IN MOST DEMOCRATIC COUNTRIES, the pace of creation of independent bodies charged with regulatory and oversight functions that had previously been entrusted to “ordinary” bureaucratic departments increased in the last two decades of the twentieth century. In the United Kingdom these new institutions are called “nondepartmental public bodies” or “quasi-autonomous nongovernmental organizations” (quangos). In the United States, they are “independent regulatory agencies”. In France, they are “independent administrative authorities”.
Although quite diverse in character, all of these organizations share a certain hybrid quality: they have an executive dimension even though they also exercise normative and judicial functions. The traditional concept of separation of powers has had to be stretched to accommodate them. The scope of the change has been considerable. In many countries vast areas of government intervention have increasingly been entrusted to these new organizations, clearly reducing the scope of administrative-executive power. Indeed, in the United States, this was explicitly cited as the reason for creating them.
The United States was the first country to set up independent authorities. Indeed, it did so quite early on: it was at the end of the nineteenth century that an institution of this type, the Interstate Commerce Commission, was created to regulate the railroads.
In other words, this first step outside the administration was taken even as administrative power itself was being strengthened. Both the growth of the administration and the need for extra-administrative bodies reflected a common concern: to develop institutions that would serve the general interest.
In the 1880s the federal government in the United States was still relatively weak, and its responsibilities minimal. The West was still being conquered, and the federal government’s main concern was to organize this expansion and secure the nation’s frontiers. The state was built on such basic functions as defense and postal service.
The situation of the U.S. government thus stood in striking contrast to that of European governments, which already boasted of administrative hierarchies active on numerous social and economic fronts. It was during this period that the welfare state was built in Germany, the civil service was rationalized in Britain, and the Third Republic was consolidated in France.
By contrast, in the United States public institutions had not evolved much despite the country’s rapid social and economic growth. Hence there existed what Stephen Skowronek has called a “governance vacuum”.1
After the end of the Civil War, the political parties and the courts had rushed to fill this vacuum and establish themselves as the forces that really had the ability to shape the country’s future.2
At first real power lay with the “party machines”. Coordination and initiative at all levels of government belonged to them. Party “bosses” were more important than elected officials, especially at the local level. It was the boss who pulled strings behind the scenes, especially when it came to awarding public works contracts. The word party became synonymous with “corruption,” and corruption became the major political issue in the United States in this period.
A new type of journalism emerged in response: “muckrakers” exposed the corrupt activities of politicians whom they accused of behaving like “new czars”. But the most effective counterweight to the parties’ “privatization” of the general interest was the courts. To battle party hegemony and compensate for the weakness of the state, various federal courts tried to exert their influence. Their decisions were intended to fulfill a social regulatory function and reaffirm a notion of the common good.
The influence of the judiciary had been strengthened by the dismantling of the war administration after the Civil War. The Supreme Court saw its role expanded in this period. In the aftermath of the conflict, it had laid down new principles that would constrain the bureaucracy in the years to come. But the judges were so disgusted by the incompetence and corruption of the politicians that they became champions of business and laissez-faire capitalism. In their eyes, it was progress to deny a corrupt political system control over society’s activities. The Supreme Court went so far as to attempt to make laissez-faire a constitutional principle. Paradoxically, the actions of the courts thus converged with the influence of the parties to prevent the creation of a truly public system of regulation.
It was in reaction to this that an important new reform movement began to emerge in the 1880s. The need to regulate the railroads served as a catalyst. The issue was in many respects central. Economically, the railway system played an obviously essential part in the development of trade and industry in a country as vast as the United States. In terms of governmental structure, the problems turned out to be too big for the states to deal with on their own.3 In the 1870s and 1880s, a variety of social movements had spread among farmers, who sought to pressure the railroads to end discriminatory pricing practices in the transportation of grain. Circumstances thus conspired to make railway regulation the ideal place to introduce an exemplary reform.4
The Interstate Commerce Act of 1887 was therefore a major milestone, the symbol of a new approach to public administration. First, the new law laid down rules governing railway freight tariffs and prohibited discriminatory pricing. More than that, it set up an independent agency, the Interstate Commerce Commission, to implement these rules and regulate the railroads generally. The commission marked a break with traditional ideas about the role of the bureaucracy.
The decision to stake out new territory was not taken lightly. Congress had devoted much thought to the idea of creating an ad hoc commission rather than delegating responsibility to an existing department of government.5 The lawmakers’ major concern was to “keep politics out” of the business of regulating a sector of the economy of vital importance to the nation’s general interest. They also wanted to “nationalize” as well as “depoliticize” the railroad issue, which by its very nature could not be dealt with effectively at the state level and which was further complicated by corrupt party influence. In short, Congress recognized that the federal bureaucracy of the day was not inherently well suited to defend the general interest. It also doubted the ability of the executive branch to serve the common good.
The “political” argument proved decisive in America, where the party system had gone awry. But it was not the only argument in favor of change. Three other factors played an important role.6 First, regulation required a high level of expertise. The existing bureaucracy lacked the requisite knowledge of railroad pricing structures, competition in the railroad industry, and relevant safety standards. Lawmakers also saw a need for a more flexible system of regulation, one capable of evolving in the future as railway technology developed; a bureaucracy used to applying fixed rules in a mechanical fashion would not do.
Finally, it was important that the new institution possess a capacity for arbitrage: it would be required not only to resolve disputes between railroad companies but also to protect customers dispersed over a wide area. It would need to find legal means to shift the existing balance of power between railroads and their customers, taking the interests of all parties into account. And it would need to go beyond a merely formal view of the situation and somehow bring the diversity of social interests into its calculations. It was thought that a commission could “serve as the poor man’s court”.7 Thus the usual boundaries between political, administrative, legislative, and judicial powers were blurred.
To be sure, doubts and criticisms were heard. Some feared that the law and the bureaucracy would be weakened. But the arguments in favor of an independent commission easily carried the day. Although the term independence does not appear in the text of the 1887 law, the idea was nevertheless central to the system. It was embodied in the principle that the new commission must be bipartisan. Of the five commissioners, no more than three could belong to the same party. Appointed by the president with the advice and consent of the Senate for a term of six years, the commissioners could not be removed unless found guilty of corruption, negligence, or incompetence.
In other words, the commission was to be protected from the influence of the party in power. Steps were taken subsequently on several occasions to increase the autonomy of the Interstate Commerce Commission and expand its prerogatives.8 The election of Benjamin Harrison as president in 1888 encouraged Congress to reinforce the independence of the new institution. Harrison had worked as a lawyer for the railroads, and this fact made the senators and representatives even more wary of executive power.9 “Political” independence therefore required “functional” autonomy with respect to the executive branch.10
The ICC subsequently became the model for a number of other independent regulatory agencies.11 During the New Deal, Congress approved the creation of several new agencies that would play a key role in later years: the Federal Communications Commission (1934), the Securities and Exchange Commission (1934), and the National Labor Relations Board (1935). The contingent character of each of these institutions is undeniable. In each case the action can be explained by weakness or suspicion of the president, Congress’s inability to solve a problem on its own, or a desire to curtail the power of an existing department of the bureaucracy.12 Ultimately, however, what was established was a novel system for expressing and managing the general interest.
The independence of the new agencies was confirmed in 1935 when the president, Franklin D. Roosevelt, sought to remove a commissioner of the Federal Trade Commission whom he deemed hostile to the New Deal. In a unanimous opinion, the Supreme Court denied that the president had the power to remove a member of the commission ad nutum, ruling that the FTC exercised quasi-legislative and quasi-judicial functions and was in no way subordinate to the president.13
In 1944, a government task force on regulatory commissions issued a report that for the first time offered a theory of the independent agency model. It emphasized four features: isolation from political pressure and independence of the executive; impartiality; ability to implement long-range policies not subject to the vagaries of elections; and ability to formulate coherent, rational policies. The report also emphasized the fact that the agencies enjoyed a positive public image and that the public accepted their decisions more readily than it did the decisions of the ordinary bureaucracy. A few years later, another report, this one by the Hoover Commission, confirmed these findings.14
In the United States, the weakness of the bureaucracy and the prevalence of political corruption created the conditions for the establishment of the first independent agencies. A century later, similar circumstances attended the creation of similar institutions in a number of countries.15 Elsewhere, especially in Europe, events took a different course. Independent authorities initially proliferated in response to demands for regulation in areas where existing bureaucratic structures had run into difficulty owing to technical complexity, overlapping competences, multiplication of affected parties, and/or diffusion of responsibilities. There was also a need to overcome a deficit of democratic legitimacy.
Significantly, these independent agencies took root even where the democratic sovereign or “Jacobin” state seemed most deeply rooted and legitimate. For that reason the French case is an interesting point of comparison for understanding changes affecting democracies in general.16 The first agency of this new type was the National Commission for Computers and Freedom (CNIL), which was created in 1978. The Senate suggested that it be designated an “Independent Administrative Authority”. This suggestion won out over the proposal by the National Assembly to make the new organism a simple department of the Justice Ministry.
This was a political choice, which followed an announcement by the government that it intended to issue an identification number to each citizen, to be used on all official records. This announcement alarmed the public, and the opposition denounced it as an “attack on freedom”. The government, seeking to prove its good intentions, appointed a commission of wise men, who were charged with finding a solution to the problem that everyone could accept.
After much consultation and debate, this commission suggested that a new independent agency be created, and this recommendation became the basis of the bill ultimately submitted to the legislature. In short, it was because the executive recognized that the public would not accept the legitimacy of a regular bureaucratic department in this capacity that it proposed passing the responsibility to an independent agency, and the legislature eventually followed suit.
Although it was never made explicit, this decision reflected the idea that a suspicion of partiality amounted to a denial of legitimacy (the charge that the new authority represented an “attack on freedom” indicated that the government was suspected of partiality). Hence in practice, the legislation amounted to making a distinction between electoral legitimacy and the legitimacy of impartiality. It was also a recognition of the insufficiency of the majority principle.
Distrust of an executive deemed to be partisan also appears to have played a decisive role in the creation of certain institutions. This was the case with the High Authority for Audiovisual Policy, which was established by the law of July 29, 1982.
Here, too, there was a problem of regulation. The broadcast media were evolving rapidly, with the creation of numerous “free (i.e., non-government-controlled) radio stations,” the introduction of new technologies (cable and satellite), and the resulting challenge to the state monopoly and the conception of the broadcast media as a public service.17 The heart of the issue lay elsewhere, however: there was no escaping the fact that an elected government could not be trusted to be an impartial agent of the general interest in an area as sensitive as news and information.
The executive branch therefore voluntarily renounced one of its powers in order to shore up its tottering legitimacy. By shifting responsibility for the regulation of the broadcast media to an independent authority, it hoped that some of the virtues of that independence would redound to the benefit of the government itself. That is why the decision was touted as an extremely important one and widely publicized, including a speech by the president of the Republic.
To be sure, establishing the High Authority was not easy. It took three governments and no fewer than three bills, each modifying the composition and powers of the new agency. Eventually the name was changed to the Superior Council for Audiovisual Policy. In the end, the break with the previous administrative-political model was dramatic.
For another example, consider the Commission on Stock Exchange Operations (renamed the Financial Markets Authority in 1996). Here the primary goal was regulatory. It was felt that the existing bureaucracy lacked the technical capabilities needed to monitor the markets in order to protect the public, prevent insider trading, regulate mutual fund advertising, and so on. But once again the heart of the matter lay elsewhere, because the bureaucracy could of course have acquired the necessary capabilities. The real reason for preferring an independent agency was the desire to avoid accusations of partiality in a sensitive regulatory domain.
In this case, the concern was inherent in the Colbertist-corporatist French model of administration rather than a consequence of government influence over the bureaucracy. Historically, France had often relied on quiet bureaucratic arbitration of conflicts between the government and various pressure groups. The fact that high-level civil servants were drawn from the same segment of society as most of the top managers of private and public enterprises and financial institutions had facilitated this type of regulation, which relied on working out compromises and avoiding public scandals rather than on strict enforcement of rules.
Each independent authority has its own distinctive characteristics. Each was established for specific reasons, with no overall plan or general model in mind. Taken together, however, the new agencies represent a shift away from the traditional power structure. The creation of these new institutions reflects new social expectations. People want a more open decision-making process and more input from conflicting points of view. They have had enough of centralization and secrecy.
In the end, the rigid “Jacobin” hierarchy, which meshed well with a certain form of corporatism, therefore gave way to a more open and interactive process of policymaking. The “technical” requirements of certain types of regulation have thus converged with “democratic” aspirations for more public justification of policy decisions, greater openness, and above all, greater impartiality. To be sure, tensions between these two motives for reform often manifested themselves, but in the end they combined to foster the emergence of independent agencies in democracies around the world.
It is striking, however, that the new independent agencies have drawn criticism from all sides even as they spread in apparently inexorable fashion. In the United States, the creation of independent regulatory authorities was seen as posing insurmountable constitutional difficulties. The new institutions were accused of constituting “a headless fourth branch of government” and “a haphazard collection of irresponsible agencies and uncoordinated powers”.18 They were seen as disrupting the “normal” structure of government in violation of both the principle that government should be responsible to the people and the sacrosanct tripartite separation of powers.
Indeed, the constitutional status of the agencies was ambiguous from the start. They were created for pragmatic reasons, and little thought was devoted to their theoretical status. As they grew, no doctrine emerged to explain their position in the system. Some scholars saw little reason for concern about the hybrid character of the agencies, preferring to judge the tree by its fruits.19 But American constitutional scholars on the whole remained suspicious of institutions that did not fit easily into their usual categories. They acknowledged that the agencies were in practice perfectly integrated into the national political culture, but despite this they continued to see them as “legal anomalies” of dubious constitutional status.20
In a more directly political vein, the Republican right would make criticism of regulatory agencies one of its warhorses in the 1980s. To Republicans, the agencies, with their certainty of representing the public good against special interests, stood as symbols of the technocratic and bureaucratic arrogance of the Left. They were inextricably intertwined with that hated enemy, the “East Coast liberal elite”.21 Conservative calls to shrink the state reflected a surge of right-wing populism, which created a climate of suspicion but in practice did little to thwart the power of the agencies.
The French situation was not all that different. Originally, as we have seen, the idea of independent administrative authorities flew in the face of the Jacobin idea that sovereignty is “one and indivisible,” and the idea of impartiality was alien to French political culture. During the Revolution, allegories of impartiality were extremely rare, and the word was absent from the period’s major debates.22 It was will—the force that unifies and decides—that was celebrated, and not impartiality, the seat of prudence and sign of a divided society.
In France as in the United States, it was social expectations and practical necessities that led to the opening of Pandora’s box. Hence the introduction of independent agencies had to contend with a persistent residue of doubt and hostility. Most jurists continued to have doubts about their legal status.23 Very few scholars offered reasoned approval. Politicians remained fiercely critical. Many deputies believed that the new agencies were products of “governmental cowardice”. Speaking toward the end of his term in 2006, the president of the National Assembly insisted that “the development of what people chastely [sic] refer to as ‘independent administrative authorities’ reflects the decline of legislative authority”.24
A parliamentary report from the same year allows us to gauge how heavily these doubts continued to weigh on people’s minds. Its tone was on the whole quite negative. Focusing on the most recent of the new authorities, the High Authority for the Struggle against Discrimination and for Equality (HALDE), the report concluded that it owed its inception to a “failure of the judiciary” and “that it would have been better to incite the state attorneys to do more than to give up. It would have been better to identify the jurisdictions in which problems existed and help them do better in the fight against discrimination”.
In the end, the report said, the new authorities were like little states “partially levitated above the traditional state”. As icing on the cake, the lawmakers alluded to the “fundamental ambiguity” of the new institutions and to the “legal oxymoron” they presented.25 They ended with a call to halt the creation of agencies that undermined the traditional structure of government. Yet the same legislators who drafted this report had voted regularly since 1977 to create the very kinds of institutions they now denounced. It was they who in 2003 created the new category of Independent Public Authority.26 How are we to understand this tension, which the report implicitly recognized?
The report noted that independent authorities were often created in response to dysfunctions or failures of the traditional state. The new institutions were thus a “political act indicating the legislature’s distrust of other powers or authorities”.27 This initial distrust may in some cases have reflected the government’s desire to relieve itself of some responsibility because it did not want to make the necessary choices or risk the unpopularity that would follow. Even if this was a “bad reason,” one that diminished the standing of the political authorities, it nevertheless needs to be taken into account.
When lawmakers find themselves in a bind, they may feel that they must do something to restore the credibility of government intervention. Yet even when such intervention is necessary, it may have perverse consequences: “The creation of independent administrative authorities restores confidence in the ‘brand new’ agency, which has no ties to the discredited old system, but at the same time it strengthens the belief that the old system actually deserves to be circumvented”. Under such conditions, the Parliament may be creating a self-fulfilling prophecy: by anticipating the weakness of the traditional, hierarchical, centralized state, it contributes to its demise.
This interpretation of the reasons for creating independent agencies is thus a mixture of justification and denunciation. It also avoids the fundamental structural issue: What creates the distrust and dysfunctionality in the first place? Instead, it assumes that the authority of the traditional state can somehow be restored by creating a new structure outside it.
If distrust of the executive has historically been one of the key reasons for creating independent authorities, there is another factor that should also be taken into account: namely, the existence of a certain “social preference” for institutions of this type. We see this in the United States in the 1940s, and recent research confirms its importance.
Consider the work of the American sociologist James Coleman, who sought to identify the degree of confidence in various types on institutions. This work inspired a study in France concerning the very sensitive subject of nuclear risk.28 Researchers conducted an experiment in which subjects were told that health problems had been observed in the vicinity of nuclear power plants and that an investigation was called for. The subjects were then asked to say which type of institution they trusted to carry out such an inquiry: the ministry of industry, the European Nuclear Agency, or a panel of independent scientists?
Most subjects favored the independent panel, while the administrative department normally responsible for dealing with such issues came in last. Interestingly, the outcome remained the same regardless of whether the health problems mentioned were benign (a slight increase in allergies) or serious (an increase in birth defects). Furthermore, a majority of the subjects who were initially most trusting of political leaders and government preferred to trust a nongovernmental organization when it came to dealing with a sensitive issue.
Other surveys have confirmed that citizens are much more likely to trust NGOs than political parties to come up with solutions to problems that benefit everyone. The social psychologist Tom Tyler has demonstrated the importance of impartiality in evaluating the legitimacy of any action.29 The institutions seen as most objective or impartial were also perceived as being most likely to serve the common good. Thus even elected authorities may be seen as less legitimate than others that have not been subjected to the same “ordeals of establishment”. These findings suggest that we ought to look into the reasons why independent authorities may be accepted as intrinsically democratic. If we want to understand the silent revolution that the advent of these independent authorities represents, we have no other choice.
1 I borrow the phrase from Stephen Skowronek, Building a New American State: The Expansion of National Administrative Capacities, 1877–1920, 1982, p. 41.
2 Skowronek (ibid., p. 39), refers to a “state of courts and parties”. His analysis should be compared with the most recent work of William J. Novak, especially The People’s Welfare: Law and Regulation in Nineteenth-Century America (Chapel Hill: University of North Carolina Press, 1996).
3 See “Early State Experience with Commissions and with Administrative Regulation,” in Robert E. Cushman, The Independent Regulatory Commissions (1941; reprint Oxford: Oxford University Press., 1972), pp. 20–34.
4 See Charles Francis Adams Jr, Railroads: Their Origins and Problems (New York, 1886) for the “mugwump” view of the problem.
5 On the reasons for this decision, see Louis Fisher, The Politics of Shared Powers: Congress and The Executive, (Washington, DC: Congressional Quarterly Press, 1981), pp. 147–148.
6 See Cushman, Independent Regulatory Commissions, pp. 45–61.
7 The formula is taken from Cushman, ibid., p. 48.
8 “The History of Regulation,” in Regulation: Process and Politics (Washington, DC: Congressional Quarterly, 1982).
