Leaders Are Not Alone: The Role of Advisors and Bureaucracies
President Truman had a sign on his desk in the Oval Office that read, “the buck stops here.” He referred to its meaning in his farewell address in January 1953, saying that the “greatest part of the President’s job is to make decisions—big ones and small ones, dozens of them almost every day. . . . The President—whoever he is—has to decide. He can’t pass the buck to anybody. No one else can do the deciding for him. That’s his job.”
What Truman referenced was that he had the ultimate responsibility for U.S. foreign policy. In his view, others in government could “pass the buck” to someone else up the chain of command, but once on the desk of the president, a decision had to be made. His statement nicely expresses two interconnected elements of political decision making: one, he implies that a single person bears the ultimate responsibility for making foreign policy decisions and two, that policy making is conducted through hierarchical organizations. Is this always the case? Let’s examine each point in turn. Although it is tempting to accept Truman’s contention that a single person bears the ultimate responsibility for foreign policy decisions, consider once again the story of the Argentine junta and the Falklands/Malvinas crisis. The junta made a joint decision. Later, President Galtieri’s surprise announcement after the Argentine invasion of the islands altered the group’s decision in a very public manner that was not easy to retract. A public announcement of this nature was probably the only way that Galtieri could override a decision made by the group. The expectation was that they would make decisions together, meaning that the junta expected to govern as a collective of equals. This junta’s expectation was not unique. Decision making groups may be of vastly different sizes, be less or more formally structured, and bear different names—junta, cabinet, coalition, parliament, etc.—but all make decisions in concert. That does not mean that Truman’s assessment was wrong. He did make the important decisions that shaped the latter days of World War II and the immediate post war period. More generally, American presidents do have the ultimate responsibility for foreign policy decisions. This is not true in all countries and decision making situations, as the Argentinean example shows. Whether one person or a collective of multiple individuals or even multiple organizations is responsible for foreign policy decision making depends on the structure of the institutions of a specific society’s government. Note that whether or not one individual bears the ultimate responsibility for foreign policy decision making does not depend on whether that country is a democracy. In fact, in the examples cited here, the democratic country (the United States) has one person who is the ultimate decision maker, whereas Argentina was at the time of the Falklands/Malvinas crisis a nondemocratic country with a group as the ultimate decision maker.
Additionally, within one country foreign policy decisions can be made by different decision units at different times or concerning different types of issues. An ultimate decision unit is defined as the person or the group who are in a position not only to make a foreign policy decision but also to prevent any other entity within the government from explicitly reversing that decision. Especially important with regard to the first element of this decision is that the person or group can use the resources of the government, such as its military, to enforce their decision. For instance, during the Falklands crisis, Prime Minister Thatcher’s decision to send the military to retake the islands was not easily reversed by any other person or agency within the British government.
In sum, determining who has the ultimate power to decide is not simply a function of the type of government but depends on identifying whether a single individual or a group has the ultimate authority to make a foreign policy decision. Making such a determination depends on substantive knowledge about the government in question. The bottom line is that the ultimate decision maker is not always a single individual, as Truman noted with regard to his own situation as President of the United States Truman’s farewell address also implied that foreign policy is made through hierarchical organizations. Another part of the previous quotation reads, “The papers may circulate around the Government for a while but they finally reach this desk. And then, there’s no place else for them to go.” Truman portrayed his office as situated at the top of the hierarchy and as the last stop in the decision making process. This reflected the way he organized his White House and communications with various departments. However, not all U.S. presidents, and certainly not all leaders, strive for this type of streamlined communications. Some leaders purposely build multiple channels of information into their advisory systems. Take, for instance, Emperor Haile Selassie, who ruled Ethiopia from 1930 until 1974. He had joined the imperial court in the capital city of Addis Ababa as a teenager. There, he found himself surrounded by the constant intrigues that were part of political life. In Ethiopia at the time, there were no formal institutions of government, and rule revolved around personal authority and loyalty to the Emperor.6 Although Haile Selassie sought to modernize his country, especially after World War II, the political system continued to revolve around personal loyalty to the Emperor. All his ministers reported to him on a regular basis, to ensure that Haile Selassie was fully informed. These ministers had every incentive to report everything the Emperor might possibly wish to know because he maintained multiple channels of information. If a minister neglected to tell him something, the Emperor would surely find out about it from someone else.
By telling the Emperor himself, each minister maximized control over how the affairs within his department were portrayed. Besides, ministers would be reprimanded and regarded as less trustworthy if they failed to inform the Emperor—something that would not be helpful to their political careers, which depended on remaining in the Emperor’s favor. It is not surprising, then, to know that Haile Selassie is characterized as a masterful politician who manipulated others with such great skill that “he sometimes appears to be a master of marionettes, moving in a mysterious way to determine the actions of the lesser individuals who surround his throne.” Interestingly, the description of Emperor Haile Selassie’s court shares much in common with the organization of the White House under President Truman’s immediate predecessor, Franklin D. Roosevelt. Under Roosevelt, the executive branch deliberately included overlapping jurisdictions, which led to conflict between departments, but also ensured that the President received information on policy problems from multiple perspectives. In addition, Roosevelt would also occasionally contact individuals working within each of the departments to obtain “independent advice and information.” Roosevelt thrived on the political conflict he thus created around him and “manipulated the structure of relationships among subordinates in order to control and profit from their competition.” Although no one has described Roosevelt as a “master of marionettes,” the institutional organization (or lack thereof) of Emperor Haile Selassie’s court and Roosevelt’s White House are remarkably similar. So was both men’s comfort with political intrigue, or the political game. In contrast, Truman’s White House was far more hierarchical.
