Public Audit and Post-legislative Scrutiny Committee

Meeting date: Thursday, December 10, 2020

Official Report 609KB pdf

---

Section 22 Reports

---

“The 2019/20 audit of NHS Tayside”

I welcome our witnesses from Audit Scotland this morning. Stephen Boyle, Auditor General for Scotland, is with us in the committee room; Fiona Mitchell-Knight, audit director, audit services, and Leigh Johnston, senior manager, performance audit and best value, are appearing remotely. I give a special welcome to the Auditor General; it is the first time that I have convened a meeting since he took up his post and I offer him my warm congratulations.

I understand that the Auditor General will make a brief opening statement.

Thank you for your introductory words, convener, and welcome back.

Good morning, everybody. The sixth consecutive report on NHS Tayside updates the committee on the progress that the board has made against the concerns about financial performance and governance issues that featured in earlier reports. Today’s report sets out the progress that NHS Tayside is making in meeting its financial and performance targets and some of the risks that lie ahead.

The external auditor gave an unmodified audit opinion on the 2019-20 financial statements. She highlighted the board’s effective financial management and on-going improvement in governance arrangements. In 2019-20, the board delivered £26.3 million of savings but still required additional financial support of £7 million from the Scottish Government. The current three-year financial plan from 2020-21 aims to achieve break even each year, but significant financial savings will need to be made to achieve that.

The Covid-19 pandemic has had a significant impact on the focus and priorities of NHS Tayside and the effect of that on its financial position and savings targets have not yet been reflected in its financial plans; it now needs to do so.

Overall service performance improved last year, but long-standing concerns about mental health services resulted in a highly critical independent report on mental health services in Tayside being published in February 2020. NHS Tayside also needs to implement its plan to improve mental health services.

NHS Tayside is making progress under its new executive leadership team, but there are still risks ahead. The achievement of a balanced financial position depends on the successful delivery of a transformation programme that reflects its new and evolving priorities.

I am joined today by Fiona Mitchell-Knight, the appointed auditor, and Leigh Johnston. Between us, we will look to answer your questions.

I will kick off the questioning on the section 22 report. This is the sixth year that Audit Scotland has had to use its emergency powers to write a section 22 report on NHS Tayside, and I think that this is the eighth year that NHS Tayside has needed extra cash from the Scottish Government. The report makes it clear that, although progress is being made, transformational progress has been “slow”. The NHS Tayside board acknowledges that the pace needs to increase. In your opinion, why has the pace been slow?

I will start and then invite Fiona Mitchell-Knight to augment my remarks. Our sense is that NHS Tayside was making progress. You are right that that has been long in the making, given the number of reports that my predecessor chose to make on the challenges that the board faced over many years, including those relating to its financial position. The reasons behind those challenges are well documented. The committee is, of course, very familiar with the board’s challenges in delivering a balanced financial position and tackling governance issues. There are also the underlying cost issues of its organisation.

Until the pandemic struck, the organisation was making progress. That is borne out by the fact that the board reported that some of its service performance indicators were broadly in line with national performance targets and, in some cases, above those targets. That pointed to an organisation that was making progress, and then the pandemic struck.

As the convener touched on, our report notes that the board still has to make significant financial savings in order to deliver financial balance within the next three years. To do so in the context of the pandemic, it now needs to rapidly increase the pace and, in effect, deliver a new plan that shows how NHS Tayside will look in the future. Significant challenges remain, albeit that we say that progress has been made.

Does Fiona Mitchell-Knight want to bring an additional perspective to that?

Can I intervene before she comes in? We all acknowledge that the national health service is under huge pressure because of the Covid pandemic, and I have heard concerns from quite a few doctors around Scotland that there were huge pressures on the NHS before the pandemic struck in March, as we both know. Can we allow the narrative of pressures due to the pandemic to detract from the work that we were doing before the pandemic? Although the report is before us today and was published only last week, a lot of the statistics and figures predate the pandemic. The mental health vacancy figures are from November 2019, which is more than a year ago. Before your colleagues come in, can you tell me why the progress of transformation was slow before March?

There are a number of different aspects to that. I thought carefully about the section 22 report for NHS Tayside this year very much for the reason that you outlined: the fact that there has been a history of statutory reporting on NHS Tayside. Many of the factors in the report existed pre-Covid, so it is not sufficient to say that the pandemic, in itself, overrides the organisational challenges that existed previously.

We thought that it was necessary to update the committee on the progress that the organisation had made. That was our judgment. As the chief executive of the board has said, the pace needs to increase. The report documents the continuing need to tackle the organisation’s cost base and the progress that is being made on aspects of its governance arrangements. Those challenges existed, but progress was being made. It is now clear that, because of the impact of the pandemic, there are elements of a game changer.

In many respects, NHS Tayside has pivoted from a transformation programme and now identifies its remobilisation programme as its new plan for the future. As we note in the report, it now needs to move quickly. All NHS organisations will need a post-Covid plan, but NHS Tayside in particular needs to show its progress.

You talked about cost base and governance. Why do you think that progress on those two specific elements was slow prior to March? I understand that progress will have slowed since then—there is no question about that—but why was progress in those areas not faster before March?

I will touch on the cost base and then ask Fiona Mitchell-Knight to say a bit more about some of the specifics of the governance and the progress that has been made.

As we touch on in the report, there are three main historical reasons why the board’s cost base is higher than those of other NHS organisations: its staffing model, its prescribing arrangements and its in-patient costs. In particular, we refer to the work of the assurance and advisory group and the external support that NHS Tayside was receiving from the Scottish Government and others to tackle those factors. The committee will have heard much on that.

The situation is undoubtedly complex and requires engagement from staff, patients and the general practitioner community. In the round, however, we said that progress was being made. The complexity of the situation is one factor, as that will have affected the pace of progress. Another factor is the disruption that the organisation has experienced in recent years. That is all the more reason that NHS Tayside now needs a plan to tackle the issues and set out how it will deliver services in future, especially given the length of time for which all this has been happening, and the fact that, given the organisation’s history, the effect of Covid cannot be seen simply as a brand-new factor.

I think that you are saying that progress is slow because the situation is so complex. Are we asking NHS Tayside to do too much?

Complexity is a factor, but that is analogous to the situation of every health board in Scotland; they all deliver complex services. NHS Tayside has had particular challenges that it needs to tackle, but none of them are new; our report does not shed any light on brand-new factors for the board. The important aspect is that it now needs to put in place—with pace, as we say—a new plan that sets out how it is going to deliver high-quality services and financial balance.

The Auditor General referred to the historical and wide-ranging challenges that the board has inherited in its cost base. However, there have been a number of changes in the board’s senior leadership in recent years, which I feel have had a positive impact on its position. I am more confident that the board’s financial plans are now realistic and can be relied on. The board has delivered on its financial plans for the past two years, and it projects that the underlying financial position for the current financial year is to break even, without the use of financial flexibility funding. In addition, the level of recurring savings has increased.

Of course, the increasing challenges as a result of the Covid pandemic need to be factored in, but the board anticipates that the Covid costs will be fully funded. Although we have reported—and the board acknowledges—that transformation has been slow overall, during the past two years the chief executive and the director of finance have introduced a sharper focus on improvements that has genuinely improved the board’s financial position.

The specific performance measures that have the greatest impact on the board’s financial position are monitored and reported on. In 2019-20, the board delivered savings of more than £26 million, in excess of its target. It is right to highlight that the board has historically had a high cost base, and changes in that respect will take time. However, I see an improved financial position as evidence that the approach and better focus from the current senior management team is working.

The Auditor General also mentioned prescribing. This year, we have seen a reduction in prescribing costs in primary care. Although prescribing costs in secondary care grew, there have been efficiencies and a greater focus on managing those costs. A medicines management group was set up to manage growth in prescribing across acute services. I am now seeing evidence that the new focus from senior management is having an impact. Of course, Covid will have affected that progress, but I see the plan for remobilisation from Covid as an opportunity for the board to sharpen that focus further.

