the vote and Marco Rubio places second with 15 percent, but Bush, Cruz and Kasich dip below 10 percent, that leaves 54% of the vote below the threshold. That leaves over half of the delegates unallocated.

Only, that unallocated portion is awarded to the winner of the primary. Trump would claim seven delegates for the 31 percent won to the three Rubio would win for pulling in 15 percent of the vote. However, an additional 12 delegates -- that 54 percent share -- would be allocated to Trump.

Trump, then, would leave the Granite state with 16 delegate surplus. Again, that would be 16 times greater than the cushion Cruz brought out of Iowa. This is an important point. No, not necessarily for Trump specifically, but it is these sorts rules-based differences that can be meaningful in a protracted race for a presidential nomination. If it becomes a state-to-state battle, then these delegate margins start to matter more and more.

To drive home the point about the New Hampshire rules -- and these two factors in particular -- let's go back and assume that Trump and Rubio keep their respective shares, Cruz and Kasich hold steady at around 13 percent (rather than falling below the threshold) and Jeb Bush manages to stay over the 10 percent threshold (to say, 11 percent). That is five candidates over the threshold set by New Hampshire state law with a collective 83 percent of the vote. That leaves only 17 percent unaccounted for under the threshold. Just five delegates.

After Trump takes his seven delegates and Rubio his three, Cruz and Kasich each also grab three and Bush two more. The five "unallocated" delegates are added to Trump's count for 12 delegates total and just a nine delegate surplus when compared to the 16 delegate cushion in the simulation with just two candidates above the threshold and the others below it gobbling up a significant percentage of the total vote. As the number of candidates above the threshold increases, the percentage below the threshold tends to decrease.

No, there are not a lot of delegates on the line in New Hampshire on February 9. However, the rules are in place to make New Hampshire potentially more valuable than Iowa proved to be and perhaps even some larger proportional states deeper into the calendar.


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

Saturday, February 6, 2016

2016 Delegate Allocation Over Time

With the first votes in the 2016 presidential nomination contests in, the races have moved out of the invisible primary and into a different phase. Concurrent with that change comes a switch in the basis for discussion. The same polls, fund-raising and endorsement data is still there, but added to that now is chatter about wins and losses and of the chase for the delegates necessary to clinch one of the parties' nominations.

FHQ is guiltier than most of digging down into that chase or the rules governing the allocation of those delegates. However, taking a step back to gain a broader view of the undulations in the process can be just as important.

Compared to the frontloaded calendars of four and eight years ago, the 2016 calendar is completely different. The February start to the 2016 process is a month later than both of the previous cycles and most of the action is packed into just two months, March and April. Eight years ago, both parties had allocated more than 80% their delegates after the contests on the first Tuesday of March. At that same point on the calendar in 2012, roughly one-third of the total number of delegates had been allocated.

In 2016?

For the current cycle, the first Tuesday in March will be the one-quarter point on the cumulative delegates allocated tally in both parties.

The national parties informally coordinated the later start and unlike 2012, actually got their way. But the allocation of delegates is also pushed into a smaller space on the 2016 calendar than was the case in 2012. The de facto national primary on February 5, 2008 when 23 Democratic and 21 Republican state voted was perhaps even more compressed than is the case in 2016.

But the development of the 2016 calendar is about the back end of the calendar as well. Four southern states -- Arkansas, Kentucky, North Carolina and Texas -- all moved from May dates in 2012 to dates in the first half of March 2016. It is the shifting of the delegates in those states along with the later start that has led to the compressed -- March and April -- calendar of events in 2016.

Whereas the middle 50 percent of the allocation -- between the 25 percent and 75 percent points on the calendar -- were bookended by early March and late May spots in 2012, in 2016 that middle 50 percent window in from early March to late April with a very quick ramping up of allocation between March 1 and March 15 on both parties' calendars. After March 15, the climb from the 50 percent point to the 75 percent point is much more gradual and even more subtle after that until a late flourish of allocation in early June to close the process out.

This has strategic implications. The remaining candidates -- those not winnowed by the February carve-out state contests -- have a quick burst of primaries and caucuses in the first two weeks of March. All of the active campaigns will want to develop any kind of a lead in the delegate count by then. Though the rules open up a bit on the Republican side with the closing of the proportionality window, there are limited winner-take-all states to take advantage of after that. That means that any delegate lead in either party will be difficult to close for those challenging to overtake the leader in the count at that 50 percent point.

