A typology of the constitutional limitations on privatization

The first problem one faces when dealing with privatization from a constitutional law perspective is the absence of a commonly accepted definition of ‘privatization’. There exist, of course, various definitions of the term in statutes or other legal texts. However, these usually refer to rather particularistic aspects of a significantly broader phenomenon. Already in the 1980s it has been rightly stressed that ‘privatization’ is used as a ‘portmanteau’ or ‘umbrella’ term[1]. It denotes various policies which may entail the transfer of assets from the public to the private sector; the substitution of public by private law; the delegation of state tasks to private persons; or, more generally, a responsibility shift from the organized community to the individual.

What is common in all these is that they describe a process[2], and more particularly a transition process from some point of the public realm (the ‘Public’) towards some point of the private realm (the ‘Private’). Hence, privatization is some kind of move form the ‘Public’ towards the ‘Private’.

Now, the ‘Public’ and the ‘Private’ are not absolute instances. Rather they attain their meaning by their relation to each other –that is, by their placement in a given ‘Public/Private’ dichotomy[3]. This dichotomy can have many varying contents –for instance: accessible/closed, transparent/secret, common/individual, society/individual, state/society, state/market etc. One and the same category can be seen as public in one respect, and private in another. For example, in relation to the individual, society is attributed to the ‘Public’; but, in relation to the state, society is attributed to the ‘Private’. Moreover, the distinction between the ‘Public’ and the ‘Private’ is being blurred, but also at the same time confirmed, by their respective placement in some trisection such as ‘Private – Social – Public’; or ‘Private – Social/Public – the State’[4]. The varying meanings that the privatization’s starting point (the ‘Public’) and its direction (the ‘Private’) can have indicate that there is no single notion of the privatization but many differing ones which vary according to the context in which they are being understood.

It is apparent then that a great variety of social phenomena may be described as a transition process from the ‘Public’ to the ‘Private’, and hence labelled as ‘privatization’. Moreover, the term may be used with varying degrees of generality. It may refer to the specific act of a sale or it may describe far more general trends as indicated by expressions such as ‘privatization of politics’[5]; ‘privatization of law’^[6]; ‘privatization of rights’^[7]; or even ‘privatization of the world’^[8]. One can easily assume that not all of these notions are constitutionally relevant, and even if so, not to the same extent.

Thus, in order to assess the constitutional relevance of privatization it is important to draw the boundaries of the phenomenon. And to do so, we first need to construct its defining criterion. For an economist, the defining criterion of privatization is the transfer of assets from the public to the private sector^[9]. For a political scientist, it is the transfer of power from the state to the society, especially the strengthening of the market and of the individual responsibility^[10].

Respectively, the defining criterion of a legal concept of privatization could be the transfer of legal power from the state to private persons[11]. By ‘legal power’ it is meant the power, i.e. the rights or competences, which the legal order recognizes to a person or entity. The ‘state’ is understood here as synonymous to public authority, hence encompassing not only the central government but also local authorities and other public law entities. The distinctive attribute of the state is the exercise of public power. The latter is exercised by public authorities not as a right but as a competence, provided for and organized by public law[12]. On the other hand, the private persons exercise rights instead of competences, and their relation to each other is regulated by private law. Nevertheless, this distinction is not an absolute one. The state may act not only in its capacity as public authority but also in a fashion similar to that of private persons, in which case it is subjected to private law. Therefore, as ‘private persons’ should be understood not only the individuals (natural persons) and the legal entities created by them (associations, companies etc.) but also the private law entities created by the state, such as a state-owned private law company.

It becomes apparent then that, for the concept of privatization proposed here, the distinction between ‘Public’ and ‘Private’ resembles the distinction between public law and private law, the former being understood as the law of the exercise of public power while the latter as the law of private relations.

One way to construct a typology of privatization is by asking the following two questions. Firstly, what kind of legal power is being transferred? More simply: What is being privatized? We can call this the ‘object of privatization’. And secondly, who obtains the transferred legal power? Or: Who obtains the privatized object? Let us call this the ‘subject of privatization’. (We do not need to ask the question of who transfers legal power to a private person, since this would in any case be the state in the broad sense understood here, i.e. some public authority).

