All the President's Men: Comparing the Secretary General of the French Presidency and the US Chief of Staff in an Era of Presidentialization

Abstract

This article compares the Secretary General of the Elysée Palace and the US Chief of Staff, central political advisors to the French and US presidents. Our aim is twofold. Firstly, we identify the precise roles of these advisors. By mapping their respective powers, we demonstrate their importance in presidential decision-making. By examining what the French Elysée Secretary General and the US White House Chief of Staff have in common and how they differ, we develop a comparative understanding of the mechanisms of the presidentialization of political executives. We show that the similarities of the two offices are linked to the ongoing presidentialization of the French and American political systems, which, by giving greater power to heads of state, also strengthens their advisors. This commonality does not rule out marked differences between these officials, demonstrating that presidentialization takes distinct forms, reflecting distinct political cultures as well as different balances of power within each institutional system.

The study of executive politics has long been interested in the entourages of leaders as essential instruments of executive power. In presidential and semi-presidential political systems, a comprehension of the role played by presidential advisors is essential to an understanding of changes in the prerogatives of heads of state. The aim of this article is to propose a comparison of two advisor roles: the Secretary General of the Elysée Palace in France and the Chief of Staff of the US president. Our ambition is twofold. Firstly, we want to shed light on the precise role and remit of these powerful advisors, who remain little-known because of their discretion, in a comparative perspective. By mapping their respective powers and showing some of their similarities, we want to demonstrate and analyse their importance in presidential decision-making in the two political systems studied.

The originality of our approach, however, is to draw this comparison in relation to the discussion of the presidentialization of the US and French political systems. Our aim is to understand whether presidentialization is accompanied by an evolution of the functions of the president's closest advisors on the two sides of the Atlantic, and if so, how different these developments are.

The Secretary General of the French presidency, at the Elysée Palace, and the Chief of Staff of the US president, at the White House, both function as indispensable aides to their presidents and are crucial for the capacity of the presidents to pursue their agendas and to manage the vast amount of information that flow into these offices – and the vast number of decisions that must be made. The two officials are both essential, therefore, for the effective governance of these two republics. If the two incumbents of these offices were to meet for lunch, they would have a great deal to talk about, but they would also find there were several points of difference, and perhaps several of incomprehension.

Among the group of presidential advisors in France, the Secretary General, or ‘SG’, as he is usually called, is, more than any other, the president's man (Martigny 2021b), at the junction between politics and administration, between the head of state and the prime minister, at the heart of executive power.¹ ‘At the centre of everything and aware of everything’: the formula used by General de Gaulle to define his place and missions as early as 1959, clearly shows how this senior civil servant, whose role remains mainly unknown to the public, plays a central part in the organization of Elysée power.

In charge of the French president's cabinet,² with an eye and an opinion on all the subjects dealt with by the presidency, the SG is in turn architect and manager, advisor, sometimes spokesperson for the head of state, shadow negotiator whose role alternates between acting as the right-hand man and alter ego of the head of state, and the only collaborator to have unlimited access to the latter. More important than the ‘directeur de cabinet’, in charge of the stewardship of the Elysée Palace, and the special advisor – essentially political – the SG is the right-hand man of the president and fully devoted to that office, central in the Elysée system through his triple authority on the political, organizational and administrative levels. The SG's missions mix political and administrative dimensions, contributing to the centrality of the Elysée apparatus in decision-making (Cohen 1981; Eymeri-Douzans et al. 2015).

The role of Chief of Staff (CoS) in the White House evolved from the role of personal secretary to the president. Every president since at least the time of Woodrow Wilson has had a personal confidant and ‘fixer’ to help impose the will of the president on the rest of government (George and George 1964; Neu 2014). The role that Harry Hopkins played for Franklin D. Roosevelt was perhaps even more central to policymaking than that of contemporary CoSs (Adams 1977). These personal aides to presidents often generated great jealousy from those in government who held more official positions, but they were crucial for the president's capacity to govern.

As the position has become institutionalized, the duties of the CoS have continued to expand. Like the SG in France the CoS is a chief, if not the chief, advisor to the president and is involved in both political as well as policy advice. Different CoSs have emphasized those two aspects of the role differently (see below), but they all have had to deal with both dimensions of the presidency. Further, they play a role in managing a large organization with many powerful, and would-be powerful, members. Still further, the CoS must communicate the position of the administration to other actors in government and to the public.

The aim of this article is to compare the roles of the two above-mentioned functions, in relation to the discussion on the presidentialization of political systems in the US and French presidential and semi-presidential regimes. Our main argument is that the CoS and the SG of the Elysée are mainly comparable because of their similar importance in presidential decision-making. They illustrate, in their attributes and influence, a tendency to unrestricted power among the closest advisors of the chief executives. Some developments in this office and in the biographical trajectories of its occupants are indicators of the increasing presidentialization of the French and US political systems in the last few decades. Despite some ups and downs, the power of these two presidents have grown, and with that growth the powers of these actors have grown too.

