Sabato's Crystal Ball

A Glossary of National Convention Delegate Allocation

Josh Putnam, Guest Columnist February 11th, 2016


The objective of any presidential campaign seeking the nomination of one of the two major parties is pretty simple: In some way, shape, or form, all the candidates want to win. All other things held equal, those candidates would also like to keep winning, from Iowa to New Hampshire to South Carolina and on and on. That is, a candidate wants to either (1) build on momentum established in the pre-voting invisible primary period or (2) create and sustain that momentum during primary season itself. The nominations of Al Gore and George W. Bush in 2000 fit the “building” mold, while John Kerry and Barack Obama in 2004 and 2008, respectively, might fall into the “create” category.

It might be easier to focus on the primary and caucus wins and losses in the first case. After all, that is the pattern: win and keep winning, with very few bumps in the road along the way. Presidential nominations are not often wrapped up so easily, however. Sometimes a candidate can clinch a nomination by a series of early victories, quickly choking off the support of challengers also vying for the nomination. However, if a candidate cannot do this and the hopes have faded on a quick and easy, momentum-based sprint to the nomination, the goalposts move and the focus shifts from wins and losses to the delegate count. And that can move things from a quick and easy sprint to a slow and murky slog through the arcane rules that govern the Democratic and Republican presidential nominations.

With that transition comes a fairly steep learning curve for the campaigns, which have hopefully already planned for a drawn-out affair, as well as for casual observers. The 2016 process is no different. While the Democratic National Committee retained the bulk of the delegate selection rules it used in 2012, the Republican National Committee made some changes to their process. On top of the patchwork of rules carried over, then, there are additional rules to consider.

The 2016 races have only just entered the voting phase of the process, and while there are some distant signs of order in a sea of chaos, here are some things to consider if either or both of the presidential nomination races become delegate counting affairs:

1. Proportionality

The concept of proportional allocation may be the most misunderstood aspect of delegate allocation. Basically, the problem is that the mechanics of the process are oversimplified. Often it is interpreted as “if Obama wins 60% of the vote, then Obama wins 60% of the delegates.” But that is only true in the very small number of states in the Republican process that have no requirement that candidates meet a certain threshold of support in the voting to qualify for delegates. For instance, in the recently-completed New Hampshire primary, only the five candidates who won at least 10% of the statewide vote were eligible to win GOP delegates.

On the Republican side, just five states are truly proportional with no threshold: Hawaii, Iowa, Nevada, North Carolina, and Virginia. The remaining states in the GOP process restrict just how proportional the allocation is. On the Democratic side, all of the states and territories holding delegate selection events are required to have a standard 15% threshold to qualify for delegates.

Proportional, then, is only completely proportional in a few rare instances. In the majority of states, proportional tends to mean the qualifying candidates receive a higher share of delegates (relative to raw vote share) because of a series of additional limiting components.

2. Qualifying Thresholds

All of the states in the Democratic process and the majority of proportional states on the Republican side have some qualifying threshold that must be met in order for a candidate to be eligible for delegates. Again, that threshold is set at the national party level in the Democratic process at 15%. Therefore, to be awarded any delegates a Democratic candidate must win at least 15% of the vote statewide and/or in a congressional district (see point 4).

Things are more variable for Republicans. The GOP has set up a “proportionality window” for the states that go between March 1 and March 14, ostensibly preventing them from awarding delegates in a winner-take-all fashion. States in that proportionality window and those that opt to be anything other than winner-take-all — whether statewide and/or at the congressional district level — the Republican National Committee allows a qualifying threshold up to 20%. A limited number of states — those described above — have no threshold. But among the other 28 states that fall in the Republican proportional category, the thresholds range from 5% to 20%. The higher that marker is set, the smaller the number of candidates qualifying for delegates is likely to be.

The true impact of the qualifying thresholds on the ultimate delegate allocation also hinges to some degree on the amount of winnowing that has taken place before any given contest. While a proportional allocation method with any threshold limits the number of candidates who qualify — making the distribution less proportional — the impact is different depending on the number of candidates still in the race. For a large field, a qualifying threshold not only limits the number of candidates who end up with delegates, but also potentially makes the allocation less proportional.