A leader’s personality is likely to affect how she or he organizes the executive. Some leaders gain insight from hearing their advisors debate issues in their presence, while others like to ponder the policy options their advisors provide to them in solitude. Some leaders are intent that their preferences shape policy, whereas others want policy choices to reflect a consensus among various viewpoints. It also matters whether a leader wishes to be actively involved in foreign policy making, actively seeking out information and shaping the policy options, or, conversely, prefers to rely on the expertise of trusted advisors who help define issues and gather information. Leaders are more likely to actively seek out information when they feel knowledgeable about (and comfortable with) foreign affairs and when they trust the bureaucracy.
In sum, there are many aspects of a leader’s personality that influence how that leader treats information—and how much information she or he requires. This has implications for the organization of an effective advisory system.
Just like the largest part of an iceberg rests below the surface of the ocean, much of that
advisory system is not readily visible. The tip of the iceberg consists of the leader and her or his immediate advisors. The rest of the iceberg is the socalled permanent bureaucracy on which leaders rely for the information that shapes their policies and the implementation of their decisions. Although we know the bureaucracy is there, we are not always sure of what those working within the bureaucracy do or how their work influences foreign policy decisions.
The people with whom a leader surrounds her- or himself matter. It is through the leader’s conversations with the immediate circle of advisors and associates that policy decisions take shape.12 Although Truman portrayed himself as the final arbiter in the decision making process, he would have admitted that his advisors shaped his policies in significant ways. There is no such thing as a perfect advisory system: each system has its own pitfalls. Leaders function best if the advisory system suits their personality, and they always need to guard against the potential problems of the specific advisory system they choose. Let’s take a look at some of the ways in which leaders have structured advisory systems. One categorization scheme, derived from the organization of the White House under several U.S. presidents, identifies three different approaches to organizing the advisory system: formalistic, competitive, and collegial. The following paragraphs discuss each in turn. As you read them, consider how a leader’s personality might predispose her or him to organize the executive along these, or perhaps yet other, lines.
The formalistic approach to organizing the executive emphasizes a hierarchical structure with a clear chain of command. This does not mean that the executive office of every leader who has employed this type of organization could be depicted with the same organizational chart. Rather, it means that leaders who employ this type of organizational structure endeavor to create an orderly decision process. Advisors each provide the leader with information on those aspects of a problem that is within their area of expertise and under the jurisdiction of their departments. These advisors, in turn, obtain information and advice from the individuals who work in their department or agency. Some leaders will want each department head to provide them with advice, while the leader synthesizes the information, as did Truman. More recently, U.S. presidents have employed their White House staff to synthesize information and advice for them. Other variations are possible within the scope of the formalistic model.
They all share in common that the flow of information and the spheres of competence of various advisors are clearly delineated. The emphasis is on analysis and on making the “best” decision possible. Although such a hierarchical structure appears orderly and efficient, it may not be possible for a leader who sits at the top of such an advisory system to know whether information has been left out or distorted as it made its way up the organizational ladder, because leaders who employ this type of organizational structure seldom, if ever, circumvent the official chain of command. This drawback of the formalistic approach to organizing the executive is the strength of the competitive approach. As the example of Ethiopian Emperor Haile Selassie shows, the leader who organizes the executive along these lines actively uses multiple channels of information. There is little cooperation between advisors in this type of advisory system. Instead, all are keenly aware that the leader can access information from a variety of sources, including the subordinates of the department heads, which creates an atmosphere of competition and conflict. Advisors all vie for the leader’s ear and rush to be the first to convey new information, either so they can present the information in a way that portrays their department favorably or so they can play a crucial role in the framing, or representation, of the policy problem. As a result, advisors are likely to present partial, incomplete, or biased information. Leaders arrive at a complete, or at least balanced, view of issues as a result of reconciling these various viewpoints. The internal competition can be hard on the leader’s advisors and may result in high staff turnover. It also demands a lot of the leader’s time and attention. When used well, it does place that leader at the hub of an extensive informational network. In doing so, this approach can generate creative solutions, because there is a confluence of many different ideas and viewpoints at the center of government. Furthermore, this system is also very good at generating solutions that are feasible: ideas are modified and tempered as a result of the interplay with other ideas, as well as the need to defend ideas in debate with others. Hence, the competitive system, if managed well, can generate solutions that are at once creative, politically acceptable, and bureaucratically doable.
There are leaders who see the advantages of the ability to ponder multiple perspectives and divergent information but who are not comfortable with the high level of internal conflict that the competitive approach is likely to generate. Yet they also wish to avoid the potential loss or distortion of information that is inherent in the formalistic approach. A third alternative takes advantage of the benefits that flow from obtaining a multiplicity of views but endeavors to cultivate a spirit of teamwork rather than competition. This alternative is called the collegial approach. As in the competitive advisory system, the leader sits at the center of an extensive informational network. Advisors do not provide their information to the leader individually but debate policy options with one another as a group. The objective of such discussions is to achieve a frank exchange of ideas—but without the conflict that accompanies the competitive system—and arrive at innovative policy proposals. The leader communicates directly with advisors but at times also reaches out to the subordinates of department heads and obtains information outside of the formal chain of command.
In the collegial approach, the emphasis is on teamwork rather than competition. Of course, differences of opinion can always spin out of control, and advisors may become competitors. On the other hand, there is also the risk that the team starts to think too much alike and that the open exchange of ideas turns into too much mutual agreement. The difficulty in making the collegial approach work is that it requires a delicate balance of diversity of opinion, mediating differences, and fostering a team spirit. Not all leaders have the skills to manage the interpersonal relations between their advisors to successfully maintain a collegial system across time. Each of the three approaches to the advisory system has its own advantages and disadvantages. The four questions that are implied in the description of the pros and cons of each approach to organizing the executive:
- How likely is it that the advisory system will distort information? The formalistic system has the highest risk of doing so.
