​Reviewing Regulatory Review

Regulation is a hallmark of modern government. Societies can help achieve collective aims through informal social norms, by making laws, by spending money from taxes or borrowings, or by regulation. Increasingly, regulation has become the instrument of choice. Regulation is chosen often, but the question is how far it is chosen well.

The main technique for providing oversight of regulation and for ensuring the quality of regulatory actions is what is labelled ‘regulatory review’. This blog looks at regulatory review as it is practised in the EU.
 
Regulation: The instrument of choice
 
Regulation has become the instrument of choice for governments for a mix of negative reasons around other instruments of government and for positive reasons: 
 
The negative aspect of the law is that it is formalised, costly, and relies on deterrence as the means to have future effect. It follows a sequential logic that limits its capacity to act pre-emptively. Its case-specific procedures means that the judicial branch faces difficulties in mobilizing the specialised knowledge derived from regular non-adversarial involvement in areas outside the law itself.

The negative aspect of public spending is the limit imposed by taxation or by borrowing.

The negative side of informal social norms is that they are often community, ethnicity, or identity specific. They may be geared to defensive and protective ends rather than to bridge social divides.
 
By contrast, regulation can support informal social norms, can be administered in more flexible ways than the law, places much of the costs of compliance on the targets of regulation rather than on the public purse, and can make use of specialised knowledge that comes from long engagement in sector specific fields.
 
Each of these reasons to resort to regulation applies in spades to the EU itself. In particular, the EU’s ability to spend is limited, notwithstanding the launch of the European Recovery Programme. Therefore, the question of how to ensure that regulation is well conceived is hugely important for the authority and credibility of the EU. Some form of oversight and quality control over the impact analysis carried out at the inception and formulation stage of new regulations is essential. In the EU, that role is performed by the Regulatory Scrutiny Board.
 
Choosing well
 
The core discipline in preparing regulation centres on impact analysis that assesses the costs and benefits of a proposal. In addition, there are several techniques that can be called on for the specific purpose of oversight, correction and lesson learning. They include the retrospective evaluation of how a regulation works after it is implemented, the monitoring of the total costs and benefits of regulation imposed in a year (sometimes referred to as regulatory ‘budgeting’), the use of ‘sunset’ clauses so that a regulation will expire unless renewed, and various forms of ‘one in - one out’ procedure intended to ensure that burdens do not become cumulative. 
 
Each of these techniques for controlling regulation has a place. Each is subject to well-known flaws. In practice, therefore, much of the burden of oversight and correction depends on regulatory review. The review process itself provides a means to monitor the effectiveness of these other techniques.
 
The features of review
 
Regulatory review, as carried out in the EU and elsewhere in the world, was pioneered in the US under the Clinton administration. (EO.12866,1993). It has four main features. 
 
First, regulatory review takes place within the context of the political priorities established by those with policy making authority (the White House in the case of the US and the European Commission in the case of the EU). For example, in Jan. 2021 President Biden called for the modernisation of regulatory review including the need to give much greater weight to distributive concerns. The current EU Commission is pursuing policies to implement a ‘Green Deal’. 
 
Secondly, the body in charge of regulatory review is set at arm’s length from those with operational responsibilities for framing regulation but remains part of the executive. In the case of the US, the responsibility rests with OIRA in OMB that is an executive office of the President. In the case of the EU, the Regulatory Scrutiny Board reports to the President of the Commission and to the Vice President for Inter-institutional Relations and Foresight. The aim is to provide the review body with an authoritative status for carrying out review, without creating antagonistic relationships within an administration. In the Case of the EU’s Regulatory Scrutiny Board, the Board has the authority to block proposals being prepared by Commission services. 
 
Thirdly, the essence of the review relies on ‘replication’ as the key test of the soundness of a proposal. This means that what is proposed must be based on evidence, quantified where possible, that would lead an independent outside reviewer to the same conclusions.
 
Finally, the review process is meant to identify and ensure the appropriate treatment of any externalities. These are of two sorts – the wider social effects of measures (such as distributional, environmental, gender and human rights impacts) and the possible impact on other jurisdictions.
 
Sources of error: Behavioural
 
Even with regulatory review procedures in place, regulation can still go wrong for many different reasons. Quantification of costs and benefits can be difficult; replication is a necessary but incomplete measure of epistemic soundness; externalities can get overlooked. The main source of error however is likely to stem from behavioural bias. 
 
Behavioural bias comes in two different forms. One form of bias flows from the general dispositions of those with regulatory authority. For example, they may be generally over-ambitious in the aims and scope of a regulation. The second class of behavioural bias centres on specific cognitive shortcomings. For example, a regulatory authority may look for evidence that confirms its pre-existing beliefs. 
 