That is because the states after the middle of March are mostly going to distribute their delegates to multiple candidates in various ways. Both catching up to a delegate leader and closing out a nomination become steeper climbs in such an environment.

Below is a look at the how cumulative growth in delegates allocated progresses in both parties in 2016.

Of note is that the Democrats are about a week behind the Republican allocation after March 1. That has much to do with there being a bevy of southern conservative states early in the process. Both parties have some form of bonus delegate calculation that that rewards party loyalty (as measured by past voting history). There are more Republican bonuses up front than on the Democratic side. Those states carry more weight in the Republican process than for the Democrats.

Note also that there are two figures for the Democrats. The first accounts for the cumulative allocation of all of the Democratic delegates. However, that includes superdelegates that will not be pledged to candidates based on the results of primaries and caucuses. When the 712 superdelegates are backed out -- see the final figure -- the picture looks largely the same among the 4051 pledged delegates. The pattern is virtually the same.

--
The Republicans
Click Image to Enlarge

You can also find this image included with the delegate allocation information for the Republican process.


--
The Democrats
Click Image to Enlarge



Click Image to Enlarge



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

Wednesday, February 3, 2016

The Iowa Delegate Count

FHQ has fielded some questions in various forms about the delegate count coming out of Monday's Iowa caucuses. There already appear to be a few details about the minutiae of it all that are not exactly clear. Let's add some clarity to this as best we can.

The Associated Press is reporting a delegate count in the Hawkeye state that looks something like this:
  • Cruz 8
  • Trump 7
  • Rubio 7
  • Carson 3
  • Paul 1
  • Bush 1
And then it stops. Only, the problem here is that adds up to just 27 delegates when the Iowa Republican Party was apportioned 30 total delegates by the RNC. And as FHQ has explained (via the Republican Party of Iowa), the state party is proportionally allocating all 30 of those delegates. That means that there is a discrepancy of some sort.

Many -- well, many in my Twitter feed and inbox anyway -- have speculated and/or questioned whether FHQ just has it wrong. That there are three outstanding, unallocated delegates would, on the surface, suggest that perhaps the three unallocated delegates are intentionally unallocated. As the argument goes, those three are the three party/automatic delegates. And perhaps those delegates, as is the case in a number of other states are very simply unbound by the results of the caucuses.

It is a somewhat persuasive argument, but this is a situation where appearance and reality are at odds. Here's why:

First, FHQ spoke with Charlie Szold, the Republican Party of Iowa (RPI) Communications Director, back in August about Iowa's placement on the our primary calendar. But I took the opportunity to ask him about the newly adopted delegate allocation plan as well. Recall that Iowa's was a non-binding caucus in 2012. Adding a binding element and a proportional allocation, then, constituted a pretty large change from four years ago. Mr. Szold shared with me the new rules, I glanced them over and seeing that the Iowa rules seemed to suggest that all 30 of the delegates would be allocated and bound, followed up for assurance. After double checking with the party's expert on the delegate process, Mr. Szold confirmed that all 30 would be proportionally allocated based on the caucus results, including the party delegates.

This was a process that FHQ repeated yesterday given some pushback. The answer from RPI was the same: all 30 of the delegates, including the automatic delegates, will be allocated and bound to candidates based on the results in Monday's caucuses.

Well, here we are, two days after the caucuses, and the same 27 delegates above are the only delegates that have been allocated (unofficially by the AP). Why?

As I speculated yesterday it could be because not all precincts were in in Iowa. There is such a cluster of candidates around the 1.9 percent mark that any additions or subtractions of votes to/from the total could affect whether Fiorina, Kasich, Huckabee or Christie receive a delegate. But 100 percent of precincts are now reporting. There is now an unofficial tally. Yet, there still is not a full allocation out of Iowa.

The reason for that is that the tally is unofficial. Until the results are certified by the RPI there will not be a full delegate count out of Iowa. Let us not forget that four short years ago, Mitt Romney was declared the (narrow) winner in Iowa late on caucus night. However, a little more than two weeks later, Romney's 8 vote win became a 34 vote loss to Santorum. There is, then, a cautionary approach to the vote certification and delegate count in Iowa this time around.

