Episode 5: Cue-Taking and the Early Legislative Process with Alison Craig

Hello and welcome to Episode 5 of the Policy Agendas podcast. My name is E.J. Fagan. Today I’m

joined by my wonderful co-host, Christine Berg. Hey. And Zach McGee, hello. And our

guest, our guest is Allison Craig, assistant professor of government here at the University

of Texas at Austin and author of the new article just out in the American journal Political Scientist

Science with to call with her coauthors, Janet Box, Stefan Stefan’s Maya. I

apologize if I’m pronouncing correctly, Andino Kristiansen. Kristiansen. The article is called q- Taking

in Congress Interest Groups Signals from Dear Colleague Letters. Allison, welcome to the podcast. Thanks

for having me. So can you just tell me a little bit about this article queue taking in Congress, interest groups signals

from Dear Colleague letters. What’s the argument you guys make? So our argument is that

members of Congress are dealing with fairly limited resources,

both in terms of time and their ability to just process the sheer amount of information

that they’re dealing that they get on a daily basis. And so one of the ways

that they deal with that kind of information overload. There are a number of ways that are discussed

in the Congress that are true. But one of them is specifically by taking cues from them, from other

actors, whether they are from other members of Congress or interest groups,

which is what we focus on specifically in this paper. And we argue that cues

from interest groups provide a strong signal to members of Congress when they’re making decisions about legislation.

And specifically, we argue that in the earliest stages of the legislative process,

it’s actually cues from groups that have a high degree of what we call social

power that have the most influence rather than

the sheer number of groups. All right. So why is this important? So why why

is it. Is it useful that is useful to know that these cues are being taken by members of Congress? Sure.

So, I mean, I think, you know, in political science, we spend a lot of time trying to identify the

ways that interest groups and affect the legislative process. We have this sort of prior

expectation that the interest groups are actually very powerful and they’re very influential in the research on exactly

where their influences is a little bit mixed. And so I think one of the

important contributions of this paper is we are showing a very clear way that interest groups are influencing

the legislative process. But we’re also showing kind of the specifics about how

those groups are influencing the legislative process. So if you think about our interest group, social

power measure, what we’re saying with that is that groups that are more well-connected within the interest

group community are more effective. Their endorsements for legislation

carry more weight. And so that’s actually I mean, we want to think

about like prescriptive implications. That’s actually something that interest groups can think about when trying

to figure out how to be effective. So the primary means

that you’re measuring this influences through dear colleague letters. And for somebody might not be familiar, we can explain

what are these dear colleague letters where the data look like and all. Sure. So, dear colleague letters are

a letter sent by members of Congress within the House. Every member Congress has the ability

to send these letters. Almost all of them do. They are letters that are typically used to

introduce legislation. About half of them are used to request co-sponsors for

a bill. The remainder are used to solicit support for legislation

or verbs or excuse me for policy letters or

events, invitation events. We focus very specifically on the letters that pertain to legislation

because we’re interested in the endorsements to bills. And so we

start with a little about about 100000 letters that Salala is

a lot of letters that consumed much of my career as a graduate student was coding

and dealing with 100000 dear colleague letters. Oftentimes,

what do they look like while you’re coding them? What’s the data format? So oh, that’s a funny story. So

you see the data format, the original data format is that the letters?

Because I actually archived them when I was still working on the Hill. The letters were saved in a Microsoft Office PSC file.

That is not a file format that is remotely useful to anyone in any way, shape or form.

So the first day they’re all in e-mail form and that’s why this

outlook, BSD. It’s basically a giant outlook mailbox. And so I had to take the individual

e-mails and shift them over into the

M box format because there is a Python script that can actually handle inbox emails.

And so I ran my script through all of these individual

M box files, extracted unnecessary information and ended up with a massive spreadsheet

of data. So they’re not using latex over there? No, they are not using

high tech uncap, though they’re still using WordPerfect and some.

So I see you CVT Carlat, dear colleague, letters. And in some sense you have a very kind of clean asker. I hear you

say they they have their requestion co-sponsors. Most of them are or at least say they’re involving legislation

and you can evaluate that legislation. How many co-sponsors it gets. So I guess the really complicated part is that the

independent variable interest group Social Power and how is how is that measured? So the interest groups

social power actually draws on data that my coauthors, John Bux, Efron’s

Meyer and Dana Kristiansen have been working on for a while and put together. And it’s based

on interest groups signing amicus curiae briefs together. So they’ve actually identified

the interest groups who sign these amicus curiae briefs before the Supreme Court

together and then are able to create a network out of that.

Those signatures where two interest groups that sign that amicus curiae brief

together are connected within this larger interest group network. So if you envision

the network is going to the larger interest group community, two groups that’s signing

letters together are connected. And then we measure it with something called the Eigenvectors

and Tragedy’s Scores, which is a really fancy way of saying

that we are accounting for not only the number of

connections that an individual group has. So if I were to sit here and say, all right,

well, I’m connected to the three of you because your graduate students are U.T. and we’re in the Policy Agendas Projects Group together.

And that would be a very simple way of measuring

interest group or connections within a network. What eigenvectors and truly does it allows me to actually

then wait how well connected you are. So my score goes up if I am connected

to other well-connected groups. And so that’s why we argue that it’s a measure of interest

group social power. Well, let’s stop and ask more specific questions about that now. So I think I think

there’s one. So when we’re talking about amicus curiae briefs, you state in the paper that

the three of you think that this is a similar measure to what we would see in these interest

groups in Congress. So but the groups that would be signing an amicus curiae

brief are going to be in the certain group that would be bringing cases before the Supreme Court.

And then you have people like the soybean interest group and

they’re probably not going to be filing an amicus curiae brief. So how do you pull them in to

this larger network that you’re talking about? So they actually do cite amaz amicus curiae briefs. That’s actually

the graph that we have in the paper that shows the network of the American Soybean Association is

actually their connections through signing these amicus curiae briefs. So

we have all of these groups and a lot of them are. That was one of my kind of early questions. I’m thinking about

this project is I think that the groups that are going to sign these briefs are going to be very like justice oriented.

And those certainly are in the groups that we air in the list, like the ACLU is prominent in

the network. But, you know, so is the National Education Association

and the National Organization of Women. OK. So so it’s y’all’s position that there is

a lot of overlap. And you’re comfortable with that? That’s that’s great, because that means a lot less data

collection for somebody else to do. Yes. You heard it here. Advocacy

is just as important as I’ve heard. I mean, the main part of our argument is that the relationships

that are formed between groups signing amicus curiae briefs together persist. The idea is

that if two organizations are going to work together and coordinate on, you know, if you agree in the language,

they have to work out the details. If you agree on the argument, that is they’re building a relationship. These are two groups that

are building a relationship. And there’s not really reason to expect that it will just vanish once

we move on to a different project than that. Any errors that are left over from there, we’re probably

probably tend toward randomness. And so you probably actually making harder to reject the null hypothesis here. Yes,

very much so. Let’s move on, talk a bit more about some more specific theoretical questions.

So, Zach, I think you want to ask about position taken. Yeah. I mean, one of the things

that I thought was really insightful that you guys talk about, right? It’s like if there are interest groups

taking positions on every single bill in Congress, these signals would be significantly weaker.

I think that’s really important to keep in mind. But it also may be wonder what is the variance

of position taking among interest group populations? Right. Like are some

interest groups just taking a handful? But the mean interest group is taking fifteen positions.

Do you know anything about that side? So we do. We do look at interest

group count like the number of within the data set. We look at the

number of times. That’s interest groups endorse legislation. And

I don’t read the exact statistics, but I think the range goes up to about 50. But the median is one.

You know, most of these groups are signing, you know, one, maybe two, putting their name on one, maybe two. The ones

that are signing or that are endorsing multiple pieces of legislation are really the big names

that you would expect. The National Education Association, the U.S. Chamber of Commerce, National Association of Manufacturers.

They are signing up when it’s only like 20 bills that they’re endorsing,

though. So this isn’t within the context of these dear colleague letters still. That’s that’s a that’s interesting.

Yeah. Because it’s it’s a matter of these are endorsements that are solicited at the earliest age of the legislative

process. So I think that’s really kind of the key distinction. We’re not saying that the National Association of Manufacturers only

endorses, you know, 20 bills in a given Congress. But if we if groups are if members are approaching

them at the earliest stages to get their support when they’re first introducing something,

it’s much more limited. Do you think the members will occasionally hold back endorsements efforts? Say

it’s a lefty group or a right leaning group just cannot signal that

that this is a left leaning or right leaning piece of legislation. It’s possible,

but for the most part, these are actually endorsements that are solicited by the members. So that

sometime I should say that the process can go in two different ways. So typically

what you end up having is a member of Congress has an idea. They put together a proposal and then

they can perhaps drop it out to some relevant interest groups to see if they’ll support them. So if I’m a member

and I want to do if I’m a Democratic member and I want to introduce a major education bill, I’m

going to shop it to the National Education Association and see what they think. And

maybe they’ll endorse it. At the very least, they want to make sure that they’re not going to oppose it. So that’s, you know, the

endorsement kind of a high ask. And a lot of cases in other cases, you’ll have interest groups that approach

members of Congress with, you know, here is the thing that we would like you to do. And then you have kind of a built-In

endorsement, assuming that the member does as they ask. But. So this

process is really interesting, I think potentially counterintuitive for people. So this

almost looks like a reverse lobbying. So this is members of Congress asking interest groups for their endorsement

rather than the other way around. And is that what you find or do you think that’s that this process

leans more on one way or the other? So we don’t really. I won’t say that our paper

finds. Because that, you know, that’s not something that we address in the paper. We don’t do the research that would be needed to actually demonstrate

that include inclusive way. But I will say based on qualitative data, I certainly think that’s true.

Based on my own experiences, I work in the Capitol on Capitol Hill for a while. I think that’s true. But also

we do a number of interviews with staff and and they describe a similar process of, you know, you

have an idea. And one of the first things you want to do is see. You can get on board.

Have you read about Poindexter’s book on I’m Lobbying 1963 old the old book? No, I say

they they did like a big go serve a base kind of semi quality of study of.

It was like a piece of terrists largest legislation. And they they went in expecting and

they say, isn’t the book kind of an iron triangle relationship? Interest groups are really telling members of Congress what to do and

all that. And they’ve got this great quote from an anonymous senator who says, like no other interest groups aren’t chasing

me around. I’m chasing them around like I’m trying to I’m trying to corral this coalition together. And and that’s that’s kind of part why I’m doing.

I I imagine that’s something that these data in another another study could really illuminate. Yeah.

All right. So, Christine, you want ask you about heuristics? Yeah, I think there’s a really

interesting normative consideration here about using dear colleague letters

as a heuristic for whether or not other people are going to support this bill. So there’s the potential

that they support the bill from the dear colleague letter alone. There’s the potential that they read the whole

bill. Is this just a filtering mechanism to make people’s lives in the House easier?

Or is this an abdication of responsibility?

I mean, I do think it I think it’s a filtering mechanism. And I think that when you consider

the sheer number of bills that are introduced in a given Congress and so on the House

side, for example, if you have six or seven thousand bills, generally speaking, introduced in

a two year period and members of the House of Representatives are limited to 18 staff members,

half of what you’re typically out in your district, you have three or four legislative staff members who have massive

portfolios covering a wide variety of issues. And I mean,

the reality is, you know, they’re not going to be able to read every single bill that’s introduced. And so

I think that one of the real values of these dear colleague letters broadly and

specifically when these interest group endorsements come into play, is that when members send out these letters,

it’s a way to grab attention from a staff member who then continues to do research.

So it’s basically it’s not as if they’re not going to ever read the bill before

they signing where they signed on to it. We’re not suggesting that that they’re just signing on based group

based purely on these dear colleague letters, so much as it’s a way of drawing attention

as you’re filtering through dozens of these today. So I think this connects back

to the main theme of this paper, is that this is all about a signaling mechanism, which again

relates to what your coauthors talked about in their work on amicus brief. So this is a very similar process as

before the Supreme Court. The people who are signing onto amicus briefs, whether it’s the solicitor general or

these really powerful social power, heavy interest groups are showing

support or not support for a piece of

legislation or whether it be in the in the court’s context as a potential decision to

be made. That’s also a filtering mechanism. And so you can really see how social power

can translate across these these different institutions. And I think that’s really,

really interesting contribution. Absolutely. Thank you. One of the excuse me. Interesting descriptive

things that you talk about with regard to the letters that I picked up on. You say there’s a negative

correlation between letters that are listing interest groups and that are explicitly seeking cosigners

or co sponsorship. No cosigners. And you attribute this to the

quick turnaround of the letters. But it. Are you? Yeah.

I mean, I was surprised by this bill. Why do you think a letter inviting a member to a reception or a letter to

the administration is a stronger place to have a queue than a request for

a cosigner specifically? So I don’t think we do suggest that. I think

if I. I think what you’re seeing is that in the earliest. So first of all, we actually filter out

the letters that have to deal with policy letters and. Invitation’s

anything that’s not dealing with bills. So what we’re left with are the letters that are recruiting co-sponsors.

And the letters that are urging support or opposition on the floor, which is actually surprisingly

small number of letters in the whole dataset. So these letters really are used

to draw attention at the earliest stages of the legislative process, more so than that kind of end-stage lobbying

where you have a lot more you know, once once the bill has been filtered down, once the bill has been survived,

the winnowing process in is one of the few things that is actually being considered on the

floor. Members have a lot more time to do

the research and that would be necessary to make decisions on the piece of legislation. They’ll be more active.

They’ll be lobbied definitely by the groups. And so we do find that in the later stages, the legislative process,

having a larger coalition from that earlier stage, still having a larger coalition actually

does predicts some measure of legislative success, which makes sense if you think about

just the size of the lobbying coalition, that would be active on that piece of legislation. I think more broadly, I

think what the the real interesting beginning of this research project and I know we’re going to be seeing

more of is you’re eliminating an early phase of the legislative process. Now, at

the Policy Dennis project, we can do some of the early phase stuff. We got hearings, we’ve got bills, we’ve got we’ve got some things like that.

This is occurring before all that. But ignoring that for a second. Most congressional scholarship is done

on roll call votes or later stay in later stages. What do we learn by shifting that

window from the roll call stage to the earlier stage? How’s the process look different from where you’re

looking at it versus when most Congress scholars looking at it? Yeah, I think it’s a different

set of questions. If you’re looking at roll call votes, if you’re looking at how members

of Congress vote on the bills that come out of Congress, you’re looking at, you know, what

is the legislation that Congress passes? I think we’re looking in the earliest stages of the legislative process.

We’re looking at the bills that are introduced by looking at the bills that get early support. What we’re able to

do is look at the ideas, the bills. And,

you know, even if we go more broadly than just looking at bills, but also looking at these policy letters, the ideas

that individual members of Congress are trying to advance without

having to worry about the constraints of the leadership’s agenda control. So the bills

that come up to a vote on the floor are heavily dictated by what the leadership

is willing to allow come up to a vote, whereas there’s no such constraint on what bills are introduced

or what bills have dear colleagues associated with them. When you when you gave your job talk almost two years

ago, at this point, you had you show this, this you were showing other parts of this work

where you where you show that there’s a lot of cooperation at this early stage. And then even in the modern Congress, when

the modern Congress, once we get to the roll call stage, we get highly polarized. Lots of this. And I asked you a question

and even you didn’t answer it the way that I thought you’d answer. And I was so much more. Your return is much more interesting. I said, so why?

Why why do we. Why is the break down? Whereas a break down the process. And your answer was, I expect you to

answer it. It’s all genociding. It’s all it’s all other other kinds of stuff. All public pressure.

But we said was that a lot of stuff passes in omnibus legislation and voice votes. So is.

Are we showing Congress being more productive than we think they’re being? Is Congress

at this stage actually getting a lot of stuff done? And it’s really the high profile stuff that we observe polarization on.

So I think that this is an area that requires a lot more study. Well,

my intuition is that, yes, that is the answer. That is my

one of the challenges that we have in political science is or is specifically in legislative and such in Congress

is that these bills are frequently incorporated into larger pieces of legislation. And it’s not always an omnibus

bill I can frequently be. So the Veterans Affairs Committee does have takes

and collapses them into a single piece of legislation and sends it to the floor. If you look at the legislative

record for those individual bills, they show that they didn’t pass because the major

the kind of collection of them did so include those that were just starting

to get into kind of tracing these bills through the legislative process when they get incorporated into

other pieces of legislation. John Wilkerson at the University of Washington is doing some great work with

that. I went to a talk at Southern by a couple of

scholars at the University of Michigan and BYU graduate students who are working on something similar.

And I think it’s really important work. And I think that having that

ability to trace bills through other afet possible avenues of legislative success

will allow us to make stronger empirical claims that, yes,

Congress is actually getting more done than we think it is, because right now the main statistic that we use

to kind of try to make that argument is that the length of bills is actually getting long. Even while the number of bills passed

reduces, which is messy, I’ll put in a plug for our titles dataset, which we released about five

years ago, four years ago. The first that I worked on here at the Policy Agendas Project, that is it’s used

a little bit, not that much when it separates us. Public laws into individual work were in laws called titles,

which can sometimes be hard to contain. Individual bills can sometimes contain pieces of bills. Families looking for a project.

