Bergdahl, Benghazi, and Beyond: The Politics of Congressional Investigations

Iran-Contra-HearingsIs Bowe Bergdahl the new Benghazi?

It would certainly seem so.  Several Republicans are calling for investigations into the now infamous prisoner swap.  Calls for impeachment exist.  And, of course, Hillary Clinton’s name has been linked to the Bergdahl scandal.

In an even broader sense, we might link the Bergdahl scandal with the IRS targeting investigation, which is still simmering, and the VA scandal, where Congress is almost certain to investigate further.

But which investigation is a “warranted” examination of legislative-executive authority, which is motivated by legislators’ “problem-solving” motivations, and which investigation is simply “political theatre?”

I think most would agree that continued investigations into the VA are warranted.  Even the IRS targeting scandal deserved at least a cursory look in my mind.  Are the Bergdahl and Benghazi investigations warranted?  It depends on your political leanings; scandals are highly subjective creatures.

Does this mean we can’t systematically examine the politics of congressional investigations?  Should we be surprised at the volume and intensity of congressional investigation in the modern Congress?

In both instances, the answer is: No.

In an excellent article appropriately titled “Divided We Quarrel,” political scientists David C.W. Parker and Matthew Dull examined the dynamics of divided government, congressional polarization, and congressional investigations of the executive branch.  Parker and Dull collected an impressive dataset of congressional investigations from 1947 to 2004 including their frequency, duration (measured in days), and intensity (measured as the number of published pages).

Here’s what Parker and Dull found:

  • Divided government is indeed associated with more frequent and longer investigations of the executive branch.
  • Legislative polarization is also associated with more intense investigations.
  • However… divided government and polarization increase the frequency and duration of congressional investigations in the House but not the Senate.
  • And finally, when the presidential is popular, Congress initiates fewer and less intensive investigations of the executive branch.

I don’t think Parker and Dull’s results could fit the current Congress’s political environment better.  Indeed, the president is unpopular, Congress is polarized by any measure, we have a situation where both parties are sharing power, and investigations in the House far outpace investigations in the Senate.

So, why is this relevant?  For starters, while we can’t point to a specific investigation or scandal and say it’s “obviously political,” we do know in a general sense that investigations are political endeavors. From a theoretical perspective, this research highlights the critically important role that parties play in the U.S. Congress (for better or worse).  While it may seem surprising to non-political scientists, for decades the prevailing wisdom was that parties didn’t matter all that much.  (In fact, the title of Parker and Dull’s article is a response to David Mayhew’s famous “Divided We Govern.”)  Lastly, while some of the dynamics uncovered are indeed intuitive, others (like presidential approval and the bicameral differences) are not.  In this sense, Parker and Dull’s work helps situate contemporary congressional investigations in a broader institutional framework (including the transition from the era of strong committees to the current era of strong parties) and highlight key determinants of congressional behavior.

Posted in Congressional Absurdity, Legislative Politics, Polarization, Political Parties | Leave a comment

Could Boehner be the first Speaker to Win Seats and Lose Job?

The Fix recently wrote about how “A 2015 rebellion against John Boehner would be unprecedented.” In the piece Philip Bump argues that “no speaker has overseen a pick-up of House seats and subsequently lost his job.”

Setting aside problems in closely connecting congressional elections and the speakership election across this period,* this statement really hangs on the definition of “losing the job” in this context. Bump is essentially arguing that aside from losing their majority, Gingrich is the only sitting Speaker to be pushed out of office. That is not entirely true. In fact, most retiring speakers face revolts of varying degrees and momentum at the time of their retirement. No sitting speaker was literally forced out through a vote. But as I argued before, most retire before a revolt comes to fruition.

For example, Speaker Reed stepped down in 1899 amid multiple rumors that many in his caucus were plotting against him. Henderson faced overthrow rumors throughout his tumultuous four year term. McCormack faced serious challenges to his speakership from young liberal Democrats. He and O’Neill actually faced challengers for the top spot in their final term. Though they won those battles it hastened McCormack’s retirement the following year and likely reaffirmed O’Neill’s decision, since he already announced he would retire in 1984 after one final two-year term. And to a lesser extent, many Democrats were frustrated with Albert’s anachronistic post-war (as in WWII) leadership style, though he did not endure overthrow rumors to the same degree.

