Alternative futures: comment on Terry Moe's "The Revolution in Presidential Studies".

Terry Moe is one of our most influential and well-respected scholars of the presidency. For a long time, he has been a vocal advocate for more a more rigorous and scientific approach to studying the presidency. To further those ends, he has pushed the virtues of rational choice and game theory. His review of the literature in his essay "The Revolution in Presidential Studies" (Moe 2009) details the payoff to presidency scholarship from the adoption of rational choice approaches. Not only have we learned much from his own work, especially his seminal article "The Politicized Presidency" (1985), but also the work of Charles Cameron (2000), Brandice Canes-Wrone (2006), William Howell (2003), David Lewis (2003, 2008), Andrew Rudalevige (2002), and others has left an indelible mark on how we see and understand the presidency. We have all profited handsomely from the contributions of this brand of research.

Yet in his essay, Moe argues for another approach to studying the presidency and strongly critiques the rational choice approach. He cites several limitations of rational choice, including the unrealistic assumptions of optimization and information processing, the centrality of equilibrium when many political behaviors and decisions are made out of equilibrium, and the oversimplification of reality. Moe argues that computational and agent-based models, which allow for greater complexity and which are rooted in more realistic assumptions about human behavior, offer a superior alternative to rational choice. As Moe has been quite prescient and thoughtful in his predictions of the payoff rational choice, we need to take his advice on how to study the presidency seriously. But it is important to note that, at this time, we just do not know whether the direction he is pointing us to take will be fruitful.

On the one hand, it is pointless to criticize an approach without seeing what it produces. At the same time, two criteria are necessary before a research program should be abandoned. First, there must be obvious limitations. Moe discusses the limitations of the rational choice approach. Second, there need to be findings and results that the reigning approach cannot explain. As the rational choice approach to presidency studies is still in its infancy and has not been applied to many important problems or pushed to its limits in areas where it has been applied (e.g., presidential-congressional relations), we do not yet know the full limitations of rational choice as an approach to studying the presidency.

As someone who was trained by students of the Simon-Cyert-March bounded rationality approach, I have a lot of sympathy with Moe's exhortation to move in that direction. While studies in that school have had an immense impact on political science, in particular the public policy subfield (e.g., Kingdon 1984), I am not sure that we can characterize that research program as being progressive. (1) Perhaps the agent- based and computational models will invigorate those studies, lead to testable hypotheses, and increase our store of knowledge.

In the remainder of this essay, rather than engage in a fruitless discussion of whether Moe's prescription is correct, I want to touch on some considerations for the future study of the presidency that Moe did not discuss. First, I think we need to incorporate Moe's institutional understanding of the presidency with studies that look at the public presidency. Like the rational choice approach to studying aspects of the institutional presidency, much time and effort has gone into studying the public presidency, and to great benefit, I would argue. Second, I think that presidency scholars are too parochial in their fixation on studying the presidency. We may profit handsomely from expanding our horizons to the comparative study of political executives. Third, despite advances made in the collection of data on the presidency (and other political executives), we are still empirically impoverished, and we need a different approach to collecting data on the presidency than the one we have so much relied on, that is, lone scholars collecting data for one-shot studies.

Too Narrow a Focus on the Institutional Presidency?

Moe focuses almost exclusively on the presidency as an institution. I, too, think that it is important to view the presidency from an institutional perspective. Any understanding must take into account the institutional character and traits of the presidency, and this, I think, applies equally well to the traditional, pre-FDR presidency as it does to the modern presidency since FDR. Yet there has been another major research program under way on the presidency, one with obvious ties to the institutional presidency, but which barely receives mention in Moe's essay: the public presidency.

Samuel Kernell's (1986) "going public" and Jeffrey Tulis's (1987) "rhetorical presidency" concepts are the seminal ideas motivating much of the research on the public presidency. Moe recognizes Kernell's contribution, and that Canes-Wrone's (2006) work has roots in Kernell's work, but he cites little else of this vein. There has been considerable research on the public presidency in recent years, for instance, by George Edwards (2003, 2009), Lawrence Jacobs and Robert Shapiro (2000), and B. Dan Wood (2007, forthcoming), as well as my own work (Cohen 1997, 2008, forthcoming), in addition to Kernell and Canes-Wrone. Despite eclecticism in research design and data, all of these works are scientific in Moe's sense of focusing on a generic presidential type rather than individual presidents, and testing hypotheses. While other than Canes-Wrone, little of this research develops formal rational choice or game theory models, most of it at least implicitly assumes rational presidents, and thus can be viewed as friendly to the rational choice program.

The public presidency literature is concerned with issues of accountability and representation. Substantively, we learn that presidents try to lead the public, but not always successfully. Edwards (2003) raises a puzzle from this research program for us to grapple with: presidential "going public" has increased in volume, but with decreasing effectiveness. Why would a rational president engage in an activity with so little payoff?

I want to offer two points about the public presidency. First, rather than studying the public and institutional presidencies in isolation from each other, we need to integrate the concepts and research from these two literatures. Second, the new brand of political psychology, and its methodology, has much to offer in understanding the public presidency.

Theoretically Integrating the Public and Institutional Presidencies

We need to integrate our concepts and research on the public presidency with the concepts and research on the institutional presidency to produce a fuller and more complete understanding of the presidency. The only place that I can see the institutional aspects of the presidency discussed side by side with the public presidency is in the modern presidency literature (e.g., Greenstein 1988). But the modern presidency literature is not so much a theory as a description of a moment in the history of the presidency. Proponents of the modern presidency concept stress special circumstances, primarily the confluence of the emergency of the Great Depression with the arrival of Franklin D. Roosevelt, to explain this modernization moment. Then modern presidency scholars list the characteristics that distinguish the modern from the earlier, traditional presidency, such as the president's increased unilateral policy-making capacity, the accumulation of staff resources, the president's role in setting the national policy agenda, the greater involvement of the president in the legislative policy-making process, and the greater public visibility of the president. But many of these characteristics had been developing for decades prior to the 1930s, leading John Woolley (2005) to question how we have studied the historical evolution of the presidency, cautioning that we need to pay more attention to identifying causal mechanisms and operationalizing variables, and that what looks like major shifts in the development of the office may be the result of the accretion of incremental changes along numerous aspects or dimensions of the office. To my knowledge, few have picked up on Woolley's suggestions (but see Woolley and Peters 2008).

Returning to the public and institutional presidencies, we know that they both seemed to arrive or develop at roughly the same time. The modern, post-FDR president is both more institutionally developed and more publicly active. I think that the public presidency is one aspect of the institutional presidency, but the primary variables that institutional presidency theory deals with, politicization and centralization (Moe 1985; Rudalevige 2002), do not appear to be directly related to the public presidency. What are the causal connections between the rise of the institutional and public presidencies? I can see several possibilities. Let me briefly sketch out some causal connections and their logic, merely as an exercise in thinking about the nature of the problem.

First, and to me the most unlikely, is that their historical coexistence is merely coincidental. Second, a third factor may have given rise to both--that is, any relationship between the institutional and public presidency is spurious, reflecting other forces operating on the presidency. For instance, both the institutional and the public presidency may be reflective of social and economic modernization, with modernization forces pressuring government and the presidency to both rationalize (institutionalize?) and democratize (the public presidency).

Third, the institutional presidency may have stimulated or caused the public presidency. For example, part of the logic driving the institutional presidency is control over policy. Presidents centralize and politicize the presidency and its ties to the bureaucracy in part to control policy. But presidents do not possess much in the way of political authority in a separation of powers system, and thus they seek to supplement that deficit with other types of resources, such as public support. Thus presidents seek public support to further their control over policy.

Fourth, the public presidency may drive the institutional presidency. It appears, at least from Tulis's work (1987), that the public presidency, what he calls the "rhetorical presidency," began to emerge before the institutional presidency. What is the logic that leads the imperatives of the public presidency to promote an institutional executive? Hypothetically, electoral (and related partisan) motivations lead presidents to need a supportive public/electorate. Presidents and their parties, in other words, want to gain and stay in office. One way to build support in the public--the prerequisite for staying in office--is to produce policies that voters like. To build those policies, presidents need some input and influence over the policy-making process. The institutional presidency provides presidents with resources to influence policy outcomes.

Admittedly, the foregoing causal mechanisms are quite sketchy, and those presented earlier do not exhaust the possibilities. In building this broader understanding of the presidency, one that incorporates both the institutional and public presidency, we need to go beyond the mere identification of the casual processes that led to the development of each of these two aspects of the presidency to consider how the institutional and public presidency interact and the implications of their interaction for the development of the office and the behavior of presidents. Thus, we need to raise these types of questions: Is pursuit of the public presidency complementary to the institutional presidency? Are there trade-offs and tensions between these two aspects of the presidency? Or do they travel in disconnected worlds, irrelevant to each other? (not very likely to me).

Political Psychology and the Public Presidency

One of the implied messages of Moe's essay is that scholars of the presidency need more than substantive knowledge of the presidency in order for research to progress, a point I heartily agree with. Presidency research requires theoretical and methodological skills, as well as deep substantive understandings of other parts of the political system. The presidency does not sit in isolation from the other branches of government or the larger political system, but interacts with them. Most of the work that Moe cites explicitly takes this interactive approach--Cameron's (2000) veto bargaining is as much about Congress as it is about the presidency, to take one example. We need to recognize explicitly that the presidency is but one branch of government in our Madisonian system of separation of powers and checks and balances. The presidency is not an isolated branch (Jones 1994).

Similarly, the public presidency work that I cited earlier is also as much about the public as it is about the president. (2) Still, most of the best work on the public presidency focuses on presidential behavior, with less attention given to understanding the public response. Other than a few exceptions (e.g., Edwards 2003; Wood 2007), much of the public presidency research seems unaware of the recent advances in political psychology. Political psychology represents another research program that has made important contributions in recent years of potential importance to presidency scholars.

The new political psychology research finds its intellectual roots in Herbert Simon and in Daniel Kahneman and Amos Tversky, and priming is emerging as the most powerful and central theoretical construct among political scientists studying mass behavior. (3) Political psychology also brings with it a powerful methodological tool, experiments, especially embedded survey experiments (but see Gaines, Kuklinski, and Quirk 2007 for a critique of experimentation).

For the most part, the political psychology literature has not been dealing with topics of central concern to presidency scholars, but neither have presidency scholars been using the theory or tools of political psychology in their research very much. Yet this approach offers us one way to deal with Edwards's (2003) puzzle: why are presidents going public so much if it seems so ineffective? Currently, we just do not know very much about how the mass public responds to presidential going public activities. Integrating what we know about going public with mass political psychology may help us better understand the conditions and behaviors that foster or limit presidential leadership of the public.

A recent essay by James Druckman and Justin Holmes (2004) in Presidential Studies Quarterly provides a template for future research. Rooted in priming theory, Druckman and Holmes suggest that we need to look at the content (and other characteristics) of presidential messages to determine their effectiveness in shaping public opinion. With few exceptions (Cohen 1997; Wood 2007), most of the literature on the public presidency does not distinguish characteristics of presidential public messages, such as content, tone, setting, audience, and so on. Druckman and Holmes find that presidential emphasis on foreign affairs seems effective in affecting approval of the president, at least for the case they test (George W. Bush). And they do so with an elegant experiment, confirmed with analysis of survey data.

Where Druckman and Holmes focus on the classic question of opinion change and formation, Jason Barabas (2008) looks at learning. Although Barabas does not employ an experimental design, he uses a sophisticated within-survey/within-subjects comparison to ask whether the public learns about the president's policy proposals from watching the State of the Union address. Like Druckman and Holmes, Barabas relies on recent political psychology as a theoretical foundation, and he finds that the public learns, but not necessarily from watching the address. News coverage seems critical for public learning.

