Democracy in today's world is synonymous with representative systems. Yet the majority of representative democracies institutionalizes mechanisms of "direct" democracy: referendums, plebiscites, citizens' initiatives, or recalls. The persistence of these institutions suggests that giving citizens the ultimate say in some matters of national public policy adds to the legitimacy of representative systems. Some recent referendum results -- the Brexit vote in the UK and the disapproval of peace accords in Colombia -- have signaled for many the risks of direct democracy. In this paper I first explore arguments in favor of limited direct agency for voters suggested by earlier theorists of representative government. I then explore the strategic reasons why real-world political leaders sometimes decide to delegate important decisions to voters. In some instances, though probably rarely, politicians decide to hold referendums for reasons that theorists would approve of.

Why does the public support or oppose military alliances? Although public backing for promises to defend other countries shapes the credibility of alliance commitments by democracies, we know little about the foundations of public opinion towards alliances. In particular, existing survey evidence cannot determine whether alliance attitudes are the top-down result of elite cues, or a bottom-up result of individual concerns and perceptions of alliance obligations and partners. In this article, I identify three potential determinants of public opinion towards alliances: elite cues, individual considerations, and alliance characteristics. I then use two conjoint survey experiments to assess the relative importance of these factors for public attitudes towards forming and maintaining international alliances.

For decades scholars have thought about the climate change problem as one that involves collective action on a global scale. That logic has animated the creation of treaties like the UN Framework Convention and the Kyoto Protocol that have failed spectacularly in having any real impact on the underlying problem. The Paris Agreement arrived with the great hope it would change everything; so far, its impacts have been modest too. This talk will explain what we know, as international relations scholars, about why the diagnoses of troubles with climate cooperation have not led to real solutions and where more effective strategies, building on the Paris Agreement, could be forged. It will be based partly on work done with the UK government (which hosts the next Conference of the Parties) and a forthcoming book (Princeton University Press) with Charles Sabel on Experimentalist Governance.

What are the downstream political consequences of state activity explicitly targeting an ethnic minority group? This question is well studied in the comparative context, but less is known about the effects of explicitly racist state activity in liberal democracies such as the United States. We investigate this question by looking at an important event in American history—the internment of people of Japanese ancestry during World War II. We find that Japanese Americans who were interned or had family who were interned are significantly less politically engaged and that these patterns of disengagement increase with internment length. Using an identification strategy leveraging quasi-random camp assignment, we also find that camp experience matters: those who went to camps that witnessed intragroup violence or strikes experienced sharper declines, suggesting that group fragmentation is an important mechanism of disengagement. Taken together, our findings contribute to a growing literature documenting the demobilizing effects of ethnically targeted detention and expand our understanding of these forces within the U.S.

Credit card debt is increasingly common among poor and inexperienced borrowers – thus de facto a financial inclusion product. However, it remains relatively under-studied. We use detailed card-level data and a product that accounted for 15% of all first-time formal loans in Mexico and show that default rates are high and ex-ante unpredictable for new borrowers – suggesting an important role for ex-post contract terms in limiting risk. However, using a large nation-wide experiment we find that default is unresponsive to minimum payment increases, a commonly proposed policy remedy. We provide evidence that the zero result is driven by the offsetting effects of tightened liquidity constraints and lower debt burdens. Surprisingly, we also find muted default responses to large experimental changes in interest rates – suggesting a limited role for ex-post moral hazard in our context. Finally, we use job displacements to document large effects of unemployment on default, highlighting the centrality of idiosyncratic shocks as a barrier to the expansion of formal credit among poorer populations.

Why, despite high rates of reported political discussion, do so many Americans dislike talking about politics? And how do the mixed considerations people hold about discussion affect the way that they communicate? We argue that we need to consider the psychological experience of political discussion as navigating a social process that is rife with potential challenges to our sense of self and our relationships with others. Variation in the cognitive resources of political conversation, such as interest or knowledge, or in instrumental goals related to learning and persuasion cannot fully explain people’s motivation to seek or avoid discussion, although considerations related to information certainly are part of the story. Discussion is an inherently social behavior, and we argue that without assessing the social factors influencing the decision to talk about politics, we can’t fully understand who talks about politics, with whom, under what conditions, and with what consequence. 

This book is an effort to open the lid on the processes that lead up to a political discussion and the implications of the conversations that do happen. Our approach is to build on what we already know about political discussion, focusing on the gaps in our knowledge resulting from untested assumptions and limited methodologies in previous work. We apply new measurement techniques in order to better understand the decision-making processes that lead to the initiation of discussion, the nuances of the interactions that do occur, and the consequences of those conversations on a wide set of political and social outcomes.