Governments don’t create wealth and employment – the private sector does. How can this best be achieved?
I am biased – at least on this topic. Government’s core competence should be to develop and implement policy and provide essential services to their citizens. Owning and running businesses is not a core competence of government. State Owned Enterprises (SOEs) are predominantly not as well managed as privately owned businesses and often a drain on the fiscus.
Where to from here?
Governments own two categories of commercial enterprises:
- Category A: Government SOE ownership in competitive sectors. These should be privatized.
- Category B: Organisations that mainly pursue public policy objectives should be government owned and managed.
The process to privatise Category A enterprises however be carefully considered. Government should not privatise an SOE before an appropriate regulatory framework for the entity has been established. This implies that two separate, but related, regulatory frameworks are in place:
- An adequate competition, or anti-trust, regulation backed by effective enforcement mechanisms is a necessity. Privatised entities involved in any activity where competition on general market terms is feasible should be made subject to competitive pressures. This leaves a residual role for sectoral regulation (for instance, third party access regulation) of activities that will necessarily involve an element of monopoly subsequent to privatisation. Safeguards must be taken to ensure the independence of the relevant regulatory agencies, in general, and vis-à-vis any remaining ownership function that the government may retain.
- An additional issue is the degree to which the SOE considered for privatisation – or part of it – can be subjected to competition. In the network industries, this brings up the question of “structural separation”. SOEs in this sector generally consist of many parts, some of which will remain monopolies. Through the process of structural separation, the latter parts are separated from those that are capable of operating in a competitive environment. Governments may, of course, decide to privatise vertically integrated SOEs en bloc, but in practice this complicates the process of introducing competition and may be best suited if the privatised entity is intended to continue as a regulated monopoly. In individual cases, policy makers may perceive a temptation to pursue this line of action because the monopoly elements can boost the privatisation revenues. However, in the longer term these conflicts with the ultimate goal of maximising economic efficiency through privatisation.
A course of action pursued by most countries is to separate “commercial” parts of a utility, typically including the retail side of business, as part of the corporatisation process. What then happens should depend on the national preference. These units may be privatised subject to standard competition policy safeguards, or they may be retained as independent subsidiaries of the SOE whilst their competition with private entities is introduced.
The final choice is whether the hard-core monopoly elements of utility SOEs should themselves be privatised. The retention of such “natural monopolies” under public control is widely supported by economic theory, and in practice the main argument for transferring them to private ownership is normally a need to address major operational inefficiencies that have accumulated due to the public ownership. Privatising such enterprises more than anything else highlights the need for independent and well-resourced regulation. A case in point is the so-called universal service obligations (USOs) that will normally follow this part of the enterprise. A means of treating USOs post privatisation will need to be established. If the SOE is corporatised, and perhaps sequentially privatised, then this will imply that this will involve an identification process, a costing process, a funding process and, ultimately, a contractual process. In the case of en bloc privatisation the state may of course simply impose USOs on buyers and let the competitive sale process sort the costing and funding out.
In conclusion:
- Governments should normally not privatise SOEs before an appropriate regulatory framework has been established. This framework includes anti-trust regulation to ensure competition where feasible, and specialised regulation to oversee activities where an element of monopoly is likely to persist.
- Existing laws and regulation (including anti-trust and takeover rules) should apply to the privatisation itself. If exemptions to this principle are granted, such exemptions would need to be fully disclosed and motivated in advance.
- The regulatory functions whose domain will be affected by privatisation need to be separated from the privatising unit, the state’s ownership unit and the executive. This can be obtained through the creation of an autonomous regulatory body outside the control of the executive powers or through, at least, a complete functional and legal separation within the state.
- Good practice calls for exposing as much as possible of an SOE’s activities to competition no later than at the time of privatisation. If monopoly activities necessarily remain the government faces a choice:
- Break up the company, sell the competitive parts and make specific regulatory arrangements for the rest;
- If the company is to remain vertically integrated during and after privatisation then the need for independent and well-resourced regulation is further underscored.
Am I biased? On reflection, I don’t think so. But an unstructured approach to privatise SOE’s could be damaging.