The purpose of this article is to explain the process of large-scale policy development in a presidential administration by using an analytical tool presidency scholars have tended to neglect--organizational learning, a branch of organization theory.(1) Given the nature of the presidency in the political system, with its contracted time horizons, unsurpassed media attention, and high public expectations for success, both the individual occupying the office and those who serve at his pleasure (i.e., staff and most political appointees) are in the position of having to "learn as you go." That is to say that they are not often afforded the luxury of systematically studying the methods by which a task is approached and disposed of, making informed decisions as to its efficacy, and prescribing carefully constructed alternatives for achieving that same task again. These methods are available to other organization inhabitants but often are denied presidents and their administrations. Indeed, one analyst of presidential policymaking, in offering suggestions for coping with what he calls a "no-win" presidency, argues explicitly that "learning must wait."(2) Presidents do learn, but it often is on an ad hoc basis, with policy and political consequences of change uncertain and sometimes risky. A central task for research on presidential organization is to understand the conditions and processes conducive to learning, be they systematic or ad hoc, and to explain the consequences of learning or the failure to learn. This article examines organizational learning in President Jimmy Carter's administration to evaluate the usefulness of a learning perspective for understanding presidential organization and its role in the policy process. A collateral interest is to explain Carter's development of a comprehensive energy program from an organizational perspective.
The structure of this article is as follows. First, I briefly review the arguments for analyzing the presidency as an organization. The fact that the presidency is occupied by individuals, each with varying approaches to the office, does not mean that it lacks continuity in important areas, particularly staff structures and functions. Next, I propose an analytic framework for understanding policymaking, drawing on insights from the organizational learning literature. I then argue that this framework is useful for studying presidential policy choices, focusing specifically on information acquisition and use, integrating policy and political imperatives. Next, I employ this framework to describe Carter's choice to drastically alter his mode of policymaking when developing his comprehensive energy program. Because this is an example of a president explicitly changing his approach to developing policy in midstream, I derive lessons for analyzing presidential decisions. In the final sections, I assess the applicability of organizational learning to presidential administrations and suggest refinements to be used in future research. Thus, this article constitutes a modest effort to advance our understanding of the way in which Carter was able to learn on the job, to help us understand what lessons the Carter experience might hold for subsequent presidents, and (most important) to refine and adapt organizational learning theory to the study of the presidency.(3)
Presidency as Organization
To promote the proposition that organization theory can advance our understanding of the presidency is to beg the question of whether the presidency itself is an organization. If it is not, then the application of a framework derived from organization theory is useless. But if the presidency can be rightly conceptualized as possessing entities or traits that continue from one administration to the next, then it can be argued that it exhibits characteristics of an organization and can be studied as such.(4)
Certain aspects of the presidency, particularly several organizational entities, have become institutionalized over time.(5) This contributes to the ability of the analyst to think of stable organizational arrangements across administrations, the sum of which make up the bulk of the institutional presidency. For a disconnected set of tasks and individuals who perform those tasks to constitute an organization in the sense that those tasks become routine, identifiable, and stable over time, it must be institutionalized. I define institutionalization as the temporal development of stable patterns of interaction based on formalized rules, laws, customs, and rituals. This is similar to and builds on the way in which Matthew Dickinson describes institutionalization as "the positions and related incentives [that] are grounded in statute, executive order, or some other formal foundation."(6) Once these positions become stabilized or institutionalized, they can be identified as traits specific to an organizational entity within the larger context of the Executive Office of the President. Indeed, since at least the Reorganization Act of 1939, certain functions have become more or less permanent features of the modern presidency. Expectations that the president will devise and submit a budget, manage an expanding executive branch, provide direction and coherence in domestic policy, and propose legislation aimed at fine-tuning the economy have contributed to an increased reliance on the statutory entities of the Office of Management and Budget (OMB), the Office of Policy Development (hereafter referred to as the Domestic Policy Staff [DPS]), and the Council of Economic Advisers (CEA) for the purposes of carrying out these tasks. A president can use them as much or as little as he wishes, but he would be loathe to get rid of them entirely. Even if a president wanted to abolish one or all of them, the need to meet or exceed expectations heaped on the office would make it politically and administratively dangerous for him to do so. This illustrates what Hugh Heclo terms the "deep structure" of the presidency, referring explicitly to those elements in the presidency that remain over time and more or less intact from one administration to the next. These organizational structures remain in place in spite of (and not necessarily because of) the variations in managing the office that are attributable to the personal characteristics and idiosyncracies of the individual president occupying the office at any one time.(7)
The office of the presidency also has perquisites of power that remain largely constant over time, although with some variation. Regardless of the degree of variation, the office carries powers independent of those who exercise them. As Martha Feldman writes,
The office [and] the person have power. After all, Ronald Reagan was charismatic before he became president. He did not, however, have the power to affect public policies in the ways that he did and the degree to which he did before he occupied the office of the president. Furthermore, other presidents have been charismatic, perhaps equally so, and they have not had the powers that had accrued to the presidency over the years that gave Reagan ... so much control over appointments. Therefore, we need to separate the person from the office.(8) Taken together, these perspectives argue strongly that the presidency can be conceptualized as an organization separate and distinct from the president who occupies the office. Whereas the individual certainly is important for understanding the institution, the organizational entities existing in the White House, particularly those of the DPS and other institutionalized structures, provide an important basis for studying the presidency as an organization and, therefore, are amenable to analysis based on various branches of organization theory. This is especially true when we consider that the White House office, particularly the individual support staffs such as the DPS, is arranged hierarchically, with the president at the apex, the director below him, and the various assistants and staff aides distributed down through the remainder of the organization. In the next section, I develop an analytical framework for explaining presidential learning as it pertains to its organizational foundations.
Organizational Learning and the Presidential Organization
In this section, I briefly describe what organizational learning is about and what it might suggest for the study of the presidency as an organization. In a cogent analysis of agency and organizational learning, Julianne Mahler writes, "In general terms, organizational learning refers to the capacity of organizations to interpret and use information they have about their results to change ways they go about their work. That is, they learn over time from experience."(9) Thus, there is a temporal dimension to learning that alters the character of the task or process to be changed. Even though learning does not necessarily enhance effectiveness or increase the likelihood of policy or political success, it does become part of the culture of the organization. As Mahler adds,
If ... new ways of thinking about how to be effective are absorbed into formal procedures, rules, routines, informal communities of practice, the collective memory, and the agency culture, these new ideas are learned by the organization and become part of the organizational institution. This last step establishes the learning as organizational.(10) Conceptual and Methodological Issues in Organizational Learning
Conceptually, learning theory gives us a starting place to think about the dynamic processes of organizations, and there are at least a few ways in which to think about organizational learning, especially as it pertains to the presidency. The first mode of learning is trial and error, which is the process of taking a particular action, receiving feedback based on the action, and interpreting that feedback. The second is imitation, which is gaining information from...