Judicial review judgments possess multi-layered value. For the parties to a case, they are an authoritative record of the outcome of a legal dispute, providing reasons for that outcome. For lawyers and judges, they support the provision of advice about the law and become legal authorities that can be used to argue and resolve future disputes. As a part of the wider constitutional system, they bring a degree of transparency to the judicial process. For researchers, they form part of the primary materials that can be subjected to different methodologies that seek to advance the understanding and analysis of public law in the courts.  

Who can access which judicial review judgments, and which judgments can be accessed, therefore matters a great deal in various ways. Access to judgments can shape legal advice given to private and public parties. It can also influence and shape how cases are litigated and how judgments are crafted. Similarly, the availability of judgments can shape the cases that researchers draw upon for their work, and thus the broader understanding of public law. 

At present, free, public access to judicial review decisions is provided by The British and Irish Legal Information Institute (BAILII). While BAILII provides free access for the public to a significant number of judicial review judgments through its website, the BAILII database is not comprehensive. If one needs access to judgments not included on the BAILII database, the only way to secure access is by paying a subscription fee to a commercial provider, such as vLex Justis, Lexis Library, WestlawUK, and ICLR Online.  Given the multi-layered value of judicial review judgments, this reality raises an important question: what exactly is the extent of the gap between freely accessible judicial review judgments on BAILII and those where commercial subscription is required? 

To answer this question, we examined a sample of Administrative Court judgments from 1 January 2015 to 31 December 2020. We started with a dataset of 5,408 unique judgments provided to us by vLex Justis, which is regarded as the most comprehensive dataset available. The dataset was built by vLex Justis to serve their legal research platforms. The judgments in their collection were compiled from data supply contracts with the relevant court transcription agencies and judgments circulated by the courts themselves. In order to test whether a given judgment in our dataset was available on BAILII, we took the neutral citation assigned to the judgment (e.g. [2020] EWHC 3076 (Admin)) and used it to construct the link to the corresponding judgment on BAILII (e.g. https://www.bailii.org/ew/cases/EWHC/Admin/2020/3076.html). We wrote code that visited the constructed BAILII link and checked the resulting webpage for the presence of a message BAILII uses to tell its users that the judgment they are looking for was not found. Instances where that message was detected were counted as instances where BAILII did not have the judgment. To verify our results, we then checked the relevant years on BAILII, such as this page for 2015 Administrative Court judgments, and compared the neutral citations listed against those in the dataset provided by vLex.

Our analysis found that, of the 5,408 unique judgments given during the period covered by our dataset, only 55% (3,001) are available to the public on BAILII. The remaining 45% (2,407) are presumed to be available only via subscriber-access platforms. Our findings are presented in full at Table 1 and Figure 1 below. It is clear that there is a significant gap between the number of Administrative Court judgments that are freely available and those which can only be found behind a paywall.

Our research suggests those judgments not freely available generally represent the quantity of judgments that were given ex tempore. That is, where, in the interests of expediency, judges aim to give judgment immediately after argument has come to an end, particularly where the factual and legal matrices of the case are sufficiently straightforward. In practice, several steps occurs when a judgment is given ex tempore. First, the judge literally speaks the judgment aloud, formulating it in her mind as she goes and recording equipment in the court will record the judgment as audio. Sometimes, when there is advance warning, a stenographer employed by the transcription agency is brought into court to transcribe the judgment in real-time. 

Either way, the resulting transcripts then need to be purchased from transcription agency. This is expensive and extremely cumbersome to do on a case-by-case basis. To deal with this, the commercial legal publishers agree annual contracts with the transcription agencies, which are private companies, for the bulk delivery of new judgment transcripts as they become available. BAILII simply does not have the spending power to obtain the transcript of ex tempore judgments in the same fashion. Crucially, there is currently no infrastructure in place to enable HM Courts & Tribunals Service to acquire the resulting transcripts from the appointed transcription agencies even though the contracts with these agencies are paid-for from public funds. 

Accordingly, BAILII’s coverage generally appears to be limited to judgments that were originally handed-down. Handed-down judgments take the form of Word documents that are emailed to the parties (and, most of the time to BAILII and other legal publishers) by the court clerk at the listed hand down-time. BAILII receives the handed-down judgment as an attachment, which they then proceed to convert from a Word document to HTML for publication on their website.

It is important to caution against any suggestion that ex tempore judgments are somehow inherently less important or otherwise matter less. While some judgments may be recognised immediately as important, it is, particularly given the multi-layered value of judgments, not always apparent which judgments may be deemed valuable at some future point. For instance, a string of seemingly simple decisions may be used to build important arguments and judgments decades later or support the provision of advice on a complex matter. Similarly, it is impossible to tell which judgments may be important for future research—cases that may seem routine may make up an important dataset in the future.  

Ultimately, access to judicial review judgments is something that everyone has a stake in, but public lawyers, whether in research or practice, should be paying close attention to this issue, particularly as the government is reshaping how judgments are managed and made available through its initiative to make judgments available via The National Archives. The gap in the availability of judgments we have examined in this post shows that an important part of public law is not meaningfully public, and ongoing changes ought to be seen as an opportunity to progress from that unsatisfactory position rather than entrench it.

Cassandra Somers-Joce is a Researcher at the University of York

Daniel Hoadley is Head of Litigation Data at Mishcon de Reya

Editha Nemsic is a Data Scientist at Mishcon de Reya

Dr Joe Tomlinson is Senior Lecturer in Public Law at the University of York

We are grateful to vLex Justis for making this research possible by making their data available to us. This research was carried out as part of a larger project that seeks to explore the use of machine learning techniques to analyse judicial review judgments.

(Suggested citation: D. Hoadley, J. Tomlinson, E. Nemsic and C. Somers-Joce, ‘How public is public law? Approximately 55%’, U.K. Const. L. Blog (25th February 2022) (available at https://ukconstitutionallaw.org/))