What set Gingrich apart from his predecessors was timing. He faced the same overthrow rumors leading up to the 1998 Election. The difference between Gingrich and the rest is he chose to wait until after the election to announce his retirement. In sum, “losing the job” through early retirement is hardly unprecedented. In fact, it is more the rule than the exception.

The second part of Bump’s argument states Boehner would be the only speaker to lose his job after his party picked up seats. This isn’t entirely accurate either. The chart below lists House Speakers since 1900 who did not die in office, did not seek higher office, did not lose their majority, and did not resign due to scandal. This narrows the field to only those speakers who had the option to stay and run for speaker again. In the 20th century, this reduces the list to five speakers.

Speakers who did not die in office, seek higher office, or resign due to scandal
Announced Retirement House Seat Change in final Election
Henderson September 1902 +7
McCormack June 1970 +13
Albert June 1976 +0
O’Neill Oct/Nov 1984 – Retired 1987 +4
Gingrich November 1998 -3

Since 1900, three out of the five speakers who had the option to run for another term gained seats in the election following their retirement. In Albert’s last term as Speaker, the parties drew a dead heat. Again, Gingrich is the exception. His party lost three seats and he waited to retire until after the election. For the rest of the Speakers, their party either drew even or won seats. However, their electoral fortunes did not stop them from preemptively stymieing revolts.

In short, both the nature in which Boehner may lose his job and his party’s electoral fortunes before his retirement have plenty of historical precedence. That said if Boehner wants to make his mark on history (some sarcasm), he could wait until after the election to retire. Given the possible change in power in the Senate, it is more likely than not that he waits until after the Election to decide whether to run for another term in the top spot. Regardless of his decision, this is not uncharted territory.

*There are several issues not addressed in this hypothetical: since 1900 speakership elections were internal battles with little national electoral consequence. For most of the 20th century congressional elections were local, not national, affairs. For most of the 20th century individual candidates were more important than national party platforms in congressional elections. The power of the speaker to dictate policy fluctuated enormously in the last 114 years. Therefore, speakers’ ability to pass policies – or take/avoid blame for failed platforms – has also fluctuated. Parties’ internal makeup/coalitions have changed dramatically. In short, congressional elections are just one factor among several that affect choosing a House leader. That said, the increasing national character of congressional elections is more prevalent today than in previous decades.

Posted in Elections, Political Institutions | Leave a comment

Assessing Congressional Productivity: Getting it Right

The Washington Post’s Wonkblog had a write up on congressional productivity not long ago. Its central thesis is Congress is more productive in election years than in non-election years. This is a good counter-intuitive point. Many pundits discuss congressional productivity only in terms of major bills passed. However, Congress often passes laws that fly under the radar.

There are a lot of reasons why Congress is more productive in election years than in off-years.  Each new Congress must start over from scratch. Therefore, bills that failed in the previous Congress must be re-introduced, committees must re-hold hearings, and the process of reconciling differences between the two chambers must also begin anew. Basically, the first year of a Congress is filled with preliminary steps in the legislative process. In the second year of a Congress, many bills require fewer steps to become law. Also in election years, many bills are passed for campaign fodder. The Senate’s attempts at equal-wage and minimum wage legislation the past few weeks are perfect examples of this strategy.

However, while Wonkblog’s central argument – that the House and Senate pass more bills in election years – is correct, there are a bevy of problems with the article.

First, it uses the wrong metrics. It employs the number of public and private bills “enacted into law” for each chamber. This is a pretty big problem since a single chamber cannot pass a law. So what does this line in the Resume of Congressional Activity actually measure? This line reports the number of laws that originate in the House (bills beginning with H.R.) or Senate (S.) that eventually became law. So really this piece is not capturing productivity in either chamber. Instead, it is measuring how many bills originally introduced in each chamber actually became law. The measure used doesn’t capture productivity–it captures bill origination.

