Public Audit and Post-legislative Scrutiny Committee

Meeting date: Thursday, September 7, 2017

Official Report 485KB pdf

---

“Common Agricultural Policy Futures programme: further update”

Agenda item 3 is an evidence-taking session on the Auditor General’s report on the common agricultural policy futures programme. I welcome to the meeting Liz Ditchburn, director general, economy, and Elinor Mitchell, director, agriculture and rural economy, both of whom are from the Scottish Government. I invite Liz Ditchburn to make an opening statement.

Thank you, convener, and good morning, everyone. Elinor Mitchell and I welcome the opportunity to talk to you again, some nine months since we last gave evidence to the committee. I am sure that you will have questions about what has happened since then and in response to the Auditor General’s update report, and I look forward to answering them. In my opening statement, I will try to capture some summary headlines. As members will know, the committee has been receiving detailed monthly reports on progress since January.

However, I want to start by putting on record again my thanks to two groups of people: first, Scotland’s farmers, crofters and other rural businesses for their continued patience during this challenging period; and secondly, all staff for their dedication, determination and tenacity in seeking to put the CAP payments system on to a more stable footing and to make the system work as we want it to work.

When I gave evidence to you this time last year, we had just come through the intensity of the June payment deadline; we were still within the penalty-free period offered by the European Commission, and were seeking to maximise payments before the 15 October deadline. I had started to make changes in the senior leadership team, and we had made the first changes to governance and processes. When Elinor Mitchell and I returned in December with the permanent secretary, the full depth of the challenge and the cumulative impact of the difficulties experienced in previous years were becoming ever more apparent.

Since then, there have been positive and important developments, as I think the Audit Scotland report acknowledges. Despite that, however, 2016 payments started later than we had originally intended, which squeezed the period in which the bulk of the payments needed to be made. We reached 90 per cent of payments by value at the end of June, which fell short of the target. You will rightly want to hold us to account for our approach and our actions over that period. For example, you might ask why, if we were doing the right things, there was not more improvement in the headline payment performance number.

Let me try to answer that by summarising what has changed over the past year. First, I want to set out some positives. Elements of performance have improved. The application window performed well and closed on time, and the percentage of online applications increased to 78 per cent. The payment performance after June—with regard to what are, in effect, the hardest 10 per cent of cases—has been significantly better. The latest data shows that we have now made more than 99 per cent of our payments by value. Payment letters now go out at the time of payment and we offered loans to farmers earlier, making loan payments from November and giving cash-flow certainty.

Critically, the way in which we operate has improved, too. Our new governance arrangements work well; we have made significant changes in the leadership team; and effective operational working processes, such as regular control room meetings, are now the norm, enabling us to track progress and take corrective action quickly. We have more and better data, management information and reporting to inform our decisions; we have more focused disciplines around software releases and information technology delivery; and quality is improving with reducing numbers of defects. Communication and engagement with staff have improved, with daily meetings and visits by a range of staff to area offices. Indeed, Elinor Mitchell has now visited all the area offices.

Finally, with regard to the technical assurance review, we have now bottomed out the underlying challenges in the system. However, other factors meant that we did not make the progress that we had hoped for. We will be happy to talk about that in more detail, but I will summarise the situation for you. The whole 2016 processing year started later because of the—very welcome, I must say—extension of the penalty-free period and the earlier extension to the application window. However, that pushed the whole timeframe forward. Moreover, while the 2015 payments were being processed, the day-to-day updating of land changes was minimised, and in order to make the 2016 payments, we had to process and validate significantly more land changes than had been expected.

Another factor is the system’s dynamic nature. Because 2015 is the base year for entitlements, the 2015 application has to be at what is technically called ready-to-pay status before the 2016 payment can be made. Even small changes can knock that out, and the added complexity was new to area office staff who had to deal with clearing the anomalies. Finally, we continue to suffer from some delays in IT delivery, with slippage in critical elements of functionality impacting on our ability to make payments. Some of those issues were known to us last year, but their scale and impact was perhaps less understood.

You might ask what we are doing to avoid a repetition of the experience. We continue to make improvements in the way in which we work, and Elinor Mitchell will want to tell you much more about that. We are responding to the recommendations in the technical assurance review; we are responding to the Audit Scotland recommendations; critically, we are looking to get ahead with 2017 requirements through, for example, parallel working on different elements; and we are starting to tackle the longer-term strategic issues as well as the short-term demands. We are not complacent, and none of this is easy.

Thank you for your time and for listening, and I look forward to your questions.

Thank you.

I will kick off the session by asking you about financial penalties. I note from the papers that we have received that you made something like 90 per cent of payments by 1 July but the European Union requirement is for the United Kingdom as a whole to make 95.24 per cent of payments. Given that one would assume the 90 per cent figure to be accurate, what financial penalties have you been notified of? What is their scale? Have you accounted for them in your thinking?

The process is still on-going, but as part of it, the European Commission formally confirms final numbers to the UK, and the UK has to decide how any penalties that are applied are distributed across the UK and the different nations. Our understanding of the current situation is that performance across the UK as a whole might well have been sufficiently good to mean that, as far as the main schemes are concerned, there are no penalties. Obviously that has not yet been confirmed and we should always be cautious until we have final numbers. There might be a small level of penalty for some of the smaller scheme elements, but Elinor Mitchell might wish to say more about that.

Yes, I can add some details. As Liz Ditchburn has said, we believe that at the UK level the 95.24 per cent target was reached for the basic payment scheme and greening scheme, so we are fairly confident that we will not face penalties in that respect. Indeed, the BPS and greening payments make up 98 per cent—the vast majority—of the pillar 1 payments.

We anticipate facing some modest penalties for the voluntary coupled support schemes, because they are Scotland-only schemes and we did not meet the 95.24 per cent requirement. The same is true for the young farmers scheme across the UK, so we anticipate modest penalties for that. However, as Liz Ditchburn has said, the EC has not yet notified us of any penalties.

You say that you are anticipating modest penalties, but what kind of scale are we talking about?

We anticipate penalties of between half a million pounds and £700,000 across both schemes.

Let me just tease out the issue of our performance relative to that of the rest of the UK. You seem to be telling me that this is one of the benefits of being part of the United Kingdom: if other areas outperform Scotland, we all avoid penalties. Is that correct?

It is correct that we achieved payment performance of about 90 per cent while the other paying agencies achieved percentages in the higher 90s and that that evened things out across the country.

Do you have information on that that you could send to the committee?

