Jim and Margaret Cuthbert | ukwatch.net

Lifting The Lid on PFI

Tim Holmes — Mon, 03 Dec 2007 15:39:36 +0000

This article explains the basics of why the Private Finance Initiative, (PFI), is a bad deal for the taxpayer. We describe here some of the major PFI errors which can now be identified, and we quantify the effects. The results, in terms of the burden being placed on each and everyone of us, are staggering: as we shall see, PFI can well be described as “one for the price of two”. There is now an urgent need for a full public inquiry into the way PFI has operated: there should be a moratorium on the signing of any future PFI deals until required changes have been made: and any past PFI deals shown to involve excessive and unreasonable profits should be reopened.

Although introduced by the last UK Conservative government, PFI was enthusiastically taken on by Labour. It involves private sector suppliers not only designing and building, but also maintaining and operating, major items of public sector infrastructure like schools, hospitals and roads over a large number of years. Instead of the public sector itself borrowing money to pay for the capital expenditure, what the public sector pays is an annual unitary charge for the use of the facility over its lifetime. All of the individual costs borne by the PFI supplier, such as capital repayment and interest charges, maintenance, and service provision, together with supplier profit, are bundled up into the single unitary charge.

The scale of PFI is huge. In Scotland alone, PFI deals in operation or signed cover capital expenditure of £5.1 billion, almost all under Labour. A further £1.7 billion future deals are in preparation.

Serious concerns about PFI emerged very quickly. One concern, for example, was whether risk was actually transferred to the private sector. Another concern was that the PFI approach also seemed to have an effect on the type of project being undertaken: there have been many examples where a public authority has started off with plans for a fairly modest refurbishment, but has ended up having been convinced that what was actually required was a much more expensive new build project. Yet another concern was with the scale of PFI projects: many projects are so large and complex that the degree of competition has been limited. As one supplier noted, an advantage of PFI from the supplier’s point of view was that “tender costs and complexity reduce competition”.

But the major worries about PFI have related to cost. For example, it became clear that many new hospitals were having to be planned on the basis of reduced bed numbers, in order to make the forecast unitary charge payments affordable: (for example, to make the New Royal Infirmary of Edinburgh PFI scheme affordable, a 24 per cent reduction in acute bed numbers was required across the Lothian Health Board area.) Suspicions really started to crystallise when several PFI suppliers were able to extract large capital gains from their PFI projects, at an early stage in the life of the project, effectively by capitalising their anticipated future profit streams. In several cases, the capital gains extracted were several times the original inputs of capital by the PFI contractor.

Concerns such as the above have been articulated by a number of critics, notable among these being Allyson Pollock, now at Edinburgh University, and her colleagues. The Government, however, (both at Westminster, and the previous administration at Holyrood), have countered criticism by their usual technique of unsubstantiated assertion: and they were greatly helped in this because the required detail about projected costs and finances was hidden away under the cloak of commerciality in confidence.

Recently the situation has radically changed: what has happened is that detailed financial projections for the operation of PFI schemes have started to become available, because of the operation of the Freedom of Information Act, (FoI). These are the actual financial projections produced by the consortia running the PFI schemes: and are normally included as confidential annexes to the Final Business Cases of PFI projects. Unison has recently mounted a concerted campaign using FoI to request copies of large numbers of full Final Business Cases. Most of the responses still contain large blanks where the financial information has been removed: however, the vital financial detail has been provided for a number of important PFI projects. This has allowed us to analyse a number of the financial projections, and this provides the basis for the key findings on which we report below.

Before moving on to the detail however, we should record our appreciation both of Unison and the FoI Act which has enabled this information to be brought into the open.

As noted above, the PFI consortium charges an annual unitary charge to the public sector client. This unitary charge has to pay for all the costs incurred by the provider, and yield a profit. Typically, to finance the initial construction, the PFI consortium will borrow the necessary capital. Most of the finance has tended to come from banks at a “competitive” rate. The remainder is usually made up of a mixture of what is called subordinate debt, that is money lent by the members of the consortium itself at a higher rate of interest, and of equity, where members of the consortium put share capital into the venture. Usually, the amount put in by the consortium by way of subordinate debt and share capital would be around 10 per cent of the capital value of the project.