10:15  

Thank you, Fiona. That is good news on prescribing. That was the subject of one of my questions.

Graham Simpson has a supplementary question.

Auditor General, what is “expensive” about the operating model? You referred to that in your report. What, specifically, makes it more expensive than other boards?

I will start in broad terms and then ask Fiona Mitchell-Knight to offer some additional examples of the ways in which the board’s cost base differs from that of other boards in Scotland.

As we set out in the report, three main areas are outliers. That is historical. Much of the context lies in the board’s operating model, which has been set up over many years following the merger of NHS trusts and boards, and in the pace at which and the way in which it delivers services to the residents of Tayside. The three factors are its staffing model, its prescribing costs and its in-patient costs. Those have all been somewhat stuck when it comes to progress and to delivering that pace, and they differentiate it from other boards.

I ask Fiona Mitchell-Knight to set out a bit more detail on how that has happened and the board’s plans for reducing that difference.

As the Auditor General has said, workforce, prescribing and acute services are the three main areas in which we have in the past identified the board as an outlier. I talked briefly about prescribing, and the board has a very clear focus on that. I will touch briefly on workforce costs, and then I will ask Leigh Johnston to provide some information about comparison with other boards.

NHS Tayside, in common with other boards, is struggling with workforce costs. There is increased activity and flow through the services, which is a good thing, because it means that people are getting treated. However, the costs for that are expensive. There is a high number of vacancies, and the cost of temporary staff is of course greater than that of permanent members of staff. That area is a specific area of focus in the board’s remobilisation plans.

Will you be a little more specific on the staffing? Where are the vacancies? One of the questions that I wanted to ask was about psychiatrists. I think that, in November last year, NHS Tayside had only one third of the psychiatrists that it needed according to the most recent figures in your report. In talking about staff vacancies across the board, do you include doctors, nurses and support staff? Will you be more specific, please? Also, how does it compare to the rest of Scotland?

Convener, I do not have that information. Leigh Johnston might be able to help you.

I do not have the exact figures to hand, but we could get those for the committee.

With reference to the previous question, which was on costs, again I do not have to hand the exact figures for comparison with other boards across Scotland, but Audit Scotland has such figures and, again, I can provide the committee with them. The reason that we have commented particularly on prescriptions, staffing and in-patient costs is that our analysis shows that they are above the Scottish average.

I will bring Graham Simpson back in. We have known for a long time that in-patient costs are higher in Tayside. I think that that has something to do with estates and, in part, the rural element. I understand that, and I know that the board is working on that. I also think that residents in Tayside want to know about those staffing vacancies, because that is of real concern to me as a resident and as a citizen. Does Tayside have fewer doctors or nurses per head than the rest of Scotland? That is the kind of information that I would have expected from the report, so if you could get back to the committee with that very soon—perhaps within the week—I would be much obliged.

I agree with that, convener, because I am still not clear in my head about what makes the staffing model more expensive than it is in other boards. Maybe that breakdown will help the committee. I assume that, at some point, someone from the board will come before the committee. We need that breakdown, because we need to know where the costs are. It is not enough just to say, “It’s expensive; there are vacancies.” We need a bit more detail.

Leigh Johnston, can you clarify that you have that information but that you just do not have it in front of you?

Yes, we have that information, and, of course, we will report, as we always do in our “NHS in Scotland” report, which is due out in February, on vacancies across Scotland.

It is important to point out that NHS Tayside is not alone in having these vacancies. It is well known that health boards across Scotland are having trouble recruiting workforce in certain areas. I can send that information to the committee. For the different boards in Scotland, I can send the comparator with the costs as well as the vacancies.

Let me reiterate my concern about the mental health situation in Tayside. We know that there was a report by David Strang, which was welcomed, and I believe that the board is having weekly meetings with the Scottish Government. I am following that up, but I am really concerned about the level of staffing in that service. Auditor General, your report says that, as of November 2019, NHS Tayside needed 23.6 whole-time equivalent psychiatrists but was employing 7.85 full-time equivalent psychiatrists. That is a third of the number of psychiatrists that we need, and we know that, since the pandemic started, the need for mental health services has increased massively. Does Audit Scotland have more recent figures on the current staffing situation in mental health services?

Convener, you are right that these are stark numbers. As you outlined, the pandemic has increased the pressure on mental health services, which is noted in the report. The team and I can check with NHS Tayside whether there are more up-to-date numbers. On the statistics that we quote in paragraph 42 of the report about the number of whole-time equivalent psychiatrists relative to the required establishment for psychiatrists, those numbers were supplemented with locum psychiatrists. That is not ideal, and the report from Dr Strang noted the use of locum psychiatrists as a factor in the quality of care in mental health services. These factors were noted as part of the recommendations for NHS Tayside. We can check the numbers, and I am sure that the board can provide an update through some of its publicly available information.

I would be much obliged if you could come back to the committee within the week on those figures, on staffing, on where the vacancies are and on the situation in mental health services, including children’s mental health services.

Colin Beattie is next.

Thank you, convener. Auditor General, it has been mentioned already that we have had six years of section 22 reports, and I have sat here and read them all over those years. Every report says that there has been improvement, and your report is no different. If we have had six years of improvement, why are we seeing exactly the same problems as before? There is nothing new here. Nothing has moved on. Why not?

I am not sure that I agree entirely with your analysis, Mr Beattie. We have seen progress this year. It has been a volatile situation over the years, and the committee is well versed in some of the challenges that the board faces. For some of the reasons that are captured in the report, which Fiona Mitchell-Knight touched on, we have seen examples of progress. The financial position is one of those. There have been improvements in financial management and in delivering financial balance in the round. The board received financial support of £7 million from the Scottish Government during 2019-20, but that was within the typical framework that the Government established for the degree of financial flexibility. We have also seen the board take steps to improve the quality of its leadership and fill leadership vacancies, and we have seen examples of service performance improvements. In our analysis, all of that pointed to a board that was making improvements.

Part of the rationale for preparing a section 22 report is to support the committee’s scrutiny and also public understanding. I have set out the progress that the board has made. Given that, as you rightly said, section 22 reports have been issued in six consecutive years, it is an important step in such scrutiny to evidence the progress that has been made. However, we say that there is a continuing need for that progress to be sustained. Particularly now, with the pandemic and the remobilisation plan, it is really fundamental that the pace and quality of that plan are put in place.

You seem to be putting a little bit of a rosy turn on this one. Paragraph 42 of your report says:

“Staff had raised concerns about the safety of patients”.

That does not sound like an organisation that has embraced transformational change and is making a difference. As I have said, every single item that we have discussed at this meeting has been mentioned in every report over the past six years. Do we now have in place senior management and proper governance that can actually drive such a change through?

In a moment I will ask Fiona Mitchell-Knight to come in on the changes in leadership. I do not agree that I am giving an entirely positive picture. Fundamental challenges remain, which the board still has to tackle, some of which are common among health boards across the country. You mentioned mental health services. Dr Strang undoubtedly produced a serious report, which contained many recommendations and steps that the board now needs to take, including engagement with users of mental health services and their families. An update report on the progress that the board has made will come out next year.

I would not understate the challenges that the board needs to tackle regarding the impact that Covid will have on its plans and the quality of its services. A fundamental message in our report is the need for an effective, deliverable plan that provides high-quality services and financial balances.

I ask Fiona Mitchell-Knight to say a little more about the leadership developments that have taken place.

I come back to the same point: not one single item that was covered in the first report, six years ago, has been fully resolved. Even matters such as the high levels of non-recurring costs, which the board treats as savings, have been mentioned in every report. We have talked about the high levels of vacancies in senior posts. Previous reports discovered that the board was managing the financial gap by delaying appointing people, and evidence of that was found. I do not know whether there is any such evidence in the current report. We have seen continued problems with governance. Staff continually express their concerns about patient safety, which also go right back to the first report. That is not terribly encouraging.

