Official Report (Hansard)

Session: 2013/2014

Date: 22 January 2014

PDF version of this report (351.27 kb)

Public Accounts Committee


Inquiry into Account NI: Review of a Public Sector Financial Shared Service Centre (Department of Finance and Personnel)


The Chairperson: Mr Stephen Peover, you are very welcome to our meeting today.  When you are ready, will you introduce your team?


Mr Stephen Peover (Department of Finance and Personnel): OK.  To my right is Paul Wickens, the chief executive of Enterprise Shared Services, and to my left John Crosby, who heads up the Account NI element.


The Chairperson: Thank you for taking the time to come here today.  It is much appreciated.  You have been here for some of our other witness sessions.  It is encouraging that the implementation phase of the project was delivered only eight months later than planned.  That was in direct contrast to what happened in the earlier project stages, when there were very lengthy delays.  What improvements did you make to the management of the project to achieve that?


Mr Peover: I am not sure where to start with that.  We view this as not being a single project.  There was a stage where the accounting service view was undertaken to look at the best way forward for Northern Ireland Departments, particularly in the light of devolution, and how we would best secure services to the 11 Departments, as it was then, and make sure that the accounting platform would work in the future.  It was at the point of creaking and collapsing.


There was a review, which went on for a couple of years.  There was then an outline business case.  The real starting point for the project was the full business case in 2006.  That is when we regard the project as starting.  The decision to go with the solution that we now have was taken in 2006 and delivered in 2009.  That is the success story; it is, by and large, a good story.  There are things that we would improve; we will come back to that in the Committee discussion.  Once the decision was made, it proceeded well.  The Audit Office acknowledged in the report that there was a delay, but it was not a significant delay given the size of the project — £170 million, £180 million or £200 million, depending on what way you want to count the numbers.  We can come back to that, too.  It was a well-delivered, well-managed project.  It has delivered the results we wanted it to deliver.


The Chairperson: At this stage, I commend you for the prompt payment and timeliness of it.  The report is not all bad.


Mr Peover: No, it is not.  Thank you.


The Chairperson: That has to be acknowledged.


Mr Peover: I do not often hear that in Public Accounts Committee meetings.  I usually hear Mr Dallat telling me that this is the worst report he has ever read.  Thank you for that. [Laughter.]


Mr Clarke: Do not get carried away.


Mr Peover: Is that coming?


The Chairperson: It is early stages yet.  I suppose I want to soften the blow.


For the benefit of the public, who might not be aware of the finer details of Account NI, can you give a background as to why the decision was taken to create the Account NI shared service centre?


Mr Peover: There were two pieces to the rationale, one of which I have just described to you:  we had a new situation where we were having 11 Departments and we needed to ensure the provision of accounting services to all those Departments. We were also at a stage where our systems were creaking and getting to the point of collapse.


Secondly, as you know, we had a wider Civil Service reform programme, which this became wrapped up in.  We have HR Connect, the IT Assist system and a whole range of areas where we are modernising and revising our back-office systems to try to give us the capacity to deliver a modern service.  So, it became part of a wider reform stream.  It needed to be done anyway, but it became part of a broader programme.


It has largely achieved its objectives.  It has replaced and modernised our financial processes and provided a common accounting system across the Northern Ireland Civil Service and beyond.  We now have 30 bodies on Account NI, which manages 35 bank accounts.  If not unique, it is fairly unusual in providing a shared service to a range of bodies rather than to a single organisation.  Most centralised systems provide for one organisation or one small family of organisations, whereas we are providing for quite a wide range — 30 different organisations.  That has its own challenges.


You mentioned prompt payment.  The 10-day prompt payment was not envisaged when Account NI was being planned:  30 days was the standard.  We have managed to accommodate 10-day prompt payment and are doing about 91% of our payments on time.  The point that Paul reminds me of occasionally is that that is cycling £2·8 billion through the Northern Ireland system.  That is money that is going to businesses — our suppliers — and it is supporting the economy by delivering public money quickly and directly to suppliers through our systems.  There is no way we could have achieved 10-day prompt payment with the old systems.  So, it is a success story.


There are things that we would do differently if we were doing them again.  There are messages in the Audit Office report that are well taken, but, by and large, I agree with you that there are good things in this report.  It is a good project overall.


The Chairperson: Before I ask my next question, I have a message from Hansard:  could members point their mikes towards them?  Could members' devices also be moved aside from the microphones?  Hansard is having difficulty in picking up the sound today.

Stephen, the Audit Office estimates that the implementation of Account NI has cost at least £213 million.  DFP cites the lower figure of £186·7 million, but could a transaction service like Account NI not have been provided for much less?


Mr Peover: We do not think so.  The full business case quoted £167·4 million, and our job was to deliver the proposed system for that cost, and we by and large did that.  I will come back to the costings in a minute.


This is quite complicated.  You referred in your introductory remarks to the earlier stages of the project — the accounting services review and so on.  The £213·1 million in the report refers to the implementation costs, as in the full business case, the cost that went before that in the accounting services review, and the subsequent costs involved in bringing new organisations into our systems, such as the Department of Justice, that did not exist when we were designing the system.


In a sense, it does not really matter what figure you quote as long as you compare consistently across the period of time in question.  The right figure to use is the full business case cost, because that is what we were given, that was our budget, that is what was approved for the project and that is what we delivered.  There was a cost overrun, but it was not that large:  we net it off at 8·4%, which is within the 10% tolerance that is normally allowed in business cases.  So, we think that it is right.


I do not want to give the impression that I am disputing the figures:  there are different figures.  My colleague used the example of buying a car.  If I decide to buy a car, I might do a lot of work before I get to the point of making a decision, such as going round car dealers for hours, reading brochures and looking online.  Then, I negotiate with the salesman and get a car for £10,000, £20,000 or whatever it is. To me, the cost of the car is the £20,000 that I pay for it, but the time that I invested up front in making the decision about the sort of car that I wanted and where I wanted to buy it and so on comes at some cost.  Up to the point when I shake hands with the salesman, I can walk away from that transaction and say that I have changed my mind and I am not going to do it.


The decision point is the point of the full business case being approved by Ministers and then moving onwards.  For us, the real cost is the cost from that point on — from the point of decision.  If you do not make the decision, the only costs are sunk costs; they have been incurred and they are, in a sense, past costs, but in implementing the project the real costs are the costs from that point forward.


At the other end, when the project is up and running fully, as it was in 2011-12, and you start taking on other bodies such as the Public Prosecution Service (PPS), the DOJ, the Driver and Vehicle Agency (DVA) and so on, those are additional costs.  They were justified by their own business cases and, from our point of view, they are cost-neutral because they cost us money, but we get money from the Departments that supply the services to cover those costs.  Those are additional and extra; if I go back to the car dealer and ask him to put go-faster stripes down the side of the car or fit alloy wheels, those are additional costs after the event and I have to justify them in their own right.


If you look at it one way, £213·1 million is a final cost that covers the whole lot, but given what we are being asked to do and given the performance of the Department against its full business case, we think that we have delivered within cost and with only a slight cost overrun.


Mr Paul Wickens (Department of Finance and Personnel): May I add a couple of things?  There is a perception that Account NI is primarily a payments system, and that is a big chunk of it, but if you go back to what Account NI was trying to replace, which was a whole set of accounting systems, the primary rationale in the first place was to replace, modernise and provide a common system and then to improve and standardise the service beyond that.  It underpins the financial management and accountability across the Northern Ireland Civil Service (NICS).


To put payments into context, it is worth taking a second to go through some of the other things that it does.  It deals with procurement, travel and subsistence, invoicing, cash and debt management, bank reconciliation, general ledger (GL) reporting, asset management and stores inventory, as well as managing the whole system itself — including the service desk that supports it and all the rest of it.  Then you can see the 750,000-odd payments that we paid last year in the overall context of the wider Account NI.  It is important to recognise that overall context.


The Chairperson: Before I let members in, I want to go back to what Mr Peover was saying.  Would it have been possible to provide what is, basically, a transaction processing service for a much lower figure?  Going back to your day at the garage or the showroom, did we pay for a Rolls-Royce system?

Mr Peover: It is one of those things.  Do you take the opportunity, when you are putting a new system in, to look at the full range of services that Paul described?  The decision was taken at the time to do that.  It might have been possible to buy a cheaper system but it would not have done the things that Account NI does.  The system has 21,500 online users and every payment that is made now, with minor exceptions, comes through Account NI in the NICS.  We are, as I mentioned, cycling £2·8 billion through the economy, and we are doing it in an effective and efficient way.  We are also providing other services to the Departments and agencies.


Mr Girvan: I just want to go back to Paul's point about how the organisation was set up and what it was there to deal with.  Probably it is Stephen who needs to answer this.  In relation to the Departments that have bought in and are using Account NI, there should have been a definite audit trail to show the reduction or the lack of requirement in areas where that resource was being used previously and where they were dealing with it on their own.  Have you been able to identify that those savings are measurable in the first place, and that the Departments are not duplicating?  There is evidence that some are definitely duplicating what Account NI is doing.  That is, perhaps, not something that Account NI has to answer for but it is something that, as Northern Ireland plc, we should be looking at to ensure that we are not duplicating, double-checking or bean-counting twice.


Mr Peover: That is quite a complicated question, and I will try to answer it as cleanly and clearly as I can.  The justification of the project in the full business case included a calculation that there were 293 people doing transaction processing in the Departments.  The aim was that 198 of those posts would be created in Account NI — my friends can keep me right on the numbers — 29 would remain in Departments to act as backup but might be reduced over time, and 66 posts would be saved.  This is where it gets quite complicated, and I wrote this out for my own benefit.  The 198 transferred across to Account NI and, as of 31 March 2013, there were 191·43 staff in Account NI.  It was projected that, with the 198 that we started off with, we would save another 40 posts through efficiencies over time.  So, that would have got us down to 158.  So we have 191·43 at March last year, but within that there are 18 who are on there because of the 10-day prompt payment requirement, which was not the original part of the calculations, and 41 are there because we took on the Department of Justice, PPS and DVA.  Our reckoning is that if you leave aside those things that changed during the past couple of years, against the 198 that we started with and the 158 that was our target for reductions, we have 132·43.  So, we reckon that we have overachieved on our savings target in Account NI against the target that we were set as part of the full business case.  It is hard to disentangle those figures because things are added on and taken away, but we are quite clear that we have overachieved on the savings target for Account NI.


Your second point was about Departments and the 29 staff.  In the report, there is a table that quotes figures of 50 and, in the case of the Audit Office, 113.  Let us start with the 50.  The 29 was on the assumption that a thing called project accounting would be included as a module in Account NI.  It turned out that that was not needed or appropriate.  So, 15 posts could have been saved by Departments but were not saved.


Mr Girvan: Was that in the original business case to get the spend of £167 million for the set-up?


Mr Peover: No, they were not crucial to that business case.


The difficulty is getting people to tell you how many staff they have doing things.  If we are going to be purist about this, as far as we are concerned in DFP and Account NI, nobody can process a transaction outside Account NI.  We do the processing; it is the staff in our organisation who do that.  There may be people outside who regard themselves as doing transaction processing, but they cannot actually make a payment and handle the transactions in the way that our staff do.  The table in the report shows that DFP has been one of the outliers in the Audit Office figures, having many more than we are supposed to have.  My finance division tells me that we have 5·75 staff in what you might loosely call transaction processing.  They are not processing transactions but are chasing things up, overseeing things and sorting out problems, and there will always be a need in a Department for somebody to relate to Account NI and provide those sort of backup services. 