Our examination of detailed financial projections for a number of PFI projects confirms just how hugely profitable PFI can be for the private sector consortia. But the important thing which our work suggests is that a large part of this profitability stems from two elementary mistakes made by the public sector in their negotiation of PFI contracts. We will start by looking at the nature of these mistakes first, and then consider their financial implications. Further details of some of the work reported on here can be found on our website, at www.cuthbert1.pwp.blueyonder.co.uk

The first key mistake we uncovered related to the way in which the unitary charge payment to the PFI provider is uprated, or indexed, through time. Normally, some simple uprate rule is applied to the unitary charge, whether as a whole or to each of its constituent parts, so that they increase in line with inflation, or some fraction of inflation. However, using a simple uprate rule like this neglects the fact that a major part of the supplier’s financing costs, namely, debt charges, will not be increasing, but will be declining through time: this is because, as debt is repaid, the payment of interest each year will decline. The difference, between that increasing element of the unitary charge which covers financing costs and profit, and the declining cost of servicing debt, is available as a large profit: a profit which has not been earned through excellent performance, but through what amounts to financial sleight of hand.

This, however, is where we come to the second major mistake the public sector has made in the way it considers PFI projects. One of the key measures which the public sector uses in assessing the profits made by PFI providers is what is called the Internal Rate of Return, (or IRR), on the finance put into the project by the consortium. This is called the IRR on equity, (though here “equity” is interpreted as provision of finance by the consortium by means both of subordinate debt and equity proper.) The IRR is, effectively, the projected rate of interest earned on the consortium’s initial financial input to the project. The public sector regularly quotes the IRR on equity for individual PFI projects, usually as a justification for the reasonable rate of return being earned by the operating consortium. The big mistake is that IRRs are typically quoted on their own: but in fact the definition of IRR implies that there is a notional outstanding debt, on which this IRR is being earned. IRRs on their own are meaningless: the average debt on which this rate of return is actually being earned must also be quoted. A little thought tells us that, if the consortium is earning an IRR of 15 per cent to 20 per cent per annum on equity, (which the Treasury seems to regard as reasonable), but postpones taking dividend payments until near the end of the project, then the outstanding debt on which the IRR is being earned will rapidly snowball: so the actual financial return to the consortium, which is determined by the IRR and the average outstanding debt, can only be calculated if both the IRR and the average outstanding debt are quoted together.

So what does the data tell us about the size of the profits earned? Let us start with one example, a hospital project in England with a capital cost of just under £70 million. To finance the building the consortium borrowed over £60m from banks, at an interest rate of just over 6 per cent: the consortium itself provided almost £10m subordinate debt for the project, for which it received a more generous 15 per cent, and the consortium also put in an equity stake of £1,000: (no, we have not misread the decimal point: we genuinely mean one thousand pounds). The project shows the classic signs of inappropriate indexation, with the senior debt being paid off quickly, and hence senior debt charges declining rapidly – but with the whole unitary charge being indexed over the full thirty year life of the project at 3 per cent per annum. As a result, the projected returns to the consortium are eye-watering: the £1,000 equity input is projected to earn dividends totalling to more than £50m. Taking account of projected undistributed reserves at the end of the project, the consortium’s own financial projections indicate that the consortium is expecting to reap a cash return of more than £90m in total on its investment, (by way of subordinate debt and equity), of less than £10m.

Cash returns on this scale are staggering. But simply aggregating cash returns accruing over a long time period is misleading, since a given amount of cash today can be invested, and is therefore worth more than the same amount of cash at a future date. It is also useful to have a standard approach to looking at different PFI financial projections – all of which are slightly different in detail. To get round these two problems, we developed two standard summary measures for comparing different PFI schemes.

First, we split the projected sequence of unitary charge payments for a PFI scheme into its service element, (that is, that part covering operations and maintenance), and what we have called the non-service element, (that is, covering interest and repayment of principal on senior and subordinate debt, taxation, and profit). We then discounted the non-service element of the unitary charge by an appropriate discount rate to give the Net Present Value (NPV) at the start of the project: and we compared this NPV with the original capital cost of the building and equipment. The discount rate we chose was the interest rate which the public sector would have paid if it had borrowed the resources itself: so the comparison of the NPV with the capital cost gives us an indication of how much more the public sector is paying for the building and equipment under PFI than if it had been able to go out and purchase these directly.

For the English hospital example discussed above, the NPV of the non-service element of the unitary charge is, in fact, almost double the original capital cost of the hospital. In other words, for this hospital, it could truly be said that the taxpayer has got “one for the price of two” through using PFI. But this is by no means a unique example. We have analysed in detail three Scottish projects. For one of these, a PFI funded major educational establishment, the NPV of the non-service element of the unitary charge was more than double the capital cost – so this example is actually marginally worse than the English hospital example. For a recent significant schools project, the NPV of the non-service element of the unitary charge was more than 50% above the capital cost.

