Wednesday, March 7, 2018

2020 Republican Rules Changes, Part Three: A Reflection on Delegate Incentives

Part One: Setting Expectations for the Next Round
Part Two: Early Proposals
Part Four: A Caucus-to-Primary Incentive?

The history of incentives programs
Whether the intention of a caucus-to-primary incentive is applied to a narrow list of targets or anything broader, they all belie the fact that such bonus delegate incentive programs have not historically been effective (or effective in the absence of certain conditions).

    Early experimentation
The RNC first used a bonus delegate system to entice states to later dates on the primary calendar for the 2000 cycle. To curb frontloading, the goal was to provide a bonus of five percent to states in the March 15-April 14 window, a 7.5 percent addition for contests in April 15-May 14 window, and a ten precent bonus for states with primaries or caucuses scheduled from May 15 through the third Tuesday in June. However, the experiment was met by a collective cold shoulder from the states. Only three states moved back beyond the March 15 point on the calendar from 1996-2000.
  • South Dakota shifted from a late February primary in 1996 to its traditional early June position for 2000 after three consecutive cycles in February. 
  • Oregon, too, moved back for 2000, from a mid-March 1996 primary back to its traditional position back in mid-May 2000. 
  • Finally, Wisconsin pushed back from a mid-March position alongside neighbors Illinois, Michigan, and Ohio -- a Great Lakes subregional primary -- in 1996 to its traditional spot at the beginning of April in 2000. 
All three moved to "old" positions where the contests in those states have been more often than not in the post-reform era. And Oregon and Wisconsin reverted to those positions after a one cycle break in 1996 from those traditional scheduling patterns.

But by far the biggest beneficiaries of the 2000 cycle bonus delegates program were states that were already scheduled on or after March 15 in both cycles. In other words, states that did nothing from 1996-2000 got the most out of the rules change. Given the overall response -- or lack thereof -- the RNC scrapped the program for 2004 and has increasingly focused on penalizing non-compliance in the time since. Sticks rather than carrots, then.

Despite the failed Republican experimentation in 2000, Democrats devised their own bonus delegates regime to counter the frontloading impulse among the states beginning with the 2008 cycle. The differences across the two plans were twofold. First, the Democratic system created two groups of beneficiaries: 1) states that held their later positions on the calendar and 2) states that moved back from their earlier 2004 positions to later 2008 positions. Additionally, the 2008 system the Democrats utilized amplified the delegates added. Depending on how late the contests were scheduled, states in the first category -- holding steady -- got five or ten percent bonuses tacked on to their base delegations (not including superdelegates). That was pretty consistent with the Republican system from 2000. But it was the second category that saw the true increase in incentives. Early states from 2004 that moved their primaries or caucuses back for 2008 saw either a 15 or 30 percent addition to their base delegations, again, depending on how far back the contest was shifted in 2008.

However, even with a more aggressive bonus system, the results remained about the same for Democrats in 2008 as they had been for Republicans in 2000. Ten states gained Democratic bonuses in 2008, but only two of those ten -- Guam and North Carolina -- moved back. The remaining 80 percent of states that received a delegate bump were awarded the smaller bonuses for not moving at all, but remaining late.

The scorecard for early experimentation in bonus delegate incentives, then, just did not show much success. Although, much of that can be attributed to timing rather than any specific failure of the two programs. The primary calendars in 2000 and 2008 were arguably the two most frontloaded of the post-reform era. On one hand that makes the incentive systems look even more ineffective. The system was intended to combat the frontloading of presidential primaries and caucuses, but saw the trend accelerate instead.

But on the other hand, additional factors may have been driving actors on the state level to ignore the potential bonuses. For starters, California's 1996 shift from June to late March and subsequent 2000 move to the front end of March changed the decision-making calculus in state houses across the country. With Florida, New York, and Texas among others already in early March, California's moves pushed the total number of delegates available early in the calendar to a level not seen in the post-reform era.

Such severe frontloading had the potential to resolve nomination races earlier. And that was the fear in some states: that the race(s) would be over by the time the process got too deep into the calendar. Influence over the nomination process is not possible if the the race is resolved, whether by the viable candidates other than the presumptive nominee withdrawing and/or one candidate winning a majority of the total delegates. It was on that latter route that delegate-rich California joining a growing number of states in early March had the greatest impact.

Moreover, the relatively quick pace with which both the Democratic and Republican nominations were settled in 2000 confirmed that state-level anxiety -- fear of missing out -- for subsequent cycles. It increased the likelihood that states would consider a jump to the early part of the calendar. And in fact, the DNC widened its window in which states could hold primaries and caucuses to include February for the 2004 cycle in the hopes of deciding on a nominee faster. In other words, the DNC traded for 2004 the desire to combat frontloading for the potential to resolve more quickly the party's nomination and set their sights on defeating a Republican incumbent.

While some states shifted into February for 2004, it was not until the 2008 cycle that a mass of states sought to move even closer to the beginning of the calendar year. Why? As was the case in 2000, the 2008 cycle had competitive nomination races in both parties. That can and did open up the floodgates to increase frontloading decisions on the state level.

Very simply, then, the motivation for going early in most cases far outweighed the incentives offered by the parties in either 2000 or 2008 to not do so. And bear in mind also that the parties did not offer a united front on incentives to go later in the calendar. At the national level the Republicans walked that road alone in 2000 and the Democrats did likewise in 2008.

    Successes in and after 2012?
Elements of these relationships began to change following the 2008 cycle, giving at least the impression that the tide had turned on incentives to beat back the frontloading trend. In both 2012 and 2016, there was an expansion of incentives-based success stories. However, the overarching picture is more complex. There were more states that took advantage of the revised incentives the Democratic Party offered in 2012. But again, conditions unique to the cycle may have contributed more to state-level decisions than the incentives themselves.

First, the DNC altered its incentive structure for 2012, dropping the distinction between states that were merely holding a pre-existing position late in the calendar and those that actually moved back to later dates. In lieu of that system, the Democrats created two separate bonuses. The first of these was a 10 or 20 percent boost granted to states that held primaries after late March. Those in May or later received the largest bump.

Additionally, however, the DNC attempted to further encourage the adoption of later calendar positions by offering a bonus for three or more contiguous states clustering their contests at points on the latter half of the primary calendar. And this latter bonus could be combined with the timing bonus. In other words, a state like Montana in early June could get as much as a 35 percent increase to its base delegation for 1) holding a June primary and 2) doing so alongside neighboring North and South Dakota.

On the surface, this new structure appears to have worked in 2012. 33 states and territories took advantage of some combination of the two incentives, and only one-third of them -- 11 states and territories -- benefited by doing nothing more than retaining their positions later on the calendar from previous cycles. That left 22 additional states and territories that made decisions to shift back their primaries or caucuses from 2008 to 2012.

But while that looks like a win for the incentives Democrats employed for 2012, those bonuses were only part of a broader array of factors driving state-level decision making that cycle. At the national party level, the early February collection of contests coupled with the even earlier start to primary season that those February contests had at least partially triggered, brought on some further reflection on the factors motivating frontloading. Obviously, the DNC saw enough success in or some cause to maintain some form of incentive system after 2008. Yet, both parties were forced to enforce their respective penalties for calendar timing violations in 2008.1 And when the national parties informally, yet collectively, closed February off to states other than the four carve-out states (Iowa, New Hampshire, Nevada and South Carolina), the parallel rules changes across parties left a significant number of states in need of a change to state law to comply with the new national party rules.

That unified front from the national parties -- on 2008 enforcement and a contracting of the window for non-carve-out state primaries and caucuses for 2012 -- created a negative inducement on the states with February or earlier primaries on the books as the transition into the 2012 cycle continued. Those 18 primary states had to make changes or otherwise risked sanction from the party/parties. Despite a handful of states again flaunting those rules in 2012, most states complied with the calendar rules changes.2

And there was a pattern to the movement or non-movement. States that were non-compliant with January and February contests or those that shifted into March tended to be Republican-controlled while states that moved to April and later spots on the 2012 calendar were more likely to be Democratic-controlled. That outcome was driven in large measure due to the differing stakes across parties. The competition and stakes were higher for Republicans. They had an active nomination race. Democrats, on the other hand, with an internally popular and ultimately unchallenged incumbent in the White House had less at stake during the nomination phase of the process. While Republican states had an incentive to have an earlier voice in the Republican nomination process, Democratic-controlled states could afford to slip deeper into the calendar, enticed by delegate incentives, an ulterior motive of affecting the Republican nomination process, or some combination of the two.3 The need to be early, as established during the frontloading era, was not there for Democrats in the way that it was for Republicans in 2012.

There were breaks in that partisan/competition pattern for 2012. Often that deviation was driven by budgetary constraints that forced a number of states to reconsolidate their formerly early presidential primaries with later state and local primaries (see Arkansas, California, and New Jersey) or moving an already-consolidated primary back to a traditional spot on the calendar (see Illinois). Other states had lingering state-level disputes over redistricting that forced a reconsideration of positions (Ohio) or an outright delay to when the primary could be held (Texas).

In general, though, that partisan pattern held in 2012. States that moved back the most tended to be Democratic states. Delegate incentives may have played a role in motivating how far states shifted from 2008 to 2012, but that operated alongside the unified threat of penalties in both parties and the competition-based stakes across them. And there is a strong argument to be made that the penalties motivated the move while the combination of delegate incentives, state-level partisan control, and competitive stakes influenced how large the shift was.

And the tale was similar in 2016.