- Is the leader exposed to a lot or to very little conflict, either substantive or interpersonal? The formalistic system seeks to shield the leader from both types of conflict, whereas the competitive system exposes the leader to both. The collegial system stakes out a middle ground by guiding members of a team to debate the issues.
- How responsive is the advisory system? Does it emphasize optimal or feasible solutions? Here, again, the formalistic and competitive systems are opposites—emphasizing optimality and feasibility, respectively—with the collegial approach staking out the middle ground.
- What conditions are required for thorough consideration of alternatives? How well each works is likely to depend on whether the leader can effectively manage the chosen arrangement.
Each system has the capacity to do well and also runs the risk of performing abysmally. Above, we introduced Emperor Haile Selassie of Ethiopia, whose advisory system approximated the competitive approach. This masterful politician has been described as an “African Louis XIV,” a reference to the centralization of power that took place during his rule. He was born in 1892 as the son of a cousin of his predecessor, Emperor Menilek.
Succession to the throne was not predetermined by birth order and family ties (the way it is in “modern” monarchies), but sons of cousins were certainly not the most likely successors to the throne. Haile Selassie’s ascension to the throne was itself a testament to his political skill. How, then, was it possible for Haile Selassie’s government in the early 1970s to deny the existence of a widespread famine within the borders of the country? The government claimed that international media reports of the famine were “misinformed and exaggerated” and continued to deny its existence until May 1973. Is it possible that provincial administrators “obscured the magnitude of the tragedy”? Is it possible that officials in the capital initially “did not even inform the emperor”? Could it be possible that the Emperor did not know of the immense tragedy that was unfolding within his own country? This would seem strange given the Emperor’s multiple channels of information that kept him “in touch with the least event,” although he did occasionally use “the claim of ignorance” to be able to deny responsibility for specific policies. On the other hand, the entire system operated through personal ties and connections, and there was no framework for systematically analyzing information. What advisor would, in such a political system, risk being the bearer of bad news? It would obviously displease the Emperor to hear of such problems within his country.
Of course, it is difficult to establish conclusively what the Emperor did or did not know at a specific point in time. Although he had always aimed to be the fully informed central hub of government, by the early 1970s the Emperor was in his early eighties, and some described him as “too old and senile” to lead effectively. He worked fewer hours and was less actively involved in the policy making process than during his younger years. Although he may have managed a competitive advisory system well at one point in time, it appears that Haile Selassie was less skillful at manipulating his advisors as he aged. In fact, it appears that he may have become the captive of his immediate advisors, who took an active role in the day-to-day administration and who carefully controlled access to him. What became visible at that point was the weakness of a fluid system based on the ability of the leader to manage (and mediate) conflict. Just as Roosevelt’s administration “spawned inefficiency” when he could not personally manage the conflict between individuals with overlapping responsibilities, when Haile Selassie’s attention to detail diminished in his later years, his advisory system suffered. With little formal (bureaucratic) structure, nor an emphasis on analysis, his advisory system increasingly distorted information. Because it was built on interpersonal connection, there were no mechanisms to prevent, or correct for, such distortion. Under these circumstances, it is entirely plausible that no one informed the Emperor of the famine: doing so would have weakened the position of the advisors closest to him—precisely the ones who also controlled access to him. The organization of the advisory system has consequences for decision making. In the Ethiopian case, the deterioration of the competitive approach meant that an aging Emperor was increasingly out of touch with events inside his own country. The “master of marionettes” had become the father of a cohort of Pinocchios. It created an opportunity for opposition groups, who ultimately took power and deposed Haile Selassie (and thereby ended the reign of the Ethiopian monarchy).
The story of the last Ethiopian emperor demonstrates the difficulty of managing a competitive advisory system. At its best, it demands much of the leader’s time and attention. It also requires a personality that is comfortable with political conflict, as both Haile Selassie (in his younger days) and Roosevelt were. Leaders who are less comfortable in such surroundings will choose either to emphasize teamwork, if they seek a hands-on role in decision making, as did John F. Kennedy, or a more formal system with gatekeepers, if they are more comfortable pondering advice and options in solitude, as was Truman.
The discussion of the advisory system has thus far largely focused on instances where there is a single leader with substantial control over the design of the advisory system. Depending on the political system of a specific society, the leader may have more or less leeway in structuring the advisory system and choosing her or his advisors. The more a leader has the ability to place his or her stamp on the organization of the executive, the more his or her personality will factor into the organizational structure.
In a presidential system of government, for instance, the executive branch of government is separate from the legislative branch. The president is elected independently and does not owe her or his position to the support of the legislature, although a troubled relationship with the legislature can render policy making difficult. In a presidential system, the president usually has substantial leeway in organizing the executive to suit her or his decision making style, just as she or he has great autonomy in the selection of her or his advisors.
In a parliamentary system, on the other hand, the prime minister owes her or his position directly to the support of the legislature. If the legislature withdraws its support, for instance through a vote of no confidence, the prime minister is forced to resign. In a parliamentary system, the composition of the executive is less clearly determined by a single individual, depending in part on the electoral system of the country. In cases where a single party tends to win a parliamentary majority, a prime minister may exercise somewhat greater influence over the composition of government and the advisory system. In cases where governments are composed of several political parties, such as in coalition cabinet government, the advisory system as a whole is less likely to be structured to suit a single personality.