Our analysis is that there are signs that the board is making progress on its financial position, its leadership changes and its service quality statistics. That is our overall assessment, but that does not detract from the fact that it still faces significant challenges, as all health boards doubtless do.

Turning to the specific points that you mentioned, such as the board’s treatment of recurring and non-recurring savings, it is true that the numbers in the financial plan referred to tens of millions of pounds that needed to be saved in the course of the next three years. It is an encouraging sign that the board delivered financial balance in 2019-20. Nonetheless, through its remobilisation plan, it needs to set out how it will deliver high-quality services and how it will do that—

It is not just about financial balance.

No.

There are more fundamental issues than that, although that is important.

I agree. It is about financial balance, service quality and addressing the mental health service challenges in NHS Tayside, along with ensuring stability through effective leadership and governance arrangements across the organisation.

The assurance and advisory group has completed its work. What support is the Scottish Government giving NHS Tayside, given the position that it is still in?

I will ask Leigh Johnston to say a little more about how the group’s work has progressed, along with the wider support that the Government is providing to the organisation.

The committee will remember that, a few years ago, NHS Tayside was a level 5 board in terms of the escalation and support that it was receiving, but it has since dropped down to level 4. It has received many millions of pounds of brokerage over the years, too. The financial support of £7 million that it received in 2019-20 was within the financial flexibility framework that was set out for boards each year.

I ask Leigh Johnston to update the committee on the specifics of the Scottish Government’s current role.

10:30  

I think that Fiona Mitchell-Knight wants to come in.

I will refer to some of the positive improvements that have been made by the board over the past couple of years that we have reported on. Over the past two years, we have concluded that the governance arrangements in relation to the board are effective, which was not the case historically. There have been definite improvements in financial planning, budgeting and the reporting of financial information, and there have been improvements to the committee structures, with the audit committee being extended to become an audit and risk committee. Those are some of the governance improvements that I can point to. There are on-going improvements still to be completed, but there has been definite progress in that area.

The Auditor General referred to the senior leadership team. Although we reported in our section 22 report that there are a small number of vacancies, some of which have subsequently been filled, there have been a number of important appointments to the senior leadership team as part of the restructuring of that team, and we are seeing positive impacts from that—particularly from the chief executive and the director of finance in delivering against the financial plans, which I referred to earlier.

In the past, we could not rely on those financial plans, because they were unrealistic and there was no evidence that the board delivered on them, but that is not the case now. For the past two years, the board has delivered on its plans—in fact, it has overperformed and given a slightly better position than anticipated. That looks to be the case this year, too. Despite the large number of savings that it has to achieve, the board is on target to overperform again.

On the support that the board is getting, when the assurance and advisory group finished its latest report, the board was being moved down to escalation level 4, so there was less formal support from the Scottish Government from that point, but the Government is providing specific assistance to the board for its mental health services improvements. In relation to the leadership of those mental health improvements, a particularly strong appointment has been the interim director of mental health, who has led the board in producing the action plan in response to the mental health report. The recent mental health strategy “Living Life Well: A lifelong approach to mental health in Tayside” has been published in draft and will be finalised in early 2021. The board recognised the importance of taking swift action to improve its mental health services, and, although Covid has clearly had an impact on how it is going about that, it has not lost that focus.

I hope that that is helpful.

You said that the appointment of the interim director of mental health has been a particularly strong appointment. How do you know that? Have you seen improvements in the service already since his appointment?

Yes. It is a lady—Kate Bell. It is about the pace of the response to the inquiry. The action plan has been put in place in response to the report and the strategy has been published. We have yet to see the impact of that on the services—they will not be fixed very quickly—but I was referring to the momentum and focus when I said that that has been a strong appointment.

Good morning. I thank the witnesses for participating this morning. I will go a bit deeper into the mental health issues, which have been touched on in a number of questions already. It is extremely worrying that, in the report, we have to read comments such as “a breakdown of trust”, “a lack of respect”,

“poor service ... patient care and outcomes”,

“long-standing” challenges and

“a culture of blame and bullying”.

How far along are we with the 51 recommendations? I hope that the answer is, “Quite far,” if there will be a report in the early half or the spring of next year. Are there any challenges that mean that any of the recommendations will not be fulfilled?

We agree that the findings of Dr Strang’s report are very serious. The quality of the mental health services that the residents of Tayside received, and the challenges, are set out in the report, which makes for very difficult reading.

We understand that Dr Strang will provide an update early next year on the progress that has been made. The most recent update that we have shows that 10 of the 51 recommendations have been identified as being completed and 38 are in progress, and that there has been some slippage against a very small number of them. There are signs of progress, but it is for Dr Strang to form a judgment, as he is expected to do early next year. We will, of course, keep track of that, and we will report progress in the round in our audit reporting in June 2021.

Can you give an indication of the recommendations on which there has been slippage and of what is being done to ensure that that does not happen?

I am not sure whether we have the specific detail on the numbers. I will check whether my colleagues Leigh Johnston and Fiona Mitchell-Knight know the specifics.

I do not have any further information.

I do not, either. The latest update to the board was in October. I can provide that to the committee, if that would be helpful.

NHS Tayside might be able to share the progress that it is making against the recommendations separately, if the committee requires an update in advance of Dr Strang’s report in a few weeks.

We know that the Covid-19 pandemic is causing nationwide issues but, as the convener said, it will hit mental health services pretty hard. What measures have been put in place to ensure that the service will be able to cope with the increased number of people who are being referred?

We will try our best to give that detail, if we have it, but I suspect that NHS Tayside would be better able to set out the steps that it has taken in response to the pandemic.

The only thing that I will add before I see whether my colleagues can supplement my answer is that, in dealing with the impact of the pandemic on mental health, NHS Tayside starts from a more difficult position than that of other boards. The convener set out the numbers in relation to the availability of psychiatry services, and there will be the wider impact on child and adolescent mental health services and so on. All of that emphasises the importance of implementing the 51 recommendations that are set out in Dr Strang’s report. At that point, we will have a better feel for how well placed the board is to support the community of Tayside.

Have the welcome improvements in financial management had an impact on the quality or level of services? In particular, have you looked at whether, during the time that those financial improvements were being made, there was any deterioration in the eight key performance indicators, such as waiting times, cancer treatment statistics and accident and emergency department turnaround times? I am delighted that the financial situation has improved, but I do not want that to be at the expense of service performance.

I will ask Fiona Mitchell-Knight to set out some of the detail of the KPIs. My initial response is that both financial performance and service performance matter. It is not acceptable to have improvements in financial management resulting in deterioration of service quality. We do not think that that is the case with NHS Tayside. We saw improving financial management and the delivery of financial balance, but also improvements in the service performance indicators. In particular, we have seen improvements in some of the A and E performance rates and some of the cancer treatment statistics, but I will ask Fiona to say more.

I do not have any detail to add other than the fact that overall performance against standards has generally improved since last year. However, of course, performance is mixed. Leigh Johnston might be able to give you more detail, but I can provide some reassurance in that the focus on the financial position is always balanced with the impact on clinical care. The two are not dealt with in isolation.

Leigh, can you provide more detail?

The board has made improvements compared to last year on many of the waiting time standards that Mr Neil spoke about. It is worth noting that, in 2018, the Scottish Government introduced the waiting times improvement plan. As part of that plan, rather than looking for all boards to meet the standard, the Government agreed performance goals or trajectories in those key waiting times targets. There were nine targets, and NHS Tayside met seven of the trajectories that it had been set. Obviously, the waiting times improvement plan has now been paused, and we do not expect it to continue. There will be a new way of trying to deal with the backlog in waiting times, which will be about clinical prioritisation. That goes back to the earlier question about mental health services. People with clinical priority will be given priority.