If we had a completely distributed system and every one of the 21,500 staff was doing, from time to time, a requisition to buy whatever it might be, pay for a service or organise a conference, those staff will spend a few minutes a day or a week inputting, and if you want to count the time, you have to count their time.  In some parts of my organisation, they have centralised that so that a group of people do that on a regular basis and, therefore, do it more effectively and efficiently, in their view.  I am not trying to throw dust in your eyes, but it is very difficult.  Our view is that nobody outside Account NI is doing transaction processing in the strict sense of the term.  There are people employed in finance divisions doing associated work, which is probably necessary and valuable work.  We counted 50 of those:  the 29, plus the 15 for the project accounting module and a shortfall of six.  So, according to our calculations, there are six too many against the business case.  The Audit Office says that there are 113 rather than 50.  The numbers are so out of line for two Departments — my Department and DARD — that I think that there is something wrong with the way our Departments have returned their figures to the Audit Office.  I have spoken to my finance director about that and she has given me a breakdown of all the staff in the Department who are loosely described as transaction processing.  Of course, Enterprise Shared Services (ESS) did not exist, so there are staff in ESS who have been added to the system from the start. 


That is an attempt to explain it.  I am not sure that it is as clear as it could be, but it is as much as I can do.


Mr Girvan: I am going to move back to the mention you made of 18 additional staff who were brought in to achieve the 10-day payment window that was set.  I agree that it has been a great help to get money through to people quickly.  It helps their cash flow and keeps businesses flowing.  My understanding — I look at things from a business perspective — is that you make an initial investment to catch up so that you are dealing with payments in that window, but once you have achieved that you will be paying the same number of invoices as you were when you were meeting the 30-day target, because you are paying them earlier.  Were those 18 additional staff permanent, or were they only there to meet the requirement?  You will find that I see things from a background of running a business, so I know that every penny counts.  You are saying that we will make the investment now to catch up, but once we achieve that, we should be paying the same number of invoices, no matter whether it is within 10 days or 30 days.


Mr Peover: It is a good question.  There are a couple of things.  To use the car analogy again, if you are driving a bit faster, you use more petrol, so if you are churning things through in 10 days rather than 30 days, you need more staff in order to support that.


Secondly, it is not just a static position.  It is not just that we have 500,000 invoices year on year.  Our numbers have been rising significantly.  In 2013-14, we will process around 30% or 40% more invoices than in previous years, and we have increased the staff to do that, but by only half of that percentage increase in invoices.  There are a number of factors to explain the staffing.  It is not just the investment to get the 10-day prompt payment; we need to maintain it.  John may be closer to that.


Mr John Crosby (Department of Finance and Personnel): I take the point that you are making.  I remember, back in my days at the Driver and Vehicle Testing Agency, when MOT test times fell behind, it took the same amount of resource to carry out the same amount of tests going forward, but we are looking at a new turnover every day.  The best analogy to use is probably that of a post office counter or supermarket:  if you want to get through a till at a supermarket quickly, you need more operators on, and you need those operators on all of the time.  If you clear a backlog, say in the post office or the supermarket one Friday night, that backlog will be there again tomorrow if you do not have extra operators and extra staff.  Prompt payment is inherent —


Mr Girvan: I do not necessarily agree with that, but I will come back to you on that.


Mr Crosby: Processing quickly is inherently less efficient than processing in slower time.


Mr Wickens: You asked about the overall cost benefit at the very start of the project.  The numbers on the departmental side were not part of the cost benefit, so the opportunity to pursue those benefits —


Mr Girvan: They are savings to the Departments.


Mr Wickens: They are savings to the Departments, so they were not part of our business case.  That is correct.  If you then look at the 18 staff that we have doing prompt payments, a simple answer is that, if we did not have to pay within 10 days, we could have 18 fewer staff.  That is the very simplistic view of the thing.  The whole system, which is inherently complex — I have gone through some of the areas that it covers — was designed and built around a 30-day statutory obligation that we have.  We measure that as one of our key performance indicators and have a very high performance against it.  It is an extra cost and it means that we have to incur extra staff.  I suppose that the debate from an economic perspective is whether that is a cost worth having to be able to pay the money out into the economy in the way that we have described.  That is a matter for discussion, perhaps.


Mr Girvan: There was one wee point, and that was on the use of consultants.  I do not know whether they were brought in on a temporary basis to achieve certain things or whatever, but there seemed to be quite a high use of consultants to deal with some areas.


Mr Peover: We have not used consultants since 2009.  They were used at a point when we could not get staff from Departments deployed to the project to get the project to the point of delivery. 


Sorry, I missed the second point.  I am not trying to avoid it.  You asked whether there should be commonality in Departments and whether we should be pressing Departments. I will speak for a second in a personal capacity rather than as DFP permanent secretary and accounting officer.  When the issue was raised originally with the permanent secretaries' group, I was and still am a strong advocate of common retained systems.  It ought to be possible to evolve a model that says that, with x amount of business, you need y staff.  We should try to achieve a common model.  That was not the view of Departments.  Departments are very different:  you have everything from DCAL, with 400 staff, to DSD, with 8,000 staff.  Their accounting officers, quite naturally, want to be able to determine the structure of their Department and satisfy themselves that their structures will deliver the business that they are responsible for.  Having somebody like me or one of my predecessors coming to you and saying, "You don't need six staff; you need only two" is hard.  At the end of the day, I am not accountable for their business; they are accountable for their own business.


We think that there is scope for rationalisation.  A working group has been set up to look at how we might do it.  We would do it via agreement with Departments because it is their business, and they have to satisfy themselves that they can stand over what they are doing and how they are doing it.  There should not be duplication of systems, and they should not be running parallel systems.  They certainly cannot make payments, but people are cautions and like to run their own system sometimes.


Mr Wickens: I will give you an idea of how that is governed:  in enterprise shared services, we have a strategy board chaired by the permanent secretary of DETI.  On that board, we have a number of deputy secretaries.  One of those chairs a finance shared services board, which helps us to shape and influence what goes on in Departments.  Two of the big areas that we have been looking at recently through that board are collaborative procurement, which will be the subject of another discussion on another day, and integrated reporting.  The discussion has been moving more and more towards the concept of what we can learn from how one Department does things and whether there is such a thing as best practice that could be shared, rather than enforced, on a basis that could help to create improvement and achieve even more efficiencies.


Mr Clarke: I am glad that I came back in before Paul Girvan asked all my questions.  Just as I was coming in, Paul Wickens was responding to his question about cost invoices.  I think that you justified your cost by saying that the 10-day turnaround makes your cost much higher.  I appreciate the point that you are making.  However, the report compares the top performer at £2·75 with Account NI at £9·73, albeit that you justified that by having more staff.  The top performer turns around invoices in 17 days and not 10, whereas others take, as you said, the statutory 30 days.  Some take 29 days.  Given that the difference is only seven days, what accounts for the variance between £2·75 and £9·73?


Mr Peover: This is a complicated one.  The figures quoted are taken from a benchmarking exercise that the Audit Office commissioned by getting figures through consultants from the American Productivity and Quality Center (APQC) and by contacting some organisations. We do not know the make-up of those figures.  We know the make-up of our £9·73:  it is made up for the purposes of the notional costing of Account NI's services.  The direct staff costs on transaction processes in the accounts payable side is £2·05, and that is quoted in the report.  In addition, we have —


Mr Clarke: May I just stop you, Stephen?


Mr Peover: Sure.


Mr Clarke: What you said poses a very good question.  You say that you do not know how the Audit Office made up its figures, but did anybody ask?


Mr Peover: Yes.

Mr Clarke: Did you get an answer?


Mr Peover: We were told that the consultant contacted APQC.  We contacted APQC, but it could not or would not tell us what figures it gave or how it gave them.  We also contacted the local organisations quoted, such as Belfast City Council, and asked them for figures.


This is not a criticism of the Audit Office.  First, it is inherently difficult to benchmark shared services.  Not many shared services are like ours, which cover 30 organisations.  Most shared services are for a single organisation.  Secondly, some of them are private sector; others are public sector.  The private sector, by and large, tends to pay over long timescales — sometimes two months and maybe longer.  They also work for a single organisation.  What about the counting of overheads?  How do you count the cost of your shared service?  Our £9·73 includes senior management salaries, running costs, system costs, licence costs, depreciation —


Mr Clarke: Are you suggesting that they are all paid too much?


Mr Peover: No.


Mr Clarke: What you are saying is —


Mr Peover: There is a DFP overhead of £1·13, an ESS overhead of 35p and a depreciation of £2·01.  There are lots of factors in that £9·73.  I am not saying that that figure is wrong.  That is the figure —


Mr Clarke: It is also three and a half times more than the top performer in the report.  I presume that you accepted the report.


Mr Peover: I do not know what the top performers cost.


Mr Clarke: I presume that you accepted the report, Stephen.


Mr Peover: No.  We commented — if you read the report, you will see that, in that paragraph, we refer to our costs of £2·05 as being the direct cost of our services.  How could we accept a report when we do not know the make-up of the benchmarks?  The report acknowledges that benchmarking is inherently difficult.


The Chairperson: Mr Peover, you mentioned that it was incredibly difficult to benchmark, but the Audit Office seemed to be able to do so easily.  It took the information off a shelf.


Mr Peover: It depends whether it is useful.  You can take things off a shelf, but do they give you hard information that you can benchmark your services against?  I do not know whether they do.


We contacted the education and library boards, the health sector and Belfast City Council, and the figures that they gave us were very different.  The direct cost of their services is more or less the same as our £2·05.


Mr Wickens: I again want to try to give a bit of background information.  I have been liaising with shared services chief executives in the Whitehall Departments in Great Britain for a number of years.  On a number of occasions, they have tried to work out how to benchmark between them.  The National Audit Office has also given a commentary on the issue and made certain statements.  Benchmarking is inherently difficult.  We brought over an organisation called The Hackett Group to talk to us.  It is recognised as the world leader in shared services benchmarking, and it has access to a much larger database than the APQC would have. 


You do not get much for a benchmarking exercise that lasts 10 days.  That gives you only a very rough skim off the top.  That is not an overt criticism of the way that the benchmarking exercise was done.  It was useful as it allows us to see where the areas are that we need to dig into to really understand how we should benchmark in the future. 


At this point, based on the data that has been generated, the only thing that we can compare is the £2·05 of direct cost.  At the level that the benchmarking has been made available to us, it is almost like using apples, cucumbers and chestnuts as comparators.  We need to work out how to get the apples to apples comparison.  We accept that there is a need for us to benchmark much more effectively and drive towards a good cost benchmark to provide transparency of costs and allow a proper and robust challenge.  I think that the £9·73 figure hides some of the detail underneath that.


The Chairperson: The Comptroller and Auditor General may want to come in.


Mr Kieran Donnelly (Comptroller and Auditor General): Mr Wickens makes the very valid point that it is important to benchmark.  When we came on the scene, there was no benchmarking, so it is very pleasing that there is an acknowledgment of its value.  In the absence of anything, we had to make some sort of initial stab at it.


It is worth saying that, for the other organisations that we benchmarked against, overheads are included in the mix.  Irrespective of that, the bottom line for performance metrics is that there should, we think, be a target for the average cost of processing an invoice — that is basic.  The fact is that it was not there.  That is a very simple measure to assess efficiency.  Going forward, I think that it would make sense to have a target for the average cost of processing an invoice and a consideration of how to bring that cost down over a period of years, irrespective of what overheads are in there or not.


Mr Peover: We entirely agree with that.  As far as a consideration of how things would have been done differently is concerned, a more rigorous and earlier approach to benchmarking would have been good.  It needs to be proper benchmarking in which we are entirely sure that we are comparing like with like.


Mr Donnelly is also right in saying that some sort of key performance indicator for the cost of processing an invoice would be good, and I think that it is desirable.  All that we can say is that we can demonstrate efficiency improvements year on year, and the number of invoices turned over per member of staff is increasing significantly.  In that sense, we are comfortable that we are improving.


On Mr Clarke's point, we want to satisfy ourselves that we were doing as well as we could and that we were absolutely sure that we were comparing with, if not the best in class, at least the median or better.


Mr Clarke: You were not benchmarking.  You did so only on the back of the report, not prior to it.


Mr Peover: No, we were not.  That was recommended in the gateway 5 report.  The Audit Office then undertook benchmarking.  We did not do that but waited to see what the outcome of its work would be.  However, we agree with the Audit Office that we should be benchmarking.