Rather, each member of the executive structures only a small circle of advisors in the department over which she or he presides. A cabinet government is a group of ministers who jointly constitute the executive of a country. They usually have collective responsibility, which means that each minister is expected to publicly support all cabinet decisions. Personal disagreements with collective decisions may not be voiced publicly. When the cabinet is made up of a coalition of political parties, meaning that two or more political parties jointly form the government, the collective responsibility for political decision making is borne by ministers who are affiliated with different political parties and have different political views and priorities. The interactions of coalition governments. The significance of the structure of the advisory system derives not only from the fact that it is often a function of the leader’s personality, but also—and perhaps more importantly—from its implications for the decision making process. In the next sections, we first examine the functioning of the government bureaucracy as a whole and then turn to an examination of the smaller decision making groups that form the immediate advisory circle around a leader.
The Rest of the Iceberg—the Government Bureaucracy
A thread that runs through the discussion of the organizational structure of the advisory system is that inaccurate, incomplete, and biased information makes its way through such policy making bodies. In some cases, information is not accurate simply because someone made a mistake or did not research thoroughly enough to discover (through consultation of alternative sources) that their information was not reliable. As the discussion of information distortion makes clear, not all failures in policy making can be blamed on such problems. That does not mean that distortions are deliberate efforts to misinform. No matter how well the advisory system works, it remains a political system. Advisors have their own perspectives on the world, as well as their own interests and ambitions. Even advisors who are appointed by the leader will not always perceive their interests to be perfectly aligned with that leader. Conversely, members of the permanent bureaucracy are not necessarily antagonistic to the leaders’ political agenda. In the end, policy choices are the result of a “dynamic influence process” in which advisors do much more than “merely collecting, processing, and interpreting information.” Each of the three approaches to organizing the advisory system addresses these issues in its own way.
Both the normative model of rationality and the formalistic approach to the advisory system emphasize finding the “best” solution on the basis of thorough analysis of the problem and the available policy options. Both downplay the role of politics in decision making. Neither is intended to describe the actual practice of policy making. Instead, the normative model of rationality outlines how policy ought to be made, whereas the formalistic approach organizes the relationships between the various individuals who are employed as members of the leader’s advisory system. To achieve a better understanding of the inner workings of the advisory system, we will need to delve into efforts to describe the actual advisory process. It is tempting to assume that foreign policy decisions are the result of a rational process in which the various agencies, departments, and offices that collectively constitute the government jointly serve an agreed-upon national interest. If this were the case, the rational policy model might provide a fairly accurate description of how foreign policy is made. It assumes that foreign policy is made as if a single, rational decision maker analyzes a strategic problem and, once the problem is defined, selects a policy response from among the available options. The process by which the policy response is selected starts by outlining the options, investigates the likely consequences of each, and settles on the option that promises the biggest benefit at the lowest possible risk and/or cost. Fundamental to the analysis, as well as the judgment of cost and benefit, is the desire to serve the state’s interests.
This rational policy model does not take into account the possibility that information could become distorted in a complex advisory system made up of many individuals, offices, and agencies. Neither does it take into account that identifying the national interest is not necessarily straightforward. Here, we delve into two alternative descriptions of the decision making process, the organizational process model and the bureaucratic politics model, which were originally created as critiques of the rational policy model. Both models take into account that there are usually multiple perspectives on any given policy problem, but they stress different reasons for the existence of those multiple perspectives.
The organizational process model envisions the government as a collection of organizations, centrally coordinated at the top, each with their own specialties and expertise, but also its own priorities and perceptions. Each organization, moreover, has its own customary ways or standard operating procedures, which is often abbreviated as SOPs. Although it is efficient for organizations to act according to such standard procedures most of the time, rigid adherence also robs them of flexibility when they confront a novel or unusual situation. According to this model, organizations respond to such situations by adapting rather than reinventing their standard operating procedures. Adaptation consists of small and incremental changes to standard procedures. Such changes are easier to implement, even if they are not an adequate response to the problem they are intended to address. And that is the key to this model: it describes government as a large conglomerate of organizations that, singly and collectively, pursues policy responses that permit them to stick as closely as possible to well-worn routines that they know to be feasible rather than to fashion policy responses that best respond to the problem. According to this model, then, inadequate policy responses do not result from a failure to objectively evaluate the risks and benefits associated with various options, but from the inertia of established organizations.
Interestingly, the competitive style of organizing the executive also predicts an emphasis on feasible policy decisions, but it stresses the interactions among individuals rather than the inertia of organizations as the reasons for that emphasis. To understand the difference between organizational inertia and the interactions among individuals within the advisory system we need to first delve into the details of the bureaucratic politics model.
The bureaucratic politics model focuses on the role of individuals within governmental organizations. Individual advisors within the government occupy specific roles within it:
- They lead, or work within, a specific agency or department. Each agency and department has its own mandate. The specific expertise and policy interests of that agency are bound to color the perceptions and opinions of the individuals working within that agency.
- Advisors are also placed at a specific location within the hierarchical structure of that agency or department. The individual who serves as the head of the agency is dependent upon her or his subordinates to provide information, analysis, and policy options.
Agencies are typically hierarchical structures and run the same risks for distortion of information that we discussed as part of the formalistic style of organizing the executive. The direction provided by the agency head influences that agency’s functioning. How much so will depend on both the quality of the leadership and the degree to which follow-through is monitored. In addition, those working within the agency are not robots that simply follow orders, but individuals with their own interests and career aspirations. Some may seek to help their superiors by highlighting information that supports their point of view and downplaying the information that contradicts it. Others may seek to advance their own career by establishing (and getting their superiors to notice) their expertise in a specific policy area. The relationships between superiors and subordinates sketched here provide only a small sampling of the many ways in which these relationships affect the flow of information, and thereby the policy options that are proposed and the choices that are ultimately made.
