Written evidence from Margaret Stephens (LCC 30)
Public Administration and Constitutional Affairs Committee
Sourcing public services: lessons to be learned from the collapse of Carillion inquiry
SCOPE OF THE INQUIRY
Following the news of the collapse of Carillion, the Public Administration and Constitutional Affairs Committee launches an inquiry into how the Government and public sector manages the risks of outsourcing the delivery of public services.
For many years the Government has been experimenting with different models to partner with the private sector in delivery of public services. From considering the questions set by the Committee, there appear to me to be three main lessons and also actions that can be taken.
There is a great bank of experience to drawn on within the respective work forces, contracting history, various enquiries and reports written. But, evidence has not been gathered systematically and organised to compare performance of different Models. Inevitably, the quality of Government decision making must be suboptimal. In particular, news on problems and failures has more impact than evidence of successes and on measuring assumed benefits. There is a real risk of unintended consequences with measures taken to deal with some problems, causing new problems. There is also a risk that Models which have delivered benefits are abandoned.
Government should decide what information it needs to make good decisions and set about collecting and managing this knowledge. This is especially important as civil servants change roles quickly.
These models have been used for large projects like the Olympics and Crossrail. There are various factors at work such as high calibre procurement specialists, experienced and empowered management and cross-party political support. The use of such models could be extended especially for services and infrastructure which require long term investment in people and assets.
PFI has been a major factor in the breakdown in trust in private sector outsourcing. This is notwithstanding that it is a relatively small element of public spending. All of the main outsourcing suppliers have had involvement in PFI. More recent high-profile failures in major outsourcing contracts have further eroded trust.
Models for procuring public services where the sole, direct, private sector counterparty makes excessive profits or is exposed to significant losses, run the risk that the story obscures the underlying question of performance.
The lesson is that where a delivery body is seen to act in the public interest, either by being publicly owned, not for profit or perhaps owned by UK pension Funds, it will enjoy much wider trust and public confidence, which is a great asset in fulfilling its purpose.
At the risk of stating the obvious, there will always be some interface between public and private sectors. Even with traditional procurement of capital projects, construction firms are used. The private sector supply chain of Crossrail is considerable, though the project is publicly owned. The outsourcing giants of Capita, Serco, G4S, Balfour Beatty and Interserve employ large numbers of people themselves and are responsible for even larger numbers in their supply chains. It is hard to imagine the hundreds of thousands of people, even millions, in the Government’s supply chain becoming public sector employees. Therefore, the Government’s experiment with different procurement models for delivery of public services must continue. This is because new and even greater challenges lie ahead; a growing population, more elderly people, higher expectations as to health and care, different and more education and training, more work-force mobility, less reliance on unskilled workers from abroad. At the same time, resistance to high taxes and the need to contain public sector borrowing. Recommendations on lessons to be learned, need to take account of these factors.
The Author
As a tax adviser working for one of the Big 4 accounting firms until 2016, I advised various models of private sector provision of public services, including PFI, from 1998 until 2016. I advised on many projects including acute and psychiatric hospitals, GP practices, schools, courts, MOD accommodation, training facilities, flight simulators, military vehicles, roads, bridges, light rail systems, tramways, offshore and onshore wind, power supply systems, nuclear and other power stations, solar farms, energy from waste, waste management, waste water systems and street lighting. I advised both the public and private sectors and across the whole asset lifecycle from option appraisal by the public sector, through procurement, bidding and bid evaluation, construction, operations, refinancing, variations and terminations. I advised investors on acquisitions and disposals of equity and debt and on fund structuring. I have represented the infrastructure industry on taxation matters in responding to HM Treasury consultations on taxation changes since 2000. I now have my own independent consultancy.
1. Does Government make effective decisions on how to source the delivery of public services?
Government should use its own guidance in The Green Book (2). It is not for a lack of a framework that there are failures in effective decision making. Broadly:
The Green Book also includes guidance on adjusting for optimism bias, which is an evidenced feature of public sector procurement, proper risk assessment and sensitivity analysis.
The preferred option should be identified, balancing costs, benefits, risks and unmonetizable factors. Then, the guidance recommends collecting data during and after implementation to improve current and future decision making.
It is hard to fault this as a good framework in terms of high level principles and had it been followed and well executed consistently cross Government, then no doubt many failures would not have happened. A continuing failure to collect evidence and use this to improve future decision making is perhaps the most glaring failure. There have been lots of enquiries and reports and some do note this. But many recommendations in these have not resulted in learning and change.