It needs more analysis. If anything, we hope that this podcast encourages

you to come up with new and creative data. And that’s what this is.

And I think that’s the biggest contribution here. That theory is solid. The empirics are really, really sophisticated.

It’s using really new ways of measuring things. But mostly this is a data set that is not roll

call votes and we should all celebrate. I love dunking on rollcall.

So given all this unorthodox lawmaking that’s going on. What does that mean for credit claiming for the interest

groups and what might that mean for the ways that these interest groups are going to endorse in these letters moving

forward? I mean, I think that the idea

that credit claiming when it comes to either interest groups or members of Congress requires that a bill with

that person’s name or organization’s name attached to it has to pass, just isn’t true. I mean,

if we think about especially I think especially when we’re dealing with the rise of social media, for example,

one of the things in the larger project where I’m looking at the collaboration between members of Congress is

I draw out a number of tweets and Facebook posts and press releases and other forms of media

where members of Congress are claiming credit for having passed

legislation that they weren’t the like main listed sponsor on

because they were the collaborative partner. But they’re still claiming it as their own. They’re still going out to their constituents

and saying, I passed this bill. And I mean, I don’t think you can get away with doing it

on something that you had no involvement in because someone at some point is going to fact check it. But if you have

if I’m working with you on a piece of legislation and I’m helpful to you, and then I go back

and it passes and I go back to my constituents and said I passed this bill, presumably

if the hometown newspaper decides to like fact check this because they noticed it didn’t match up

and they go to you, you would confirm as the bill sponsor. Yes. This was a collaborative effort

between the two of us. That’s right. Suddenly, I think no.

Makes no sense. I worry, though, right? Like, don’t the interest groups want to be

able to to fire up their base, too, though? Right. Like, it’s not very exciting to say, like,

oh, this one section in this omnibus bill was for us, as it is to say,

like we endorse this bill. It made it all the way through the process. Everyone appreciates how difficult

that is. And like that was really good for all of us. But I think that the messaging tends to be

much more on the frame of. We got this thing done. So it’s less

about did we pass the bill? Did we not pass the bill? It’s like we have done this thing and delivered this. Good to you, our

members. So, you know, the American Farm Bureau. You know,

they’re not gonna be able to claim credit for every single piece of a farm bill, for example.

You know, they’re going to highlight the points that are of particular importance to them and their members

and ignore the parts they don’t like when they talk about it. Fair enough.

All right, so I think that’s that’s a wrap up questions for for for this podcast, so as we wrap up the podcast itself,

I would love to hear the recommendation from you for a work of political science that you think our listeners

should should go out and read. So I just finished rereading for the

fifth time. E Shatz Snider’s the somersault semi sovereign people I I’m teaching

this semester. And one of the things I mean, I just love that book. But one of the things

I was really struck by going through it, it’s it’s looking at kind of addressing

this question of, you know, who has power in American democracy and the relationships between,

you know, the role of individual voters and interest groups and leaders all working together.

But one of the things I was really struck by going through this older book is the

number of times I was like, wow, that really applies for explaining what’s going on right now. It is, I think, an

incredibly applicable work in a lot of ways. The chapter in that book, I think is just so under-discussed

is the chapter where he he key. Can he ask about how some these very, very small

interest groups can get policy changed? An interest group? I think it was one of the examples he used. It was

like the Shoreline Association of Rhode Island or something like that with 300 members and they somehow managed

to get a piece of legislation changed shirt and showing that it did. That actual number doesn’t matter.

What matters is the fight that they’re able to draw on the allies they’re able to seek. Do you think that do you

think that this project you’re working on, that the AGP historical we’re talking about now is is

an extension of shash night or I mean, are you are you showing, I think, a much more beside him in here? Yeah, I’m I’m

a youth boss. I mean, theoretically. Are you are you are you in that tradition where you’re looking for

essentially groups attracting a crowd for a fight? You know, I didn’t think about it in exactly

those terms, but I think that’s a fair. I think that’s certainly a fair summary. All right. Now, I wish

I could reframe, you know, kind of the interloper. You know, there’s always the next article.

There’s always something in print at the American Journal of Political Science. Allison, thank you very much for joining us.

Zach, Christine, thank you for joining us. Liberal Arts Instructional Technology. Thank you for producing this podcast. Until

next week, this has been your Policy Agendas podcast.