Take Iowa, for example. If the threshold in the Hawkeye State had been set at 20%, then only Ted Cruz, Donald Trump, and Marco Rubio would have qualified for delegates. Rather than Cruz claiming nearly 28% of the delegates, as the Texas senator did because Iowa Republicans set no threshold, Cruz would have won 36% of the Iowa delegates under a set of rules that included a 20% threshold. The allocation becomes less proportional with the threshold and a group of candidates below the threshold claiming a significant share of the vote. Nearly a quarter of the vote was won by candidates below 20% in the Iowa Republican caucuses.

In a scenario where there are fewer candidates, though, the outlook is different. If there are just two candidates and the race is reasonably competitive, then a qualifying threshold is less restrictive on the eventual delegate allocation. On the Democratic side, then, so long as Bernie Sanders and Hillary Clinton exceed the 15% threshold, the allocation will continue to be pretty close to mathematical proportionality. There are no other Democratic candidates credibly vying fora sizable chunk of the vote that would come under the threshold. That is in contrast to the more crowded Republican field. As the GOP field winnows, however, the delegate share of the remaining candidates above the threshold would tend to be a more proportional reflection of the vote share. And that will likely be true regardless of where a threshold is set. The candidates not qualifying for delegates will tend to take up a shrinking share of votes as primary season progresses.

3. Winner-take-all Thresholds

Though there is only one type of threshold on the Democratic side, the Republican process allows for an additional threshold. Proportional states can also set a threshold that, if surpassed, can give a candidate all of the delegates from the state or congressional district. In other words, even states in the proportionality window in early March could allocate all or a greater fraction of their delegates to one candidate and still qualify as proportional under the RNC rules.

By rule, a state can set that threshold no lower than 50%. If, however, a candidate wins a majority of the vote either statewide and/or in a congressional district, then that candidate would win all of the delegates, all of the at-large delegates, or all of the delegates from a congressional district. (More on these distinctions below in point 4.)

Assuming there is a crowded field, there is a smaller likelihood of a candidate meeting such a winner-take-all threshold. That is particularly true in the earlier states on the primary calendar. As a nomination race stretches deeper into the calendar, though, and as the field winnows, the odds of one candidate hitting that threshold increase. The decreasing field is key. As a field winnows (or if the field only ever had two candidates to begin with), then triggering a winner-take-all threshold becomes more likely. This is most often relevant toward the end of the process on the Republican side. Remember, the Democrats have no winner-take-all triggers.

4. Pooled delegates

Another difference across parties concerns how delegates are grouped for the purpose of allocation. State parties under Republican Party rules have the ability to allocate all of their delegates — whether proportionally or winner-take-all — based on the statewide vote. However, all Democratic states and a group of Republican states also split that allocation up across units. That is to say, state parties can allocate a portion of their delegates — the “at-large” delegates — based on the statewide results. But those same states can award “district-level” delegates based on the vote within each congressional district in the state. Democrats actually require 75% of a state’s base delegation to be made up of delegates allocated at the congressional district level or smaller — Texas Democrats, for example, use state senate districts — and 25% to be elected at large.

The former group — having all the delegates allocated based on the statewide vote — pools all of the delegates. The latter group of states separates the allocation, basing it on district-specific votes. This, too, has strategic implications for the aggregation of delegates based on what kind of allocation system a state employs: usually winner-take-all or proportional.

First, under a winner-take-all plan there are obvious benefits. If the delegates are pooled, then a truly winner-take-all allocation, like what Florida and Ohio Republicans are using in 2016, is a potential boon to any candidate in the delegate count. But if a state uses a separated allocation, like the winner-take-all by congressional district method South Carolina Republicans traditionally use, multiple candidates can get their hands on delegates. Under such a system, the statewide winner typically wins the most delegates, but other candidates who may be stronger in particular regions within the state — rather than statewide — also have an opportunity to win delegates if they can finish first in a district.

But on the Republican side, this pooled versus separated distinction occurs in proportional states as well. On the one hand, some states utilize a separated allocation system that doles out at-large delegates proportionally based on the statewide vote, and district-level delegates in a proportional fashion based on the congressional district vote. Georgia Republicans, for example, allocate at-large delegates to candidates who win at least 20% of the statewide vote, and allocate each congressional district’s three delegates to the top two finishers in each — first place receives two, second place gets one, with no threshold to qualify. On the other hand, some Republican contests, such as Virginia’s, pool their at-large and congressional district delegates and proportionally allocate all of them (minus the three party/automatic delegates) based on the statewide vote alone.