Experimentation is not the only methodology that one can use to explore the mass response to the president. Sung-youn Kim, Charles Taber, and Milton Lodge (forthcoming) use agent-based computational models, which Moe recommends, to compare "motivated reasoning" and "Bayesian updating" models of citizen presidential vote choice. They find that the motivated reasoning model fits the data better than the Bayesian updating model. My larger point, however, is that the public presidency is an important research program in addition to the institutional research program that Moe identifies, that research on the public presidency may profit from the "revolution in political psychology" and that we need to integrate these two perspectives (institutional and public presidencies) for a fuller understanding of the presidency than is currently the case.

A Comparative Politics of the Presidency? (and Separation of Powers Systems)

Puzzles--that is, facts and relationships between variables that we have a hard time explaining and making sense of--are often critical in stimulating major research advances. By and large, puzzles rarely seem to motivate research on the presidency, with few exceptions (Edwards 2003; Canes-Wrone 2001, 2006). But there is a major "comparative politics" puzzle staring at us that presidency scholars have given almost no attention to: presidential systems seem less long-lived than parliamentary systems (see the review in Cheibub and Limongi 2002), but the American presidential system ranks among the most long-lived democratic systems now in existence. Why? What is it about the presidency, and our separation of powers system, compared to others, that has produced such a resilient political system?

To take as one hypothesis, by Matthew Shugart and John Carey's (1992) coding, most of the fragile presidential systems possess a strong executive and a weak legislature, but it appears to me that both the American presidency is moderately strong, while Congress is very strong. Perhaps this combination of strong, constitutionally independent branches, which can protect themselves from each other, is what produces or fosters political system stability. Comparable institutional strength, with the Constitution as one source of that institutional strength and legitimacy, forces each branch to bargain with the other. The rational choice separation of powers literature that Moe cites has focused primarily on the American case. It may be useful to extend the rational choice, game theory approach to alternative separation of powers designs, as proposed here. (4) In the process, we might learn more not only about the functioning of the American design, but about other designs as well, and discover a possible answer to the puzzle of the longevity of the American presidential system. To push the point, rather than explain the longevity of the U.S. system and our presidency as an outlier, an idiosyncrasy, we should be trying to build a theory that explains the performance of the presidency in comparison with that of other similar and different heads of government.

The Impoverished Empirical Description of the Presidency

The foregoing puzzle on the longevity of the American presidency highlights another weakness of presidency studies--empirically, we still cannot describe the presidency very well. Describing the institutional attributes of the presidency and presidential behavior, both across time and in comparison to other executives, is necessary if we are to advance our understanding of the office, and of political executives more generally. Moe states in his essay that "[i]deas can come from anywhere, and good ones can be hit upon by historians, journalists, and empirical researchers just as well as by rational choice theorists" (2009, 714), a sentiment that I agree with, but he also states that "[p]residency scholars have always been good at generating facts" (714), a point that I do not entirely agree with, or at least think this is something that we can do better than we have done.

"Facts" about the presidency come in several varieties. First, there are the many "stray facts," the minutia of the office and the behaviors of individual presidents that to some degree may be no more than clutter. Presidency followers have been very good at generating these types of facts. Obviously, some of these "facts" are more important than others for us to know about. But the facts or empirical observations necessary to test hypotheses about presidential behavior and institutional performance may be more important. I call these types of "facts" or data purposive data, because theory or hypotheses direct us to the types of data that we need in order to test propositions derived from the theory.

Purposive data can be collected in two ways, following either a big science model or the individual scholar approach. Both have their advantages and disadvantages, and we need both types for science to progress, but presidency studies have been primarily of the second type (when presidency scholars even purposively collect data, which I think is all too rare) and not the first, or big science, type. What I mean by the "big science" model is a team of researchers collecting data for the larger community of scholars, and those data become part of the infrastructure of the field, and the data collection effort continues into the future. Continuing to collect data into the future is important for presidency studies because a critical dimension of the office is that it is a historical institution--that is, the presidency has changed, and probably will change, over time. There are several such infrastructure data collection programs in political science, the American National Election Studies (ANES) and congressional roll call studies (e.g., from CQ, but more importantly from Poole and Rosenthal) among the most prominent. The more recent collection of data on the Supreme Court under the direction of Harold Spaeth, made public to the larger research community, has reinvigorated and stimulated research on the Court and judicial policy making. (5) Presidency scholars have relied on these data sources quite heavily, but these data have not been collected with the needs of presidency research in mind.