The second standard summary measure we developed relates to the internal rate of return (IRR). For each of the projects we have examined in detail, we calculated the pre-tax IRR on equity: (that is, on equity broadly defined as subordinate debt plus equity proper). But in addition, we also calculated the average notional outstanding debt on which this return was earned. In all of the three cases considered above, the pre-tax return on equity was just above 20 per cent: (which would correspond to a net post-tax return in the mid-teens, which the Treasury would probably regard as acceptable). However, the really significant point is that, in each case, the average debt on which this IRR was earned was very much larger than the actual input of capital via subordinate debt and equity: in the most extreme case, the average debt was more than 2.5 times the initial input of subordinate debt plus equity: in another case, more than twice: and in another case 1.5 times. This confirms that the IRR alone, (generous as it may seem), in fact grossly underestimates the true scale of return accruing to the operating consortia in some PFI projects.

Our analysis confirms therefore, that the returns arising from inappropriate indexation are enormous – as we have seen “one for the price of two” is an appropriate designation of some PFI schemes.

What we have described so far in this article are definite errors and mistakes we have uncovered in the practice of PFI. It would be wrong, however, for us to give the impression that we understand everything that is wrong with PFI – far from it. Our detailed examination of PFI financial projections indicates that there are other important aspects of the process which require much closer investigation. For example, once the construction stage of the project is completed, there is often projected to be a significant financial reserve left from the original project finance, which is then paid into reserve accounts to earn interest for the operating consortium. At its worst, this could mean that the public sector is paying, via its expensive unitary charge payment, to fund the consortium to borrow money which the consortium then pays into an interest bearing account for its own ultimate benefit. There needs to be a much more detailed accounting for the cost elements which are actually contributing to the cost basis of the unitary charge.

What we have shown in this article is that, once PFI is examined in the light of the facts which are now becoming available because of the Freedom of Information Act, it can be seen to be nothing short of a disaster. The effects, however, go beyond the question of costs alone – even though no country, whether it be Scotland or the UK as a whole, could long support buying its capital assets on a “one for two” basis. But in addition, PFI has also had a marked, and negative effect on the industrial structure of Scotland. Given the large size of PFI projects, local firms have great difficulty in competing. The effect, in the Scottish context, is that local firms have either been squeezed out, or, (if they did get a foothold in the PFI world), are more susceptible to being taken over by international players attracted by large PFI profits. So not only are we paying excessively for PFI services: at the same time PFI contributes to our losing control of our own economy.

So what needs to be done: we suggest that a number of actions are required. Firstly much more needs to be done to improve the availability of information. The detailed financial projections should be published for all past schemes, and also for any future schemes. In addition, standard indicators should be published for all schemes. In the light of our own work, we suggest such indicators should include the Net Present Value of the non-service element of the unitary charge, in comparison with the capital value: and also the projected internal rate of return on equity, together with the corresponding average notional outstanding debt on which the IRR will be earned.

Secondly, any future PFI projects should be unbundled into smaller projects as far as feasibly possible, so that a genuinely competitive market has the chance to become established, and to allow local firms to compete.

Thirdly there has clearly been a systemic failure in the existing mechanisms designed to secure value for money from PFI schemes. There needs to be a full and detailed inquiry to establish exactly why the reality of PFI as it is now emerging is so different from the utopian view presented by PFI adherents like Partnerships UK.

In our view the problems with PFI are so grave that tinkering is not the appropriate approach at this stage. We would seriously suggest that there should be a moratorium on all future PFI projects, until the full facts of what has gone wrong, and why, have been established: preferably this should be done through a public enquiry. Moreover, if it is established that excessive and unreasonable profits have resulted from past PFI schemes, then efforts should be made to re-open them.

Margaret Cuthbert is an economist. Jim Cuthbert was formerly Chief Statistician at the Scottish Office.

Drowning in Self-Interest

Alex Doherty — Thu, 28 Sep 2006 16:27:53 +0000

Scotland is about to be faced with an intense struggle to prevent
the privatisation of its water industry. Powerful voices have
already been raised in favour of moves towards privatisation – including a surprising statement by the Chairman of the Water
Industry Commission, that water should be freed from state
ownership. This article outlines the main pressures leading
towards water privatisation. Basic fallacies are uncovered
in the arguments used by the privatisation lobby and the key
challenges that are faced if privatisation is to be avoided. In
particular, the critical battleground will be in the campaign to
expose the flaws in the currently used method for setting utility
prices, namely the Regulatory Capital Value method.

The movement to privatise water in Scotland has been
unwittingly assisted by three crucial strategic mistakes made
by the Scottish Executive. Firstly, the Scottish Executive made
basic mistakes in the implementation of a new system of
financial control for the water industry in 2002, which meant
consumers were overcharged, (by about £1billion over the
period 2002-10). This also meant that the Executive could use
water charges as a new source of taxation, transferring funding
provision to other parts of the budget. The resulting shortage of
public expenditure provision is now impacting on the short-term
availability of capital investment for the industry, with adverse
effects on levels of service and customer satisfaction.

