Next, a discussion about National Political party nominations, scandal, and comebacks, as well as rules changes for the 2016 election. This is part of the state of the parties conference. It runs just under two hours. We have four papers to be presented. We hope you will have great interest in what they present. More than anything else, we hope you will have Great Questions because this group of individuals have not only studied what has been, but are giving us perspective of what is to come. I think youll find their perspectives very interesting. About party a paper power and the causal effects of endorsements. Seth is at the university of denver and eric is at the public posse institute of california. I will shut up and let them talk. Whats were here presenting this on behalf of our other co authors. We are trying to come up with a measurement of the impact of a Party Endorsement in a primary, which is traditionally a tricky thing to measure. Study toing a fun case do this. This is the state of california which just employed a new top two system of electing candidates to office as of last year. Every candidate of any party can participate starting in the june primary. All voters see the same ballot consisting of all the candidates of the parties per the top two vote getters go to a november runoff. This was one of the latest california innovations to make everything wonderful. What wases noticed happening and responded by issuing a series of endorsements voters insteer the the direction of one of their candidates. We wanted to see what kind of influence that might have. Is Main Research question what sort of impact Party Endorsements have. Do they help candidates in the primaries . This is a tricky area. There is a severe problem. Parties and other endorsers generally pick strong candidates. They pick the type of people likely to do well in the election anyway. If you look at every california candidate for assembly, state senate, and congress and compared those who got a Party Endorsement with those who did not, it looks like indoor sees more thanes that 54 those who did not. That is an absurd number. We do not think the endorsement is worth 54 points, but it might be worth something. The problem is it is hard to figure out what it is worth because they tend to do well in the first place. Question. Important if endorsements convey some benefit, parties can be influential in primaries. They can do some of the picking for us to help narrow the field of candidates. That would make them king or queen makers. They have an important effect on nominations. On the other hand, if there is no real value of the endorsement, parties do not have much power to pick people for us. They are just cheerleaders. They are jumping on and supporting whoever happens to look like a good candidate anyway. The answer to the question tells you whether we are living in a candidate centered or party centered political system. Get at the question using two Different Research designs within the same study. The first is a survey experiment in which we created three fictitious candidates and presented them to 1000 respondents in a survey. We had three candidates with bios of each. We randomized the endorsements between the two democrats. Isolates and controls for all other possible candidate factors. We are just looking to see what impact the endorsement had on peoples stated intentions. Second, we looked at the actual election results. We have a nice piece of data in that we have the endorsement votes done at the level of the county party. Howsee how much strong these candidates were, how much they were supported by county believes. We can compare those candidates who came short of the endorsements with those who got barely above the endorsement threshold. Theoretically, the candidates have similar strength and we can see how much the endorsement mattered. Themain hypotheses are that endorsements should provide some benefit for candidates. It should be a help. Whether the exact reason for that is something we are not sure about, it could be some sort of internalization of an elite norm among voters. The second hypothesis focused on the democratic Party Endorsement. We would expect a democratic Party Endorsement would matter more among democratic voters than among independents and republican voters. The third where we are unsure is it might benefits some types of candidates more than others. In california, you could say there are two main types of Democratic Candidates. Here is thee Traditional Democratic comes out of education or the labor union and there is the new democrat that might come out of the business world. We were wondering whether an endorsement might matter more for one type. Setfirst one provides some up for a survey experiment. I will just say these are three fictitious candidates for a status symbol he see that we came up with. The first is what we would call a traditional democrat. He is involved in the local school board and favors increased funding for schools. Candidate is a new democrat. He wants communitybased policing. Finally, we have our token republican who owns and operates a small business. He wants to cut taxes and red tape of all the standard republican stuff. Then, we randomized the endorsement in our survey. We contacted 1000 californians around the time of the june 24 primary. We gave people the short bios. 1 3 saw the traditional democrat getting the endorsement. New democrat getting the endorsement. What did the results look like . Here you go. Showse set sells the condition in which there was no endorsement for any of the candidates. Johnson, the traditional democrat, gets about 30 . Uthrie gets below 20 the other