More commonly, we have lone or small teams of presidency scholars collecting data purposively for a specific research question. There are many of these, and they have been important not only for their data collection and resulting descriptions of aspects of the presidency and presidential behavior, but also for addressing important questions. But there are problems with a subfield that relies only on this type of purposive data collection. Such data collections tend to be one-shot affairs, and they are rarely replicated or built upon. Also, data from such enterprises are not as easily accessible to other scholars, even though we have developed a norm of data sharing that is to be commended.

The institutionalized data infrastructure approach--that is, "big science" data collection--can be and has been criticized for stifling innovation. Scholars may rush to analyze those data because someone else has paid the cost of data collection. Young scholars, with the tenure clock ticking, may see this as a way to speed up their publication track record. Too much energy may be put into some questions at the expense of other, perhaps more important ones. How many more studies of voting behavior or public opinion using the ANES do we need? Should we as a discipline be putting so much intellectual energy into those types of questions at the cost of not applying ourselves to other, equally important problems? To borrow a phrase that Douglas Arnold (1982) coined a quarter century ago, the big science approach may lead to "overtilling."

Yet presidency, or more properly, political executive research, suffers not from having too much of this type of data, but from lacking such a data infrastructure. Presidency research may benefit from several such large-scale, ongoing, institutionally based data collections. Obviously, the documentary record is an important data source, especially for recovering information about the past, but not everything that we want to know about the presidency can come from documents. Interviews with Washington elites and decision makers are important for filling in the gaps of the documentary record and for providing us with the perspective of participants in government. The political science literature is full of informative and necessary studies based on interviews. We would be intellectually impoverished without these.

With regard to the presidency, perhaps no study can compare with Joel Aberbach and Bert Rockman's (2000) several large-scale interviews with Washington elites over the past several decades. From those studies, we have learned much about the presidency and presidential-bureaucratic interactions. Yet being able to interview the top- level government officials has proceeded when Aberbach and Rockman have been able to secure funding, leading only to several waves of such interviews over the past 30 years. While indispensable, we need to think about the future, about systematizing such data collection efforts for the wider community of presidency and other scholars. My vision may be pie-in-the-sky and mere wishful thinking, but to regularize interviews with Washington policy makers, a sample of top-level bureaucrats, members of Congress (or their staff), and White House staffers in real time, perhaps every two to four years, much like the ANES surveys of voters, could provide us with data that we as scholars want and need.

Similarly important is the creation of a historical database that relies on documents, newspapers, and the like, to record institutional attributes of the presidency and presidential behaviors and activities. Along similar lines, it would be useful to have a complete and easily accessible database on presidential approval and other aspects of public opinion on the president. Approval questions have been asked for executives in many countries, which should also be included, using a standardized format, one that is flexible and can be easily manipulated for many research purposes. And it would be useful in designing these data collections to do so with an eye toward collecting similar data in other nations, so that we can truly do comparative research on political executives. I am sure that others have ideas about the types of data they think are important and would be appropriate to collect in this way.

Conclusion

Terry Moe, with his essay "The Revolution in Presidential Studies," has again challenged us as presidency scholars to think about how we study the presidency and suggested a direction that future research should take. I have little quarrel with incorporating bounded rationality theory, along with simulations and agent-based modeling, into how we study the presidency. We may realize big payoffs from doing so.

But it may be risky to put all our eggs in one basket, so in these pages, I have pointed out several other things that I think presidency scholars need to be thinking about for the future of the subfield. Not all of my points are at variance with Moe's prescription. For instance, I see a comparative politics approach and more investment in data collection as completely compatible. Moreover, even my suggestion that we need to integrate an institutional understanding of the office with research on the public presidency can be taken more as a friendly amendment that does not reject the importance of an institutional approach to studying the presidency. Rather, we need to broaden our institutional understanding of the office to incorporate this important side of the presidency.