As we have explained, the numbers are not confirmed until we get to the very end of the payment period, which can take some years. For example, we have not yet finished the 2015 payment period across the UK because of things such as transfer of entitlement related to probate cases. It takes a long time for the numbers to be finalised. We tend not to publish final figures until everything is cleared, because the numbers go up and down, the denominator changes, as does amount of value and so on. I will check with the co-ordinating body about what information we are allowed to share at this point in the year, but we will, of course, share everything we can.

Thank you. We recognise that these are estimates and that the process takes years to complete, but the information would give us an order of magnitude with regard to what areas in the rest of the UK are performing better than we are. We can learn from that.

I have one more question before I move on to other members. It has been suggested that staff are involved in recovering loans, which is diverting their attention from processing the new payments. Is that accurate?

There are separate teams involved. The issue is about identifying the resources that we need to undertake the various tasks in hand, which is part of the new workforce planning strategy that we have in place. We identify the work that needs to be done at different times and we make sure that we have people involved in the different jobs during the weeks and months when we need them.

Okay.

As with so many public sector information technology projects, a big part of the problem originally seems to have been on the management side. Reference has been made to the substantial changes in personnel within the senior management team. Can you give us a bit more information about that? How have you ensured continuity between the new management team and the previous one? What is the mix of skills in the current management team, and how is that going to work in future?

As Elinor Mitchell has been more involved in the detail of the changes all the way down through the structure, I will ask her to start.

Continuity has been a really important factor for us. I have been in post for a year; a number of weeks after I started, I appointed a new chief operating officer. We split what was previously called the rural payments and inspections division into two to increase the management bandwidth, and a second deputy director took up post at the beginning of the year. As a result, there are now two senior managers across RPID.

At the level below the senior civil service—or senior C-band management, to use the civil service grades—there is an enormous depth and understanding of the systems and processing, and we rely on those staff extremely heavily. For example, a couple of weeks ago, we had a C-band conference that brought all the senior staff together in one place to talk about the year that had passed and the year that was coming. The senior people in the area might be very new, but we have worked very hard to ensure that we draw on the knowledge and understanding of the staff within the organisation who have been around for a long time.

Is there no core senior team that has the expertise within it and which has oversight over everything?

Yes, there is. There are senior management teams at all sorts of levels, but the core senior management team who look after what I would call RPID is made up of me; Annabel Turpie, the chief operating officer; Andrew Watson, director of the Scottish paying agency; and Eddie Turnbull, who heads up the information systems division. We are all new—we have all taken up post within the past year—but each of us works with a wide range of people who have been in the division for a long time, and we draw on their knowledge and understanding.

Over the course of the year, we have drawn much more heavily on the knowledge, wisdom and understanding of the day-to-day running of operations provided by the principal agricultural officers and their teams who run the area offices around the country. They are involved in the performance data collection and reporting conversations, the control room meetings and the governance arrangements, to make sure that we truly understand the local issues and impact of our decisions.

I am concerned about whether that is really a project team. Do those senior members of the project team have the skills and experience to implement an IT project?

09:15  

I stress that this is much more than an IT project. There is an IT project within it, but success for farmers depends on an awful lot more than the IT. The IT is one element of it, and we are using clear, standard protocols for managing the IT contract and our relationship with the contractor as well as the business-as-usual capability that we have within our own IT support systems. However, the policies that set how the schemes operate, the knowledge of the European Commission, the farmer-facing people in the area offices and the way in which we manage operational processes all matter as much as the IT delivery to the overall impact and outcomes for farmers.

Since we were last in front of the committee, we have transitioned from a programme structure to more of a business-as-usual structure; as a result, what we are talking about now about is an operational unit that is underpinned by IT and which very much relates to the policy. All of that has to sit together across the piece, and the responsibility for that is carried by Elinor Mitchell and the senior management team.

Surely all the problems relate back to the IT services.

I think that the Auditor General has said on record that many if not all of the problems date back to some of the decisions that were taken in 2012, and we share that view. Some of the design choices and the things that happened in those early days—for example, the complexity of the system that we designed—have had a significant impact on how we have been able to take the project forward. The IT is just one element of that, although it is a very important element.

Given that a huge chunk of the project is IT—it accounts for more than £170 million—I want to establish the capacity of the senior management team to implement it. I agree that there are other elements, but the failure has been primarily on the IT side. Does the new leadership team have the experience and capacity to oversee that?

As Liz Ditchburn has said, the CAP futures IT programme came to an end on 31 March and was brought into business as usual. The system is now owned and run by the Government, and we—along with our delivery partners, the largest of which is CGI—are responsible for ensuring that it continues to run and develop as we need it to.

The appointment of Eddie Turnbull has been critical to that, as he has experience over many years of developing a number of IT solutions both in the Scottish Government and elsewhere. Since his appointment, he has developed a transition plan for knowledge transfer from the contractors to ISD staff, and he has also introduced several industry-standard methodologies for carrying out system testing and ensuring that the project is running effectively. He has brought about many changes to the way in which we run the project now that it is part of our day-to-day working.

In putting together the management team that we now have in place, we have not only done the things through the appointment of Eddie Turnbull that Elinor Mitchell has described, but sought to increase the capability around audit and assurance. That is really important, because so much of this is about the financial impacts of penalties. We have increased that knowledge and skill within the senior management team as well as increasing through the appointment of the chief operating officer the operational capacity to run our business processes effectively. It is the combination of all those things that determines our success, so we felt that it was important to build a leadership team that combines all those skills. That is the team that we need going forward.

I recognise what you are saying about the different layers of management. How many people are in the senior management team? Is the decision making in that team now working well? Is it efficient and are decisions based on good information? I am feeling hesitant about all this, because the structure that you are describing is not a typical project team such as we might expect for an IT project of this magnitude. I keep coming back to the IT side as that is the major part of the project, although I realise that there are other parts to be delivered as well. I just do not see any familiarity as far as the structure is concerned.

I think that there are two things that we can pick up on. The first brings us back to governance, because I think that that is part of what you are asking about. The governance arrangements with regard to the project structures might look very familiar to you. We can also send you more information about the organisational structure that Elinor Mitchell has put in place.

I am struggling slightly with the concept of the senior management team, because the senior management team operates at all sorts of different levels for different things. We have a clear governance model that we can share with you, but it might help if I were to set it out now.

At the top level, we have the CAP executive steering committee that Liz Ditchburn chairs and which includes a range of people from the Government, non-executive directors and external expertise from, for example, the national health service. Below that, we have the CAP strategy and assurance board, which I chair. That is the high-level governance in the programme.

