Wednesday, May 9, 2018

DNC Rules and Bylaws Committee Tweaks Delegate Allocation Formula the margins.

On Tuesday, May 8, the Democratic National Committee Rules and Bylaws Committee (RBC) reconvened for its fourth series of public meetings of 2018. This meeting -- at this particular point in the cycle -- is typically the mark-up meeting. The RBC starts going through the delegate selection rules in sequence, offering proposed amendments along the way.

The session on Tuesday pushed through much of the rule book that fits under the purview of the mission at FHQ. For example, while the RBC punted on the substance of the superdelegates question of the cycle -- the actual reduction of the role of the unpledged delegates -- but laid the groundwork for a change. It added a new rule, creating a new class of automatic delegates by breaking that group off from the broader group of unpledged delegates formalized in the old Rule 9.

Additionally, the RBC retained the same basic framework of the primary calendar from 2016, adjusting it to dates falling on different days in 2020 than 2016.1 Under the new Rule 12 Iowa, New Hampshire, Nevada, and South Carolina will keep their positions in February and other states will continue to have a window from the first Tuesday in March through the second Tuesday in June to conduct their respective delegate selection processes.2

But there was an innovation to the DNC delegate allocation formula the RBC adopted yesterday that deserves some attention. It is not that the change will have significant effects -- FHQ is skeptical of that -- but that it is a change to a formula that has been quite uniform across all states in the Democratic nomination process for years.

How did the formula change?

The RBC was on the verge of quickly dispensing with the new Rule 14 when co-chair Jim Roosevelt interjected to inquire if the committee was behind leaving the qualifying threshold described in Rule 14.F in place. Now, typically the threshold for qualifying for delegates is set at 15 percent. Those provisions for district delegates (Rule 14.B) and at-large and party leader and elected official (PLEO) delegates (Rule 14.E) are unchanged.

Rule 14.F lays out the threshold in the event that no one gets a share of the vote at or above 15 percent. That has been set at within 10 percent of the frontrunner in past iterations of the rule.3 In other words, if the frontrunner receives 14 percent of the vote statewide, then all candidates who received between four and 14 percent of the vote qualify for a proportional share of the delegates allocated based on the statewide results.

As Roosevelt pointed out, this has not tended to be an issue, but in the case of a potentially large field of candidates, it is more likely that the vote gets split into small slivers and no candidate crests to or above 15 percent. And depending on what share of support the "frontrunner" receives, a lot of candidates from a large field could end up within 10 percent of the "frontrunner" and eligible for a proportional share of delegates.4

To mitigate some of the associated issues attendant to the formula in combination with a large field, the RBC devised an elegant solution. Under the 2016 Rule 13.F, there was a fixed 10 percent qualifying window under the share of the top votegetter. But the change for 2020 in Rule 14.F would make that qualifying window in the event that no candidate receives at least 15 percent of the vote more dynamic.

The old fixed 10 percent window will become a bit more elastic, conditioned by what share under 15 percent the "frontrunner" gets. If the leading candidate wins, for instance, 12 percent of the vote, then all those candidates within six percent -- half -- of the top finisher would qualify for delegates. If the winner got eight percent, then all those within four percent -- those between four and eight percent -- would qualify for delegates.

Again, while this is a fun little tweak that grabs the attention of delegate nerds, the change is very likely to have only marginal effects in a race with even a large field.


Winnowing is going to happen with or without this change and the bulk of that process will take place in February during the pre-window period in which Iowa, New Hampshire, Nevada, and South Carolina will hold contests. That quartet is not a series of delegate contests, where accruing and counting up delegates is the goal for campaigns. Rather than delegate contests, the four early carve-out states are momentum contests in which winning is the metric that separates the wheat from the chaff.

