class: center, middle, inverse, title-slide # Why do Agencies (Sometimes) Get So Much Mail? ## Public Pressure Campaigns and Bureaucratic Policymaking ### Devin Judge-Lord ### Slides: <a href="https://judgelord.github.io/research/whymail">judgelord.github.io/research/whymail</a> <br> Data: <a href="https://github.com/judgelord/rulemaking">github.com/judgelord/rulemaking</a> --- <!-- Hi, my name is Devin Judge-Lord, I'm a Ph.D. candidate at Wisconsin This paper is the first bit of writing I've done for my dissertation on public pressure campaigns in bureaucratic policymaking, for example, mass comment campaigns. It sets out concepts and measures for these occasional bursts of civic participation, which I will then use to assess whether it matters. I am still workshopping these concepts and measures, so your comments will be decisive for where I take this in the next few years. SLIDE My dissertation is at the intersection of two kinds of literature that have not had much contact. On the one hand, civic engagement has long been a core interest of political scientists. We are highly uncertain about the size of effects from actions like letter writing, signing petitions, protests, but there is little doubt that, at a large enough scale, these things matter. --> ## Rulemaking, ~90% of new U.S. law -- ## Whom does notice and comment rulemaking empower? -- ## If groups build the power to pass a Green New Deal, what happens when 30+ agencies write the actual policies? -- ## Will organized public pressure to carry over to agency policymaking? -- If so, by what mechanisms? --- exclude: true Rulemaking, ~90% of new U.S. law  --- <!-- this project sits at the intersection of two literatures--> ## Civic engagement ### Writing to government officials, signing petitions, attending hearings, attending protests, or donating to a political campaign (Verba and Nie 1987).  Most rules receive <10 comments -- , some receive millions. <!--  --> --- ## 1946 Administrative Procedures Act ### "opportunity to participate in the rulemaking through submission of written data, views, or arguments" - U.S.C.§553(c) ### "a bill of rights for the hundreds of thousands of Americans whose affairs are controlled or regulated" - Senator Pat McCarran --- exclude: true ## Mass engagement in bureaucratic policymaking ### "The ‘mass comments’ occasionally submitted in great volume in highly salient rulemakings are one of the more vexing challenges facing agencies in recent years." (Sant’Ambrogio and Staszewski 2018) --- exclude: true ## The debate in administrative law - **Optimists:** Notice and comment is the "purest example of participatory democracy in actual American governance." - **"Regulatory democracy" reformers** (Cuéllar 2014, Johnson 2013) argue that it could be more discursive. Agencies often discard non-technical comments, but they shouldn't (Mendelson 2011). - **Pluralists:** "The goal of e-rulemaking is to more fully capture such credible, specific, and relevant information, not to solicit the views of random, self-nominating members of the public." (Herz 2016) - **Pluralist reformers** suggest that lobbying groups disclose there membership, funding, and decisionmaking processes (Seifter 2016) - **Skeptics:** Public comment distracts from good policy. Abolish it. --- ## Bureaucratic policymaking ### Discretion (institutional design, information asymmetry, multiple principals, coalitions) -- > "The realm of durable, consequential policies is the realm of organized interests." - Hacker, Hertel-Fernandez, Pierson, and Thelen <div class="figure"> <img src="Figs/causal-classic.png" alt="A Simple Model of Interest-Group Lobbying" width="4383" /> <p class="caption">A Simple Model of Interest-Group Lobbying</p> </div> ??? The normative interest in expertise on and deliberation--in the quality of information--have kept distracted from empirical reality. These are not attempts at information or deliberation; they are attempts to gain power with quantity rather than quality. In Susan's recent Annual Review on the politics of rulemaking, she found that the literature has next to nothing in the literature about mass engagement, besides some filp skepticism. SLIDE For many reasons, Rulemaking is usually an insiders' game--people here and others have shown that to the extent, anyone can influence bureaucratic policymaking; it is the usual suspects, mostly businesses. --- ## Public pressure as *political* information <div class="figure"> <img src="Figs/causal-direct.png" alt="Direct Effects of Political Information: Incorporating Political Information into Models of Bureaucratic Policymaking" width="4383" /> <p class="caption">Direct Effects of Political Information: Incorporating Political Information into Models of Bureaucratic Policymaking</p> </div> --- exclude: true ## Public comment periods ### Usually an insiders' game, but occasionally not. Does it matter?  <!-- but SOMETIMES it is very different, or at least it looks different because lots of ordinary people are voicing their opinions, but we don't know if it matters. --> <!-- Alex mentioned the Payday loan industry funding of the Congressional Black Caucus --> --- exclude: true <!-- I argue that, if public pressure campaigns affect policy, it is through political information --> ## Political information (e.g., signals of the scale and intensity of public support)  `\(\Diamond \text{ observed actions, } \bigcirc \text{ unobserved prefrences, } \square \text{ information}\)` <!