Quantifying the Value of Nikki Haley’s Endorsement

It was leaked this afternoon that South Carolina Governor Nikki Haley will endorse U.S. Senator Marco Rubio.  Questions about Haley’s endorsement have persisted for months.  Until today, the only reliable information was that she would not be endorsing Donald Trump.

Needless to say, Haley’s endorsement is welcome news for Rubio’s campaign as voters head to the polls on Saturday for the GOP primary.  But just how much is Haley’s endorsement worth?

According my data, Haley’s endorsement should net Rubio an additional 4% on Saturday.  But first, a little background.

Perhaps the most contentious issue among election forecasters this cycle is the extent to which party elites shape presidential nomination contest outcomes.  Although primaries and caucuses give voters the power to decide who wins the nomination, and even though parties are hardly monolithic teams that act in perfect harmony, the conventional wisdom is that party elites do indeed have considerable power over who wins the nomination.

In large part this view stems from a book appropriately titled “The Party Decides.”  Written by political scientists Marty Cohen, David Karol, Hans Noel, and John Zaller, the book proposes that one of the key tools used by party elites is the power of their endorsement.  It is a testament to Cohen et al.’s work that Nate Silver–the most influential election prognosticator–has written extensively about the party decides thesis and has even catalogued endorsement data this cycle.  Silver calls it the “endorsement primary.”

Although most political observers accept the general wisdom of the “The Party Decides,” the contentiousness surrounding this book concerns how (if at all) Donald Trump fits into the thesis.  Given that most Republican elites are apprehensive (to put it mildly) toward a Trump nomination, and Trump he has secured so few endorsements, there are questions about the book’s predictive power in 2016.  For a more thorough discussion of these issues, see Nate Cohen and Nate Silver.

Back to Nikki Haley and Marco Rubio.  If endorsements do indeed affect who wins the nomination, it should be possible to quantify just how much Haley’s endorsement is worth.

I have been working on a project with two College of Charleston students that examines the factors that correlate with success (or failure) in South Carolina’s nomination contests.  Our data cover the period from 1988 to 2012, giving us an even balance of five contested Democratic elections and five contested Republican elections.

Our primary predictors of a candidate’s vote share include their performance in New Hampshire and Iowa, how much media attention the candidate received, a candidate’s demographic characteristics, and the volume of their endorsements by state party officials.

We plan to release the model’s full predictions Friday evening and discuss more fully what factors have shaped South Carolina’s election results from 1988 to 2012.  Spoiler alert: endorsements are indeed an important factor in the model.

But just how much does a governor’s endorsement matter.

According to the model, from 1988 to 2012 the governor’s endorsement has netted the endorsed candidate an additional 4% of the vote.

Clearly, 4% is not enough to vault Marco Rubio past Donald Trump (at least, not on its own).  According to most polls, Trump has a 19-point lead over Rubio (35% to 16%).  However, a 4% improvement in his vote share would put Rubio in second place ahead of Ted Cruz (who is polling around 18%).

Going forward, a second place finish would be of considerable value for Rubio’s campaign.  A third place finish, however, could dim Rubio’s long term prospects.

Posted in Elections, Political Behavior, Political Parties, Primaries, Voting Behavior | Tagged , , , , | 2 Comments

How long will the “open process” last?

During the Republican retreat two weeks ago Speaker Ryan doubled down on his commitment open the process in the House. The original pledge was offered to satisfy conservative members’ desire for greater input and influence. Anyone with a deliberative-democratic bone in their body should welcome this change and the pledge. However, it comes with some major risks. And if history is any sort of guide, this commitment may be short lived.

The tweaks to the Republican Steering Committee haven’t changed much in the House. House majority leaders still control the process. Any bill of consequence will need to go through the Rules Committee, which is still firmly in the hands of Speaker Ryan. This means the Speaker is voluntarily opening the process to his rank-in-file and the minority for amendment, deliberation, etc.

In the past these types of voluntary commitments have had a limited shelf life. Since the reforms of 1975, restrictive rules – that limit opportunities to amend legislation on the floor – have been the norm. The speaker puts a bill under a process that prohibits members (either totally or in part) from amending its language. Members can only vote for or against the bill. Restrictive rules were used a whopping 92% of the time in the 113th Congress. That’s down from the record high of 98% in the 111th, but dramatically up from previous Congresses. For example, restrictive rules were used 75% of the time in the 108th, 55% of the time in the 101st, and only 16% of the time in the 95th Congresses.