With this article, we examine the role(s) of the most senior staff in two presidential (or semi-presidential) systems. By doing so, we want to reflect on the issues the driving forces behind political decision-making (Blondel and Müller-Rommel 2016), the extension of executive prerogatives (Lowande and Rogowski 2021) and the politicization of the senior administration (Peters 2013). The central importance of these top members of the presidential team supports the idea that having strong, personalized aides is a significant aspect of a strong presidential system. Although different presidents may use their top advisors differently, they are crucial for shaping a presidency.

We also argue that the powers of these top advisors and managers for the president have been expanding, and therefore to some extent both of these officials may be becoming more important, if not more similar. We propose the argument that such an evolution of roles and missions says a great deal about the process of presidentialization that is underway on both sides of the Atlantic, yet with nuances that are the product of each political culture.

Framework of analysis

In support of this assumption, we turn to three strands of research: (1) the general research on the role and powers of the US and French presidents; (2) research on ministerial and presidential advisors; and (3) research on the presidentialization of political systems. First, long-standing scholarship has demonstrated the importance of the prerogatives of the presidents of France and the US, two countries in which the presidency holds a power unrivalled in the world's democracies. In France, the overwhelming weight of the president of the Republic in the French institutional system has long been described as ‘presidential monarchy’ (Duverger 1974), whose influence is unparalleled in other democracies. This role has also increased over time, particularly since the so-called ‘hyper-presidential’ Sarkozy presidency in the mid-2000s (Duhamel 2008; Haegel and Grunberg 2012; Rosanvallon 2018), which led to a more extensive use of power by French presidents (at least in the case of Nicolas Sarkozy and Emmanuel Macron), causing the gradual marginalization of the prime minister and even the government, without any constitutional reform (Dulong 2021; Martigny 2019; Thiébault 2016).

In the United States the question of the ‘imperial presidency’ has come and gone since at least Richard Nixon. Richard Nathan (1975) described Nixon's approach as the ‘administrative presidency’, meaning that the president sought to control the rest of the executive branch more completely than had other presidents. This style has reasserted itself several times, notably with Presidents Ronald Reagan and Donald Trump. The administrative presidency can be associated with the assertion of the ‘unitary executive’, meaning that the executive branch should be entirely under the direction of the president and have no autonomy (Driesen 2020). Although there has not been the steady accretion of power as in the French system, the role of the president remains the central political power.

The growing interest in the workings of political decision-making has generated a large and burgeoning literature on ministerial advisors (Eymeri-Douzans et al. 2015; Shaw 2023), which provides useful insights into the roles played by aides in general. The more limited literature on advisors of chief executives points to the distinctive roles of advisors to presidents and prime ministers, be it their capacity to provide policy and political advice, manage presidential teams or protect the politician they serve (Andeweg 1999; Esselment et al. 2014). It should be noted that the literature devoted to the CoS is substantial in the United States, demonstrating the long-standing interest of government studies in the workings of the White House (Cohen and Hult 2021; Cohen et al. 2012a, 2012b, 2016; Hult and Walcott 2004; Patterson 2008; Pfiffner 1993, 1996; Sullivan 2004; Walcott and Hult 1995, 2005; Walcott et al. 2001; Whipple 2017). In France, the study of the entourages of the prime minister and the president have been the subject of a seminal edited book in the last decade (Eymeri-Douzans et al. 2015), or works have focused on the Secretary General of the Government (Eymeri-Douzans and Mangenot 2019). Yet, little research has focused on the SG of the Elysée Palace – the studies that exist were mainly conducted by law academics (Foucaud 2010; Magnon in Eymeri-Douzans et al. 2015) – political science literature remaining rare on that matter (mainly Cohen 1980, 1981; Coutrot et al. 1982; Martigny 2021a, 2021b, 2023; Rouban 1998).

The correlation between the powers of heads of state and the role given to their advisors has been established by studies of the presidentialization of political systems, especially Thomas Poguntke and Paul Webb's seminal work (2005) analysing the strengthening of the role of heads of state and government in several Western democracies. Against a backdrop of objective growth in the scope of executive power, and the emergence of populism in the early 2000s (Kriesi 2014, Kriesi and Pappas 2015; Mudde and Kaltwasser 2017; Müller 2017), presidentialization scholars have attempted to understand the driving forces behind the growth of that power in a comparative perspective. The analysis of the presidentialization of political systems put forward a general model to explain this phenomenon, proposing criteria that allow a dynamic assessment of its importance. These criteria include what the authors call ‘power resources’ of chief executives. Growth in the resources available to presidents is partly due to the extension of the number and prerogatives of the advisors who surround them. It leads presidents to hire reliable, loyal and responsive advisors to whom they entrust part of their power. Scholars such as Sue Pryce (1997), Michael Foley (2008) or Robert Elgie and Gianluca Passarelli (2020) have also emphasized the development of staffing as a significant feature of the presidentialization of parliamentary systems. Even in presidential systems, the powers of advisors appear to be increasing (Helms 2015). To analyse the roles of the first of these in both an analytical and dynamic way is to understand what the power of the advisors says about the power of presidents.

The four roles of the Secretary General and the Chief of Staff

Although much of the literature on advisors is descriptive, there have been efforts to employ theoretical and analytical approaches on roles. For example, Anna Esselment et al. (2014) argued that the familiar principal–agent model could be applied to the relationship between political or bureaucratic leaders and their advisors. This principal–agent model for the primary presidential advisors is perhaps not quite so simple as that found within a bureaucracy, given the multiple goals being pursued and the many agents that the president may have (ministers, other advisors, the party, etc). The idea of the CoS and SG as agents can be linked to the idea that these officials are means of magnifying the power of the president or prime minister (Pickering et al. 2023). Those executives have legitimate power within their governments, but that power must be put into action.