Again, the issue is that four candidates are vying for three remaining delegates. Fiorina, Kasich, Huckabee and Christie all are eligible to round up to a delegate, but there are only three delegates. Christie is at the bottom of the order and would appear to be the odd man out. But it is too close to call until the vote is certified. How close? The four candidates' fractional delegate shares range from .527 to .559. That is four candidates separated by just .032 delegates. In other words, the shifting of a few votes here and there is consequential to the final delegate count.

This is a wait and see sort of thing and nothing more. The count at the top of the order -- where it matters -- is not going to change.

--
One more thing to add to this:
If one calculates the allocation of 27 delegates, the distribution would look like this:
  • Cruz 7
  • Trump 7
  • Rubio 6
  • Carson 3
  • Paul 1
  • Bush 1
  • Fiorina 1
  • Kasich 1
Cruz would not have 8 delegates in an proportional allocation scheme that intends to award 27 delegates. The Texas senator would just miss out on rounding up to 8. Actually, if only 27 delegates were being allocated the outcome would be clearer. Fiorina and Kasich would round up, but Huckabee and Christie would not. The four would still be clustered but the math would be clearer starting with 27 delegates.

--
Another astute question FHQ has received concerns that Huckabee delegate. And now that Rand Paul has also withdrawn, another delegate can be added to that mix. What happens with those delegates?

They stay bound to Huckabee and Paul regardless. The only out for Iowa delegates is if only one name is placed in nomination at the convention. If only one candidate is placed in nomination, as has been the case throughout the post-reform era (save one contest, the Republicans' first under the new system in 1976), then all of the delegates are bound to that candidate.

Unlike some other states, Iowa does not permit the release of delegates nor for them to become unbound in any way.

--
There are also some questions out there concerning the delegate count on the Democratic side. Much of this seems to be the result of a lack of clarity concerning how many delegates Iowa Democrats actually have. The Iowa Democratic Party delegate selection plan suggests that there are 54 delegates, but the Democratic National Committee count (still being reviewed, so not final) has Iowa with just 52 delegates, 44 pledged and 8 unpledged (superdelegates).1 It is difficult to speculate on the tentative delegate count in Iowa anyway since the delegates will not be chosen until the later stages of the process. But that process is made even more difficult when there is some doubt about how many delegates the Iowa Democratic Party has been apportioned by the DNC.

--
More post-Iowa delegate count thoughts here.

UPDATE: Certified Republican Party of Iowa caucus results and official delegate allocation.

--
1 While the link to the IDP delegate selection plan was to a draft FHQ had stowed away, I doubled checked it against the plan on the party website at the time of the original posting. It seems to have been updated as of Thursday, February 4 and now reflects the 52 delegates the DNC has apportioned to the Iowa at this time.


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

Monday, January 25, 2016

2016 Republican Delegate Allocation: IDAHO

Updated 3.8.16

This is part twenty-two 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. 

IDAHO

Election type: primary
Date: March 8
Number of delegates: 32 [23 at-large, 6 congressional district, 3 automatic]
Allocation method: proportional
Threshold to qualify for delegates: 20% (statewide)
2012: proportional caucus

--
Changes since 2012
The date is much the same -- it amounts to the same first Tuesday after the first Monday in March date -- but the mode of delegate allocation is different in Idaho in 2016 than was the case in 2012. Republicans in the Gem state traded in their early caucus for an earlier presidential primary (moved from the traditional mid-May point where it was consolidated with the primaries for other offices).

Idaho Republicans also switched out some funky caucus allocation rules for a set of rules that is more in the mainstream of how other states are allocating delegates in 2016. Gone is the more or less instant runoff method that netted Mitt Romney a winner-take-all allocation. That has been replaced by a proportional allocation based on the statewide results of the March 8 presidential primary in Idaho. As has been witnessed in other states, though, what constitutes proportionality varies quite widely.


Thresholds
Much of that variation in the application of proportionality is attributable to a couple of factors. First, is the delegation pooled or separated for the purposes of allocation? The former typically means that a state will end up closer to a true proportional allocation of delegates than not. The latter opens the door to the bulk of the wild delegate allocation plans out there. Secondly, however, is there a threshold that the candidates must meet to qualify for delegates? With a threshold comes limitations in terms of how many candidates end up with delegates.

But that fact is tempered by the distinction raised in the first question. States that have separated allocation -- distinctions made between the allocation of at-large and congressional district delegates -- and a threshold have been the most complex. Many of the SEC primary states fit this description. Idaho fits somewhere in between states with that type of allocation and those with a proportional allocation of pooled delegates without a threshold.