Leaders use restrictive rules for a variety of reasons. But one of the most important reasons is to limit minority party mischief. When in the minority, both Democrats and Republicans have used open rules to disrupt policymaking. They intentionally offer amendments to drive wedges in the majority party coalition, making final passage of the bill difficult or impossible.

In the late-1980s, Republicans became a very effective opposition by offering amendments that undermined policy on the floor. Democrats responded by increasingly closing opportunities for amendment. Republicans retook the majority in 1995 promising to achieve what Democrats couldn’t: more open debate and amendments for the rank-and-file. Democrats reacted predictably, offering hundreds of political amendments that undermined the new majority’s bills. Republicans leaders reigned in their commitment to an open process just a few months after taking office.

Polarization and partisan warfare have undermined serious efforts to reopen or maintain amending processes in the House and Senate. This week the Senate is struggling to pass an energy bill through a maze of amendments ranging from small adjustments to charged language relating to the Flint water crisis.

Given the polarized environment of the House, if Speaker Ryan wants to keep his pledge to open the process he may also need to commit to failing. An open process will undoubtedly force votes that many of his members would rather not see on the House floor. It’s downright likely that an open process will cost the majority some major policy victories at the expense of more deliberation. Given the size and safety of the House majority, Ryan has some freedom to take these hits and retain the majority. However, Speaker Ryan’s commitment to open deliberation will likely come at a cost that include policy wins and possibly seats in the House.

Previous leaders changed course after flirting with open procedures. A couple months of costly political votes were enough to renege on their promises. The extent the process is opened, or remains open, in 2016 will depend on how much risk Speaker Ryan and his colleagues are willing to shoulder. That willingness will be closely tied to the number of vulnerable Republican seats and presidential odds in 2016.

| Leave a comment

Is Nikki Haley a Strong Vice Presidential Candidate? Evidence from Research on the “Veepstakes”

After seven years in office, last night President Obama gave his final State of the Union address.  Immediately following Obama’s speech, South Carolina Governor Nikki Haley delivered the GOP’s primetime response.  Naturally, journalists and pundits are debating whether Haley’s speech helps her chances of securing the vice presidential nomination.

In a half dozen studies , political scientists have examined the factors that increase a candidate’s chances of being chosen in the so-called “veepstakes.”  Based on this body of research, it seems to me that Haley matches a lot of the criteria found to predict who is selected to join the ticket and should be considered a strong candidate.

What follows is based on four studies: two by Baumgartrner (2012 & 2008), one by Sigelman and Wahlbeck (1997), and another by Hiller and Kriner (2008).  Although the selection of a running made would seem (on its face) to depend on the preferences of the presidential candidate and the context of the election, these studies have a fairly high degree of predictive accuracy over a fifty year period.  For example, Baumgartner’s (2012) model correctly predicted 70% nominees from 1960-2008 while Sigelman and Wahlbeck’s model correctly predicted 68% from 1940 to 1996.

One of the most important factors is the candidate’s age.  On the one hand, Sigleman and Wahlbeck (1997) and Hiller and Kriner (2008) find that presidents seek an age balance on their ticket.  Older candidates were more likely to select a younger running mate while younger candidates were more likely to select older candidates.  At the same time, Baumgartner (2008, 2012) finds that younger candidates have a higher chance of being chosen (all else equal).

At just 43 years of age, Nikki Haley is younger than all of the Republican presidential candidates (Marco Rubio, the youngest, is 45, and Ted Cruz is 47).  Aside from Rubio and Cruz, Haley is at least ten years younger than the rest of the Republican field and would thus help balance the ticket (she is Donald Trump’s junior by 27 years).

Another consistent factor in these studies is whether the potential running mate ran against the party’s presidential candidate in the primary .  As you might expect, rivals are less likely to be selected (Sigleman and Wahlbeck 1997; Baumgartner 2008).  Another point in favor of Nikki Haley.

Significant media exposure is a third factor that seems to increase the probability of being chosen as a running mate (Baumgartner 2008, 2012).  Needless to say, Haley’s GOP rebuttal will almost certainly go a long way toward increasing her national media exposure.  And although Haley is not as well-known as other Republican governors (Chris Christie, Rick Scott, Scott Walker), she gained significant national exposure in the aftermath of last year’s shooting at Emanuel AME Church and for her leadership in removing the Confederate Flag from the South Carolina Statehouse grounds.