The study of the historical emergence of presidential advisors reveals this ‘putting power into action’ trend. Both functions emerged almost in the same years, which correspond to the reorganization of the American and French political systems in the immediate post-World War II period. The CoS began to be institutionalized during the Truman administration, and Sherman Adams (1953–1958³) in the Eisenhower administration is usually called the first genuine CoS. He remained in office longer than any other CoS has since (almost six years). The experience of the New Deal, World War II and the continuing Cold War made it clear that the president was to be even more central in policymaking that required more support within the White House.

The degree of institutionalization of the office, and the Executive Office of the President (EOP), has varied over time (Cohen et al. 2016), depending upon the style of the presidents and the ‘meta-cycles’ of governing (Rockman 1984). There appears to be a learning process, during which presidents of one party learn from the perceived successes and failures of previous presidents of the same party, or even the other party. This learning is facilitated by the frequent reappearance of officials of the previous administrations in the staffs of incoming presidents. For example, a deputy chief of staff from the Carter administration argued that the best way to learn the job was to talk to people who had done it before, not read about it (Walcott et al. 2003: 112).

Although the French SG first emerged in the 19th century, it is really when General de Gaulle returned to power in 1958 that the office experienced a political and administrative rise. Like in the US, the expansion of the SG's role marked a reinvestment of the presidential office. De Gaulle decided to make the SG his closest collaborator. This choice indicates the conception of the qualities that he expected: obedience, discretion and unfailing loyalty, which made the SG a civilian collaborator with virtues comparable to those of a soldier. President Georges Pompidou modernized the function in the 1970s, and the SG later became an essential part of the entourage of the president of the Republic after the political alternation in 1981. The SG became a presidential control tower, increasing its missions and lowering the watertight barrier that had separated the political dimension from his hitherto administrative role. He became the conductor of the political and administrative organization of the Elysée, leaving the director of the cabinet with the sole task of managing the presidential administration.

It can be argued that both officials play the four fundamental roles identified for the CoS by Karen Hult and Charles Walcott (2004). These four roles – guardian/custodian, administrator/manager, political/policy advisor and proxy – are played out in different ways over time, depending on the nature of the president, the challenges being faced by the government at the time, and the capabilities of the CoS or SG. Comparatively, some of these roles may be more important in one country than another – for example, manager in the United States, or proxy in France.

Guardian/custodian

The SG of the Elysée and the American CoS share the common responsibility to protect the chief executive. Being the principal aides to the presidents, they are the only advisors to have unrestricted access to the presidential office. Part of the role of protection, however, is internal – what R.A.W. Rhodes and Anne Tiernan (2014) term ‘pest control’. They therefore act as gatekeepers to the office of the president (Whipple 2017). Everyone wants a little bit of the president's time, and these little bits add up and risk impinging on decision-making on major issues. At the same time, however, both the CoS and the SG must ensure that the right people do get access and that the executive does not become too isolated. They receive people that the president does not have time to see, be they friends of the president or important members of civil society (chairmen of large public and private companies, trade union leaders, heads of charities and NGOs etc.). This task of protecting the political interests of the president partly implies managing the president's relation with the main media, alongside the press secretary/conseiller de presse. They can meet with political journalists to gather as much information as possible for the head of state, select those who would require an appointment with the president and explain the president's agenda.

This gatekeeper role extends to the president's team of advisors, of which the SG and the CoS are the managers. They decide whom and what the president needs to see. Most presidential advisors rarely see the president, and their role is to organize the visibility of staffers, to enhance talents but also to avoid the frustration of being kept away from the centre of power.⁴

The task of the CoS in the White House is, in this matter, more difficult. There are more advisors, and significant policy advice organizations such as the National Security Council, within the presidential office. Further, while the policy remit of the president of France is limited, that of the president of the US extends across the whole federal government. More policies, more people and potentially more conflicts among them, must be managed by the CoS. Therefore, the CoS has a staff of their own, and that staff must balance their loyalty to the president with that to the CoS.

Administrator/manager

The literature on advisors somewhat undervalues the managerial role played by officials such as the CoS and the SG. They are not just providing policy and political advice, they are also managing staffs of other advisors, legal counsels and perhaps leaders of operating agencies as they seek access to, and influence over, the president. This idea is summed up by a CoS quoted in Patrick Weller (2018: 88) stating, ‘I manage so he can lead, because if he has to manage, then actually something is probably getting a bit wrong.’ Rhodes and Tiernan (2014) also point to the importance of the CoS as a manager on behalf of the chief executive. As such, an important mission common to both the CoS and the SG is to organize and manage the team of presidential advisors. It implies selecting senior White House or Elysée staffers and supervising their offices' activities. They must also control the impressive flow of information directed to the head of state, attempting to check the messages flowing out of the office, but also of managing the daily emergencies that interfere with the head of state's agenda. For instance, the collaborators attached to the General Secretariat of the Elysée Palace participate in the daily interdepartmental meetings convened by the Secretary General of the Government and the prime minister's chief of staff at the Matignon Palace, meetings where draft legislations and regulations are processed.