In its 2018 report on PF1 and PF2 (PFI), the NAO explains that over the last 20 years capital investment using PFI has averaged around £3bn per annum in comparison to publicly financed expenditure which currently amounts to £50bn a year (4). However, the use of PFI has had a huge impact on perceptions of the private sector. This NAO report explains that HM Treasury takes the view that the risk transfer to the private sector can result in benefits which can outweigh the higher financing costs. The NAO finds that some of these perceived benefits can be obtained without using PFI and notes evidence has been gathered which indicates that PFI projects may be significantly more expensive than direct procurement. However, the most important point is that the Committee of Public Accounts and NAO have highlighted the lack of data to assess and understand the benefits of PFI. This data would assist the decision makers in future projects. Some of these benefits can only be understood over the long term and we now have many years’ experience. Some data exists but is not necessarily all in the hands of Government or the public-sector procurer, nor is it in a readily usable format. However, the failure to collect and understand this data is in breach of the requirements of the Green Book.
I worked on the 2003 Green Book to find a methodology to capture the different corporation tax collection outcomes for Government as an input factor to the discount rate prescribed for calculating the public-sector comparator to decide whether using PFI to procure a project is value for money. At that time hundreds of PFI contracts had been signed but HM Treasury did not make the financial models which could have been used to harvest the data for this purpose available. I understood that it did not have these readily available and may not have had the rights to access them. From the start, Government could have decided what information it wanted to collect when it embarked on PFI and also secured rights to updated information it wanted. Had it done so it would have had a bank of information to inform decisions about contracting and to produce implementation guidance and training on Green Book Guidance.
I do not know if Government has systematically gathered the information, in relation to outsourcing in order to inform improved decision making, but I would guess not. Government should have a great, well-organised, easily accessible, bank of procurement options, contracting solutions, monitoring data, etc. widely available to enable Government Departments and other public-sector bodies to fulfil the requirements of the Green Book. Detailed and comprehensive supporting guidance should be developed and regularly updated.
For anyone it’s hard to know what you don’t know and this is probably the position of Ministers, senior civil servants and public-sector managers procuring services. Gary Sturgess’s report for the Business Services Association (3) produces evidence from the outsourcing industry and other sources that the public sector does lack sufficient skills and experience, lessons have not been learned, information not gathered and examined properly and reports not read. Assuming this is true, at least in some areas of Government, the question is why?
The incentives need to be looked at. If the incentives were right then the skills, information and willingness to learn lessons would be in place.
The main driver for PFI was the attraction to Government of having new schools, hospitals etc. without actually spending all the public money up front. This is an oversimplification (see 5 below) but it is the main reason why so many projects happened. As the on-going public expenditure committed has become more visible, along with issues such as inflexibility, higher costs, excessive profits and a failure to collect evidence of the benefits, the use of PFI by the public sector has fallen sharply. This may well be the wrong answer for Government.
With outsourcing, an important driver was reducing public sector expenditure and after 2010, this probably became the main driver of policy and the manner of execution. The problems now arising are set out in Gary Sturgess’s report. They are overwhelming and will be hard to deal with, even with a real cross-party will. There is a real conflict between short-term political drivers and actual delivery of value for money public services which require longer term investment, innovation, the building of skills and experience. There are many victims of getting this wrong - the users of public services (especially the elderly and the young), the employees (again many lowly paid, part-time, with dependents) and tax-payers, who do not want their taxes wasted.
Moving from PFI and direct outsourcing, for major projects we can see use another model. The Government owned, but independently managed, Special Purpose Project Delivery Vehicle. The Olympics Delivery Authority is considered a success, as is Crossrail. The model is being used for HS2 and with the recent set up of the Submarine Delivery Authority. Admittedly, these are major infrastructure projects which have attracted the highest calibre leaders and procurement specialists. There has been enormous political will to get these projects right and we can see that where the procurement method enjoys broad political and media support and management is independent of political interference, this helps.
In contrast, outsourcing of public services has not enjoyed broad cross party and media support. The major outsourcers have been vulnerable to Government incentives to reduce their margins and take on unprofitable contracts. If they were highly profitable this would also be a problem for Government and attract criticism. The benefits in terms of innovation and efficiency have not been measured. This has been bad for Government delivery of public services and for these businesses, their investors and lenders. I am sure some lessons can be learned to improve the position but other models should also be looked.
Incentives are improved where models are independent of short term political interference. Their funding must be stable and transparent. They must also be seen to be acting in the public interest in terms of spending enough but also innovating and achieving value for money, over the medium to long-term. They probably don’t even need to be Government owned as long as the profits have a public benefit. UK pension funds are very keen to invest in long-term stable, inflation protected investments to match their liabilities. With well thought out governance and protection for these investors, it should be possible to design bodies with the right incentives to deliver public services.
Does the public sector have the capacity to deliver services in-house when that is the most appropriate route?
Sometimes yes and sometimes no. If it is the most appropriate route, the capacity can be built with time. This is one of the factors to be considered using the Green Book Framework.
2. What lessons need to be learned from the collapse of Carillion about how Government and the public sector manages the risk from suppliers throughout the life-cycle of outsourcing a public service?