Sitting beneath that, we now have the paying agency strategy board, which is chaired by the director of the Scottish paying agency, Andrew Watson; and the delivery board, which is chaired by Annabel Turpie, chief operating officer of RPID. Underneath those boards, we have the accreditation committee on audit as well as the business design authority, which keeps a tight grip on the change processes that were among the problems that we had last year. Below all that, there is a range of working level project boards.

In portfolio management terms, there is a clear structure. Sitting alongside that are the senior management team arrangements for the whole directorate, in which the various bits of RPID are involved. I am confident that we have a robust governance system that is taking the right decisions at the right time and that we are making a difference.

Convener, it might be helpful to understand a little more about the structure, because I am a little concerned about it. If we were to see it set out on paper, we might understand it a bit better.

I see that we are getting the nod that that will happen.

Yes, we can share that.

We would be happy to do that.

I want to stay in the same area but take a more retrospective approach. There have been significant changes to the leadership team, and the Auditor General’s report and some of the answers that you have given suggest that the tanker might be starting to turn. However, is that not a tacit admission that the old leadership team failed, because it was that team that got us into this mess? In any event, who has been held accountable for the situation?

We have been over that ground in the committee previously and our retrospective view has not changed. As the Auditor General said, the roots of the challenges go right back to some of the choices that were made early doors.

By whom?

As we have previously said in the committee, some of those choices were made by ministers, some were made by officials and some were made by the European Commission. There was a whole set of issues: we had to design a system before the European Commission had finalised its regulations, and we made some decisions to achieve clear policy aims that meant that we arrived at a system for the schemes in Scotland that had a higher level of complexity than those in the other parts of the UK.

As is well known, those decisions were taken by ministers at the time, in consultation and discussion with stakeholders including NFU Scotland. The difficulty of trying to design a system while the European regulations were not yet formalised is also well known. All those factors contributed to a very challenging situation for the management team that was in place at the time.

As we have also described to the committee, the project began from difficult roots and had a series of cumulative problems. Given that we are dealing with live payments, we do not have the luxury of being able to say, “Let’s pause everything for a year and try to get the system back up and running”; we have to try to provide the payments to farmers. We are always trying to manage a live system and get payments to farmers as quickly as we can—as we need to—at the same time as improving and stabilising the system, and that was the challenge that the management team faced.

That tells me why, but it does not tell me who. Who has been held accountable?

As we have described, there was no single point of failure.

So nobody has been held accountable.

I think that you also asked that question of the Auditor General, who has really considered the points. The roots of the problem go all the way back.

I accept that.

We are all being held accountable all the time.

I presume that you accept that the contractors are at least in some way liable. Nevertheless, the Auditor General’s report says that the CGI contract to which Ms Mitchell referred has been extended for two years at a cost of £29 million. A contract with another supplier has also been extended at a cost of about £3.5 million. Do you have any comment about why we are extending the contracts of companies that have apparently failed to deliver what they were asked to deliver? Does that represent an appropriate spend of public money?

As the audit report tried to describe, one of the challenges of managing such a contract is that, as the committee has heard in evidence, the project has gone through different phases. There were phases in which it was led by the Scottish Government, with the contractor operating only as a supplier of personnel and the Scottish Government being accountable for what the contractor did and how it did it, and there were periods in which the contractor was under much more normal contractual circumstances and was held to account for that. We have had different payment processes in place during different phases. Elinor Mitchell might want to say a bit more about that, as I might not have all the detail.

There is no single thread. The business model that we used has changed, and that has changed the accountability of the contractor. We can talk about how we have increased the commercial accountability in the contract and about the penalty systems that we have had in place. We had those systems in place at the beginning but changed the business model so that the contractor supplied only personnel. We then went back to a system in which we were able to apply penalties.

That normal discipline is in place but one of the challenges has always been that, unless the Government can specify very clearly what we are requiring the contractor to deliver, we need to work with time and materials contracts rather than pay according to output. We have had to maintain time and materials systems for longer than we wanted because of the nature of the challenge that I outlined to you—the fact that the regulations were not clear before we had to design the scheme. Therefore, we have always been playing catch-up to get a clearly defined requirement to hand on to the contractor, for which the contractor can be held fully to account. We are trying to move to that situation and have made significant progress, but the position continues to be made more complex by the interdependencies of what the Scottish Government needs to do in the business processes and what the contractor needs to do in IT delivery. That comes back to the original point that it is not just about IT delivery but is much more complex than that.

We commissioned an independent technical assessment review from Fujitsu and worked with the company over a number of weeks to examine whether the system was working. The baseline question was whether it was a system that we could ever develop and build on or whether we would have to write it off and start again. That would be the catastrophic solution. The Fujitsu report said that the basic, fundamental architecture was sound, although it acknowledged that there were a number of defects in the system that needed to be fixed.

It was fairly early in my post when we went through that process, and one of the first questions that I was faced was whether we should continue working with the contractor that we had worked in partnership with for a number of years and that had a real depth of knowledge and understanding. We all recognised that things needed to change, and CGI recognised that as well. It has invested heavily in new management structures, just as the Government has, in order to move things on to a different footing. We have now moved on to a different footing and I can give some statistics about the quality and timeliness of the releases that we get.

There is a new way of working with CGI that enables both sides to understand the business and IT requirements. It involves an agreement that we are looking for a high-quality product that we can implement and that will work first time.

Does that not suggest that something went wrong at the contractor level before? If it does, what steps have been taken to recover the public money that was paid to the contractor?

09:30  

I am not sure that we could categorise it as something going wrong. As Liz Ditchburn said, the history of the CAP project and the catalogue of decisions that were taken mean that it would be extremely difficult to identify at what point a wrong or a bad decision was taken, because decisions are taken on the basis of the best information that is available at the time.

So we launched a project—the cost of which is now £178 million—without knowing whether it was going to work.

The system works. It worked last year—it got 90 per cent of payments out by June and 99.7 per cent of payments out by 15 October. Although the system does not work as well as we would want it to, it does work.

It is rather difficult to suggest that the model is working.

Last year, there was the loan scheme. I am pleased about that because, ultimately, we are talking about people’s livelihoods. The Government has a finite budget, so the loan money must have come from somewhere. If that is correct, what has not been procured or funded as a result of the Government’s having to cross-fund the loan scheme?

As I think Audit Scotland’s report makes clear, the financial transactions budget from which the loan scheme came funds a number of programmes across the Government. That budget is utilised by a range of programmes, some of which are demand driven, and, across the year, it is always difficult to predict what they will need. For example, some of the programmes that involve housing depend on conditions in the housing market and how much demand there is for such schemes. At the beginning of the year, our finance colleagues project what they imagine the demand might be and how the forecasts will play out over the year. It is not a case of having a clear set of projects that are on particular paths; the budget line is much more fluid than that in the sense that the demand can change over the year.

