Audit Committee, 16 Sep 2003

Meeting date: Tuesday, September 16, 2003

Official Report 253KB pdf

---

“Individual Learning Accounts in Scotland”

We move to agenda item 4, which is the inquiry into the Auditor General for Scotland's report on individual learning accounts.

I welcome the witnesses to this meeting of the Audit Committee. I understand that we will be considering the facts contained in the report "Individual Learning Accounts in Scotland". We will be asking questions about the design and management of the ILA scheme, the action that has been taken since the ILA scheme closed, and the development of the successor individual learning account scheme.

I ask the witnesses to introduce themselves, starting with the nearest to me in the front row and moving out and back. I invite Mr Frizzell to start.

I am the head of the Enterprise, Transport and Lifelong Learning Department, as it has been called since May this year. On my left is Laura Barjonas, who is head of the team in the division of the department that is dealing with the aftermath of the original ILA scheme and the design of the proposed new scheme. On Laura's left is David Stewart, who was the head of the division when the original scheme was designed, although he has now moved to another part of the Executive. On David's left is his successor as the head of the funding for learners division in the department.

I am a senior director responsible for skills and learning at Scottish Enterprise. My colleagues and I will be democratic about introducing ourselves.

I am director of finance at Scottish Enterprise.

I am project manager of ILAs at Scottish Enterprise.

We have apologies from Mr Robert Crawford, who is in the far east.

Good morning. I am the chief executive of Highlands and Islands Enterprise.

I am director of skills for Highlands and Islands Enterprise.

I am the head of internal audit and compliance for Highlands and Islands Enterprise.

I am the chief executive of the Scottish university for industry, which is known to the public as learndirect Scotland.

For the benefit of members, I point out that, during the previous agenda item, the official report staff had some difficulties because, as a result of the dynamics of the chamber, members were turning round to speak and were not speaking directly into their microphones. If members would speak into their microphones, that would also help me, because I have an ear infection and it sounds to me as if we are in a biscuit tin.

Mr Frizzell has said that he would like to make an opening statement before we ask questions. I invite him to do so.

I welcome this opportunity to consider the Auditor General's report on the individual learning accounts scheme in Scotland. Overall, the scheme achieved its objective of facilitating access to lifelong learning, but, as with the parallel scheme in England, there were flaws, which, as the accountable officer for the department, I very much regret. Nevertheless, I believe that we can learn lessons from those flaws, which, in due course, will enable us to devise a better successor scheme in Scotland.

ILAs were a UK-wide flagship initiative of the Government that was elected in 1997, which was developed by the Scottish Executive after devolution in 1999. We can learn a number of lessons in Scotland, both from the experience of the scheme in England and from how the scheme operated in Scotland.

ILAs were an innovative attempt to broaden awareness of and involvement in lifelong learning. As the Auditor General's report notes, the scheme proved popular with genuine learners and learning providers. An independent evaluation of ILAs in Scotland that was carried out in 2002 provided positive feedback on the benefits of ILAs from learners, learning providers and other stakeholders. The success of the scheme in achieving its main objectives makes it all the more regrettable that we had to close it down, and all the more important that we find out what went wrong and learn the lessons before we introduce a new scheme.

ILAs are an important plank in the Executive's lifelong learning strategy, which was launched in February. In the partnership agreement that was published after the May elections, ministers committed to introducing an improved ILA scheme. Since the original scheme closed, the department has been working hard to understand what went wrong and why it went wrong and to identify the lessons to be learned so that the same problems do not arise again. We are ensuring that the risks are properly assessed so that, as well as avoiding a repeat of what went wrong before, we can anticipate and avoid new problems that might arise next time round.

Ministers have still to decide the precise form of the new scheme, but the intention is that it will be more targeted than the first one to achieve a better focus on non-traditional learners, which is where additional funding is likely to have the most impact by encouraging increased participation in learning. Under the new scheme, we intend to use the Student Awards Agency for Scotland rather than Capita Business Services Ltd to administer the scheme. The Student Awards Agency will work in partnership with the Scottish university for industry—which is known to the public as learndirect Scotland, as Frank Pignatelli said—to deliver the new arrangements.

I am sure that the committee wants to understand what went wrong and what lessons we have learned for designing and implementing the new scheme. The department is determined to ensure that the shortcomings of the previous scheme, which led to its closure, will not be repeated.

You touched on many issues in your opening statement, but it was more about the history and mechanics of the scheme. What did the ILAs scheme seek to do in creating a culture of lifelong learning in Scotland, which had not really existed prior to the scheme?

The scheme attempted, in a relatively non-bureaucratic and easy way, to broaden the opportunity for people to get into learning for the first time or, for people who had some post-school learning experience such as evening classes, to get back into learning. The principle was that the Government would give a subsidy towards the cost of courses but people would be required to make a personal commitment. The scheme was an innovation and had not been tried before. It aimed to give learners control over what they wanted to do. The scheme used the Scottish university for industry as an intermediary to connect learning providers or opportunities with people who wanted to embark on learning.

The Scottish Executive's target was for 100,000 accounts to be opened by 2002. Why were there not other objectives for the scheme, for example in relation to the overriding desired outcome of getting more people into learning?

That outcome was consistent with the Executive's wish to have a properly skilled and qualified work force. It was hoped that an outcome of the process would be a more skilled population and, therefore, a better-qualified work force that could meet the future challenges of a competitive economy. The combination of the lifelong learning and enterprise agendas in one department connects skills and economic development. The outcomes sit firmly in that area.

What did you do to ensure that the training offered by learning providers met minimum quality standards?

One of the shortcomings of the original scheme was that although we had a registration process with SUFI, we did not have an accreditation or quality control policy for learning providers. We will seek to rectify that in the next scheme.

I will deal with the interaction between your department and the Department for Education and Skills. As you said, the scheme was introduced by the 1997 UK Administration and driven forward by the 1999 Executive. However, it appears that your department was not represented on the DFES programme board. Why was that? Was the reason historical? What effect did that have? Is the situation on-going in any shape or form?

With hindsight, I think that it might have been helpful if we had been represented on the programme board because we would have been closer to what was going on in the DFES. However, the arrangement that we adopted was not particularly unusual. It was consistent with the way in which government was done before devolution. If there was a Whitehall lead, as it was called, it was often literally that: the project would be led from down south with appropriate participation from the Scottish end. In the case of ILAs, that participation was through home countries meetings involving the Scots, the Welsh, the Northern Irish and the DFES to keep everybody up to date. That was not an unusual way of proceeding across a range of policies. Some economies of scale were involved—in Whitehall, more resources are generally brought to bear on such projects than we can bring to bear in the Executive.

The project was DFES led, and there was a feeling that, if we were on the programme board, we would be likely to find that staff were going up and down to London to meetings that were entirely about internal DFES administration and not necessarily about matters that would affect all four Administrations. Therefore, we were content—as were the Welsh and the Northern Irish—to go for the home countries meetings as a means of keeping in touch. As was normal, we also relied on the DFES to let us know about any significant changes that happened and any developments about which we ought to know.

I appreciate that the civil service required to go through a period of change to recognise the advent of devolution, just as parliamentarians did. It was suggested that one problem of devolution would be the existence of two conflicting power centres acting combatively. However, in some instances—including, apparently, in the case of ILAs—something seems to have fallen through the middle between the two conflicting power centres. Did the DFES exclude you, did you feel it inappropriate to be included in more than the home countries meetings, or did the situation simply arise because neither realised what the dynamics of devolution were causing to happen?