That sounds good. One of the points that was made in the introduction was that there has been a reduction in the prescription drugs bill in the primary health sector, although it has gone up in the secondary health sector. When the committee was in Tayside, taking evidence from the broad two years—[Inaudible.]—points it made about its relatively high prescription bill was that spending more on drugs kept people out of hospital and that, if it had not spent as much on drugs, the number of people in hospital would have gone up. Conversely, is there any evidence that the reduction in the primary care prescription drugs bill has led to any deterioration in the percentage of people being hospitalised?

We might need to come back to the committee on the specifics of that, although we heard Fiona Mitchell-Knight say that some of the challenges exist in the cost base in the acute setting, too. Although I accept NHS Tayside’s assertion that increased prescribing costs in primary care might result in a reduction in activity in the acute setting, I am not sure that that is entirely borne out in some of the high-level numbers in our annual report. I will check with my colleagues whether we have seen that relationship in detail, Mr Neil, because it might be a reasonable explanation.

No, I do not have any evidence on that.

10:45  

If we decrease the drugs bill in the primary sector and that leads to more hospitalisation, that could be a contributing factor to the increased drugs bill in the acute sector. I am basing that on the converse evidence that was given to us two years ago in Dundee. It would be useful to check that out.

The report comments that costs overran by £2.5 million. I have two questions about that. First, was that specifically social care costs, or was it the cost of running the health and social care partnership? The two things are not necessarily the same. Secondly, an increase in social care costs can sometimes lead to a decrease in health costs, because people are being treated at home, rather than being hospitalised. Was there any link between the increased cost of social care and hospitalisation rates, for example?

I will pass over to Fiona Mitchell-Knight again in a second. Fiona can draw on her experience as the auditor of Dundee City Council and of the Dundee integration joint board with regard to the relationship that you outline. The points that you make about investing in health and social care in the community setting and about the expectation of avoiding costs in an acute setting all seem reasonable.

I ask Fiona Mitchell-Knight to cover the specifics of the £2.5 million overspend.

I do not have any more detail on the overspend, other than the reporting that it relates to social care. I do not have any information on the interaction between those different costs. I am sorry, but we would need to come back to you on that question, Mr Neil.

If you could come back to us on both issues, that would be helpful. The whole point is about the interaction with prescriptions and social care. One of the key drivers is clearly to reduce the level of hospitalisation of patients, and social care and prescription drugs in the primary sector can make a significant contribution to achieving that. Any information that you have on that interaction would be extremely helpful.

I have two final questions, the first of which concerns estates. One reason why the cost base of NHS Tayside was higher relative to the cost bases of other territorial health boards was the legacy on the estates front. The board never really cleared that. A lot more buildings were being used to deliver services in Tayside, relatively speaking, compared with other health board areas. Some of those buildings were being vacated and marketed. Has there been any progress in streamlining the estates portfolio?

Asset management and the development of asset management strategies across the board is an area where we have on-going focus, and we monitor that through our annual audit report. A regional asset management plan was presented to the board in 2019, but we reported that it needed better alignment with the board’s transformation plans and a better link to the clinical strategy and financial plan. Although there has been some progress, there is still further progress to be made.

No doubt we will get an update on that in your next report.

Turning to the second of my final questions, I notice that the chief executive has a salary of £160,000, and that a contribution of £140,000 has been made to his pension fund, if I read the report correctly. What is the justification, in particular, for that contribution to the pension fund?

I will ask Fiona Mitchell-Knight to set out some detail in a wee second.

All those numbers are audited—both the remuneration that the senior executives receive in the year and the pension contributions. The specifics of the pension contributions can vary between individuals, depending on their age and length of service, and it is all subject to audit evidence over the course of the year.

I will ask Fiona if we have any detail on the size of the number and on why it differs from what you might have expected.

I do not have further information on that. I simply reiterate what the Auditor General said, in that the remuneration report figures are audited by us, and we have not commented on any such issues in our report for this year.

Even though the figures are audited, there might still be concern. For example, Ferguson’s shipyard has audited accounts, but no one would argue that it has a particularly good record, given where we have ended up on that, for whatever reason.

The fact that the figures were audited does not answer the question of what the justification might be for that level of public expenditure—in this case, on a pension contribution. I am happy for you to come back to us on that, but I think that we require justification for that level of spending. Someone on a salary of £160,000 has been given a pension contribution of £140,000. I am sure that that would enormously delight the low-paid people who work for NHS Tayside and who struggle to get by. I would appreciate it if you could provide justification for that.

The figures that appear in the remuneration report are in accordance with the pension scheme, so I suggest that the issue is with that, rather than with the individual entitlement of the chief executive.

I am not in any way casting aspersions on the chief executive, who, according to your report, appears to be doing a good job. What I am asking for, as a member of the committee, is an explanation for that huge pension contribution having been made. I will be happy to receive that in writing, but I think that it merits justification.

I will add to Fiona Mitchell-Knight’s comments. NHS senior managers are members of the NHS superannuation scheme, which sets out the employee and employer contributions for individual members. As Fiona said, we will come back to the committee with any additional information that we have. Factors that might have influenced the size of the numbers that have been disclosed include whether there have been transfers in from previous pension entitlements. Other factors such as age and length of service can lead to variations in pension entitlement.

We offer the committee an assurance that the numbers have been audited and will have been set out from information that NHS Tayside has received from the Scottish Public Pensions Agency. However, if we can provide the committee with any further information, we will be happy to do so.

To go back to the earlier question on the overspend and the relationship between primary care and acute services, NHS Tayside might also have further updated information on that.

I want to discuss certain financial matters that your report covers. A couple of them have already been raised by Alex Neil, and I will return to those in a minute. In the meantime, exhibit 2 on page 8 gives projected financial savings. Just for clarity, have those figures just been lifted from NHS Tayside’s information, or have you put them together or changed them in any way?

The figures in exhibit 2 are from NHS Tayside’s strategic financial plan, which covers from 2020-21 through to 2022-23. I will pause here to say that the scale of the savings, at £85.2 million, was identified before the Covid pandemic. The key point in the recommendation is that there is a real need for a revised plan, with an accompanying realistic and deliverable savings target, to be made at some pace.

Is the board doing that?

Yes. It is in the process of updating its remobilisation plan and is taking steps to do so. It is important that that is done correctly and with the right level of diligence, but with the pace that should be associated with it.

It would be useful for the committee to see that.

Going back to the way in which the figures were put together, the numbers all come down to a deficit of zero. That suggests that a balancing number exists that brings everything back to zero. Do you know which number that might be?

As you can see in the exhibit, there is some detail on the numbers that will be delivered. It is not uncommon for a savings plan to have that level of analysis in it. We do not recognise the balancing figure. One of the key roles that we carry out in our audit work is to ensure that there are robust savings. There is a role for health boards in relation to governance. They should be probing, analysing and ensuring that there are sufficient levels of detail in any savings plan.

As we have touched on, some savings plans will include specifics, which are typically associated with recurring savings, but some savings will be more led by activity that happens during the year. The committee has heard many times about the use of vacancy management arrangements to deliver non-recurring savings. There will be a balance between the two arrangements, as ever. There is a far stronger basis for recurring savings than there is for non-recurring savings.

It is a nice presentation, but we cannot see the detail from it. Is that what you are saying?

I am saying that NHS Tayside has the detail, and it is incumbent on it to ensure that it does not provide just a high-level analysis. Any credible savings plan has to be supported by detailed plans and steps. In many cases, it should include specifics relating to estates, prescribed savings, workforce and so on. Without that level of data, it will not be a credible and robust plan.

My next questions come from looking at the financial statements, which came to us very late yesterday, as you might know, because the annual report was not put on the NHS Tayside website for some reason. If I have missed something as I skimmed through the report, I am sure that you will forgive me. In relation to integration joint boards, which Alex Neil mentioned, your report says that there was an overspend of £2.05 million on social care and an overspend of £0.35 million on health. The board’s annual report says that it has a 50 per cent stakeholder interest, let us say, in the joint integration boards. Does anybody control the IJBs?