When we explained to our finance colleagues the demands that a CAP loan system would place on that budget, they looked at their most recent forecasts, their forward projections and the end-year flexibility that they had. The money that we are talking about is, of course, money that comes back to the Government—it is not expenditure. The issue involves the profiling of the money that goes out and the receipts that come back in. The needs of the loan scheme fitted with the updated forecasts and the projections that we had for the second half of the year, so we did not require any other programmes to be curtailed as a result of the CAP loans. The normal in-year fluctuations in the demands on that budget meant that there was sufficient room for the CAP loan scheme to happen.

When we carried out the loan recovery for the 2016 basic payment support schemes this year, we worked closely with our finance colleagues to meet the 31 March deadline. They gave us a number—off the top of my head, I cannot remember what it was but I think that it was around £115 million—and told us that that was what we needed to recover. We worked with them on a daily basis to make sure that the payments went through so that they could hit the right number on loan recovery.

The money comes back into the system—it is a loan—and I share Liz Ditchburn’s understanding that there was nothing that was not done because we required a CAP loan scheme.

Before I ask some questions on the IT side, I want to return to the issue of disallowance. It is not possible to find out much information about the system down south. However, from the information that I have read, I believe that the Department for Environment, Food and Rural Affairs was preparing for substantial disallowances and fines of up to £180 million because its system had substantially overrun its budget. Liz Ditchburn mentioned that the system in Scotland is much more complex than the one that has been adopted down south, and I am pleased to hear that substantial progress has been made on both systems to avoid the substantial fines that have been mentioned. Is that the case? Has there been a late run of improved performance in both systems in an effort to avoid fines from the European Union?

I understand that most of the challenges that the English paying agency has faced date back a while. I am not sure which reports you have seen, but it was well known in the media that the English paying agency faced significant challenges with its system and that there were significant cost overruns.

That was some time ago, and one of the biggest shifts that we have had to make in implementing the CAP payments programme has been the move to area-based payments. England moved to an area-based system slightly earlier and so went through that pain in an earlier CAP period, and its current performance has been better and more stable. It faced more challenges in 2015, but we hear that 2016 has been a good year for the English paying agency. It is by far the largest component and the most significant agency, in numeric terms, in the UK performance, which shifts the overall balance. If the English agency does well, the overall UK performance is likely to be high; if it does badly, the overall UK performance will be brought down.

The figure for the Scottish component, which you—or perhaps it was Elinor Mitchell—mentioned earlier, was potentially £500,000. Figures that the committee heard previously were as high as £60 million. Why is there such a huge gulf in the estimates of that additional risk?

Those are two different numbers: late payment penalties, which are incurred when we are late in making payments, as the name suggests, and disallowance, which is the £60 million estimate from the Auditor General. A disallowance estimate is based on an assessment that is made when auditors review systems and processes such as the inspection regime and the controls that are in place. They assess whether an organisation might be disallowed on certain factors.

Is the other number—the £60 million—still potentially in the system?

The Auditor General and Audit Scotland have made a disallowance assessment of up to £60 million and have put that figure in their report. We do not make any assessment of disallowance, as we believe that we follow the regulations to the letter of the law. We do not consider that we will face any disallowance. Of course, every year we do face disallowance, but there is a long, drawn-out process of negotiation with the European Commission about that, so we do not know its level yet.

For full disclosure, I will go back to the late payment penalties figure to make sure that there is no confusion about the years. The year that Elinor Mitchell spoke about with regard to the most recent payment performance is the current year, in which we hope and believe that the UK has done well enough. We still have outstanding late payment penalties arising from 2015. The committee will remember that we previously gave an estimate of £5 million for potential penalties. That remains an estimate, as we still do not have final information of that penalty. I want to be sure that members are aware of that information as well as of the figure for 2016.

Before Willie Coffey moves on to IT, I want to pursue that issue further. I understand that the 2015 figure of £5 million is an assessment of financial penalties for late payments.

Yes.

Have you made an assessment of the disallowance financial penalties for weaknesses in control?

No, we have not.

Why not? One clear committee recommendation was that you should conduct a full assessment of financial risk.

We have conducted a full assessment of financial risk, and we take risk into account much more fully in decision making. However, with regard to disallowance, when audits are carried out by the European Union, the auditors come over and then write a letter to us that gives their assessment. There is then a negotiation. The auditors may have interpreted the rules in a particular way and we may have interpreted them differently—that is a point of discussion and negotiation between us. It would be inappropriate for us, as a paying agency, to assess that we would be liable for any disallowance, because we do not believe that we have implemented the rules incorrectly until we are told otherwise.

Is the Auditor General’s report wrong?

No. The report uses a different methodology. It looks at the previous audits and makes an assessment of the issues that the auditors have raised. It may say that something might attract a particular level of penalty, and that is a perfectly reasonable thing for it to do. However, it would be inappropriate for us, as a paying agency, to say up front that we believe that we will be disallowed because of the methodologies that we have used.

I will pursue this point and will then let Liz Ditchburn in.

You are saying that £60 million is a reasonable assessment but that, because you are still in negotiation, you will not confirm or deny whether there will be a disallowance.

The Auditor General has made that assessment of £60 million. I will not comment on whether it is reasonable. Undoubtedly, the Scottish paying agency will suffer disallowance at some level, although I think that £60 million is at the high end of what it will be. We have never had such a disallowance before, but I guess that time will tell.

This is the first time that you have operated a system for CAP futures payments.

Yes, that is true.

So there is no history.

There is no history.

Okay.

As Elinor Mitchell says, a lot of it is about whether we understand the rules and apply them appropriately. You are right in saying that there is no history on this particular CAP system, but there is a history of many years of Scotland applying CAP rules, whatever the CAP system was at the time, and suffering lower levels of disallowance than other parts of the UK. We have a track record of understanding whatever the European scheme is at the time sufficiently well to apply the rules effectively in ways that we are able to defend to the European Commission. Sometimes, the Commission might have one interpretation and we might have another, but that allows us to put forward evidence and so on.

We have a track record of being able to demonstrate that we can apply the rules appropriately. The rules might be different this time round, but the capability to understand those rules and apply them effectively throughout our systems is still pertinent.

Okay. Given that you have all that ability to apply the rules effectively, what was the failure in 2016 down to?