Receiving much of the attention in debates of Haley’s likelihood of being the GOP running mate is her ability to balance on the ticket in a demographic sense.  As a woman and an Indian American, she would certainly add diversity to the ticket.  In two studies, gender and ethic balance were found to increase the likelihood of selection (Baumgartner 2012; Hiller and Kriner 2008).

Lastly, there is some evidence that a candidate’s experience matters.  Generally, the longer a candidate has served in national or subnational office, the more likely they are to be tapped as a running mate (Baumgartner 2008; Hill and Kriner 2008).  Having served as South Carolina’s governor for five years and in the South Carolina House for six years before that, Haley certainly has some notable political experiences.

Although Nikki Haley certainly fits the above criteria, there are some potential negative effects that should be discussed.

Conventional wisdom says that presidents select running mates from large states in an effort to increase their vote share.  However, the effect of state size has yielded mixed results in the above studies.  State size matters in Sigleman and Wahlbeck’s (1997) analysis but not in Baumgartner’s (2008) or Kriner’s (2008) analyses.  Notably, the two most recently published articles (with data for the most recent elections) find no effect of state size.  In contrast, there is no evidence that presidents are more likely to seek a regional balance on their ticket, picking a running mate from a different part of the country (Baumgarnter 2008; Hiller and Kriner 2008; Sigelman and Wahlbeck).

So, South Carolina’s relatively small population size may not be as big a deterrent as most people think.  And even if Marco Rubio or Ted Cruz win the GOP nomination, an “all south” ticket does not seem to hurt Haley’s chances either.

Another common claim is that presidents select candidates in an attempt to boost their vote share in pivotal “swing” states.  One study, however, suggests that running mates do not add very much in terms of additional votes (Dudley and Rapoport 1989).  In this study, presidential candidates gain just 0.3 percent in their running mate’s home state.  In all likelihood, the small bonus in the in the running mate’s home state is simply not enough to overcome other strategic considerations.  As evidence, consider that the last two vice presidents were from solidly blue and red states (Delaware and Wyoming) as were the last two running mates on the losing side (Wisconsin and Alaska).

In the above studies, only one factor is a clear negative for Haley’s likelihood of selection: she never served in the military.  In both studies by Baumgarner (2012 & 2008), veterans were significantly more likely to be chosen.

All in all, Nikki Haley seems like a strong contender for the Republican Party’s vice presidential slot.  Based on studies in political science, she matches most of the criteria identified as key predictors of who gets selected.  Of course, the running mate will depend on who wins the GOP nomination in the first place.  But Haley looks like a strong choice at the present time.

Posted in Elections, Political Parties, Primaries, Voting Behavior | Tagged , , | 1 Comment

What’s on Tap in Congress in 2016?

This was originally published at the Government Affairs Institute at Georgetown University.

The 114th Congress was a whirlwind of activity compared to its predecessors. Accomplishments like trade promotion authority, a Medicare “doc-fix” solution, a two-year budget deal, and the highway funding act were legislative highlights in a productive first session. In all, the 114th Congress passed 115 laws, the most in a first year of Congress since 2009.

Speaker Boehner’s last order of business was the “clean out the barn.” Several hurdles that impeded productive lawmaking in the past were cleared. The debt ceiling is not an issue again until 2017. Congress ensured the budget debate is more symbolic than substantive with a 2-year budget deal. The highway trust fund has finally been resolved. Speaker Boehner’s departing gift for Speaker Ryan was, essentially, a clean legislative sheet.

Yet despite fewer obstacles, significant lawmaking will be hard to come by in the Second Session. Majority Leader McConnell has a strong incentive to protect his moderate colleagues. Seven Senate Republicans up for reelection in 2016 represent states that Obama won in 2012. On the other side of the Capitol, Speaker Ryan is tasked with navigating bills through a three-faction House: Freedom Caucus Republicans (the most conservative), establishment Republicans, and Democrats. He may be forced to find votes from different factions depending on his objective. For messaging purposes the most conservatives Republicans are obvious allies. However, on spending policy he may be forced to appeal to Democrats in order to craft a majority that can pass a bill that can also get through the Senate. If he wants to remain Speaker after the 114thCongress, he can’t lean on Democrats too often. These dynamics do not favor a productive 2nd Session.

