the first two parts of this series, we looked at the some background information that applies to both parties (namely the roles of the national parties, the state parties) and state legislatures and some of the basic rules that the Democratic National Committee requires all state parties to follow. As noted in the previous post, the Democratic National Delegate Selection rules recognize four separate categories of delegates — district level delegates, at-large delegates, pledged PLEOs (party leaders and elected officials, and unpledged PLEOs (a/k/a super delegates). This post will look at how those delegates are allocated to the states (and then to the individual categories within the states).
The Democratic Party begins delegate allocation with a base of 3,200 district level and at-large delegates for the fifty states and the District of Columbia. Each state’s initial share is determined by a two-part formula. The first part averages each state’s share of the national Democratic vote over the last three Presidential elections. The second part looks at each state’s share of the total electoral vote (e.g. California has 55 electoral votes out of the national total of 538, or slightly over 10%). Those two numbers are averaged together to get a final ratio. (The effect of this formula is that the more Democratic states do slightly better than a pure allocation based on electoral votes and the Republican states do slightly worse. For California, rather than the 10.2% that California would get from electoral votes, this formula gives California 11%.) When this formula leaves a state with a factional delegate, .5 or higher gets rounded up, anything less than .5 gets rounded down.
After the base allocation, the Democratic Party gives a bonus for holding primaries or the first stage of a caucus (for those states using caucuses to award delegates) later in the process. For states holding a regional primary (at least three adjoining states) on or after March 28, the states get a 15% bonus. If a state begins in April, the state gets a 10% bonus. If a state begins in May or June, the state gets a 20% bonus. The bonuses for going after April 1 or on top of any regional primary bonus.
For the states and the District of Columbia, their “base” allocation is split between district-level delegates (75% of the delegation) and at-large delegates (25% of the delegation). In addition, each state gets 15% of their base delegation as pledged PLEOs. Outside of the states and the District of Columbia, the remaining territories and Democrats Abroad are granted a specific number of delegates by the Call for the Convention. (Most of the territories get 6 at-large delegates, but Democrats Abroad get 12 at-large and 1 pledged PLEO and Puerto Rico gets 44 base delegates.) In addition to the delegates, each state gets one alternate for every 12 delegates.
For the purposes of district-level delegates, states use either congressional districts or some smaller district. A state with one congressional district does not need to split the state for the selection of the district level delegates although, in the past, some have. For the purposes of allocating the district-level delegates, states are given three basic options: 1) a formula based on population and the average vote for President in the last two elections (similar to the formula used by the DNC to allocate delegates to the states); 2) a formula based on the Democratic vote for President and Governor in the most recent election for those offices; and 3) a formula based on Democratic Party registration in the state and the average vote for President in the last two elections.
Going back to the 3.200 base delegates, with district delegates representing 75% of that number, there are approximately 2,400 district-level delegates which (with 435 Congressional Districts plus one for DC), the average number of delegates per congressional district is between 5 and 6 delegates. However, once a State has opted for one of the three formulas and allocated their district-level delegates between the districts, the reddest districts tend to get 4-5 delegates and the bluest districts tend to get 7-8 delegates. Because the Democratic Party requires these delegates to be awarded to candidates proportionately, it is possible (in advance) to determine what percentage of the total vote gets a candidate an additional delegate. For example, in a four delegate district, a candidate wins her first delegate with 15% of the vote, a second delegate with 37.5% of the vote, a third delegate with 62.5% of the vote, and the final delegate with 85.1% of the vote. In a seven delegate district, a candidate wins his first delegate with 15% of the vote, a second delegate with 21.5% of the vote, a third delegate with 35.8% of the vote, a fourth delegate with 50% of the vote, a fifth delegates with 64.3% of the vote, a sixth delegate with 78.6% of the vote, and the final delegate with 85.1% of the vote. A candidate with good polling data and good field operations can use these numbers to choose a handful of districts were a small increase in the vote would secure an additional delegate.
Beyond getting to choose the mechanism for allocating district level delegates to districts, a state party (within limits) gets to choose when/how delegates are awarded. There are two basic mechanisms used to award delegates.
First, and most common, is the binding primary. On a day typically scheduled by the state legislature, voters go to their typical polling place and cast a secret ballot. Election officials count the ballots and the results of those ballots determine how delegates are awarded with the results in each district determining the district-level delegate split and the state-wide result determining the at-large delegate and pledged PLEOs splits.
The other mechanism is the caucus/tiered convention system. In this system, a mass meeting is held in a local unit and delegates are elected from the local unit to go to a regional convention and/or state convention. For states that opt for caucuses, there are some differences in how the states award delegates. Some states use the initial vote at the local level (either by sign-in sheet or other mechanism) to award delegates to the national convention. On the other hand, some states use a later level to award delegates. For example, in the past, Iowa only awarded district-level delegates at the Congressional District conventions and at-large delegates at the state convention.