Remember that the project started before devolution, so quite a lot of work was done on it prior to May 1999. Nobody would have regarded the way in which it was approached then—a Whitehall lead working with the departments from elsewhere in the United Kingdom—as unusual. It would not have occurred to anybody that having anything other than home countries meetings was an appropriate way to proceed. We were not specifically excluded from the programme board, but the view was taken that there was no particular need to be on that board, which would to a large extent deal with matters that were of interest to the DFES and its administrative procedures alone, rather than of interest to everybody else. The decision was a matter of how best to use our resources.

When devolution came along, we took over the project, picked up from where we had left off and took the scheme forward ourselves. The basic arrangements were conditioned and set down pre-devolution.

Are things now dealt with differently?

We have had to learn not to have the Whitehall lead in such matters. We go our own way. The new ILA scheme will be very much an ILA Scotland scheme. There are no current plans for a parallel scheme in England. The scheme will be designed entirely by us and we will implement it in our way.

Was the Scottish viewpoint adequately represented and was there an appropriate level of communication with the DFES?

With hindsight, I think that the communication could have been better. In my experience, over a long time pre-devolution, communication was never perfect and we had to work at it. On the occasions in question, communication at certain times might have been better than it was.

It has also been suggested that the DFES did not always keep you fully briefed on the implementation of the national framework. What action did your department take to ensure that your viewpoint was taken into account?

Matters were raised in a variety of ways. They would be raised bilaterally in day-to-day communications and the home countries meetings were an opportunity to raise issues. We were always faced with not knowing what we did not know. However, once we found out what we did not know, we always sought to ensure that, if the DFES had not told us about something on one occasion, they would tell us the next time.

I ask my colleague David Stewart, who was heavily involved in the interaction with the DFES, to add to that.

Eddie Frizzell has made the main point. We used the home countries meetings to try to keep in touch on all aspects of the project but, outwith the meetings, we had regular contact with the DFES by e-mail and telephone. The difficult issue was those developments of which we were not aware. One of the key tasks was to find out about those and follow them up with the DFES. We also discussed bilaterally with Wales and Northern Ireland the interests of the Celtic fringe, as it were.

I understand why such bilateral meetings take place. It is important that there should be amicable interaction. Is that the way in which matters are still dealt with, or has there been a change in procedure to ensure that the Executive's viewpoint is taken on board? Is there anything more than what, with all due respect, comes across as a fairly clubby meeting? Such meetings might be the best way—perhaps somebody will comment on that—but they might not. If they are not, how have we changed the pattern of interaction between the Executive and the United Kingdom department?

Since devolution, formal concordats have been drawn up between the Executive departments and their counterpart departments in Westminster. I think that those are published documents, which lay down a pretty firm basis for consultation, interaction and keeping each other informed where necessary. However, as I said, the issue does not arise in the next phase of ILAs, because we are doing our own thing entirely this time round. By definition, since devolution we do more in our own way across the piece. The interaction with Whitehall has therefore been significantly less for many of us in a number of departments.

I will deal with the evaluation of risk, risk assessment and the risk management that was put in place. ILAs are one of the worst cases that any of us has ever seen of fraudulent activity in a Government scheme. I bumped into one of the providers of ILAs on the ferry one day about two years ago. He said that he had enrolled six people, done some training for them and then sent a bill to Capita. The cheque came to him, but no one checked whether he had delivered the learning or even whether he was able to deliver it.

Paragraphs 3.21 and 3.23 of the Auditor General's report set out the risk management arrangements under the scheme. Who had overall responsibility for the security of independent learning accounts? [Interruption.]

The reason why I consult my colleagues and stop to think about your question is that that is one of the issues and one of the lessons for the future. There were a number of players in ILAs, and responsibility was spread among them. Capita was responsible for operating the scheme, but we did not require it to quality control providers. Scottish Enterprise paid out money on the basis of information that it got from Capita. We had designed the scheme. With hindsight, I think that it was hard to pin down overall security for the scheme, as nobody had responsibility for checking whether a learning provider was genuinely providing what it purported to provide. That was a quality control problem that undoubtedly existed.

Risk assessment was not high on the agenda, because the risk assessment that KPMG carried out in 1999 did not identify learning providers as a likely risk. There was an awful lot of focus on whether people would pay their £25 and whether there was a way in which learners might abuse the scheme, but nobody made the judgment that there might be an issue with learning providers. If the risk assessment had come up with the possibility of learning providers abusing the system as a serious risk, more attention would have been paid to providing a way of dealing with that.

The overall policy context for ILAs was that the scheme was innovative and we were not to overload it with bureaucracy. The scheme was meant to be administered with a light touch so that people would not be deterred from wanting to open an individual learning account. It is important to understand that background. Such schemes are often beset by bureaucracy, and there are often all sorts of forms to fill in. The policy with ILAs was to try to avoid that and to make the scheme as accessible as possible.

We appreciate that. However, the touch was clearly so light that it went too far the other way.

I want to return to the fundamental question of who had overall responsibility for the security of ILAs. Surely, as accountable officer, you had that responsibility.

Yes, the department is responsible for ensuring that the relevant players get things right, which means that the accountable officer has to answer for that.

So you accept that you had overall responsibility.

I do not accept that I was personally responsible for people making fraudulent claims. It is important to understand that the people who make such claims are responsible for doing so.

The department appointed consultants to complete a risk and issues register for Scotland in order to inform the procurement of the customer service provider. What specialist advice or information did the department seek when assessing the risk of fraudulent activity in the scheme?

KPMG was asked to make that assessment during the early stages. I think that David Stewart can give the committee a more accurate answer than I can to that question.

After examining the DFES register and deciding that it did not provide significant risk assessment in the area, we asked KPMG to carry out an assessment for Scotland. Because that assessment came up with very little, we did not actively pursue the question of risk and fraudulent providers. With hindsight, we know that that was obviously a mistake.

How was the risk of fraudulent activity reviewed as the detailed design of the scheme was developed? Was any further work done on that matter?

Because the matter was not flagged up as a key risk area in the earlier stages of the scheme, it did not come back into mainstream focus until the scheme was running and issues began to arise on a case-by-case basis.

You are basically saying that the department took the consultants' report at face value and put aside any analysis of whether a light-touch approach would have left it exposed to fraudulent abuse. Surely that must have been a consideration for the department.

The other aspect of proceedings was the intention to monitor through the enterprise network payments made under the scheme and the system itself. We felt that that further work would allow us to keep a check on the scheme's development and the processing of payments.

You have quite rightly drawn parallels with other activities in the department that are often criticised for being over-bureaucratic. However, are you saying that there was no discussion within the department about whether a light-touch approach would be open to fraud and abuse?

By virtue of the assessment initially by the DFES and then by KPMG that the area was not significant or problematic, it was felt that monitoring payments represented the way forward. Obviously, with hindsight and in light of the report under discussion today, it is clear that that turned out not to be the case.

Given the number of players involved, I want to clarify a point about responsibilities. Mr Frizzell, during this period, did advice to ministers come directly through your own department or did it come from any of the other players involved?

It was the department's responsibility to advise ministers. As a result, any interaction with ministers was by the department.

Am I correct in recalling that, following the suspension of the scheme in England, ministers assured members across a range of different forums—certainly in the chamber and in other communications, including answers to parliamentary questions—that it was not felt that similar problems existed with the scheme in Scotland? Presumably that was based on advice from within the department. Even without the benefit of hindsight and in light of the evidence that was available at that stage, do you feel that ministers should not have been advised to give such assurances?