half of the sample goes to robinson, the republican. The move from the middle to the left where johnson gets the endorsement, you see the dark goes up. It gets a bump of about seven points. The traditional democrat getting the democratic endorsement seems to be worth seven points. If you go from the middle to the right where the new democrat gets the endorsement, that is only two or three points. This is suggestive of the idea that endorsements seem to be of greater help to the traditional democrat. I am going to turn you over to eric for the exciting conclusion of our talk. Stay tuned. We have one more graph about the experiment. As your Party Identification ofomes weaker, the effect whether or not your candidate is weaker, whichgets kind of makes sense. As you become less of a democrat, you will not be as responsive to democratic use cues. The second way we look at this because the survey experiment is sort of artificial and would not apply in the real world. We took observational data, otherwise known as real election. Ata of how muchasure showing up to conventions and caucuses, how much they supported a particular candidate. Imagine as support among elites got higher that the performance in the primary would be better because the elites are in part you can candidates who are good. The red line is the endorsement. This is roughly what you would expect to see. As party elite support grows, performance also grows. If the endorsement have some effect, you would expect those ,hat barely got the endorsement even though there is an underlying quality issue here, that those who barely got the endorsement would do better than those who barely did not get the endorsement. That is the logic of the discontinuity. If youre only looking at those near the cut point, you can make a case it is like flipping a coin whether they are on one side or the other. Sense, they are experimentally identical. Discontinuity at the point where zero is where you get the endorsement. There is a discontinuity of about 15 points, which is remarkably similar to the size of the effect we found in the survey experiment. We are not making any claims, but that is kind of cool anyway. What is always an issue with regression discontinuity design you are looking above or below, the candidate who is more skilled and who has more of some quality or resource will be able to manipulate the outcome. Maybe they have contacts with people making the vote, some kind of insider track, that they would always come out ahead. They would be able to manipulate the results and it would not be truly like flipping a coin. A would be awaited weighted coin. We looked at whether those above and below were different on other dimensions that do not have to to do with the underlying support measure. One. Is it turns out in the original graph going back one slide, there are more incumbents just above and below. There is some indication they might be manipulating. We are doing what we have. It turns out the effect is still there. We also went through and coded Democratic Candidates according to whether they had business experience or not. Very crude but roughly comparable to what we had in the survey experiment. We see basically the same difference. I am sure many in this crowd will look at the values and say a significantuals value. Es to stop using pvalu evaluate science. It is basically the same pattern. E saw before we also did the randomization inference test. Ofs is a very similar way catching leading the values. It is a similar idea. It tests more directly for the ,otion of the internal validity not worrying quite so much about whether they are representative of a broader sample, but then saying to the endorsement make a difference on this group of people in front of us . You see basically the same going down that table. You see placebo tests where we take something that should not differ above and below the threshold and compared those as well. We do not see any difference for those, but we do for the actual vote share. That is what you would want to see if it is having an effect. The experiment shows the endorsement matters. It helps the traditional democrat more than the business democrat. That is counterintuitive but interesting. All of our versions of regression discontinuity show around the same effect. Not a 54 point effect, but we never believed it was true anyway. Showing parties do not rule the roost, but they can have an important influence on so outcome of these races, we calling them constrained kingmakers. Thank you very much. Going a little further on the were going to, talk about the role of rules in the 2012 nominations, which means pretty much republicans. In 1968, the Democratic Party embarked on what is often referred to as one of the greatest Party Reforms in with thehistory commission that was the inning of an overall hall overhaul of how nominees were selected. Prior to this, party elites were choosing the candidates. Commission, the role of selecting nominees was shifted to voters. Ies and caucuses start to matter where voters can express preferences for a candidate and that culminates in the delegate selection at a National Convention. Starting with the commission, the Democratic Party made numerous reforms to the process. Tinkered with it almost every election cycle until the 1980s. The Democratic Party had no fewer than eight Reform Commission starting in 1968. While this was happening, the Republican Party was very hands off in the process and did not become actively involved in reforming the president ial nomination process. One of the most recent activities by the Republican Party is the creation of the temporary delegate Selection Committee in 