Dr Nathan Canen joined the Department of Economics in April 2023. Here he discusses his work in the areas of political economy and econometrics, and what attracted him to join the Department.

What research projects are you currently involved with?

My research can be broadly categorised into two fields. The first one is political economy, which studies the role of politics in shaping economic policies and outcomes. The second one is econometrics, which studies statistical methods for economics problems.

Within political economy, my main research projects focus on the organisation of political parties and the use of information by citizens, voters and politicians. For example, I have been studying sources of political polarisation – which is the divergence in political behaviour across party lines. This is a fascinating question given its relevance and implications for policymaking. But it is also deceptively hard to answer: politicians may diverge in their behaviour because of their ideologies, or because opposing parties are effective at convincing them to behave in a certain way. Disentangling these explanations requires careful use of economic theory and data. With new methods, we find that a party’s effectiveness at corralling its members has been an important driver of polarisation in politicians’ behaviour in the U.S. over the last few decades.

As researchers, we typically do not observe what drives legislative or citizens’ behaviour. Instead, we must infer those explanations from outcomes (e.g. how politicians vote on bills, what committees decide to approve, what news citizens view on TV). Quantitative answers to these questions often involve nonstandard datasets (e.g. network relationships among politicians), computationally intensive methods or theoretical properties making them hard to quantify (e.g., our theoretical models may have multiple solutions). Most importantly, we cannot easily ’randomise’ treatment to evaluate alternative political systems and institutions. My research in econometrics addresses these challenges by providing new statistical tools designed for the political economy problems I work on. By carefully thinking through the type of assumptions in such contexts, we can develop better suited estimators and statistical tests.

Why did you choose this research field?

I have always been fascinated by the interaction between politics and economics: it was my main interest during my undergraduate degree! As I continued within the field of economics, I learned more about political economy and became even more passionate about it. I particularly liked its emphasis on thinking through the feasibility of desirable policies. This led me to my PhD at the University of British Columbia, where I worked with my amazing supervisor and co-authors. My first paper studied networks in Congress: how politicians choose who to work with to craft and pass bills.

While I was working on those topics, I found myself facing methodological challenges that required new tools. For example, data about politicians’ networks cannot rely on commonly used statistics, which are built on assumptions of independence across observations. So, I continued to take econometrics classes and think about how I could improve the statistical approaches in my political economy papers. I was then very happy to be able to co-author with econometricians interested in the same types of statistical problems. The rest is history!

What are you planning to work on next?

On the political economy side, I am starting a research agenda on the role of deliberation in decision-making in committees. Many important regulatory bodies use advisory committees when making decisions. For instance, regulatory agencies in healthcare (e.g., the Food and Drug Administration (FDA) in the US or the Medicines and Healthcare Products Regulatory Agency (MHRA) in the UK) use such committees when deciding whether to approve a new product. Congressional and parliamentary committees also perform similar information gathering.

These committees deliberate before making a final recommendation. Their deliberations can be based on existing information, but committee members can also request additional information. Hence, they may act strategically to learn more information, with a goal of influencing (e.g., persuading or convincing their peers), thereby influencing the committee’s decision. In these projects, we use transcript data from FDA Advisory Committees. We transform this data in a way that can be quantified within a theoretical model of learning in committees. Then, we evaluate the role of individual preferences, information and strategic incentives in driving committee-level outcomes and compare their performance to alternative institutional designs. We are hopeful this will lend insights into evaluating the performance of these institutions.

On the econometrics front, I have become interested in the assumptions needed for counterfactual analysis (i.e. the evaluation of policies that have yet to be implemented). For example, how do we evaluate the effects of a future minimum wage, when that minimum wage has yet to be observed? Making these predictions is a hard statistical problem. Typically, researchers may use a theoretical model and assumptions to provide reasonable estimates. However, some of my recent results suggest that researchers may not need all of those assumptions, making their results more ’robust’.

What impact do you hope your work will have on society?

Broadly speaking, I hope my work shows the need for nuance in policymaking, and how theory and data are helpful. It is often tempting to make a conclusion based on a striking figure or descriptive evidence. However, when analysing individual actions – such as in politics – those answers may not faithfully represent actual behaviour or mechanisms. Hence, even well-intentioned policies may have different effects depending on what type of incentives are given.

For example, whether political polarisation is being driven by party organisation, rather than politician ideologies, matters for which policies get approved. Whether voters are demanding too little information, or just not having access to it, further changes the way we think about similar questions.

On the regulatory committees side, we hope that our work will help inform the way such committees are designed. This may include how the committees are composed, rules on what type of information they receive, and recommendations on when they can call a vote. While actual changes to committee design may take time, our findings have potential to help improve decision-making: for example, have products been approved that should not have been, and is that due to the way committees are designed?

Why did you join the Economics department at Warwick?

The Economics Department at Warwick is an incredibly dynamic, stellar, and vibrant community for economics. In fact, I was already familiar with much of the work being done at Warwick, especially within my research fields. There is world-leading research being done across the Department and in many fields. I can interact with cutting-edge theoretical and empirical, applied and technical research, with amazing colleagues and staff. This dynamism is also present through the ambitious work conducted by both staff and students, active visitor series, opportunities for presentations and conferences, among others.

An essential part of Warwick’s edge, though, is its collegiality: staff and students are incredibly warm, providing an environment that is ideal to collaborate. This allows us to exchange ideas across fields and methods which, I think, really helps break ground on new contributions. I am particularly grateful to all colleagues, support staff and students who have made me feel so welcome. All in all, Warwick is a very exciting place to be in and I am looking forward to contributing!

Nathan Canen is an Associate Professor in the Department of Economics. View his staff profile.