This is hugely problematic for a few reasons. Bills that originate in one chamber are many times merely vehicles for the other chamber’s bill. For example, according to the House’s reading of the constitution, all revenueand spending bills must originate in the lower chamber. However, there are several revenue and spending measures the Senate would like to consider and pass. So rather than wait for the House to initiate the spending bill the Senate would like to pass, it will substitute its text on an old House bill and claim that the constitutional requirement has been met. The same is true vice-versa. A situation like this happened this past February when a Senate bill originally naming a Nashua, NH air traffic control tower after Patricia Clark turned into a bill raising the United States debt ceiling. The House changed the title of the bill to reflect the change.

Chamber productivity should be analyzed by the number of bills passed in the chamber per session of Congress. Or, it should capture the total number of laws. (See below)

bills and laws

As these numbers show, the trend described in the piece holds but not to the same degree. Bills passed in each chamber increases during election years, but only modestly. The number of laws passed, however, is nearly double in election years rather than in non-election years, suggesting that one chamber is likely finishing the work of the other in the second year of the Congress. Regardless, stating that “the volume of bills passed during election years nearly doubles, in both House and Senate” falls well short of reality.

These numbers also fail to distinguish between significant and non-significant bills, which is a big distinction. When discussing congressional productivity abstractly most people are not referring to inconsequential bills such as renaming Post Offices. Yet, to a large extent these numbers capture exactly that. In any given year, most bills are passed with little to no fanfare. The overwhelming majority of bills (over 88% from 1891 to 1994 according to Clinton and Lapinski (2008)) do not even receive a roll call vote in either chamber. In any given Congress the overwhelming majority of bills are passed under super-majority procedures such as unanimous consent or suspension of the rules (which requires 2/3rds majority to pass the House). Just because there are more bills “passed” does not mean Congress has been productive on any substantial front. In this Congress, the notion that Congress will get nothing done of much substance in this election year holds true.

Another problem is that the article includes the passage of private laws. Private laws are fundamentally different from public laws (note: the above chart graphs measured passed in each chamber minus private laws). Public laws apply to all citizens, states, etc. Private laws are bills that directly affect one or a few private citizens. These bills often deal with immigration cases or claims against government. In the more distant past private bills were the means used to dole out government benefits (Jenson 2003). Private laws frequently bestowed disability benefits to wounded veterans and their widows as far back as the Revolutionary War.

private laws

Over the past several decades the practice of passing private bills has declined dramatically. Some of the functions private bills served have been automated through programmatic legislation. Other research suggests new avenues of constituent service emerged, making private bills an antiquated means for representatives to address constituents’ problems. Regardless, conflating public and private bills can overestimate the “productivity” in many of the mid-20th century Congresses.  However, the general trend that Congress is passing fewer laws still holds.

The article was an interesting and counterintuitive look into congressional productivity. However, this is one instance in which the thesis was not resting on the right evidence.

Piece was re-posted from

Posted in Legislative Politics, Political Institutions | Leave a comment

Nickelback, Herpes, and Obama’s Vote Share in 2012

Spurious relationships are a serious problem in social scientific research.  But… they’re also fun!  For example, did you know that ice cream causes crime?  Also, global temperature is inversely related to the number of pirates.

Seth Masket provides an excellent example of a spurious relationship asking: Are Democrats perverts?   As he shows, it turns out that there is a strong relationship between porn use and Obama’s vote share in 2012!  But as Seth aptly explains:

For one thing, there’s a big potential ecological inference problem here. We’re making assumptions about individual level behavior by examining data aggregated at the state level… Second, chances are that even if there is an individual relationship here, it’s not a direct one. Porn usage may correlate with something else that also correlates with partisan voting patterns. 

Seth notes that porn pageviews explain 16% of the variation in Obama’s vote share.  In an attempt to outdo him in the “spurious-relationship-a-thon,” I looked at the predictive power of Google search traffic at the state-level for: (1) the rock band “Nickleback” and (2) “cure for herpes.”  Here are the results.*

HerpesNickelbackAccording to the first scatterplot, “Cure for herpes” has a negative relationship with Obama’s vote share in 2012, supposedly indicating that herpes caused people to vote for Mitt Romney in 2012.  However, the R2 is just 0.08, which indicates that herpes has little predictive power (in fact, the relationship is not significant (p=.12; n=31)). Nickelback also has a negative effect, supposedly indicating that fans of the Canadian rock band were more likely (!) to vote for Romney in 2012.  Notably, in the second scatterplot, the R2 is 25%, indicating the fully a quarter of the variation in the 2012 election outcome is explained by the state-level variation in Nickelback fans.

While it’s difficult to take the above “findings” seriously, social scientists dedicate their lives to distinguishing between correlation and causation.  In practice, it’s much harder than most people appreciate.  What the world gives us is correlation, so it’s easy to draw faulty conclusions from observational data.  In an undergraduate research methods class of mine, we cover countless examples of correlation and causation.  For example, congressional candidates who spend large amounts of their personal wealth on their campaign actually receive fewer votes.  Also, humanitarian aid is associated with various negative consequences such as higher rates of infant mortality.  Are these causal relationships?  Of course not. Always remember: correlation is not causation.

* Don’t tell my department chair I spent 30 minutes working on this…

| 1 Comment

Senatorial Courtesy, Blue Slips Caught in the Fallout

Ian Millhiser has a very good piece on judicial nominations and blue slips over at Think Progress. It covers a lot of ground and is a wonderful read.

However, I do have some bones to pick with his take. At the core of Millhiser’s argument are blue slips and their place in Senate history. He contends that blue slips, the process by which home-state senators grant approval to president’s judicial nominees, has little rooting in Senate process and history. There are a couple problems with this.

First, blue slips are the Judiciary Committee’s extension of senatorial courtesy. While the formal practice of blue slips, with letters from the Committee to the nominee’s home state senators, extends back to at least 1917, senatorial courtesy extends to the 1st Congress (1789-1791). There is no formal rule outlining the practice but it stems from a very rich history of “advice and consent.” In this way, blue slips are not unlike many Senate practices. The Senate is heavily guided by customs and precedents. Blue slips are not an aberration to this principle. Rather, they are a modern example of that history.

Second, it is improper to compare today’s blue slip process with those in prior congresses. Given my training is in congressional development and history, this is not an argument I normally make. However, the advent of the post-nuclear Senate has fundamentally changed the importance of the blue slip process.

For most of Senate history, the blue slip process has not been absolute. Failure to approve of a nominee did not prevent floor consideration, though it often meant the demise of the nominee. Today, floor consideration for judicial nominees is fundamentally different. The minority party no longer has the ability to filibuster judicial nominees (except SCOTUS nominees). Because of this, the majority can run roughshod over minority opposition, effectively ignoring their concerns as they approve judicial nominees by majority vote.

This has shifted minority opposition from the floor to the committee room. With no other place to stop or slow the nomination process, it is not surprising that blue slips are now in the crosshairs.

Patrick Leahy (D-VT), chairman of the Senate Judiciary Committee, is the first Senator since James Eastland (D-MS) to use blue-slips as an effective veto on a nominee. As Forrest Maltzman and Sarah Binder point out, blue slips were never intended to be a veto. However, Leahy’s position as arbiter of the blue slip is much more precarious post-nuclear option. And in all likelihood it is untenable. Arguments like Millhiser’s illustrate Leahy’s predicament. Either Leahy adheres to his current blue slip practice and hopes Republicans honor it when they retake the majority (either in 2015 or later), or he can ignore it and ensure its demise under Republican majorities.

The nuclear-option has stressed the blue slip process. With no other avenues to object, the minority has turned to its few remaining tools. Unfortunately, the nuclear-option’s collateral damage includes the minority’s role in “advice and consent” on nominations, and now appears to threaten advisory practices that stem from senatorial courtesy.

In a polarized atmosphere it is easy to become frustrated with bipartisan practices, precedents, and customs. However, eroding these customs compromises the Senate’s function as the check on overambitious majorities.

Posted in Legislative Procedure, Senate | Leave a comment

Liberal Bias in the Classroom: College of Charleston Edition

imagesYou’ve heard about it.  College professors, who are disproportionately liberal, skew the views of their impressionable young students by assigning leftist books and poking fun at conservatives in the classroom.  Rick Santorum famously called colleges “indoctrination mills” and anecdotes about liberal bias abound.  But is there systematic evidence of this?