An IJB is a body corporate. As an individual public body, it is subject to audit and scrutiny, and it has its own board. There is a joint relationship—hence, the narrative refers to 50 per cent—between the NHS and the local authority, and the membership and make-up of the board is split between those two organisations.

Is there a deadlock arrangement if both organisations vote against each other?

I will pause for a moment and ask Fiona Mitchell-Knight to set out how the governance of decision making works.

If she could do so briefly, that would be helpful.

The three relevant integration joint boards—Angus, Dundee and Perth and Kinross—all have their own governance structures and board arrangements. Their boards are responsible for budgeting and for monitoring performance and financial activity.

So NHS Tayside has a 50 per cent interest in each of them. Where is the £2.05 million? If it has a 50 per cent interest, I presume that the more it charges to the joint board, the more it gets back—it gets half of it back. Is that how it works?

I will ask Fiona Mitchell-Knight to answer that. The arrangements can be quite complicated. There are various factors, including set-aside, which relates to the flow of funds between the board and the scrutiny, but I will—

Is the £2.05 million net, or is it the real cost to NHS Tayside?

Would you mind repeating that?

Is that a net cost? Does NHS Tayside charge that to the IJB and then get a percentage charged back to it? I presume that it would be 50 per cent.

I will ask Fiona Mitchell-Knight to give an explanation. NHS Tayside and the three local authorities for the three different IJBs are all financial contributors to the IJBs. The services then return. On the specifics of how the financial position is arrived at at the year end, I will ask—

Is there a settling of balances at some point between NHS Tayside and the IJBs?

11:00  

I am not sure that we would say that it is a settling of balances. The framework agreement between the organisations sets out clearly information about their respective contributions to the IJB, what services will be delivered thereafter—

For the sake of colleagues, I do not want to get into too much detail about the accounting, but maybe you can come back to us about whether that is a real number, and a cost to—

I think that Bill Bowman is touching on one of the key issues about accountability of those joint boards, all over Scotland: how the funding arrangements are arrived at and how we can scrutinise them. Certainly, that was in the last question for you on my list today, Auditor General. Tayside is a good example. There are three different boards and, I think, three different funding arrangements. Maybe in answering Bill Bowman’s question you will also tell me how we get accountability, given that the contributors are split over your jurisdiction and that of the Accounts Commission. How do we get accountability for expenditure in the IJBs?

You are correct, convener, that there is a range of accountabilities between me as the auditor of NHS and central Government bodies, and the Accounts Commission, which oversees local government.

IJBs are local government bodies, and their activity is therefore subject to scrutiny by the Accounts Commission. Those arrangements are working well. When issues have arisen in individual IJBs, the controller of audit has prepared statutory reporting on those organisations. There is a mechanism in place for that. Through my work, as I think the committee will be familiar, we have prepared a number of reports about how health and social care integration is working in Scotland, and it is still part of our planning. As you know, we are thinking about our forward programme, and we can consider carefully how best to set that out in more detail, in order to support the committee’s scrutiny of IJB arrangements.

There are three joint boards in the NHS Tayside area. Is that unusual? I live in Lanarkshire, in which there are two councils, one health board and one IJB rather than two. Why are three IJBs needed in Tayside? It is not for you to justify, but is it a normal arrangement?

Yes, it is. It reflects the fact that three local authorities operate within the NHS Tayside boundary. In Lanarkshire, there are two local authorities. There are 31 IJBs in Scotland. As the committee has previously heard, one of those—NHS Highland, alongside Highland Council—operates a lead agency model.

Bill Bowman, I interrupted you.

Thank you, convener. I was just going to ask Mr Boyle to explain—perhaps not now—where that £2.05 million hits the accounts of the various bodies, and how much of it flows round and comes back.

I am happy to ask Fiona Mitchell-Knight to explain that in a bit of detail, if that would be helpful just now.

I have some other questions to ask. Is that okay, convener?

Yes, if we can be tight.

I ask Mr Boyle to come back to us on that later.

I return to the accounts. On page 62 of my copy, there is an item headed “Fair Pay Disclosure”, which compares the pay of the most highly paid director to the median—or average—remuneration of the employees. The total remuneration for that director is given as £190,000, whereas, earlier, the accounts state that the total remuneration for Mr Archibald is £300,000, which would give a much higher multiple. Has there been some disconnect between the two tables?

The short answer is no, I do not think that there has been a disconnect. The NHS accounting manual will have set out the way of disclosing fair pay in the accounts—whether it is gross salary and excludes or includes pension contributions. My understanding is that it excludes pension contributions. That will be the same for the most and the least highly paid employees, and will arrive at the ratio that is reported.

Is it just sloppy wording, then, to talk about “total remuneration” in both places?

The wording that will be used will be set out in the NHS accounting manual, so when I say—

But we are not looking at that. We are looking at the financial statements, or at the notes in them.

I am sorry. I should say that the NHS accounting manual sets out prescriptively the format and content of NHS accounts in Scotland. If it wishes, the Scottish Government, which sets the manual, has the option to broaden out the definition to include pension contributions.

I am just pointing out that the words “total remuneration” appear in two places and seem to mean different things.

That is helpful feedback; we can highlight that to the Scottish Government to see whether there is anything that it wishes to improve and provide additional clarity on.

Auditor General, I thank you and your colleagues from Audit Scotland, Fiona Mitchell-Knight and Leigh Johnston, very much for your evidence on NHS Tayside this morning.

---

“The 2019/20 audit of Social Security Scotland”

Item 4 is an evidence session on the section 22 report, “The 2019/20 audit of Social Security Scotland”. I welcome our witnesses from Audit Scotland: Stephen Boyle, the Auditor General for Scotland; Mark Taylor, audit director; and Carole Grant, senior audit manager, audit services. Mark Taylor and Carole Grant are appearing remotely.

I understand that the Auditor General will make an opening statement on the report.

The report updates the committee on the findings from the 2019-20 audit of Social Security Scotland. It brings to your attention the appointed auditor’s qualified opinion on the regularity of the 2019-20 accounts. It also provides an update on the progress that Social Security Scotland has made on establishing its error and fraud management arrangements. The appointed auditor has given an unqualified opinion on the information that is reported in the financial statements, which means that he is content that they show a true and fair view and follow all relevant accounting standards and rules.

The qualification relates to the estimated levels of fraud and error of £14.8 million in carers allowance expenditure, which is 5.2 per cent of the total benefit. That expenditure is not in line with the relevant legislation.

Carers allowance is being delivered by the Department for Work and Pensions on the Scottish Government’s behalf, through an agency agreement. Social Security Scotland is reliant on the DWP’s estimates of error and fraud. It is important for Social Security Scotland to understand the levels of error and fraud in all the benefits that it administers. With a range of benefit delivery arrangements in place, the impact of error and fraud will remain an audit risk.

Social Security Scotland has strengthened its arrangements for the prevention and detection of error and fraud over the past year and has increased staffing in line with workforce plans. Further development is needed in some key areas, including on-going development of digital systems, guidance on determining residency and establishing the agency’s own arrangements for estimating and reporting on error and fraud within the Scottish system.

My report also looks ahead to issues and challenges for Social Security Scotland due to the Covid-19 pandemic. The agency is now responsible for a range of complex and high-value benefits, and the continuing impact of the pandemic increases uncertainty and financial risk in an already challenging environment. The implications of changes to delivery timelines, arrangements for benefit delivery and benefit eligibility and uptake rates will be complex to manage. Social Security Scotland has adapted quickly to the immediate operational challenges of the pandemic, allowing it to maintain the administration of benefits while addressing the safe working of its staff. We will continue to assess the detail of those arrangements through our audit work.

Mark Taylor, Carole Grant and I are happy to answer questions.

I would like to get more clarity around the error and fraud team. The section 22 report is about the levels of error and fraud in the system. Are we more or less comparable with the rest of the United Kingdom in terms of benefit fraud levels?

The picture is undoubtedly complex. I will ask Carole Grant to give more detail about the progress that the agency has made in increasing the number of people whom it employs and how that is working in relation to error and fraud work.