The presidential race makes things worse. Presidential campaign politics complicate congressional politics. There is no formal mechanism linking the current Congress to presidential candidates. Nonetheless, Congress tends to reinforce their party’s nominee at the expense of their opposition’s nominee, and worry about down-ticket races. This means potential legislative goals brewing in the 114th Congress will drop by the wayside should it create friction with the nominee’s position or message. We are much more likely to see the Republican majority adapt their policy positions to the Republican nominee while also avoiding success so as to not steal their thunder. This means we are far more likely to see messaging bills than significant policy overhauls.

This does not mean Congress will go dormant. Here are some areas where policy will move forward or at least has the potential to see significant action. Those areas are outlined below with very equivocal certainty.

Most likely will definitely happen:

ACA Repeal

This has been in the works for the better part of a year now. The March budget resolutions in both the House and Senate outlined instructions to repeal Obamacare.  Republicans spent the last 8 months crafting a path to use the powerful reconciliation process (exempt from Senate filibuster) to repeal Obama’s signature accomplishment. For the first time in 5 years and over 50 attempts, Republicans will place a major repeal of the health care law in front of the President. The bill will die there.


Late in 2015 the administration completed negotiations, which started under the Bush Administration, on a 12 nation trade agreement called the Trans-Pacific Partnership (TPP). This is among the largest trade deals ever struck in terms of the number of countries involved and the size of their economies. Congress authorized trade promotion authority earlier in 2015, giving them the final say in an up or down vote that will occur later this year. This is an issue that divides ideological extremes against the middle. While the left and right have serious critiques of the deal, a sizable bipartisan coalition are at least interested in finding a way to support it. Despite rhetorical equivocation, Congress may enact one of the largest trade deals in history and in doing so, will likely hand President Obama his final legislative accomplishment of his presidency in early 2016.

Appropriations: Defense, Milcon-VA, E&W, Agriculture

One thing is certain, Congress will almost certainly go another year without passing all 12 appropriations bills (last time it was done was 1996). Appropriations is divided up into two camps. Those camps are the individual bills that will almost certainly make it through the process on their own, and those bills that will be lumped into an omnibus package at the end of the year. Of the twelve, Defense, Milcon-VA, Energy & Water, and Agriculture are the most likely to make their way through the process. However, that would be ambitious and we shouldn’t expect such ambition in a presidential year. In all likelihood, only two will actually pass both chambers with an outside chance at passing three.

The reality is appropriations still face a “double reverse filibuster” of sorts. In addition to the filibuster, a sizable portion of the House majority are noncommittal votes when it comes to appropriations. Only 150 of 246 House Republicans supported the omnibus appropriations bill last month. If Speaker Ryan wants to pass appropriations bills, he’ll almost certainly need Democrats. This gives Democrats leverage in negotiating policy riders and funding levels in both chambers, which could further inhibit support among the Speaker’s and Majority Leader’s more conservative wing. In sum, some action on appropriations is likely; just not a lot.


The National Defense Authorization Act will likely continue its march as the lone authorizing bill Congress actually completes. There are rumors that acquisition reform and DoD overhaul could be included in the next NDAA, which would present significant roadblocks. However, it’s pure speculation at this point. The NDAA may also include a major overhaul of Tricare, the health care plan for the military. Regardless, there is enough bipartisan momentum pushing the NDAA onward to continue its reauthorization streak of over half a century.

Possible but not likely:

Tax Reform

While comprehensive tax reform will remain elusive, a more focused international tax reform has a better chance this year than most. December’s tax extenders passage made more than a dozen permanent, while giving longer extensions for others. While the Presidential race will complicate negotiations, repatriation of funds from abroad, possibly for infrastructure, is an ongoing item of interest. Republicans have their eye on lowering the top corporate marginal rate of 35%–a higher marginal rate than other nations, though US effective rates are similar, and often less than other OECD countries.  While the top of the GOP’s wish list also includes a move towards territorial taxation (where profits of US multinationals abroad are not subject to US tax), they are far more likely to get a reduction in the corporate rate in exchange for reduced tax expenditures or better enforcement. That said, this would be a heavy lift in a presidential year that still features upwards of 10 Republican candidates with their own tax plans.

