Editor’s Note: Earlier this week Policy Exchange’s Judicial Power Project published a report by Dr Jason Varuhas on Judicial Capture of Political Accountability. The Project has invited commentators with a range of views to reply to the report. Some might be expected to be more sympathetic to the report and others much less so. The first reply is by Dr Richard Kirkham from the University of Sheffield.


Recently, the Judicial Power Project posted a list of 50 problematic cases, to which others have responded critically. The list of cases was intended to provoke debate, the more important task though is to understand and reflect upon the theoretical assumptions that underpin the list. In an impressive paper, Jason Varuhas has done this with regard to one of the cases that the Judicial Power Project identified as ‘problematic’: Bradley v Secretary of State for Work and Pensions. There is no room in this post to take on his full critique, instead I point to some issues that require further debate.

The problem with legislation short of detail

How does a judge deal with a statute that endows a public body with a significant amount of discretionary power? The Parliamentary Commissioner Act 1967 is a commendably short statute. It empowers the Ombudsman to accept a complaint, investigate and then report. Where fault on the part of the investigated body is found, the Act is silent about the need for a report to lay out findings and recommendations, the form of recommendations that can be made, the duty of the investigated body to respond, the right of a complainant to challenge the report and the legal grounds that the judge might apply in any legal proceedings. All of the above non-specified practices, however, represent important aspects of the Ombudsman’s operation. On at least two of these practices Varuhas argues that in Bradley the Court of Appeal expanded the impact and meaning of the law beyond the explicit wording of statute. Varuhas must be right.

The crucial issue here is our ability to agree whether and when the judiciary’s subtle expansions of the law are appropriate. In 1993 the High Court ruled in Dyer that it had the constitutional authority to review a decision of the Parliamentary Ombudsman. It did this notwithstanding the silence of the 1967 Act on the matter, the fact that the Ombudsman’s report did not create a legal enforceable remedy, or the position of the Ombudsman as an office of Parliament. For many of the same reasons laid out in Varuhas’s paper, this move to allow the courts to oversee the office was long resisted in the ombudsman sector and viewed with nervousness. It was said that the courts would not understand the very different form of institution that Parliament had created; judicial oversight would add cost, delay and uncertainty to the office’s work; the judiciary would impose inappropriate procedural standards on the institution; and the judiciary would be tempted to intervene in the substance of the ombudsman’s decisions.

In Dyer, the overall architecture of the Ombudsman’s operation was examined alongside the significance of its work to administrative justice and the Court came to the view that the rule of law required it to test the office’s decision-making for various tests of legality. The logic of the Varuhas paper would seem to imply that this was an inappropriate intervention of the judiciary into an arena that should be considered as belonging solely in the political realm. But the High Court ruled:

“Many in government are answerable to Parliament and yet answerable also to the supervisory jurisdiction of this Court. [The Court] see[s] nothing about the Ombudsman’s role or the statutory framework within which he operates so singular as to take him wholly outside the purview of judicial review.”

In Bradley, the Court of Appeal was faced with another novel legal scenario, a challenge to an investigated body’s rejection of the Ombudsman’s findings. Again, the Court examined the overall architecture of the Ombudsman’s operation, including the power of the office to submit a special report to Parliament, as laid out in the legislation. The Court also considered the powerful autonomous status of the office, the purposes for which it was created and the impact of the case on thousands of individual citizens. For multiple reasons, including constitutional principle (the Ombudsman is situated in the political sphere) and pragmatism (the legal case obstructed the political process to no clear effect other than cost), Varuhas argues that Bradley is an example not just of the judiciary expanding the rule of law, but doing so inappropriately. The Court of Appeal though came to the view that where a specialised body is established by Parliament with a specific function to report on maladministration, then it is incumbent on the investigated body to address its output with respect. In Bradley the government’s response largely repeated its original version of events, failed to address directly the Ombudsman’s findings and deliberately misinterpreted the Ombudsman’s position. In order to prevent the position of a statutory office being undermined, the Court expanded the reasonableness test to integrate a duty on the investigated body to provide cogent reasons for its response to the Ombudsman’s report.

Managing the risks of judicial overreach

Varuhas calls for Bradley to be overturned by the Supreme Court and then, on the basis of a brief review of related case law, rounds off by inferring we may have already arrived at a legitimacy crisis in the judiciary’s functioning. I would suggest there are preferable ways forward.

The Bradley extension

As regards the decision in Bradley, a tried and trusted constitutional technique is available: legislative amendment.

One way to interpret what has occurred is that Parliament has introduced a constitutional innovation (the ombudsman) and deliberately chosen to do so through ambiguous legislation. As naturally happens with such legislation, the courts are then left to make sense of the legislative gaps in the light of unforeseen developments and ongoing practice. This is a particularly appropriate function for the court to perform with autonomous public bodies such as the ombudsman which are implicitly charged with both promoting administrative justice and trust in administration. The alternative – that the courts should not concern themselves with that which the legislation does not specify – would leave a whole raft of administrative activity outside the ambit of the rule of law. Such judicial practice is part of the inherent checks and balances of the UK constitution, and if Parliament is not content with the judicial solutions established then legislation can be amended. As luck would have it the current government has promised new legislation involving the scheme that was involved in Bradley.