In a UK context, error and fraud have always been a feature for organisations that make benefit payments. For example, the committee will be familiar with the fact that the National Audit Office has qualified the DWP’s accounts for 31 successive years due to levels of error and fraud in a range of its benefits.

Our assessment is that Social Security Scotland has made progress in relation to upskilling, recruitment and error and fraud activity, but that risks remain, particularly as—because of the pandemic—there is some interruption to the pace at which it becomes responsible for some benefits. However, error and fraud is a key factor that it needs to challenge. It would be useful for Carole Grant give a bit more detail about how that is working.

Before she does so, I am unclear from the report whether we have a firm figure for the level of fraud. Paragraph 25 of the report states that:

“Targeted sampling indicated that around 75 per cent of errors had already been identified and corrected.”

Paragraph 20 states that

“DWP and HMRC estimates of error and fraud ... range from 3.9 per cent to 9.4 per cent”,

which is a heck of a margin, given the figures that are being handled. Do we know what the level of fraud in Scotland is?

No, we do not, although we would all like that level of precision. There will always be estimates of the level of fraud in relation to any benefit, and it is undoubtedly a complex area. However, we have better estimates this year, particularly in relation to carers allowance. As Mr Beattie and the committee may recall from our discussion last year, the estimated levels of error and fraud in carers allowance payments dated to back to activity in the mid-1990s. We have seen progress on that front this year, which allows us to be more certain about the levels of fraud in the Scottish system, and which has a bearing on the qualification that we put on the accounts.

Although we say that progress has been made, we have tried to quantify elsewhere in the report that, as the Scottish social security agency assumes more responsibilities during 2021, the levels of error and fraud are still estimates. That is important in relation to the work that Social Security Scotland has done with its own team to provide estimates that are reliable for the agency and around which it can target its activity to tackle levels of error and fraud.

I am just trying to get my head around that. Targeted sampling indicates that 75 per cent of errors are being identified, and the DWP and HMRC are looking at error and fraud rates that range between 3.9 and 9.4 per cent. I presume that that estimate comes after the sampling that indicates that 75 per cent of errors are being picked up. That would indicate that about 25 per cent of the errors and fraud that are taking place are rolled up somewhere in that very wide range of 3.9 to 9.4 per cent.

The 3.9 to 9.4 per cent range covers the estimates against the individual benefits that are set out in paragraph 20; the estimates therefore vary depending on the benefit. I think that it would be helpful to pause for a second and allow Carole Grant to set out in a bit more detail how the intervention team is working and how that ultimately translates into the reliability of the estimates and what that means for—[Interruption.]

I would like at some point to better understand the wide range of figures and its implications for the sampling.

I am happy to come in on that point. As the Auditor General said, the arrangements for dealing with fraud and error in Social Security Scotland have been strengthened. Specifically in relation to the 75 per cent figure that has been quoted, the interventions team that has been established did a specific data mining exercise within which it found an error rate of 43 per cent; however, of the errors, 75 per cent had already been identified and were being addressed. That shows the difference between detailed data mining and the error detection rules and processes that are already in place.

11:15  

We are seeing—and we need to see this—a move away from detection and towards prevention, in order to ensure that the amounts are paid correctly in the first place. Social Security Scotland is working on developing that approach further.

For clarification, your starting point is that there is a 43 per cent possibility of error and fraud, which is eventually whittled down to between 3.9 per cent and 9.4 per cent. That seems extraordinary. Is that what you are saying?

No—you are picking up figures from different parts of the report that do not necessarily come together in that way. The 43 per cent that I mentioned related to a specific exercise in Social Security Scotland; it refers to the real errors that were identified.

The rates in the other paragraph of the report that has been mentioned are DWP estimates for the overall benefits system. We are talking about some specifics that sit alongside estimates for what happens elsewhere in the system.

Putting the other figures aside, you said that 43 per cent of benefit payments—the figure is presumably based on some sort of sampling—were found to involve potential error or fraud, and that proactive action had to be taken to bring that number down. That indicates a colossal level of error and fraud. Carrying out an on-going exercise to bring the figure down to what we would hope would be an acceptable level would be a significant task.

The starting point is that, at this stage, Social Security Scotland does not measure its overall error rates. We highlight in the report the need for it to build up the capacity to do that and start publishing the information.

To go back to your initial question, we do not know from our overall assessment whether error and fraud rates in the benefits that Social Security Scotland administers are in line with, or vastly different from, the DWP’s experience.

For the benefits that Social Security Scotland administers for the DWP, it relies on the United Kingdom-wide figures that the DWP prepares. That is the basis for our qualification in the report. Importantly, Social Security Scotland does not yet do the necessary work to undertake its own estimates. It is building up the capacity to do that, in the way that we have set out in the report.

Separately, Social Security Scotland does some targeted work to look at high-risk areas. It goes in to see whether the front-line processing has picked up what needs to happen in those areas, and, where there has been error or suspected fraud, whether that has been dealt with at the front line. If it has not, the agency corrects at that point. That is targeted, risk-based work, but we are unable to extrapolate the results to cover the overall population because it is focused only on high-risk areas.

From our perspective, the key point is that, at this stage, Social Security Scotland cannot tell the committee or Audit Scotland what its overall error rates are for the benefits that it administers directly. We make it clear in the report that it is important that Social Security Scotland begins to build up the capacity to do that.

What about the 43 per cent figure?

When the agency has done its targeted work, it undertakes the corrections for the instances that it has come across. From those areas, it will learn what is happening around the processes so that it can correct and strengthen them. The figure of 43 per cent, with 25 per cent of errors not identified, is a small amount in the overall sense and helps the agency to understand—

You have now brought in a figure of 25 per cent.

I was referring to the fact that Social Security Scotland found 75 per cent of the errors at the front line, on the first go, which leaves 25 per cent—

So there is the 75 per cent within the 43 per cent, and the 25 per cent—

I am sorry—I am not following where the 43 per cent comes in.

It was mentioned just now.

If I may interrupt, there seems to be quite a lot of confusion here about the figures, so I wonder whether Carole Grant could write to the committee and explain the matter a bit better. Might that be helpful?

Yes—I am happy to do that.

Yes—we can do that, convener. I guess the point is that there is still a wide range of estimates within the level of benefits within Social Security Scotland and those that are provided on its behalf by the DWP. Social Security Scotland is making progress, but it is early days, in terms of its own capability, for it to undertake the level of fraud and error investigation that it will ultimately need to, once it assumes responsibility for all the benefits in its gift.

We are happy to come back to the committee on the specifics of how the methodology is developing.

Thank you, Auditor General.

Your report indicated that the error and fraud team was being expanded. You said that it would be up to 47 full-time equivalents by March 2020, and eventually 190 FTE. Did the agency achieve 47 FTE by March?

We report that the team increased from 17 to more than 40 full-time equivalents by the end of March, with plans for further expansion. However, those plans will depend on how the pandemic has impacted on the agency’s ability to recruit over the period.

That work is a key part of the agency’s business, so it needs to, and has plans to, increase its capability. It is taking steps, although it has much to do to get to a place where it has all the ability and authority itself to undertake the level of error and fraud prevention work that it needs to do.

Are you satisfied that the training and policies that are being put into place are adequate and appropriate for what is required, and that the staff are properly supported and equipped for the job?

I have a two-part answer to that question. The agency has made progress. At paragraph 22 of our report, we mention the strategies that it has implemented: the error control strategy and the counter-fraud strategy, as well as new arrangements for investigations, all of which are welcome. The agency had planned to engage with the UK Cabinet Office, drawing on its expertise around error and fraud prevention during the year, but those plans were interrupted as a result of the pandemic. Such skills transfer and knowledge transfer are key if Social Security Scotland is to build up its own capacity around error and fraud prevention during 2021.

Good morning, panel, and thank you for joining us.

The report says that there are “ongoing weaknesses” within the agency’s core case management system. Could you tell us what those weaknesses are?

