Presidents’ abilities to connect with the public are of utmost political importance. As the focal leader of the nation, presidents can leverage their unique connection to this nation-wide constituency to influence their negotiations with the legislative branch. In pure presidential systems, the constitutional separation of origin and survival of the political executive demands constant negotiation and compromise across independent branches of government, incentivizing the president to rely on this unique connection with the public.

Modern American presidents live and die by their connection to the public: their electoral campaigns take root years in advance, and many attempt to maintain said political momentum with ongoing direct public appeals throughout their administration. Whether it’s FDR’s radio broadcast Fireside Chats to Donald Trump’s ubiquitous use of social media, leveraging public support has become a tool in the president’s arsenal to strategically wield when deemed necessary.  Indeed, many scholars and political spectators attribute President Obama’s campaign success to his effective use of social media and then note the continuance of this strategy of direct public appeals throughout his presidency, taking the form of speeches and weekly YouTube addresses throughout his administration. Direct public appeals, so the story goes, enable U.S. presidents to apply indirect pressure on members of Congress, thereby improving the chances that the presidents’ preferred policy would be adopted into law.

Although the notion of ‘going public’ has its origins in U.S. presidency, we have little sense of how direct appeals to the public fit into the broader portfolio of presidential powers. Our research situates presidents’ direct public appeals in the broader portfolio of comparative presidential powers. Rather than construe populism and presidents’ plebiscitarian orientation as a personality trait or leadership style, we consider how a president’s propensity to appeal to the public may vary in response to changes in the bargaining environment, which may vary both across countries and over time as a function of institutional, personal and political factors. Our forthcoming article inPresidential Studies Quarterly shows that the frequency of presidents’ public appeals varies with both their partisan support in the legislature, their status as a newcomer to the political system, and electoral and legislative institutions. Further, we also debut an original dataset and archive that is available to the public, such that other scholars may also investigate this sort of question in future research.

The Presidential Speeches of the Americas (PSA) is a dataset and archive of appearances and speeches made by 24 presidents across 18 pure presidential systems of the western hemisphere. These data contain the records of presidents’ speeches and public appearances as advertised on the official websites of the presidency, most of which contain the transcript of the presidential address. Our aim was to collect as much information as possible, harvesting presidential speech archives for as long as they were made available online. Most sitting presidents maintain an online archive of presidential activities and speeches, and in several countries online archives were also available for previous presidential administrations through the WayBack Internet Archive. An overview of the data contained in the PSA dataset is shown in Table 1. This dataset and archive include records of (and in most cases transcripts of) more than 12,500 presidential speeches, made by 24 presidents in 18 pure presidential systems throughout the western hemisphere. It is available to the public, may be found on the website https://www.psa-dataset-archive.com.

Our research team brings together a diverse group of faculty, graduate and undergraduate students from the College of Social Sciences and Public Policy. Co-authors Amanda Driscoll (Assistant Professor of Political Science) and Alexandra G. Cockerham (now Assistant Teaching Professor of Interdisciplinary Social Sciences) were inspired by a shared interest in comparative democratic institutions, and began collaborating when Professor Cockerham was a Ph.D. candidate in the Political Science department. Together, they extended one theoretical model of presidential power in the United States, deriving the logical expectations for alternative institutional and political environments for the pure presidential systems throughout the western hemisphere. Undergraduate Joan V. Joseph (now graduate student at Massachusetts Institute of Technology) joined the project in the summer of 2016, and was centrally responsible for collecting and organizing the presidential speeches dataset as part of her research internship for the Research Intensive Bachelor’s Certificate in the Department of Political Science. As the project evolved, we were elated to work with multiple undergraduate research assistants, who aided in the coding and classification of the presidential speeches as part of both the Research Intensive Bachelors Certificate Program and the Undergraduate Research Opportunities Program at Florida State University. In sum, our resulting publication, dataset and archive represent the fruits of collaborative efforts of many members of the COSSPP community.

We set out to fill an important lacunae in the research on comparative presidentialism, to systematically consider how presidents’ direct public appeals serve as one resource among many that presidents may use to advance their policy agendas. To that end, we also introduce and publicize a new dataset and archive of presidential speeches, the Presidential Speeches of the Americas dataset and archive. Our statistical analysis of a subset of the PSA data suggests that presidents’ direct appeals to the public might serve as a substitute for other sorts of presidential powers, either those derived from their support in the legislature, or those granted to the executive in constitutional texts. These results underscore the advantage of considering ‘going public’ in a comparative perspective, wherein variance in institutional and partisan support can be empirically considered.

For additional information, or to find our forthcoming research at Presidential Studies Quarterly, please visit the PSA Dataset and Archive website at https://www.psa-dataset-archive.com.

The authors would like to thank FSU undergraduates  Lauren Pentrack, Ashley Ravins, Deanna Rodriguez, and Harrison Weeks for their excellent research assistance.