In 1967 much was unknown about what is needed for an ombudsman scheme to operate effectively. The legal issue resolved in Bradley, and many other previously undetailed aspects of the ombudsman’s work, should now be clarified. To ensure that the Parliamentary status of the ombudsman is observed in the relationship between the ombudsman and the government, when eventually passed, the new legislation should specify (a) that the final decision to implement an ombudsman’s recommendations should be the investigated body’s because it is the body accountable for performance (a point confirmed in Bradley); and (b) that the investigated body should be under a duty to provide reasons for its response to an Ombudsman’s report, reasons which directly acknowledge the Ombudsman’s findings. At the same time, Parliament’s Standing Orders should be amended to require a government rejection of an Ombudsman’s report to be voted on and debated on the floor of the House.

Judicial review of the ombudsman

As for the bigger question of the judiciary’s role, Varuhas’ concerns, and indeed those of the Judicial Power Project, go further than the ombudsman sector. But Varuhas cites the three Balchin cases around the turn of the century and the recent Supreme Court ruling in JR55 in support of his wider claim that the judiciary are ‘increasingly willing to review’ an ombudsman. I will return to JR55 shortly, but want first to suggest that a more complete review of the evidence across the whole ombudsman sector provides a much more optimistic narrative than the conclusion drawn by Varuhas.

Mostly the ombudsman sector operates solely in the administrative realm, sometimes its work attracts political attention and very rarely it becomes a judicial matter. In the main, the public sector has given up reviewing decisions of the ombudsman and few users of the ombudsman would concur with Varuhas’s inference that judicial review offers them realistic hope of redress. Indeed, if we leave the Pensions Ombudsman scheme aside which operates on a very different foundation to the others, relatively few ombudsman findings have ever been quashed. Once you factor in the tremendous institutional shift towards ombudsman schemes over the last 15-20 years across multiple jurisdictions, there is not much evidence of an increase in judicial oversight of the sector. Far from it, most cases are refused permission and even where an ombudsman scheme has lost, one or two rogue cases aside (eg Cavanagh and Others v Health Services Commissioner), the input of the courts has been largely deferential to the discretion of the ombudsman. Beyond the cases listed in this post (three of which involved the same complaint), it is difficult to identify any cases in which the reports of either the Parliamentary or the Health Service Ombudsman have been quashed.

Prior to the Supreme Court’s judgment in JR55 last month, a more balanced assessment of the case law on the ombudsman sector would have been that not only does it provide little evidence of a legitimacy crisis, but arguably it provides model guidance for how a judge should demonstrate institutional restraint to avoid all the concerns that Varuhas raises. But where Varuhas’ paper is certainly correct is to highlight the ever present risks of one or two high profile cases creating uncertainty and fertile ground for fresh litigation, particularly where the out of step judgment is provided by the Supreme Court. To manage these risks I suggest a few strategies here for the ombudsman sector and beyond.

To begin with, the answer cannot lie purely in the interpretation of legislation. Two immediate problems follow from this strategy. First, as evidenced in the JR55 judgment, ombudsman schemes confer wide discretionary power on the office and any attempt at textual interpretation requires an expansive set of reasoning and much room for judicial overreach. Second, where the Court claims the right to apply restrictive interpretations to a discretionary power without clear guidance this creates uncertainty, a potential chilling effect on decision-makers and acts as a catalyst for experimental litigation. Thus, as Varuhas recommends in respect of ministerial power, before a judge chooses to find meaning in legislation which is deliberately ambiguous she should be obliged to state very clearly why and how an intervention is to be justified. Until JR55, this principle of institutional comity between the courts and the ombudsman was engrained in the case law. In ignoring this background principle, the Supreme Court decision in JR55 has unfortunately confirmed some of Varuhas’ fears.

The problem is, however, that for many people ombudsman schemes are the only realistic route to justice, and hence their work does require scrutiny. Unless that scrutiny should be confined purely in the political realm, then the residuary role of the judiciary needs to be explainable. Very occasionally, following a balanced interrogation of legislation complete with a consideration contextual factors, a suitable construction of legislation may be available and necessary. But otherwise, in the absence of explicit statutory reference to the judiciary’s role, the solution must lie in the incremental and deferential application of administrative law doctrine. A recent paper from the Office of the Independent Adjudicator for Higher Education illustrates how this focussed adaptation of legal doctrine can produce a healthy relationship between the courts and the ombudsman.

To conclude, whilst I agree with Varuhas that the judiciary have sometimes made mistakes and could do more to adopt consistent strategies of decision-making, when it comes to accountability institutions such as the ombudsman the judiciary can and does perform a powerful service in retaining the integrity of the model created by Parliament. Further, because Parliament retains the last word, we should not be too concerned about the risks of systemic judicial overreach.

Richard Kirkham
University of Sheffield

Share This