2008 wave that reformed the most recent process. I look at the rules set in place by the National Parties and effect of those rules. I wanted to take a step back from that project and look at why republicans became involved after being so uninvolved for so long and what the consequences of their involvement were in the most recent nomination. There are four main reasons the Republican Party has been less involved in the president ial nomination process than the Democratic Party. The first is they have been more content with their success at. He president ial level another reason is in the Democratic Party, there was a strong section calling for reform of the process. That faction did not exist to the same extent in the Republican Party. The Republican Party is known for their position on states rights and limiting federal involvement, so it did not see the need to become involved by issuing national mandates as the Democratic Party did. Has been more it complicated to change the rules on the republican side than on the democratic side. Changesepublican side, have to be approved by four different bodies. That meant if republicans wanted the 1976hanges for nomination season, they had to implement reforms at the 1972 convention. It was a longer process and when they did not are taken like the democrats did. Regardless, the republican process did change along with the democratic process, in large part because of changes at the state level. As a result, many states switched to Holding President ial primaries. That was often the result of changes in state laws. When the states created these laws, they also did so for the republicans at the same time. Even though there was not the same involvement, the Republican Party process also changed. The Republican Party did not resist those changes. Again, for several reasons. This move was seen as a popular performed reform. The Republican Party was not willing to go against that reform the people had embraced. 1976, the first time the Republican Party was hosting a competitive nomination in the postreform era because in 1972 resident nixon was being the public accepted these reforms because they had seen it play out in 1972 on the democratic side. In 1976, there were two popular republican candidates running and party elites were willing to accept input from the public on which candidate they preferred. Finally, the Republican Party had seen in 1972 the National Attention from the media and voters that the democrats received and they were not willing to quietly nominate a candidate in 1976 while the Democratic Party got the spotlight. To thehanges happened republican process as well, but not because of their direct involvement. We see the first major attempt at reform on the republican side with the creation of a task force on primaries and caucuses at the 1996 convention. The republicans wanted to combat frontloading, which is states moving their contests up earlier in the nomination season to gain influence on the process. The Republican Party believed frontloading prevented voters from having meaningful participation and also harmed abilities to fund raise over the compressed calendar. To combat frontloading, the Republican Party offered bonus candidates to the states. That was supposed to be an incentive to the states to hold a later contest, have more delegates, and make the contest worth more to the candidates. We see the 2000 republican nomination was frontloaded. It started three weeks earlier was the 1996 calendar and more frontloaded. The Democratic Party was more Successful Holding states back and saying you cannot hold your contests in february than the Republican Party was. These reforms were seen as a failure. That brings us to the republican second major attempt at reform with the selection in many formed in 2008 to make reforms for 2012. The creation of this committee is significant not only because it is one of the rare instances where the Republican Party has become involved in the last 40 years, but also because they allow changes to go into effect for the previous nomination. These changes did not need to be approved by the National Convention and just needed to be approved by the National Committee in 2010 and would go into effect. The republicans once again tried to reduce frontloading, create a longer nomination season, and allow more voters a say in selecting the nomination. The nomination quickly. Obama democratic side, and clinton battled it out for months and we saw Media Attention at all time high. The goal was to create a more exciting, link the nomination that would pull voters into the process. They tried to achieve these goals in two weighs per the first was by regulating the calendar. The earlys said on or afterote february 1 and before the first tuesday in march where all other states have to go after the first tuesday in march. They also said states voting before april 1 had to use allocatenality to delegates. This was supposed to be seen as making the states more influential. If the state held a contest later, it could give all delegates to the winning candidate. In than relying solely on incentives, the Republican Party said we will enforce a penalty if the states do not abide by our rules. It said we will take 50 of your delegates away if you break these rules. The penalty, the republicans were not able to prevent states from ignoring the rule and gladly accepting the penalty in scheduling early contests, as they should have inected given the actions 2008. In 2012, we see movement in moving earlier and then creating a Ripple Effect where others are going to move their contests earlier to preserve their early status and influence in the process. As