Most attempts to share services in Whitehall have been pretty disastrous. Joshua Chambers meets Paul Marriner, the man charged with demonstrating that shared services schemes don’t have to be expensive white elephants
“It’s my first interview, be gentle with me,” jokes Paul Marriner, as I sit down and switch on my dictaphone. I have no intention of being rude, but he does need to answer some tough questions. After all, as executive director of shared services in the Cabinet Office, he’s working on a vast, pan-Whitehall project in a field where government has had little previous success.
Indeed, in the main, departments’ past efforts to share back-office services have met with failure. A string of reports by the National Audit Office (NAO) show that most such projects have ended up spending far more than they ever save. Just last year the NAO analysed five shared service centres, finding that while they took seven years to develop and cost the government £1.4bn – including an overspend of £0.5bn – only one of them had broken even. None of them had managed to actually save the taxpayer any money at all.
Marriner represents a fresh start for government shared services, in more ways than one. Unlike many senior civil servants, he left school at 16, working his way up in finance and accountancy before becoming an expert on shared services in the private sector. In 2011, he was seconded into government to help it learn the lessons of previous public sector shared service programmes; he’s now employed as a civil servant to work on the latest iteration of Whitehall shared services.
The services that departments are trying to share are those dubbed “back office” functions, meaning the regular transactions that keep departments ticking over: payment of salaries, automated procurement systems, and general accountancy processes such as keeping books and ensuring invoices get paid.
“It’s standard stuff, there’s nothing new,” Marriner says. Departments have tried to share these services before; and there’s little logic in duplicating very similar processes in separate centres, he argues, when departments’ resources could be pooled and costs reduced through economies of scale.
Making a case
The coalition wasn’t put off by Whitehall’s past failures in the field, and in July 2011 published a strategic vision calling for all departments to join some form of shared service. The Next Generation Shared Services Strategy, published last December, filled out the gaps, providing timescales, benchmarks and risk assessments.
“Very significant savings” can be made by pursuing this strategy, the Civil Service Reform Plan claims. Meanwhile, the strategy says that if savings lie within the upper range of the benchmarks set for them, the new approach could deliver savings of between £400m and £600m a year. But given government’s track record, what’s the evidence that this is possible in the public sector, with its complex departmental structure, dispersed decision-making powers, and differing terms and conditions? “There’s evidence building all the time,” Marriner responds, citing shared service centres run by the Ministry of Defence and the Department for Business, Innovation and Skills; he claims that since the NAO’s last report, both have managed to stabilise their programmes and show progress. “They’ve got the base to work on and drive out costs,” Marriner says: while the NAO hasn’t publicly released anything on shared services since its critical 2012 report, he says, it “agrees that things are moving forward now”. The MoD shared service is established on a risk-reward basis with a private sector contractor, he adds, so the contractor took the initial hit and the centre is now saving money.
However, the NAO also said last year that the Cabinet Office didn’t have a “sufficient grip on the cost of [government’s] activities to promote shared services”. If the Cabinet Office is going to persuade departments to spend more money on this approach, does it now have a better understanding of how much is currently being spent on back-office services? “Absolutely,” he responds. “The work we’ve done, which has culminated in the publication of the strategy, has taken a long, hard look at the cost of operations across all the shared service centres, and across departments as a whole. We believe we understand the costs; and when we procure solutions and operators, we know what we’re buying.” Has the Cabinet Office also factored in government’s previous optimism bias when it comes to sharing services? “We have looked at [the sums] very hard, and I think we have been cautious in our targets,” he says.
Proposing a plan
The plan envisages two large, independent centres which departments will commission to run their back office services: Independent Shared Services Centres one (ISSC1) and two (ISSC2). It’s intended that ISSC1 will serve the smaller regiments, such as the Department for Culture, Media and Sport. Meanwhile, ISSC2 will be geared towards the armies of employees at the Department for Work & Pensions and the other larger departments.
There are also a few stand-alone centres that won’t form part of ISSC1 or 2, either because departments have particular needs – the Ministry of Defence has specialist security requirements, for instance – or because these centres are already close to achieving the desired results. The Ministry of Justice, HMRC, and Department of Health centres are going to remain separate because it’s likely they will achieve savings on their own, but Marriner hopes that all other departments will join the ISSCs.