I ask Carole Grant to set out the progress that the agency has made on its digital infrastructure.

In our report, we pick up on the fact that there are manual controls within the system, which form part of the process of a minimum viable product in system development. I can confirm that audit testing during the year confirmed that the system controls were operating effectively. Manual controls are more reliant on individuals, and there is more risk of them not being consistently applied. That is why we would look for system controls to be in place. However, we identified the control environment to be operating effectively.

Are you expecting the system controls all to go digital at a certain point? Do we have a timescale for that?

We do not have a specific timescale. The system is being continually improved as part of the agile approach to system development and as new benefits come in. There is a continual learning cycle within that.

We would like to see some of the manual controls replaced with system controls. However, we are content with the overall control environment.

Does the digital system have any cost overruns? I am thinking back to other public sector information technology systems that did not go as well as planned.

I do not have any information on that. Does Mark Taylor have anything to add from the programme side?

The committee will be aware that we have reported on the progress of the overall programme a number of times. When we last looked in depth at that, the overall cost levels were not a challenge, but we said that it was important that the programme was able to set out its longer-term plans—there has been significant progress on that—and translate that work into cost plans, on which there has also been progress. As the Auditor General mentioned, we are considering our future work programme and we expect that continuing work on social security will be part of that. We will continue to update the committee as we do that work.

There seems to be good progress being made. Do you foresee any challenges or issues coming down the line in relation to the digital system?

It is still a big job; clarity about the order in which things are done and the arrangements that are in place is fundamentally important in what is a very complex set of relationships, including the relationships with the DWP and a range of suppliers. There is much work still to do on all that, but we are not aware of any red flags at the moment. We will continue to do the detailed work that we have planned on that issue and will report to the committee in due course.

I will briefly come in on that issue. At last week’s meeting, the committee took evidence from the Scottish Government on its major IT infrastructure projects, in which the social security digital programme features prominently. Social Security Scotland is included in the infrastructure that the Scottish Government is deploying around its assurance activity. There is a clear focus on the need for that to go well, but assurance activity is built into that process.

I want to be clear that, in relation to error and fraud in carers allowance, Social Security Scotland is reliant on the estimates from the DWP—is that still the case?

Yes, that is still the case. Those estimates were updated during 2019-20. I mentioned the reliability of the estimates that had been in place, which dated back to the mid-1990s. Those were updated during the past financial year, which allowed it to be a more accurate assessment of the level of error and fraud.

Earlier, Mark Taylor said, I think, that Social Security Scotland is not yet in a position to publish its own error rate. Am I right in thinking that that is down to a lack of staff?

That is a planned project and a planned growth that assumes responsibility for the range of benefits that are coming to Scotland. Part of that has been interrupted by the pandemic. We set out at exhibit 1 in the report—it may be helpful for the committee to look at the exhibit—the waves of benefits that have come to Scotland, what the anticipated timeline was before the pandemic and what it is now. Alongside your question about Social Security Scotland’s capacity to undertake error and fraud evaluation work, the pace of that work, as it has received competence to deliver the benefits, has also been slightly interrupted. Those things will happen in tandem. The agency has made progress and it has increased the number of people in its error and fraud team. The training and support that it had planned did not take place and will now happen during 2021. As the agency’s capacity grows, it will assume responsibility from DWP for that work.

If it has all those new people, what are they doing if they are not producing an error rate? What are they working on? Are they constantly doing training?

11:30  

It is more than that. We set out in paragraph 23 the main responsibilities that the team is undertaking. Investigations, training and recovery activity are all factors in the team’s work. As ever, if the committee wants more specifics, Social Security Scotland is well placed to provide an update on how the division of responsibilities is operating.

To be fair, that question is not for you; it is for Social Security Scotland.

At paragraph 11, the report says that there has been an estimated £14.8 million of overpayments in relation to carers allowance. How was that figure arrived at?

I am happy to start, but I will ask Mark Taylor to say how we arrived at that figure. That builds on your previous question about use of the DWP’s estimates of the level of fraud and error related to carers allowance. The report talks about the split that is assumed between error and fraud rates. I will ask Mark Taylor to say a bit more.

The DWP’s fundamental process of measuring and estimating its error rates is based on a review of a statistically valid sample of cases. It investigates the circumstances of the individual cases for each benefit, including carers allowance, and then tries to confirm retrospectively the evidence base for the award of the benefit. On many occasions, that includes talking to the individuals concerned and obtaining proof of employment, income levels and other factors. There is quite a detailed review of that sample, from which the figures for the wider population at UK level are extrapolated.

As the Auditor General said, that had not been done for carers allowance for a long time; it has been updated only in the past year. We have relied on that estimate to provide an appropriate view on the level of error in Scotland. The figure of £14.8 million is about 5.2 per cent of the total spend on carers allowance this year. It is an estimate based on a UK-wide figure. We think that there is a clear enough case for that to be brought to the committee’s attention, because there is a material and significant level of non-compliance with the underlying benefit legislation. That is what we have reported in the audit and certificate.

I will ask about the accounting. If you paid out £100 and estimated that £5 of that related to fraud or error, would you set up a recoverable amount of £5 and reduce that as you collected the money back or decided that it was irrecoverable. Is that what happens?

Effectively. Mark Taylor will say more about that.

In a very broad sense, you recognise the expenditure of £100 until you identify the specific error or fraud of £5. You then identify that as a debt and collect it. Of course, not every individual case is identified so, alongside that, there is a set of provisions and estimates for the overall effect of errors. All that plays through into the aggregate spending figure. We have looked at the details and are happy with the overall accounting approach that sits behind the figures.

Could you clarify that? If you pay out £100 and then estimate that £5 should not have been paid out, do you set up an amount recoverable for that?

To clarify, you account for the £100, because that is what you have paid out. That is the very short and simple version of what happens. Separately, when you identify specific overpayments, you set up a debtor for that overpayment and, in effect, reduce the £100 for the specific overpayments, but not for the overall—

Why would you not rely on your estimates of fraud and error rates and do that when you pay out, rather than deal with it on a cash basis when you get the money back later on?

It is largely because you do not expect to recover all of that.

In which case, you would have a different—

You can only recover the specific debt that you have identified for individuals who have been overpaid, .

It seems that there is something missing. If you know that there is an error, based on your best estimates, the person looking at the financial statements would be interested to know how much of that you think that you will get back. You would judge that some of that will never be recovered, so you would reduce the amount. However, it seems that what is happening is an easy way of kicking the issue down the road until you find something and then taking the cash you get when you get it.

In essence, what you described is what happens. However, it happens in a very detailed way; I am happy to go into that.

The key issue that has been recognised is the amount of public expenditure cash that has been paid out—irrespective of whether that was paid out as a result of error and fraud. That figure is recognised as the cost to the public purse. When you recover that, you would recognise the recovery.

That means you are losing control over the management, are you not? Social Security Scotland does not have to account for what it thinks the error rate is; it only lets the amount go down until it is collected—at which point, everyone is happy because some money is coming in.

That is absolutely the case, and it is why we are strongly of the view that it is important to measure and report on overall error levels. Our qualifying the accounts is one measure of that, but the most important one is the agency’s own assessment of error rates and what it is doing to manage them, and similarly what the DWP’s assessment is and what approach it is taking. We look to continue with that message.

We have just discussed the impact of Covid on eligibility for and uptake of benefits. The number of households in Scotland that receive universal credit has increased from just under 250,000 to just under 500,000.

It is important to understand the impact of increased eligibility and uptake on Scottish benefits. That remains uncertain, but it would be helpful to understand the problems more fully. Could you talk us through, in detail, the difficulties in assessing the impact of that and also indicate when we might get a clearer picture of that impact?

I will start, and in a moment I will ask Mark Taylor to come in to set out in a bit more detail the pandemic’s possible implications for the Scottish budget.

