Immigration judges ‘named and shamed’ by Sunday Telegraph [updated]

Updated | Today the Sunday Telegraph (ST) has named and shamed the “three judges who allowed the most appeals” in cases involving the deportation of foreign criminals (Judges who allow foreign criminals to stay in Britain).

Apparently:

The investigation looked at all 184 appeals against deportation by foreign criminals in the 12 months up to June 1 which were brought under Article 8, in whole or in part, in the Upper Tribunal of the Immigration and Asylum Chamber.

David Barrett (who has form on this blog for poor reporting) and Josie Ensor, who wrote the piece, do not reveal the source of the investigation. Why ever not?

Whomever has produced or leaked the study clearly has their sights on the Parliamentary debate surrounding the new immigration rules – see Article 8 and a half – wider than thought, but will it work?

But some words of caution for Parliamentarians considering this article. However the Home Secretary’s changes have been presented, foreign deportations account for only a tiny fraction of the cases in which Article 8 plays a role (the ST identifies 184 in 12 months, although as you might guess, they are not listed). By contrast, the proposed changes to the immigration rules will affect all immigration cases, involving thousands of people each year. What is proposed isn’t a simple tweak, but rather a completely different approach to the courts’ role. The key paragraph is 39:

Where the rules have explicitly taken into account proportionality, the role of the Courts should shift from reviewing the proportionality of individual administrative decisions to reviewing the proportionality of the rules.

In other words, it is proposed that the courts will have an initial role in assessing the fairness of the rules, but save for exceptional cases, will then not have a role in individual cases. This is arguably against the grain of the Human Rights Act, which allows individuals to seek a judge’s opinion in relation to any decision by a public authority which arguably contravenes  their fundamental rights.

Another point is that there were 184 foreign deportation appeals last year, apparently. How many were successful, according to the ST? 96, or “just over half”. But the ST does not mention how many of those 96 were won on Article 8 grounds. For example, how many cases were won because of a UKBA mistake? Or a basic error of law by a court? Since we cannot see the original data, it is impossible to find out.

Without this information, statements such as “In 11 he ruled in favour of the criminal bringing the appeal, with just two going in the Home Office’s favour” are meaningless. To say there is a “startling variation in the “hit rate” under different judges” when the sample size is so small (it is assumed), exposes the lazy logic behind the claims.

The article has an unpleasant tone, which is explained in part by the crimes committed by the men. But full  biographical details are provided for the judges, including the fact that one of the judges is a Methodist lay preacher. What exactly will this achieve?

As the Judicial Office’s fairly anodyne statement to the ST rightly says, “Judges make a decision in each case according to the evidence before them.”

This means it is dangerous, particularly when there are relatively few cases each year involving foreign deportations, let alone won under Article 8 grounds, to use this kind of ‘league table’ or ‘name and shame’ approach. And the ST has loaded the dice rather by not saying where the data came from in the first place, meaning it is impossible to challenge. Shoddy stuff indeed.

Update, 19 June 2012 | Josie Ensor, one of the article’s co-writers has tweeted the following in response to me asking where the data came from:

Sign up to free human rights updates by email, Facebook, Twitter or RSS

Related posts

22 thoughts on “Immigration judges ‘named and shamed’ by Sunday Telegraph [updated]

  1. My anxieties about the underlying study here are that it is not clear how many judges were looked at or what factors were controlled for when comparing judges decisions. In looking to compare I’d expect the researchers to conduct mutlivariate analysis to control for (as far as possible) differences in underlying case type. I note here that some of the judges in the story are referred to as senior immigration judges and others are not. I do not know how such cases are listed but the likelihood is that Senior Immigration Judges get different kinds of case from their less experienced colleagues. This may well explain, in whole or in part, their different outcomes. One judge may be getting a batch of cases with a lots of mistakes on them (perhaps from one bit of the UKBA) and others with less such problems.

    I also note that the numbers of deportation cases considered by judges is small (up to 11 or 12 deportation cases). I suspect many judges only looked at even smaller numbers of deportation cases. If there are particular types of deportation case with very different ‘success’ rates then one would not expect the study to be able to control for these with sample sizes for judges of this kind of size.