The bureaucratic politics model stresses that advisors’ perceptions and priorities are shaped by both the organizations that employ them and their personal ambitions and interests. As a result, policy choices become the end result of complex bargaining at multiple levels: hierarchically between superiors and subordinates with their own individual agendas and horizontally between heads of agencies that represent different interests within the government. Both the competitive and collegial style of organizing the executive recognize that this occurs and seek to harness the power of multiple perspectives, but both require considerable skill and involvement by the leader. The formalistic style does not take either the complex advisory relationships or the distortion of information that can result from them into account. On the contrary, it emphasizes the efficiency of formal bureaucratic structures and shields the leader from much of the conflict that is likely to occur within the advisory system. Although it is important to recognize that the advisory bureaucracy rarely functions as efficiently and dispassionately as the rational policy model assumes, it is also important to recognize that correctives can come from the leader’s advisors as well as from the leader her- or himself. The advisory system surrounding the leader can either mediate or aggravate the distortions in the flow of information from organizations and individuals within the bureaucracy. It is the dynamics of this smaller advisory circle to which we now turn.
Back to the Tip of the Iceberg—Decision Making in Small Groups
Leaders and their advisors depend on government agencies, and the individuals working in those organizations, for information and advice. In the end, however, foreign policy decisions are made closer to the tip of the iceberg: by leaders and their small circle of advisors, or by groups of policy makers. It is in these small groups where policy makers meet face-to-face that decisions are fashioned on the basis of the information and analysis provided by the various agencies and departments. Such groups may consist of only a few people or encompass an entire cabinet in a country with parliamentary government. Some scholars even include groups as large as the entire parliament. Larger groups will require more rules and direction to function well than small ones, which can remain more informal in their interactions. The important distinction is that the members of the group speak directly with each other as a collectivity.
Here, we are primarily interested in groups that are no larger (and perhaps smaller) than a cabinet government. A cabinet government is a group of ministers who jointly constitute the executive of a country. Officially, cabinets usually have collective responsibility, but the prime minister can become a dominant figure within the cabinet rather than simply one of the collective. This is especially true in electoral systems that yield governments dominated by one political party, rather than a coalition of several parties. British Prime Minister Margaret Thatcher, for instance, was such a dominant prime minister.
Although some scholars include parliaments in their definition of small groups, we will not: the members of parliaments do meet face-to-face as a group, but their deliberations are governed by highly formalized rules and protocols. The kind of group dynamics that are the subject of this section do happen in (subgroups of) parliaments, but they are not usually part of the formal sessions of a parliament. Some of interactions described here could occur in larger groups, others require the more intimate setting of a small group.
Small groups serve a variety of functions in foreign policy decision making. Most popular are two images of the small group: one portrays the advisory group as a think tank, where top advisors use the available, but incomplete, information to jointly construct a representation of a foreign policy problem, determine its importance among other foreign policy problems, and debate how best to respond to it. The basic assumption is that decision making in a team is “demonstrably superior to [single] individuals when it comes to processing information about novel, complex, and unstructured problems.” In other words, groups or teams are especially good at making sense of the sort of ill-structured problems that characterize most foreign policy decision making situations because the creative interplay between their individual efforts to make sense of the available information yields greater insight than a single person could achieve within a short time frame. That is, provided that the group functions as intended. The other popular image of the political decision making group is that of the command center, where the group jointly determines the foreign policy actions. In this role, the group builds on the think tank role to develop options, evaluates them, selects the most viable ones, and ultimately makes a decision.
Most of the time, even in groups of supposed coequals, there are subtle and informal hierarchies that shape the relationships among the individual members of the group. Consider, for instance, that the Argentine junta accepted the analogies to Suez and Rhodesia proposed by their Foreign Minister, Nicanor Costa Mendez, and estimated the British response on the basis of that assessment. This estimate was crucial to their decision making. Because Mendez was seen as more knowledgeable, he exercised greater influence over the decision than his colleagues. In other situations, such as the example of the U.S. Presidency so aptly described by Truman, groups of advisors may make recommendations, but the responsibility ultimately rests with one individual leader, who may or may not have taken part in the group’s deliberations. If the leader accepts the recommendation, the group appears to be the command center without having the ultimate authority. Hence, the accuracy of the image of the group as a command center, and whether we should interpret it as one that makes decisions or one that simply advises, depends on the powers invested in that collectivity. In the case of both the think tank and the command center role, the group is involved in the actual work of decision making; that is, it works at identifying the contours of the problem, discusses policy options, and so on. Formal and informal small groups perform additional functions as well: One, group decision making can help a government present itself as a unified team that works together in the national interest. The fact that decisions are made or recommended by a unified team helps justify and legitimize these decisions as reflecting the government’s, or even the society’s, core values. This same emphasis on values can be turned inward: reference to the government’s or society’s core values can help shape the norms of the group and its members. If such an emphasis is very strong, it can squelch dissent within the group and make the group function less effectively. Two, a small, informal group can also serve as a sanctuary.
Such a group becomes an emotional support system for a leader, which can help her or him deal with the pressures and stresses of leadership. It can also devolve into a circle of sycophants, making a leader overconfident in the appropriateness of her or his judgments rather than providing a gentle “reality check” when appropriate. Three, groups that are formally part of the government structure can serve as “smoke screens” behind which informal groups do the actual work of decision making. Such informal groups can be places where differences of opinion can be aired out more readily than in more formal settings. Consider, for instance, foreign policy making in societies that are governed by coalitions of parties with very different ideas about the policies that will best serve the state’s interests. Policy makers may prefer to sound out their colleagues in other parties informally and work out major differences before publicly debating the issues with one another. The use of informal networks to work out agreements is common in other situations as well. Once differences of opinion become public, it will be harder for any of the decision makers to change their opinion. Doing so could damage their political reputation or the party’s reputation. It could make either look like they are just “blowing with the winds” or compromising for the sake of expediency or their own advancement. This discussion of the different roles played by small groups in foreign policy decision making is not exhaustive, nor is it meant to be. Rather, it serves to highlight that small groups perform important functions in addition to their role in the decision making process itself and that those additional roles are not wholly separate from that decision making process but influence it.