The object of privatization can be either a state-owned asset or a state task. What is an ‘asset’ is of course readily apparent, for example real property, an enterprise, shares etc. Less obvious is the meaning of the term ‘state tasks’[13]. For our purposes, suffice is to say that a state task is an activity for which some public authority is responsible and which involves either the provision of a public service, such as social security, or the performance of a public function, such as policing. A more neutral, although not always accurate, term that is sometimes used is ‘public activities’[14]. It is important to note that the problem of the state tasks is essentially a constitutional problem[15].

The respective privatization types can be referred to as ‘proprietary privatization’ and ‘functional privatization’. Proprietary privatization is the transfer of property rights from the state to private persons. This usually involves an act of sale. The most prominent and most common case is of course the sale of public enterprises. Functional privatization, on the other hand, takes place when the state delegates the exercise of a state task –say, the provision of health services– to a private person without any property rights being transferred. This involves some kind of an act of delegation[16]. Given that state tasks are characterized by the responsibility of some public authority, the central issue here is what kind of responsibility (if any) the state retains after the delegation[17].

There is a crucial difference between proprietary and functional privatization. In the former case, the state simply transfers property rights, which are private rights. These rights as such are not in any way altered; the only thing that changes is their subject: a private person instead of the state. In the latter case, the state delegates public competences exercised by public authorities. This is a more subtle process, since it entails not only a change of the subject (a private person instead of the state) but also the transformation of the object, the formerly public competence, so that it can be exercised by the private person as its right. To put it simply, a public activity is transformed into a private one.

The subject of privatization can be either an enterprise of the private sector or a private law entity of the public sector. In the former case, legal power is being transferred from the state to what we can call a ‘genuinely’ private person, i.e. a natural person, an association of such persons, or rather, in the standard case, a private enterprise. Indisputably, these are subjects of constitutional rights and of the private autonomy. In the latter case, legal power is being transferred to entities, such as a private law company of the public sector, which may be seen as emanations of the state. However, the state acts here not in its capacity as public authority –it does not exercise public power– but rather at the same level as private enterprises, that is as a market actor. Thus, a private law company of the public sector (e.g. a state owned commercial bank) enters into common market transactions and acts more or less in the same manner as the respective private enterprises (e.g. a private bank). Nevertheless, it remains questionable to what extend the state-owned companies are subjects of constitutional rights[18].

The respective privatization types can be referred to as ‘substantial privatization’ and ‘formal (or organizational) privatization’[19]. Substantial privatization is the transfer of property rights or the delegation of state tasks to ‘genuinely’ private persons; that is from the public to the private sector. Formal (or organizational) privatization, on the other hand, is the transfer of property rights or the delegation of state tasks to private law entities, such as companies, of the public sector. These are established by the state in order to perform economic activities, which are essentially private. The state acts through them no longer as public authority but rather it uses private law forms of organization and action instead[20]. Whereas the substantial privatization involves a transition from the public to the private sector, in the formal privatization the transition is within the public sector. What is common in both cases is that legal power is being transferred to an entity organized by, and subjected to, private law.

The aforementioned typology, constructed on a double axis, according to the object (proprietary and functional) and to the subject (formal and substantial) of privatization, produces the categories presented in Table 1^[21].

Needless to say, this is a purely logical schematization. The privatization practice has produced a great variety of legal techniques, which do not always fit into those categories. Moreover, various combinations and refinements of the respective types exist. The point to be made here is that there is no single notion of privatization but rather many differentiated ones. Thus, it is useful to discern at least three broad categories:

– The most obvious type of privatization, and the most commonly perceived as such, is of course the sale of public enterprises, or, more broadly, the property transfer from the public to the private sector.

– A different type of privatization is the shift from public to private law in the implementation of public policies by the state. Using private law to re-organize public services and public functions the state substitutes its pubic offices by public enterprises, a process which could also be labelled as ‘de-administration’[22]. This makes the government more flexible but at the same time –which is a crucial issue– less restraint by public law obligations.

– Another type of privatization then is the delegation of state tasks to the private sector. However, whereas public authorities are bound by constitutional constraints and by the principle of legality, the private undertaker stands under the protection of constitutional rights and of the principle of private autonomy. This raises the danger that such a delegation could undermine constitutional accountability[23].

Do constitutions matter for privatization? A possible answer could be simply: no. According to this, constitutions are principally neutral towards privatization, in the sense that they usually neither prohibit nor prescribe a certain privatization. The decision for a privatization then would be just a matter of governmental policy whose implementation depends on the legislature, while the constitution confines itself just to posing the general framework for the transformation of policies into legal acts. After all, constitutional texts rarely make mention of privatization[24].

This perception, however, seriously undermines the political and normative magnitude of the constitution. A constitution is the act by which a state is constituted, i.e. the act of the political organization of a society, which implies the separation between the political sphere and the sphere of civil society[25]. To put it simply, through the constitution are drawn the separating lines between the public and the private[26]. Now, if privatization represents a shift in those separating lines, then there is an inevitable constitutional dimension to it[27]. The argument goes as follows: The constitution permits a certain margin of alteration in the relationship between public and private; privatization involves such an alteration; the question then is whether privatization remains within the constitutionally permitted margin.

Therefore, constitutions do matter for privatization. The central issue is to draw the boundaries of the realm that, according to the constitution, has to remain public. This is of great importance, since a constitution is able to guarantee a realm of freedom and autonomy, a private realm, only when –and indeed because– it entrenches at the same time a public realm, a realm of politics and authority. The constitution guarantees quite as much the private as well as the public realm. And, as the public cannot exist without the private, similarly the private cannot be maintained without the public.

A privatization requires a political decision, which is often formulated as a programme of governmental policy (privatization programme). Law has a double function here[28]. On the one hand, law is the necessary means for the implementation of a political decision. The legislature transforms the political programme proposed by the government into a legal act. On the other hand, law has a relative autonomy towards politics, so that it may impose external constraints to the implementation of governmental policies. Certain political choices may be unacceptable to law, which then sets substantive or procedural limitations in order to prevent (or repeal) their transformation into legal acts. Therefore, law functions at same time as an implementation means and as a limitation to governmental policies[29].

Ordinary law is of course at the disposal of the legislature –and also, to a great extent, of the government through the governmental majority in Parliament. Generally speaking, ordinary law may be seen as a variable that, at least up to a point, adapts to political choices. On the contrary, constitutional law –as well as EC law[30]– is not at the disposal of the legislature. Hence, political choices have to adapt to constitutional law, which poses the constant upon which political decisions have to be reviewed. While ordinary law functions mainly as an implementation means for privatization programmes, constitutional law functions mainly as their limitation. Therefore, the constitutional law of privatization is mainly the law of constitutional limitations to privatization[31].

Certainly, constitutional law may also function in a way that facilitates privatization, for instance by providing justification for the relevant political decisions and inspiration for the formulation of privatization programmes[32]. But this is clearly a secondary role of the constitution in relation to privatization. A short note has also to be made here. While constitutional law is mainly restrictive to privatization, on the contrary, EC law is to a great extend facilitative to it[33]. A possible –although rather schematic– explanation for this is that, while a constitution constitutes a political union, and hence defines a public sphere that is being separated from the private sphere, EC treaties constitute, at least at the present stage of their development, primarily an economic union. The creation of a common market with free competition (still) forms the defining element of EC law. Therefore, EC law has an extended private sphere –and only a limited public one– to guarantee[34].

Generally, we can discern two sets of constitutional limitations on privatization. These are, firstly, substantive and procedural limitations and, secondly, explicit and implicit ones[35].

The substantive limitations initially deal with the question of whether a certain privatization is constitutionally permitted in the first place, or not (we can call this the ‘Whether’ question). To mention but two examples, art. 16 par. 5 of the Greek Constitution prohibits the establishment of universities by private persons; and arts 87d – 87f of the German Grundgesetz prescribe that certain public services, especially utilities infrastructure, be provided by federal administration (bundeseigene Verwaltung). Certainly, more often than not such limitations are not explicitly stated in the constitutions but rather derive from constitutional principles. Thus, the democratic principle may prohibit that certain so-called ‘core’ activities of the state be delegated to, constitutionally unaccountable, private persons[36]; or the social principle may prescribe that certain fundamental social services should be provided by the state[37]. Furthermore, the substantive limitations deal with the –practically, much more important– question of the conditions under which a principally permitted privatization may take place (the ‘If so, under which conditions’ question). This is the problem of the constitutional guarantees that should accompany a privatization, including those imposed after the actual act of privatization has taken place[38]. A major issue here is what kind of responsibility (if any) the state retains towards the privatized activity. It may be seen as a constitutional requirement, for instance, that, when the state delegates a public service to a private enterprise (hence privatizing the responsibility for the provision of the service), it nevertheless retains the responsibility for safeguarding the service to the citizens by way of regulation[39].