As already mentioned, Democratic contests proportionally allocate delegates by both the state and district-level vote — though states with just one congressional district can essentially be pooled states if they don’t use even smaller districts to allocate (e.g. Montana uses east-west districts based on when the state had two congressional districts before the 1990 census). So in a state like Vermont, at-large and district-level delegates are proportionally allocated by the same vote.

5. Backdoor Winner-take-all Triggers

A crowded field of candidates also opens a potential backdoor to a winner-take-all allocation. Based on the combination of the Republican National Committee rules and the state-level rules of delegate allocation, if only one candidate surpasses the qualifying threshold, then that candidate could win all of the delegates from the state, or in a districted system, all of the at-large delegates (based on the statewide results) or congressional district delegates (based on the congressional district results).

While this is possible among a number of the states with contests in the March 1-14 proportionality window, it is dependent on not only a large number of candidates, but a large number of candidates who are able to continue winning a fairly large percentage of the vote. Given the pace of winnowing thus far during the 2016 cycle through just two contests, the likelihood of a backdoor winner-take-all trigger being tripped is decreasing as well. And as the chances of a backdoor winner-take-all scenario decrease, the odds increase for the types of true winner-take-all threshold allocations described above.

6. Winner-take-all

The Democratic Party does not allow states to allocate their delegates in a winner-take-all fashion, and while the Republican Party does, states with contests outside of and after the proportionality window have not moved en masse to adopt winner-take-all rules. There are only five truly proportional states, as stated above, and only 10 truly winner-take-all states and territories. That is more than the six winner-take-all states in 2012, but it is still a limited part of the delegate allocation puzzle in 2016. The Republican Party has always left the decision up to the states, and few actually opt for a winner-take-all method.

What the system is left with in between truly proportional and truly winner-take-all on the Republican side and across the Democratic process is a majority of states with rules that might best be described as winner-take-most or winner-take-more. But the winners aren’t the only beneficiaries of this system. Those at the top of the order in a state win delegates while those further down in the tally do not. Who finishes at the top, though, has a lot to do with how quickly the field winnows. That input says much about how the limitations discussed here affect the cumulative delegate count.

7. Released Delegates

Once the delegates are allocated, there is at least one additional factor to consider: What happens to a candidate’s delegates when that candidate withdraws from the race? As is the case elsewhere, there are differences across the two major parties. There is variety on the Republican side and a more uniform process in the Democratic system.

Republicans leave the question of how to release delegates up to the states. In a state like Iowa, the delegates are bound through the national convention’s first ballot regardless. Both Mike Huckabee and Rand Paul were allocated delegates in Iowa, and those delegates will stay bound to those candidates despite both having since dropped out of the race. The only exception is if only one candidate is placed in nomination at the GOP convention. The Iowa delegates can move as a bloc to that candidate, the presumptive nominee. Other states are not as rigid.

At the opposite end of the spectrum are states like New Hampshire. Four years ago in the Granite State, Jon Huntsman won two delegates, but immediately dropped out of the race after finishing a disappointing third. Those two delegates were free to move to another candidate of their choosing after Huntsman’s withdrawal.

Somewhere in between are states like Kentucky or Texas. Both allow for the delegates of candidates no longer in the race to be reallocated to active candidates still competing for the nomination. Normally, regardless of being released, these delegates tend to end up with the presumptive nominee at the convention.

Within the Democratic rules, the process is standardized across states. At-large delegates are reallocated similar to the rules used by Kentucky and Texas Republicans. On the other hand, congressional district delegates are free to shift to another candidate following the withdrawal of their candidate, as is common in the New Hampshire Republican Party process. John Edwards’ delegates were not released at the time of the suspension of his campaign in 2008, but were when the former North Carolina senator formally withdrew from the race and endorsed Obama.

Some of the rules behind these concepts are more consequential than others and even then only under certain circumstances. Yet together they provide valuable insight into how candidates in both the parties accrue delegates on the way to the nomination.

Josh Putnam is a lecturer in the Department of Political Science at the University of Georgia who specializes in campaigns and elections. He is the author of Frontloading HQ, a blog about the presidential primary process.