The actual decision making role of a group of policy makers is twofold, as previously introduced. One, groups play the role of think tank when they gather, organize, and process information to gain an understanding of the problem, i.e., construct a representation of it. Two, groups act as a command center that develops and evaluates options. The command center is responsible for arriving at a decision. The same group of individuals may fulfill both the think tank and command center roles. The distinction roughly mirrors the two stages of the poliheuristic theory. At each stage, group dynamics influence the process in a somewhat different manner: a more even distribution of power among the group members makes it likely that a wider range of options will be considered and debated as the group fulfills its think tank role, but may also make it more difficult for the group to reach a decision as it moves to its command center role.
Colleagues and Competitors
Advisors are both colleagues and competitors. As we saw earlier, the competitive executive management style aggravates and utilizes the elements of competition among the policy makers who surround the leader. The collegial style, on the other hand, seeks to foster collegial interaction while acknowledging the multiplicity of viewpoints. The formalistic style pushes competition and conflict away from the tip of the policy making iceberg. The thread than runs through each of these three descriptions of government decision making is that individual and organizational factors influence problem representations and decisions. More importantly, you have probably noticed that the policy making process is rife with opportunities to advance the cause of the (perceived) national interest, one’s organization, one’s superior(s), or oneself. A policy maker can act collegially and loyally to achieve her or his ends or can choose to subvert the career of a superior or a colleague. The small advisory groups at the apex of government are no different from the government as a whole: the members of such groups can interact in a variety of ways to serve a mixture of interests. Their interactions can be summarized into four main interaction patterns: bargaining, concurrence, deadlock, and persuasion. We discuss each in turn.
The bureaucratic politics model specifically mentions bargaining, which implies that the decision making process involves give-and-take and that the preferences of any one policy maker never fully shape the decision but often partially do so. This suggests some form of compromise as the likely result of the bargaining process. A compromise would suggest an integrative solution, defined as a result that represents the preferences of all members of the group, albeit modified to some degree. This may be easier to achieve in smaller groups than in larger ones. A small group of advisors close to the leader may be able to arrive at an integrative solution easier than, for instance, a coalition cabinet made up of representatives of several different political parties. The former are much more likely to share assumptions about the core values of the government. In larger groups, bargaining can easily lead to a subset solution, in which one faction’s ideas end up dominating the preferences of other members or factions within the group. This situation can emerge in small groups as well, especially if the members of the group are of unequal status. In larger groups, the subset solution could favor smaller as well as larger factions, depending on the politics surrounding the specific issue at hand. In either case, the members or factions that emerge on the losing end of the bargaining process must at the very least acquiesce in the group’s decision.
When such acquiescence emerges quickly and without much debate, the decision process is more accurately characterized as one of concurrence. In this case, other options are barely discussed, or not at all. Instead, the decision makers quickly settle on an option they jointly perceive as a reasonable solution. This can happen for three reasons: One, there may be a dominant solution. This means that only one option is credible. In terms of the poliheuristic theory, it may mean that only one option met the
noncompensatory criteria in the first stage of decision making. In other words, other options are sought and considered, and are quickly disposed of because they either are not politically feasible or simply not adequate policy responses to the problem. Two, and likely only when a decision is not of vital national importance, it is possible that the policy makers satisficed, or accepted the first option that met their threshold of acceptability. Depending on the situation and on other simultaneous problems that demand the time and attention of policy makers, this may be a reasonable way to deal with a situation. Three, concurrence can be evidence of groupthink, or the premature closure of the search for options.
The problem with groupthink is not just that there is little or no consideration of alternatives (something that is also true when there is a dominant solution). The primary problem is that the decision makers fail to critically examine their problem representations and the option(s) before them or to ask themselves whether there might be other options that they have not yet considered. This may happen as a result of strong cohesion among the members of a small group: they perceive the world in such similar ways that none of them is able to offer an alternative point of view or think of alternative ways of confronting the situation. Think, once again, of the failure of any of the members of the Argentine junta to suggest that British Prime Minister Margaret Thatcher might perhaps respond differently than her predecessors had during the Suez and Rhodesia crises. None of the decision makers in Argentina at the time seems to have considered the possibility. And if they did, they kept quiet because they all perceived Foreign Minister Mendez as more expert than themselves. In essence, groupthink is the problem of a distorted and one-sided problem representation based on incomplete or faulty information that no member of the group questions because none of them can conceive of any other way of understanding the problem or no one dares suggest that the person they all see as more expert might be wrong. Hence, no one plays “devil’s advocate.” Interestingly, U.S. President John F. Kennedy purposely left the deliberations of his small circle of advisors during the Cuban Missile Crisis to avoid precisely this problem. He was aware that his presence might influence the nature and content of the discussions among his advisors. Remember that Kennedy organized his executive along the lines of the collegial style, which routinely takes multiple perspectives into account but which can also devolve into a closed system of mutual support.
When the system works well, as it did during the Cuban Missile Crisis, the decision making process can avoid the pitfall of groupthink. When the system does not work well, it is likely to fall into precisely this trap. The formalistic style has no built-in mechanism to counteract the distortion of information. When it works well, the emphasis on analysis helps it avoid the problem of groupthink, but when it does not work well, this style of organizing the executive may also suffer from groupthink without ever realizing that its distorted view of the issue is not an accurate representation of the problem.