The procedural limitations deal with the question of how a political decision for a privatization will be legally implemented (the ‘How’ question). The central issue here is whether a privatization should be implemented through statute, or whether (and if so, to what extend) this can be made by governmental decree or by an administrative act[40]. Generally, the involvement of the Parliament should be regarded as a major guarantee not only of the rule of law principle but also, through the publicity of the parliamentary procedure, of the democratic principle[41]. A somewhat different issue poses the case where a constitutional amendment is required before a privatization may take place[42]. This, in turn, raises the problem of the limits on the power to revise the constitution, notably the problem of a constitutional core, within which possibly also a ‘domain réservé’ of the state[43], which cannot be amended.

Now, as has already been indicated, explicit limitations, i.e. limitations expressively provided for in constitutional provisions, are rather rarely to be found in constitutional texts[44]. Moreover, it is highly unlikely that we find any constitutional clauses, such as those provided for in many constitutions for nationalization[45], relating to privatization. The reason for this may be quite simple. Contrary to nationalization, which represents a direct interference with constitutional rights, the state needs no constitutional justification for privatization. To put it simply, the state may not nationalize unless this is allowed by the constitution, while it may privatize unless this is prohibited by the constitution.

Therefore, the limitations that a constitution poses on privatization are for the most part implicit limitations. This means that they are not readily apparent in the constitutional text but need to by discovered, namely to be derived from some explicit constitutional provision or, in most of the cases, from some constitutional principle. It becomes apparent, then, that the identification of the constitutional limitations on privatization requires a great deal of effort by way of constitutional interpretation.

In the remaining of the paper, I will propose a classification of the constitutional issues related to privatization, which might be helpful to the undertaking of identifying the constitutional limitations on privatization.

Generally speaking, we can discern two broad types of state activity. An activity that the state performs can be regarded either as an economic activity or as an activity that entails the exercise of public power. In the former case, the state participates in the process of production and distribution of goods or services; it stands at the same level as private persons and therefore it is subjected to private law. Contract is the main legal form though which the state performs its economic activity. In the latter case, the state acts in its capacity as public authority and it is therefore subjected to public law. The exercise of public power takes the legal form of a legislative, an administrative, or a judicial act.

Certainly, this is not a straightforward distinction. It is not always obvious whether an activity that the state performs is of the former or of the latter type, and indeed many mixed forms exist. Nevertheless, this distinction provides the basis for the classification of state activity in four main categories:

a) Provision of (ordinary) market goods or services;

b) Provision of public services of an economic nature (or economic services of general interest);

c) Provision of public services of a non-economic nature (or non-economic services of general interest)[46];

d) Activities of policy-making, lawmaking and law enforcement.

Privatization raises different constitutional issues for each of the above categories. These can be summarized as follows:

a) Provision of market goods or services. The state provides ordinary goods or services through public enterprises –mostly, in the form of a private law company. These usually operate within a competitive market (e.g. in the industrial, agricultural, commercial, credit sector etc.), along with similar private enterprises and in the same manner as any market actor. We can regard this as a purely market activity of the state. In most of the cases, there would probably be no particular constitutional reason (although there may exist political or economic reasons) for the state to undertake such an activity. Hence, there would also be no constitutional obstacle if the state decided to dispose of it by selling the enterprise. However, the constitutional issues raised here have to do with the conditions upon which this sale takes place[47]. We can assume that, even in its decision to sell an enterprise, the state remains bound by the constitutional rights of the citizens as well as by a public law obligation to pursuit the public interest. If that is the case, then the state has no power to sell an enterprise to any buyer and at any price it wishes. An arbitrarily choice of the buyer or a pricing below the real value would entail an unfair advantage to the buyer and would thus be detrimental to the public interest[48]. Hence, a possible constitutional limitation on the sale of public enterprises would be the requirement for an open and objective privatization procedure and for a fair pricing.