The larger underlying point of my essay is that the presidency field has been all too parochial. Much presidency research is overly concerned with the presidency and the historical minutia of the office and its occupants, at the expense of seeing the similarities and differences of the presidency to other national chief executives and lower- level executives such as American governors. Much presidency research fails to appreciate the place of the presidency within our separation of powers/checks and balances system--that is, that the presidency is but one actor in a system in which institutional actors were meant to interact in order to produce public policies. Much presidency research seems uninformed by research from political scientists studying other components of the political system (e.g., legislative scholar, public opinion) or by research currents outside political science, research that may bear on our knowledge and understanding of the presidency. Finally, more presidency research needs to speak to larger questions, such as accountability, representation, decision making, and behavior, that animate much of political science and the social sciences more generally. As Moe details, the revolution of rational choice studies of the presidency has moved us in all of these directions. Moe's new prescription may also move us in these directions, but the points I have offered here, I think, will also move us in these directions.

References

Aberbach, Joel D., and Bert A. Rockman. 2000. In the Web of Politics: Three Decades of the U.S. Federal Executive. Washington, DC: Brookings Institution Press.

Arnold, R. Douglas. 1982. "Overtilled and Undertilled Fields in American Politics." Political Science Quarterly 97 (Spring): 91-103.

Barabas, Jason. 2008. "Presidential Policy Initiatives: How the Public Learns About State of the Union Proposals from the Mass Media." Presidential Studies Quarterly 38 (June): 195- 222.

Cameron, Charles M. 2000. Veto Bargaining: Presidents and the Politics of Negative Power. New York: Cambridge University Press.

Canes-Wrone, Brandice. 2001. "A Theory of Presidents' Public Agenda-Setting." Journal of Theoretical Politics 13 (April): 183-208.

--. 2006. Who Leads Whom? Presidents, Policy, and the Public. Chicago: University of Chicago Press.

Cheibub, Jose Antonio, and Fernando Limongi. 2002. "Democratic Institutions and Regime Survival: Parliamentary and Presidential Democracies Reconsidered." Annual Review of Political Science 5: 151-79.

Cohen, Jeffrey E. 1997. Presidential Responsiveness and Public Policy-Making: The Public and the Policies That Presidents Choose. Ann Arbor: University of Michigan Press.

--. 2008. The Presidency in the Era of 24-Hour News. Princeton, NJ: Princeton University Press.

--. Forthcoming. Going Local: Presidential Leadership in the Post-Broadcast Age. New York: Cambridge University Press.

De Figueiredo, Rui J. P., Jr., Tonja Jacobi, and Barry R. Weingast. 2006. "The New Separation of Powers Approach to American Politics." In The Oxford Handbook of Political Economy, eds. Barry R. Weingast and Donald Wittman. New York: Oxford University Press.

Druckman, James N., and Justin W. Holmes. 2004. "Does Presidential Rhetoric Matter? Priming and Presidential Approval." Presidential Studies Quarterly 34 (December): 755-78.

Edwards, George C., III. 2003. On Deaf Ears: The Limits of the Bully Pulpit. New Haven, CT: Yale University Press.

--. 2009. The Strategic President: Persuasion and Opportunity in Presidential Leadership. Princeton, N J: Princeton University Press.

Gaines, Brian J., James H. Kuklinski, and Paul J. Quirk. 2007. "The Logic of the Survey Experiment Reexamined." Political Analysis 15 (Winter): 1-20.

Greenstein, Fred I., ed. 1988. Leadership in the Modern Presidency. Cambridge, MA: Harvard University Press.

Howell, William G. 2003. Power Without Persuasion: The Politics of Direct Presidential Action. Princeton, NJ: Princeton University Press.

Jacobs, Lawrence R., and Robert Y. Shapiro. 2000. Politicians Don't Pander: Political Manipulation and the Loss of Democratic Responsiveness. Chicago: University of Chicago Press.

Jones, Charles O. 1994. The Presidency in a Separated System. Washington, DC: CQ Press.

Kernell, Samuel. 1986. Going Public: New Strategies of Presidential Leadership. Washington, DC: CQ Press.