On the opposite end of reaching agreement too readily is the failure to achieve any agreement at all. This is called deadlock. In countries such as the United States, where the President selects the members of the executive, the occurrence of irresolvable differences among small groups of top advisors should be rare. The leader selects these individuals to suit her or his worldview, policy preferences, and decision making style. Contrast this with the situation of a coalition cabinet in a parliamentary system of government: who becomes part of the group in such cases is not determined solely by the prime minister but is dependent on negotiation and dynamics internal to the political parties that participate in the coalition. When coalition cabinets are truly deadlocked, this may result in the resignation of a minister or even the fall of the government. The frequency with which this happens varies, depending on the issues as well as the political traditions within a specific country. Even in countries where the members of government serve at the pleasure of a prime minister or president, and are presumably selected because they suit this individual’s priorities, is it possible for advisors to have very strong differences with regard to specific policy problems. For instance, during the Carter administration a stalemate developed between National Security Advisor Zbigniew Brzezinski and Secretary of State Cyrus Vance with regard to the direction of foreign policy. The two men had very different worldviews. If such problems are allowed to fester, and especially if they become public, they can have negative consequences for a government’s ability to respond to foreign policy problems and lead to perceptions of ineptness.
Deadlock may result when several members of a decision making group each have strongly held opinions about the course of action that should be taken in response to a specific foreign policy problem. Such strongly held opinions can also lead to efforts at persuasion. In fact, one might imagine that mutually unsuccessful efforts at persuasion could result in a standoff between the members of the group as each advisor digs in his or her heels and refuses to compromise. Efforts at persuasion do not inevitably lead to a deadlock. If they are successful, others in the group come around to seeing the situation from the point of view of the persuader. More likely may be the subset solution mentioned previously, which is a solution that is preferred by a subset of the advisory group rather than the entire group.
The difference between persuasion and a subset solution is that the latter does not require that the persuader have fully convinced his or her colleagues, only that he or she have achieved sufficient agreement among them that they are willing to go along with the proposal. The difference between persuasion and compromise is that the former is not an integrative solution that reflects (aspects of) the starting preferences of all members of the group. Instead, the preferences of one member or a subgroup become dominant. Hence, a subset solution can connote either a partial compromise or partial persuasion.
Comparing the preferences of the advisors at the outset of their deliberations with the decision could help evaluate whether a specific subset solution is best characterized as compromise or persuasion. Information about the initial positions of decision makers is not always easy to obtain. Decision makers often cloak their statements about decisions in terms of bargaining, even if that’s not exactly what happened. Hence, if foreign policy analysts accept decision makers’ own characterization of the decision making process, it is likely that they will overestimate the degree to which bargaining plays a role in decision making.
Political Games, a.k.a. Strategies of Influence
The four decision processes described in the previous section define that process primarily by the manner in which the outcome is achieved. Much more can be said about small group interactions, as each of the policy makers involved is likely to be engaged in efforts to manipulate the decision process to increase the chances that the decision she or he favors will dominate or significantly influence the decision. Political manipulation is defined as the effort(s) made by one or more individuals to influence a situation in which a group is making a decision in a way that increases the chances that the outcome will reflect their preferences. There are a variety of strategies a decision maker can employ to improve her or his chances to significantly influence the decision. Such strategies can be divided into
(1) efforts to influence the composition of the decision making group so as to reduce the impact of opposing viewpoints;
(2) efforts to influence the beginning stages of the decision process, such as the framing of an issue or perceptions of its relative importance among the various issues the government confronts simultaneously;
(3) efforts to manipulate the dynamics of interpersonal interaction within the group.
This means that knowledgeable and well-informed policy makers can differ greatly in their assessments of the same situation. Each of them operates on the basis of partial information and their own political instincts—which are in turn heavily informed by the policy maker’s personal operational code. The impact of these differences of interpretation on the decision making process will depend also on the personal characteristics and ambitions of the individuals involved.
What tactics might decision makers use to improve their chances to influence the decision? Let’s examine each of the three types of strategies previously mentioned in a little more detail:
- Group composition. Policy makers sometimes try to influence the composition of the decision making group. This can be achieved either by excluding a colleague whose opinions contradict one’s own or by trying to include additional members into the group who will support one’s position. Exclusion can be achieved formally only by policy makers who have the clout to play a role in determining membership in a particular decision making group, but it is also possible to schedule meetings at a time when the individual with the opposing opinion happens to be out of town or to hold informal meetings with select group members apart from the officially scheduled ones. Including additional members into the group can sometimes be justified on the basis of their expertise and can be useful in providing additional support for one’s position. It is also possible to strengthen the credibility of a viewpoint by claiming to speak for one’s superior. Doing so not only includes that person’s opinion into the debate, it serves to lend greater weight to one’s own point of view.
- Framing. Efforts to influence how an issue is framed are especially effective at the early stages of a decision making process. Individual decision makers are likely to frame problems each in their own distinctive way. Once they join one another in a group to deliberate how to respond to this problem, they will each operate on the basis of this individual problem representation, unless the group first deliberates the contours of the problem before moving on to outlining and discussing options. By influencing the group’s collective problem representation, an individual policy maker can manipulate which options will then have a greater likelihood of being chosen.
- Interpersonal relationships. In addition to manipulating who participates and how the problem is framed, policy makers are likely to use a variety of tactics to influence how they and others in the group are perceived. In addition to bolstering one’s position by claiming to speak for a superior, a policy maker might seek to discredit the expertise of their opponent. Another tactic is to get others to agree in stages. This is called the “salami tactic.” It requires a lot of planning and patience to structure a debate in such a way as to get colleagues to agree with you on minor points to build to the inevitable conclusion you set out and that they cannot escape once they have agreed with you on the smaller points. Still another tactic is to leak information. This is a risky strategy, because leaks can easily backfire and have negative implications for the reputation of the source of the leak. It can damage the advisory system because its members can no longer trust that their deliberations will remain confidential. Yet it can also be an attractive strategy for someone who cannot get his or her voice heard within the group or to circumvent a rival.