9 On this key episode, see Marver H. Bernstein, Regulating Business by Independent Commission (Princeton: Princeton University Press, 1955), p. 23.
10 On the history and various meanings of independence, see James W. Fesler, The Independence of State Regulatory Agencies, Public Administration Service no. 85 (Chicago : Public Administration Service, 1942), p. 13.
11 The Federal Reserve System was instituted in 1913 and the Federal Trade Commission in 1914.
12 In this vein, see the interesting history of the SEC by Joël Seligman, The Transformation of Wall Street: A History of the Securities and Exchange Commission and Modern Corporate Finance, 3rd ed. (New York: Aspen, 2003).
13 Humphrey’s Executor v. United States 1935 (295US602).
14 Task Force Report on Regulatory Commissions (Washington, DC: Government Printing office, 1949). On the Hoover Commission report (1949), see Louis Fisher, The Politics of Shared Powers: Congress and the Executive, 4th ed. (College Station: Texas A&M University Press, 1998), pp. 150–151.
15 See the studies collected in Larry Diamond, Marc F. Plattner, and Andreas Schedler, eds. The Self-Restraining State: Power and Accountability in New Democracies (Boulder, CO: Lynne Rienner, 1999).
16 For an overview, see Catherine Teitgen-Colly, “Les autorités administratives indépendantes: histoire d’une institution,” in Claude Albert Colliard and Gérard Timsit, eds., Les Autorités administratives indépendantes (Paris: PUF,1988).
17 Jacques Chevallier, “Le statut de la communication audiovisuelle,” Actualité juridiquedroit administratif, 1982, pp. 555–576.
18 The quoted phrases were employed by the Brownlow Committee in 1937, as quoted in Bernstein, Regulating Business by Independent Commission.
19 One such was Robert E. Cushman, the first historian of the new institutions. See his pioneering article, “The Constitutional Status of the Independent Regulatory Commissions,” I and II, Cornell Law Quarterly, vol. 24, no. 1 and 2, December 1938 and February 1939.
20 See, for example, Peter L. Strauss, “The Place of Agencies in Government: Separation of Powers and the Fourth Branch,” Columbia Law Review, vol. 84, no 3, April 1984; and Geoffrey P. Miller, “Independent Agencies,” Supreme Court Review, 1986.
21 See Eugene Bardach and Robert A. Kagan, Going by the Book: The Problem of Regulatory Unreasonableness (Philadelphia: Temple University Press, 1982); and James V. Delong, Out of Bounds, Out of Control: Regulatory Enforcement at the EPA (Washington, DC: Cato Institute, 2002).
22 I have found only two revolutionary engravings celebrating the quality of impartiality (BNF, Estampes, Collection Hennin, nos. 11069 and 11070). The De Vinck Collection, the usual standard of reference, contains none.
23 The first work to offer a comparative overview of the problem was Colliard and Timsit, eds., Les Autorités administratives indépendantes. In this book, Guy Braibant, a member of the Council of State, referred to institutions “contrary to republican tradition” (p. 291).
24 Jean-Louis Debré, quoted in Le Monde, November 28, 2006.
25 Article 20 of the French Constitution does indeed state that “the government is in charge of the administration”.
26 In 2008, four agencies fall under this designation: the Financial Markets Authority, the Insurance and Mutual Fund Control Authority, the High Authority for Health, and the French Agency for Controlled Substances.
27 This and followings quotations are taken from the chapters on the conditions under which independent authorities are created. See Patrice Gélard, Les autorités administratives indépendantes, évaluation d’un objet juridique non identifié (Paris: Rapport de l’Office parlementaire d’évaluation et de législation), June 15, 2006, vol. 1, pp. 28–32, and vol. 2, pp. 20–28.
28 Nonna Mayer has done similar work in France, as described in “Les dimensions de la confiance,” in Gérard Grunberg, Nonna Mayer, and Paul M. Sniderman, La Démocratie à l’épreuve (Paris: Presses de Sciences-Po, 2002).