Mr Wickens: We should be benchmarking against other shared services, which the report does not do.  It compares a shared service with, effectively, an accounting department.  There is a radical difference.  Albeit both have overheads, they will be different.


The key point is that we need to do more on that.  Doing proper benchmarking will cost us money, but it will also be difficult given that Whitehall shared services have not been able to agree on how they should do it.  We contacted them very recently to establish what progress had been made, and they are still struggling with exactly what should be there.


Mr Peover: None of that detracts from the point that we want to benchmark.  We want to prove that it is a good system — as good as it can be — and keep the pressure on.  We are a service organisation that serves 30 organisations — 29 others and ourselves.  They are entitled to expect a high-quality and cost-effective service.  If we can demonstrate that to them, so much the better.


Mr Clarke: I am interested in what you did differently to improve efficiency.  You say that the ratio is up, so something must have changed.  To my mind — the suspicious mind that I have sometimes — Departments said that there were no efficiencies to be made there.  Stephen, it is interesting to note that you said that your processing is better than it was.  Something must have changed to cause that, and that would have had a knock-on effect on the cost of processing an invoice anyway.


Mr Peover: I want to quote some figures.  The number of payments made per full-time equivalent member of staff in 2010-11 was 7,513.  In 2012-13, it was 8,818, and, this year, we project that it will be 11,066.  That is a 47% improvement in the number of invoices processed.


I will let John talk about what was done.  He is close to the specifics of how we made those efficiencies.


The Chairperson: Before we move on, two members want to come in on Mr Clarke's previous point.


Mr Dallat: We seem to be going all over the place.  The analogy of the car came up a few times.  It was a bit of an old banger, was it not?


Mr Peover: No.  It was a very effective car.


Mr Dallat: Do you think so?  The key performance indicators do not seem to suggest that she was top of the range.


Mr Peover: That depends on whether you accept those key performance indicators.


Mr Dallat: Maybe I should ask a question about that.  The report seems to have been very professionally prepared — it is one of the best that I have read.  The Account NI content is not necessarily the best, but it is a very professional report.  Why do you not accept it?


Mr Peover: We do not know the basis on which the benchmarking figures were calculated.


Mr Dallat: Do you not think that that smacks of a level of arrogance well beyond what you should be doing?


Mr Peover: No.


Mr Dallat: Perhaps we should move away from the car analogy because I do not think that it was a good idea.


Account NI is top-heavy with grade 7 civil servants doing what I would have thought required only supermarket check-out skills.


Mr Peover: I do not think that it is.


Mr Dallat: This was a right old cash cow for you at the top end, was it not?


Mr Peover: No.  John will keep you right with the figures.  We have 180-odd or 190-odd processing staff and 16 management staff.  Bear in mind that we have not engaged any consultancy since 2009, despite the fact that we have taken on substantial new organisations.  Part of the reason for the senior management overhead is to avoid the need for consultancy.  We are criticised sometimes for not learning lessons and not being able to acquire the skills.


Mr Dallat: You would never be criticised for employing consultants if you needed them.  You do get criticised when consultants' contracts start off at £0·97 million and spiral to £9·6 million.  Do you accept that that is a valid criticism?


Mr Peover: No, I do not.  We have discussed this before, and you know my views on it.


Mr Dallat: Mr Peover, I am not sure that you should have bothered coming here.  It looks like you will not accept anything.


Mr Peover: It is acknowledged in the report that the consultancy was well managed and approved in accordance with the procedures with the Minister, and so on, and with business cases.  We have been over that topic before, and you know my views on it.


Mr Dallat: I am so sorry for boring you, but it is my job to ask you the question.


Mr Peover: No, you are not boring me, but you know that we have been over the topic before, and I do not accept the point that it was uncontrolled expenditure.  The point that I am making to you now is that since 2009 we have not had to engage any other consultancy.  Part of that is because we have a reasonable number of senior management.


Mr Dallat: There are so many senior civil servants that you do not need consultants.  It is six of one; half a dozen of the other, is it not?


Mr Peover: No, it is a lot cheaper.


Mr Dallat: I am not so sure about that.


Let us look to the future.  Where do we go from here?  If the ordinary man in the street were to be told that every cheque that he wrote would cost him £9·60, he would say that that was not right.  What is wrong with all these other Departments that have not been brought into this wonderful circle of friends?


Mr Peover: All the Departments are on the system, as are a number of their agencies and some non-departmental public bodies.


Mr Dallat: A pile of them are not.


Mr Peover: Forty-five of them are not.


Mr Dallat: Is that not a failure?


Mr Peover: No, I do not think that it is.  Each organisation has to decide for itself what systems it wants.  Organisations in that group of 45 have varying numbers of staff, from 30 to hundreds and, perhaps, thousands of staff.  A system that is appropriate for a Department may be too sophisticated for a small organisation.  They have to make their own decision on whether the system is value for money for them.


Mr Dallat: So it is just a matter of do it or do not do it.  We have created this golden cow, spent £213 million on it and you can decide for yourself whether you want be part of it.  Economies of scale do not matter, and the way in which public money is used is taboo.  You come here, reject the report and argue against anything that is in it.


Mr Peover: For a start, I have not rejected the report.


The Chairperson: The benchmarking element is, I think, what has been rejected.


Mr Wickens: The full business case was designed to build a shared service, and everything that emanates from that, for 21 organisations.  That is the base that we go back to.  Over the past couple of years, we have increased the number of organisations to 30 — nine additional organisations have joined us — and achieved significant proven economies of scale on the back of that.


If you take the benchmarks that have been referred to — all of them, not just the cost benchmark — we are above the median in transactions per full-time equivalent (FTE), so we are performing very well on a productivity basis.  We are at about the median in matching invoices to purchase orders.  Those are good, valid comparisons.  We are world-class in prompt payment and in keeping money flowing in the economy, paying our suppliers within 10 days. 


There is a big question on costs that we need to come back to.  We need to go back and say, "Now, having received the information from this benchmark, it is appropriate for us to revisit that and really understand the cost of doing all those things that are well accepted:  our suppliers like it, and our users like it."  It is a good system.


Mr Dallat: I know that we want to move on, Chairperson, and I am conscious that some people outside the Building might watch the coverage of this session, but, if we do not make some progress on how we will improve this, the only people watching us will be those who suffer from insomnia.


Companies such as Amazon handle millions of transactions a day.  If it had those costs, what do you think it would do?  It would be bankrupt within a week.

Mr Peover: I just do not accept your point, Mr Dallat.


Mr Dallat: I know.  You never do, but anyway —


Mr Peover: That is because it is wrong in this case.  The £9·73 is a notional cost for a charging model.  It is not necessarily comparable with the other figures quoted in the report.  We do not know whether it is or not; we think that it is not.  We would need a proper benchmarking investigation to decide whether those figures are valid comparisons.


We are doing quite well on the other benchmarks that Paul talked about.  Our direct costs are £2·05 per invoice processed, which is similar to those of other comparable organisations.  We may be wrong on that, and you are right that we should be benchmarking more effectively.  We will benchmark more effectively — that is the real change that will come from the report.  It is not that we do not accept benchmarking; we do not accept that the comparisons are valid and are wary of accepting that we cost twice as much or four times as much as other performers.  We need to know the basis.


Mr Dallat: I know that other members need to get in, but I have spent some time trying to get my head around the idea that the chief accounting officer does not accept benchmarking.  Who makes up the rules that you go by?  Do you make up your own rules?


Mr Peover: I did not say that I did not accept benchmarking; I said that I did not accept these benchmarks.


The Chairperson: We should move on.  I will take the last supplementary in this section and then move on to Mr Rogers's question.


Mr Copeland: Is it in order to put a question to Kieran for clarification?


The Chairperson: Yes, but I was going to move on to Mr Rogers.


Mr Copeland: All I wanted to clarify was whether Kieran considered the benchmarking to be adequate and robust.


Mr Donnelly: It is the best available in the market.  In the absence of anything else, it was important that we started the debate.  It is acknowledged that more work needs to be done on this.


Let us go back to basics.  A key purpose of shared service centres was to reduce costs and move resources to the front line.  That was part of the philosophy.  Therefore, the surprise to us was that there were all sorts of metrics on Account NI but nothing on cost, so we had very little to work with.  It is very encouraging that there is now an acceptance that we need to open up the cost debate.


The Chairperson: That is what the report allows us to do.


Mr Wickens, there was a strategy board meeting, and you acknowledged the governance arrangements that exist for management to manage operations from here on in and how they would be implemented.  You said at that meeting that the arrangements were to "consolidate" and improve systems. 


Mr Peover, you would not accept the benchmarking.  I am sure that you acknowledge that you were found to be four times more expensive than others.  This report has opened the door so that we can all see in, and there is room for improvement.


Mr Peover: I accept Mr Donnelly's point that we should have a key performance indicator on cost processing.  I also accept that we should be benchmarking and that we should have been doing so before now.  I do not accept that we can draw valid comparisons between the costs quoted.  They may be valid, but I do not have enough information to make that judgement.


Mr Clarke: The point is that, if it costs you £2·05 to process, which is what Paul said, what is the other £7 made up of?


Mr Peover: The figures —

Mr Clarke: To a degree, I take your point that we do not know what the £2·75 represents in the case of the top performer.


Mr Peover: The rest of it comprises £1·53 for other salaries —


Mr Clarke: What does that mean?


Mr Peover: It means senior management, development, training, finance, contract reporting, services —


Mr Clarke: That takes us back to John's point about too many senior officials.


Mr Peover: No, it does not.  We have a shared service that covers a range of functions.  We attribute 47% of the management cost of that service to transaction processing because 47% of the workload is in that bit of the organisation.  That may or may not reflect an accurate allocation of the management overheads to that part of the business.  For the purpose of doing a notional costing of Account NI, that is how we have to apportion the overheads.  So there are all of those things — the running costs, stationery and so on, system costs, which we —


Mr Clarke: May we have a breakdown of those costs?  If you call them out, I will write them down.


Mr Peover: Yes.  We mentioned the processing figure of £2·05; other salaries cost £1·53, running costs are 11p, system costs for contracted-out services are £2·39, and licences and application software support cost 16p.  Depreciation, which is non-cash and not charged to the public, is £2·01.  The ESS overhead recovery is 35p, and the DFP central overhead recovery is £1·13.  That all adds up to £9·73.


Mr Wickens: That is why I said at the start that it was important to recognise that Account NI is much, much more than a payment system.  All of the other things that I mentioned earlier attract costs, and the issue is how you apportion the costs and overheads.  That is where there is a useful and valid debate to be had.


Mr Clarke: When did the £2·39 system cost disappear?


Mr Wickens: That runs the whole way through.  It is our contracted-out cost to BT for running and managing the underlying technology systems and support behind them.  It was part of the business case.


Mr Clarke: Which business case?  The first one or the last one?


Mr Wickens: The full business case in 2006.


Mr Clarke: I am somewhat sceptical because the company charging £2·75 is bound to have many of those other costs as well.


Mr Peover: We may be performing very badly.  The problem is that we do not have the information to draw those comparisons.  For example, I cannot say that it costs us £2·05 for that bit of it and we are attributing so much of the other cost but somebody else is managing those other costs for half the price.  We need benchmarks.  I have no problem accepting that point.  It is very valid.  We accept that point that the Audit Office makes.  The question is —


Mr Clarke: If you look at the processing of invoices, then given the numbers involved, surely efficiencies should be found in the high number.  Your costs are going to escalate if your invoice numbers were fewer.  It is a good job, Paul, that you got some of those other people on board, or we would be looking at a higher figure.  If you proportion the cost given the number of invoices you are processing, and if you have reduced that number, your costs are still going to be equally high.  Your processing costs are higher.


Mr Wickens: If you look at the trend over the past couple of years, we are, effectively, processing more invoices with fewer people.  More invoices per person are going through the system.  We are getting proven efficiencies on that basis.