It is that kind of allocation that ends up being perhaps the least complex. For example, then, Idaho Republicans will pool all of their delegates rather than invite complexity through a separate allocation of at-large delegates and the six delegates in just two congressional districts. The rationale is similar to the that of other small states. Power comes through allocating a small bloc (or blocs) of delegates rather than a much more decentralized proportional allocation.

Gem state Republicans will accomplish this through pooling their delegates but also by layering in a couple of thresholds. First, to qualify for any of the 32 delegates, a candidate must win at least 20% of the vote (before rounding) in the Idaho presidential primary. That will have the effect of limiting the number of candidates who receive delegates to likely four or fewer. If no candidate clears the 20% threshold, then the allocation would be carried out as if there was no threshold at all.

Not expressly prohibited by the rules is a backdoor to a winner-take-all allocation. While the "if no candidates clear the threshold" contingency is described, the "if only one candidate clears that qualifying threshold" is not. What that means is that there is a backdoor winner-take-all option on the table in Idaho. And since the delegates -- all 32 of them including the automatic delegates -- are pooled Idaho is the one state before March 15 where just one candidate surpassing the qualifying threshold could net that candidate all of the delegates from that state. The others to this point on the calendar that have allowed for a backdoor winner-take-all option, but the allocation was split into statewide/at-large and congressional district delegates. That makes Idaho a potentially powerful, pre-March 15 piece in the allocation puzzle. Things would have to fall right -- a crowded field with little winnowing or separation between the candidates, for example -- but a potential +32 coming out of Idaho could be more valuable to a campaign than a split of, say, the 155 delegates in Texas.

Finally, Idaho also has a true winner-take-all trigger as well. Should one candidate win a majority of the vote statewide, that candidate would also be entitled to all of the delegates from the state. This may or may not be likely in a less crowded field. The last two winners in Idaho received more than 50% of the vote.1


Delegate allocation (at-large, congressional district and automatic delegates)
The Idaho Republican delegates will be proportionally allocated to candidates based on the outcome of the March 8 primary in the Gem state. Based on the last poll conducted on the race in Idaho (the late August 2015 Dan Jones and Associates survey2), the allocation would look something like this3:
  • Trump (28%) -- 32 delegates 
  • Carson (15%) -- 0 delegates
  • Bush (7%) -- 0 delegates
  • Cruz (7%) -- 0 delegates
  • Rubio (6%) -- 0 delegates 
  • Paul (5%) -- 0 delegates 
  • Christie (4%) -- 0 delegates 
  • Fiorina (3%) -- 0 delegates
  • Kasich (2%) -- 0 delegates
This captures the perfect storm sort of scenario for a backdoor winner-take-all allocation of the Idaho delegation. Trump would barely get half of the total necessary to win a true winner-take-all victory, but because he is the only candidate over 20% (in this simulated allocation), he would win all of the delegates from the state; a backdoor to a winner-take-all allocation.

The arrival at such an outcome is a bit unconventional in Idaho. The allocation calculation calls for the candidate's share of the vote to be divided by the total statewide vote. That would mean that Trump, in the above example, would only receive a 28% share -- or 9 -- of the total number of delegates. That is not winner-take-all by any account. However, the rules also call for the unallocated delegates -- those not allocated to Trump above -- to "be apportioned proportionally among candidates who clear the 20 percent threshold." That would award the remaining 23 delegates to Trump, giving him the full allotment of delegates.

Were the backdoor winner-take-all allocation trigger not tripped multiple candidates crossed the 20% barrier -- or none did -- the Idaho Republican Party rules are not clear on how the rounding would be handled (though the allocation would be proportional without any threshold). The only reference to rounding is that a candidate must have 20% of the vote to qualify. 19.9%, for instance, would not round up to 20% and qualify any such candidate for delegates.

It is not clear, then, if the rounding is always up to the next whole number or simply to the nearest whole number. Additionally, there is no outline for a sequence to the rounding procedure or a contingency for unallocated or over-allocated delegates. That contingency coming into play is a function of how many candidates clear the threshold. Anything more than two candidates raises the likelihood that there will be an under- or over-allocated delegate.