Revising USA Freedom Act

The events in San Bernardino revived debate on domestic government surveillance. In 2015, Congress passed the USA Freedom Act, which curtailed some NSA activities in the wake of Edward Snowden’s revelations. Following San Bernardino, many lawmakers, including the Senate Majority Leader, have suggested they may need to re-examine their revisions. This would be a large undertaking full of political complications that are only amplified in a presidential election year. While it’s possible Congress revisits their new policy, it’s firmly in the unlikely category.

Prison reform, energy legislation, revising Dodd-Frank

Each of these areas enjoy significant bipartisan support. Trimming back Dodd-Frank regulations continues to occur, though much more frequently in the House than in the Senate. Energy legislation continues to affect regions of the country differently, lending itself to unusual political coalitions. And finally, prison reform is an area where Republicans and Democrats appear to agree but fail to come together on the solution. Major overhauls in any of these areas are unlikely. These can be hot-button campaign issues and will lose momentum in Congress. Regardless, minor bills that affect these areas could make their way through Congress, giving the institution some low-profile legislative victories as the presidential campaigns heat up.

Posted in Legislative Politics | Leave a comment

Regular order: Republicans’ risky venture into open debate

Members in the House are calling for regular order. If you have no idea what “regular order” means, don’t worry. You’re not alone. In fact, you’re probably in the company of many members of Congress.

Calls for regular order are almost as old as the institution itself. In theory, regular order is open, deliberative processes that often mimic lawmaking in Congress’s committee era. During that time, committees began the process. Once a bill hit the floor and debate and amendments flowed freely. Any member was typically free to offer an amendment and as a result members contributed policy language on the House floor.

Keep in mind this era died in roughly 1975 and has remained thoroughly dead ever since. Today, fewer than 9 House members (four Republicans, five Democrats) have been around Congress long enough to have actually experienced some version of it while in office. Compared to today’s House, it’s an alien way of doing business. It’s been so long since anything like regular order has been practiced that long-time Congress observer Walter Oleszek wrote in the Congressional Research Service’s centennial study, The Evolving Congress, that it’s essentially become an empty term. Legislative process has been a lot of things in the last 40-years but “regular” is not among them.

Regular order is anachronistic because today’s politics are so different. Attempts to reincorporate “regular” processes in modern day have failed miserably. A big reason has to do with increasing partisan warfare in the chamber. Opening up the process to mimic regular order requires cooperation between the parties. This has not been Congress’s strong suit for some time. Each time the leadership has attempted some version of regular order in the last 30 years, partisans in Congress have used it for political advantage.

The best example came during the Republican Revolution of 1995. Prior to the 1994 election, Democrats increasingly prohibited amendments and shut down debate on the House floor. In the late 1980s and early 1990s, Republicans had become adept at offering poison pill amendments. These amendments divided the Democrats and at times prevented passage of the legislation. Democrats responded by increasingly preventing amendments, closing down the process. In 1995 Republicans returned to the majority on a platform that promised more open debate and amendments in the House.  Democrats responded by offering the same poisonous amendments that Republicans had hurled at them for years. After a few months, Republicans predictably reversed course and shut down the amendment process.

The House has polarized even further in the last 20 years. At this very moment, two poison pill amendments, Syrian refugees and Planned Parenthood, threaten the omnibus spending bill. In fact, seemingly every spending bill the past few years have included riders that threatened a government shutdown. Political and messaging amendments prevail in both the House and the Senate. And as a result, cooperation has plummeted to new lows. Given this climate we have to assume that opening the process will be met with predictable, partisan results.

In this context nobody truly knows, members and observers alike, how the new leadership will assuage members calling for “regular order.” Those calling for a return to regular order – today personified by the House Freedom Caucus – are typically those not in power who want a greater voice.  But in today’s polarized environment the desire to be heard can quickly deteriorate, erupting in more mud-slinging than inclusive deliberation. Some version of regular order would be a more inclusive process. But recently including more members in policymaking has resulted in more politically charged votes. How Speaker Ryan and his leadership team deal with this risk while also allowing for more robust deliberation is unknown. But one thing is certain, the balance between open debate and political risk is a very delicate one.

| 1 Comment

Ohio’s Issue 3 Could Influence Federal Decriminalization Efforts

[An earlier version of this blog post was published by The Washington Post’s Monkey Cage blog.  See here.  Thanks to John Sides for his support.]

Last Tuesday, voters in Ohio rejected a constitutional amendment that would have legalized both recreational and medical marijuana usage. Naturally, folks are wondering about the implications of Issue 3’s failure for similar measures in other states.