I find it difficult to get worked up about the “liberal bias in academia” mantra because of it’s prevalence.  However, recent events at my institution, the College of Charleston, merit some attention.

At issue is the book “Fun Home” by Alison Bechdel.  While the book explores various “coming of age” issues, the central themes are sexual orientation, gender identity, and Bechdel’s complex relationship with her closeted gay father.  Also, the book is written in the style of a graphic novel, which, you know, nakedness!

Anyway, because “Fun Home” was supposedly purchased with “tax payer money” (it wasn’t) and was “required” reading (it wasn’t), the South Carolina legislature voted to strip $52,000 out of the College’s budget (the overall cost of the reading program).  Rep. Gary Smith, one of the originators of the proposal, was quoted thusly:

I understand diversity and academic freedom. This is purely promotion of a lifestyle with no academic debate.

Colleagues at the College of Charleston have addressed various dimensions of this issue (see excellent articles by John Warner, Alison Piepmeier, Tim Carens, and Chris Korey).

I felt it important to address the bigger issue: Can college professors sway the views of their impressionable students?  Indeed, this is the underlying issue.  It is implicit in Rep. Gary Smith’s claim that the book “promotes” a lifestyle.  Its also a social scientific question that can easily be tested.

So… is there any evidence that college processors turn their students into liberals?

The answer: No.

In an innovative paper, political scientists Matthew Woessner and April Kelly-Woessner sent surveys to 200 political science professors and their students, asking questions about party identification, ideology and the appropriateness of discussing personal politics in class.  Notably, Woessner and Kelly-Woessner conducted the survey twice: once at the beginning of the semester and once at the end.  Armed with this data, they were able to assess if students became more liberal or conservative over the semester and whether changes in the students’ ideology correlate with the professor’s views.

Here’s what they found:

  • Despite instructors’ best efforts, most students are able to accurately guess their professor’s party identification (though most though their instructors were more moderate on average).
  • According to the result of the survey, 28% of students reported a change in their partisanship during the semester.  Notably, much of this shift is toward the Democratic side of the aisle.
  • But most importantly, this movement in students’ views does not correlate with either (1) the students’ assessment of the professor’s beliefs or (2) the instructor’s actual political beliefs.

As Woessner and Kelly-Woessner explain:

While student views do shift over the course of a semester, they tend to move somewhat randomly, usually regressing toward the mean.  Thus…the changes are hardly what one might expect if faculty members were systematically indoctrinating their students.

What’s often lost in the debate about bias in college classrooms is the larger question about where our political attitudes come from in the first place.  According to published studies, children experience the greatest growth in their political learning during adolescence (not college) and by about 9th grade children have an ability to think in ideological terms (Erikson and Tedin 2011, pg. 128).  In terms of party identification, the biggest influence on preadult socialization is not their school or even friends (which have almost no effect), but that of the family.  According to one estimate, 76% of children have the same partisan attachment as their parents (Erikson and Tedin 2011, pg. 132).

Simply put, in addition to the lack of systematic evidence that professors “indoctrinate” their students, the assumption that college students are “impressionable” is not supported either.  In fact, overly political college professors are likely to generate one kind of outcome more than any other: negative evaluations!  Seriously.  In yet another study, Kelly-Woessner and Woessner show the political bias in the classroom yields negative student evaluations and a general dislike of the subject matter.

But how are we to make sense of the fact that students’ political attitudes do move in a modestly Democratic direction during college?  Well, it’s important to be clear what we’re talking about.  Studies have shown that education is strongly correlated with greater tolerance, which in turn has a liberalizing effect on social issues (Erikson and Tedin 2011, pg. 139-142).  It’s also true that college students are exposed to a large volume of media and interact with a more diverse segment of the population and are therefore at the forefront of changing social norms.  But on economic issues, there is no aggregate change over the course of a college career.  In fact, the better educated are typically more conservative on economic issues (Erikson and Tedin 2011, pg. 139-42).

