“Experience is accumulating that remoteness between ownership and operation is an evil in the relations among men, likely or certain in the long run to set up strains and enmities which will bring to nought the financial calculation.” John Maynard Keynes, National Self-Sufficiency, 1933.

In Scotland, the public sector has contracted the private sector to build and run many public sector assets through PFI and its later non-profit distributing, (NPD), variant. The key private sector vehicle for doing this is known as a Special Purpose Vehicle (SPV). As shares in the SPV are bought and sold, the actual ownership of a PFI may quickly come to bear little relationship to the original shareholders. In this paper, we show who does actually own the SPVs which run Scotland’s PFI schemes: and we also show why the question of ownership matters.

There are eighty five PFI projects in Scotland (excluding projects reserved to Westminster), with a typical project life of 30 years. They have a total capital value of £5.706 billion, and in 2011-12 an annual charge to the public sector of £865.8 million.

Does it matter who owns PFI companies? The question of ownership certainly matters to the public sector client. For example, in Angus schools PFI, credit agency checks were used in assessing the financial standing of the bidders, “as they are providing the financial guarantees for the construction and facilities management sub-contracts”. It is clear that the Council was not only concerned about the construction phase but also about the consortium being able to handle long term facilities management of the PFI.

Among the public, there is still a widely held view in Scotland that part of the value of the public sector is that it is not-for-profit, and that it matters a great deal who owns our basic public infrastructure. One clear indicator was the resounding ‘No’ to the privatisation of the water industry. For the general public, the large returns which are made by many owners of PFI companies are not acceptable.

Moreover, public concerns over health and safety, and inconvenience, were widespread when the Seafield PFI sewage works failed in 2007 and pumped effluent into the Forth. Further, who owns the special purpose vehicle determines, for example, the PFI companies’ attitude to extending opening hours in schools for community use. And undoubtedly, the return to our economy can be much larger if the SPV owners are taxpayers in the country and not registered for tax in a tax haven.

The Treasury acknowledges that authorities might have legitimate interest in placing conditions on the companies who can own PFI contracts, but it has not taken any steps to enable this to be done effectively.

So who does own Scotland’s PFI companies? This is not an easy question to answer. The information published by the Treasury on ownership is seriously flawed; it is not always up to date and it just lists the owners’ names as given, with no clue of who may ultimately have the controlling interest.

We have researched company information and company reports to determine ownership in the four main PFI areas in Scotland: water, education, health and transport. A significant finding is how often equity stakes in PFI companies can change hands. Take, for example, the Aberdeenshire schools PFI1 project, with a capital value of £14.3 million. The original shareholders were the PFI Infrastructure company (20 per cent), Bank of Scotland (30 per cent), and the Robertson group (50 per cent), all companies based in Scotland. In 2004, the PFI Infrastructure company increased its shareholding and the Bank of Scotland reduced its. In 2005, I2, a company owned by Barclays and Societe Generale, bought out the PFI Infrastructure shareholding. In 2007, 3i, another PFI player, bought a substantial share in I2. In 2009, Barclays bought out 3i and Societe Generale, so that the shareholders of the school’s SPV are now, Barclays, Robertson group, and Lloyds.

The outcome of our research is a set of detailed tables showing the latest information we have been able to determine on the ultimate owners of each PFI scheme in Scotland, where these owners are registered, the capital value of the scheme, and the annual unitary charge (the tables can be accessed at www.cuthbert1.pwp.blueyonder.co.uk). The most striking features to emerge are:

The types of companies involved. A typical PFI project consists of a consortium of companies each bringing specialisms to the project – like expertise in construction, facilities management and finance. Hence, these types of companies constitute the original ownership of the SPVs. Their owners today are typically large banks, investment companies, and the investment arms of large construction companies. Among the banks, Barclays is dominant. Among self-standing investment companies, a dominant player is Innisfree, which has a seven per cent stake in Scotland’s PFI projects, including three schools projects, a hospital, and a road. The CEO of Innisfree is David Metter, a non-dom who is believed to control around three quarters of the company. Seventy two per cent of the shares in Innisfree are held in Guernsey. Among the construction companies which have investment arms are John Laing (Channel Islands), Balfour Beatty, and Hochtief.

Despite the dominance of these large non-Scottish companies, there are one or two exceptions. In particular, there are two significant Scottish companies, Robertsons and FES, which have used the opportunity of PFI in their respective base areas of Grampian and Stirlingshire to grow, and to build up significant portfolios of SPV shares.