I cannot recall the assurances that you are referring to, but I am sure that they will be a matter of parliamentary record and can be checked. If I could see that material, I could probably recall exactly when advice and assurances were given.

However, I should mention that the situation became serious in England before it did so here and that, at the point at which the DFES was becoming very concerned about the matter, we were not experiencing the same level of complaints about the scheme in Scotland. Although we had received some complaints, which we told the DFES and others about, we were not experiencing the same difficulties at that point. When we began to experience those difficulties in Scotland, we responded accordingly and ministers decided that we should wind up the scheme. However, we wound up the scheme later because it took longer for the problems to appear in Scotland.

I appreciate that this information is available to committee members, but I wonder whether Mr Frizzell could confirm for the record the date on which the English scheme was suspended.

I think that that information is contained in the report. I am sure that we can look it up.

The English and the Northern Irish closed their schemes on 23 November 2001.

Will you also tell us for the record when the Executive publicly acknowledged that there might be problems with the Scottish scheme?

I would have to check that, but it would have been around that time that Scottish Enterprise began to hold back payments. From recollection, I think that that would have been the point at which representations about the withholding of payments would have been made to MSPs and others. That would have happened around October or November 2001.

The announcement that payments in Scotland were being suspended was made in November 2001 and the scheme itself was closed in Scotland on 20 December 2001. There was therefore a period of suspension in Scotland.

Has KPMG subsequently been asked to provide a rationale for its failure to identify a risk?

We have not asked the company to provide a rationale, but it might have been asked to do so by the DFES.

Given my fixation on consultancies, I want to follow on from Robin Harper's question. Why was KPMG procured and what was its remit?

KPMG was procured initially by the DFES. I cannot remember the exact details of the procurement, but I have no doubt that the proper process would have been followed in that respect.

The DFES used KPMG quite extensively to assist it in developing the scheme and to undertake for it a number of tasks such as the risk assessment. We drew on the contract with KPMG as part of the overall UK arrangements for proceeding with ILAs.

Will you expand on KPMG's remit as far as security was concerned? If not, will you provide it for the committee in due course? I am curious to know what the company was directed to do because after all, if we go to outside companies for specialist advice, they will no doubt receive high rates of remuneration. Following on from my colleague Robin Harper's question, I think that once we know what KPMG was directed to do, we will find out whether it carried out those directions.

The DFES certainly contracted KPMG to carry out a range of jobs, which I cannot immediately reel off. However, I can say that it was asked to examine the question of a risk register, hence that particular piece of work. I also recall a discussion during one of the home countries meetings about the presentation by KPMG of a UK risk register and we then carried out the subsequent piece of work on a risk register with the company. That key piece of work was discussed at home countries meetings.

Can we see that documentation?

No doubt it is held by the DFES and I presume that there is a copy in our files.

If it is at all possible to furnish us with the documentation, we will deal with it after this meeting.

We will try to get it.

The Scottish Executive's internal auditors concluded that there was a risk of learning providers submitting exaggerated or fraudulent claims for payments. The auditors believed that that risk was greater than the risk of fraudulent activity among individual learners. What action did your department take on receiving that advice from the internal auditors?

That advice was contained in an e-mail that covered a number of other matters—it was not highlighted as a big, flashing risk. As far as I can determine, the advice was not acted on because it was felt that other controls were in place, such as a registration process for providers—English providers were required to register with Capita and we required providers to register with SUFI—and that continuing monitoring of the scheme would be carried out by the enterprise networks. The presumption was that there were probably enough safeguards in place.

I put the same question to Mr Carmichael as I did to Mr Frizzell: was Scottish Enterprise not concerned about the risk of fraudulent activity, given that the scheme was set up with such a light touch?

Yes, we had concerns at the outset of the scheme and we raised them with the department at the time. Our responsibilities—

Your concerns were specifically about fraudulent activity.

That and other issues. Our responsibilities were twofold: to make payments and to carry out monitoring. At the time of making payments, it was not possible to check those payments. We would receive a payment file from Capita and then pay against it. By its nature, the monitoring was retrospective.

Is it correct to say that you had concerns about the scheme being open to abuse because of the lack of an inspection process and proper risk assessment?

Yes.

Were you concerned that KPMG said that there was no risk when it carried out the risk assessment?

I do not think that we knew that at the time.

Was that information not handed back to you by Mr Frizzell's department when you raised your concerns with him, as you claim that you did? Surely that was the answer the department gave to that question today.

We had concerns and that was why we asked for clarification from the Scottish Executive. It responded by letter and allayed our concerns about the controls over the scheme.

In that letter, did the Scottish Executive refer to the KPMG risk assessment?

It did not respond on that point, but it responded on issues that arose as a result of our consideration of the risk.

Can we have a copy of that letter?

Certainly.

Can we also have a copy of your original correspondence with the department?

We raised the issue at a meeting originally, but the letter refers to the meeting.

I put the same question to Sandy Cumming.

We had similar concerns. We discussed them with the department and we received reassurances at meetings that the systems had been fully tested, although our role was not to monitor Capita.

As I understand it, Scottish Enterprise said that it had some responsibility for monitoring. Is that correct?

Yes, we were responsible for monitoring, but it was retrospective monitoring of claims made by providers. We did not undertake systems checks at the outset.

Therefore, are you saying that HIE had no responsibility for monitoring and that it was concerned with payments alone?

We had a responsibility for monitoring payments—a similar role to that of Scottish Enterprise. I was instructed in February 2001 to arrange auditing to meet the requirements of my role as accounting officer. We took a risk-based approach. The advent of high-risk providers in the north of Scotland came much later. The level of high-risk providers was minimal in September 2001. The payments that we had made amounted to less that £20,000.

Did you raise your worries about the risk of fraudulent activity among learners in writing to the Executive?

I do not think that we corresponded formally, but perhaps Sandra Dunbar will comment.

The risk of fraud was discussed at meetings in autumn 2000. We obtained assurances at that time that the systems at Capita to detect fraudulent activity either were already in place or would be put in place.

What type of fraudulent activity were you concerned about at that meeting—fraud by individual learners or fraud by the providers? There seems to have been a different assessment of the different levels of risk.

Both were discussed in September 2000 when we obtained assurances via the Executive that, in respect of fraudulence by the provider—where learners had not undertaken activity—spot checks would be made by Capita.

Do you have a written assurance from the Executive or did you receive the assurance only in discussion at the meeting?

We have a letter from the Executive dated September 2000 confirming that spot checks would be undertaken.

Can we have a copy of that letter?

There have been several requests for written evidence. For the benefit of witnesses, we will write to them after the meeting to request documentation so that there is a record of what they have been asked to provide. It is still early in the meeting and they have already been asked for a number of documents.

Given the points raised by Mr Carmichael and Ms Dunbar, I ask Mr Frizzell or one of his colleagues to say on what basis they gave assurances that the matters referred to were not a problem or were being addressed. Were they monitoring the situation themselves, or was information coming from the DFES? Why did Mr Frizzell's department give assurances that the matter was being addressed?