Mr Clarke: Are you trying to say that that £2·75 is only the processing cost?  Is that your assumption?


Mr Peover: The £2·05 is the direct staff cost —


Mr Clarke: No, the £2·75 of the top performer.


Mr Peover: I do not know what the top performer's costs are.


Mr Clarke: I thought that you said that you contacted it to try to find out.


Mr Peover: We contacted the APQC.  I do not know whether it knew how the consultant derived the figures.  We just do not know.


Mr Wickens: It could be against an offshore private sector organisation doing payments for only one organisation.  We do not know.  It is a valid question that needs to be answered in order to see whether we are comparing apples with apples or apples with chestnuts.


Mr Clarke: It is processing 16,500 invoices as opposed to your 12,300.


Mr Wickens: Again, that is something for us to strive towards as we continue to improve the number that we are doing per person.  We have proven that we are doing very well.  We would like to continue to increase that number.


The Chairperson: We will move on now.  We have been here for an hour and we have just gotten into the first part of the meeting.


Mr Rogers: You are welcome, gentlemen.  Will you tell me briefly what was included in the 2003 outline business case?


Mr Peover: John would know the detail of that better than me.  The original project started with a review of the accounting service.  The proposal was to create, effectively, another accounting system of the type that we had — [Interruption.]


Mr Dallat: I am sorry.


Mr Peover: — but with four processing centres rather than six.  It was just an accounting system.  The further development looked at what more could be done and the options for moving towards shared services.  That was the genesis of the outline business case.  We are slightly hampered because none of us was personally involved at the time, so our knowledge is derived from talking to colleagues and reading, but John may be a bit closer to the initial concept.


Mr Crosby: The accounting services review (ASR) looked at replacing the disparate systems across the six legacy Departments pre-devolution.  The outline business case looks speculatively at the range of options available, including the option to deliver a shared service.  An outline business case will always have estimates in it; for example, the cost of a contracted service will have to be estimated because there is no procurement at that stage.  The outline business case covered 10 years rather than the 12 that we subsequently agreed.  It took a very limited view of how to deal with the legacy systems, manage them, wind them down and manage the migration.  There were quite a few things in the outline business case that were not ultimately in the full business case.  The outline business case is the initial thinking for the full business case.


Mr Rogers: It is hard for me to comprehend how the whole scope of the project was not included in the outline business case.


Mr Crosby: It was there, substantively.  There was very little missing from the outline business case.  What points do you have in mind that were not included?


Mr Rogers: Basically, between 2003 and 2006, the scope of the project changed.


Mr Wickens: One of the big things that happened was that Gershon published his report in 2004, which was the first time that shared services had been mentioned in the context of the public sector as a way of doing things differently.  I was not there and have not read the full details of the outline business case, but one massive difference that creeps into the scope is that it takes a shared services approach to this as opposed to just replacing a set of systems.


Mr Crosby: And, of course, by 2006, prices had moved on considerably.  The increase in the duration of the period obviously increases the bottom line, but the increased period was deemed to be the optimal period in which to provide the necessary return on the investment.


Mr Peover: As John said, when you go beyond the outline business case, start advertising the proposal and get bids, you then begin to see the market price.  I know that Mr Dallat does not like the car analogy, but, I might want to buy a car for £10,000 and find that, when I look for the sort of things I want, I will have to pay £15,000.  You have to go to the market and test it to see what you can get. I think we had five bidders for the project.  We got a preferred bidder and went into detailed negotiations with them about how the model would be developed as part of the full business case preparation.


Mr Rogers: How did this change in scope; what effect did it have on the estimated costs, and what services were not included in the outline business case that turned up between 2003 and 2006?


Mr Peover: As I said, I do not know, Mr Rogers.  I will need to go back to the papers from back then and write to you.  We do not have the detail of that.


Mr Rogers: Do you not think it is important learning?  We are talking about moving forward and perfecting a much better system.  Do you not think that all those things should be well scrutinised?


Mr Peover: Yes, and I hope that they were scrutinised at the time, because the process moves, as always, to a full business case.  That was considered at the time and tested for affordability, cleared with Ministers and approved, and we then moved forward on that basis.  The decision point was the full business case.  What was possible before that, or what people thought was possible, is one thing.  We can find out and see what the difference is between the costs, if we can compare them, but I am personally not au fait with what happened back in 2003.


Mr Rogers: It took three years for the full business case to be approved.  Why?


Mr Peover: The outline business case was in October, and, if you go through the process set out in appendix 2, you will see that the formal proposals were submitted in October 2004; so, it took a year to get to that point.  The preferred bidder was selected two months later, in December, and then there was the gateway review and exploration with the preferred bidder of the nature of the contract.  Things were discussed, added in and taken out and there was a fairly robust set of negotiations with the contractor, as I understand it, because once you have got to a single contractor, you want to be sure that you are getting value for money from them.  The full business case was approved against the actual costs as estimated at that point in May 2006.


Mr Crosby: It is fair to add that there was a lot of work going on, in parallel, connected with the business transformation required; for example, agreement on a common chart of accounts was required.  When you bring 11 disparate Departments together, all of which have individual accounts, getting agreement on something like that takes time.  That would all have been going on in parallel, so that it would have been in place ready for the implementation to begin.


Mr Rogers: Another point I am making is that it took quite a while to get to the full business case; but why did the full business case go ahead when, at gateway 3, you talked about an amber status and that there were things that were not completely ironed out?


Mr Peover: Amber is what it is.  It is flagging a warning that you need to keep an eye on particular management issues.  It does not mean that you should stop.


Mr Rogers: This is unbelievable.  You can see why there was an amber, because the next thing we see is that, in December, which was six months later, the contractor gave notice that the June 2007 go-live date could not be achieved.


Mr Peover: The bottom line is that the project was delivered more or less on time.  For a £180 million to £213 million contract, it was delivered with only a few months' delay, which is fairly unusual in contracts of this size.  We can pick apart the bits of it, but the bottom line is that this was well managed.  The Audit Office report recognises that this was a well-managed contract, that any delays were well managed and that we soaked up the contingencies in the programme in order to try to deliver it more or less on time.  So, it was delayed by only a few months and was well within the tolerances that you would expect in a contract of this size.


Mr Rogers: Time does not permit us to do anything other than pick at bits.


Mr Peover: We can certainly go back, have a look and see if we can define some of the stages that you are talking about, but the overall position, and the Audit Office has acknowledged this, is that the thing was well managed.


The Chairperson: May I come in, Mr Rogers?  Were you content at the time that the preferred bidder selected in November 2004 was still the appropriate selection in June 2006?  What processes were in place for the selection of that preferred bidder?


Mr Peover: It was a normal competitive contract.


The Chairperson: Were you content that the same applied in 2006?


Mr Peover: I assume so.  We were not directly involved, but the decision was made.  It seems to have worked.  The thing has been delivered.  Nobody has challenged the award of the contract.


The Chairperson: Who was involved in the selection at that time?


Mr Peover: Do you mean which individuals were involved?


The Chairperson: Yes, who was involved in the process?


Mr Peover: That is a good question.  It was before our time.


The Chairperson: Could we get that information?


Mr Peover: John Hunter was my predecessor, and there was Bruce Robinson.  This spanned quite a time, from 2003 to 2009, so it is a period of six years.  A variety of people were involved.


Mr Wickens: Going back to the full business case of 2006, I accept the point you are making about the length of time it may have taken to get to that point.  We would not have reached contract signature and full business case approval had we not demonstrated value for money at that point in time.  The fact that we did not get a challenge then, or since, and the fact that we are living within the tolerances would lead us to indicate that we are broadly content with that.


Mr Rogers: The Chair is making a point.  Was it value for money based on the November 2004 figures, or when they appointed 18 months later?  It was not signed until 18 months later.


Mr Wickens: The only baseline we can go back to with regards to testing value for money is the 2006 baseline.  So, it would have been an evolving situation working with the preferred bidder to get to a contract that was acceptable to both parties.  That is the way it was.


Mr Clarke: If we go back to the discussion that we had earlier about the baseline, how can you work out value for money?  You have acknowledged that it is difficult to do that.  I appreciate that you would not have been there at that time, Stephen, but, in respect of that business case, how can you judge something as being value for money if you have nothing to measure it against?


Mr Peover: It is normally just a comparison with the existing system; on the 293 staff who were processing, for example —


Mr Clarke: Following your previous point, do you accept my point that if you cannot draw a comparison with how others do it how can you draw the conclusion that this actually was good value for money?  I think we are sitting here today, still not knowing how others have drawn up their costs.


Mr Peover: Yes, but the full business case is not approved against other people's systems, it is approved against whether it is going to be value for money.  Is it going to deliver the savings to cover the costs of the system, or are you getting benefits in addition to the savings that are sufficient to justify investment in the project?  That is the way all business cases are managed.  They are rarely managed on the basis of a direct comparison with another organisation and an approach of, "They could do it for x; can we do it for half of that?"


Mr Clarke: I think that that is an unfair assumption, because if another company can do something more cheaply, that should be borne out in whether something is value for money.


Mr Peover: Yes, but that is tested through the procurement process.  You go to the market; you ask people to tender against the specification.  In this case, five companies tendered and the best tender was picked on the basis of the assessment criteria for the competition.  Again, I am slightly hampered in that we were not there.


Mr Clarke: The problem may have been in the criteria.


Mr Peover: I have to assume that the people who were there at the time applied the criteria appropriately.  Certainly, there was no challenge to the procurement, and the system has been delivered.


Mr Clarke: Nor am I saying that, but if whoever drew up the criteria did not look at drawing the right comparison, and setting aside that the criteria were followed, the criteria might have been wrong in the first instance.


Mr Peover: Yes, but it is difficult to go back eight years and say whether we would apply the same criteria now.  The system works.  The Audit Office report acknowledges that it works.  It has improved the quality of financial management and processing in the NICS.  There is more work to be done.  We could benchmark better; we could set proper key performance indicators on cost.  Going back to your point about the number of invoices processed per person, we could up our game.  There is significant room for improvement, but, fundamentally, the thing is working.


The Chairperson: Before Mr Rogers finishes, Mr Copeland indicated that he wanted in.  I will then let you finish.


Mr Copeland: We sometimes struggle to ask the most basic questions, such as this:  what did we actually get?  I presume that it was goods and services and stuff.  Who owns the intellectual copyright?


Mr Peover: That is a good question.  I do not know.


Mr Crosby: The system is largely based on proprietary software.  It runs on the Oracle e-business suite.  We deploy Real Asset Management, ReadSoft for scanning, and we use Cognos for reporting solution.  Layered on top of that is the configuration, which we apply to that to meet our business needs.  That then, is the specially written software, if you like, that pulls that configuration together, and that is what we have responsibility for.  That is deposited in escrow and, should there be any issue with the contractor continuing and so forth, we can get that back and get access to it.


Mr Copeland: I understand that.  However, some of the systems you mentioned are off-the-shelf and have been tried and tested elsewhere.  What I am really driving at is this:  we have paid someone a proportion of £200-odd million to weave it all together into a layer that specifically suits us.


Mr Crosby: Absolutely.


Mr Copeland: However, we are not totally unique in that we are a devolved Administration and have a number of Departments.  Who owns the intellectual copyright to the work that was done on our behalf and that was unique to us?

Mr Crosby: We own the work that is unique to us.  However, we do not own the actual —


Mr Copeland: Copyright?


Mr Crosby: I am not sure whether copyright is the right expression, but certainly the intellectual rights.


Mr Copeland: I have no real reason to ask, but you know as well as I do that the first time you do something is the most expensive time to do it.  Once you have it done, is there anything to prevent the system that we have paid for and that has been made for us suddenly appearing on offer to another organisation structure or body?


Mr Crosby: Not as such.  However, there is nothing to stop us using and deploying what we have learnt from developing the system to suit the Northern Ireland Civil Service business needs.  Knowledge will be obtained in the course of developing any system.  It is important to recognise —


Mr Copeland: John, we are not in the business of selling computer systems.