And finally, in a few cases, the PFI owners are very obscure and small, yet own important PFI projects. For example, two PFI hospital companies, (Stonehouse and Crosshouse hospital, Ayrshire), are owned in entirety by a company called George Street Capital. The accounts for the company show that there are only two shareholders, both private individuals, and the company’s registered address is care of a lawyer’s office in Glasgow.

Concentration of Ownership. There is considerable concentration of ownership. The 85 SPVs running PFIs in Scotland have in total 51 shareholders. Seven companies own more than 50 per cent of the stock and subordinate debt of the PFI companies. These are, Barclays, Innisfree, Bilfinger Berger, Balfour Beatty, ABN Amro, John Laing and HISL. At the other extreme, SCRA, Care UK, Filetek and Stirling Gateway have a very small shareholding amounting in total to less than 0.1 per cent of the total.

Where these companies are located. The location of the SPV owners is as follows:

Location	Number of Companies Owning stakes in Scottish PFIs	Number of Scottish PFI projects in which stakes held	Total Capital Value of stakes held £m	Percent share of capital value
Channel Islands	8	33	1660.15	29.1 per cent
Netherlands	3	9	629.13	11.0 per cent
France	3	6	200.54	3.5 per cent
Germany	3	8	561.9	9.8 per cent
USA	1	3	43	0.8 per cent
Hong Kong	1	2	78.15	1.4 per cent
Australia	1	1	146.5	2.6 per cent
Worldwide: Barclays	1	25	776.66	13.6 per cent
: other	4	9	102.94	1.8 per cent
Rest of UK	13	29	1061.37	18.6 per cent
Scotland	13	31	445.89	7.8 per cent
Total	51	156	5,706.22	100 per cent

Note: the column giving total capital value shows how the total capital value of PFI projects splits down in proportion to the equity shares owned by companies in each location. The classification of owners by country is, to some degree, a matter of judgement. In most cases, the location is the head office of the company. A number of companies have moved their PFI financial business offshore to the Channel Islands, and are therefore classified as “Channel Islands” companies in the above table. This includes the John Laing group, the PFI spin out parts of HSBC, 3i, Semperion, Henderson, and Saltaire. Innisfree has also been included in this group as, although the company is registered in the UK, its principal shareholder is Guernsey-based. Barclays is by far the largest individual corporate shareholder, has been given its own entry, and also the classification “worldwide”.

Overall, what is absolutely clear from the table is that PFI ownership in Scotland is dominated by companies which are located off-shore in the Channel Islands or are foreign based. Just over one quarter of schemes by value is located in the UK.

The dynamism of the market. As illustrated in the example given earlier, it is common for equity stakes in PFI companies to change hands. But the secondary PFI market is dynamic in another sense as well: the return on the sale of equity to the original owner can be very large. For example, in 2008, Kier sold its 50 per cent stake in Hairmyres hospital PFI to the other shareholder, Innisfree, for £13.8 million. Its original investment in the project in 1998 was an equity injection of £50 and a subordinate loan of £4.2 million, (part of which had been repaid). This gain was after Kier had made an £8.1 million refinancing gain on Hairmyres in 2004. According to Kier, in total, this represented “a return of around five times our original investment”.

So what are the implications of all of this? The original proponents of PFI were very keen that a healthy secondary market in PFI equity should develop. The idea was that this would reduce the cost of PFI capital, and increase the attractiveness of investing in PFI to people like construction companies, who would want to extract their capital early on for use in other projects. These advantages should not be dismissed.

On the other hand, the features of the PFI secondary market which we have actually observed indicate that there are some very real potential drawbacks. These include:

• Excessive scale of returns. If the construction phase of a PFI project has been successfully completed within the original budget, then the allowance for the construction risk built into the original contract would be available to inflate the price in an equity sale. Both the National Audit Office (NAO, 2012) and a House of Commons Select Committee inquiry (Public Accounts Committee, 2012) examined the profits being taken in the secondary PFI market. Both enquiries concluded that there was some evidence of excess profits, over and above the return to equity owners from successfully bearing construction risk. For example, the Public Accounts Committee found that: “There is evidence from the amounts being realised by investors selling shares in PFI projects of excess profits being built into the initial pricing of contracts.” Research which we ourselves have undertaken also clearly points to excess profits in a number of PFI deals.
• No benefit to public sector from equity sales. When profit is taken out of a PFI scheme by means of an equity sale, then none of this comes to the public sector. This contrasts with the situation when profits are taken out through re-financing.
• Location of PFI owners in tax havens implies significant loss to public sector. PFI companies operate in the UK and the SPVs which run them are subject to UK corporation tax. In a large number of cases, as seen from table above, the ultimate shareholders in the SPVs tend to be based outside of the UK. Those based in the Channel Islands have no further taxes, such as income tax, stamp duty or capital gains tax to pay. Even those based in other tax jurisdictions where they do have to pay taxes will not be contributing to the revenues of the UK Treasury. There is a potential for further loss to the UK Exchequer if the SPV owners engage in transfer pricing, so that profits generated in the UK are actually booked as occurring in low tax jurisdictions elsewhere. It would be surprising if this were not happening given the wide occurrence of transfer pricing in multinational business.
• The public sector has no control over who owns PFI companies when equity stakes are sold. As we have seen, public sector bodies commonly carry out extensive checks on the initial PFI consortium members in order to make sure of their reliability, and in order to check that they have sufficient financial resources to meet their obligations. However, the public sector effectively has no control over who subsequent PFI owners might be. Allowing a free secondary market in shares of PFI companies without carrying out similar checks on social commitment, competence and financial robustness makes a mockery of much of the earlier pre-contract scrutiny.
• Bundling of safe PFI projects with riskier projects increases risk to the public sector. To give an example, 3i Infrastructure, which owns 1.77 per cent of the PFI market in Scotland, in part through a portfolio called Elgin, wrote “Exposure to social infrastructure (UK PFI) is helpful in providing the Company’s portfolio with lower risk, index-linked cash flows, which are counterweights to some of 3i Infrastructure’s higher risk investments, for example those in the 3i India Infrastructure Fund. The Elgin portfolio is fully operational, and is delivering a robust yield.” This example illustrates how PFI projects are liable to end up as the low risk components in a mixed risk portfolio of projects. If some of the higher risk components in such a portfolio then experience problems, this could well lead the investment company to seek to extract every last drop of potential profit from the PFI scheme, leaving the public sector exposed to increased risk of the project under-performing or even failing.
• Increasingly heavy gearing of some PFI companies. The ‘gearing’ of a company is the percentage of the capital invested which is represented by debt. It can now be seen that more and more secondary market purchases of PFI projects are debt financed. So, for example, when Barclays Integrated Infrastructure Fund (BIIF) bought out all non-Barclays investors in its sister fund, Infrastructure Investors (II), for £558.6 million, the transaction was funded in part by a £346 million acquisition debt facility provided by a banking group comprised of Calyon, Lloyds TSB, Nationwide and RBS. The effect of such debt financed secondary market sales is therefore to increase the amount of market debt which the SPV ultimately has to service: in other words, the effect amounts to increasing the SPV’s market debt gearing. This is likely to increase the risk of collapse if the SPV suffers a setback. Again, this increases the risk being born by the public sector.

This exercise has shown that there are indeed legitimate concerns about the way PFI companies in Scotland are owned, and about how the secondary market in PFI equity operates. There is a real need for improvements to be made. In particular, we would urge that:

• There should be much greater openness about who owns PFI companies, including their ultimate controlling ownership. Ownership details should be available without going through the kind of laborious exercise which had to be undertaken for the present study. The Treasury should be put under pressure to do this job properly, maintaining a record which is accurate, up to date, and which goes beyond company name to reveal controlling ownership.
• Future PFI or similar contracts should embody conditions which give the public sector adequate control over equity sales and subsequent ownership of PFI projects. In particular, the public sector should have a right to share in profits realised on sale of equity: they should be able to veto equity sales which place extra burdens of debt and risk on the PFI project: similarly, there should be a veto on equity sales to companies (or their parents) registered in tax havens.
• As our study indicates, only two Scottish companies have benefited from the Scottish PFI market in the sense that they have been able to grow into significant players with substantial PFI equity holdings. And in fact there is evidence to suggest that the more usual situation is that the PFI market has increased concentration among construction and facilities management companies, to the detriment of Scottish companies. There is an urgent need to investigate the effect of PFI and the NPD model on Scottish companies: and to take remedial action where it is called for. The PFI/NPD model may not be something that is generally liked: but as long as it exists, we should at least be making sure that it is the Scottish economy that benefits as much as possible.