What’s happening so far
ISSC1 – the centre for small departments – will be built out of the Department for Transport’s existing shared services centre. This centre has an abysmal reputation and hasn’t ever saved money. It is currently run by civil servants, but the plan is to outsource the whole operation to the private sector. As Marriner says, “it was costing an awful lot of money previously. That is the reason why DfT made its decision to divest: it was too expensive”. The outsourcing will be completed by March, and should “significantly change the cost of services without impacting on quality,” he adds.
The transport department won’t thereafter have any control over the service; it will simply become one of the centre’s customers. “One of the learnings of my strategy – our strategy – is that [the centre’s] independence is quite important to people joining it,” Marriner says. “People don’t like the thought of going into services run by a big department because they feel that the priority will be on the big department, and not on them as an equal customer.”
The plan for ISSC2 is still being written – although it’s known that it will be built out of the Department for Work and Pensions’ centre, which is currently run by civil servants. Detailed plans will be submitted to the Treasury in March. “We’re looking to go to the market, because we do believe we want an operator,” says Marriner. “How that operator works – whether that’s a joint venture or a strategic partnership, or we buy in an operator for a set period of transformation – is still be to be decided.” The DWP centre is now under the executive management of the Cabinet Office, and – given Treasury approval – the Cabinet Office intends to approach the market in April.
Still to come
The strategy’s timescale is very ambitious, as both the NAO and the Cabinet Office itself have noted. Following the publication of the initial vision and consultation with departmental permanent secretaries, the strategy was extended by six months: the final migrations into shared services are now scheduled for 2015, and will – the strategy says – be continuously monitored for progress. Meanwhile, Marriner says “there’s always flexibility, but we believe that a longer timescale brings more risk into the programme; and if you know what you want to do, if you come to the right conclusions in terms of options, and if you prepare properly and execute that as quickly as you can, then you de-risk the programme”. He believes that “the time for talking is over. We need to move forward now and get this done. All ministers have agreed it; all ministerial departments have agreed this strategy – so now we just need to get on with it”.
Moving fast “de-risks” the process because “you take out any indecision”, Marriner explains. “You don’t allow people to think about [decisions] any more than they’ve already thought about them. This strategy’s been thought about for 12 months. It’s gone through a lot of rigour; it’s gone through a lot of sharing with departments and taking their input in. It’s now time to get on with it.”
However, while departments have committed to the broad strategy – they have accepted that shared services are a good idea – they haven’t, in fact, promised to join ISSC1 or ISSC2. “They’re all in for a planning assumption that they would go there, but we still have work to do with them. We still have to go through the business case with them,” Marriner says. Didn’t he just minutes ago (or two paragraphs above) say that departments have agreed the strategy? “They’ve agreed the strategy that two independent centres are the right thing to do,” he replies carefully.
Let the games begin
The fact is that no department can sign up to the strategy yet, because the price of joining a shared service centre hasn’t been determined. Only once ISSC1 has been outsourced will the Cabinet Office know the prices for departments using the centre. “We were in a chicken and egg situation last year where departments were keen to join, but until we can do a business case with prices, you can’t sign up,” Marriner says.
Meanwhile, many departments pencilled in to join ISSC2 already have their own independent centres. “The only ones that don’t are Cabinet Office and the Department for Education, who are currently customers of the Department for Work & Pensions; and the Department of Energy & Climate Change, which is a customer of the Defra shared service centre. So apart from those, they are all in shared services at the moment,” says Marriner.
Doesn’t this mean that departments will have to close down their existing shared service centres to move into ISSC2 – which is essentially the DWP centre? “No, this is not a DWP takeover,” he responds. “What DWP brings is scale – a huge amount of volume – because they are a big employing department. We are completely open as to how these centres are built. So an operator, whoever we pick, could just as easily build it out of the Department for Business, Innovation and Skills’ centre as they could build it out of the DWP centre – as long as the volume is in place. They may decide that the Swindon office is better than the Cardiff office, for instance. Or they may say there’s room for both.”