The safeguards in question were those that we understood—incorrectly—to be in place via the general scheme process agreed with Capita through the DFES. We thought that it was safe to rely on the payments system that was in place throughout the UK and had been agreed by Capita and the DFES. At the other end of the process, we expected Scottish Enterprise and Highlands and Islands Enterprise to monitor the payments at local level and Scottish Enterprise to satisfy itself in its accountable officer role—to the extent it thought necessary—that the process that Capita had in place was sufficient for its responsibilities for making payments in Scotland. The problem was that the processes that were in place between Capita and the DFES turned out not to be as robust as we had understood them to be. The monitoring at the other end between the enterprise network and the providers took place over a time scale that was delayed for longer than we had originally hoped. The combination of those two situations created the problem of detecting the fraud.

Was it not the case that the security and fraud theory of Capita and the DFES was not borne out in reality in the practice of SE and HIE? From what you say, I understand that you preferred the theory to the practice, although I could be wrong. Despite what you were told about the implementation of the scheme, there seems to be a perception that the theory was right.

Those discussions were held at another stage. I am talking about discussions that took place before June 2001 and the swell of complaints around the scheme, and were about the theory and the processes that were in place at that early stage. An important part of the theory related to the putting in place of the monitoring arrangements so that any problems in the system could be indentified.

What about September?

September?

I believe that Mr Carmichael talked about September. What happened in September, when the theory period had ended and problems had begun to be flagged up?

I think that Mr Carmichael was talking September 2000—that is, the previous year—when I wrote a letter to the enterprise networks about the monitoring that we would expect to be carried out. The letter of February subsequently amended that information.

Mr Frizzell, paragraph 3.26 of the Auditor General's report talks about

"a number of concerns about the audit and accountability aspects of the contract with Capita."

How did your department respond to those concerns? This morning, Scottish Enterprise told us that it had concerns, which it relayed to you. However, we have also heard that your department weighed up concerns while waiting for complaints to be made. Could you explain exactly how that process worked?

The chief executive of Scottish Enterprise personally raised concerns with me by telephone. Those concerns were not specifically about the risk of learning provider fraud; they were about our approach to the issue and the fact that Scottish Enterprise would have preferred to use its own scheme. I think that that letter was seen by the Auditor General—

A letter? Initially, you said that he contacted you by telephone.

He followed up the telephone call with a letter.

It would be interesting to see that letter.

I took the letter extremely seriously and asked for a review of our progress and for advice on the terms of the reply. The letter was examined carefully in the department and the wording of a detailed, two-and-a-half page reply was agreed. That response sought to reassure Scottish Enterprise on a number of points and asked the organisation to get back to us if there were continued concerns. When no further communication was received, I assumed that we had settled the matter.

David Stewart is better placed than I am to talk about the day-to-day dialogue with Scottish Enterprise.

That exchange took place around September and October 2000. We then had further discussions in the steering group about the way in which we would work with Capita and decided to make one change to the arrangements. We had said that we would expect Scottish Enterprise to monitor the entire contract with Capita but, in light of the fact that the DFES was raising a number of policy issues over Capita and because we, rather than the enterprise network, had the policy lead, we decided that we would meet regularly with Capita in relation to the monitoring of its contract. However, we reaffirmed that we would continue to expect the enterprise network to monitor the payments that were made under the contract and its accountable officers to satisfy themselves that the Capita system was sufficient.

Mr Frizzell, in August 2001, after you had received a number of complaints concerning the activities of some of the learning providers, your department wrote to all providers drawing attention to the detailed rules and requirements of the scheme. Was not that detailed guidance provided from day one? Were there any alterations in the subsequent issue of guidance in August 2001?

General guidance on the operation of the scheme was available to learning providers and anyone else who wanted to see it on the website. Clarification was issued to learning providers in August 2001 in response to the issues that were being raised with us at that point.

Was there a significant difference between the two sets of guidance?

We clarified a number of issues in August. I cannot, off the top of my head, give you a line-by-line comparison, but I can say that the clarifications were made in August in the light of issues that were being raised with us.

Was Scottish Enterprise satisfied with the clarification that was issued in August 2001?

Yes. That guidance was essential for the monitoring process that we were carrying out. We needed to monitor the activity against the set guidance.

Mr Frizzell, paragraphs 4.3 to 4.6 of the Auditor General's report indicate that there was confusion between your department, Scottish Enterprise and Highlands and Islands Enterprise over responsibility for the auditing of Capita's internal control systems. Why did that confusion arise?

I am not entirely clear why the confusion arose. As the Auditor General records in paragraph 4.4, in our view, the letter of February 2001

"provided definitive clarification of the role and responsibilities of both enterprise bodies in respect of audit and monitoring."

However, you have indicated that the internal controls for Capita were included in the contract that was awarded, which was held by the department.

I am not quite sure what your question is.

You have indicated this morning that your colleagues in Westminster were leading the negotiations and were involved in the awarding of the contract to Capita. In that case, why should the difficulty that was experienced have arisen?

You have indicated that everyone should have been aware of their roles and responsibilities but you have also told us that the systems were not robust and the other witnesses have told us that they were unaware of them. Why was the system not working?

One of the problems that arose was that, as it turned out, Capita was not carrying out the registration checks that we expected it to be carrying out. However, in our view, the letter of February 2001 set out what Scottish Enterprise and Highlands and Islands Enterprise were expected to do with regard to the monitoring of payments.

Was Capita not carrying out the registration checks as expected because it was not adhering to an instruction or because it had not been so instructed?

I would have to go back to the documents to tell you exactly, but the extremely detailed contracts—about 100 pages long—between Capita and the DFES and between Capita and us set out everything that was expected of both parties signing the contract. To that extent, we had expectations that Capita would do a range of things. As we went through the monitoring process, it became clear to all parties that the ILA scheme was bigger and trickier than we had thought. At a number of meetings, both with the DFES and with Capita, other issues arose relating to how best to monitor and develop the scheme. One of the problems was that the DFES had Capita do a number of things under its contract, and Capita tended at that stage to give preference to dealing with the larger DFES requests rather than the smaller Scottish requests. We pursued those matters through the monitoring process, but the situation had overtaken us by August 2001, when the problems were significant.

I want to clarify the extent to which the department in Scotland had its own monitoring processes with Capita. You said that there were problems, in that it was difficult to get information from Capita.

It took time to develop the management information required under the contract. We had several exchanges with Capita in order to ensure that we were getting the range of information that was specified in the contract in the terms in which we wanted it. However, that was primarily around numbers of learners, types of programmes and activity that was under way. We looked to Scottish Enterprise to develop the monitoring of payment under its monitoring arrangements.

You stated that Capita had not carried out the registration checks that the 100-page contract seemingly set out. My understanding is that one of the central aims of a private finance initiative is to try to transfer risk from the public sector to the private sector. Under the terms of the contract, did Capita have liability for some of the losses that the public sector suffered and if not, why not?

We are talking about registration checks—not accreditation checks—on English providers operating with the DFES and through Capita. Those checks were distinct from the separate registration checks on learning providers operating under the Scottish scheme through SUFI. Distinctive Scottish checks were in place and were separate from the checks that Capita would have made under the English scheme.

What bearing did Capita's failure to carry out proper checks south of the border have on the scheme north of the border, given that there were two separate checking systems?

There was an element of reciprocity in that, initially, learning providers that were accepted through the scheme down south could be accepted into the Scottish scheme. When that was identified as a potential weakness, the arrangements were changed such that learning providers had to register directly with SUFI.

Was that done retrospectively?

It was done when the weakness was identified.

But was there a requirement to check retrospectively the learning providers that Capita had checked already?

As I understand it, at that point, the providers remained within the system.

I ask Mr Pignatelli to comment on that.