Mr Crosby: I know, but the Oracle system provides you with a shell or framework.  It is like any accounting system; it comes off the shelf; it does not have your chart of accounts in it; it does not have your suppliers; and it does not have your users in it.  Those have to be put in.  Therefore, it is not of any use to anybody until it is configured for your business.


The Chairperson: It is unique to each individual business.


Mr Rogers: With regard to the bidder, etc, in November 2004 you moved quickly to get the preferred bidder but, by your own admission, you said that it was really the June 2006 contract when he signed up.  In other words, the other three or four bidders had no opportunity to tender for the 2006 figure.  Can you clarify something for me?  When you were asked why you went so quickly to move to the preferred bidder, you said that it was to remove competitive tension.  As an ordinary man on the street, I would have thought that a little bit of competitive tension would be good to ensure value for money and that the contract you signed in June 2006 was a competitive tender.  How could you ensure that it was a competitive tender when it was so different from the one 18 months earlier?


Mr Peover: I do not think that it was hugely different.  The competitive tension was at the point of procurement.  We had five bidders and we chose a preferred bidder.  That was at the competition stage.  Thereafter, we had to work with the preferred bidder and flesh out the system and put pressure on him as part of the discussions.  However, the competitive tension was there.  It was not a single-tender action; it was not a single bidder; it was five bidders in competition with one another.  Whatever the criteria, British Telecom (BT) came out as the preferred bidder.  A decision has to be made at some stage, and it was made at that stage.  We had five bidders, and we chose one.


Mr Rogers: You said that there was not a lot between 2004 and 2006.  However, if you look at the costings, you will see that, in October 2003, the costing was about £113 million, but the costing in 2006 was £169 million.


Mr Peover: That is the outline business case.  That is a different stage in the process.


Mr Rogers: There must have been a big change in the contract over that period.


Mr Peover: Again, we need to look back at this, as none of us was involved.  Where there are five bidders involved, the decision on a preferred bidder is made in the competition between the five.  Presumably, it was made on the basis of value-for-money criteria as well as quality, but we need to go back and have a look at it.


Mr Rogers: Do you not agree that, for the ordinary man on the street like me, it is hard to believe how a bit of competitive tension would not help to ensure value for money?


Mr Peover: Competitive tension was there at the point of the award of contract.  There were five bidders, and we chose the best of those five bidders.


Mr Rogers: That was in 2004.  It was signed 18 months later, but the goalposts had moved a bit.

Mr Peover: In April 2004, four of the five bidders were invited to submit outline proposals.  The invitation to negotiate methodology was approved in July 2004.  The formal proposals were submitted in October 2004, and the preferred bidder was selected in December 2004.  So, the process was completed reasonably speedily.


Mr Rogers: I am with you there, but it was not signed until nearly 18 months later.


Mr Peover: That was because we were in negotiation with the preferred bidder.  Fiona might want to comment.  She has personal knowledge of the issue.


Ms Fiona Hamill (Department of Finance and Personnel): I have personal experience.  I was the director of technical implementation for the project.  The reason for the delay between the original contract and the signing of the contract was the detailed description of the service and what would be delivered.  If you look at the Account NI contract, it runs to over 1,000 pages of very detailed specification for each module of the system.  Each bit of that was subject to clarification and negotiation between the Account NI development team and the contractor.  That is what took all that time.  We had to be very clear with the contractors exactly what we meant when we said that we wanted a purchasing system and a reporting system.  That explains, between selecting the bidder and getting to physically sign the contract, what was going on in that time.  It was the fine-tuning of that huge contract from the very detailed issues such as how many reports will be produced, how many days it will take and how frequently we will back things up.  That is what was going on in that period.


Mr Rogers: The fact was, and it has already been admitted, that the scope of the project changed over the period as well.  So, my interpretation is that the November 2004 contract was not what was signed in 2006.


Mr Wickens: As I understand it, there would have been an outline contract in 2004, which would have been part of the preferred bidder evaluation process.  The normal process after that is to get into the detail.  Some cases take longer than others.  This was a long one, given the detail, as Fiona said.


Mr Crosby: As I mentioned earlier, the changes are not substantive.  The ones that I am aware of are as follows:  extending the contract term to 12 years instead of 10 years; and adding in something that had not been there in the first place, responsibility for the legacy services that were with another company, the running down of those services, and the maintaining of those services until they were ready to cut over, and to keep those services live, in so far as people using the legacy systems needed them to complete their accounts and have their accounts audited.  These were all costs that increased during that period as we began to understand more of what was actually needed.


Mr Rogers: OK.  I will move on.  I am sure that somebody else can pick those things up.  Paragraph 2.12 states:


"DFP also removed a number of design elements on the basis of business need, for example, Project Accounting; Enterprise Planning and Budgeting; and Sales Order Processing."


How do you reconcile ensuring effective and efficient delivery of projects and value for money, and, at the same time, reviewing a project accounting module?


Mr Peover: Our customers did not need the system that we might have provided for them.  With regard to sales order processing, we do not have that many sales.  It was of no major value to the system, so there was no point in having modules that people did not need.


Mr Rogers: Surely, a project accounting module ensures that all costs are attached to a project and you know exactly where your money is going.  How did you ensure that?


Mr Crosby: That is absolutely right.  During the 2003 work on the outline business case — again, this predates me — it was envisaged that project accounting would solve the need for those parts of the organisation that needed to do what you just described.  When we looked at project accounting retrospectively, once we got the rest of the system in, we discovered that, first, it was impossible to retrofit and, secondly, that the off-the-shelf product was designed for a manufacturing business that wants to attract costs to projects.  That is not the way government accounting works in that responsibility for expenditure sits with the finance directorate and not with project managers.  So, it emerged down the line that there was not a perfect fit for the product and the business need.  We have other means in the system and in the chart of accounts to charge costs associated with projects to those projects, and that is what most people use.  As a result of that de-scoping, a credit was put back into the account to reverse that requirement out.


Mr Rogers: Was the Department of Finance and Personnel (DFP) given a particular budget within which to work that or was it like a blank chequebook?


Mr Peover: The cost in the full business case had to be justified.


Mr Rogers: If you leave out a module such as enterprise, planning and budgeting, how do you ensure that you are managing budgeting effectively and are forecasting effectively?


Mr Peover: Departments do that as part of the routine course of events.  That is part of our normal business in managing budgets, and Account NI supplies lots of financial information that helps finance directors manage their budgets more effectively.  I am not quite sure what the connection is.  The costs of the project set out in the full business case decided the cost envelope that we had to work within.  That is how we dealt with it.


Mr Easton: I notice that the removal of a module from the report resulted in 15 out of the 50 planned savings being unachievable.  What was the rationale for that?


Mr Peover: John has just described it in the sense that it was deemed not to be a particularly necessary element of our business and would have been difficult to fit into the way in which we do our business.  As John said, it is designed for a more commercial environment where there is a project and a project manager and you have the budget and deliver the project.  Our budgets are managed differently in the public sector in that they are managed through finance directors.  That was the rationale for it.


Mr Easton: Even though 15 were not achieved.


Mr Peover: There were 15 staff savings, yes.


Mr Easton: What was the result of those 15 not being achieved?


Mr Peover: The business case still stacked up.  They were not crucial to it.  The savings I described earlier were still made and are still being made.  We are still demonstrating a good performance against the costs in the business case.


Mr Easton: How does Account NI assess its performance against comparable organisers?


Mr Peover: We have difficulty with that because, as Paul described, we have sought to find good comparators for our performance.  We are unusual, if not necessarily a unique organisation, in that we cover 30 different organisations and provide services.  We interface with them, and they all have different requirements.  We manage 35 bank accounts and make payments on behalf of lots of different organisations from small to large.


Ideally, you want a directly analogous comparator so that you can say, "Here is an organisation that does more or less exactly the same job as we do and is doing it for half our cost or twice our cost".  That could be our comparison.  We would like to be in that position, and the report has given us the stimulus to pursue benchmarking ourselves and see if we can find sufficient detailed information to allow us to do Paul's apples and apples comparison rather than apples and chestnuts — I have never heard that comparison before.  You are right: those comparisons need to be done.  We do not have adequate comparisons, and we will be spurred on by the report to try to achieve them.


Mr Crosby: Since the report came out, I have approached the health and education sectors in particular and have got some high-level figures from them.  The only figures that they can give me at this time are direct staff costs, which come in at about £1·95, or just under £2·00, and that equates fairly readily with our £2·05, given that prompt payment in those sectors is nowhere near as good as ours. We certainly have plans to work with other people in Northern Ireland who provide a similar service in a similar context to Account NI.  A difficulty with some of the organisations is that they are single entities.  Some of the Northern Ireland comparators in the report are single entity organisations, which do not give a valid comparison with a shared service.


Mr Peover: Your point is correct.  We would like to have those comparisons because they would provide us with reassurance that we are performing our duties effectively.


Mr Easton: Mr Crosby, out of the 22 planned project benefits from Account NI, the Department claims that 16 were achieved.  Is that right?


Mr Crosby: Most of the benefits were planned in the original benefits realisation plan, and things have moved on since that plan was created.  We accept that there were a lot more benefits in the plan that there ought to have been and that they were not all capable of being managed in their own right.  One of the benefits in the plan, for example, was to improve staff morale.  It is very difficult to separate staff morale after the introduction of a shared service from the impact of the recession on pay restraint, pension restraint, and so on.


We have to accept the criticism that there were too many benefits.  However, if you go down the list of benefits in figure 8, most of them were delivered.


Mr Peover: The point that I should have made at the start is that we do not think that the benefits realisation plan is a good model.  It consisted of 22 items, some of which were measurable and some of which were not, and some of which were baselined and some of which were not.  Those items provide a context for the project and a rationale for doing it, but the idea that we might have measured 22 different things over a 10-year period was not a good one.  That is why the 22 items were refined down to the four key benefits.  I would not want to stand over the benefits realisation plan that was originally set out.  However, if you regard some items as intangible, ask yourself whether we have a more flexible system than we had previously.  If you ask the finance directors that, the answer will be yes.  Do we do things more effectively?  We could certainly not have done some things without the new systems.


There is an argument that the benefits were achieved, but it is very difficult to quantify them.  There was a long list of benefits that might have flowed from doing this:  most of them did flow, but I would be hard pushed to say that there was improvement of x% or y%.  They are not like that.


Mr Easton: Do you accept that, for 17 of the 22 non-monetary benefits, there was a lack of evidence to support the assertion that you achieved them?


Mr Peover: Yes, I do.  There is some evidence that they were achieved.  We are a business that serves customers, and our primary customers are the finance directors in the Departments and agencies.  We asked them whether they were basically content, and they said that, although there are problems, they are basically content.  We can probably stand over the fact that the benefits were achieved to some extent, but I would not stand over any attempt to measure them or say that they were hard-edged SMART targets that were quantifiable.


I will go back to Mr Donnelly's point about key performance indicators.  If you are looking at a project of this type, what you really want are four, five or half a dozen key performance indicators that are at the core of what you are trying to do.  You then assess your performance against those key performance indicators.  Twenty-two is excessive.  We could never have measured them, and it would probably never have been worth the effort as we would have had to have baselines and follow them right through.  With the benefit of hindsight, the right thing to do would have been to have half a dozen benefits and to have pursued them.


Mr Easton: So you should not have gone for 22.


Mr Peover: No.


Mr Wickens: Across all the shared services, we have tried to simplify by agreement all our governance structures to make sure that we are measuring only the things that make sense.


Mr Easton: Who decided that you should go for 22 benefits, which could not be achieved?


Mr Peover: I would need to check that.  I assume that, like most decisions in the public sector, it was made by a group or committee rather than by an individual.


Mr Wickens: They were part of the full business case benefits in 2006.  That would have been part of the overall business case.


Mr Copeland: So the business case, which was considered and accepted, included benefits indicators that were deemed to be unachievable?