There should be enough suppliers for Government to be able to terminate contracts on agreed grounds and to re-award them to others. There is always the risk of a badly managed supplier and also Government itself can do a bad delivery job. If Government wants to take a contract in-house it must maintain the capability to do so. Government can decide to be a competitor itself.
Ideally, the market should also be deep enough to avoid over dependency on a small number of contractors. The view would be that this is bad for contractors, as if they are over dependent they may take on work that they shouldn’t. It’s bad for Government because services may become overpriced and there is less incentive to innovate. Also, the bar to new market entrants may become too high. There should also be a competitive labour market for the employees of these major suppliers and also for their subcontractors.
However, there is a strong view emerging that a highly competitive, adversarial approach is not serving us well. That in order to improve our productivity, which is essential to quality of life we want, we need more highly performing, infrastructure. “The model we use to deliver and operate much of our infrastructure is broken. Too often it produces assets and networks that are expensive, perform poorly and fail to exploit the advances in technology that are transforming other industries. Too often the supply chain that delivers our infrastructure seems locked in a cycle of low margins, low investment and dysfunctional relationships.” (1) The Institution of Civil Engineers is, “supporting a community of owners and suppliers committed to change.” The focus is more collaboration for creating long term value for money. This chimes with the broad findings of Gary Sturgess (3), that longer term partnerships and a relational approach will deliver better value and better services.
Finally, I do think that participation of foreign suppliers is desirable. This brings best international practice and pulls up domestic standards. Lessons learn in other countries can be brought to benefit UK. Also, these are not dependent contractors so their success indicates that there is good governance, proper processes are followed and the UK can be trusted. This fits with our broader agenda of being a good country to do business in and attracting foreign investment.
Government has followed EU procurement rules and, in my experience, given great attention to ensuring a competitive market. One example, was the awarding of Offshore Transmission Contracts when EU rules required offshore wind energy generation to separate from transmission. Great efforts were made to encourage new entrants to this new market so that there would be competition.
As mentioned above, the use of Government owned, independently managed, delivery bodies have been successful and this would be especially so in managing SMEs.
No doubt Government could do better, but to manage its own supply chain would be a mighty task and itself need funded. Also, it might have undesirable consequences such as over interference in business to no purpose. It would be hard to do better than the market in regulating contractors. The private sector will learn lessons and take action more quickly than Government can. The point about the private sector is that it must be able to succeed and fail and Government should only interfere with great care.
Special purpose Government entities appear to do well managing their part of the Government supply chain. The development of the 2012 Olympics site was a very complex procurement, on behalf of three Central Government Departments, and pioneered a method of mapping and tracking its contractor exposure as part of managing the progress of the whole project, timing, contract interfaces, costs against budget etc. Crossrail and other projects have developed this approach and the Government Infrastructure Projects Authority (IPA) understands and advises on best practice. If Government were a business, the failure of a major supplier would be on its risk register, but Government has so many risks to manage that the approach chosen is more that of flexibility and capacity to respond. These are skills which the senior civil service is well designed for.
A middle course would be to make more use of intermediary bodies to manage more contracts, such as Crossrail or ODA and perhaps to reduce the size and scope of contracts.
I would think, in theory, yes. The issue is more that these are overridden by more powerful drivers such as budget costs. There are also issues to do with recruitment, knowledge and skills training of people-see below. There have been failures resulting from people following processes without having the experience to really understand the risk implications and the authority and confidence to act appropriately.
This will differ greatly depending on the public-sector body. Over the years I have met really competent and committed teams and individuals. However, I can make some general observations. Both the private and public-sector struggle with continuity as between the teams who negotiate contracts and those left to monitor and manage the contracts. On the private sector side, the profit motive means considerable efforts are often made to overcome this – through knowledge management, training and continuity of senior management. The situation is often more difficult for the public sector. It will often use consultants to advise and negotiate on its behalf, and many civil servants change their roles very quickly. Both of these issues have grown with the severe contraction of the civil service since 2010, notwithstanding the re-engagement of many former employees as independent consultants.
It is very difficult to turn the clock back and whereas in the past corporate memory primarily rested in people, in future it must rest in knowledge management systems. The public sector would do well to invest in the best systems it can, creating a bank of knowledge than can be accessed across the public sector and even be made available to academics and others.
3. Given the concentration of outsourced public-sector contracts into a small number of large companies do the rules on oversight and accountability of public services need to change?
I don’t think changing the rules would help.
It is core to these businesses that they have public trust and are well run. If they are not they will fail.
They have learnt from their own failures and those of their peers and will be learning from the collapse of Carillion.
4. Are there limits to what can be outsourced?
Development of policy cannot be outsourced. There are also indicators of reaching the limits as to what can be outsourced:
There could be a view that a high level of societal outcomes which cannot be expressed in money indicates a limit. I am not sure this is necessarily the case.