Most of the motivation to move to earlier spots was on the Republican side of the equation for 2016. The RNC mostly eliminated the caucus loophole (see footnote #2) and increased its penalties for timing violations. That made the price for holding a primary or caucus before March much higher in 2016, pushing the holdout states in violation of the (intention of the) rules in 2012 to resettle into compliant March slots for the 2016 cycle. And that was a group of mostly Republican-controlled states -- Florida, Michigan, and Missouri -- in which decision-makers were not factoring in the DNC delegate incentive structure. Other late 2012 states also pushed forward for 2016 as well. They too were newly Republican-controlled (Arkansas and North Carolina) or Republican-controlled and reverting to traditional positions following the resolution of redistricting disputes (Texas).

That Republican-driven calendar movement had implications for the lure if not use of incentives to states in the Democratic process. In total, eight states lost Democratic delegate bonuses from 2012 to 2016. Of those eight, five -- Arkansas, Florida, Michigan, North Carolina, and Texas -- were Republican-controlled states with primaries. The remaining three -- Democrats Abroad, Kansas, and Nebraska -- were caucus states where state parties were making strategic scheduling decisions. In addition to those eight states losing incentives-based bonuses, Hawaii and Utah -- both caucus states as well -- moved to calendar positions and into clusters, gaining them Democratic bonuses.

But that net of six states losing bonuses from 2012 to 2016 indicates more a maintenance of the status quo for the incentive structure Democrats carried over virtually unchanged into an open nomination cycle.4 If the lack of competition mattered to Democratic adoptions of bonus-producing strategies on the state level for 2012, then the increased competition of 2016 should have raised the number of states forgoing incentives in 2016. It did, but only marginally. But again, most of the change was Republican-driven. Democrats were not in a position of power in state capitols across the country, and thus, not in a position to make changes into incentives or into earlier calendar positions. Moreover, the competition that developed between Clinton and Sanders during primary season 2016, was not fully formed (or recognized) in 2015 when decisions were being made on primary and caucus scheduling. Sanders was emerging but had not yet emerged as the not-Clinton in the Democratic nomination race by mid-2015.

Democrats, then, may tout the overall pictures of incentives adoption in 2012 and 2016 as successes, but the above is an overly lengthy way of saying that the supposed effectiveness of the incentives structure can potentially be explained away by other factors. It should also not be lost that the incentives have not faced a test under truly competitive circumstances. 2020 may offer such a test. And under heightened stakes, the trade-off between incentives for later calendar positions may be outweighed by the state-level desire to weigh in before it is too late.

But what that means for Republicans and any effort to encourage caucus states to shift to primaries presents a different set of questions. The final installment in this series will tease out the lessons from the above and apply them to a potential Republican incentive for 2020.

Part One: Setting Expectations for the Next Round
Part Two: Early Proposals
Part Four: A Caucus-to-Primary Incentive?


1 The DNC Rules and Bylaws Committee penalties on violating states like Florida and Michigan in the immediate lead up to and during primary season in 2008 was a roller coaster. The RBC first stripped both states of their full delegations in fall 2007 when neither state/state parties could either force a move of their respective primaries or accept remedial caucuses. That the Democratic race between Obama and Clinton was as close as it was kept the results of the two under the microscope throughout primary season. Just before the end the 2008 calendar, the Rules and Bylaws Committee returned to the original 50 percent delegate penalty called for in the delegate selection rules, before opting to seat the full delegations from both states at the convention. That move has often been cast as the party ultimately bowing to the states, but the DNC ex post facto reasoning on that progression and how penalties could be assessed in the future is the the penalty was in place when it counted, during primary season. The RNC took a different path, penalizing violating states during primary season and through to the convention. The true intent of the party's 50 percent penalty has always been kept, but the implementation has occasionally meant the seating of a full delegation from a violating state, but reducing those delegate votes to/by half.

2 States like Arizona, Florida and Michigan demonstrated a willingness to take the 50 percent reduction in exchange for the potential for early influence over the contested 2012 Republican nomination race. But a number of caucus states also stuck with February dates but skirted penalty because the first step in their processes elected delegates to the next tier but without a concurrent presidential preference vote.

3 This ulterior motive could be described in a number of ways: helping a more conservative candidate emerge by backloading contests in bluer states and/or hurting frontrunner Mitt Romney by depriving him of wins in perceived hospitable territory, but also as lengthening the Republican process and/or stoking internal divisions in the Republican Party. The simplest explanation is that the Democrats, with no active nomination of their own, were informally playing on the periphery of the Republican process.

4 The 2016 bonus delegate regime remained the same as in 2012 with one exception. An adjustment was made to when the bonus window opened to account for there being five Tuesdays in March 2016 as opposed to the four in March 2012. The correction meant the window opened at approximately the same point on the calendar; March 20, 2012 and March 22, 2016.

No comments: