Constitution as Coordination Mechanism—and not a Contract

Constitutions are typically modeled as contracts.Footnote ¹⁰ Instead of a contractual approach, I use Hardin's (Reference Hardin1999) basic view of a constitution as a coordinating mechanism among the “interests that matter” within a polity, rather than a contract. Hardin (Reference Hardin1999, 111) explains that a “constitution does not commit in the way a contract does. Rather, it merely raises the cost of trying to do things some other way through its creation of a coordination convention. Moreover, what it commits to is open to evolution and change in a far more expansive way than is the expectation of action under a contract.” In fact, I agree with Hardin's (Reference Hardin1988) claim that “the very analogy to contract is wrong insofar as contracts require sanctions to back them”—or Gordon's (Reference Gordon1976) complaint that:

The constitution-as-coordination approach raises a serious question about the necessity of any written document. After all, if a constitution is not a contract, and is constantly shifting as a coordination mechanism among dynamic interests in the polity, why bother writing anything down? In response, Hayek (Reference Hayek1960, 180) reminds us that a constitution can be viewed as an appeal of the people sober to the people drunk on political power and raises the transaction cost of change. What is more, constitutions exist, and courts interpret them (the subject of this paper), even if the constitution is ultimately a coordination mechanism rather than a contract.

Types of Guardians

To shed light on how the constitutional system is designed to prevent government failure, I apply here a model of integrated and disassociated guardians (from Weissing and Ostrom Reference Weissing, Ostrom and McGinnis2000(1993)): an “integrated” guardian has a “motivation intrinsic to the operation of the system, and a direct interest in the operation of the system”; conversely, “disassociated” guardians “lack connections to the local setting and thus lack a personal or intrinsic interest in the operation of the system.” Within a constitution, “integrated” are those institutions with a direct vested interest in limiting the power of their fellow players in the political game—including presidents, upper and lower chambers, and states and federal governments. “Disassociated” guardians are those constitutionally created institutions that have only an interest-in-principle in limiting the expansion of the state—this list includes constitutional parchment, philosophical principles, and certain judiciary bodies, such as the US Supreme Court. Note that Wagner (Reference Wagner1993, vii) considers all three branches of government as internal safeguards. For the purposes of this analysis, most constitutional courts are “disassociated” guards, as they typically do not carry much weight in the political game because judges are nominated rather than elected.

The Constitutional Document

Although it is hardly an exhaustive measure of a country's political system or the likelihood of successful constitutional maintenance, the actual text of a constitution is a good indicator of political reality. What ethical and political philosophies does it express or imply? What does it try to establish? How much does it try to do? How much power does it explicitly or implicitly try to give to the State?

As a general rule, American-style constitutions are relatively short. They typically establish only the basic parameters of governance and essentially document, in a Lockean manner, the rights that the people relinquish to the government—as in the preamble to the US Constitution, “We the People of the United States … do ordain and establish this Constitution for the United States of America.” Such constitutions do not try to do more than establishing the basics of the playing field and typically enumerate the powers of government rather than the rights of citizens.Footnote ¹² These documents are typically binding, short, and straightforward. The US Constitution, for example, has a four-line preamble and contains only seven articles.Footnote ¹³

Within the stricter constitutionalist parameters of the American tradition, Commonwealth “constitutional systems” are not “a constitution” (Gardbaum Reference Gardbaum2001). For example, in England, birthplace of parliamentary sovereignty, no single document stands above the will of the people, as expressed by the legislator.Footnote ¹⁴ However, “even the nontextualist English Constitution has often been understood in [constitutionalist] terms: the constitution of post-1689 Glorious Revolution England supposed that some particular set of ‘founding moments,’ including 1215, 1642, and 1689, had set liberty-loving Englishmen on their special course” (Ferejohn, Rakove, and Riley Reference Ferejohn, Rakove and Riley2001, 22).

Kelsen constitutions are typically much longer than American constitutions. They are usually exhaustive documents that incorporate political philosophy, founding principles, modes of governance, fastidious rules of procedure, and meticulous definitions of the citizens' rights granted by the constitution. France's constitution, for example, contains 89 articles. In addition, it includes by reference dozens of other articles from the 1789 Declaration of Human Rights and the Preamble to the 1946 Constitution. Kelsen constitutions are not necessarily the supreme law of the land, as they often draw a distinction between the constitution and “the law,” including prior acts of Parliament or prior statutes, such as the Napoleonic code. In a telling commentary on rights, Kelsen (Reference Kelsen1942) refers to “rights granted by the Constitution” to the citizen. This stands in important contrast to the American constitution, through which the people grant certain powers to the government. Under the American founding philosophy, the notion that government could grant rights to citizens is absurd, because (in the words of the US Declaration of Independence), “all Men … are endowed by their Creator with certain inalienable rights …” (ibid). In sum, Kelsen constitutions do a little bit of everything. This is a reflection of the tempered tradition of parliamentary sovereignty from which they stem: under the Kelsen compromise, government (i.e., Parliament) has many (but not all) powers. Hence a lengthy constitution attempts to define those powers and to put some constraints on Parliament.Footnote ¹⁵

Constitutional Institutions

Systems established by American constitutions typically exhibit strong and separated guardians and keep each other in check through mutual competition. These systems flow from a deep suspicion of power (and, more specifically, unchecked or concentrated power). The branches of government—federal versus state, upper chamber versus lower chamber, or executive versus legislative—hold each other in check as integrated guardians of the constitutional order. Lee and McKenzie (Reference Lee and McKenzie1987, 46–47) explain that “the intent of the Founding Fathers was to create a mixed government in which power was dispersed vertically through different levels of government and horizontally across given levels of government. The thought was to have government power so fragmented that it would be extremely difficult for any narrowly motivated political faction or group to gain sufficient control to work its political will. Only those objectives widely shared and consistent with constitutional limits would be realized through the use of government power.” Their quests for power limit the expansion of the other branches in a healthy competition that prevents any one branch from emerging as a dominant force.

Under Commonwealth systems, government is, to a large extent, whatever Parliament says it is, and the legislative and executive are intertwined. The courts are typically weak, integrated guardians (under the authority of the executive), and Parliament has the final say over the advisory opinions of judges who interpret a relatively loose constitution.Footnote ¹⁶ Integrated guardians do exist within the government, however, in the form of bicameral competition. In a sense, the Crown acts as a disassociated guardian: while it does not jockey for political power with the other branches, it does offer stability and legitimacy to the constitution. And, fundamentally, the electorate acts as a disassociated guardian, which checks Parliament through politics.

Kelsen offers a balance between the two extremes of American judicial review and the Westminster tradition of parliamentary sovereignty. When drafting the Austrian constitution that was to become a model for many other European countries, Kelsen worked within a philosophy of parliamentary sovereignty. He sought to fix some of its perceived abuses, while squarely rejecting the American model of full judicial review and separation of powers. Instead, he established a constitution with some limited sense of competition. Thus we see integrated guardians competing for power but within a much tighter and less autonomous framework. The institutions of constitutional review are less powerful than under the American judicial review system yet significantly more present than in the loosely fettered Commonwealth system.

Constitutional Review: Philosophies, Institutions, and Mechanisms

The actual, formal institution of constitutional review plays a central role in constitutional maintenance as the following three models indicate.

In the American system, the Constitution is the supreme law of the land, and all governmental institutions are beholden to it. All US courts have authority of constitutional review. All citizens can make their way up to the Supreme Court to contest an alleged constitutional violation.Footnote ¹⁷ The US system reflects a suspicion of excessive government, and, in the prescient words of Thomas Jefferson, “the natural progress of things [is] for liberty to yield and government to gain ground.” The US Constitution grants certain powers to government (not to citizens), and the courts apply constitutional safeguards to protect the natural rights of citizens against violation by a government that may fail and encroach on its enumerated and limited powers. In the more mature American-style systems, the judiciary acts as a disassociated guardian, relying on an entrenched constitution, to bind the activity and growth of the other two branches.Footnote ¹⁸ Interestingly, in a less mature American system—such as the United States from 1787 to the groundbreaking Marbury v. Madison decision in 1803 or the first post-Soviet Russian constitutional court—the judiciary in fact is an integrated guardian and must vie for political power and legitimacy with the other branches of government. However, as the judiciary gains legitimacy and an aura of independence, it becomes a disassociated constitutional guardian. In a telling case, Russia's first post-Soviet constitutional court was established as an American-style court; acting as a disassociated guardian before establishing popular and political legitimacy, it was disbanded for its Icarian behavior and recast as a Kelsen institution (see Epstein, Knight, and Shvetsova Reference Epstein, Knight and Shvetsova2001).Footnote ¹⁹

The recent phenomenon of “Commonwealth constitutionalism” presents an interesting development.Footnote ²⁰ Indeed, how does a system of parliamentary sovereignty, in which the legislator reflects the will of the people against undemocratic interests, adopt constitutionalism and its accompanying review of the popular will? Recently attempts have been made to entrench certain basic rights into Commonwealth systems. In sum, the courts have the power to declare whether a legislative act is constitutional, but Parliament maintains its prerogative by deciding whether to accept the judiciary opinion or ignore it. Masterman (Reference Masterman2011, 5) explains that “though seeking to preserve the idea of sovereignty, [commonwealth constitutionalism] clearly envisages distinct roles for each of the three branches of government while also extending the range and depth of the court's powers of judicial review of both executive and legislative activity.” The “government has repeatedly stated its belief that the normal course of action would be that … a declaration [of unconstitutionality] will almost certainly prompt the Government and Parliament to change the law” (Gardbaum Reference Gardbaum2001). Parliament, however, has the final say, under a system that places “the onus squarely and unambiguously on Parliament if it wishes to act inconsistently with a [constitutional requirement], a price that [Parliament] may not be willing or able to pay” (ibid)—and a system where courts typically defer to Parliament in political matters (Masterman Reference Masterman2011, chapter 5). Parliament ultimately will have to answer to public opinion through the game of politics. In sum, “the political expectation, if not the legal duty, is that [the constitution] will ultimately trump” Parliament's actions, when the courts declare them unconstitutional; the constitution is “expected to be politically, though not legally, supreme law” (Gardbaum Reference Gardbaum2001).