b) Provision of public services of an economic nature. Public services differ from ordinary commodities in that the state assumes responsibility for (ensuring) their provision to every member of the community. Hence, they are addressed not to consumers but rather to citizens[49]. There exist economic and non-economic public services. In the former case, the provision of the service takes place through an act of market exchange (consumers enter into a contract with the provider); however, the state intervenes in the market process, in order to ensure that all citizens will be able to enter such a contract –that is, to guarantee equal and universal access to the service[50]. The main example here is the utility services, such as telecommunication and postal services, electricity and gas, water and sewerage, public transportation, and broadcasting. Traditionally, most European states have provided utility services through public, indeed nationalized, enterprises. In the standard case, these were excluded from competition; and they were operating under a special legal regime. The state then assumed both the responsibility for the provision (which is an economic activity) as well as for the ensuring of the provision to the citizens (which, as an intervention to market mechanisms, entails the exercise of public power). During the past two decades and a half, extensive privatization has taken place in the utilities sectors. This involved three main trends: the opening up of markets through abolition of monopolies (liberalization), which, in a broad sense, may be understood as delegation of (part of) the activity to the private sector; the transformation of public enterprises into ordinary private law companies, also known as ‘corporatization’[51]; and, eventually, the sale of public enterprises. The main problem here is how the state may retain its guaranteeing responsibility, while delegating the responsibility for the provision –which makes the case for regulation[52]. This is a constitutional problem, insofar as the guaranteeing responsibility of the state is considered as its constitutional obligation. The obligation to regulate the utilities markets would then be the main constitutional limitation on privatization and liberalization of economic public services.

c) Provision of public services of a non-economic nature. Non-economic are those public services whose provision is excluded from the market. This is the case notably with some fundamental social services –which in many constitutions are also protected as social rights– such as public education, public health, and social security. In most cases, the provision of these services does not entail any market exchange and does not establish any contractual relationship; rather they are addressed directly to citizens as such, and they are subjected to public law. The state then acts in its capacity as public authority and indeed these services are mostly provided by typical public offices, such as schools and universities, national health systems, or social security organizations. It is important to not here that similar services are also provided by the private sector as economic services, e.g. private education, private health services or private pension schemes. Privatization here shows two main trends. Firstly, market elements are introduced in the provision of social services by public authorities, which increasingly tend to operate within markets or quasi-markets[53]. Secondly, and perhaps more importantly, by cutting down social services the state increasingly urges citizens to shift to the private sector, in order to meet their social needs[54]. This, so to say, de facto delegation is usually facilitated by public policies, such as a tax policy that “discourages people from relying upon government programs or services…, but encourages them to rely instead upon private resources obtained through the market, or, if necessary, from family or charity”[55]. The constitutional problem here is a rather complex one. To put it simply, it has to do with the question of whether, and if so, to what extend, the constitution –possibly, through a social clause or through the recognition of social rights– guarantees equal and universal access to a system of social services as public services (as opposed to market services). This, in turn, raises the question of whether, and under what conditions, the private sector may be incorporated into this system.

d) Activities of policy-making, lawmaking and law enforcement. The exercise of public power is the distinctive feature of the state. Private persons may not and can not exercise public power; this would be a contradictio in adiecto. Thus, the delegation of public power as such would be impossible, indeed unthinkable[56]. This means, then, that the state may not delegate to private persons activities that are performed as public competences –that is, by means of exercising public power. However, the state might transform such competences into private activities, so that they can be exercised also by private persons. For instance, it has already been mentioned that social services may be provided either by public authorities as non-economic public services, or by the private sector as market services. The problem here is a rather different one. Are there any state tasks that can only be performed by means of exercising public power and, therefore, necessarily by public authorities?[57] Generally speaking, activities of policy-making, lawmaking and law-enforcement require the means of public power. Even here, however, within the bundle of activities of which each state task consists there may be included some that can be exercised as private activities. Thus, many supporting or auxiliary aspects of even the so-called ‘core’ state tasks could be delegated to private persons[58]. The constitutional challenge here is to ensure that such delegations will not undermine the democratic responsibility of public authorities and, therefore, constitutional accountability.

---