But there’s a potentially larger question. Could the defeat of Ohio’s Issue 3 influence marijuana policy at the federal level—for instance, efforts to eliminate federal felony charges for non-medical marijuana, or to influence related questions like DEA funding, drug sentencing guidelines, and so on.

It certainly could.

In 2011, Dan SmithJosh Huder, and I published an article  in American Politics Research that explored whether the passage or failure of statewide ballot measures affects how members of Congress vote. In other words, do state ballot measures influence policy outcomes at the federal level?

Our theory was that that when a statewide ballot measure either passes or fails, it tells lawmakers precisely what their constituents want, down to the district level. That helps reduce “policy shirking,” in which lawmakers vote contrary to the wishes of their constituents.

Lawmakers care about one thing above all else: winning reelection. Reelection-minded lawmakers try their best to behave as pure delegates–and to faithfully represent their constituents’ views.

But a number of studies have shown that lawmakers can be poor judges of their constituents’ preferences on specific issues. While they may know how liberal or conservative their district is, lawmakers have incomplete knowledge of voters’ beliefs on more nuanced issues, like marijuana legalization.

And so lawmakers search for reliable information about what their constituents want. Public opinion polls are an obvious resource. But polls are not always accurate. For example, a handful of polls–including one at my alma matter, Bowling Green State University–suggested that Ohio’s decriminalization measure would pass. In fact, a Kent State University poll suggested Issue 3 would pass with a whopping 58 percent support. What’s more, people who respond to polls are not always the same people who turn out to vote, which is one reason polls are inaccurate. But for obvious reasons, lawmakers are especially interested in voters.

And so we theorized that a ballot measure’s results offer the best possible information lawmakers can use to figure out what their constituents want. It offers information that’s specifically about a single issue, as opposed to a lawmaker’s more general intuition about how liberal or conservative constituents are. And it reveals what voters want, not just what poll respondents want.

[A number of readers of the Washington Post article have pointed out that some voters opposed Issue 3 because of concerns about monopolies.  I think it’s a fair point, but here’s my response.  First, I still think Ohio’s congressional delegation will look at Issue 3’s defeat as a piece of information when trying to decide what their constituents want.  I maintain that the vast majority of the votes for or against Issue 3 were ultimately about marijuana legalization.  Indeed, it seems unlikely to me that 36% of Ohioans voted IN FAVOR of monopolies.  Second, if you’re a lawmaker trying to decide where your constituents stand, it’s telling that pro-legalization voters were persuaded to vote against that issue because of something unrelated to legalization.  I think that gives lawmakers some good information how much pro-legalization constitutions care about the issue.]

Here’s how we tested our hypothesis. We took the results of statewide ballot measures on three issues–regulating campaign finance, raising the minimum wage, and banning same­-sex marriage– and compared that to how lawmakers voted on those issues.

Our hypothesis appears to be partly correct. Members of the House do vote in ways that appear to be influenced by how their constituents voted on ballot measures.

But that’s not true for Senators—which makes sense. Senators are more insulated from public opinion than representatives, since they generally have a larger constituency and only have to go up for election every six years.

We also report in this paper that representatives are most likely to respond to what they learn about their median constituent–the constituent a lawmaker needs to win over in order to get at least half the vote.

In other words, the statistical results show that it matters little if the ballot measure passed with 70 percent of the vote or 51 percent of the vote. Both show a member of Congress where at least half of his or her constituency stands.

So how could the defeat of Ohio’s Issue 3 hurt federal efforts to decriminalize non-medical marijuana use and so forth? First, it could influence Ohio’s 16 representatives. Second, it could affect how lawmakers in similar Midwestern states vote–although that’s a theory we do not test that in the paper.

Ohio’s decision on marijuana, in other words, has national implications.


When Does Congress Repeal Legislation? A New Dataset of Major Repeals from 1877-2012 Provides Answers

Our understanding of Congress—and how we evaluate the institution—is shaped by the laws it enacts.  Yet Congress often performs the opposite of law creation: repealing landmark laws.  But despite the regularity and importance of repeals, we know very little about when and why repeals happen.  One reason for the lack of research on repeals is methodological.  Unlike lists of bills voted on or laws enacted, no government publication catalogues these important outcomes.