Another question would be: are there limits to what should be sourced from the private sector directly by Central Government? Examples of limits here are probably:
5. What lessons can be learned from PFI?
There are important lessons to be learned from PFI. This was a Government initiative dating from the mid to late 1990’s which resulted in the building of many new hospitals, schools, roads and other public infrastructure. As a methodology, it has proven attractive to other Governments and been taken up by several other European countries, Australia, Canada and others.
In the UK, it has been heavily criticised by politicians and in the media and been much less used since 2008. For some years this was because of the Global Financial Crisis increasing the cost and reducing availability of private finance. Now the explanation is more likely that the PFI model has become discredited and Government Departments and other public bodies do not feel confident about using it. The inflexibility of long-term PFI contracts can be hard for public sector bodies to manage. Other models for public private partnerships building major infrastructure have evolved and been perceived as successful. However, PFI has left a damaging legacy in terms of the private sector being trusted to deliver public services. How has this happened and what can be learned by Government?
The NAO Report on PFI and PF2 issued on 18 January 2018 is the most up to date, “briefing on the rationale, costs and benefits of the Private Finance Initiative…. Programme as a whole.” (4) It highlights as a key point that, “As part of PFI reform HM Treasury considered removing incentives, unrelated to VfM, which have driven the use of private finance, but it chose not to. If capital and cash budgets are insufficient, private finance may be the only investment option for public bodies.” There have always been incentives for the public sector to use PFI.
The NAO Report sets out the potential benefits which can justify the higher financing and other costs of PFI as, certainty over construction costs, improved operational efficiency and higher quality and well-maintained assets. It finds that some of these benefits could in theory be achieved with publicly financed procurement, some NAO work on hospitals has shown no evidence of operational efficiency, but on the positive side PFI contracts do guarantee maintenance spending and standards. However, the evidence base so far is not firm as to whether PFI projects deliver sufficient benefits to justify their additional costs.
The private sector suppliers of these construction, maintenance, financing and other services bid for and delivered the contracts Government asked for, in competition under EU procurement rules. The concerns of politicians and the public that PFI contracts have cost more or that benefits have not been measured or evidenced, that the contracts are inflexible and lock the public sector into expensive long-term commitments, is the responsibility of Government. Nevertheless, they have damaged confidence in the private sector to deliver public services.
What lessons can be learned?
PFI contracts are complex and so are the financial and investment markets created by them. The evolution of these has also played a major role in the loss of trust in the private sector.
Refinancing gains were a relatively early source of criticism. In the early 2000s, interest rates were falling and, as PFI projects moved from construction to the less risky and more stable operational phase, project finance could be refinanced on highly advantageous terms. Shareholders made substantial gains and Government acted to enforce gain sharing for refinancing gains on new projects.
The Committee has heard evidence that building contractor margins have always been tight, services margins have been greater but also tightened in recent years. In order to manage risks transferred, the public sector required private contracting firms bidding for PFI contracts to take equity and provide long-term shareholder loans to PFI special purpose companies. Bidding for PFI contracts was much more expensive than for publicly financed contracts. If a bid was successful these costs would be borne by the project but if not then the contractors would have to bear these abortive costs. A success rate of one in four bids would be considered good. A market in selling PFI equity and loans in operational projects developed. The gains initially funded equity and loans in new projects and abortive bid costs but after the global financial crisis these gains had become substantial and a major contributor to the overall profits of the major listed suppliers to public sector.
Equity and shareholder loans in PFI projects became valuable in a way which could not have been foreseen by Government or the major suppliers of outsourced services at the outset of the PFI initiative. Some financial entrepreneurs saw the opportunities from the early 2000s and aggregated equity holdings, bringing in new investors including many UK public sector pension funds. These were Government backed, long-term, inflation protected, steady income yielding investments and a good match for pension liabilities. As interest rates fell these became increasingly attractive and valuable. This market was useful to the PFI programme as it allowed the original building contractor owners to realise their gains and bid for and invest in new projects. The 2018 NAO report sets outs the process by which Government sought to reform PFI from 2010-11 and the main changes which have resulted have been a requirement for the public sector to take equity, for equity returns to be published and equity competitions encouraged. These arise from criticisms over the size of equity gains.
Equity and refinancing gains have contributed to a suspicion and loss of trust in the private sector. It is easy to see how in the minds of Parliament, the media and the public these gains are in some way related to the additional costs, the inflexibility and uncertain benefits of PFI. That because investors have made gains, the public sector has lost out. That the private sector has taken unfair advantage, and is therefore not to be trusted as providers of services to the public sector.
What lessons can be learned?
Government created PFI and the private sector has created the PFI investment market. They are both complex, understood by few and have contributed to a distrust of private sector outsourcing contractors as fit partners to deliver public services.
April 2018
Footnotes *