Quis Custodiet Ipsos Custodies on the Internet? Self-Regulation as a Threat and a Promise
Jonathan Cave,Christopher T. Marsden
DOI: https://doi.org/10.2139/ssrn.1366723
2008-01-01
SSRN Electronic Journal
Abstract:The ICT domains have always been subject to degrees of technical, economic and/or societal regulation. The traditional basis for these interventions was a 'governance gap' between the economically-motivated activities of key stakeholders and the external consequences for other firms, end-users, public services, etc. Recent changes in European market and societal context, and policy initiatives such as the Lisbon and 'Better Regulation' agendas, have triggered a reconsideration of this basis. Four developments in particular are particularly challenging: enterprise convergence and divergence that reshape market and sector boundaries; the evolution of 'converged' regulators along sectoral or network-industry lines; new regulatory concerns (IPR enforcement, RFID, net neutrality); and changes in the European policy context. These have combined to lay the foundation for cross-cutting reviews and rebalancing of regulatory roles and responsibilities which can have profound structural and dynamic implications. This development has been largely confined to formal or statutory regulation, while much of the governance in these domains is provided by a spectrum of self- and co-regulatory organisations (hereafter referred to as XROs). We consider what sorts of self- and co-regulatory arrangements exist, what issues they address, what other impacts they produce and how their existence affects regulatory assessment. At a minimum, regulatory impact analysis needs to take into account: * The pre-existence, structure and performance of XROs involving key stakeholders and/or addressing the issues addressed by the proposed regulation * The ongoing role and activities of XROs as part of the context for both 'laissez-faire' and statutory regulation; and * The advantages and risks for strategies that seek to achieve regulatory objectives through explicit reliance on or support for XROs (e.g. by delegating authority, endorsing XRO-produced standards and Codes of Conduct, or providing monitoring and enforcement support). The research reported in this paper analyses the roles, functions and impacts of these organisations in various ICT-related domains and considers their implications for developing a regulatory posture that is more supportive of overarching policy objectives, more transparent and accountable, more flexible in response to technological and other changes, less burdensome to those regulated and less likely to distort market outcomes and evolution. The research reported is based on a full-year European Commission-sponsored project to March 2008: * a review of the literature surrounding self-regulation (in a wide range of contexts, including financial services and professional self-regulation), * 21 extended case studies of Internet XROs, * an analytic treatment of the determinants and impacts of XRO formation, agenda-setting, rules, monitoring, enforcement and compliance; and * a policy analysis of the scope for regulatory engagement with XROs and methods option development and ex ante (and to a lesser extent ex post) evaluation. Particular issues concern: the degree to which XROs are formed around specific issues, market segments, personalities or types of action (e.g. standardisation); whether different types of statutory or XRO governance are likely to adopt more stringent or more cost-effective rules; whether different arrangements are more vulnerable to capture or corruption; and whether compliance will be higher under specific types of arrangements. These can be related to a number of topics of current interest. An example can be seen in recent European legal actions to force Internet Service Providers (ISPs) to monitor and enforce intellectual property rights. The preferred mode is a form of co-regulation; ISPs are supposed to do this voluntarily, but face fines if they fail to perform this regulatory function. While it may be argued that ISPs are best-placed to do this, it is clear that the benefits accrue mainly to content owners. The market solution would be a sharing of the gains between the two sides, but this potentially conflicts with 'net neutrality' proposals to prevent ISPs from discriminating on the basis of content. Public policy and the peer-reviewed literature converge on the recognition that there is always a price to be paid for regulation in the form of distortion, cost, institutionalisation, agenda creep and so on. This needs to be offset against the justifying benefit, which may mean extending or shrinking regulation in various areas, rebalancing rule-making and rule-enforcing, delegating or clawing back responsibility, etc. It is necessary to reassess not only how to regulate, but also whether and even why (if some needs become more pressing or cease to be relevant). Generally, this calls for some evolved form of, or alternative to, regulation. The paper presents six specific findings and associated recommendations for policy formulation. There is no 'magic bullet' in Internet regulation; resolving contested policy claims among competitiveness, innovation, public safety and security concerns involves continual political judgment. However, Impact Assessment provides a tool for clarifying potential costs and benefits even where political judgment necessitates a decision that may impair competitiveness and/or innovation. This may help in the design of regulatory policies to mitigate e.g. the anti-competitive effects of an otherwise necessary increase in regulation. We by no means discount the need for more regulation in policy areas where the benefits of market-based XROs may be outweighed by distortions of competition, free rider problems, lack of compliance incentives, extra costs in self-regulation or other drawback. We do conclude that, given suitable options for assessment and a full understanding of the conditions that lead to XRO success, the Impact Assessment methodology may be expected to yield satisfactory results when applied to XROs in the Internet domain. It should therefore be employed more broadly.