We agree that many of the benefits of universal credit are, effectively, passported benefits that flow through to eligibility for some of the benefits that will be captured in Social Security Scotland’s accounts. In the report, we touched on the scale of the numbers that the relative increase in universal credit claimants, caused by the pandemic, has had.

We flowed through what that might mean for the Scottish child payment, as an example. The Scottish Fiscal Commission’s estimates say that in 2021-22 alone there will be an increase in child payment amounts from £65 million to £77 million. The forecast that the Fiscal Commission made and what that means to the impact on the Scottish budget will be captured in the budget when it comes to Parliament in January.

It is undoubtedly true that the situation introduces more volatility into the Scottish budget and is important with regard to the distinction between the Scottish-only benefit arrangements and those that are UK-wide and administered on the behalf of the DWP. Social Security Scotland will bear the risk of the Scottish-only benefits through the implications on its budget.

In answer to the question on when we will know more and what we already know, we start with the expectation—as you do, Mr Bibby—that the increase in universal credit, which has been hugely significant in volume, will ultimate play through into passported benefits. It is fair to say that, based on our information from Social Security Scotland—it will be able to provide more up-to-date information on this—we are not seeing that significantly being the case so far. Of course we would expect a degree of lag where lots of people who perhaps are unfamiliar with the benefits system are applying to it and taking advantage of the support that is available for the first time. It will take a little time to work through that and understand what else might be linked to it. Our information is that that is not pulling through in high volumes yet.

As the Auditor General says, the Scottish Fiscal Commission has as a key part of its remit the responsibility for forecasting likely benefits expenditure. It has been unable to update its forecast as a result of the pandemic, but we would expect that, when the Scottish budget is published in the new year, the Fiscal Commission will provide the updated forecast of both what the impact has been in the current financial year and what the likely impact is in the future. At that point, we will get more information and more of a sense of the overall impact of the pandemic.

Lastly, as the Auditor General has said, the extent to which that hits the Scottish budget depends on whether there is an equivalent UK benefit. For the Scottish-only benefits, all the costs fall within the Scottish budget, so, if demand goes up, there is more cost to be borne in the Scottish budget. For the benefits that have a UK equivalent, what matters is the relative uptake—the relative cost per head at an England and Wales level and at a Scottish level. That is much harder to predict, so, where you expect benefit costs to go up, the funding that is available through the block grant adjustment might well go up to the same extent, or indeed it might go up more or less than that. The Fiscal Commission forecast will give more insight into that.

That is helpful. Thanks for that answer.

I have a quick question. When I was social justice secretary, between 2014 and 2016—perhaps I should declare this as an interest—I had responsibility for initiating the discussions on the process that would be used for transferring benefits. The single biggest issue was the computer systems that were used by the DWP. They were like a penny farthing bike, when a Lamborghini was needed.

Auditor General, where do we stand in terms of Social Security Scotland’s future IT capacity and the DWP’s capacity? Both are needed to make the proper transfer of responsibility. If we get additional benefits, could they be much more easily and timeously transferred now than then, without IT being a barrier to doing so?

I will start and then ask Carole Grant and Mark Taylor to comment on how that is progressing in the agency.

The committee will know that we have touched on the importance of effective IT arrangements—across the piece, but undoubtedly for this project. One finding in our report from 2019 that provided an update on progress was on IT systems and the sense that the arrangements between the Scottish Government project team and the agency were not as cohesive as they needed to be. We have undoubtedly seen progress on that front during the year. Importantly, users in Social Security Scotland are shaping and are more involved in the process of how IT will work within the agency. That is welcome, and more on that is captured in the report.

Carole Grant can update the committee on progress relating to the process of benefit transfer and how those connections between DWP and Social Security Scotland are working.

We are seeing a development in the relationship. As part of the performance audit work, which was paused as a result of the Covid-19 pandemic, we did some direct work with the Department for Work and Pensions as well as Social Security Scotland. We heard from both sides that the relationship was in a good place and that both were actively involved in discussions about how they can move forward and what will make the case transfer work. Some of that was covered in what the Auditor General said about improvements in the relationship between the programme and Social Security Scotland. We have also seen improvements in the other, wider relationships that need to be in place as we move down this road.

11:45  

I will add to that briefly, to recognise the point that Mr Neil made. The interplay between DWP systems and Scottish systems is absolutely still part of the critical decision-making process. We are aware of, and have reported on, the reliance that the agency and the Scottish Government continues to place on DWP systems.

We are also aware that, for example, when the cabinet secretary, Ms Somerville, made an announcement about disability benefits in November, a key judgment that she would have made, and a key part of the programme’s work, would have been to agree with the DWP the path to readiness on that. The committee will see from her announcement some areas in which she added caveats, because that process has not yet been fully nailed down. In other areas, though, it has been clear that both parties are ready to go, which is an important part of the process.

I have a couple of questions on that area. Social Security Scotland is a new agency so, in addition to the work that it does, which we have been discussing today, it is of interest in itself. It is of particular interest to me because it is based in Dundee. How many people are currently employed by the agency?

I will definitely have to pass the specifics of that over to my colleagues. You will have seen that the agency has been visible about setting up its headquarters in Dundee, with its new waterfront building. I will check with my colleagues whether we know how many people it currently employs.

Yes, I will come in on that one. To quote the numbers from the audited accounts at the end of March, 666 individuals and a full-time equivalent staff of 647 people were employed at that point. There were also around 75 full-time equivalent interim workers.

We know that since then, after pausing it for a short period in March as a result of Covid, the agency has been able to resume its recruitment process. We also know that, during that period, around 400 staff have been recruited in addition to the 666 since that period. Therefore, despite the pandemic, there has been a real gearing-up. The convener will know that there is still a long way to go on employment overall, but the agency has continued to build up its capacity. That is largely in anticipation of the further benefits that it will administer. There is a timetable for the disability benefits that have already been planned. Most pressingly, applications for the Scottish child payment are already open and payments will begin in the new year.

You are telling me that more than a thousand people are employed by Social Security Scotland. Is that correct?

Yes.

Can you tell me how many are employed in the Dundee area?

I am afraid that I do not have that information, although perhaps I should have had it with me today, convener. However, we can certainly investigate it with the agency and confirm it to you.

Okay. I would really appreciate it if you could come back to me on that.

I am also interested in where the senior jobs are located. One concern is that we were promised such jobs in the local area, but some of us understand that many of them are located in Glasgow. I would quite like to have a breakdown of where they are located. Obviously, that becomes more complex in light of the Covid pandemic, because people are now working from home. One of the reasons for my question is to get an idea of how that affects our local economy in Dundee. I would like to know how many of the jobs that we were promised are located locally and how many have been spread across the country. If you could come back to me in writing on that, I would be very grateful.

We will investigate that with the agency and come back to you.

You are right, convener. I was just going to say that the situation is currently much more fluid and so it is much harder to say that with any certainty, in light of the pandemic and the working-from-home arrangements that have been in place. We have seen that organisations—Social Security Scotland included—have been able to function through the use of such arrangements. However, there is less certainty in their being able to predict where people’s bases might be, and that information might not then translate into what their being in a city could mean for the local economy. It is probably a matter on which Social Security Scotland rather than Audit Scotland should update the committee, because it would be able to do so with a bit more certainty.

Perhaps Audit Scotland could give me the information that it has. I understand that the figure of 1,000 jobs represents the total across the country, but I would be interested to know how many of those are held by people who have addresses in the Tayside area.

There is also a huge question about site 6, which was the flagship building for the country. As you said, Auditor General, it is directly across from the Victoria and Albert museum and was extremely expensive for the local authority. The unforeseen circumstances of the Covid pandemic have meant that that building is probably not being used much at the moment, but I wonder what the agency’s plans are for making use of it in the future, when things return to normal. However, perhaps we could pick up all those questions with the agency.

No member has indicated that they wish to raise any further points on the section 22 report with the Auditor General and his team. I thank all our witnesses for their evidence. I now close the public session of our meeting.

11:51 Meeting continued in private until 12:08.