-- Before we can assess effects, however, we need to know what exactly we are assessing the effects OF and WHEN we expect to see these effects. This is my present task. --> --- class: inverse ## Political information (e.g., signals of the scale and intensity of public support) ### 1. What is it? ### 2. How do we measure it? ### 3. Why does it occur? (selection problems) --- exclude: true <!--## Selection process --> ### Which rules see public pressure? - Significant policy changes at less-expert agencies (Moore n.d.) - Ones people dislike? (negativity bias) - Controversial, salient? (both endogenous) <!-- indeed, we often measure salience this way --> ### -> Opportunities for mobilization --- exclude: true background-image: url("present/tweetBenWilkerMoveon.png")  --- <!-- First, I want to assert that what we care about here is MOBILIZED commenting. Sure, a few of us in this room may peruse open dockets on regulations.gov, but this is just not what ordinary people do. By ordinary, I mean not professional policy influencers.--> # *Mobilized* civic engagement -- ### Who mobilizes? <!-- Tricky concepts! --> <!-- In the paper, I suggest that many conflicts that mobilize thousands are best seen as over public and private goods rather than cultural issues --> - Interest groups/ "advocacy" groups (Shapiro 2008; Potter 2017)/ "societal constituencies" (Balla et al. n.d.) - Unions, industry associations, 501c3s **on both sides** <!--sometimes people that don't fit that concept--> <img src="Figs/EPA-protest-kills-jobs.jpg" width="45%" /><img src="Figs/EPA-protest-nurses.jpg" width="45%" /> -- # -> Coalitions --- # Why mobilize people? ### Outside lobbying #### 1. Expand the scope of conflict (Schattschneider 1975) vs. #### 2. Leverage resources into an impression of public support (Skocpol & Hertel-Fernandez 2016) ### Not lobbying #### 3. Meet member expectations or recruit new members (Carpenter 2020) --- exclude: true ## Comments evoke public support "Along with 80% of the American people, I strongly support ending commercial trade in elephant ivory in the US." "petition signed by 67,275 self-proclaimed United States residents" and “more than 110 East Coast municipalities, 100 Members of Congress, 750 state and local elected officials, and 1,100 business interests, all of whom oppose offshore drilling." --- class: inverse middle # Tasks ## 1. Identify coalitions ## 2. Measure mass engagement ## 3. Identify which campaigns ought to matter --- ## Identifying Coalitions ### Text reuse 1. Remove text copied from proposed rule or request for comments 2. Exact match = duplicate (cosigned) comments 3. 10-word moving "window" = part copy vs. entirely unique ### Cluster comments by word frequency and sentiment ### Inspect and hand-code clusters ("coalitions") <!-- EPA has contractors do this by hand! --> --- exclude: true <!-- Second, we need to measures what campings are doing that might be influential. I call this political information, following Nelson and Yackee --> ## Political Information ### "Coalition lobbying can generate new information and new actors—beyond simply the ‘usual suspects’—relevant to policy decision makers. Thus, we theorize _**consensus**_, coalition _**size**_, and _**composition**_ matter to policy change" (Nelson and Yackee 2012) --- ## Measuring Political Information* ### 1. Number of comments per coalition ### 2. Average effort per comment Co-sign an org's form comment <--> Partially unique <--> Unique Lower effort <------------------------------------------> Higher effort ### 3. Potential contagion (share of unique comments) -- *Agencies may not process comments in ways that deliver this information to decisionmakers --- exclude: true ## Selection issues ### Which campaigns ought to matter? ### How to tell them apart? | | Inside lobbying | Outside lobbying | |:---|:---:|:---:|:---:| | "Normal" lobbying | High | None | | "Going public" | High | High | | "Disrupting consensus" | High | Low | | "Going down fighting" | Low | High | --- class: inverse middle # Does the institution enable grassroots contestation of powerful interests or astroturf legitimation of them? --- class: inverse middle center # First cut --- class: center ## Most comments are mass comments <img src="https://github.com/judgelord/dissertation/raw/master/Figs/comments-mass-1.png" width = 700 /> --- ## Who mobilizes? - A few public interest groups -- (more "grassroots" than "astroturf") -- - More "conflict expansion" than groups with the most resources "buying" the impression of public support <img src="Figs/toporgs-1.png" width="50%" /> 100 orgs mobilized 39m of 48m regulations.gov comments on proposed rules, ~87/100 lobby in the same in coalitions. --- ## When does mass engagement occur? ### Asymmetric mobilization > negativity bias <img src="Figs/comments-mass-support-vs-oppose-1.png" width="80%" /> --- class: center inverse  # Thanks! Unified Agenda + ORIA reports + Regulations.gov data: GitHub.com/JudgeLord/Rulemaking or DevinJudgeLord@fas.harvard.edu --- exclude: true Political Information in Bureaucratic Policymaking  `\(\Diamond \text{ observed actions, } \bigcirc \text{ unobserved prefrences, } \square \text{ information}\)` --- - 20 orgs mobilized nearly half of comments on the top 10 most commented on rules - CFPB's Payday Lending Rule received 80,000 partially unique comments - DOI's National Monuments Review received 200,000 unique comments  --- exclude: true background-image: url(Figs/ej-second-order.png) background-size: contain --- exclude: true background-image: url(Figs/commentor-race.png) background-size: contain