    Quantitative studies of judicial decision making have a very important place. And I would fully support the Tribunal Service looking at this kind of data with their judges as part of their process for understanding their decisions. But such research needs to be properly designed (I am not saying this study was not, we simply do not know) and sensitively handled. One cannot of course expect that sort of sensitivity from the Telegraph, but I would hope such sensationalisation does not discourage research. Roger Hood’s study Race in Sentencing is one of the most impressive and well known studies of judicial decision making of an analagous kind: (http://prb.sagepub.com/content/40/2/106.extract). Even then it was strongly criticised by the judiciary who argued that it oversimplified sentencing; but it provided powerful evidence of differences in judicial sentencing that appeared to be related to ethnicity. The Telegraph study may or may not really suggest underlying differences in the way judges approach cases; we simply cannot know on the available information.

  2. One other thing. The fact that judges are named, suggest that this was not a study done by academic researchers. Ordinarily, research subjects would be protected by anonymity. This suggests to me that the purported differences between the judges may not have been subject to the kind of proper scrutiny I’d expect.

  3. What an appalling article.

    Judges, an their decisions, should not be immune from criticism. But this “investigation” simply looks at raw numbers without going into detail on any individual case.

    The only British paper from which one can expect anything approaching high journalistic standards is the Financial Times.

  4. I presume this is original research by Telegraph staff, not because it names judges, but rather because it seems not to consider at all other factors.

    One thing which makes it incredibly difficult to assess is that the cases they mention explicitly do not seem to be reported on either on BAILII or on http://www.ait.gov.uk/

    I tweeted at Josie Ensor, but haven’t heard anything back yet.

    • Chris- unfortunately not all Immigration Tribunal decisions are on Bailli. And the Telegraph in particular seems regularly to be sent Tribunal decisions which are not even available to the general public.

      My guess is that this is not an internal Daily Telegraph report, but rather by a think tank such as Migration Watch, or possibly Policy Exchange, although the latter tend to be a bit more rigorous.

  5. So no one gets confused – all Upper Tribunal (Immigration and Asylum Chamber) judgments supplied to BAILII by the tribunal are published. A rough calculation of judgments received this year shows we are only getting about 20% of these judgments – so if any statistical analysis is done using BAILII as a data set it would be very misleading as would any analysis of individual judges decisions (given that the individual judge may have made five times the number of judgments for a given period than the number found on BAILII). The actual criteria for deciding which judgments are sent to BAILII and appear on the tribunal’s website “Reported judgments” is not given,

    Decisions of the First-tier Tribunal (Immigration and Asylum Chamber) are not published by BAILII and as far as I know are not generally available for publication.

  6. The Telegraph’s article has to be one of the most shoddy articles I have read in recent times. It obviously has no fear of breaching data protection laws in respect of the “criminals” (who Esnor omits to mention would have already served their sentences in respect of their crimes) or indeed of the immigration judges who are doing their jobs to the best of their abilities. Where did the details come from? The Immigration Tribunal does not, save for exceptional cases, promulgate decisions publicly.

    The judges have had the benefit of assessing all the evidence and have had the benefit of oral evidence, something which the poorly educated author of this article had not. One would not dream that a newspaper of the Telegraph’s pedigree would declare that independent juries regularly had reached the “wrong verdict”. Are the writers mates of Theresa May? It serves the political aims of the Home Office’s proposals this week rather too neatly.

    The tone of the article suggests that only Home Office “yes men” should be recruited to the Tribunal (lets do away with the Tribunal altogether then?) The part that did make me laugh was in respect of “.. a Jamaican drug dealer who made two successful claims under the human rights law.” Well, if one claim was successful there would be no need to make another human rights claim?!! The article displays the ignorance and arrogance of the writers contributing nothing but abhorrence of any public debate.