Do you mean the failure in terms of late payment penalties?

Yes. Did you learn from the control weaknesses?

There are many different things going on. There is a very clear system around late payment penalties—it is a simple test of whether we did or did not make a payment on date X. That sits within a very clear band. As the committee might remember, there are then questions about whether we have a 5 per cent allowance that we can use after date X or a 2 per cent allowance that we can use after another date. However, it is black and white, because the payment is either made or it is not.

The categories of disallowance are broad and varied given the nature of the questions that the European Commission can ask. Disallowance includes a lot of other things that are far more qualitative and are not so black and white as whether we made a payment on date X. The questions may include whether the scheme rules have been applied appropriately, and there may be points of interpretation that we will debate with the Commission. The Commission might also ask about control weaknesses and whether the inspections were carried out effectively enough. It is a very different set of categories and issues. It is not as simple as saying that we were late with payments, therefore we will have disallowance. Two quite different things are going on.

Okay.

Can I clarify that point? You asked whether we had learned from the past. For the certifying audit in 2016, which looked at the CAP payment year, the overall conclusion of the report was that there were no findings of major weakness in the Scottish Government’s administration of CAP funds for the 2015 scheme. However, we are not complacent about that at all, and a number of intermediate findings are being addressed. Within the governance structure, we have a number of working groups that cover each of the main areas in which issues were found, and one group is focused on making sure that our performance is better this year than it was last year. The 2017 certifying audit is now under way, and I hope that it will recognise that we have made some significant improvements in the areas in which both internal audit and Audit Scotland found that there were improvements to be made. We are absolutely not complacent about that.

Paragraph 60 of the Auditor General’s update report mentions some of the weaknesses in the controls. The report was published in June, so it is very current. Do you accept all the failings that are set out in paragraph 60, given what you have just said? Maybe I should read some of them out for the benefit of the Official Report. Paragraph 60 states:

“The incremental nature of developing and implementing the rural payments system has affected the quality of the audit trail during 2016. The paying agency’s focus on delivering core compliance functionality has meant that the audit trail is not always easy to see or access ... Delays in delivering the new system impacted significantly on the paying agency’s planned programme to make BPS payments to farmers, crofters and rural businesses.”

I could go on. Serious control weaknesses have been identified by the Auditor General in paragraph 60 of her most recent report.

09:45  

I recognise that, of course, and we are not complacent. We are working hard on all those issues. I should say that the European Commission confirmed that no financial corrections were required as a result of the certification audit.

The bottom line is whether the Scottish paying agency can continue to make payments to Scottish farmers. Yes, we can. The certification audit identified no major findings or weaknesses. Some serious issues were raised and we are working on them. We are working on a daily basis to improve our position.

I will let Willie Coffey back in, but I want to pursue that point. The Government has said that

“the assessment of risk from financial penalties is now an embedded feature of our governance arrangements”.

What does that mean, and when will we be able to see that publicly?

For example, we have introduced a new commissioning process for change requests or new pieces of functionality that are required, and that new process goes through a number of governance boards and eventually comes to the CAP strategy and delivery board, where a key piece of information that is sought is whether the change will improve our audit and assurance processes or whether a risk is attached to it. That is a key part of our decision-making process.

Will that be publicly available?

I would happily share the commissioning process forms.

We are interested in the risk assessment in particular. You may have made an internal assessment of the financial penalties for disallowance, but you are not sharing that with us because it is part of the negotiation. What disallowance did you budget for in 2015 and what are you budgeting for in the future? Surely the risk assessment leads into that.

I come back to your question on whether we disagreed with the Auditor General. It is important to say that we absolutely agree with the Auditor General and the Audit Scotland report about the importance of risk assessment and the need to maintain systems and mitigate the risk of financial penalties, and we have embedded that in our Government processes, as the Audit Scotland report recognises. We are doing a set of things that enable us to form the risk assessment; if there is a difference of approach, it is that we are not seeking to quantify the assessment as a single number or a range of numbers, particularly in respect of the disallowance report—rather than the late penalties—for the reasons that Elinor Mitchell outlined.

We will not be able to provide you with a single number, but you are right to challenge us on whether we are clear that the risk assessments are informing the decisions that we are making. We believe that they are. You are asking us whether we are aware of the potential implications, and we are.

You also ask whether we are budgeting for the implications appropriately. The budget does not work in that way, in the sense that it is a long, drawn-out process, and we are still nowhere near the conclusion of the final liabilities for the 2015 scheme. There is further audit work to be done, and then subsequent negotiation. I understand from my finance colleagues that the point at which the risk has crystallised into an absolute number and a request from Europe, when we have exhausted the opportunities to reduce it through further evidence, is when we would need to bring it into our budgeting, but it would come into a future year, not the current year.

Elinor Mitchell may want to say more about the financial aspects.

On absolute clarity about the number, the only thing that I will add is that I understand that the accounts that will be published this year include an allowance of £1 million for the CAP 2015 year, because of the actual late payments that we made.

I am confused now. You had a risk assessment that said £5 million, but you are saying that there will be something—

As Liz Ditchburn said, the question is to do with the status of that figure and when a risk materialises.

So there is £1 million in the current accounts.

The accounts are currently draft.

In the draft accounts.

In the draft accounts. So there is £1 million in the draft accounts but more to come.

Possibly.

In future accounts.

Yes.

And we do not know the number.

Yes.

Wow.

I call Willie Coffey.

I have almost forgotten what I was going to ask, as it was so long ago that I spoke.

Sorry.

I will ask the witnesses about looking ahead from this point. Elinor Mitchell mentioned that we were at 99.7 per cent. Liz Ditchburn mentioned critical IT functionality. I will explore that a wee bit more and try to understand what the critical factors are that will manage the process in years to come. Does that involve IT? Are there IT components that need to be in place, correct and working to ensure that we do not face such issues in the next few years?

There is the IT and the business process—the annual business cycle. We have been working hard this year to match the agricultural cycle with the single application form, the application processing cycle and the requirements of the broader business cycle—inspections and the like. We need a combination of all three: making sure that we have the right staff in place, with the right skills to do the jobs that we need them to do; making sure that the IT system is working to its optimum level; and finalising getting the additional functionality that we need in the system at the time when we need it.

Where are you in saying to the committee, the public at large and the farming community that the offer is completed, that it is ready for next year and that it will be on time? Are you still finalising key components of the IT system?

Yes.

When will that be complete?

We have a detailed IT delivery schedule, which we are more than happy to share with the committee. It sets out when the various bits are likely to be delivered in respect of the CAP pillar 1 and CAP pillar 2 payments.