If, however, the threshold is lowered to 15% to include Ben Carson in the allocation, then Trump would round up to 9 delegates, Carson would round up to 5 and the remaining 18 delegates would be uncommitted. That is due to the allocation equation laid out in the Idaho Republican Party rules. Only candidates over the threshold qualify for delegates, but a candidate claims delegates based on their share of the statewide vote divided by the total statewide vote for all candidates, not just the qualifying candidates. If the threshold is lowered to 7% to pull in Bush and Cruz, then Trump and Carson would be allocated the same 9 and 5 delegates respectively and Bush and Cruz would both round down to 2 delegates each. 14 delegates would be unallocated and would remain uncommitted. While the rounding in Idaho retains some operational question marks, the combination of simple rounding rules -- rounding up above .5 and down below it -- and not allocating delegates for those under the 20% threshold simplifies things to some degree.


Binding
Though the party rules use pledged rather than bound, the intent is the same with regard to how the Idaho delegates act at the national convention. Delegates are bound to the candidate who "proposed them on their list" or were pledged to by the Idaho Republican Party Nominating Committee on the first ballot of the roll call voting at the national convention. Like Hawaii, the candidates have some say in who their Idaho delegates will be. Candidates submit a list delegate candidates  -- a slate essentially -- equal to 80% of the total delegation (26 delegates). Should a candidate win all of the delegates and/or fail to submit a list (or the requisite fee), then the Nominating Committee would select and pledge delegates to that candidate. In the event of a winner-take-all scenario, that would mean the remaining 20% of delegate slots not covered by the 80% slate. In the case of a candidate not filing a slate, that would mean how ever many delegates allocated to that candidate. In a straight proportional allocation of the delegates, the 80% slate is likely to contain enough delegates to cover the candidate. Again, that is 26 of the 32 delegates.

It is a little quirky, but it does highlight that the candidates would mostly retain some power over who their delegates are. That has implications for the national convention should it go beyond just one vote. Most of those delegates would be more loyal to the candidate than delegates selected and bound to a particular candidate but who prefer another.

Additionally, a candidate's withdrawal from the race or suspension of their campaign makes their delegates uncommitted. This is clear if the withdrawal/suspension occurs after the selection but before the convention. In that case, the Idaho Republican Party rules clearly release the delegates. However, if the withdrawal occurs before the selection process described above, then the Nominating Committee may fill those slots with as many uncommitted delegates as the withdrawing candidate was entitled to.


--
State allocation rules are archived here.


--
1 Granted, McCain had already wrapped up the nomination by the time Idaho voted in May 2008 and Romney benefitted from the aforementioned, quirky caucus allocation rules in 2012.

2 Though the 17% of respondents who fell into the "Don't Know"/undecided category in the Dan Jones poll would also not have qualified for delegates, distributing that fraction to other candidates in a primary could have pushed others over the 20% qualifying threshold.

3 This polling data is being used as an example of how delegates could be allocated under these rules in Idaho and not as a forecast of the outcome in the Gem state primary.



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

Thursday, January 21, 2016

2016 Republican Delegate Allocation: HAWAII

Updated 3.9.16

This is part twenty-one 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. 

HAWAII

Election type: caucus
Date: March 8
Number of delegates: 19 [10 at-large, 6 congressional district, 3 automatic]
Allocation method: proportional (statewide and in congressional districts)
Threshold to qualify for delegates: No official threshold
2012: proportional caucus

--
Changes since 2012
The Hawaii Republican Party rules for delegate allocation have not changed in any meaningful way as compared to the method the party operated under in 2012. The date -- the second Tuesday in March -- is the same and though Hawaii Republicans lost one at-large delegate from 2012, it is a change of just one delegate. But the small number of delegates at stake in the March 8 Hawaii Republican caucuses will be proportionally allocated based on the caucus results at both the state and congressional district level. Rather than pooling the 16 at-large and congressional district delegates, Hawaii Republicans will allocate both types separately. At-large delegates will be proportionally allocated based on the statewide results while the congressional district delegates will be awarded based on the outcome of the caucuses in each of the islands' two congressional districts.


Thresholds
There is no threshold in the Hawaii allocation process at either the state or congressional district level to qualify for delegates. However, the rounding method used by Republicans in the Aloha state both advantages the winner/top finishers and limits the number of candidates who will likely end up being allocated any delegates (see below for an illustration of this).


