Are the Public Accounts Committee sufficiently across the detail of PFI?

Following the Public Accounts Committee’s (PAC’s) recent report titled ‘Government’s Use of Private Finance for Infrastructure’, Geoffrey Clifton-Brown, Chair of the PAC, published an article in CityAM expanding on some of the committee’s observations, and the recommendations made.

However, there are five points in particular Clifton-Brown makes that are worth scrutinising in further detail, which are set out below.

1. ‘Pipelines have not been comparable year on year – indeed, nothing was published at all in three of the years between 2019 and ’22.’

   Mr Clifton-Brown is conflating the NISTA (formerly IPA) PFI data containing information about operational PFI Projects, which is generally published annually but was not between 2019 and 2022, and a pipeline of future infrastructure projects. While the National Infrastructure Commission (NIC) pipeline did not publish a pipeline in 2019 or 2022, they did for 2020 and 2021.

2. ‘The badly-built hospital needing millions in maintenance or remediation fees’

   Mr Clifton-Brown talks about ‘the failings of PFI’ and gives generic examples, of which this is one. However, maintenance under PFI is the sole responsibility of the Project Company and its supply chain, and any costs over and above what has been modelled remain the private sector’s responsibility; if that responsibility is not discharged, deductions from their fees are incurred. In the event a hospital was badly built and remediation is required, this also falls to the private sector at no additional cost to the public sector client.

3. ‘the school forced to trim its playing field grass perfectly down to the centimetre for fear of charges from a private firm’

   This example, also ostensibly of the ‘failings of PFI’, has the actual situation entirely the wrong way around. Any requirements, the length of grass included, will have been specified during the procurement design by the public sector client. The Project Company then must adhere to these standards, or else they will be charged in the form of deductions from their regular service payments.

4. ‘While some infrastructure would not have been built without PFI, public bodies are now on the hook for £136bn in fees to PFI companies up to 2053’

   Things don’t get built and maintained and operated on an availability basis that is far more stringent than non-PFI standards for free! PFI built over 700 social infrastructure assets, from schools to hospitals, roads to prisons, and in addition to the buildings themselves, and the borrowing costs, these fees also cover:

   • individual maintenance of all building components;
   • replacement of assets like heating systems, ventilation systems and in some cases complex medical equipment such as MRI scanners;
   • In many cases ‘soft Facilities Management’ services such as cleaning, catering, and portering;
   • any remediation work to a building’s fabric in the event it is required, including where changes in law require it, such as replacement of external cladding.
5. ‘Our report identifies a misplaced belief – that transferring risk to the private sector means that the risk now lies elsewhere…’

   Risk transfer is a universal concept in all contracts, and talking about it in the way Clifton-Brown does here is misleading and problematic for any and all contracted arrangements. He seems to imply that risk will inevitably crystallise, and at a point when it sits with the government. On the former point, risk can be managed, and the transfer of risk should always follow the principle that risks are transferred to those best placed to manage them. It is entirely legitimate for entities to be paid to take responsibility for a risk, with appropriate reward for managing the risk to ensure it does not crystallise. On the latter point, there have been many examples, in both the context of PFI and beyond, where risks have crystallised and the cost of remedying them has fallen to the private sector, demonstrating that a risk transfer away from government has worked.

   In PFI, this has included the replacement of cladding required as a result of recent reforms in response to the Grenfell Tower fire, and the risk the private sector initially bore in not being paid until PFI assets were built and operational; even where delays occurred, this almost exclusively remained the case despite a significant increase in private sector costs.

   This is not to say that the government should not have robust contingency plans for all kinds of issues, supplier insolvency included. Clifton-Brown is right to say that supplier insolvency is made more likely by over-transferring risk, but the idea that all risks will eventually fall back to the public sector if they crystallise is wrong.

What should we take away from this?

While we welcome the PAC’s interest in this important topic, in order to learn the lessons of the past to improve the infrastructure landscape in the UK, we need to understand the details fully, engaging with them and iterating on current models and practices to ensure repeats of these mistakes are much less likely. Some of the points made by Clifton-Brown fall some way short of this aspiration, and risk perpetuating over-simplifications of a complex area in a way that harms the stated aspirations of the report.