In new research, Nate Birkhead and I describe a new dataset of major congressional repeals enacted from 1877-2012.  For this project we catalogued repeals by culling historical newspapers, historical reference volumes, and period-specific texts.

Needless to say, the importance of this research was elevated in 2011 when Republicans took control of the House and promised to repeal the Affordable Care Act (a.k.a. “Obamacare”).  Like the effort to repeal the Affordable Care Act, the repeals in our dataset represent some of the most contentious, salient, and long-running disputes over national policy.  Examples include the repeal of multiple New Deal statutes following the 1994 Republican Revolution, dramatic statutory changes in monetary policy in the 1890s, the repeal numerous tax statutes in the 1920s, and the repeal of the Chinese Exclusion Acts in the 1940s.

Beside the fact that they are understudied, repeals provide congressional researchers a unique perspective on Congress.  Analytically, repeals allow us to compare lawmaking in two time periods: the enacting Congress and the repealing Congress.  And theoretically, one of our central claims is that the causes of repeal differ from those which explain law creation.  In particular, we believe that although shifts in Congress’s membership—known as changing “pivot points” in the literature—are a key cause of law creation, we believe the ebb and flow of party strength is critical to explaining law reversal.

Figure 1 (below) presents our data.  We can see that repeals do not occur uniformly over time: there are clear “spikes” in repealing activity.  In brief, the spikes are generally consistent with our theoretical claims.  We see increases in repealing activity when (1) the majority party is ideologically unified and (2) the majority came into power after a long stint in the minority.


But because there are other factors that can cause repeal, we developed a statistical model that predicts when repeals happen. Our specific model is known as a “survival analysis.”  Survival analysis is frequently used in epidemiology to understand how long patients survive some illness and the effects of surgical procedures and drugs.  In the context of our study, the model predicts how long laws “survive” before being repealed.

Our model shows that some of the “usual suspects” explain when and why repeals happen.  When the enacting coalition is voted out of office over time, and the distribution of preferences shift to the left or right, repeals are more likely to occur.  We also find that the larger the distance between the House and Senate, the less likely repeals are to happen.  Both findings fit well-established conclusions in the congressional literature regarding the creation of laws.  We also find that repeals are more likely in two policy domains.  First, tax laws are among the most likely to be repealed.  And second, laws created for war purposes are often repealed soon after the war ends.

Although these findings help us understand when repeals are most likely, we find that the partisan factors in our model have the greatest overall effects on the probability of repeal.  Like the descriptive results in Figure 1, the model reveals that repeals are most likely when the majority party is ideologically unified and the majority came into power after a long stint in the minority.  We also find that these two effects are not just additive, but instead have an interactive effect.  Figure 2 presents this interaction effect.  On the x-axis is the measure of how long the majority was in the minority prior to winning control of the House and Senate (coded as the number of chambers in the minority over the past 10 years).  The y-axis represents a higher or lower likelihood of repeal.  And the dots represent the estimated effect of the majority party’s ideological cohesion.


Figure 2 (above) shows that when the majority party was in the minority for an extended time, and members of the majority are ideologically unified, there is an increased likelihood of repeal beyond what is predicted by these factors individually.  As an example, when Republicans regained control of the House and Senate in 1995 after being the so-called “permanent minority,” a standard deviation increase in their ideological cohesion is predicted by the model to have increased the probability of repeal by 208 percent.  And indeed, the Republican controlled congresses from 1995-2000 enacted a number of major repeals (most notably, the repeal of Glass-Steagall).  At the other extreme, Figure 2 shows that when the majority has been in complete control of both chambers over the past decade (coded as zero chambers in the figure), there is no effect of greater ideological cohesion.

While there are various ways in which statutes can be “undone” (invalidation by the Supreme Court, defunding, sunsets, etc.), Congress regularly voids its own statutes via repeals.  Given our findings about the importance of political parties to when and why repeals happen, in particular the ebb and flow of party strength, we characterize repeals as long-term contests between two great “teams” over national policy.  Simply put, when Congress enacts a new law, it is hardly the end of the game.  Anyone observing Republican attempts to repeal the Affordable Care Act will recognize this feature.  However, our results suggest that these kinds of partisan contests are an enduring feature of our politics.

A future post will discuss the implications of our research for Republican attempts to repeal the Affordable Care Act (a.k.a. “Obamacare”).


Another version of this post was publisher on the London School of Economics USAPP blog.  See here.


| Leave a comment