I’ve been intending to follow up my previous post by looking at the outcomes of judicial review (JR) applications. I’m very grateful for a query from Richard Doughty from The Chartered Institute of Legal Executives, which has prompted me to complete this analysis.
Judicial review in England and Wales is a two-stage process. The first-stage ‘application for permission (or leave) to apply’ is either granted or refused (a chance to re-apply (‘renewal’) is given to those initially refused). Those applicants granted permission to apply may proceed to the next stage, the final hearing, at which their case may be allowed (found in their favour) or dismissed.
It is tempting, therefore, to use the percentages ‘granted permission’ (at the first stage) and ‘allowed’ (at the second stage) to assess the merit of JR applications. One should be cautious, however, about interpreting those figures simplistically, as they take no account of other possible outcomes that are not reflected in the official statistics. An application is classed as ‘withdrawn’ if it is settled at any stage before the final hearing, whether or not it is settled in favour of the applicant. ‘Granted’ and ‘allowed’ percentages therefore do not necessarily reflect ‘merit’. Indeed if a case is very likely to go against the defendant, it may well be settled at an early stage and therefore classed as ‘withdrawn’. I recommend reading the work of Prof Maurice Sunkin and colleagues (e.g. ‘Dynamics of Judicial Review Litigation’) for a detailed discussion of these issues.
The following analysis, therefore, should be read with these caveats in mind. (Do bear in mind, too, that I am not a legal expert. If I have got some of the terminology wrong, or have misunderstood something, please correct me).
Considering only the JR cases not classed as Immigration and Asylum (IA) and using the Ministry of Justice (MoJ) database linked in my previous post, I’ve looked at the outcomes of non-IA cases in terms of percentages granted permission and those proceeding to, and allowed at, final hearing. For the first three graphs below, I divided the defendants into ‘Central government’ (i.e. named Whitehall departments, including HMRC) and ‘Other defendants’ (other public bodies such as Local Authorities, the wider criminal justice system, NHS, disciplinary bodies, universities and schools). Central government departments were the defendants in about a fifth of the non-IA JR applications each year since 2007. Finally I compare cases received by the MoJ (over half of all cases received by central government) with those received by other Whitehall departments.
Outcomes of First Stage Applications
About a quarter of the ~2500 non-IA JR applications received each year were granted permission to proceed to the next stage (either initially or on renewal). Dividing the defendants into ‘central government departments’ and ‘other defendants’, Figure 1 shows that cases involving central government departments were slightly more likely to be granted (in 5 of the 6 years covered).
Cases proceeding to a final hearing
Figure 2 shows that only about half of the applications granted permission actually made it to a final hearing (termed ‘substantive decision’ in the database). Where central government departments were defendants, the proportion reaching a final hearing was slightly higher. (2012 is lower because not all the cases in that year had been closed by the time the database was released). So what happened to those eligible cases that were never heard? It seems likely that many were settled in favour of the claimant, otherwise why would they withdraw their case?
Cases ‘allowed’ at the final hearing
Of those JR applications that received a final hearing, about 40 per cent were ‘allowed’ (i.e. found in favour of the applicant). Figure 3 shows that the proportion was rather lower for JRs received by central government departments than other defendants.
So, non-IA JR cases involving central government departments as defendants seem to be slightly more likely to reach a final hearing, but were less likely to be ‘allowed’ when they do so, than cases involving other public authorities as defendants.
Ministry of Justice compared with the rest of Whitehall
As shown in my previous post, the Ministry of Justice was the defendant in over half of the central government cases (which averaged about 450 cases a year over this period). So in the final part of this post I compare the outcomes of MoJ cases with those received by other central government departments (the numbers of final hearings are too small to divide up the departments further).
There’s no obvious difference in outcomes at the permission stage, as shown in Figure 4.
And although there is considerable variation from year to year, there also seems little difference between the proportion of those eligible that actually proceed to a final hearing (Figure 5), until the latest two years, when MoJ cases were less likely than those involving other departments to reach a final hearing (this does not reflect a lower completion rate, in fact MoJ cases were slightly more likely to be classed as ‘closed’).
But once they reached a final hearing, there was no clear pattern as to whether MoJ cases were more or less likely to be ‘allowed’. The numbers here are small – an average of 70 central government cases reached a final hearing each year from 2007 to 2011 (fewer in 2012, as the database did not include all outcomes for that year), and the number of cases allowed never exceeded 30 in any one year.
Applications for (non-IA) JR were slightly more likely to be granted at the permission stage for cases that involved central government departments as defendants than for those involving other defendants. In contrast, the cases that made it to a final hearing were somewhat less likely to be allowed (found in favour of the applicant) when the defendant was a central government department. Although the Ministry of Justice was the defendant in over half of the cases received by central government departments, there was no systematic difference between the MoJ and other government departments in terms of outcomes, though considerable year-to-year variation.
An interesting finding is that only about half of the cases ‘granted permission to apply’ for JR were actually settled in a final hearing, despite being eligible to proceed to a hearing. This may well reflect a high level of pre-hearing settlement (in favour of the applicant) if a defendant considers that the case is not worth defending. It is a pity that the database does not give any further details of why eligible cases were withdrawn. The proportion of cases ‘allowed’ in the final hearing should therefore not be considered in isolation, as a substantial number of cases that do not receive a final hearing (and are classed as ‘withdrawn’) may nevertheless have been equally (or more) ‘winnable’ had they received a final court decision.
Finally, if anyone likes to look at the raw numbers on which these graphs are based, here is an Appendix showing all the numbers.
The numbers in this post were calculated from a database of over 57,000 JR applications made between 2007 and 2012. As part of my study with Christopher Hood on long-term changes in UK government administration, I also collated the aggregate JR numbers shown in Judicial Statistics (and more recently in Court Statistics Quarterly). The graph shows the number of final hearings as a proportion of those eligible. These aggregate numbers are consistent with my calculations for 2007 to 2012 shown above. But the graph shows that it was not always the case that fewer than half of eligible cases proceeded to final hearing. During the 1990s, pretty well all of the cases granted leave to proceed did in fact reach a final hearing, but the proportion fell dramatically after 2001. Varda Bondy and Maurice Sunkin attribute changes in the grant rate to the publication of the Bowman Report (‘Review of the Crown Office list‘) in March 2000. The report encouraged early involvment of defendants, which may have led to earlier and more frequent settlements. The change in final hearing rate may have the same cause.