In the English system, the contract with Capita allowed learning providers to register direct and to self-certify. It was clear to us that any English registered learning provider could offer provision in Scotland. The Auditor General noted in paragraph 3.16 of his report that that negated the gateway approach that we had in Scotland. Although it has been suggested that we did not accredit learning providers in Scotland, given the nature of Scottish provision all the providers on our register had track records. If they had been working in national programmes they usually had the Scottish Quality Management System kitemark. If they came from the formal system, they had the Quality Assurance Agency for Higher Education kitemark. We felt that our system was robust in relation to recording and requiring providers to go through the process.

When it became obvious around June 2001—six months after the scheme took off—that there was a problem, we issued to all existing providers a code of practice. On the point about retrospection, 932 providers were contacted and asked to sign the code of practice, which was about their meeting quality standards and providing proof of that and the financial arrangements that they had in place to process funds. Of those 932 providers, 865 responded immediately and complied with the code of practice. We continued discussions with the 67 providers that did not comply. Some had gone out of business, some had moved on and some did not submit information.

The legal advice that we took at the time was that it would be difficult to push the code of practice, given that the scheme arrangements did not require providers to sign one. Within two or three months of the publication of the code of practice, the Executive's Enterprise and Lifelong Learning Department issued a strong letter to all providers. As the Auditor General pointed out, there is no doubt that the distance learning providers from England were a serious weakness in the scheme. The fact that self-certified learning providers with no quality audit were able to offer provision in Scotland was a significant weakness that we have identified as requiring serious attention in the new scheme.

I have questions for Mr Carmichael and Mr Cumming about the process of checking the payments to learning providers. Scottish Enterprise and Highlands and Islands Enterprise did not start to check payments until the scheme was at least nine months old. What were the main factors that contributed to the delay in introducing monitoring? Would the earlier introduction of monitoring of learning providers have helped to detect and deter improper and fraudulent activity in the scheme?

First we clarified the various organisations' responsibilities in the letter of February 2001. There was a lack of meaningful management information from Capita and a low level of activity in the scheme until August or September 2001. We accept the concern about delays in monitoring, which the Auditor General raised in his report. It is almost impossible to quantify the impact that earlier monitoring would have had. There was always an expectation among providers that there would be monitoring. By definition, monitoring is retrospective so the expenditure had to be incurred before we could carry out monitoring. Earlier monitoring would have highlighted weaknesses in the scheme a bit earlier, particularly as guidelines were not issued until August. If they had been issued earlier, that might have made a difference to the scheme, but it is impossible to quantify the exact difference.

We carried out monitoring from the start of the scheme. We carried out our own monitoring by analysis of activity in comparison with other information. We did not undertake monitoring visits from the outset, but we monitored the information available. When I received the letter in 2001 about arranging auditing to meet the requirements of my role as accounting officer, we identified the need to appoint additional resources to Sandra Dunbar's team. Those arrangements were in place in May 2001.

The situation was different in the Highlands. The number of high-risk providers was low right up until September 2001. I believe firmly that the action that we took at the time was appropriate.

In its report in October 2002, Audit Scotland said:

"On the basis of our review we have formed the opinion that the relatively late arrival in the HIE area of potentially fraudulent practitioners operating elsewhere throughout the UK, together with the monitoring structure installed by HIE and the body's prompt reaction in investigating and visiting suspect providers, has helped to mitigate the level of potentially irregular claims."

Mr Pignatelli, you talked about your concern that the distance learning providers in England were getting access to the scheme without basic checks of their eligibility being carried out. Did you register those concerns at an early stage? If we in Scotland had designed such a scheme, how might we have avoided such problems?

The concerns to which I referred were registered around June. The scheme began to operate towards the tail end of 2000, so by June 2001 it was clear that a number of approaches had been made via MSPs and by people writing to SUFI and to the Enterprise and Lifelong Learning Department.

The big problem for us was that, having identified that there was a distance learning issue, we had no control over it. We knew to some extent all the people on our database, because they had entered information in it. There were also follow-up questions and, in some cases, follow-up visits; we knew many of the providers. Looking back, I think that it is apparent that distance learning providers who had an intention to defraud the system would probably have been able to do so. We can put in place very strong mechanisms to ensure quality for direct provision, but the distance learning element was a weakness. In June, when we issued the code of practice, we picked up all the issues that had been raised and required people to register with us.

Distance learning is a new, exciting and innovative approach that involves an element of risk. In June 2001, when we saw what was happening, we examined some of the mechanisms that should be put in place. Organisations such as the British Association for Open Learning and the Open and Distance Learning Quality Council have mechanisms for ensuring quality in the provision of open learning. We immediately got in touch with both organisations and asked what kind of mechanisms we should put in place for distance learners.

Because of the aggressive marketing and overselling of distance learning, sometimes providers sent out materials to people but failed to offer tutor support. As soon as we became aware of that, we insisted that providers should not simply send out CD-ROMs to people and that there should be robust tutor support, linked to the BAOL and other quality marks. On reflection, if we had anticipated the sort of thing that happened we could have taken steps to prevent it. However, as soon as the information was made available to us, we took steps to ensure that risk was minimised.

Given that ensuring quality in distance learning providers is a well-recognised problem, should the issue not have been picked up in the design of the scheme?

Our role was always straightforward—we acted as a broker. If someone wanted to learn, they phoned our helpline. We should not forget the popularity of the scheme in the first two years of its operation. We launched our services in October 2000 and our helpline received 700 calls a day, seven days a week. The calls were not simply about ILAs. Some people who want to learn want face-to-face learning, some want part-time learning, some want full-time learning and some want correspondence courses. We had no difficulty in handling all those transactions. We referred people to correspondence or online courses, usually run by providers that we knew.

Between October 2000 and June 2001, we were unaware that a number of unscrupulous distance learning providers down south were registering and self-certifying on the Capita database, which they were able to access, and gaining access to clients in Scotland. That would not happen in Scotland, where the minimisation of risk in the quality of learning providers was significant. However, Rhona Brankin is right to say that we should have given more attention to the problem. As soon as it was identified, we took steps to minimise the risk.

Did Capita provide weekly or monthly activity reports on the number of accounts that were opened and on expenditure? Did the department have numbers on complaints and performance against agreed service targets?

The contract contained arrangements for monthly reports but, in addition, regular communications were made by e-mail and telephone and in other forms on other issues as they arose. In general, reporting was monthly.

Did you not receive weekly reports, although they were outwith the requirement of the contract?

As I said, throughout the course of the scheme, much more regular contact took place over and above the monthly reports, which were top-line reports.

Did you have sufficient resources to study those reports to ensure that they reflected the position in Scotland?

We studied each report carefully. The reports formed the basis of our monitoring discussions with Capita, at which we talked about any concerns with the material. Over several meetings, those discussions led to an improvement in the management information for monitoring that we received from Capita under the contract.

You were in a different situation from that of your colleagues from the DFES, who told the House of Commons Public Accounts Committee that they did not have sufficient resources to review the reports. You say that you had appropriate resources.

We certainly studied the reports and discussed them at the meetings, so on the narrow question whether we had time to consider the reports, the answer is yes.

We have exhausted our questions on risk control, so we will move on to ask about recovering overpayments from learning providers.

The Auditor General's report reproduces data from Scottish Enterprise and Highlands and Islands Enterprise from which the suspected level of irregular claims was extrapolated on the basis of sampling. The estimate was that irregular claims had a value of almost £4.5 million and represented 24 per cent of total claims received. Will Mr Frizzell tell us the current estimate of the total value of irregular payments and how much the Enterprise, Transport and Lifelong Learning Department has recovered?