  7. Pingback: The Telegraph’s witch hunt for lenient judges | Free Movement

  8. Thanks for highlighting this, Adam. I’ve put up a blog post on Free Movement about the article. Using Upper Tribunal decisions for this exercise is pointless because of the error of law jurisdiction. Blog post here:

    http://www.freemovement.org.uk/2012/06/18/the-telegraphs-witch-hunt-for-lenient-judges/

    It isn’t about really finding the lenient judges, though, it is about judges as a class and the fact that some of them allow these appeals. The named individuals are just very unfortunate to have their names thrown up in a spurious ‘research’ exercise.

  9. I acted for one of the successful appellants referred to in this article. The information given about his personal circumstances is so limited as to be misleading, as is the summary of the proceedings, and even the reference to the judge (the judge who is ‘named and shamed’ is not the one who wrote the decision). If the feeble quality (or worse) of the reporting is replicated in the other cases referred to, then this article misleads, not only by giving a false sample of Article 8 cases, but also by giving misleading and misleadingly incomplete information about the cases which are referred to.

    • Have you thought about lodging a complaint with the Telegraph? I would certainly consult the Bar Council. Very surprised the Judicial Communications Office has not taken action in at least issuing a statement as to the (ir)relevance of judges’ backgrounds to the decisions they take…

  10. Thanks IC. Yes, we have thought about lodging a complaint, but we are going to apply for anonymisation first, before further irresponsible and inaccurate journalism puts our vulnerable client at further this.

  11. I shall read all the above again for my book as I truely beleive some judges need to be sent to Coventry and called to Parliament…..once it starts by internet there is no where for judges to hide, other judges need support.

    Giving refuge to Talaban…. giving refuge yo xyz rapists…..let these judges pay all out of their own money.

    Their judgemens are as political as my comments to come.

  12. Dear Adam and others,
    To correct some of the most unfortunate misinformation in this blog, and subsequent comments, here are a few points which I hope you will find helpful.
    The source of the research was the Upper Tribunal’s own website, where all of that court’s judgments are published. The analysis was completed by examining ALL judgments over a 12 month period, and isolating those which concerned foreign national criminals appealing against deportation wholly or partly on Article 8 grounds.
    A tally was kept of whether the claimant lost or won at the Upper Tribunal, and which judge heard the application. Where more than one judge was on the bench, the case was included in the tally under the name of the most senior judge only.
    One commentator’s blog questions whether we appreciated the difference between an appellant’s appeal and a SSHD appeal. I can assure him or her that we do.
    This process was rigorous and took many working days to complete, as you may imagine.
    Commentators have questioned whether it was “misleading” to analyse the relatively small number of cases which involved foreign national criminals and Article 8 grounds. But it is this type of case which is specifically addressed by the Home Office’s Statement of Intent document, and which will see new limitations imposed on the use of Article 8. Why do writers here effectively suggest broadening the scope of our research beyond the type of case which is at issue, and which the government is about to restrict?
    I fail to understand why there are conspiracy theories here suggesting that the data must have come from a think tank or lobby group. Sunday newspapers carry out investigations using their own reporters and this was such a case, involving deployment of not inconsiderable resources in terms of our time and effort.
    The Judicial Communications Office was specifically asked a number of questions which could have thrown further light on the wider issues. For example, we asked what auditing processes of judicial decisions may be in place to ensure consistency. The JCO chose to provide only a brief statement, which was included in full, and did not answer our specific points.
    Information about the judges’ backgrounds was taken from their Who’s Who entries.
    Adam writes that I have “form on this blog for poor reporting” (in his view) but neglects to point out that I also have “form” on these pages for accurate reporting;

    http://ukhumanrightsblog.com/2012/01/09/are-some-rights-to-private-life-just-not-cricket/

    Indeed, the contributor of that piece concluded that I was “near enough spot on” in my article.
    So I fear Adam is guilty of exactly the kind of selective reporting of which he accuses me.
    The Sunday Telegraph has highlighted a series of worrying developments in this kind of Article 8 case, examining flaws in UKBA procedure, questioning ministers’ decision making and drawing attention to cases involving serious, violent criminals which have attracted widespread public debate.
    I appreciate this is a blog written by and for human rights lawyers. One cannot expect (without intending to be rude) “turkeys to vote for Christmas”, but when new and important information is put forward about the operation of our immigration tribunals it would be more productive, in my opinion, to embrace that information rather than dismiss it out of hand because it challenges one’s preconceived ideas.
    Sincerely,
    David Barrett
    020 7931 3518
    david.barrett@telegraph.co.uk

    • “A tally was kept of whether the claimant lost or won at the Upper Tribunal, and which judge heard the application. Where more than one judge was on the bench, the case was included in the tally under the name of the most senior judge only.”

