DON CHIPP FOUNDATION EVENTS
Human Rights Forum
NSW Parliament House
9 December 2004
"RefLaw
Australia: Federal Court Refugee and Migration Decisions Database"
Nitra
Kidson
Thanks everyone. I’ll try to be fairly brief today. I’m well aware that, although we at RAILS think the project that we’re doing is incredibly exciting and will be really useful, it doesn’t make for an exciting presentation! I prepared a paper for everyone to take away, not because I intend to speak from it, because I’m fully aware that many of you will go somewhere else while I present and nod off, and I wanted you to have something to take away with you so you didn’t have to worry about that.
Many, many, many thanks to the Don Chipp Foundation for funding this project. It’s really from that funding we were able to decide to go ahead with it and we’ve also managed to get together an army of volunteer lawyers who are helping out as well.
I thought that perhaps the easiest way for me to explain to you what we’re doing is to show you how it started, and why we decided to do it. I guess a quick synopsis is that we’re developing an online database, which is meant to capture Australian court decisions in the area of refugee law, so it will cover all High Court and Federal Court decisions, and ultimately one day may include Federal Magistrates Court, we just don’t have the capacity for that at the moment.
The idea is to capture it all in a way to enable lawyers working in the field to find out what the law is very quickly, but also to have all that information capture in a way which would reveal trends in decision making which would reveal systemic issues. So that as well as being a legal research tool it would be a really good advocacy tool and a policy or social research tool.
The way in which it started was that at the beginning of the year the RAILS project was, and continues to be, involved with assisting Afghan TPV holders (and any TPV holders, it’s just that the Afghan case was predominant) to prepare their appeals to the Refugee Review Tribunal. And one of the difficulties that people were experiencing is that it was very time consuming trying to find out what was happening at the Refugee Review Tribunal, because when you prepare a case you like to be able to find comparable cases to what you have and to be able to present, to any Tribunal member, a picture of successful cases that are similar to your own.
And it was a very time consuming task trying to find Tribunal cases that were similar to your own. Not because they weren’t and not because they weren’t publicly available, but because it was very difficult amongst the mass to isolate what you wanted.
So for example, you probably know that so many refugee cases revolve around having to look at a very specific region and what’s going on in that region, and the problem with, although a lot of decision are published online, when you try to actually, for example, isolate decisions involving people who come from Jaghori district, is that when you punch in the words “Jaghori” you get every single decision that’s ever been published in relation to Afghanistan, because all of those decisions cite a lot of human rights reports that happen to mention Jaghori. So it is very difficult to try to isolate those decision dealt with a person who came from Jaghori.
Now you could do it, it was just very time consuming. And so what it meant that people and advocates all over the country were all spending a lot of time sifting through this body of information. And it seemed to us that it was probably better, given the lack of resources in the sector, for one person to do it, and then to put that information and that knowledge in some kind of tool or database that was then readily accessible to other people.
So I put together a very basic database and amateurish database, which I’ll show you because that’s really been the genesis of what were doing now. But as we did it we realised that it was a really useful thing, not just for case preparation, but because it became very easy to spot systemic trends. Because the way in which we do it is to set up a series of criteria and to classify cases it means we can prepare profiles based on those criteria.
So for example, a lot of the refugee legal centres, the not-for-profit centres around the country, are part of email lists where we all share information. And people would be getting on there and saying “has anybody had any decisions in this area?”, “do you know what’s happened there?” and “what are the success rates like?” and “how have people been going with claims of this kind” and “has anybody had this member” – and what people were really doing was trying to exchange information to give them profiles, or trends of what was happening.
So I’ll show you what we did…
I just want to say, and assure the Don Chipp Foundation that what you’re about to see is just for internal purposes Excel Spreadsheet. It’s not what you’re funding – what you’re funding is far more glamorous and far more impressive, but this is where it started. And it’s useful to show you what we did.
Does everyone know the story of how Apple computers started, with the two guys in the garage? This is like the workbench in the garage, but the Foundation’s funding the “Apple Mac”, so it will be much more impressive.