As with any major IT system, on-going maintenance is needed every year. With the SAF, there is what is called the rollover—for example, we need to change the dates. Rules change—for example, greening rules changed in Europe this year. Every year, we need to implement changes, amendments and upgrades.

On the base functionality, we have talked before about the cost of the CAP futures programme, which is £178 million. About £11 million was brought forward from last year into this year. The completion of the bits that cost us £178 million—the cost of the system—will be in this financial year.

I am pleased to see in your progress update that new testing processes are in place, that quality management is going on and that fewer defects are emerging from the software. Are you becoming more confident that the system is settling down and stabilising to enable the process to be carried out correctly, properly and on time for the next round of applications?

I am cautiously optimistic. We have the independent technical assurance, which has identified that there is a significant amount of technical and production debts. We have a joint team in place, with members from the Scottish Government’s information systems division and from CGI, which is triaging those debts. If they are affecting payment performance, we are dealing with them.

In the past six months, we have reduced the technical debt by something like 40 per cent. We are not adding to it any more because, when we put new releases into the system, we work to make them defect free. I am cautiously optimistic, but there is a lot of work to do. I am really proud of what the team has done and continues to do to make inroads into improvements in the system, but this is a journey.

It is important to highlight the technical assurance review. This time last year, when we sat in front of the committee, some members asked us whether the system would ever be fit for purpose, and we said that that was a really important question that we, too, really needed to know the answer to. We cannot go forward just in hope and faith; we absolutely have to assess the system, warts and all—its strengths and its weaknesses.

That is what the technical assurance review was about. It was good to hear that the system’s design and infrastructure are fundamentally sound. There are many areas for improvement, and there are problems and legacy from previous changes that we need to sort, but the system’s design provides a basis from which we can continue to improve and develop the system.

As a new member of the committee, I am picking up things as we go along. Before I ask my question, I will make a couple of comments on the penalties that I have heard about. You talked about the disallowance penalty of £60 million. I understand that your negotiating position would be that you have done everything correctly but, in the real world, we have to assess where we stand and what may hit us. I would like to think that you do that, although I am not asking you to put the information out there.

When you talked about the late payment penalty of £500,000 to £1 million, you used the word “modest”. That would be quite a lot of money to many people, so that may be the wrong terminology to use in speaking about a payment from which we get no value.

You have submitted a paper on the update.

The latest update?

Yes. It is the one that, in the last sentence at point 1, under the heading “Update on progress”, states:

“There are appropriate disaster recovery solutions”.

I am sorry—do you mean the letter that we sent to the committee on 30 August?

Yes. It states:

“There are appropriate disaster recovery solutions in place for our two separate platforms ... The Disaster Recovery position is constantly developing and forms a feature part of infrastructure planning and application support.”

I want to clarify what that means. Have the disaster recovery solutions been fully tested? What does “constantly developing” mean? That does not sound like something that is stable and working. What is a “feature part”?

I will say a couple of things by way of introduction. Elinor Mitchell may wish to give more detail.

The new system, which is the futures system for RP and S—RP stands for rural payments; I always forget what the S means—has a full disaster recovery system in place that meets the required standards, and we expect restoration of service through that system within four hours. You asked whether it has been fully tested. It has been partially tested, but it has not yet had a full, go-live, take-everything-out-and-see-whether-it-works scheduled emergency exercise test, because of challenges of timeframes and fitting everything else in. However, it has been tested to a significant extent and was designed to the standards that the industry would expect of a full disaster recovery system. That is the main system—the new system—that is working to make the payments.

As the committee will know, we are still using a number of legacy systems, because some of the challenges in the early years meant that we had to delay the implementation of parts of the new system. One of them is the land parcel information system—LPIS—and another is the Scottish integration admission and control system—SIACS. Those legacy systems do not have the same full disaster recovery system as the new system has, but they have full data and system back-up, and back-ups are held off site so that we can recover and restore all the data and systems if we need to. The time that those legacy systems would require to be recovered is much longer, but we have tested to ensure that those back-ups restore from tape.

The team is working through the levels of risk that we are prepared to take, and those legacy systems will be replaced. The question that we have to ask ourselves is what it would cost to put in place a full disaster recovery system for the legacy systems when some of those systems will be working for only a couple of months. We have to assess the risk of not doing it and the cost of doing it. The team is also assessing what mitigations can be put in place to improve the disaster recovery of the legacy systems.

I do not know whether Elinor Mitchell wants to say more about that. I have described the overall way in which we are approaching the issue.

I do not want to add more about disaster recovery.

Bill Bowman remarked on the late payment penalty. I used the word “modest” to reflect the fact that the payment was modest in comparison with the hundreds of millions of pounds that we pay out as an agency. That is not to detract from the fact that we should be making payments on time. I accept that we should be making payments on time, and I regret the delays that farmers are experiencing in relation to the moneys that we have not got out of the door yet.

Thank you for that comment. To go back to disaster recovery, what part is constantly developing? Is it the legacy part?

10:00  

Yes—it is the risk assessment of the legacy. We need to risk assess not just in the abstract but against particular payment processes. The risk may depend on whether the system is exposed to particularly high volumes, for example. The teams are undertaking dynamic risk assessment and assessing their options. That is an appropriate way of dealing with the risk that arises from the legacy systems.

We need to be clear that we are taking risk-based decisions and that we are explicitly accepting a level of risk in deciding whether we can tolerate that and whether the cost of mitigating it is too much. It is also critical to consider the length of time that we are exposed to the risk. If the legacy systems were going to be running for 10 years, we would without doubt choose to do things differently. However, as we know that the systems will be replaced, we have to assess the trade-offs between carrying a level of risk now and when the risk exposure stops.

You said that the futures system has been tested to an extent.

We would still like to be able to carry out a full scheduled emergency exercise on the futures system. We have not yet done that, but we need to do it at some point. We are confident that the new system—

Can you give us a point in time?

I do not have a date.

We do not have a schedule for that, because, as Liz Ditchburn said—

Are you talking about doing it in a month, three months or a year?

We have carried out desk exercises and we have done what we can reasonably be expected to do without taking the whole system down and doing full disaster recovery. It may sound melodramatic, but we have not yet done that because the system is operating all the time—we are always working on the system and getting payments out.

As Liz Ditchburn said in her introductory statement, the 2016 payment year has been fraught. This is perhaps a good point to pay tribute to the tremendous dedication and work of all the staff throughout the country who work on the process every day. We started late, we had a number of changes to make in the system and we have been literally making payments on a daily basis the entire year. It would be inappropriate to take the system down to do a full disaster recovery.

