Joshua C. Fjelstul, Ph.D.

 Welcome!

I’m a political economist and data scientist. I received my Ph.D. in Political Science from Emory University in 2019. I’m currently a postdoctoral research fellow at Washington University in St. Louis.

You can contact me at joshua.fjelstul@wustl.edu.

website2.jpg

I’m a political economist and data scientist. I received my Ph.D. in Political Science from Emory University in the spring of 2019. I am currently a postdoctoral research fellow in the Department of Political Science at Washington University in St. Louis. I also have an M.A. in Political Science from Emory; a B.A. in Government and History, with Special Honors in Government; and a B.B.A. in the McCombs Business Honors Program from The University of Texas at Austin (UT).

I study comparative political institutions, political economy, and political methodology. My substantive interests span legislative, bureaucratic, and judicial politics. My work focuses on inter-branch political conflicts over policy implementation in multi-level political systems, including the European Union. I am particularly interested in implementation of economic regulations. My methodological interests including using machine learning (deep neural network architectures) to measure public position-taking by individual legislators and judges based on the text of legislative speeches and judicial rulings. My applied research uses a wide array of methods, including game theory and causal inference techniques. My research has appeared in the American Political Science Review, European Union Politics, and International Interactions.

My current research agenda has two main strands. First, I study the strategic behavior of actors within legislative, bureaucratic, and judicial institutions in inter-branch political conflicts over policy implementation in multi-level political systems, where policy is made by a higher government and implemented by lower governments. I also study how the internal structure and organization of bureaucratic and judicial institutions shapes their external behavior in these political conflicts. My work explores the conditions under which government agencies implement policies that differ from what policy-makers intended; the effectiveness of institutionalized pre-trial bargaining procedures at resolving disputes over policy implementation; the conditions under which the strategic behavior of courts and prosecutors deters government agencies from implementing policies that run counter to the preferences of policy-makers; and the strategic override of adverse judicial rulings in disputes over policy implementation by legislators. I study these dynamics in the context of the European Union (EU), which provides a data-rich environment for theory testing.

Second, I am developing a widely-applicable methodology that uses cutting-edge machine learning techniques (self-supervised deep neural network architectures) to measure public position-taking by individual legislators (over time and by policy domain) based on the text of legislative speeches. I am validating the methodology using a new corpus of over 300,000 floor speeches from the Canadian House of Commons. This corpus ties into a larger data-collection project I am managing, consisting of over 25 interrelated datasets, that maps together all bills, votes, floor speeches, and committee speeches in the Canadian House of Commons (2006-2018). I am also theorizing the processes by which these speeches are generated to identify and address sources of selection bias. My in-progress book project, mentioned above, will use this methodology to test theories of why roll-call votes are called and to explore how selection bias in the generation of roll-call vote data affects what we can infer about the preferences of legislators from observable voting behavior. In another project, I am using this methodology to measure public position-taking by judges at collegial constitutional courts, where voting behavior is unobservable. I plan to integrate these two research strands by using text-based measures of position-taking to test spatial theories of bargaining and voting behavior in legislative and judicial politics.

I employ a broad set of technical skills in my work. My quantitative work uses Bayesian statistics and causal inference techniques, including difference-in-differences designs and matching. My skill sets include quantitative text analysis, machine learning, structural equation models, computational simulations, data visualization tools, and advanced programing techniques. I have taught workshops on quantitative text analysis for the Institute for Quantitative Theory and Methods (QTM) at Emory University and the Emory Center for Digital Scholarship (ECDS). I have used these skills to create a number of large, original datasets to facilitate my work, including the recently published EvoEU Dataset, which models the complete network of EU law, 1952-2015 (over 365,000 legal documents and over 900,000 connections between them).

 Publications

Read my publications.

website3.jpg
  • Fjelstul, Joshua C., and Clifford J. Carrubba. 2018. “The Politics of International Oversight: Strategic Monitoring and Legal Compliance in the European Union.” American Political Science Review 112 (3):429-445.

Abstract. States often violate international agreements, both accidentally and intentionally. To process complaints efficiently, states can create formal, pre-trial procedures in which governments can negotiate with litigants before a case ever goes to court. If disputes are resolved during pre-trial negotiations, it can be very difficult to tell what has happened. Are governments coming into compliance? If so, are they only doing so when they have accidentally committed a violation, or even when they are intentionally resisting? Or are challenges simply being dropped? This paper presents a formal model to address these questions. We develop our theory in the context of the European Union (EU). To test our model, we collect a new data set of over 13,000 Commission infringement cases against EU member states (2003-2013). Our results suggest that accidental and intentional noncompliance both occur, but that intentional noncompliance is more common in the EU. We find that the Commission is an effective, if imperfect, monitor and enforcer of international law. The Commission can correct intentional noncompliance, but not always. It strategically drops cases that it believes it is unlikely to win. 


  • Fjelstul, Joshua C. 2019. “The Evolution of European Union Law: A New Data Set on the Acquis Communautaire.” European Union Politics 0 (0):1-22.

The European Union legal system is one of the most complex and sophisticated in the world. This article models the Acquis Communautaire (i.e. the corpus of European Union law) as a network and introduces the Evolution of European Union Law data set, which tracks connections between European Union primary law, European Union secondary law, European Union case law, national case law that applies European Union law, and national law that implements European Union law. It is the largest, most comprehensive data set on European Union law to date. It covers the entire history of the European Union (1951–2015), contains over 365,000 documents, and records over 900,000 connections between them. Legislative and judicial scholars can use this data set to study legislative override of the Court of Justice of the European Union, the implementation of European Union law, and other important topics. As an illustration, I use the data set to provide empirical evidence consistent with legislative override.


  • Fjelstul, Joshua C., and Dan Reiter. 2019. “Explaining Incompleteness and Conditionality in Alliance Agreements.” International Interactions 0 (0):1-27.

What explains the form of international alliances? Conventional wisdom forecasts that audience cost-averse states like democracies demand that their alliances include conditions, such as limitations to specific regions, to reduce the likelihood of paying the audience costs of treaty violation. This paper builds on the conventional wisdom by proposing that democracies are especially likely to demand that an alliance include conditions that make the treaty more incomplete, that is, make the compliance requirements more ambiguous, in comparison to including conditions that do not make the treaty more incomplete. Treaty incompleteness gives leaders another tool to reduce domestic political audience cost risks. Incompleteness makes it easier for a leader to argue to a domestic political audience that a wider array of actions, including actions such as nonintervention on behalf of an embattled ally that other signatories might view as noncompliant, is actually compliant, reducing domestic audience costs for these actions. Examining all international alliances since 1816 using seemingly unrelated regression, the paper finds empirical support for the theory: alliances with democracies are more likely to have conditions making the treaty more incomplete, but not more likely to have conditions that do not make the treaty more incomplete.

 Working Papers

Read the latest versions of my working papers.

website1.jpg
  • Fjelstul, Joshua, Policy Implementation, Noncompliance, and the Judicial Impact of Courts.

Abstract. Policy-makers depend on government agencies to implement laws, and courts adjudicate disputes over incorrect implementation. This paper analyzes how the politics of adjudication affect the implementation of policy. Under what conditions do government agencies more closely implement laws, and when do courts deter incorrect implementation? I model the strategic interaction between an implementing actor, a plaintiff, and a court over the implementation of a law. In equilibrium, implementing actors make concessions by implementing more complaint policies in order to avoid litigation. I find that courts are most effective at deterring noncompliance for intermediate levels of preference divergence between policy-makers and implementing actors. I test the observable implication of the model in the context of the European Union (EU) using a novel dataset of implementation opportunities and noncompliance cases.


  • Fjelstul, Joshua, The Politics of Noncompliance in Common Markets.

Abstract. States create common markets to accrue consumer welfare gains. Given incentives to cheat to protect domestic firms from foreign competition, they create international regulatory regimes to manage noncompliance. But the politics of compliance generates systematic bias in the types of noncompliance cases that get litigated. I develop a formal model that explains how the politics of compliance in regulatory regimes systematically distorts the welfare gains that states accrue from developing common markets. The model predicts the sectors in which regulatory regimes are effective at reducing trade barriers — those with intermediate levels of homogeneity in terms of firm productivity — and the downstream consequences for the performance of individual firms and consumer welfare. I show that if we do not take into account the politics of compliance, our predictions about the distributive consequences of common markets and the conditions under which consumer welfare will increase will be systematically biased.


  • Fjelstul, Joshua, Rationality in the Public’s Evaluation of Economic Policy: Evidence from the European Sovereign Debt Crisis.

Abstract. The increased use of direct democracy to set economic policy raises an important question: To what extent are aggregate preferences over complex economic policies consistent with individual rationality? Even if we do not believe that individuals have the cognitive sophistication or information to rationally evaluate policies themselves, can robust public debate over policy lead individuals to hold preferences that are consistent with rationality? This question has far-reaching implications for the coherence of economic policy. I theorize how a rational individual would evaluate the following policy: Commission monitoring of member state compliance with the Stability and Growth Pact (SGP), which governs the European Economic and Monetary Union (EMU). Then, I present empirical evidence from the European Sovereign Debt Crisis that, in a high-information environment, where public debate is robust, individuals express preferences over Commission monitoring that, in the aggregate, are surprisingly consistent with individual rationality.


  • Fjelstul, Joshua, Life After Brexit: The Impact of Common Law Judges at the Court of Justice of the European Union.

Abstract. How will Brexit impact the jurisprudence of the European Court of Justice (ECJ), given that the United Kingdom (UK) is one of the EU’s two common law member states? This paper uses a new dataset of over 10,000 ECJ cases (the universe of cases, 1956-2016) and quasi-experimental techniques to estimate the causal effect of common law judges on case law citations. To assess the likely impact of Brexit, I identify three specific causal mechanisms by which common law judges can push the ECJ to cite more case law, the conditions under which each mechanism can operate, and how common those conditions are. Based on my findings, I run a simulation to assess the likely impact of losing the UK judge. The impact of Brexit is likely be minimal in the short term, but more substantial in the long term.


  • Carrubba, Cliff, Joshua Fjelstul, Matt Gabel, and Dani Villa, The Internal Organization of the European Court of Justice: Discretion, Compliance, and the Strategic Use of Chambers.

Abstract. Overseeing the complex legal system of the European Union (EU), the Court of Justice of the European Union (CJEU) is an extraordinarily productive international court. To help the Court manage its substantial caseload, member states have given it discretion over which cases it can hear in small Chambers. The rules governing the Court’s use of Chambers have evolved substantially over time. The Chambers System may have a significant impact on the application of EU law, but we know remarkably little about it. This paper provides the first comprehensive description of the rules governing the Court’s use of Chambers and their evolution. We assess the degree to which the Court has complied with these rules, evaluate the Court’s strategic use of Chambers, and simulate the Court’s workload under counterfactual rules to demonstrate the impact of the Chambers System on the efficacy of the Court.


  • Cheruvu, Sivaram and Joshua Fjelstul, Do Administrative Deficits Cause Noncompliance with International Law? Causal Evidence from the European Union.

Abstract. What are the causes of noncompliance with international law? A rich scholarship argues that limitations in administrative capacity is one of these causes. We provide the first quantitative causal evidence of this relationship. Employing a dataset of European Commission infringement cases launched against member states and leveraging member state qualification for the FIFA World Cup and UEFA Euros as an exogenous source of variation of member state administrative capacity, we use a difference-in-differences design to provide evidence that administrative deficits cause a substantial delay in Commission infringement proceedings. We conclude by discussing how international organizations can address this noncompliance problem.


  • Cheruvu, Sivaram and Joshua Fjelstul, Can International Institutions Help States to Comply with International Law? Encouraging Evidence from the European Union’s Pilot Program.