In sum, college certainly affects the views of college students, and one of those is a liberalizing effect on social issues (but not on economic issues).  But the evidence does not support the view that it’s “indoctrination” or the “promotion of a particular lifestyle” by individual faculty.  Rather, the move to the left on social issues stems broader educative effects, an exposure to more diverse groups of individuals, and greater media consumption by young adults.  In all seriousness, the “solution” would be to simply avoid college altogether (not avoid specific textbooks or professors)

Finally, I’ve had this conversation before.  One response is: “So two college professors did a study of college professors and found no evidence of liberal bias?  I’m skeptical of this study’s objectivity.”  Well, for starters, the paper was peer reviewed in a leading political science journal, which should stamp out faulty research methods.  But second and perhaps more importantly, Matthew Woessner is a conservative himself!  I’d also direct skeptics to Mike Munger’s excellent book “Scaling the Ivory Tower” which was published by the conservative Institute for Humane Studies.  In this book, Munger (a noted conservative) explains that while bias in academia certainly exists to an extent, it’s much smaller than conservatives realize.


Robert Erikson and Kent L. Tedin “American Public Opinion” (2011, eighth edition).

| 1 Comment

A Caveat on Congressional Productivity

On Thursday, Chris Cillizza examined an Obama statement in Texas: “This has become the least productive Congress in modern history, recent memory. And that’s by objective measures, just basic activity.” Cillizza agrees and extrapolates this a little too far, saying this Congress is the least productive in history. By the numbers many would argue this isn’t exactly accurate, but I’d argue he’s ultimately right.

First, if we take Obama’s statement at face value, he is right. This Congress is on pace to be the least productive in modern history (modern, in this instance, referring to the post-WWII period). However, it is decidedly not the least productive in history. The current Congress has thus far passed 98 laws. This is awful but it is not the fewest in American history. The 113th Congress, though it still has a couple months to legislate, is already more productive than several congresses prior to the Civil War. So strictly speaking this Congress will not be the least productive in history.

If you think that is a low-bar to evaluate policy productivity, you’re right. To fully examine policy productivity, we need to account for the need for policy. In short, what should Congress be doing. Reauthorizing existing policies and programs, addressing pressing social problems, or fixing broken policies are at the core of congressional responsibility. Sarah Binder examined policy demand in Stalemate to examine gridlock since 1947. She found that even when controlling for policy demand gridlock has still increased over the post-war period. This reaffirms what we already knew but doesn’t evaluate policy demand in the antebellum period (mostly due to data constraints), the last time we saw productivity levels this low.

We already know there were several antebellum congresses that passed fewer laws, so the 113this not the least productive in history. However, when put in the context of policy demand, it very well may be. The Federal Government today has more responsibilities than it did 150 years ago. The Industrial Revolution, and the economic and social problems that accompanied it brought enormous challenges to society and Congress, spurring a slew of labor legislation, economic regulations, and other policies addressing those problems. Modern shipping practices have facilitated interstate commerce. This has boosted economic potential but also made monitoring corporate financial and economic activity and controlling things like food borne diseases more difficult. Technological advances have brought the privacy versus security debate to the forefront of government functions as monitoring private conversations and access to private financial information has become easier.

Social and technological advances have created new problems for governments to address. As a result Congress has more to do. One could argue the best government governs the least, but in many ways that shirks social responsibilities already under the federal government’s purview. Legislators should oversee and direct existing programs toward the proper ends. Unfortunately, unproductive congresses often fail in this capacity. This Congress’s basic budget and spending squabbles have overshadowed this fact. According to CQ, almost a third of discretionary government spending in 2014 funded programs that have not been reauthorized by Congress. Government programs are potentially wasting money on outdated projects because Congress has failed to perform its legislative function.

Pending lawmakers do not want to return to a period where employees were required to work 60 and 80 hour weeks for extremely low pay, food quality standards were lax, basic infrastructure needs were not met, or wiretapping and information hacking goes unchecked, these problems demand attention. However, given the current policy environment these vital functions are largely ignored.

In short, the 113th Congress is not the least productive in history. But relative to policy demand it very well may be.

Posted in American Political Development, Legislative Politics, Policy Agendas, Political Institutions | 1 Comment