The extrapolation still produces a figure of about £4.4 million that might have been irregular claims. We think that we have overpaid £1.2 million. We have calculated irregular claims of £4.4 million, but we are withholding £3 million. We are concentrating on recovery of £1.2 million.

What is the department doing to recover other outstanding overpayments or irregular payments to learning providers?

We have not yet pursued recovery from the 11 providers in England that remain subject to possible criminal proceedings—they are being investigated for fraud. About £500,000 of that £1.2 million might be attributable to those providers, which leaves about £700,000 that we seek to recover in Scotland.

Initially, 10 cases were referred to the Crown Office for consideration as prima facie fraud as opposed to simple irregular payments. When we first became aware of problems, the definition of irregular payments ranged from an instance of a form not being signed to serious situations such as potential fraud. The 10 more serious prima facie cases were referred to the Crown Office, which decided, after due deliberation, that there was insufficient evidence to pursue prosecutions for fraud. In fact, no one in Scotland will be prosecuted for fraud; the position in England has still not been resolved.

We are now seeking recovery through administrative action and are writing to people, saying that we believe that there were irregularities and that we would like money back. Up to this point, we have recovered only about £2,500, but now that the Crown Office decision is out of the way, we are able to pursue recovery.

I want to clarify a couple of points in that response. Are you or any of the witnesses able to indicate when we might expect the English courts to reach decisions about providers against whom criminal prosecutions remain outstanding?

I really do not think that we are able to answer that question. It could be a very long process. After all, prosecuting a fraud is more complicated than most other prosecutions.

Although I will ask this next question, I will understand if you do not want to go any further than you have in the comments that you have already made. Are you prepared to estimate the amount that the public purse will ultimately have to write off?

The figure will be no more than £1.2 million, and I will want to get that down as far as is humanly possible.

I want to ask about the other side of the coin: outstanding payments to learning providers. I appreciate that you have already touched on the matter. The Auditor General's report points out that the value of outstanding claims at 31 March 2002 stood at £5.6 million. However, subsequent payments to learning providers reduced the level of outstanding claims to £1.7 million by the end of November 2002. How much currently remains outstanding in claims for payment?

We are still holding £3 million against claims. We have not paid out that money.

How many providers are continuing to pursue claims for payment, or did you cover that in your earlier answer?

Altogether, we are withholding payments from 40 providers.

Is it fair to say that you do not expect to be pressed hard to make payments out of the £3 million for those claims?

There has been a fair bit of pressure, which has diminished as time has gone on and as we have found a way of releasing funds by applying error rates and working out what might be reasonable claims. However, the pressure is not off us. We are still receiving letters from MSPs on behalf of constituents. One problem is how we ensure that we do not penalise bona fide providers who have delivered learning and are entitled to payments. We have tried to strike a balance by paying out where we think it is reasonable to do so and not paying out where we really feel that that would not be reasonable.

We have made part-payments of about £440,000 to a number of providers. However, the figure is included not in the £3 million that has been withheld, but in the money that we have already paid out.

One might wonder why we are not being sued for a substantial payment, for example. That is an interesting question.

It has crossed our minds.

It raises the issue whether some providers realise that they have been clocked and so will not push the issue, but I do not know. We are by no means through all that yet.

What evidence do you have about the impact that the suspension of the scheme had on learning providers? You indicated that in some cases the impact was quite profound, especially for smaller providers.

The impact on some providers was serious. Irregularities were uncovered in relation to some major providers, including some that would be quite surprising to you, but it was quite clear that there was no fraud and that they just had not done things properly. It was clear to us that there would be a major impact on some small providers. That is why we sought a way through by calculating how much it might be reasonable to pay—so that we did not put people out of business unreasonably. Perhaps that is not a sufficiently detailed answer, but I do not know that I can give a more detailed one. Laura Barjonas might be able to say more.

In the period immediately after closure, when the validation process was on-going, a large number of learning providers were under review. With the enterprise network, we sought to move through that as quickly as possible, while recognising that we had to ensure that we were as robust as possible in examining possible irregular payments. Things have moved on considerably since that time. The number of learning providers in relation to whom payments are being withheld and have been withheld for some time is much smaller now. They are the ones in respect of whom we have continuing concerns about the irregularity of the payments.

With the benefit of hindsight—to use a phrase that has cropped up more than once this morning—and knowing what you know now, is there anything that could or should have been done differently following the suspension of the scheme? Could the process have been expedited more effectively, in a way that might have had less impact on training providers that had payments outstanding?

I do not think that we could have done anything differently. If we had paid out, we would have found ourselves in the impossible position of risking more public funds—and being criticised for that—and of rewarding possible fraud. We worked on the basis of the information that we had at the time.

I know that we were all keen to strike the balance between not throwing more good money after bad—as it were—so that taxpayers' money was safeguarded, and paying out where we felt satisfied that there was justification for some payment. I think that we would not have acted any differently. A lot of consideration was given to having criteria that could stand up and provide a basis for making payments. The department was very conscious that we could not just withhold all the money and treat everybody as if they had deliberately tried to defraud the taxpayer when that was clearly not the case.

I want to return to the 40 providers in relation to which there are outstanding claims. First, how many of those 40 were registered through Capita and through the self-certification process south of the border? Secondly, how many of them were based in Scotland and registered through Mr Pignatelli's code of practice? Lastly, what criteria are used to determine whether the outstanding payments will be authorised?

The KPMG report into what went wrong is reported as having stated:

"In 19% of learning episodes sampled there was no enrolment documentation available to the consultants. In a further 27% of the sample the standard enrolment statement had not been completed".

Furthermore,

"Signature comparisons could only be undertaken in respect of 61% of the sample due to the absence or illegibility of a signature".

Indeed, in 7 per cent of cases,

"there were indications that the signature on the enrolment form was probably not the same as the signature on the ILA application".

The report goes on to state:

"In 53% of the cases sampled, no telephone number was available"

for students. Of those who did participate in the sample,

"32% said they did not receive the training for which learning providers had claimed. A further 293 of the 1,583 students who said they did receive training stated they did not make any contribution towards the costs of that learning episode."

Of all the different criteria, which ones are you emphasising in rejecting claims for payment? How many claims have already been paid for training that was not what the Executive expected?

There are a lot of questions there. It would be fair to start with the Capita ones.

The first question was about—

The 40 providers. How many were self-certified through Capita in England and Wales, and how many were Scottish companies registered through the SUFI code of practice?

I will need to ask my colleagues about that. I did not bring the information with me. I do not know whether someone did.

The figure of 40 is evenly split between Scotland-based and England-based providers. I do not know how many of those subsequently signed the code of practice. Frank Pignatelli might be able to say more on that. A number of them will have signed the code of practice.

If you can supply written evidence in that regard after the meeting, we would be perfectly happy to accept it.

Could you give us the split in cash terms?

Is that available?

Do you mean in terms of the balance between the England-based and Scotland-based providers?

Yes.

The proportion of Scottish providers is slightly greater. Do you mean in terms of the amounts that are currently being withheld?

Yes.

Your question is how the £3 million that has been withheld breaks down. I do not think that I have that figure here. I have got just about every other figure. We will be happy to provide it.

Does Mr Pignatelli want to add anything?

Just to say that a learning provider from Scotland could be on the database quite legitimately. Mr Frizzell pointed out that some of the cases result from slipshod administration and poor monitoring and follow-up of documentation. We must not assume that some of the people were actively involved in fraudulent activity. The point is that no system was available to make checks. When the signatures did not coincide, someone in the system should have been able to point that out. The 20 or so Scottish providers may be legitimate providers of learning opportunities to people, and have good-quality provision. They may have been involved in slipshod administration and, in some cases—I do not know whether the Crown Office would want to prosecute them or not—they were involved in things that should have been picked up by the system.

I asked a question about the criteria that you are using to make or reject payments. There is a list of criticisms. Which criteria are being used in determining whether to make a payment?

First, you referred to the KPMG study. It is important to say that the percentage figures relate to a sample that was skewed against the high-risk providers. The sample was not representative of all learning providers. As time has gone on, we have worked to validate more and more claims, so we have a much better basis for working things out.

I refer you to appendix 5 of the Auditor General's report, on "Policy guidelines on the release of funds and other action". We gradually worked through, increasingly releasing funds. As it says in the appendix, we had to "minimise losses", to

"take due account of the legal implications"

and to

"take … account of the cost/benefit implications".

The title at the bottom of the page is:

"Indicators of potentially irregular payments".

We had regard to what the errors might have been. We can give the committee more detail on that; it is written down. However, we were clear about what we would regard as a serious breach and what would be a less serious breach.

In the light of the information that you have and, in a sense, have not received, you are able to say categorically that there is no question that there will be any prosecutions. Can you envisage any development that might result in a prosecution?

I dare say that, if new evidence were uncovered in relation to a case that had not already been referred to the Crown Office, that would be conceivable, but I rather doubt it. The cases in which it looked as if there might be a chance of a prosecution were referred to the Crown Office and the prosecution authorities made decisions regarding them.

We shall move on to deal with the lessons that have been learned and the development of a successor scheme.

The Auditor General's report and Mr Frizzell's comments today indicate that a replacement ILA scheme is being developed as part of the Scottish Executive's overall strategy for lifelong learning. Can he tell us about the development of that replacement scheme?

We have learned lessons, one of which is that we have to be more rigorous about risk assessment—that is the starting point. As I said at the beginning, the judgment that learning providers were not a high risk was wrong. I apologise for that on behalf of the department. We were partly guided by KPMG in reaching that decision. We need to be a lot more imaginative about the scams that people can pull, basically.

That is happening. There is much more rigorous risk assessment now, not just in relation to the ILA scheme but across the Executive, where the approach to risk assessment has developed significantly in the past two or three years.

We have developed a new process. We are project managing the development of the scheme in a systematic way, using a project manager. We have a gateway review process, so that, at particular points in the development of the scheme, people—not those who are driving the scheme forward—can come in and assess whether everything has been taken into account, whether the risks have been assessed and whether the process should proceed to the next stage. We are moving forward. Our progress is slow, but that is a result of our stopping at certain stages to ensure that we have got the process right.

We have learned a key lesson in relation to learning providers. We need to have some quality control—it is not enough just to have organisations register with SUFI; there has to be a system whereby SUFI can check whether the learning providers have accredited quality systems. That is in hand. Linked to that, we have to make an assessment of which providers might be regarded as high risk or low risk. Intuitively, further education colleges or higher education institutions would be lower risk providers and providers with no history of learning provision or those that have only conducted distance learning would be higher-risk providers. Last time around, the absence of a quality control system was a clear weakness.

We have also learned that we need to have more of a contractual relationship with learning providers; it should not simply be a matter of issuing guidance. Last time around, there was guidance on the website and we clarified the guidance when we began to get complaints in the course of the following summer. Under the new scheme, we will require providers explicitly to accept the operational rules of the scheme. That way, if something goes wrong and we want to get our money back from learning providers, we can say that, in effect, they signed a contract. They should be aware of the deal.

The situation as regards learning providers will be a lot tighter. A lot of learning providers will find the new system a lot more bureaucratic than the old system, but that is how it will have to be. There will not be a light touch this time around. That is a significant difference compared with last time.

We will require providers to keep better records and to furnish information on request.

I have told you how we are advancing the project through the gateway reviews and how we will deal with risk. The audit and budget management of the scheme is another area in which we need to do better.

As members can tell, many players were involved last time round, including Capita. We will make the process much simpler this time. The Enterprise, Transport and Lifelong Learning Department will design the scheme and lay down the policy guidelines. SUFI and the Student Awards Agency for Scotland will be the other two players. We will not use Capita or anyone else—we will do things in-house and the SAAS will take on responsibility for payments. It is developing properly tailor-made information technology systems to allow it to do that.

We are tightening up in all those key areas, which is one of the reasons why it is taking quite a long time to launch the new scheme. We had ambitions for the new scheme to be launched rather earlier, but it is important that we get it right—that is why the process is taking rather longer than we had hoped. Nevertheless, the firm commitment in the partnership agreement that we will launch a new scheme remains.

If you look back, to what extent did the involvement of several organisations in the administration of the scheme contribute to its problems and errors?

It has become obvious that the more people who are involved, the more important it is to get communications right. In the old scheme, communications—whether with the DFES, with Capita or between the department and Scottish Enterprise—could have been better at several junctures. The involvement of fewer players and the fact that the administration of the new scheme will be much more of an in-house effort means that we should avoid the communication problems that we experienced last time.

Would you accept that the absence of a formal system of accreditation was a major control weakness?

With hindsight, I think that it was. We did not want to make the system too bureaucratic—there was meant to be a light touch. The absence of a formal system of accreditation was a problem, but it will not be that way this time.

I would like to clarify what you mean by "a light touch". Initially, I got the general feeling that the importance of that phrase was more to do with the experience of the student. The implication is that part of the problem was that that light touch went into the learning provider system as well. If we remove the light touch on learning providers by having more stringent quality checks, what are we doing about the light touch for the student? Are we still trying to achieve the open access that was key to the whole system?

Yes, we are. I did not mean to imply that the light touch related only to students' experience. We did not want the scheme to be too bureaucratic; we wanted learning providers to be innovative and to be prepared to deliver in ways that were different to the customary 9-to-5 method that colleges used.

We will provide improved guidance to learners to explain how the scheme works and we will explain how they can use their accounts. In addition, there will be guidance for learners on the need to ensure that they choose the learning provider and the learning that they want; SUFI will be very helpful in that regard. We will also send out a strong message to learners about looking after their ILAs—they will be told that they own them and that someone else cannot appropriate them for any purpose. Clear guidance will also be provided on how learners can make complaints if they have concerns about the system. The aim is not to tie them down to all sorts of rules—although they will still be expected to make a financial contribution—but to give them more guidance.

We want to target non-traditional learners and people who find it hard to get into learning, so we are conscious that we need to work with organisations, such as voluntary bodies, that might work with those people in order to ensure that such bodies have the necessary guidance and can help the learners through the experience. We are trying directly to cover the learners and those who might help the learners to get into the market.

Robin, do you have any points to make?

I presume that the department will assess the risk of fraudulent activity before the successor ILA scheme is introduced. What steps will you take to ensure that a robust control system and security arrangements are in place for the successor scheme?

That will be effected largely through the Student Awards Agency for Scotland, which is an agency of the department; we own it and it reports to ministers. People who have expertise in information technology security are advising us and we will arrange checks and auditing much more at our own hand. The quality control of learning providers will be a key issue in that. Payments will be effected through the SAAS system.