Surely, though, BIS could just stick with its own shared services centre? After all, it has sunk a large amount of money into it. In 2011, the NAO looked into the Research Councils’ shared service – owned by BIS – and found that it went £51m over its £79m implementation budget, underachieved against expected savings by at least £73m, and lost £13m on terminating a dysfunctional IT contract. Since then, Marriner says, BIS has expanded the centre, and is “actually making that a cheaper and better-run shared services centre”. Why, then, would they move? “This is not a mandated programme. But what we do have as part of the programme is the Crown Oversight Function, which is there to drive and benchmark performance,” he replies, “all departments must meet benchmarks, and if they don’t hit those benchmarks, then they will have to join a shared service centre.”
So unless the BIS centre starts to meet benchmark levels – sourcing back office services at costs below those central departments can typically achieve – its customers will have to join another shared service centre. However, that doesn’t mean they’ll have to join one of Marriner’s two independent centres, he admits. “They could join the MoD; they could join the MoJ... ISSC1 and 2 are there to pick up those people that do not really have major [shared service] programmes.”
Marriner is negotiating with departments at the moment – and while departmental accounting officers won’t sign up to ISSC1 or ISSC2 unless they clearly offer the best value for money, he’s confident that the projects’ scale will produce the lowest possible prices. “How can you justify continuing to pay more?”, he asks. “If people are proven to be getting a good service at a particular price – which is lower than the price you can achieve, mainly because of scale – then why would you stay on your own?”
However, he accepts there may be complications for some departments: “They may have very long [back office service provision] contracts they have to look at. If they’ve got a long system contract, for instance, it may be very expensive to get out of it, so how do they do that? Every department will be in a different position, but what accounting officers need to think of is the wider public sector value for money test.”
Buying problems, solutions
Many departments are in a good position, Marriner says, because “some of the bigger contracts are coming to an end in the next year or so.” And while some departments have won exemptions, Marriner believes the majority should join one of the two common IT platforms that will be used by the independent centres.
One reason departments have been reluctant to move to shared services centres is their fear that shared services systems won’t accommodate their unique sets of terms & conditions – but Marriner says it’s possible to handle them all on a standardised platform. Anyway, he says, in future the Civil Service Reform Plan will be standardising performance management and other admin arrangements, reducing this complexity.
On such technical matters, Marriner certainly knows his stuff. He designed the NHS’s payroll system ten years ago, which caters for 6,000 different terms and conditions and still produced savings, he says. Marriner also has experience when it comes to big IT schemes, having worked for some major contractors into government (including Fujitsu, which has had a chequered history with the NHS). But now he’s in the public sector, he says he’s determined to get the best deal for taxpayers – even if that means squeezing the big players. The DfT’s service – soon to become ISSC1 – is currently using a system run by a big supplier, SAP, but “they’re just too expensive. Too expensive to run, too expensive to migrate to,” Marriner says: he’s looking for a cheaper, open-source alternative. The Cabinet Office also wants to involve SMEs as much as possible, and is currently using two small businesses to assist with the implementation of the shared services strategy.
ISSC2, however, needs to retain its existing, relatively expensive IT system, he says, because departments have already invested in these systems and need to recover that investment before looking at other solutions. A Prior Information Notice (PIN) has been placed to alert the market that there are plans to procurement an Oracle IT system at a value of £250-750m. Some departments are already using this system and their service contracts are going to expire soon. While Marriner wants to move smaller departments onto other software, the big players will need to maintain and support this system for the next few years, he says – though he will save 20 per cent of the potential cost by procuring as HMG rather than as a set of individual departments (see news, p2).
A few more significant obstacles remain, however – including the fact that some departments that outsource their back office services may end up paying VAT on the cost, adding 20 per cent and wiping out any savings. “We are working with the Treasury and HMRC to see whether we can remove that barrier, because actually it’s just moving money around the system... It doesn’t get any more revenue in,” says Marriner. “I think we can do something about that, and we’re trying to get the Treasury to come back and extend their relief scheme, which most big departments already have.”
No-one denies that these plans are ambitious. Indeed, there are even thoughts – albeit vague ones – of extending shared services past back-office functions and into procurement advice, Marriner says. Our interview ends with him still enthusing about the programme – determined to explain that, while there have been past failures, and some teething problems remain to be tackled, the programme represents good value for money.
With his years of experience in this field, he should know. This may well be the job he’s been working towards his entire life – and it’s an important one: trying to save the country billions of pounds by joining up inefficient processes, and thus allowing departments to sink the greatest possible proportion of their budgets and management into the frontline services that make a tangible difference to people’s lives.