Mr Peover: No; they are intangible and difficult to measure rather than unachievable.  I honestly think that some of them were achieved, but it is difficult to prove that because we do not have hard-edged measures.


Mr Copeland: The other issue, Chair — with your forbearance because it is perhaps slightly off track — is that the Civil Service is no different from any other large organisation.  It is a business, but it is not a business in the traditional sense.  Nothing frightens big organisations more than change.  In many ways, a cultural change could have been seen as desirable.  Did that have an impact on the culture of the Civil Service?  Was it negative or good?  More importantly, how was it assessed and measured?


Mr Peover: As I said, it was part of an overall programme of Civil Service reforms.  There was a lot of concern about the impact.  There is resistance to change in any organisation when there is a major reform programme.  Many of the 293 staff would have had to cease what they were doing — perhaps what they had been doing for many years.  On the HR side, it was even more radical because we were shifting functions from the Civil Service into the private sector and reducing the size of the retained HR functions in Departments.  So people were going to —


Mr Copeland: Were there any redundancies?


Mr Peover: No.  We redeploy staff rather than make them redundant.  Bear in mind that we have a system of 20,000-plus staff.  We have a turnover year on year from natural wastage, retirement and resignations of around 500 or 600 people — perhaps more than that — even in these times when people are hanging on to their jobs.  Very rarely is it necessary to make people redundant.  We can usually redeploy them.  If people have been doing a job and enjoyed doing it, and suddenly someone says, "Sorry.  We are not doing that any more.  We are handing that out to a contractor", or "We are centralising that in another organisation", it causes them concern on a personal level.  We did not have to make anybody redundant.  However, it still meant that people had to move out of their usual roles.


Mr Copeland: Salaries are always a primary cost driver.  If we have established that we did all that and that anybody who was previously employed to do that work, which is now being done by outside contractors, presumably at a cost, and account headings were just shifted about —


Mr Peover: What happens is that we have the same number of people, and we do things differently.  Those people can do other things that, hopefully, are useful, so we redeploy staff to more productive activity.  People used to do back-office work that is now done largely through automated systems.


We did not really talk about how we have managed to achieve further savings.  We have been streamlining our systems and looking at quality improvements in Account NI and the other systems, and trying to make ourselves more efficient.  Savings are being made in the system.  We are reducing costs and transferring them to other, hopefully more productive, activities.


Mr Wickens: I want to come back to your point about change and culture.  It was a big thing for the Northern Ireland Civil Service to move from separate accounting organisations towards a single organisation doing things on their behalf.  One way in which we tested that was with the gate 5 process, which is all about benefits realisation.  It is run independently and is chaired by somebody from outside Northern Ireland.  The gate 5 report recognised specifically that it is clear that the significant change in financial processing and the change in culture have been achieved successfully.  I think that that is also recognised in user satisfaction reports and in Audit Office surveys.  I think that they state that the finance directors are broadly satisfied with the way in which it is working for them today.


Mr Girvan: I know that a lot of emphasis is on what everyone thinks is wrong.  I measure a service on the feedback of those who actually use it.  Figure 10 shows that overall satisfaction goes from 42% in 2008 to 55% in 2012, so it is going in the right direction.  Do you see that as good feedback?  Almost half of the people who use it have had some difficulties or are perhaps not satisfied.


Mr Wickens: It is a very positive measure based on where it has moved from.  To give you some idea of the fact that we will never be content, certainly in my organisation, I have a team that is specifically responsible for user experience.  I have somebody whose title is "head of customer experience" and whose role is to understand, through independent surveys conducted by NISRA, what kind of feedback we are going to get and, on the back of that, drive change plans and change programmes.  For each of the shared services, including Account NI, we have a continuous improvement plan that specifically seeks users' views on the things that they do not like or the improvements that they want, and then that is fed in.  Some of those will inevitably have associated costs, and some of those costs will be too much so we will have to get on within our cost envelope.  However, we take that feedback very seriously and use it to drive further change.


Mr Peover: Mr Copeland made the point that people do not necessarily like change.  When you ask people whether they are happier now with a centrally managed system than when they ran their own show, they may be reluctant to say that it is better.  We are trying to provide commonality, consistency of approach and value for money.  We want to improve our customer feedback.


Mr Girvan: To ensure that you get a satisfactory report, there has to be accuracy in the processing.  Have you done benchmarking on the accuracy of the processing?  By that I mean that, if, for argument's sake, I send in an invoice for £15,554 and get paid £15,550, is that seen to be accurate?  Has there been any feedback on that aspect?


Mr Wickens: I think that that goes beyond user satisfaction and into audit scrutiny.  We are heavily audited on the basis of the very things that you are saying.  As a public sector organisation, it is incumbent on us to prove absolutely that we have accuracy.  We have a very strong audit trail in that respect.


Mr Dallat: I want to try to move the discussion on a bit.  We began by discussing the car that Stephen was going to buy.  Can we now assume that she is MOTed and ready for the road?  Will there be no more consultants?  Will it no longer be top-heavy with senior civil servants?  Will unit costs be lowered?  Will the 45 bodies outside main Government embrace Account NI so that economies of scale can be achieved?


Mr Peover: I do not know.  I am not the accounting officer for all those bodies.  If I can give an example —


Mr Dallat: Sorry, I know that it is rude to cross over you, but you probably underestimate the value that I place on you [Laughter.]


Mr Peover: Thank you.


Mr Dallat: You are the boss cat.  Do you not have any influence over those people at all?


Mr Peover: I hope that I have some influence —


Mr Dallat: Thank you.


Mr Peover: — but not in the sense of making decisions for them.  There is an issue here about —


Mr Dallat: Is it a problem that you do not have the power to drag people in and, as Paul said, break down the culture of writing with a quill and not embracing new technology, and so on?


Mr Peover: Hopefully, not many of them are writing with quills.  Some of the smaller organisations presumably have off-the-shelf accounting systems that, in a small organisation with a smallish budget and a few staff, is adequate for their needs.  They do not need the sophistication of a system that is designed for £2·8 billion of expenditure.  I do not think that we can prescribe that every organisation on that list of 45 should be a customer of Account NI.  I think that many of them should, and I encourage my colleagues to use their influence as accounting officers in their Departments to push people towards Account NI.  We have demonstrated success.  We successfully brought in the DVA.  John knows this well as he used to work there.  The DVA —


Mr Dallat: I hope that it is still there after next week.


Mr Peover: Yes.  The DVA was a difficult one as it involved two organisations:  the Driver and Vehicle Testing Agency (DVTA) and Driver and Vehicle Licensing Northern Ireland (DVLNI).  It had two accounting systems, and part of it, the DVTA, was a — what do you call it?


Mr Crosby: A trading fund.


Mr Peover: Yes, and it was the only one in Northern Ireland.  It also had two different platforms for its accounting systems.  So we brought in the DVA, which was a typically difficult one to do.  We also brought in the DOJ, which again was a big complex family of organisations, and the PPS.  We did that successfully and reduced the unit cost.


We think that we have a good service.  We think that it is worth people buying into it, and we would encourage our partners in Departments and organisations to buy into Account NI when it suits their needs.  They will have to make that decision.  If you have only 30 staff and a budget of £1 million a year, it may not be worth buying into a system as complex as ours.


Mr Wickens: Can I just add a little bit to that?


Mr Dallat: I find that very useful.  I accept that there are small organisations for which it is not worth it.  However, what can you do to make sure that those that are big enough embrace this — I was going to call it a "monster", but that would be grossly unfair — embrace Account NI?


We live in a small part of the world and cannot afford the luxury of setting Account NI up at a cost of £213 million, while others have the luxury of running their own wee department or spot and do not contribute to the economies of scale that it will achieve.  From today's session and the report that we will draw up — as you do not agree with part of the Audit Office report — could you help us with our report so that we do not have a repeat?


Mr Peover: I would very much welcome the Committee's encouragement to my colleagues to buy into Account NI.  Those that bought into it — the three that I mentioned — each did a business case to justify using the system rather than doing their own thing.  So each organisation would have to do a business case to satisfy itself that the cost of buying into Account NI is value for money.


As things stand, we think that the system delivers value for money.  We are delivering against the business case and the savings that we were tasked to make.  However, obviously, the more organisations that buy into our system, the better, and I certainly welcome the Committee's support of the wider extension of the Account NI system.


Mr Dallat: That is a good enough point for me to stop.  I will go home happy in the knowledge that these little homelands that have existed in the Civil Service are things of the past and that Mr Peover has our full support in exercising his influence and authority to make this thing work.


The Chairperson: Mr Copeland then Mr Hussey.


Mr Hussey: No, he has been like a budgie all afternoon. [Laughter.] Mr Dallat is miffed when you mention cars because, of course, he drives a Morris Minor Traveller, and you are not really having it.


Mr Dallat: It did not cost £231 million.


Mr Hussey: It may one day be worth that.


Mr Peover: My first car was a 1953 Morris Minor with a split windscreen.


Mr Hussey: But you two are a lot older than me.


Mr Dallat: I am glad that we have something in common.


Mr Peover: So am I.


Mr Hussey: Somebody may have asked this question, but why was there a 12-year period for the contract?  Where did the number 12 come from?  Why 12 years?


Mr Peover: That is a good question.  I assume that it was decided that it would deliver the best value for money over a 12-year period.  I would need to go back and look at the justification.


Mr Hussey: It is a very strange figure.  I have never heard of any contract being awarded for 12 years.  With the technology that we are dealing with, which is changing every day — I said that emails would not catch on, so I am the expert on this — I feel that a 12-year contract is very difficult to justify.


What Departments or groupings in the Northern Ireland Civil Service do not use Account NI?


Mr Wickens: Can I pick up on the first question and maybe move on to the second.  It is a 12-year contract.  I think that 10 years was the original term that was considered, but it was seen that it was better value for money to spread the cost over 12 years.  The contract was not for a static system, and it was not that we were putting in a system that we would use for the next 12 years.  In fact, in the past six months, we have been through a complete technical refresh of the whole infrastructure and the hardware and software that sits underneath.


Mr Hussey: I accept that, but —


Mr Wickens: That was a big part of it as well.


Mr Hussey: As technology moves forward, the provider that you use may not necessarily be the one that is most up to date.  I think that a 12-year period restricts that.  Although I am sure that BT is wonderful, there are others that may be taking steps ahead of them.  That is my concern about the 12-year contract.


Let us move on to who in the Northern Ireland Civil Service does not use Account NI.


Mr Peover: All the Departments use it and a number of the agencies —


Mr Crosby: All the agencies —


Mr Peover: All the agencies and a number of other —


Mr Wickens: All Account NI customers are listed in appendix 3.


Mr Crosby: The NDPBs not using Account NI are listed in appendix 4.


Mr Hussey: I am aware of that, but I am asking whether anybody does not use it.  The list clearly shows us who uses it, but I want to know whether we are aware of anybody who does not use it.


Mr Peover: The only one that left was AFBI.


Mr Hussey: Who?


Mr Peover: The Agri-Food and Biosciences Institute.  It left on the basis that it had a different set of needs because it is more commercial.  It sells its services and attracts costs.  All the agencies are in.


Mr Hussey: In the review of public administration, we are clearly going to do away with quite a few councils and go down to 11.  I sat on various committees about that.  Does Account NI have any intentions or hopes that, when the RPA happens, perhaps councils will use Account NI?  Have any of the councils even suggested that they are interested?


Mr Peover: There are issues about the scope.  When the original project was tendered, it was for the core system.  Councils and health bodies are outside the scope of the contract, and we could not extend it to them without running the risk of challenge.  The contract is due for renewal in 2018, and we have already begun the process of looking at how we get from here to a new contract.  I reassure you that we will take advantage of technological advances between now and then, and we will certainly look at the scope.  There are, however, limitations about how far we can extend Account NI's coverage.