Abstract. Do policy solutions exist that can help member states comply with international law? We answer this question by examining a management school solution for noncompliance — proper rule interpretation. In this short article, we provide the first causal evidence of this management school solution to noncompliance by showing that European Union (EU) Pilot — a program designed to facilitate negotiations between member states and the European Commission before formal infringement proceedings through clarifying rule interpretation — improved the infringement procedures' efficiency. By resolving accidental noncompliance with EU Pilot, the Commission has streamlined the process of implementing of EU law by using the formal infringement procedure to pursue strategic noncompliance instead of accidental violations.


  • Fjelstul, Joshua, The Political Economy of State Aid in the European Union.

Abstract. Business regulations in the European Union (EU) are extraordinarily complex, making it costly for small and medium enterprises (SMEs) — responsible for most capital investment in the EU — to infer member states' pro-/anti-investor preferences, which can vary by sector, from their de jure and de facto policies.  How does uncertainty about member states' preferences affect firms' decisions to undertake capital investments? What can states do to credibly communicate their preferences to firms? In this paper, I explore how EU member states can promote investment by using state aid to credibly communicate their preferences to non-recipient, resource-constrained SMEs. I present a formal model to identify the conditions under which signaling can induce non-recipeint firms to accept prospective investment projects that they would have rejected otherwise. I take advantage of new EU reporting guidelines to compile a new dataset of approximately 4,000 state aid awards granted in 2016 and 2017. I use this data to test the model’s main prediction: That aid-granting agencies can attract the same investment projects from non-recipeint firms with smaller state aid awards when their member states are more sensitive to reputational costs stemming from insincere signals.


  • Carrubba, Cliff, Braden Dauzat, Joshua Fjelstul, and Matt Gabel, Bureaucratic Expertise in Judicial Decision-Making.

Abstract. Under what conditions can we expect bureaucracies to choose to provide international courts with third-party briefs? Under what conditions can we expect these briefs to influence courts’ rulings? Carrubba and Gabel (2015) find that the European Court of Justice is more likely to rule in favor of a litigant when the Commission submits a brief on its behalf — a finding that is robust to controlling for the position of the Advocate General. Motivated by this empirical puzzle, we develop a case-space model that examines the strategic interaction between the Commission and the Court. We analyze this model in order to examine how differences in preferences and differences in expertise drive how the Commission chooses to participate in court cases and when the Court defers to the Commission’s position.

 Data Projects

Check out the data projects I’m working on.

website4.jpg

I'm currently working on several large datasets to support some of my current and future research projects. If you're interested in any of my data, please contact me about availability. 


Compliance in the European Union (EU):

I've created a dataset of over 16,000 infringement cases opened by the European Commission against member states, covering 2003–2016. The dataset includes all cases active at any point during this period and tracks every case as it moves through the stages of the Article 258 infringement procedure. The dataset also includes the directorate-general of the Commission that initiated each case (which indicates the policy content) and the EU directive that the case concerns, if applicable. For directives, the dataset also indicates whether or not a case deals with the noncommunication of a directive (as opposed to incorrect transposition). 

This dataset features in The Politics of International Oversight: Strategic Monitoring and Legal Compliance in the European Union, published in the APSR. It also features in two of my working papers: Do Administrative Deficits Cause Noncompliance with International Law? Causal Evidence from the European Union and Can International Institutions Help States to Comply with International Law? Encouraging Evidence from the European Union’s Pilot Program.


Composition of the Court of Justice of the European Union (CJEU):

I've created a dataset of 10,350 ECJ cases from 1952 through 2015 (the universe of cases). For each case, the dataset includes the chamber of the Court, the composition of the panel of judges, the advocate general, the legal procedure (reference for preliminary ruling, action for failure to fulfill obligations, action for annulment, action for failure to act, or appeal), and citations (to case law, secondary law, and the EU Treaties), among other information.