Both sources of behavioural bias can lead to regulatory error. Since the review mechanism is placed within the same machinery of government, the key question about the effectiveness of review is how far reviewers are likely to be subject to the same behavioural biases as those responsible operationally for initiating and preparing the regulations. Reviewers may share the same general dispositions. They may share similar cognitive shortcomings. OECD's 2021 Regulatory Policy Review highlighted the importance that governments and regulators apply the insights of behavioural analysis to their own behaviour.
 
Dispositions
 
There are three general dispositions of policy makers that can contribute to policy failure.
The first disposition is over ambition on the part of the policy maker. The second, is when there is unresponsiveness to changing circumstances or changing behaviour of the targets of regulation. The third source of error arises from any tendency to over-generalise and, for example, to attribute a problem to a single cause with a single solution.
 
In the case of the EU there are safeguards against each of these general dispositions:
 
The key defence against overambition, in the case of the exercise of any powers that lie outside the exclusive competence of the EU, is provided by the need for measures to pass a ‘subsidiarity’ test. Those nearer to where policy is implemented are more likely to be able to appreciate the context and potential obstacles. 
 
The key defence against unresponsiveness to changing circumstances is provided by the retrospective review of past regulation in the same area and the concept of the regulatory ‘cycle’ where retrospective review is built-in as an integral part of the cycle. 
 
The key defence against over-generalisation lies in the provisions intended to ensure ‘proportionality’. The tests are intended to ensure that the need has been identified clearly and that the proposed measure is ‘fit for purpose’.
 
General dispositions that are weakly defended against can themselves lead to poor quality regulation. They also provide a permissive environment for other more specific cognitive errors. 
 
Cognitive error
 
The Commission, including the Regulatory Scrutiny Board, operates in a collegial setting. In this setting, officials have a sense of sharing in the common cause of furthering European integration and where ambition to come up with new proposals for EU measures is a ‘good’ and where ‘more’ is ‘better’. Each new Commission wants to make its mark with ambitious new objectives. This collegial sense of shared purpose is beneficial in many ways. However, it can also be the source of the more specific cognitive errors.
 
First, a belief in the rightness of a chosen path is itself a source of bias (action induced bias) and can lead to bias in the way problems are presented (availability bias).
 
Secondly, collegiality can lead to ‘group-think’ where common notions of validity lead to overconfidence, thinking ‘within the box’, and overreliance on others believed to have superior knowledge or information (herding).
 
Thirdly, shared principled beliefs among those engaged in pursuing a common cause can lead to what is called framing bias. In this form of bias, the initial formulation of problem and the starting point of the analysis is set out in ways that steers the subsequent analysis and can lead to failure to weigh and assess alternative courses of action and alternative approaches to the analysis.
 
Fourthly, those engaged in a common endeavour often share causal beliefs. They can be inattentive to contrary evidence (Confirmatory bias) and pay more attention to their shared causal assumptions than to diagnostic data. (Attribution bias).
 
Guarding against error
 
In principle, those initiating regulatory proposals and the reviewers themselves, are likely to be well aware of these behavioural sources of error. In relation to correcting general dispositions that can lead to regulatory error, the Scrutiny Board has given attention in particular to retrospective review, scrutinising about 13 a year over the last 5 years (2016-20) and establishing the principle of ‘evaluate first’ before new measures are initiated. A key aim is to provide a check on self-interested evaluations: ‘Operational departments may have an interest in the evaluated initiative, and this can impede a frank assessment of its potential flaws’. (2019:20). Retrospective review is one of the main defences against unresponsiveness. However, there are indications that both subsidiarity tests and proportionality tests have become box-ticking exercises. (See posts of 11/1/2018 and 6/1/2021). This means that the Commission remains weakly defended against overambition and over-generalisation. 
 
In relation to the more specific sources of cognitive error, the Scrutiny Board aims for early intervention in the preparation of impact assessments so that ‘all parties develop a joint understanding of what a proportionate analysis would look like’(2019:25/26). In addition, the RSB is reviewing in greater detail the initial design of impact assessments (2020. Annex 2). However, among the 13 aspects of impact assessment given emphasis in the 2020 RSB Annual Report, particular sources of cognitive bias are not specifically identified and discussed. 
 
There is, moreover, another area of concern – the need for extra scrutiny of externalities relating to the impact on other jurisdictions. The RSB is operating in a policy environment where EU institutions want to promote a more assertive role of the EU in the world. This is leading to regulations such as the proposed digital services act and the proposed carbon border measures where extraterritorial effects are a principal aim. They too need to be proportionate and responsive rather than exercises in ambition and muscle flexing. The need for such review is not identified in either the 2019 or 2020 Annual RSB Reports.
 
Conclusion
 
The RSB has established a high reputation in the world of regulatory review. At the same time an OECD report on “Better Regulation practices across the EU’ (March 19, 2019) noted the need for ‘further analytical work’ on EU oversight bodies. At the least, the RSB needs to address the various forms of behavioural bias in an explicit way. In addition, it needs to give greater attention to scrutinising the extraterritorial effects of EU proposals. As a last resort, the institutional placement and location of the Board could be reconsidered.