Because caucuses tend to have a lower turnout than primaries, turnout operations are key to which candidates prevail and candidates can win delegates with much lower vote totals than would be required if the state had a primary (one of the reasons why total national vote can be misleading in a close race). Additionally, in states that award national convention delegates later in the process, some delegates elected to attend the later meetings will have been pledged to support a candidate who is no longer running or a candidate who does not have enough supporters at the higher level to win delegates (in most circumstances due to the candidate only winning delegates in some local meetings and falling short in other local meetings). A campaign that can do a good job at converting these de facto uncommitted participants to their candidate at the district and state level meetings can pick up a couple more delegates than their initial round of support would suggest.
Aside from the power of the state parties to designate the stage in the process at which delegates are awarded, the state parties have significant freedom to designate how the actual people who serve in delegates are elected. Obviously, in caucus states, district-level delegates are selected in a district-level convention. In primary states, some states require district-level delegates to run on the primary ballots. In other states, the state parties use a caucus/conviction system to select the district-level delegates. For primary states using a caucus/convention system, the delegate selection meeting can occur prior to the primary (in which case, each candidate has a slate of potential delegates in a priority established by the district convention and the results of the primary determines which of the “nominated” delegates is actually elected) or after the primary (in which case, the candidate’s caucus only elects the number of delegates that the candidate won). Whatever process the primary state uses, each congressional district delegation must comply with the gender balance rules. Furthermore, if a state uses a caucus/convention system (either instead of a primary or to elect delegates allocated by the primary), the state must use a formula accounting for population and past election results in determining how many delegates each local/regional convention gets to elect to the next meeting.
State parties also have some discretion in structuring the election of at-large delegates and pledged PLEOs (and do not have to use the same method for both). First, the state parties can assign the task to a state convention. Second, the state parties can assign the task to the party’s state committee. Third, the state parties can assign the task to a meeting of the district-level delegates. To the extent that a “caucus” state uses either of the last two methods to pick at-large delegates or pledged PLEOs, that state must use some mechanism (either the composition of the district-level delegates or presidential support at an earlier level of the process) to award the at-large delegates or pledged PLEOs to particular candidates.
Before leaving the Democratic Party process, two last details to note. For pledged PLEOs, the rules create a “priority” for persons to be considered (beginning with state-wide officials and big city mayors then moving to state legislative leaders then to other legislators then to other party leaders). The rule does not quire require that the slots be filled in the order of priority but does strongly suggest that the slots be filled in that order of priority. (To use my state as an example, we will have ten pledged PLEOs, four state-wide officials and maybe four big city mayors. If the suggestion is treated as binding, the election of pledged PLEOs would merely ratify these eight names and then choose two from the legislative leadership for the remaining slots.)
The other detail is the unpledged PLEOs. To some extent, this group (currently 739 delegates) was designed as a check on the primary process. If there was some reason why the leading candidate was not a “viable” candidate, these party leaders could swing the convention to the runner-up. The other thing is that the exact number of unpledged PLEOs is fluid until the eve of the convention. Delegate slots go to current members of Congress, current Senators, and current Governors (a current total of approximately 255). Vacancies can occur in any of those positions at any time, adding more delegates if a Democrat wins a seat currently vacant or held by a Republican or losing delegates if a Republican takes over a seat currently held by a Democrat. In addition, retirements by current party leaders could increase the number of slots for former party leaders (and sometimes health issues keep former party leaders from attending the convention or they simply opt not to attend). Unlike other delegate slots, there are no alternates for unpledged PLEOs. So if a vacancy occurs on the eve of the convention or the delegate opts against attending, that delegate does not get replaced.
Having summarized what is a very complex procedure, let’s get back to the original point. How do the rules impact the process. First, the unpledged delegates and the early states form a screening process. If a candidate can get endorsements from a significant number of unpledged delegates, some news organizations will include those endorsements in the delegate count, potentially allowing a candidate to survive mediocre results in the early states. Without such support, candidates need to, at-least, meet the fifteen percent threshold in the early states. A candidate who has not won any delegates after the first “open” Tuesday (March 1st in this cycle) will have trouble continuing further. Second, if the race is close, being prepared to contest every caucus state and every primary matters. Currently, it looks like there will be 8 primaries or caucuses on March 1 with another 12 scheduled later that month. A candidate who has to re-tool on March 2nd could find themselves permanently behind on March 15. Third, in close races, finding the districts were you can gain a delegate is crucial. At-large delegates are much easier for candidates to gain(depending upon the state as little as a 1-2% swing), but the candidate who can get those extra votes in the district that matters gains a big edge. Finally, any discussion of the rules depends upon the unwritten rule — it takes money to run a campaign. Most campaigns die because money dries up when the candidate falls off the pace. The use of good tactics to gain extra delegates only makes a difference if you are competitive in the first place.
Depending upon what happens with a certain former Secretary of State over the next month, Democrats may see a primary season that resembles the 2000 primary with few candidates and a clear early winner. On the other hand, we could see a repeat of 2008 in which a good tactical understanding of the rules and a wise use of data played a key role in determining which of two equal candidates just barely won the race.