Will you tell us more about how you have involved other organisations in shaping the new scheme? What consultation processes, or opportunities for dialogue, are under way?

A lot, but Laura Barjonas will be able to tell you about them in more detail.

The process has been continuing over a longer time than was anticipated initially. However, that has been beneficial in terms of our being able to get detailed feedback from learners, learning providers and intermediaries, who will be crucial. After the conclusion of the scheme, consultants carried out an extensive independent evaluation, which involved a number of workshops and feedback from learners, learning providers and intermediaries. That contributed to the initial shape of the proposals. We continued to have discussions and consultation over that period.

Over the past couple of months, we have been involved in a number of intensive consultations with a large sample of learners throughout Scotland. We have held a number of learner workshops and learning-provider workshops throughout Scotland, which have involved intermediaries, trade unions, representatives from community education and a range of others who we think will have a stronger role next time round. That recent intensive consultation is still feeding into the development of the scheme and has been useful for our understanding of the potential operational impacts on learners and providers. That will give us a balance and help us to ensure that controls are in place and that the system is robust. At the same time, we are ensuring that procedures are in place that will work for the learners and the providers.

To what extent is that dialogue affecting thinking only on the operational elements of the scheme and to what extent is it still shaping the policy of the scheme?

There will be an opportunity for us to revisit elements of the policy and the overall management approach in the light of the feedback. That is part of what we should do in any case as part of the continuing development of the scheme.

I am trying to get a sense of what stage of development the scheme is at and what opportunities remain for anyone, including us, to influence and shape the new scheme.

The responses to the consultation have supported the need to reintroduce the scheme, and the need for increased focus on non-traditional learners and targeting. There has been much support for many operational elements. A number of areas have been identified where we will reconsider specific elements of the business model and consider how that feeds into the operation of the scheme.

My question follows on from that. Which bodies in Scotland that are concerned with quality in education have you consulted on the development of the successor scheme?

Existing bodies offer accreditation, but perhaps Frank Pignatelli could answer that better.

Since SUFI was established in October 2000, we have had continuing discussions with several key bodies, some of which have signed our memorandum of understanding. Such organisations include the Scottish Qualifications Authority, which accredits national vocational courses, Investors in People, and the Scottish Quality Management System. The SQMS kitemark is also important to us. We have been talking to the European Foundation for Quality Management.

All those bodies have different loci in different parts of the system; for example, the private and public sectors vary in their emphases. We have refocused particularly on distance learning and I mentioned two bodies earlier—the Open and Distance Learning Quality Council and the British Association for Open Learning. All those bodies say that they have approaches that are consistent with what we are looking for.

To add to what Eddie Frizzell said earlier, in terms of our approach to the new scheme, there is a detailed statement of about 200 or 300 pages about ensuring that there are no risks attached. We have given a commitment that we will undertake a case-by-case study of verification of the quality marks that people already have; only people who can produce such evidence will be involved. That is a continuing process that will be quite a burden. I was struck by a point that was made by some members of the committee, which was that we should try to minimise the burden for learners and ensure that we get the systems operating. I am confident that all the bodies I mentioned would be consulted.

The last part of the loop is to ask that a formal learner-provider agreement be signed. Part of that agreement would be about feedback from individual students. We have customer relationship management software that allows people to make complaints. The process in which people feed back when they are not pleased with aspects has so far been informal—it will now be formalised.

My postscript is that all the quality systems in the world could be put in place, but many learners have come to us from traditional providers who put quality assurance agency imprimaturs on their degrees or diplomas, but those learners still feel dissatisfied. We go beyond the formal systems to the learner experience.

Absolutely. Are there issues about the quality of the quality assurance systems? I am sure that we all recognise the importance of working with a learner, but there are issues concerning existing quality assurance systems that must be considered in the design of the new scheme.

That is an important point. In setting up learndirect Scotland, ministers in enterprise and lifelong learning signed up to a 10-point pledge to learners. Every person who engages with our organisation is asked to read the 10-point pledge that says to learners that we will get them learning when they want it, where they want it and at a pace and style that suits them. That is demanding, particularly for the formal education service. I am happy to say that 427 learning centres throughout Scotland now meet that pledge. To meet the pledge, they must go through a rigorous quality process.

People who were confident because they had a list of kitemarks on their note paper were coming to us and saying, "Of course we'll pass this system." They found it to be rather rigorous because we ask for evidence to support the kitemark, not just the kitemark. So, if a provider says that they can provide learning whenever someone wants it, that to us means evenings, weekends and during the summer. Often, that is not available. The combination of the strategy that has informed the creation of learndirect Scotland, plus the quality systems, should go some way towards what Rhona Brankin is looking for.

For clarity, will Mr Frizzell rehearse the precise steps that he plans to take to assess the risk of fraudulent activity before the successor ILA scheme is introduced?

We will have a risk event next Monday—it is just one of the events. We will bring in a whole bunch of people who have experience of the type of things that went wrong last time. That will include people from inside the department, finance people, internal auditors, solicitors and people who know about information technology and security. SUFI and the enterprise networks will also be involved, as will the Student Awards Agency for Scotland.

We want to bring together as many people with an interest as possible. We will think widely about things that might go wrong and try to anticipate them. Risk assessment is not easy: my senior people and I have spent a lot of time over the past two years on compiling for the department a risk register across the whole range of our business. It is not easy, especially with something new, but this is not new; we have experience from last time. First time round, it was a brand new scheme that no one had tried. With hindsight, it is easy now to ask why people did not think of particular risks, but it is not an easy thing to do.

What is the objective of the new scheme? A policy objective of the previous scheme was to have 100,000 learning episodes. The policy was driven by the numbers game; numbers were the measure of success. How will you measure success in the new scheme?

The thing about skills and learning initiatives is that they take a while to work through. We have not been set a target of having 100,000 accounts by a given date, so we are not being driven by that. We do not have, at the back of our minds, a voice saying, "Oh my goodness, we have to make progress. We can't stop and think." The commitment is to introduce the scheme and we are working towards doing that in the first half of next year. The outcomes that we want are twofold: we want people to feel more comfortable about getting into, or back into, learning; and we want to attract people who have been especially resistant to that. That is the reason for targeting non-traditional groups. We want to raise the skills base and the capacity of an important economic resource—the work force. That, in the broadest possible terms, is the answer to the question. That answer is consistent with the overall lifelong learning strategy.

What you say is interesting. In the old scheme, you were driven by the numbers and the need to roll the policy out. Did that lead to the scheme's downfall? Were a pile of projects ignored because they failed?

We live in a target-driven world. Targets can drive performance, but there is a downside to working towards a target for a given date. The old scheme was one of 20 things in the previous programme for government that my department had to deliver on fixed dates. What went wrong in the design of the scheme was that the risk assessment did not take account of all possible risks. The light touch was too light in relation to learning providers. At the time, there were good enough reasons for doing things as they were done, but we have learned that we cannot do things that way again.

I thank all the witnesses for coming along today. Not everyone received as many questions as did the department—it was always going to be that way—but I am sure that people realise that it is useful to have everyone here at the same time so that we can all follow the different lines of questioning. I thank you for your time and commitment and wish you a safe journey home.

We will move into private session; I hope to get through the other matters on our agenda by 12.30, but that will be in the hands of committee members. We will pause to allow members of the public and the press to leave. I suggest that we resume promptly at 5 past 12.

Meeting suspended until 12:09 and thereafter continued in private until 12:57.