Kelsen constitutional review presents a fundamental tension between judicial review and parliamentary sovereignty. In an article explaining its own purpose, France's Constitutional Council explains this tension: “from the revolution of 1789 until 1958, French law never authorized an institution [of constitutional review]: for more than a century and a half, it was considered that to submit a law voted by the Parliament to the control of a judiciary authority, whatever its nature, was excluded. The Law, expression of the general will—according to the formula of Jean-Jacques Rousseau adopted in Article 6 of the Declaration of the Rights of Man and the Citizen—could not see its legitimacy contested: the infallibility of the Legislator was accepted as dogma.”Footnote ²¹ Kelsen, when reflecting on the Austrian Constitution of 1920 and its system of constitutional review, wrote: “it was exclusively the public interest protected by the courts and not the private interest of the parties which was decisive from the point of view of the procedure [of constitutional review]” (Kelsen Reference Kelsen1942). Under the Kelsen system, the constitution establishes specialized councils of constitutional review. The constitutional court is an integrated guardian, as it cannot be approached by ordinary citizens but only from within the government. Kelsen explains the basis for his compromise: “since the Constitution conferred upon the Constitutional Court a legislative function, i.e., a function which, in principle, was reserved to the Parliament, the Austrian Constitution of 1920 provided that the members of the Constitutional Court have to be elected by the Parliament… .” (ibid). Finally, the review council can only consider laws between the time they are passed by the legislature and promulgated by the executive, and only if it is explicitly approached. In this manner, Kelsen grafted a limited version of American-style review onto a tradition of parliamentary sovereignty, by significantly restricting the powers of the constitutional council, and closely intertwining it with the other branches of government.Footnote ²²

Scylla and Charibydis: Parliament's Tyranny or Gouvernment des Juges?

What do we make of these different systems, and their respective attempts at constitutional maintenance? John Marshall's categorical distinction between parliamentary sovereignty and entrenched constitutionalism still holds: “between these alternatives there is no middle ground. The constitution is either a superior, paramount law, unchangeable by ordinary means, or it is on a level with ordinary legislative acts, and, like other acts, is alterable when the legislature shall please to alter it” (Gardbaum Reference Gardbaum2001). Thus I will set aside the Kelsen compromise, and focus on the extreme cases of American judicial review and Commonwealth parliamentary sovereignty.Footnote ²³ The Mephistopheles of the State does not allow Faustian framers to get off easily. Even in their attempts to mitigate government failure, framers find themselves sitting on the horns of a dilemma of constitutional maintenance. At the extremes, we face a tension between the tyranny of an omnipotent legislator and the aristocracy of a judiciary that does not represent the popular will.

Scylla: The Tyranny of an Unfettered Parliament

Parliamentary sovereignty originated in the struggles of the people to assert themselves against the Crown (see North and Weingast Reference North and Weingast1989). Gardbaum (Reference Gardbaum2001) explains the history of legislative supremacy as the expression of the popular will:

However, replacing one unfettered body with another presents its own problems. Under the extreme of parliamentary sovereignty, the rules are too weak, and no guardian protects against the whims of any faction that can secure a legislative majority and lord it over everybody else (the fear that impelled the American founders to create a structure of diffuse government, with separate branches checking each other). Besides the obvious possibility of tyranny of the majority, and the fundamentally problematic notion of any unregulated body (regardless of its constituency), parliamentary sovereignty presents two other basic problems. First, dynamic inconsistency predicts that people will change their minds with changing circumstances: what is “good” when group A holds the reins of government can be different to group A than what is good when group B has that same power. “The problem is sometimes posed as one of continence or self-control, in which an agent may increase his capacity to achieve goals by preventing himself from yielding to temptation” (Ferejohn, Rakove, and Riley Reference Ferejohn, Rakove and Riley2001, 26), much like Ulysses who was literally bound to a mast before encountering the sirens. In a similar manner, an entrenched constitution can bind those who have not yet tasted the corrupting effects of power: “in particular moments when the people have lost their heads, the courts might step in and offer some corrections and appeal to the more sober sense of the people” (Whittington Reference Whittington2002). Second, the notion that Parliament actually represents the will of the people seems somewhat naive and romantic.Footnote ²⁴ Beyond the obvious, the principal-agent “model suggests that, characteristically, the agent will not act in the best interest of the principal” (Furubotn and Richter Reference Furubotn and Richter2000, 148–49). In the particular case of politics, the agent (the ruler) gains coercive powers over the principal (the people), thus dramatically changing the nature of the relationship. “In the hierarchy of democratic authority, ‘the people’ are sovereign and legislators are their representatives, so it only makes sense to take constituents as principals and legislators as agents … The ‘principal's problem’ is the constituents' problem of gaining control over legislators … [Thus,] while citizens are nominally the superiors in this hierarchy, it is the legislators who actually hold public office and have the right to make law. Their role, as agents, is to exercise public authority, backed by the police power of the state, in telling their principals what to do” (Moe Reference Moe1990, 233). Gardbaum (Reference Gardbaum2001) neatly summarizes the problem:

Alas, the story does not end there, quietly and simply. The opposite extreme is dangerous, too.

Charibydis: Unaccountable Judges

The American system explicitly rejected parliamentary sovereignty and the notion that only one branch of government would represent the popular will against other interests. Gardbaum (Reference Gardbaum2001) explains:

However, under a Marshall system, the rules can be too strict, archaic, difficult to change, and entrenched, and they can thwart the democratic expression of the popular will. Under extreme circumstances, “to put too much emphasis on particular powers and rights is to ossify these claims and, often, to unduly privilege people and groups that have done well in the past, against new claimants. Entrenched powers and rights tend to protect existing structures of power and wealth, and limit the capacity of a people to shape their destiny” (Ferejohn, Rakove, and Riley Reference Ferejohn, Rakove and Riley2001, 23). Ely (Reference Ely1980, 4) expresses a similar worry that the “central problem … of judicial review [is that] a body that is not elected or otherwise politically responsible in any significant way is telling the people's elected representatives that they cannot govern as they'd like.” He then (181) asserts that judicial review “can appropriately concern itself only with questions of participation, and not with the substantive merits of the political choice under attack.” In sum, he optimistically concludes (183) that “constitutional law appropriately exists for those situations where representative government cannot be trusted, not those where we know it can.” Similarly, Kramer (Reference Kramer2004, 248) opines that “the Supreme Court is our servant and not our master: a servant whose seriousness and knowledge deserves much deference, but who is ultimately supposed to yield to our judgments about what the Constitution means and not the reverse. The Supreme Court is not the highest authority in the land on constitutional law. We are.” Finally, Hirschl (Reference Hirschl2004) worries that the “global trend toward judicial empowerment through constitutionalism should be understood as part and parcel of a large-scale process whereby policy-making authority is increasingly transferred by hegemonic elites from majoritarian policy-making areas to semiautonomous, professional policy-making bodies primarily in order to insulate their policy preferences from the vicissitudes of democratic politics.” In sum, similarly, Ferejohn (Reference Ferejohn2002) worries that “the American model is easier to reject on democratic grounds than the European one.”

To Marshall's credit, a US Supreme Court Justice stated in a 1990 opinion that “to hold a governmental act to be unconstitutional is not to announce that we forbid it, but that the Constitution forbids it” (Gibson, Caldeira, and Baird (Reference Gibson, Caldeira and Baird1998), emphasis in original). But judges themselves have their limitations. Barber (Reference Barbern.d.) describes a worst-case scenario, “where differences [in constitutional understanding] are fundamental, the call for respecting law is actually a call for respecting the judgments of those who have acquired the authority to speak for the law. And on those occasions where that authority is reasonably in doubt—as it can be, for it's the law that defines it—the rule of law becomes a cover for rule by the people with guns.”Footnote ²⁶ To a large extent, this fear is alleviated by questions of court legitimacy; see, for example, Gibson, Caldeira, and Baird (Reference Gibson, Caldeira and Baird1998), who report that “roughly two out of five Americans who are completely unhappy with the [Supreme] Court's policy outputs nonetheless express support for the institution.” Still, Barber's point at least should give pause to anyone concerned with constitutional maintenance, and the rule of law over the rule of men (non sub homine sed sub Deo et lege—not under man, but under God and law, to use the classical phrase). It could be argued that “the judicial veto of legislation replaces government by the people with the gouvernement des juges regarding many of the most important and controversial issues to be resolved in a political community. It gives final decision-making power on fundamental, usually hotly-contested matters of principle and on the issue of what is and is not the law of the land to the branch of government that is least accountable and which, if it is representative at all, represents the sovereignty of the past over the present. In so doing, it disables representative institutions and displaces popular self-government” (Gardbaum Reference Gardbaum2001).