Will the 2017 payment year be any better than 2016? That might not be the case in terms of staff workload. We are introducing the new LPIS to improve our information holding on land—that is clearly very important for a land-based system. That was identified in our audits as one of the key controls that were not right and we are taking action on that this year. However, that means that the system is constantly under pressure. I know what that means for disaster recovery, so we constantly assess the risk of that. My assessment is that we should continue working with the current system and carry out a disaster recovery test when it is not under so much pressure.

I do not want to pursue the matter without having the technical knowledge. Perhaps you could come back to us on the technical guidance about back-up testing when a system is not available to be tested. What is the best practice in such a situation?

We would like to come back to you with more details. It is a technical subject in which I, too, am not qualified.

We can confirm this and give you more information, but I understand that the design of the system and the nature of the test that we have carried out meet the standard. I do not have the standard in my head and I thought that I could find it on one of my pieces of paper—it is one of those numbers. We should write to you with more information.

It probably would not help me even if you did quote it.

Do I have time to ask another question, convener?

Absolutely.

The fourth bullet point under point 2 in your submission says that you will

“Develop a benefit realisation plan to record and monitor all potential benefits and value that the system can provide”.

Is that something that you have begun or something that you have accepted that you will do? I am interested in what “all potential benefits” might be.

A business case was made at the beginning of the CAP futures programme. Normal practice is that, towards the end of a programme, we look at that, review it, update it and decide our final position before going into the gateway review process. That work started recently and is under way.

Okay. Do you know any of the potential benefits and value?

The work has just started.

Will we hear about that at some point?

Of course.

I have a couple of short questions. First, I refer to Elinor Mitchell’s updated management report, which is dated 4 September 2017. I suggest that it would be useful to add a column to the table to show percentages. If we take the BPS, for example, the value of payments made to date should also be shown as a percentage of the total, and the same for VCS.

Your submission says:

“We remain on track to issue the payment schedule to all our customers by the end of September 2017.”

Am I right in my calculation that only 3 per cent of payments under the BPS are outstanding and need to be made between now and the end of September to meet that target?

Yes. At the end of August we were on 99.3 per cent for the BPS, which includes VCS. We have completed the beef scheme, such as we can—there are always tail payments—and the VCS sheep scheme is the one that we cannot complete until we have completed the BPS payments.

The table subdivides the BPS from VCS. For the BPS, the number of eligible applications is 17,990 and the number of BRNs—business reference numbers—whose payments have been processed to date is 17,509. That indicates that there are 481 outstanding payments. That suggests that roughly 97 per cent have been made, so you have 3 per cent to catch up on.

That is on VCS, yes.

Okay.

We have 321 outstanding BPS payments.

Are you fairly confident that those will be made by the end of September?

We will do all that we can. As I said, every year there are payments that are delayed because of probate and all sorts of issues, which can be exceedingly complicated, but I am confident that we will get to the very tail end by the end of September.

Okay. That is the percentage of the number of applications. Is the percentage of the value roughly the same?

It is roughly the same.

Okay. So you are fairly confident that you will get as close as you possibly can, and reasonably close to 100 per cent.

Yes.

Right. Good. Let us now look specifically at the VCS payments that are highlighted at the bottom of the report, which include payments under the suckler beef support scheme for the mainland, the suckler beef support scheme for the islands and the Scottish upland sheep support scheme. Is each of those at a similar kind of percentage, and are you as optimistic about getting the payments completed by the end of September?

As I said, the beef scheme is now complete. We achieved more than 99 per cent overall by the end of August. We have made good progress over the summer and we remain on track to get down to the very last few tail payments across all pillar 1 schemes by the end of September.

Right. Okay. My second question is about the fifth bullet point under point 2 of your paper. One of the things that you say you will prioritise is to

“Communicate clearly the payment timescales and processes to farmers, crofters and rural businesses.”

I know what the result of a survey about that would have been up until recently, but if we carried out a survey of farmers, crofters and rural businesses today, do you think that the majority of them would think that that has been achieved?

Do you mean the visibility of when we will make payments?

I mean clear communication about the payment timescales. In addition to the core issue of the payments themselves, communication was one of the issues raised by the NFUS and individuals.

There are a few things that I would like to say on communication.

We still intend to publish a schedule of payments for next year. I have been reticent about doing that up until now because I want to be sure that when we set out a payment schedule, we can meet it. When I go around the country, what I hear from farm businesses is that what they really want is assurance on the dates. They want their money and they want it as soon as possible—that is entirely appropriate—but the main thing that they want is surety of timing. We have not been able to produce that until we had a clear IT delivery schedule, which we have only recently firmed up. We are now in the final processes of agreeing the dates that we can commit to on delivering information and we will publish that as soon as we possibly can.

One of the achievements that the team has made over the year relates to general information. In the 2015 payment scheme, payments were sent out and at some point later in the year payment letters were sent out. People received money in their bank account and, at a time that was detached from that, they received letters. However, in the 2016 processing year, we sent out payment information letters at the same time that people got their money, so we have demonstrated that the information that is available to farmers is better.

Another issue that has come up in conversation with the NFUS and individual farmers relates to understanding and knowledge of their entitlements. All entitlements are now shown online, so a farmer who logs in can see very clearly on the screen what their entitlements are. That is another piece of information that we have worked hard to make sure that people can access more easily this year.

Have you had any complaints about communication issues? What is the level of complaints about that specific aspect?

When I was round and about in the area offices over the summer, staff said that, recently, they have had very few phone calls about payments. Undoubtedly, the loan scheme has helped with that because people have had money so they were not waiting for their payments. That was very good. The level of calls to area offices has been pretty low over the summer.

Is the loan scheme being operated separately, in the sense that the loan is given pending the award of the grant?

Yes.

When you then calculate the amount of grant that a farmer is entitled to, do you deduct the outstanding loan so that they get a net grant payment—one that is net of the loan—or do you pay the full grant and, in parallel, take steps to recover the loan?

I am almost certain that we pay it net. If I am wrong on that, I will come back to you.

Okay. There might be cases where the loan exceeds the grant entitlement, so is that differential recouped?

There are two things on that. The BPS loan schemes are fixed at a certain percentage and the LFASS—less favoured area support scheme—loan schemes are up to a certain amount. That reflects variability. Quite detailed calculations are done by the loans team in advance of loan offers being made to avoid that very possibility, because no one wants to receive money that has to be recovered. However, there are occasions when recoveries have to be made, often because of penalties that have been applied to farms. The team does high levels of checks and calculations to make sure that people are not offered too much money as a loan.