What we did was we compiled a spreadsheet. So we went along and we isolated the way in which we record cases. So we had identifiers like date, the RRT reference, the member who decided it, the outcome (whether it was a set-aside, or a successful appeal or not). One of the core issues in relation to TPV holders is the legal approach that members take to the case. I don’t want to get into the details of it all, but there’s basically two ways of dealing with these sorts of cases depending on whether you view it as a brand new assessment and you just pretend that someone never had refugee status before, or whether you recognise that prior status and actually look at whether it’s appropriately brought to an end.
1C refers to Article 1 in the Convention, which is called the cessation provision. So that’s a legal approach that says you actually treat refugee status as continuing unless there’s something which brings it to an end.
1A2 is effectively a fresh assessment. You just pretend that the person has never had refugee status before and start all over again.
So we were tracking the legal approach, because there has been a huge divide within the Tribunal how they approach these cases. So then remember this is particularly for Afghan’s, so we tracked province and district where people are coming from. And then we tracked each of the Convention reasons the claims were being made on, so that’s the Convention reason, race, religion, nationality, social group and political opinion. And what we did at the end, is we just prepared a very short comment, which may just elucidate a little bit. So for example it may name the political parties that the person belonged to, or may cite something about the type of social group that was claimed, or it may just contain a really important or pertinent observation from the member, just so that when people are looking they can get an idea from that comment of whether it’s a case they want to look at in depth.
There are about 120 entries in here at the moment, and the beauty of putting it all in a database is because we can filter. So for example, if you are about to have a hearing before a particular member, and you want to find out what that member’s been like in relation to these cases, you can isolate decisions from that member. You can see immediately what sorts of decisions have been made. You can then isolate further. You can then isolate that down to particular district and see whether that member has made decisions for someone in that district, or for that political opinion, or whatever it is you want to isolate. You can also see immediately what legal approach they’re taking.
And that was where we started to realise it had really great systemic implications. So for example you can immediately success rates just by isolating all the decisions that have been successful, and if you want to, for example, see who’s taking a completely fresh approach, so ignoring the previous grant of status, we can isolate all members who are taking what’s called a 1A2 approach and we can also isolate “set-asides”, so we can immediately see on any given legal approach what the success rate is. We can immediately see, for example, are some provinces being treated differently to others, are success rates varying depending on where people are from or the type of claim they’re making. So as well as being invaluable for individual case preparation, it enables us to see very quickly what’s happening on an issues basis in the Tribunal.
I just happened to mention to other people in other states, and they said “great!” and it’s now being used by about ten legal centres around the country. So for example City Legal Aid NSW uses it, and the Refugee Advice and Casework Service use it here. And it made us think, it’s been so useful in relation to just the Tribunal, wouldn’t it be great if we could do it with court decisions.
It’s obviously a much bigger thing to do it with court decisions, because you’re having to track so many more things, there’re so many more variables involved. But it’s imminently doable, it just takes time, and a lot of resources, which is why we’re so grateful to the Foundation and really it’s only happening because we’ve got a pool of really dedicated volunteers. We’ve got about 20 lawyers who are volunteering their time doing case analysis and things like that.
What we did put in there are direct weblinks to the decisions. So you go through and isolate whatever it is you want to see, by picking whatever combination criteria you want to look at. You then can go to the commentary to quickly see whether all those cases meet that profile of interest, or only some of them, and then from the weblinks you just go back and forth, and you can have a look at the decisions, and come back, have a look, come back – it just works from one screen to another.
So what we’re doing with the court database is much grander, but the same sort of the thing. We want to not only be able to, at a micro level, to assist applicants and lawyers to prepare refugee cases really well by being able to ascertain what the law is on any given point really easily. But also, if we’re going to go to the trouble of reading hundreds and hundreds of cases and recording all this criteria, we want to use it as a broader tool, and to be able to use it to reveal, and to dispel, issues.
For example, in the handout I’ve given you a list of all the sorts of things we’re tracking and plotting, you can see it’s fairly extensive, which is why it really does need a custom built database, it’s not really amenable to this sort of spreadsheet. So we track numerous things about the actual court: which court; who the judges were; how they voted; whether people were legally represented; we’re tracking things like whether a case of substance was argued before the court; then we track the actual what’s called the administrative profile, which is really the legal basis for bringing the appeal; and then we track what’s called the refugee law profile, so we break up the Refugee Convention into all its elements and facets and we track which decisions not only decide issues but make interesting comments about issues, as well as refugee related cases, so things like detention, conditions of detention, things to do with people’s identity, that sort of stuff we’re tracking as well.
And a similar sort of thing but more expansive, as well as recording cases against criteria, we prepare a set of what ire called catch-words, which is really like a summary of everything to case deals with. So the ideal is that you will be able to go into a database, punch in a profile of what you want – you may say I only want to see cases involving the meaning of the word “race” in relation to someone from Somalia, and hopefully what you’ll then get is a series of summaries of cases coming up, which you can tell from those summaries what you want to look at in more detail.
But for example, as a research tool or an advocacy tool, what we’re hoping will be revealed will be broader trends, and I’ll give you a really anecdotal example, so don’t quote me on the actual numbers here – its very much anecdotal, it’s my feeling after having read about 300 full court cases – the Government quotes around a success rate figure for court appeals of around 3%, I think is the current figure they’re fond of quoting, is that only 3% of refugee appeals that go to court are successful, and it’s used to justify continued attempts of block access to the courts as a hideous waste of public money, for what are effectively unmeritorious claims. What’s going to be very interesting to if whether statistically what gets borne out is certainly my very strong sense from having read all the decisions in a short time frame, is that it’s not that simple, and you have to actually look at the way in which those statistics are composed. For example, anecdotally, my feeling after reading about 300 cases, was that about 30% of them were not cases that were actually prosecuted. So someone filled out a form to lodge a Federal Court appeal, didn’t really say anything meaningful in that form, possibly didn’t even turn up on the day. You don’t know whether was because there really was no case there, and the person did that purely to buy a bit of time and extend the process. You don’t know whether it’s because the person just couldn’t get access to legal advice and so had no idea whether they had a good case or not. You don’t know whether they in fact had a case but they were incapable for many reasons to actually prosecute it. So about 30% of the cases that are unsuccessful are cases where in fact the court never really looks at anything. Whether there is anything in those cases or not we’ll never know. It tells us there’s a big problem in terms of court time, but it also tells us that we really can’t use those cases when we view the overall success rates because they significantly distort what’s happening.
In the balance of cases that are left, you end up with far from a success rate 3%, you end up with a success rate of somewhere between 15 and 25%, which is when the courts actually have a case, which gets argued before them and where they really go look closely and make a decision. Those are cases where the nature of migration law at the moment is that you have to demonstrate a pretty serious flaw in the process to be able to have a successful appeal. You don’t get to just go and complain that you didn’t like the Tribunal decision, or that you disagree with it, you have to show a fundamentally serious error of law. So if you consider that that’s the bar which the government has set for a successful appeal, then there’s a very big difference between a 25% success rate and a 3% success rate. And it’s a very different reflection on the quality of the process that’s being used.
So we are really interested in things like if there are any links between things like legal representation and success. There are a lot of ideas out there that judges have various views and vote in various way, we’re interested to see if that pans out, whether you can really say that or not. So in all sorts of ways we’re hoping it will be a really valuable tool of advocacy, for research, as well as for individual case preparation, and we’re hoping that it will be available from around February. We haven’t worked out the nitty-gritty yet. It will certainly be available free to the non-profit sector, and whether we go further than that we’re not quite sure yet, we’ll see how we go. We also have an arrangement with a couple of tertiary institutions for ongoing maintenance of the database using student volunteers. So it’s a fairly massive undertaking, and I really want to acknowledge that it’s only made possible because of the thousands of hours that have been volunteered by really dedicated lawyers and students, and that’ll be the basis upon which it continues.