This dataset features in two of my working papers: Life After brexit: The Impact of Common Law Judges at the European Court of Justice and The Internal Organization of the European Court of Justice: Discretion, Compliance, and the Strategic Use of Chambers.


Evolution of European Union (EU) Law:

The European Union legal system is one of the most complex and sophisticated in the world. This article models the Acquis Communautaire (i.e. the corpus of European Union law) as a network and introduces the Evolution of European Union Law data set, which tracks connections between European Union primary law, European Union secondary law, European Union case law, national case law that applies European Union law, and national law that implements European Union law. It is the largest, most comprehensive data set on European Union law to date. It covers the entire history of the European Union (1951–2015), contains over 365,000 documents, and records over 900,000 connections between them. Legislative and judicial scholars can use this data set to study legislative override of the Court of Justice of the European Union, the implementation of European Union law, and other important topics. As an illustration, I use the data set to provide empirical evidence consistent with legislative override.

 Teaching

Learn about my teaching interests.

website5.jpg

My teaching interests include comparative politics, political economy, EU politics, political methodology, and game theory. In the spring of 2020, I will be teaching a comparative politics course on the European Union at Washington University in St. Louis. I have co-taught Emory University’s graduate-level Introduction to Game Theory course (with Cliff Carrubba). I have also been a teaching assistant for Introduction to International Relations (undergraduate), Political Science Methods (undergraduate), and Introduction to Game Theory (graduate).

Graduate:

  • Co-Instructor, Introduction to Game Theory (POLS 513), with Cliff Carrubba (Spring 2016).

  • Teaching Assistance, Introduction to Game Theory (POLS 513), with Shawn Ramirez (Spring 2018).

Undergraduate:

  • Teaching Assistant, Introduction to International Relations (POLS 110), with Zack Bowersox (Spring 2017).

  • Teaching Assistant, Introduction to International Relations (POLS 110), with Eric Reinhardt (Fall 2015).

  • Teaching Assistant, Political Science Methods (POLS 208), with David Davis (Spring 2015).

Workshops:

  • Instructor, R Application: Social Media Scraping, QTM-ECDS Fall 2019 Workshops, Coding in R Series, the Institute for Quantitative Theory and Methods (QTM) and the Emory Center for Digital Scholarship (ECDS), Emory University, February 13, 2019.

  • Instructor, R Coding Intermediate 2: Strings, QTM-ECDS Fall 2019 Workshops, Coding in R Series, the Institute for Quantitative Theory and Methods (QTM) and the Emory Center for Digital Scholarship (ECDS), Emory University, February 6, 2019.

  • Instructor, R Coding Basics, QTM-ECDS Fall 2019 Workshops, Coding in R Series, the Institute for Quantitative Theory and Methods (QTM) and the Emory Center for Digital Scholarship (ECDS), Emory University, January 23, 2019.

  • Instructor, R Coding Intermediate 2: Strings, QTM-ECDS Fall 2018 Workshops, Coding in R Series, the Institute for Quantitative Theory and Methods (QTM) and the Emory Center for Digital Scholarship (ECDS), Emory University, September 19, 2018.

  • Instructor, R Coding Intermediate 1, QTM-ECDS Fall 2018 Workshops, Coding in R Series, the Institute for Quantitative Theory and Methods (QTM) and the Emory Center for Digital Scholarship (ECDS), Emory University, September 12, 2018.

  • Instructor, Text Analysis for Social Scientists, the Institute for Quantitative Theory and Methods (QTM) Workshop Series, Emory University, Spring 2016.

  • Instructor, Text Analysis for Social Scientists, the Institute for Quantitative Theory and Methods (QTM) Workshop Series, Emory University, Spring 2017.

I've also completed the Teaching Assistant Training and Teaching Opportunity (TATTO) program conducted by the Laney Graduate School at Emory. TATTO is an innovative teacher training program designed to prepare Ph.D. students to teach undergraduate courses. This program consists of four components: a short summer course, departmental seminars, a teaching assistantship, and a teaching associateship (a co-teaching opportunity).