The use of such tactics shapes the decision process in the small group. Other strategies may be used by policy makers as they seek to influence the decision process. Whether that process is best characterized by bargaining, concurrence, deadlock, or persuasion will depend on the tactics group members use to influence one another and how those tactics combine to produce the outcome. For instance, if opposing viewpoints have been successfully excluded, the group might reach easy agreement (concurrence), whereas if each seeks to convince others of their viewpoint they may be unable to reach a decision (deadlock).
Small advisory groups, with their strategies and political games, are especially common in political systems with a strong executive, characterized by one individual who has the final responsibility for the decisions, such as is common in presidential systems. The entire advisory system ultimately coalesces at the tip of the iceberg where the president or prime minister can’t pass the buck to anybody, as U.S. President Harry Truman said. A different situation is found in parliamentary governments where coalition cabinets are common. Decision making has a more collective quality. This does not mean that the dynamics of advisory systems, bureaucracies, and political manipulation are irrelevant. It does mean that decisions are ultimately made by a group of policy makers who represent different political parties and policy agendas. It is to the dynamics of such groups that we turn in the next section.
Coalitions: Governing Together
In political systems in which the ultimate responsibility rests with a single chief executive, such as a president, advisory groups are not ultimately responsible for the final decision. Rather, it is the chief executive who bears that responsibility. In parliamentary cabinet government, the heads of the various departments share collective responsibility for policy decisions, at least in a formal or legal sense. The prime minister is in that case considered to be the primus inter pares (Latin for “first among equals”), meaning that the prime minister holds the special position of head of the collective but is not superior in rank to her or his colleagues. In practice, though, the prime minister often carries greater weight in decision making than the other members of the group. This is especially true in parliamentary systems where a single party dominates the government, such as is usually the case in Britain. There, prime ministers like Margaret Thatcher and, more recently, Tony Blair acted as the central figures of their governments, leading to the “presidentialization” of cabinet government.
When the cabinet is created out of a coalition of political parties rather than one dominant party, the situation is different. In such cases, cabinets remain closer to the principle of collective responsibility in their decision making. As discussed earlier, coalition cabinets bring together the representatives of two or more political parties for the purposes of governing a country. The power of these parties is rarely, if ever, equal: Each party’s presence in the coalition is proportional to its relative presence in parliament and also reflects its electoral gains in the most recent election. The exact distribution of power among the coalition partners, or the parties that have agreed to form a government together, is the subject of negotiations between those parties. Before signing a coalition agreement, the document that governs the cooperation between the coalition partners as they govern together, representatives of these parties negotiate not only the number of ministers each party will provide but also which ministries each will hold, as well as the general outlines of the policy agenda that will guide their government.
Coalition cabinets rest on the foundation of carefully worked-out agreements between the parties that constitute them, but those agreements can never fully specify policy decisions in advance. Hence, coalition cabinets engage in group decision making processes. They bargain and persuade, occasionally they concur, and at other times they deadlock. Such deadlock can have consequences far beyond the policy issue itself. Failure to reach agreement on one important policy problem can spell the end of the coalition government when the representatives of the party that cannot get its way threaten to abandon the coalition. If they make good on their threat, the government falls. This is a high price to pay for disagreement, because it necessitates at a minimum another round of negotiations for a new coalition agreement—with no guarantee that all ministers will be reappointed to a ministerial post—but it can possibly also lead to a new round of elections.
The fact that there can be such a dramatic consequence to the inability to reach agreement affects the dynamics within the coalition. It can give the smaller, or junior, party influence beyond its relative strength in the coalition. In essence, the junior party is in a position to “blackmail” the larger, or senior, party into agreement, assuming that the senior coalition partner would prefer to continue its government role. A more positive interpretation is to view the junior coalition partner as providing a corrective by not permitting the senior coalition partner to always get its way or push through its own policy agenda.
It is important to note that junior parties do not have significant influence on every single decision. Not much is known about the circumstances that permit junior coalition partners to place their stamp on policy, but it appears that the threat to abandon the coalition is an important strategy for exercising disproportionate power. Furthermore, when the senior coalition party is internally divided about the preferred policy option but the junior coalition is united and sides with the faction of the senior party that most resembles their own position; the junior party can have a distinct influence on the decision. Finally, the distribution of ministries matters also. If the junior party has a share of the important, or core, ministries (like foreign affairs, defense, or economics) roughly equivalent to the share of core ministries held by the senior coalition partner, then the junior coalition partner can exercise greater influence than when it is largely relegated to politically peripheral ministries—such as a ministry of culture. The politics of joint governance through coalition cabinets show that such governments are particularly vulnerable: disagreements over foreign policy can, and occasionally do, lead to the dissolution of government.
Whether policy makers function within the institutional constraints of coalition governments or guide a leader who has been able to structure the advisory system that surrounds her or him, in each case the institutional arrangements have an influence on the decision making process. No institutional arrangement is perfect; each has its own advantages and pitfalls. How well each institutional arrangement works will depend on the leader’s ability to use the system to its full potential. And this, in turn, depends on how well the leader’s personality is suited to the institutional arrangements. It also depends on wisely chosen advisors, who understand the institutional framework in which they operate and the person (or persons) for whom they are working. But even under the best of circumstances, foreign policy outcomes depend on more than good decision making. Domestic and international constraints also play a role in determining whether policy decisions lead to the desired outcome.