Two questions arise from that. First, would it not make more sense to give them an advance partial grant payment rather than a loan, for which there is a completely separate form of administration?

That would be my very strong and clear preference but, unfortunately, we are not allowed to do that because, under EU regulations, we have to do all our inspections and process them before we are allowed to make an advance. This year, we do not plan to finish our inspection processes until December, so although that is significantly better than last year, we cannot make advance payments.

This is the same EU that has not had its accounts endorsed for 30 years. That is not your fault, it is the EU’s fault.

Finally, have there been any bad debts on the loans to date? What provision, if any, is there for bad debts on the loans?

As far as I am aware, we have had no bad debts to date and no provision has been made for them. However, if that is incorrect, I will write to the committee.

Great. Thank you.

10:15  

We have talked a lot about the process and the procedures. I want to touch briefly on the people who are powering through the work either behind the scenes or in public-facing roles. You both recognised how hard staff are working; I am sure that they will be pleased that you acknowledged that on the record. However, can you reassure the committee that the welfare of the staff is being looked after? We asked previously about the impact on staff wellbeing. Was there any restriction, particularly over the summer, on staff taking holidays and, if so, was that difficult to manage? Is there capacity in the team to cover holidays and sick days?

I cannot pay enough tribute to the staff in Saughton house and every area office around the country, because their dedication to the task is something that I have never come across before in all the jobs that I have done in the Scottish Government.

I take the welfare and wellbeing of my staff very seriously. I am based in Saughton house and I have been around all the area offices regularly to talk to staff about how they feel about things. I look regularly at our stats on absence management and staff flexi levels. Nothing in those cause me concern at a global level. Our absence levels, flexi balances and carry-over of annual leave are all roughly the same as those in the rest of the Scottish Government.

I pay tribute to the principal agricultural officers, who work closely with the staff in scheduling work across the year and who produce very clear timelines. Some members of the committee and of the Rural Economy and Connectivity Committee visited our offices in Ayr and Hamilton over the summer. They will have seen the detailed office plans that show the tasks that need to be completed and the staffing resources that are required each week. Everybody is encouraged to take their holidays at the times that they want to take them—they are planned for. Of course we still ask people to work overtime at some points in the year. Staff do so willingly and gladly, because they know that it has an impact.

One of the things that has been better this year than it was last year is that we have worked in collaboration with area offices to make it clearer to them what the peak times are and when staff might be needed to process applications. That is partly because we have been much more rigorous in engaging with, talking to and valuing the knowledge and understanding of area office staff about what they can do. There are periods of time when staff cannot do much to process the forms, so they use that time to manage their other work. The fact that we have been clearer about work processing and work planning is really beginning to help.

You mentioned overtime. Has there been an increase in overtime? How does your use of overtime compare to that in other parts of Government?

Our overtime levels have remained pretty static over this year and last year. Overtime is well managed locally by the area office staff.

You said that everyone is working really hard. Is that within the norms of a normal working week? I got the impression that people perhaps had to work around the clock and go beyond the call of duty. Are you saying that things are regular?

People are working very hard. It is a stressful environment because staff know that they will have to take phone calls from farmers if payments are late or if they have not understood the letters that they have received. I recognise and understand that that can be stressful.

I have asked the principal agricultural officers to make sure that systems are in place for staff to come and talk to people if they feel anxious and stressed about the conversations that they are having or if they are finding things difficult. I do not underestimate that. I have never worked in a small office, but people tell me that it can be very stressful. For example, you might be one of eight people in the office in Tiree, all of whom also live and work in the community. You will meet people who have been affected in the supermarket and when you pick up your kids from school. If there are problems with the system that you are working in, that can take over your life.

I understand that. We have had people suffering from stress because of work. We take that very seriously and we offer a high degree of support as an organisation. We do our very best to manage that. I hope that everyone in the organisation would recognise that I take wellbeing and welfare very seriously indeed.

One of the risks that was identified when we heard from the Auditor General previously was the idea of burnout and what could happen as a result—it would probably mean bringing in more agency staff, which would be another cost. The working environment is not really a pleasant one for anyone. However, you are not overly concerned—you say that it is a stressful environment—

It is.

Are staff impacted by that stress or are you not seeing any real evidence of that?

One thing that we have done this year is that we have looked at vacancies and at skills. We have looked at jobs where people have been on temporary promotion for a long time. We have a resourcing plan in place; we have taken steps to recruit a number of key posts. There has been an influx of agricultural officers in land management teams and that has had a really positive impact on staffing numbers in area offices.

As Liz Ditchburn mentioned, we have also looked at the skills that we have around audit, finance, business management and business skills and we have recruited staff into those posts. We have taken a very proactive approach in asking what skills and capabilities we need across the organisation.

I have been very up front and open with all my deputy directors and all the C-band staff, and I have asked them to be very clear about what resources they need to do the job in hand. I am not asking people to work 24 hours a day; I want people to tell me what resources and skills they need to do the job and then we go out and try to source the resources for them. Our first port of call is always to look within the Scottish Government.

Thank you for that. I wish all the staff well and for those who have not had a holiday yet, I hope that they get one soon.

I would never be in the business of denying somebody a holiday.

There was a key recommendation on knowledge transfer in the Auditor General’s report. In your busy work environment, how are you ensuring that key contractors transfer their knowledge to Scottish Government staff? What is the end date for that?

First, we absolutely agree on the importance of knowledge transfer and the risk that would come from not doing that well.

I think that one of the reports mentions that we have already identified some of the most critical contractors—the people who hold the most knowledge personally. Work shadowing arrangements are in place for those critical contractors so that that knowledge is starting to be passed on.

I understand that all contractors now have a plan in place for knowledge transfer. Those plans are at varying stages of implementation. That is partly because there needs to be someone to whom that knowledge can be passed. That is related to the recruitment schedule that Elinor Mitchell talked about.

It is fair to say that we are absolutely on the case on this; there has been some good progress but there is still some way to go. The way in which the contractor team and the Scottish Government team work together has really developed and matured over the year. Elinor and I were at an all-staff session with all the CGI staff contractors and it feels like one team now. It feels as though these are integrated teams, with the contractors and people from the Scottish Government working side by side. Of course there is formal knowledge transfer but the knowledge transfer through working together and through working on common goals is also significant.

If you could give us an update on that as the work progresses, that would be much appreciated. I thank you both for coming along to give evidence to the committee.

10:24 Meeting continued in private until 11:01.