Mr Wickens: As for the process for getting more bodies on board, I completely concur with the fact that there is an opportunity for the wider public sector in Northern Ireland.  We can get better economies of scale for Account NI, deliver better value for money and get better visibility of costing for scrutiny purposes.  Everybody accepts that it is a good thing to do.  There are procurement obstacles to prevent our going to organisations such as councils.  In appendix 4, another bunch of some 45 organisations is listed, some of which are having active conversations with their parent Departments and saying, "Do you not think that you should be encouraging this organisation to join?"


In the next few days, I have a meeting with one of the smaller Departments, which has a number of arm's-length bodies and NDPBs, to have that very discussion.  I had a discussion with one of the bigger departmental boards yesterday on the same topic of how we can get its set of arm's-length bodies and NDPBs to join Account NI.  It is an ongoing, active process, and we welcome the Committee's encouragement.


Mr Hussey: I find it difficult to accept why we have not had a service designed for Northern Ireland for all our government — local and regional — and why computers in district councils cannot speak to computers in the Departments that they deal with regularly.  However, that is an argument for another day, but it concerns me.  I am still not happy with the answer on the number of staff savings.  It seems that there have been no savings.  There was to have been a reduction in staff, but we still have the same number of staff.  Is that the case, or did I misunderstand what you were saying?


Mr Peover: It depends on how you look at it.


Mr Hussey: It either is or it is not.  That is how you look at it.


Mr Peover: There were 293 staff:  198 of those transferred to Account NI, 29 were retained in Departments, and 66 posts were saved.  Those are the basic figures.  I did philosophy at university, and in philosophy, there are counterfactual conditionals.  If nothing had changed, we would at least have achieved that, and we have, in fact, achieved more than that.  There was a further reduction of 40 staff in Account NI, which would have brought us down to 158 from 198.  So we would have saved the 66 who were in the original numbers and saved the 40 against that 158.  In fact, it is back to apples and apples.  If you try to get apples and apples and say, "Where do we stand today, leaving aside prompt payment and the other bodies that we have brought in such as the DOJ and the PPS?"  If we leave those aside, we think that we have 132 staff in Account NI as opposed to the 198 that we started with.  So we think that we have saved those staff.  Some of them have been reinvested, because we have put staff back in to deal with prompt payment and those other bodies being brought on board.  Like for like, staff numbers have gone down from 198 to 132.  That is our view of the figures.


Mr Hussey: Being very philosophical, that went right over my head.  We have not really saved anything.  It is not just apples and apples; you are putting in apples, pears, grapes and everything else.  The answer is that either we have reduced the number of staff or we have not.


Mr Peover: The answer is that we have reduced the number of staff.


Mr Hussey: Is it in line with what you suggested in the business case?


Mr Peover: We have exceeded the business case.  If you take the additional issues such as the new departments that were taken on and leave aside prompt payment, we have exceeded the target set for us in the business case.


Mr Wickens: We are happy to write to the Committee to break the figures down, because it is difficult to get across in a meeting.  There was a target of 158 staff — this is purely on a like-for-like basis — in Account NI.  We are sitting today at 132 on a like-for-like basis.  We have lots more staff than that because we have decided to do lots of extra things.  We have been asked to do lots of extra things.  Therefore, we have effectively reused the staff savings we made to do other things.  We have netted off 26 additional savings beyond our target in the full business case.  If you compare that on the departmental side, they are about five down once you do a netting-off, like-for-like comparison.  We firmly believe that we have overachieved.  We are happy to break that down and write to the Committee to show how we work that out.


Mr Hussey: They say that seeing is believing.  I would be happy to see that.  We will see where we go from there.


Mr Peover: We will certainly do that.


The Chairperson: In 2006, the final business case predicted that 72 members of staff would be released from the Department to take the project forward to the implementation phase.  The total released at that time was 42.  The Audit Office found that a number of individual consultancy contracts were properly approved and managed, but a number of consultancy requests at that time did not seek DFP supply or ministerial approval.  Obviously, you would look for the right levels of authorisation, and you would want to cover that.  The reason cited in the revised business case in 2011 was that in-house staff were unavailable.  How was that the case?  Will you clarify that?


Mr Peover: Departments were reluctant to release staff.  It is like anything else:  we were looking to them to give us good, well-motivated staff to help us to introduce Account NI.  They wanted those good, well-motivated staff in their own organisations to carry on delivering the legacy systems and do whatever they wanted to do.  We were effectively coming with a begging bowl and asking for resources.  To be fair, we got 42 out of the 72, but we did not get the 72 whom we thought that we needed.  That is why.


The report acknowledges the fact that Departments are understandably reluctant to give up staff to help a new project.  They have their own priorities and pressures and simply do not want to give away resources.  That is why we had to buy in consultants.


The Chairperson: Why did some of the consultancy requests not seek DFP supply or ministerial approval?


Mr Peover: That was the subject of a previous report.  We talked about this issue during the review of consultancy.  There may be some exceptions, but, in the main, those things were approved.  Some of them may have been approved retrospectively, but they were approved.  Ministerial and DFP supply approval were received.  I need to check to be absolutely sure.  The Audit Office report suggests that some of them were not but most, if not all, were.  I think that I acknowledged previously that nobody wants consultancy to increase from £973,000 to £9·6 million, £9·7 million or whatever.  However, that was required because of the inability of Departments to support the project to the extent that it needed to be supported to allow us to do it.  The positive point is that, since 2009, we have not had to employ consultants.  We have carried the process forward and skilled up our own staff.  There has been a transfer of skills, which is also acknowledged in the report.  We are managing the contract and bringing on new bodies without consultancy support, which is desirable.  My Department's consultancy has dropped dramatically to a few hundred thousand pounds rather than millions of pounds, which it might have been in the past.


The Chairperson: Before I make my concluding remarks, does any other member wish to speak?


Mr Rogers: It is interesting that, on the basis of that, you needed 72 and were able to get 42.  How have you taken on board, for example, the PAC report from April 2012 on the use of consultants and the 2009 report, which recommended:


"departments should always undertake a review of their capability to carry through ... projects. Those plans should consider and address the skills and resources available"




"properly identify the need for consultants to fill any ... gaps."


When you meet the C&AG and agree, partly agree or do not agree a report, how are those recommendations filtered down to the Departments to the level of SMART targets in Departments to achieve the recommendations?


Mr Peover: As a prologue to that, this was a new departure for us in engaging with the Audit Office as part of this project.  It was very helpful and talked to us at various stages through the work, and we had a very constructive engagement with it.  If I have appeared to be critical, I am not; it was a very good process and a model for how these processes should be done in the future.  That was a very positive development.


You asked about the dissemination of lessons.  The first part of that is the memorandum of reply, which is drawn up by the relevant Department and comes back to the Committee, giving the Minister's responses to the recommendations that the Committee brings forward.  Those are disseminated to Departments, and Departments get to see the memorandum of reply.  Quite often, a system is created.  Let us take the consultancy example: in DFP, we now get annual returns from Departments on their consultancy spend so that we have overall sight of what is being spent.  We also do a bit of analysis.  We do a bit of deep diving; we ask them for the documentation in relation to some of the specific projects that they have let, to see whether they did a business case and whether they had ministerial approval, supply approval and so on.  We do checking.  Not only do we trust, we check.  That is how we ensure that the lessons are carried forward from the comments and recommendations that come from the Committee and from other sources.


Mr Rogers: Say, for example, that they are way over budget for consultants when you review that.  Would you, then, require them to sharpen up their targets for the following year in those areas?


Mr Peover: That is a matter for Ministers.  A Minister in a Department has control of his or her own budget, and it is for them to decide how they want that budget spent.  We would check the procedural side of it to see whether they have ministerial approval, whether they have done a proper business case, whether they have the documentation to show that the work that they commissioned is justified, whether they have done a post-project evaluation and whether, hopefully, they have had some transfer of learning to their staff as a result of the process.  We cannot intrude on a Minister's decision on how to spend his or her own budget.  If a Department seeks approval from its Minister to do something and gets it, that is the decision.


Mr Rogers: Yes, but after all, you are the accounting officer.


Mr Peover: Not for that Department.  If the Health Department or the Department of Culture, Arts and Leisure, for example, want to engage consultants to do work for them, their accounting officer has to justify that decision.  We would check the process to ensure that they have gone through the required stages.  We would require the information in order to keep a bit of a challenge in the system.  At the end of the day, it is not for me to tell the Culture Minister, the Agriculture Minister or the Enterprise Minister that they cannot do something.  We work for Ministers, and it is their decision.  They and their accounting officers have to account for it.  They will be called before this Committee to account for themselves, if you think they have wasted money doing a particular piece of work.


Mr Dallat: Chairperson, I promised faithfully not to open my mouth again, but Stephen really does drag it out of me.  Recently, one Minister took another Minister to court over a financial transaction.  You are Mr Moneybags; you hold the money for the Assembly.  Are you telling me that you have no influence or control over Departments that, I think, outrageously refuse to become part of an organisation that was set up at a cost of £231 million?  Is that not the tail wagging the dog?  I do not want to make such comparisons, but —


Mr Peover: I am touched by your faith in my influence.


Mr Dallat: You have to be: I do not want to go home tonight feeling that you have no control over the Budget of the Assembly.


Mr Peover: I am not going to go into a description of the process.  Ultimately, the Executive decide on the Budget.  The Executive allocate money to Departments, and it is for the Departments and Ministers to decide on how they want to allocate their resources.  If there is a cross-cutting issue, they may need to come to the Executive or DFP for approval.  I do not want to get into specifics.  At the end of the day, if the accounting officer in DCAL or DETI decides that they need some consultancy support and their Minister agrees with them, provided they go through the proper processes, we in DFP do not say, "No, you can't do it".  That is not our role.  We can influence the overall allocation of resources, and we do that through the Finance Minister working with his Executive colleagues.  Unfortunately, I do not have the power to stop some of these people doing some things.


Mr Dallat: Maybe "power" is the wrong word, but you are bound to have massive influence to change what Paul said was a culture of writing with quills.


Mr Peover: You and the Chair know — I have said this at sessions that we have had — that we regard the Committee as an ally and an asset to us in trying to enforce good standards of management of public money throughout the system.  That is our role, and that is supported by the Assembly and the PAC.  Therefore the points we made earlier about you lending weight to our role in this and providing recommendations that we can point to and say to our colleagues, "We should be doing x rather than y" is helpful.  Constitutionally, I am not capable of intervening in the Budget management of any of the other Ministers and saying they should not do what they are doing.


The Chairperson: Are you satisfied, Deputy Chairperson?  Mr Easton, you wanted to —


Mr Easton: Finish my questions, yes.  Mr Crosby, Northern Ireland Water, Belfast City Council, the Northern Ireland Housing Executive and Invest NI seem to perform better than Account NI, from the figures that we have.  What are they doing that you are not?


Mr Crosby: That is a very good question.  This information came to us fairly late, as the report was published.  I have gone back to all four to ask them to explain the constitution of their figures.  Only two of them — Belfast City Council and Invest NI — have come back to me.  Both of them have said, "If you look at all of the things that are in your figure of £9·73, we do not include the half of that; in fact, we do not include most of it".  The comparison that we have got back from them is that their direct cost of staff per transaction is much higher than ours.


Mr Easton: So, are you saying that we are being misled?


Mr Crosby: No, I am not saying that.  I am saying what they told me when I asked them.


Mr Peover: It illustrates my point, Mr Easton, that I place no more credence in those figures than I do in the overall benchmarking.  The real issue is that we do not know the proper break-up of these figures.  They tell us things that may be true or partially true.  We need to sit down with them and, line by line, go through how they allocate resources against their spend.  The information that we have is that they seem to be in the same ballpark as us, but we do not place any great credence in that either.  I would not want to rely on it.


Mr Easton: So, the figures that we have got may be misleading.


Mr Peover: It is the point that I tried to make earlier.  We have £9·73, which is made up of a lot of things.  It is our notional costing.  We would not charge people on this basis, but, notionally, that is our charge per transaction.  It includes lots of things that may or may not be relevant and which others may or may not include.  The real point is the point that we made with the C&AG, which is that they have kicked this debate off.  We think that it is a good debate, and it is the right debate.  We need to benchmark properly, and we need to satisfy ourselves either that we are way out of line with everybody else, which seems unlikely given our performance in other areas, or that we are not counting the same things.  If we are not counting the same things, we need to count the same things.


Mr Easton: So, you will stay in contact with the others and try to resolve how people get these figures.


Mr Wickens: We have kicked off a formal benchmarking project across Enterprise Shared Services.  There are parts of ESS where we do fairly good benchmarking and on a cost basis.  We want to make sure that we do it consistently.  A fairly substantive project has just been kicked off on all of that.


Mr Crosby: In going forward, my emphasis is more likely to be on the health and education sectors to provide a more relevant comparison because of their scale, size and cross-functionality.


Mr Easton: Mine, too.  I would be keen to get to the bottom of why the figures we are getting are somewhat different to yours.


Mr Crosby: Yes, so am I.


Mr Easton: You are using key performance —


The Chairperson: I think that Mr Donnelly wanted to come in there.


Mr Donnelly: Yes, Chair, those were the best figures available when we were constructing the report.  In the absence of any other figures, it is a starting point of the debate, but it is not the end point.  The main objective is to open up a debate on costs because there is nothing happening there at all.


Mr Crosby: I appreciate that that is the position, but, once you drill below the surface, you find more questions than answers.


Mr Easton: You are using key performance indicators now.  You have 10 KPIs.  Why were you not using them beforehand?


Mr Crosby: I beg your pardon?


Mr Easton: Were you using them before 2011?


Mr Crosby: Yes, we have been using them since we went completely live in 2009.  In 2011, we had more than 10 KPIs, and we had a discussion with the finance services board about whether that was the right number and whether they were the right ones.  We agreed at that meeting that the board wanted to monitor no more than 10 and that these were the 10 that it wanted to monitor.  That is why there was a change at that time.  It was simply on the presentation and who would review the figures.


Mr Easton: Why cannot  Account NI streamline its processes to a level that matches top-performing competitors to achieve savings?


Mr Crosby: That comes back to whether we accept that the benchmarks we have been provided with are valid.  Certainly, on the non-financial ones, Account NI performs as well or better than most others.  Our prompt payment performance is exemplary, and the Audit Office has acknowledged that across other parts of the public sector in Northern Ireland.  When you compare it with the private sector, it is not a requirement that they have to meet.  They more normally pay in 60 days, as those of you who are connected with that will know.


Mr Wickens: If you compare with one or two of the others, one point that is often debated is the time to close the general ledger, which looks way out of kilter with the rest of the world.  There is a simple answer to that, but first, the private sector is trying to manage its cash and close its accounts as quickly as possible.  World class is T+2, where T is today and they try to close their accounts within two or three days.  The best that we can get agreement on from all of our stakeholders — remember that we provide a service to their requirements — is nine days.  We can be much quicker than that, but we provide the service on the basis of their demand.  Some of these things do not stand on their own.  It is not a matter of saying that somebody else can do it in this time, so we should also be able to do it in that time.  It is based on the requirements of our customers and stakeholders.


Mr Easton: Is there any reason why Account NI has not monitored wider departmental finance staff numbers?  The number of finance staff in NICS has increased substantially from 1,239 in 2006 to 3,221.  That is an increase of 160%, yet the response to the NIAO survey suggests that staff numbers have decreased from 916 to 586.  Why is that?  Does nobody actually keep proper account of staff numbers?


Mr Peover: It depends on how you count these things.  To answer your first point on whether we monitor them, the answer is "No, we do not".  The view was taken that the retained finance functions were a matter for Departments.  They needed to decide what they needed for their own purposes.  I think that that is dubious, personally, but that is the line that was taken. 


Finance staff do lots of other things: budgeting, financial control, support to boards, governance, corporate planning and business planning.  We have professionalised our finance function over the last 10 or 15 years.  It would have been exceptional to have properly qualified accountants in our system.  Now, since we have modernised our accounting systems, we have properly qualified accountants.  Fiona knows more about that than many of us. 


We have changed the nature of our finance function.  The bit that we are talking about in Account NI is not that range of other things.  There are things that Departments will want to do within their own ambit.  We provide that shared service to enable them to get the transaction side of things and the processing out of the way, and we think that we have done that effectively.  If you were asking me personally whether I thought that we should look hard at the way retained finance functions are staffed, the answer would be yes.  We are doing a bit of work on that already.  However, at the end of the day, accounting officers and Departments have to satisfy themselves that their staffing is appropriate to their needs.  If we were to say to them, "You have 10 staff; you could do with five" and they make do with five and the thing falls over, it would be our fault rather than theirs.  The accounting officer is the person who has to come here and account to you and the Assembly for the performance of his or her Department — sorry, they are all "his" now.  That is a heavy personal responsibility on accounting officers, and they need to be absolutely sure that they have got the systems and processes in place that satisfy them that they are delivering the services appropriately.


Mr Easton: But you do agree —


Mr Peover: I agree with you.  We should rationalise and streamline retained finance functions.


Mr Easton: Yes, and we should know how many staff we have.


Mr Peover: Yes.  It depends on the question you ask.  If you ask people one question, you get one answer; if you ask them another question, you get a different answer.


Ms Hamill: I want to clarify a point.  In relation to the apparent increase in numbers from 1,239 to 3,222 — 160% increase — unfortunately that is another instance of the basis of the calculation.  We complete a survey for HM Treasury every year.  Unfortunately, every year, HM Treasury changes the basis of how it wants figures reported between those who are directly related to finance and staff who have only a small engagement with finance.  They are not figures that we in the NICS have ever audited or proofed ourselves.  That is why the figures have increased so dramatically during the period.  It is just the basis on which HM Treasury has asked us to report staff engagement in financial transactions.  It is not any dramatic movement in staff numbers within the NICS.  They are unaudited figures.


Mr Girvan: I want to come back to the point about not comparing apples with apples in relation to the cost.  If we believe that we are there or thereabouts — I will accept that — there should be an operation where we go out to actively attract other Departments to come on board with this, and that goes back to a point that John made.  I am one of those who say that, if you can actually deliver it cheaper and charge for the service, the Department will make the saving for that.  I know that that is not a practice in the Civil Service, but, coming from the private sector, I look at it that way, and I think that everything should stand on its own.  If you can get it more effectively and if you can buy that service in and do it by yourselves, still as part of the Northern Ireland Civil Service but delivering it more efficiently than it is being delivered at the minute with each Department having its own people to deal with payments and everything else, you would gain.  There is merit in ensuring that everybody uses the same process and measures like with like.  We should try to sell that as a way forward.


Mr Peover: To some extent, we do that.  When we brought on the DOJ, PPS and so on, that was at a cost.  They justified it against a business case, and, to them, it was a more efficient way of doing it.  They could make savings and do it more effectively by coming within our system.  They justified it to themselves, and we justified it to ourselves by saying that we would do it at what it would cost us.  The process that you just described does work. 


There is a separate issue about hard charging.  The charging for Account NI has been problematic for a long time, and we had a lot of difficulty with Departments at the outset in trying to apportion costs.  If you think about it, we are effectively providing the same service, on a different scale obviously, to a Department such as DCAL with 420 staff as we are to DSD with 8,000 staff.  How do you charge in a situation like that?  Do you charge them all the same unit cost?  How do you manage that process?  There are issues about charging.  There is a reference in the report to hard charging and a preference on the part of the Audit Office that hard charging is a good discipline because it inculcates the behaviours that you just talked about, such as looking for value for money, driving down costs and trying to find the most effective and most efficient way of delivering a service.  When you work in our system, it requires you to create a bit of a bureaucracy because you have to estimate your costs, charge those out and issue bills and so on to your customers.  There is an issue about how you do that.  It is a policy issue that Ministers have to decide case by case, but we note the Audit Office's view that it is a good thing.


Mr Girvan: It is a good thing.  Through your charges, you are making a calculated decision on all your costs that you have to run that.  It is not just a bottomless pit.


Mr Peover: In a sense, we have brought the costs into Account NI.  We can afford to run it because we have taken the costs from Departments.  Effectively, we have charged them.  We now hold those resources and provide services.


Mr Girvan: I would like to see where you can identify that when you cross-check from one Department or body's accounts and it shows the reduction in what they are drawing down.  You rarely hear about Departments' moneys and the amount that they get.  They are always bidding for more.


Mr Peover: To put it the other way round, in our system in Northern Ireland, we do not raise our own taxes but get a block grant.  If £200 million is allocated to body X, it is in body X's interests to spend that £200 million for the benefit of the population of Northern Ireland.  If it spends £10 million of that on staff, that is £10 million lost to the public.  If it spends £8 million, that is £2 million more going on a public service rather than on staffing and so on.  It works in our system.


Mr Girvan: Unless you come from a trade union background where you think the right way to do it is to spend it all.


Mr Peover: That is our system.  We are allocated a block of money by the UK Government, and our job is to spend that effectively in providing services to our population.  The more we spend directly on services to the public and the less we spend on back-office services, the better.  The money should go to hospitals, roads, houses, nurses, doctors and so on.


Mr Wickens: We are effectively a mandated monopoly, and the Committee seems to be leading that way and saying that it wants us to be a bigger monopoly and do more for more organisations.


Mr Girvan: We are trying to get people to use the approaches that have been put in.  We have put PEDU in place, and we want to ensure that people buy in to make sure of the shared services and, in doing so, get more efficient delivery overall for less money.


Mr Wickens: That is what I was going on to say.  In ESS, I have a savings plan year on year to reduce my overall budget, and we have been meeting that so far.  Because the funding that is in place is, effectively, centrally allocated, if we were to hard-charge, I would have to send out an invoice to individual organisations with a cheque attached from my budget to theirs for them to pay me back again.  They would say, "I am not giving you all that back.  I will only give you 95% back".  So, we would then have to create a debt management unit, which is where we were when we created Enterprise Shared Services in the first place and went for the central funding allocation.  So, in a mandated monopoly, there is a slightly different argument that transparency of charging and costing is perhaps different from actual hard charging, but that one will run and run for some time yet as we continue to look at it.


Mr Girvan: I do not mean that cheques have to be transferring.


Mr Peover: We can certainly have notional charging.


Mr Girvan: Exactly.  That is the way I see it.


The Chairperson: Mr Peover, thank you.  In summing up, the key objective of Account NI and why it was set up was to simplify and rationalise the whole financial process and provide a common accounting system across the NICS.  So, in doing that, it is bringing things back into economies of scale.  I am sure that lessons have been learned from the debate today, but what procedures will now be put in place or are planned to be implemented in the light of the Audit Office report on Account NI?


Mr Peover: We will wait for the Committee's report and recommendations, which I assume will reflect at least some and maybe all of the Audit Office recommendations.  The two big points that we talked about today are benchmarking and the creation of a key performance indicator for unit cost, and we will look at those issues very quickly.  That will take that away.  When we see your report, we will have a look at the recommendations and respond to them.


The Chairperson: The report shows a slight increase in customer satisfaction, and there was a particular focus on that in the debate today.  I thank the auditor and the Audit Office, which has been acknowledged by you, Mr Peover, for flagging up the benchmarking issue.


Mr Peover: It was the first time that the Audit Office has engaged in the sort of process that we went through with it on the report, and it has been a very helpful process.


The Chairperson: That is to be welcomed, and it shows that we work effectively when we work together.  Thank you, Mr Peover, Mr Wickens and Mr Crosby for you time.  We may want to write to you for further information when the Committee does its debrief.  Thank you, members, Mr Donnelly, Ms Hamill and Hansard.

Find MLAs

Find your MLAs

Locate MLAs


News and Media Centre

Visit the News and Media Centre

Read press releases, watch live and archived video

Find out more

Follow the Assembly

Follow the Assembly on our social media channels

Keep up-to-date with the Assembly

Find out more

Useful Contacts

Contact us

Contacts for different parts of the Assembly

Contact Us