Kim, Sung-youn, Charles S. Taber, and Milton Lodge. Forthcoming. "A Computational Model of the Citizen as Motivated Reasoner: Modeling the Dynamics of the 2000 Presidential Election." Political Behavior.

Kingdon, John W. 1984. Agendas, Alternatives, and Public Policies. Boston: Little, Brown.

Lewis, David E. 2003. Presidents and the Politics of Agency Design: Political Institutions in the United States Government Bureaucracy, 1946-1997. Stanford, CA: Stanford University Press.

--. 2008. The Politics of Presidential Appointments: Political Control and Bureaucratic Performance. Princeton, NJ: Princeton University Press.

Moe, Terry M. 1985. "The Politicized Presidency." In The New Direction in American Politics, eds. John E. Chubb and Paul E. Peterson. Washington, DC: Brookings Institution.

--. 2009. "The Revolution in Presidential Studies." Presidential Studies Quarterly 39 (December): 702-25.

Rudalevige, Andrew. 2002. Managing the President's Program: Presidential Leadership and Legislative Policy Formulation. Princeton, NJ: Princeton University Press.

Shugart, Matthew S., and John M. Carey. 1992. Presidents and Assemblies: Constitutional Design and Electoral Dynamics. New York: Cambridge University Press.

Tulis, Jeffrey K. 1987. The Rhetorical Presidency. Princeton, NJ: Princeton University Press.

Walker, Stephen G. Forthcoming. "The Psychology of Presidential Decision Making." In The Oxford Handbook of the American Presidency, eds. George C. Edwards III and William Howell. New York: Oxford University Press.

Wood, B. Dan. 2007. The Politics of Economic Leadership: The Causes and Consequences of Presidential Rhetoric. Princeton, NJ: Princeton University Press.

--. Forthcoming. The Myth of Presidential Representation. New York: Cambridge University Press.

Woolley, John T. 2005. "Drawing Lines or Defining Variables: Studying Big Changes in the Presidency." Paper presented at the Annual Meeting of the American Political Science Association, September 1-4, Washington, DC.

Woolley, John T., and Gerhard Peters. 2008. "A New Data Set on Forms of Presidential Unilateral Action." Paper presented at the Annual Meeting of the Annual Meeting of the American Political Science Association, August 28-31, Boston.

JEFFREY E. COHEN

Fordham University 

(1.) Since the publication of Kingdon's book, many people have adopted his terminology, and there have been some conceptual critiques, but it appears to me that, for the most part, what we have had is an application of Kingdon's stream model to various policy areas--that is, an accumulation of policy case studies using Kingdon's approach to describe that policy, but little testing of propositions that can be derived from his theory.

(2.) We need to recognize, too, that presidents compete with other political leaders, especially members of Congress, when attempting to influence the public. This competition rarely appears in studies of the public presidency, and may be viewed as another arena of the separation of powers being played out.

(3.) Most of what I focus on here is the literature as it relates to mass political psychology. There is another literature that looks at the impact of presidential psychology on the executive's decision making. See the review of this literature in Stephen Walker's forthcoming essay. Notably, Walker suggests that both the psychological and the rational choice approaches have much to offer in combination for understanding presidential decision making: "Even if cognitive, motivational, and emotional biases structure the diagnostic stage of the decision-making process by specifying goals and eliminating some alternatives from the menu for choice, it is still likely that decision makers use rational choice procedures to make their choices" (551).

(4.) Such a movement from a rational choice perspective is already under way, although it is still preoccupied with the American case (see De Figueiredo, Jacobi, and Weingast 2006).

(5.) See http://www.cas.sc.edu/poli/juri/sctdata.htm.

Jeffrey E. Cohen is a professor of political science at Fordham University. His books include The Presidency in the Era of 24-Hour News and Going Local: Presidential Leadership in the Post- Broadcast Age.

AUTHOR'S NOTE: I wrote this paper while I was a visiting senior research scholar, 2008-2009, at the Center for the Study of Democratic Politics, Woodrow Wilson School, Princeton University. I want to thank Charles Cameron, Brandice Canes-Wrone, Milan Svolik, and Jessica Trounstine for helping me clarify the thoughts presented here.