Delegate allocation (at-large delegates)
The Hawaii delegates will be proportionally allocated to candidates based on the outcome of the March 8 caucuses in the Aloha state. There has been no polling in the Hawaii race, but rather than make up numbers, FHQ will base the simulated allocation below on the data from ISideWith.com.1 Based on that data, the allocation of the 13 at-large and automatic delegates would look something like this2:
  • Trump (44%) -- 4.4 delegates [5 delegates]
  • Rubio (15%) -- 1.5 delegates [2 delegates]
  • Cruz (14%) -- 1.4 delegates [2 delegates]
  • Carson (10%) -- 1.0 delegates [1 delegate] 
  • Paul (5%) -- 0 delegates 
  • Bush (5%) -- 0 delegates 
  • Christie (4%) -- 0 delegates 
  • Fiorina (3%) -- 0 delegates
  • Kasich (2%) -- 0 delegates
This is where the rounding method is important. Fractional delegates always round up to the nearest whole number, but that means there is a threshold under which no one receives delegates and the last one over that barrier receives any leftover, unallocated delegates. That is because the rounding has a sequence, starting with the top votegetter and working downward through the list in the order of finish.

All of the candidates, then, round up to the next whole number despite, for example, Carson having a remainder less than .5. That leaves just one delegate left for Rand Paul by the time the sequence gets to him. Now, the junior Kentucky senator would have been allocated one delegate anyway based on pulling in 5% of the statewide vote. However, that is not the proper way of thinking about the Hawaii allocation because of the sequential rounding. Once the sequence gets to Paul, there is just one delegate left. It would be his.

Such a sequential method when coupled with always rounding up means that while there is no official threshold in Hawaii, an unofficial one will force itself into the allocation of the at-large delegates. Some candidate will be the last to be allocated delegates in the sequence and those behind that candidate will be left out of the allocation process. In this example then, that threshold is at 5%, but again, not officially.

Four years ago, all 11 at-large delegates were allocated to the top three candidates and Newt Gingrich, despite winning more than 10% of the vote had nothing to show for it. The unofficial threshold to win any delegates, then, was higher in 2012 -- around the 19% that Ron Paul received.

The main point here is not so much this threshold idea, but rather to point out that though there is no official threshold, there are limitations to who and how many candidates receive any delegates. And it should be noted that part of that is a function too of there being so few delegates in the first place.


Delegate allocation (congressional district delegates)
The same type of principle used in the allocation of the at-large delegates impacts the congressional district delegate allocation as well. There is no threshold, but as FHQ has said numerous times, there are only so many ways to proportionally allocate the three delegates the RNC apportions to each of a state's congressional districts.

The key to the allocation of the Hawaii congressional district delegates is in the rounding. Just as above, fractional delegates will be rounded up to the nearest whole number. If a candidate receives more than one-third of the vote in a congressional district, then that candidate will round up to two delegates in that district. The remaining delegate would be left to the candidate in second place in that district. If no candidate clears that 33.3% threshold in a congressional district, however, then the top three candidates would all be allocated one delegate.

This is a firmer threshold than the one discussed in the at-large example above. It is akin to some of the winner-take-all thresholds that exist in other states, but this one is specific to the allocation of two versus one delegate to the congressional district winner. To receive all three delegates -- to round up to all of them in a congressional district -- a winner would have to clear the roughly 67% barrier. That seems unlikely even with a winnowed field.


Delegate allocation
(automatic delegates)
Due to the new guidance from the RNC on the allocation and binding of the three automatic delegates, the Hawaii Republican Party could no longer leave them unbound. Rather than included them in the at-large pool of delegates as other states have done, the HIGOP has opted to separately allocated those three delegates based on the statewide results.


Given the above data from the simulation, Trump would receive two delegates and Rubio one.


Binding
Hawaii Republican delegates are bound by party rule Section 216 to candidates on the first roll call ballot at the national convention. However, those same party rules allow the candidates/campaigns some discretion in the delegate selection process. Delegates, then, may not be bound after the first ballot, but would likely remain loyal to their candidate in any subsequent vote. They would be pledged but not bound unless or until their candidate pulls his or her name from consideration.


--
State allocation rules are archived here.


--
1 I have no idea about the methodology at ISideWith. Again, this is just an exercise in how the allocation might look, but probably will not. Using their numbers is perhaps marginally better than me creating results. And yes, they sum to 102.

2 This data is being used as an example of how delegates could be allocated under these rules in Hawaii and not as a forecast of the outcome in the Aloha state caucuses.




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