In the White House, advisors may be involved in policy meetings involving both the president's own staff and the cabinet. The advisors may not be called upon to give an opinion on behalf of the president on an issue, unlike the SG or CoS, who can act as the president's messenger and give opinions on subjects that the CoS has discussed with the president, or which the CoS considers do not deserve to be brought to the president's attention. As such, they have a direct influence on the orientations of presidential policy and the directives sent to the government, even to the ministers themselves.

Both SGs and CoSs also play a Spiegelprinzip to monitor other policy actors, especially those with particular significance to the president (defence, foreign affairs, economy). The SG and the CoS, and their respective teams, are responsible for liaising with the ministers' advisory teams to ensure that presidential directives are carried out and that the ministers do not emancipate themselves from the tutelage of the chief executive.

This function plays an increasing role in both political systems. In France, a first turn was taken under the presidency of Nicolas Sarkozy and his SG Claude Guéant (2007–2011). The president requested his ministers sign a ‘letter of commitment’ prepared by the SG that set out several objectives for them, thus reducing their political independence. If this principle was formally abandoned, the Elysée's control over the ministers and their advisors remained under François Hollande's presidency, and later became central again under Emmanuel Macron and his SG Alexis Kohler (2017–). More than ever, such tutelage accompanies the increase of the president's power in the French political system. The control of the US president over his cabinet is not so strong, although all clearly know they serve ‘at the pleasure of the president’, and know that the president can use his power to issue executive orders to legislate in their policy domains.

Political/policy advisor

In France and the US likewise, both the SG and the CoS also take on more political prerogatives. They participate in the process of government composition and reshuffle, either upstream by establishing contact with the ministers who are expected to join, or by comforting those who see their hopes dashed, or who are called to leave the government. In France, the SG even announces the composition of the government on the steps of the Elysée Palace, an emblematic scene that most often sums up his role for the public. The SG coordinates relations between the president's office and the prime minister's office, organizing regular meetings between the two entities. He prepares the Council of Ministers with the help of his counterpart, the Secretary General of the Government (Eymeri-Douzans and Mangenot 2019).

Alongside the CoS, the SG also plays an essential role in the appointment process for major administrative and political posts, whose influence on the transformation of the state apparatus is easily measurable. This influence on appointments is essential in that it directly guides the policies implemented on both sides of the Atlantic. Although patronage appointments in the public sector tend to be regarded negatively, all governments do fill some positions through political appointment. The two positions studied in this article tend to be the types of positions for which patronage is considered appropriate. These officials must work closely with chief executives and share confidential information – both public and personal – with those executives, and therefore may need qualities that would be impossible to assess in any merit system.

We can think of patronage appointments as being based on alternative forms of trust, and as having alternative responsibilities (Panizza et al. 2021). Patronage appointments may be made on the basis of the allegiance of the appointee to the party. Appointments in the upper echelons tend to be made on the basis of personal trust, which is definitely the case for the officials with which we are concerned in this article. Personal loyalties are much more important for the CoSs and SGs than are party affiliations, although in the US these officials have often had connections to the political campaigns of the president.

Individuals in patronage positions may be given a variety of different tasks, ranging from policymaking to political advisors to being personal ‘minders’ for political leaders (Connaughton 2010). In the typology mentioned above, policy and politics were separated, if perhaps artificially. For these two officials in France and the United States, the policy and political tasks are fused and inseparable as they direct, manage and oversee policy development. These positions are in the gift of the president, but with the honour comes an immense range of intertwined responsibilities.

Assisting the presidential appointment of the heads of public companies, the directors of the central administration, and various administrative functions in all fields lies in the mission of the SG in France and the CoS in the US. For the SG, this role intervenes in the allocation of public jobs, notably in the selection of candidates, and even in the final decision. It can be significant, depending on the SG's personality and the appetite of each president to influence appointments. Jean-Pierre Jouyet, François Hollande's SG (2014–2017), was nicknamed by the latter the ‘HR of the Republic’ because of his interest and influence in these processes.

While in France, personnel decisions are made by a larger number of executives, in the United States, the Chief of Staff's team responsible for evaluating candidates for public office shares common ground, but is much larger in number. Typically, the Presidential Personnel Office employs about 90 people, although under President Trump it was reduced to only 30. This may appear to be a very large number of people for this task, but over 4,000 positions must be filled by appointment, and it is important both for governmental and political reasons to get the personnel correct. This staff is, of course, especially busy in the early days of an administration, but the typical length of tenure for appointees is under two years, so there are always positions opening up. The role of the CoS is especially important during the roughly ten-week transition period between presidents, when a new team must be built and be ready at noon on 20 January to take over government.⁵

Proxy/confidant

The role as the president's proxy is one of the other roles shared by the CoS and the SG. It sometimes, if not often, involves a personal relationship between the chief executive and their closest advisor. In France, this was the case for the first SG in modern history, Geoffroy Chaudron de Courcel (1959–1962), De Gaulle's aide-de-camp during World War II. Often, the SG is a long-time advisor to the president, a trusted figure from life before the presidency. This was also the case with Michel Jobert (1969–1973) with Georges Pompidou, for whom Jobert had been the head of cabinet when the latter was prime minister, or with Claude Guéant, Nicolas Sarkozy's first SG, a long-time collaborator and confidant at the Ministry of the Interior. This is also true with the current SG, Alexis Kohler, who served as head of cabinet for Emmanuel Macron when he was minister of the economy and is portrayed by the media as ‘the president's twin’ (Martin et al. 2021).

Relationships can also intertwine public and private life, reinforcing professional bonds with personal ties (Martigny 2023). Socialist president François Mitterrand appointed the son of an old World War II friend, Hubert Védrine (1991–1995), as his SG, while François Hollande hired his best friend, Jean-Pierre Jouyet, to the position. The relationship can even become family, as when Frédéric Salat-Baroux (2005–2007), who served as the SG of Jacques Chirac, ended up marrying the president's daughter (after the end of the presidential term).

The same patterns can be found in the United States, although the career backgrounds of the individuals involved are somewhat different. Rather than being from the higher civil service, most CoSs have been involved politically with the president, or are personal friends of the family even from childhood – such as Ken O'Donnell (1961–1963) for John F. Kennedy. The closest the American officials have come to a civil service background is the presence of a number of ‘in and outers’ who move back and forth between the public and private sectors depending on which party has the presidency. Three retired generals have also served as CoSs. George W. Bush's first CoS, Andrew Card (2001–2006), was ‘inherited’ from his father's presidential staff. Even if the connections as the ‘nobility’ of the state may be less clear than in France, the personal connections and the ability to work with the president are as essential for the CoS. Given that many American presidents have been outsiders to the usual circles of power in Washington, this poses something of a dilemma for a president. The president wants personal loyalty and even friendship, but also needs someone who can be effective within those circles of power. Presidents who have opted for friendship over familiarity with Washington – for example, Jimmy Carter with Hamilton Jordan (1979–1980) – may be disappointed.

These similarities in roles, although real, need to be nuanced on some points. The nature of the two functions diverges in one important respect: SGs tend to remain in office longer than CoSs. If we compare SGs during the Fifth Republic (66 years) with those of CoSs beginning with the Eisenhower administration (72 years), we can see that the SGs tend to remain in office longer on average (3.25 years, while CoS remain only an average of 1.97 years). If we remove the two long-serving CoSs under Eisenhower the average becomes 1.85 years for the CoS, and if we remove the two interim presidents in France, the figure there rises to 3.66 years for SGs. Andrew Card (2001–2006) and Denis McDonough (2013–2017) are exceptions in the modern era of CoSs, each having served at least one entire presidential term of office, under presidents George W. Bush and Barack Obama, respectively.

The seemingly slight difference in the length of office of these officials masks some differences in the role of these officials. The SGs represent the continuity of the state in a more étatiste (state-oriented) political system, and their greater length of service has symbolic as well as practical significance. The SG of the Elysée Palace is at once more political than the White House CoS, and more of a mandarin – a civil servant. Conversely, the shorter terms of the CoSs reflect the ‘in and out’ nature of government in Washington, with many officials coming and going, matching changes in the political climate as well as changes in what a president may need at a particular time of his administration. For example, it has been argued that President Biden selected Jeff Zients (2023–) as his new CoS because of his ability to handle the management of the White House effectively while the president prepared for another electoral campaign (Baker et al. 2023).

That description of the differences is perhaps too stark, and to some extent appears to diminish the role of the CoS. In addition to managing the political functions, the managerial tasks imposed on the CoSs are such that they must focus much of their attention on making the office run effectively (Cohen and Hult 2021; Cohen et al. 2012b). The CoS is embedded in, and manager of, the Executive Office of the President with approximately 4,000 employees, most of whom are engaged in policy work. Simply keeping that work flowing and coordinating the various policy advice offices within the EOP is a daunting administrative task. The SG, in contrast, is content to administer the presidential team, and even then, only the technical and political advisors, without direct responsibility for the organization and daily life of the Elysée Palace. With fewer than 100 people within the Office of the President in France, and with many of them having clearly less significant roles and not being under the SG's responsibility, the SG may be able to handle the task in a more personal manner than is possible for the CoS in the more bureaucratic White House. The SG's authority is exercised over about fifty collaborators (56 in the last count in 2023), a number that has been steadily increasing in the Fifth Republic (Elgie 2000), SGs of the French presidency are very much mandarins, performing tasks that are more like those of a higher civil servant dealing with important policy matters for a minister.

The CoS in the White House is more a manager than a mandarin, although not formally designated as the manager of the presidency. He is certainly giving advice to the president, but much of the role is managing all the other sources of policy and political advice that exist within the EOP. The CoS is responsible for managing the flow of paper (or now electronic information) within the White House and the Executive Office Building. The president cannot possibly read all that information, so the CoS must manage it and determine what is worthy of presidential attention. Also, the EOP contains functions within it, such as budgeting, that do not exist in its French counterpart so that, although the CoS does not manage the budget process directly, it is clearly more connected to the CoS role than would be true for the SG. Indeed, CoSs are often major negotiators with Congress during the budget process (Cohen et al. 2016).

It is finally important to note that the nature of the roles played by these officials is to some extent dependent upon the individuals involved, and that variance is perhaps greater in the United States than in France. In France, the function is the subject of a roadmap largely defined in advance and informed by the previous holders of the function. Pierre Bérégovoy (1981–1982), who assumed the position in the early years of François Mitterrand's first mandate, largely guided the nature and organization of the SG's missions through his prescriptions. His successors, with one exception – that of Claude Guéant under Nicolas Sarkozy – did not stray far from a largely routine practice of this job. Those CoSs that have close personal relationships with the president may be more political advisor than those whose connections are more remote. For example, during the Obama administration the ‘Chicago Clan’ was a closer political clique than has been true for many other presidents (Cohen et al. 2012a). Further, presidents who are outsiders to Washington – such as Jimmy Carter or Donald Trump – have found it difficult to find CoSs with whom they were comfortable and who could be effective in dealing with the rest of the federal government.

In France and the US alike, though, the essential mission of organization of the Office of the President is accompanied over time by a rise in the means placed at the disposal of the SG and CoS. We have already mentioned the increase in the number of staffers, but the budget is another illustration of this rise in resources. In France, the budget for the Elysée Palace, which includes a significant proportion of staff, has exploded in 40 years, rising from 3.1 million euros under François Mitterrand to over 100 million under Nicolas Sarkozy, a figure that has remained relatively stable since then (Dosière 2019). At the White House, where the staff, as we have pointed out, is much larger, the budget allocated to the Executive Office of the President has also increased significantly since the early 1990s, from $194 million in 1993 to $714 million in 2022.

Key actors of presidentialization

As we have already stated, our argument in this article is to demonstrate that the power of the CoS and the SG is largely dependent on the nature of the presidential systems in France and the United States. This nature is both static – fixed by the institutions and power practice – and dynamic in that it evolves over time in the direction of further presidentialization. It is interesting to observe this phenomenon through the legal status, visibility and rivalries entertained by these two key advisors. It is also to be found in the evolution of careers of both SGs and CoSs after serving the presidency.

Legal invisibility

What also brings the SG and the CoS together is the lack of legal status. There is no decree governing the conditions of appointment, the duties or the duration of the position of SG, or that of CoS. What better sign of a power that is not limited, as in parliamentary systems, by the legislature? While transparency and accountability are becoming central issues in political life, the president's entourage continues to escape such scrutiny. Their power is all the more extensive that it feeds on this informality, and on the absence of regulatory limits. This situation has its downside for the SG and the CoS. In both cases, personnel decisions hinge on the president. Furthermore, the SG and the CoS have no power to sign decisions, not even decrees, which are simple acts. In France, the SG does not even have the power to manage the budget of the Elysée Palace, which is traditionally entrusted to the head of the cabinet.

The absence of any legal existence of the SG, and more generally of the SG's team, is due first and foremost to the desire of several presidents – notably Georges Pompidou, François Mitterrand and Jacques Chirac – not to grant a formal existence to the Office of the President, in order to avoid unbalancing the institutions of the Republic (Martigny 2021a, 2021b, 2023). It can also be explained by the presidents' desire to ensure strictly discretionary power over their closest collaborators, through a structure entirely organized around and through the president (Foucaud 2010). The personal link between presidents and SGs – often chosen from their political entourage, or even from their circle of friends – reinforces this situation. For similar reasons several American presidents – for example, Gerald Ford and Jimmy Carter reacting to the role played by H.R. Haldeman (1969–1973) in the Nixon administration – have attempted to eliminate the role of CoS, a step that proved to be impractical.

Thus, the strong discretionary dimension that characterizes the presidential function goes hand in hand with the legal unaccountability of the president's advisors (Foucaud 2010). As the influence of the president rises in time within the executive power, presidents and their staff have bypassed the role of the government and parliament. This erosion of checks and balances is one of the direct consequences of prerogatives that nothing limits, except the authority of the president themself, and that no one evaluates or sanctions in case of failure. In France, failures can therefore only be sanctioned by the president. However, practice shows that in fact such a situation never occurs. The political responsibility of Dominique de Villepin (1995–2002) in the failed parliamentary dissolution of 1997 decided by Jacques Chirac did not have any consequences on his career at the President's Office. As with the indictment in 2018 of the current SG of the Elysée, Alexis Kohler, in a case of illegal influence peddling, for the time being closed without action, it has not had more consequences.

American presidents and their staffs, however, have more constrained powers, and when they attempt to use decree powers, as President Biden has with student loan forgiveness, they may find themselves losing in court. The president does have decree powers but must be careful not to impinge upon the Constitutional powers of Congress when using them. Likewise, the CoSs may be more constrained and may suffer when they overstep the unwritten, but clear, limits of their roles. Those limits may arise in their relations with other members of the executive branch, with Congress, and even with the president. These limits are very much a function of the backgrounds of the CoS, with one with extensive congressional experience, such as Leon Panetta, having fewer barriers when dealing with Congress (Sullivan 2004).

High-levels of discretion

The extent of the power of an advisor who in France is sometimes referred to as the ‘non-elected vice-president’ or ‘the most powerful shadow man in France’ has as another point in common with the CoS – an unfailing discretion at odds with the logic of control of executive power by parliament or the courts in parliamentary systems. What characterizes the ethos of both SGs and CoSs is a mixture of absolute loyalty to the president, and discretion, even organized secrecy around their actions. In France, the nature of this invisible power has in the past generated accusations of organizing a ‘shadow cabinet’ outside of any democratic legitimacy. If this accusation is not totally unfounded, as we will argue, it does rather hinge on a sense of service to the state and a professional ethic made of abnegation, a sense of duty and discretion. The argument put forward by the holders of these functions to explain such organized invisibility of their role is precisely the desire not to hinder the direct relationship between the president and the people. When CoSs, most famously Alexander Haig (1973–1974), have made themselves more visible and seemed to attempt to assume power, their actions have not been well received.

As a control tower and head of the presidential team, the SG and the CoS think of themselves above all as cogs in the service of their president and of the nation. The CoS in the United States must manage a second balancing act with the vice president. Although the CoS may be closer to many major decisions, and certainly closer to the day-to-day management of government than is the vice president, the CoS should not place the already almost invisible at times vice president further in the shadows. Thus, some of the roles played by the SG in France may be less appropriate in the United States.

The above said, the CoS can play a political role by dealing with difficult situations within government that might alienate other powerful officials if the president, or even the vice president, were to undertake the task personally. The CoS must become the ‘abominable no man’ with the White House, conveying difficult decisions to other members of the president's staff or perhaps cabinet secretaries (Pfiffner 1993). CoSs play that role even though their position may appear weaker than others in the White House Office, given that they lack Senate confirmation.

The invisibility of these advisors has been described as an expression of unchecked presidential power and has been put under pressure from the media. The desire to increase democratic transparency goes along with a growing interest in the behind-the-scenes workings of politics, showing the increasing interest of the public in these shadowy functions. Such a development is not unique to the SG and the CoS, but concerns all the president's advisors, and even the entire political world. It does, however, raise questions about the ability of the executive branch in the future to maintain full discretion over the modalities of its action within the presidential office.

This discretion has changed over time as executive power has increased. In France, the ‘hyper-presidential’ turn taken by Nicolas Sarkozy between 2007 and 2012 led to the spotlight being shone on his SG, Claude Guéant. At a time when ‘the president is governing, governing more, more than before’ (Duhamel 2008), media attention is focused on the person of an omnipresent president. This attention benefited the SG, who was pushed to the forefront for the first time to present the president's decisions in the media, starting with the presentation of several major new programmes in place of the prime minister. The departure of Claude Guéant from the President's Office in 2011 ended an extensive practice of the function, and his successors returned to a discretion closer to the traditional conception of a shadow collaborator. However, the omnipresent power assumed by the SG at that time has become the rule, and the majority of the SGs who succeeded him have retained very strong prerogatives, characteristic of the new powers assumed by the French president.

The American president, as noted, is becoming more presidential, but that general upward trend has many ups and downs in it. Changes in the power of CoSs in the United States have therefore been less linear. Much of their role in governing depends on the style of the president. For a highly personalized president such as Donald Trump the role was almost a barrier to that personal power, while for a more corporate president such as George W. Bush there was substantial delegation. Trump has been the exception, however, in the course of a general enhancement of presidential power, and time will tell whether his second mandate will mark an evolution in presidential power.

Institutional rivalries

The nature of the French and American presidential systems, and the progression of executive power over time, lead the SG and the CoS to go beyond their roles as simple advisors. Both the CoS and the French SG must deal with rivals for power and for influence with the president, but the nature of those rivalries is different. In the US, this rivalry can come from within the Executive Office of the President. The most important of these is the director of the Office of Management and Budget (OMB). Preparing the annual budget to be sent to Congress is a central responsibility of the president, and the director of the OMB therefore has access to the president and can at times circumvent the CoS. In three cases directors of the OMB have become CoSs. Also with the EOP there are several policy councils, such as the National Security Council, that provide the president policy advice. The leaders of these policy councils are often personal friends and confidants of the president, and therefore managing them and their flow of information to the president is challenging. In some ways the CoS must be a ‘chef of staff’, trying to bring multiple ingredients to make a workable and palatable policy mix for the president.

Yet, rivalry is most of the time external, as the CoS must deal with the ultimate rival of the executive branch, Congress. While the White House is engaged heavily in developing policy ideas, most of these will have to be adopted by Congress if they are to become law. The White House CoS must therefore spend a good deal of time ensuring that the relationships between the White House and Capitol Hill are as congenial as possible, meaning that one of the CoS's missions is easier when the president and Congress are from the same political party. The CoS may also be the president's emissary to Capitol Hill, working with members of Congress directly to facilitate the passage of legislation (Patterson 2008). When presidents fail to advance much of their policy agenda, the blame may fall at least in part on the CoS, as it did for Rahm Emmanuel (2009–2010) in the Obama administration (Pfiffner 2011).

In France, the Elysée SG confronts a somewhat different set of rivalries for influence with the president. Although SGs may occasionally find themselves in competition with other presidential advisors, such as the director of the cabinet or the president's ‘special’ (read political) advisor, the SG's pre-eminence is not disputed internally. The role as a one-man band and a jack-of-all-trades in assisting presidential decision-making makes SGs both above the fray and endowed with an aura that is also linked to their technical and dispassionate knowledge of state affairs.

Yet, the French SG must contend with the prime minister. While in a semi-presidential system the prime minister may not have the political power of that office in a parliamentary system, or of Congress in the United States, he or she is a significant political actor, especially during periods of cohabitation or the absence of a clear majority in the lower house of the French parliament, the Assemblée Nationale.⁶ However, his fragility in the institutional system makes him or her dependent – except in the case of cohabitation – on the president of the Republic, who appoints and dismisses prime ministers without institutional constraint (Thiébault 2016). This situation has led to periods of competition with the SG. In theory, this situation is made impossible by the fact that the SG does not entertain any relations with the parliament, in the name of the separation of powers. The SG is supposed to act as a facilitator between the presidential and prime-ministerial cabinets. In practice, however, the rise of presidential power in the last decades has led the chief executive to use the SG as a weapon to weaken the prime minister. Under François Mitterrand's second mandate, Prime Minister Edith Cresson accused the then SG Hubert Védrine (1991–1995) of bypassing her by directly dealing with ministers who challenged her authority. But the most striking example is the notoriously execrable relationship between Claude Guéant (2007–2011) and François Fillon, then prime minister of Nicolas Sarkozy, fostered by President Sarkozy, who had decided to weaken the prime minister in order to assert the hyper-presidential turn of his term.

Life after serving the president

Another indication of presidentialization in the United States and France, manifested through SGs and CoSs, is the development of their respective careers after leaving office. Until the 1980s, after leaving the Office of the President, the SG generally continued a career in the public sector, sometimes in banking. Only a minority went into active politics, entering the government following the example of Michel Jobert (1969–1973), who became minister of foreign affairs in the 1970s and 1980s, or Edouard Balladur (1973–1974), who, years after leaving the service of Georges Pompidou, became prime minister – between 1993 and 1995 (Tricaud and Willaert 2009). The choice to enter active political life became much more frequent after 1981 – it has involved 7 of the 11 holders of the post since that date. This development reflects the more political nature of the office, and the greater temptation, if not incitement by presidents themselves, to switch to active politics. Some SGs became ministers (Jean-Louis Bianco, Hubert Védrine, Dominique de Villepin, Claude Guéant), even prime ministers – in the case of Pierre Bérégovoy, Dominique de Villepin and Jean Castex; members of parliament (Philippe Bas) or heads of local executives (Jean-Louis Bianco). Emmanuel Macron, deputy secretary general of François Hollande between 2012 and 2014, even managed the feat of being elected president of the Republic instead of his mentor!

As in France, some CoSs in the US have gone into electoral politics, with one becoming vice president – Dick Cheney, who had been Gerald Ford's CoS (1975–1977). Other CoSs have become members of Congress (Cheney and James Jones from the Johnson administration). Others have become cabinet secretaries (Alexander Haig at State, Donald Rumsfeld and Leon Panetta at Defense and James Baker at Treasury); Panetta was also director of the CIA. One of Barack Obama's CoSs, Rahm Emanuel, became mayor of Chicago. Two of President Obama's CoSs have also taken major positions in the Biden administration: Denis McDonough became cabinet secretary and Rahm Emanuel ambassador to Japan.

Conclusion

Beyond their undeniable similarities, the two functions have marked differences in their role and career patterns, linked to the singularity of each political system. While both are confidants of the president and principal advisors, the CoS, more than the SG, must also be a manager of a large organization. In addition, the CoS must deal with an independent Congress, which is often controlled by a party different from that of the president. The tasks of the SG may be somewhat similar during periods of cohabitation, but the continuing struggles to get the president's programme through Congress is a central and specific challenge for the CoS. As politics has become more polarized around the world, other leaders in central agencies will face more of these problems, but that role now appears to set the CoS apart.

Yet, we have shown that what SGs and CoSs have in common is the nature of their role in the entourage of presidents. More importantly, their lack of legal status and the rise of their powers say a lot about the evolution of the presidential function in both countries. In this respect, it is possible to compare them with other high-level advisors in parliamentary systems, such as the chief of staff to the British prime minister in Downing Street, the chief of staff to the Prime Minister's Office in Canada, the Moncloa chief of staff in Spain or the head of the Federal Chancellery in Germany. The acceleration of political decisions and the growing role of the executive branch in most contemporary democracies are thus powerful explanatory factors for the increased importance of these advisors, despite different political cultures and institutional systems.

The crux of our discussion is that, despite the differences between the SG and the CoS, the roles they play in relation to the president are singularly comparable and reflect the exorbitant and growing weight of the executive in political decision-making. From the perspective of the two key actors in the presidencies that we have examined, this phenomenon is real and demonstrates the growing role over time of the president's advisors in political decision-making, a reliable indicator of the ‘increase in power resources’ referred to by Poguntke and Webb (2005) as one of the criteria for presidentialization. This power can sometimes lead to situations of competition between these advisors and other parts of the institutional system: the prime minister in France, and Congress in the United States. The quasi-regulatory discretion of these operators in the shadows, and their unwavering loyalty to the president they serve, is an indispensable condition in this context, if the system is to avoid excessive tensions generated by presidentialization.