Even if the early states were delegate contests, there just are not all that many delegates at stake. Couple that with the fact that all of a state's delegates are not pooled and allocated together and it minimizes the impact of the allocation formula change even further. In other words, there is not one collection of delegates in a state allocated as a group based on the statewide results. Delegates are split into smaller subunits: at-large delegates, party leader and elected official delegates, and district delegates. These smaller, early states, then, already have small pools of delegates, but splitting their allocation up means that a proportional share of a subset of an already small share of delegates does not amount to all that much.

Take New Hampshire, for example.

There were 24 pledged delegates in the Granite state in 2016: five at-large delegates, three party leaders and elected official delegates, and eight congressional district delegates in each of the two congressional districts (16 total). Even if multiple candidates receive more than the typical 15 percent to qualify for delegates, there are only so many ways five at-large delegates can be proportionally allocated. After all, there are only so many delegates to go around. That principle is still true under the revised formula.

However, where this altered allocation formula can be of some concern is in the event that there is significant clustering among candidates in the results in early states. How can winning be assessed if, say, ten candidates are all clustered not only under 15 percent, but within a few percentage points of each other? In that scenario, this new rule ends up being pretty arbitrary in its now more dynamic line of demarcation between who gets delegates and who does not.

Truthfully, clustering or not, arbitrary or not, the order of finish will matter, and the next contest with come barreling down the pike to confirm in whole or in part the results from the previous contest of contests.

This is probably entirely too much of a treatise on such a small rules change. In the end, the matter is the change of just a few words in one rule in a big book of rules. However, as an aside, something should be said about the action here on the part of the RBC. First, co-chair Roosevelt exhibited the power of suggestion that the chair can have in these proceedings. Again, the committee was seemingly ready to affirm the new Rule 14 and move on. But Roosevelt raised the issue and the can of worms was opened, and more importantly a rule was altered.

That is neither necessarily good or bad. But it should be noted that it breaks the Democratic rules-making process from a pattern of fighting the last battle in which out-parties -- having just lost a presidential election if not the White House -- often find themselves. Instead, this small maneuver was forward looking; planning for the conditions of the next cycle rather than fixing the perceived problems of the last. That is no small thing, but in view of the 2020 rules making thus far, it is exception rather than rule.

1 What was interesting about the calendar discussion -- and particularly the controversial issue of the protected status of the four carve-out states -- was that no one raised the Iowa/New Hampshire question. In fact, section B of the new Rule 12 drew more of a discussion than section A, which deals with scheduling. Section B has required of states in the past that they begin and complete all steps of their delegate selection events in the calendar year of the presidential election. The rule specifically cites presidential candidate filing as an activity that falls into the delegate selection process. But that is also an activity that has stretched beyond the calendar year into the year before the presidential election. Functionally, that required of those states where that activity pushed out of the calendar year to petition the RBC for a waiver.

In the new draft rules for 2020, there was a proposed change to the calendar year requirement that granted states the latitude to make those decisions without the worry of going through a waiver process. However, the members of the RBC balked at that change, questioning whether it gave incentive to states to push activities into the year prior the presidential election. Rather than go down that road, the RBC opted to return to the original language. In turn, that kept the calendar year requirement in place and continued to necessitate a waiver from states where state law deadlines may violate the rule. The intent of the change -- of alleviating the burden -- was more focused more on the RBC than the states. But the desire to avoid motivating states to make changes that pushed parts of their delegate selection events into the year prior to the presidential election outweighed the need to relieve any undue administrative burden on the RBC. The scope of states required to petition for a waiver for this purpose was not particularly widespread anyway.

2 The new description should clarified. That does not mean that a completely new rule was created. That did happen with the new automatic delegate rule, but in the rest of the cases above where new is used it means new numbering relative to how the rules were numbered in 2016. The insertion of the automatic delegate rule bumped all subsequent rules down a slot in the sequence. The 2016 Rule 11 on timing, for example, became the new Rule 12 for 2020.

3 "Frontrunner" in this case refers to the candidate who has received the most votes; the top votegetter statewide and/or in a district.

4 And if the top candidate garners 10 percent or less of the vote, then everyone who received a vote would qualify for delegates. There would be no effective threshold in that case.

No comments: