Business and Trade Committee

Welcome to the fifth panel in today’s hearings on justice for the sub-postmasters. Amanda Pearce, thank you very much indeed for joining us. We are really grateful to you for coming along and giving evidence today. As you know, we are more and more concerned about the Capture cases. We understand that there is at least one case in the High Court, so we will be careful about the lines of questioning that we pursue with you today. At this stage, the first question is: does the CCRC have a good sense of how many Capture-related convictions there may be?

The simple answer to that is no. We have 35 pre-Horizon applications. The reason why I refer to them as pre-Horizon applications is that we know that they do not all involve Capture; some involve other computer systems. In fact, we have already referred one case, on the basis of concerns about the operation of the automated payments system. We received in January last year from DBT a list of, I think, about 480 convictions in the 1990s. That list, as we understand it, was produced by the Post Office in response to a request from DBT to try to identify Capture cases. At that point, there were 31 cases on that list that were identified as potential Capture cases. At the same time, we had 28 live cases, only seven of which were on that list of convictions. So, first of all, it is not a complete list. Some of the convictions on the list do not have names—they are post office branches or post offices, rather than named individuals. Of the seven cases of ours that were on the list, none was marked as using either Capture or any of the other software, but two of them are ones that we have since referred to the appeal courts, one on the basis that it was a Capture case and one, as I said earlier, on the basis of potential problems with the automated payment service as well. That list is not reliable.

This may be a complicated question given the information that you have just shared, but how many applications have you so far had, or think you have had, relating to the Capture convictions? Actually, we probably need to broaden that and talk about software failure or automated transfer of funds perhaps, because we are talking about automated payment systems now—is that a piece of software that is used by the bank, or is used by the post office, potentially widening the scope yet again? Can you provide us a bit of information on that?

As I said, we have 35 applications, which we have termed pre-Horizon cases—

That is individuals who have applied to you, is it? They came to you.

That is individuals who have applied to us, yes. There are 35 cases, and of those, at the moment we are certain that Capture was used in two. In the other cases, we are trying to establish whether Capture or some other system was in operation. We know that a number of IT systems were in operation prior to Horizon, but there is a real lack of reliable information about those cases, because of their age—they are about 30 years old. Some of the applications we have predate the introduction of Capture, so they will not be Capture cases, but they may be cases of other software. As yet, other than the automated payment service, we do not have information to suggest that other software was flawed, but we are mindful that that is a possibility. We are looking at that. In those cases, we are also alive to the possibility that, given the basis on which we referred the Horizon cases on—which was not just that the software was faulty, but that there was a lack of disclosure and of investigation—there are also those wider considerations when we look at such cases. It is not beyond the realms of possibility that a case does not have an IT software issue, but nevertheless has issues of disclosure and investigation. Consideration, particularly of these pre-Horizon cases, is quite broad and—

But the failure may not just be an IT failure.

That is a possibility, certainly. But the challenge is that, as I say, given the age of the cases, there is a real lack of reliable information.

When you start looking at one of these, it sounds like you have literally no idea where it will lead or how long it will take. That is something that we want to try to understand: do you have any idea what sort of length of time the investigations will take?

It is really quite difficult. They are obviously not a homogeneous group. The oldest of our open cases was received in November 2024, so that one is 13 or 14 months old, and the most recent application was October 2025. They are all at different stages, and there are differences in the amount of available information in the cases. In some cases, the Post Office has information and files—they have been able to locate information—but in others, they cannot find any information at all. In some cases, applicants have retained papers, which are able to provide some information. There is very little information from the courts. We have a power under the Criminal Appeal Act 1995 to require any information that is in the possession or control of a public body, so we go not just to the Post Office, but to the courts. We try the police, if it is a case where there might have been police involvement. We try the applicant, in case they have material. A range of inquiries need to be made in these cases to try to establish if any system was in operation and what part it played in any conviction.

Sometimes the information is just not going to be there any more, but are there occasions when there is information, but it is taking rather longer than it ought to get hold of it?

There are a number of threads to follow in these cases. When we are requesting information from the Post Office, it is trying to provide it as quickly as it can. As I say, it cannot always find information and our experience—across not just the Capture and pre-Horizon cases but the Horizon cases—is that sometimes we are told, “We don’t have anything,” or, “We’ve given you everything we’ve got,” and then some months later more information will be found. We have recently had information on one of the pre-Horizon cases that suggests that Horizon was installed earlier than had originally been thought, and that case may be one that benefits from the Horizon legislation. It is complex to try to pin down the information in these cases.

It is quite interesting that Horizon may have been installed earlier than we thought—that probably needs a bit of processing in itself. In terms of the time that this has taken, these convictions were obviously 30-plus years ago in some cases. Are you concerned that some of the people involved may pass away before this gets to a conclusion?

We are absolutely aware that these are very old convictions and that individuals are elderly in some of these cases. Around a third of the applications that we are dealing with are made by families on behalf of people who have already died. There are around a third where the sub-postmaster is over 70, and then a third where they are under 70. We are conscious that these need to be addressed as a priority, and we are working as quickly as we can to get through them and to get the information that we need to make a decision.

Realistically, even on those 35 cases, what is your best guess on how long it will take you to work through those?

With the two cases that we have already referred, one took 18 months and the other took two years. I think that we are moving faster on the cases now. What I think happens—we found this with the Horizon cases—is that the early cases involve more work in establishing the legal argument and drafting the decision, and making the decision to refer. Once we have that framework, it helps with the subsequent cases, so I would hope that the later cases will not take as long, but it is very much about trying to find that available information and making sure that we have followed every thread we need to, so that we can establish what system was in operation and what part it played in any conviction.

From the picture you have painted this afternoon, even with the best will in the world, there is not a hope in hell that we are going to be able to assemble all the information that is needed to make a good judgment.

It is very difficult, yes. There will be cases where we simply do not have the information. One of the things that we have done is speak directly to applicants, and sometimes, with the passage of time, people understandably do not always remember which system they were using. There are situations in which we can talk about the process that somebody used for managing their post office accounts, and we can infer from that what system they might have been using. Of course, we do not need to be 100% certain. The Commission’s test for referring a case is that there is a real possibility of a successful appeal. We need to get to the stage where we think there is some evidence on the basis of which the court might conclude that this was a case that relied on data from a flawed system, and that was essential to the prosecution. Our test for referral is a relatively low bar, but we need to find some information and evidence on which we can make those decisions.

Once you have made a referral, what happens then?

It will depend where the original conviction was. If the person was convicted at a magistrates court, we would refer the case to the Crown court, and that would be dealt with as a rehearing. In general, our experience of the Horizon cases is that the appeals were not opposed and convictions were quashed. If somebody is convicted at the Crown court, we would refer it to the Court of Appeal, and it then proceeds as an ordinary appeal. Counsel would be instructed, grounds of appeal would be lodged, and it would proceed as an appeal in the ordinary way.

The courts are obviously under terrific pressure at the moment. What is the best-case scenario for how quickly a case would move through the courts once you have referred it?

That is quite a difficult question to answer, because it depends on the parties lodging their grounds—

What is your professional judgment?

I think that the aim will be to deal with these cases relatively quickly. I think the Court will be as aware as the CCRC is of the need to move as quickly as possible on these cases. As far as I am aware, the two cases that we have referred to the Court of Appeal so far have not been listed yet, but I would imagine that they will try to move relatively quickly on those cases. I would really be pulling a figure out of the air if I was trying to give you a number of months.

Feel free to pull that figure out of the air, because your definition of fairly quickly might be different to mine.

That is true. Do not hold me to it, but I would hope that it would not take more than six months; it might even be faster than that. However, it is very much dependent on the two parties involved.

Given the age of these cases, given the known systemic problems with Capture, given the time that it will take you to work through the cases, given the time that it will take the courts to work through the cases and given the challenges in assembling the evidence required, is there not a case for Parliament extending the legislation regarding the overturning of convictions to include Capture cases?

I think the commission would be in favour of anything that helps to remedy a miscarriage of justice. Regarding a legislative solution, we heard earlier today that there are challenges with the Horizon legislation and the application of it, and that it is not always the neat and tidy solution that we would hope for.

No, but it has worked for—

Absolutely. A considerable number of people—

One-hundred and eleven people.

Absolutely. Equally, as I mentioned earlier, we now have a case where we think the person was turned down on the basis that Horizon was not installed, but there may now be information to suggest that it was. What I would say in terms of a legislative solution for Capture cases is that those drafting that legislation would need to bear in mind the difficulties that we have experienced in identifying whether Capture was in use at the relevant time. Also, I think the possibility that there are other flawed systems needs to be borne in mind. So, any legislative solution might need to be drafted in very broad terms. Otherwise, we could be back here in 12 or 18 months’ time talking about a different system and whether further legislation is required. So, legislation is not a simple solution; it is a potential solution, but it is not simple.

Caveat your answer to this question how you like. You have a long and distinguished record in this field, and you have seen a lot of things. As a professional public servant, in your heart of hearts do you think there is a case for extending the Horizon legislation? We have got to be careful in extending it, but is there a case for Parliament to look at extending it?

I think that there is certainly a case to look at it, yes. Absolutely.

Amanda, thank you very much indeed. That is all our questions. Thank you very much for coming along and being open with the Committee this afternoon. That concludes this panel.