      Are you saying you didn’t look at the merits of the decisions, but just did a purely numeric tally?

  13. David,

    Thank you for the reply – much appreciated. You are quite right about the cricket post, which I should have included too.

    Re the source of the investigation, it was fair to speculate that the source may have been a think tank on the basis that (a) the article began “An investigation into decisions”, not “a Telegraph investigation” – what not just make it clear? and (b) your immigration coverage regularly quotes from think tanks, and particularly Migration Watch – see http://www.google.co.uk/search?client=safari&rls=en&q=telegraph+migration+watch&ie=UTF-8&oe=UTF-8&redir_esc=&ei=foftT5_RDoe18QPs4Nn7DA

    The most important criticism of your methodology, which you have not answered despite the lengthy reply are:

    “Another point is that there were 184 foreign deportation appeals last year, apparently. How many were successful, according to the ST? 96, or “just over half”. But the ST does not mention how many of those 96 were won on Article 8 grounds. For example, how many cases were won because of a UKBA mistake? Or a basic error of law by a court?”

    In other words, no analysis at all of the merits. Phrases like a “startling variation”, “In stark comparison” are meaningless in this context. Simply stating that a judge allows lots of appeals (or more than another judge) without analysing the merits of the individual cases is no analysis at all.

    I am sure the research took some time, but this isn’t a case of turkeys voting for Christmas, its about the most basic statistical trap which (I still think) your analysis fell into.

    Court cases aren’t like football matches where you can look at which team won the most matches in a season and from that extrapolate which is the best. The circumstances are essential.

    To use an example, take the X-Factor. Simon Cowell is asked to judge a group of singers who the producers have singled out because they are particularly terrible but will make good TV. By contract, Louis Walsh’s group includes the 10 best singers who have been auditioned so far. Unsurprisingly, Cowell lets through 0 but Walsh lets through 7.

    A bare analysis of the statistics would suggest that Walsh is a soft touch, and Cowell is perhaps overly harsh. But, without considering how those singers got to the judges in the first place, that analysis is meaningless and potentially (in that example, certainly) misleading.

    This is completely different from the criminal context where statistics have been published about which judges are more lenient on sentencing. In those cases, the statistics are based on the number of sentences which have been overturned on appeal. Therefore, there is an objective (i.e. a judicial) analysis of each sentence and a conclusion – accessible to the public – on whether it was indeed more lenient. Arguably, if a judge is consistently being ruled against because his judgments are too lenient, then that might suggest he is more lenient than others.

    But in this context, it really is different. I still believe your methodology was fundamentally flawed and the article was therefore misleading.

    • Dear Adam and others,
      If anyone can come up with a better methodology, and has time to do the work, I’d be delighted to consider the results for publication in The Sunday Telegraph. (I’d rather not have any X-Factor influences in there, mind you.)
      I also look forward to Adam introducing an “accurate reporting” tag to accompany the “poor reporting” one already in use on this blog, for the sake of fair play.
      Sincerely,
      David Barrett

  14. Let’s see the ‘research’ so that it can be assessed. Without that it is impossible to describe this as ‘new and important’ as David Barrett asserts. On the contrary, it looks irrelevant and misleading. The point of the exercise was obviously very much a political one and it appears to ignore countless considerations that would profoundly affect the results.

    The failing here is not the politics — a journalist is of course entitled to make political points in all sorts of different ways — but the attempt to dress up a hunt for a certain pre-identified ‘type’ of judge as a respectable research exercise. Barrett virtually admits that this was the purpose of the exercise in his comment.

Comments are closed.