Secondly the Executive set up a centralised and bureaucratic
leviathan in the shape of Scottish Water. The intention was to
improve efficiency, and harmonise domestic charges however,
the reality has been that Scottish Water has not been capable of
responding adequately and flexibly to local requirements. It has
also, despite harmonisation, introduced a tariff
system with high fixed charges for
industry, which adversely affects
small firms. It has also
reduced local contractors
to a peripheral role in the
delivery of much of its capital investment programme, through
its approach of developing Scottish Water Solutions, an unusual
partnership with specific private sector companies, which
has had adverse effects on efficiency due to the loss of local
contractors detailed knowledge. The overall results have
been damaging to customer satisfaction and local economic
development.

Lastly the Scottish Executive sanctioned the implementation in
Scotland of setting water prices via the Regulatory Capital Value
(RCV) method, which leads, as we show below, to significant
overcharging and ultimately generates a substantial financial
surplus for the operating company. Because of the profits
generated by the RCV method, water is now regarded in the City
as being a rich source of corporate profits.

These mistakes have set in train powerful forces, which work
towards the privatisation of the water industry in Scotland.
Public dissatisfaction can be manipulated to blame problems
with water on Scottish Waters public sector status – and this
dissatisfaction can only get worse, as prices start to increase
rapidly again in 2010, inherent in the RCV approach. The Scottish
Executive in due course will come under severe temptation to
cash in on the financial surplus which the RCV method will
generate, by selling Scottish Water for a one-off gain of several
billion pounds and the City wants to get its hands on the profits
to be gathered from a privatised water industry.

These pressures towards privatisation are supported by the
widely held view that the Scottish Executive wont be able to
afford to maintain Scottish water as a public sector body in the
longer term, since water has been privatised down in England
and therefore there are no Barnett consequentials to benefit
the Scottish budget. It is unfortunate that this argument has
been too readily accepted – without, apparently, anyone
sitting down to do the arithmetic, which shows that the
argument is a myth with no validity.

At present the borrowing provision for water in
the Scottish Executive budget is £182 million
per annum. If we suppose that Scottish
Water has a continuing investment
requirement of £500 million per
annum in real terms, then
financial modelling shows that,
in the long run, Scottish Water
would need to borrow about
£120 million in real terms
and would have interest and
debt repayment charges
substantially below those
implied by the RCV method.
In other words, the Scottish
Executive could significantly
reduce its public expenditure
allocation for Scottish Water in the
long term while the industry could still afford
to maintain an investment programme on an ongoing basis at
its currently high level. In fact, the Executive would be unwise
to reduce its long term public expenditure allocation for water
as low as £120 million, since more than this would be required
if inflation increased.

A similar myth also exists that while it might not be necessary
for the Executive to sell off Scottish Water, the Executive might
judge that the money could be better spent on other priorities.
This view may have had some validity at times in the past,
when water and sewerage were unexciting topics but this view
is emphatically ceasing to be the case nowadays, as water
emerges worldwide as a limited resource of key importance. It
would be folly for the Executive to give
up the Scottish peoples ownership
of water just as the comparative
advantage, which Scotland could gain
from its water resources, is becoming
clear.

Now the issue of RCV, which reveals
fundamental and damaging flaws in
this method and is in many ways the
nub of this paper. On the face of it,
there may appear to be little ground
for questioning the RCV approach,
after all, it is the established approach
to utility pricing supported by the
World Bank and major accountancy
firms and it is applied to a whole
range of utilities, not just in the UK
but internationally. But because the
RCV method is widely applied does
not mean it is right – particularly
since, even at a superficial level, there
are a number of awkward questions
about the RCV approach.

For example, the RCV approach
is an application of current cost accounting. But why has
current cost accounting been abandoned everywhere else
in the private sector apart from the price setting for utilities?
And how, if the RCV approach is correct, can we explain the
massive profits earned, under RCV price setting, by the water
companies in England – where, for example, annual dividends
to equity owners have frequently run at a level of 30% or so of
the equity capital actually raised by the industry. Also, if the RCV
approach is indeed generating appropriate incentives, how can
the reluctance of the privatised water companies in England to
deal with chronic leakage problems?

The answers to these questions become clear when we probe
below the surface of how the RCV approach actually operates.
This modelling shows that the RCV method is fundamentally
flawed and has the effect of turning capital investment,
particularly long-term capital investment, into an activity, which
yields a substantial financial surplus for the utility.

For example, a company investing in a 30-year life asset, with
inflation at 2.5%, and interest rate of 5%, will reap under RCV
pricing a financial surplus of no less than 43% of the value of
the investment. Note that this surplus is of purely financial
origin- the company will reap the surplus whether or not the
underlying investment project actually makes an adequate real
return to the company: In the bizarre world of RCV, the company
would make a substantial profit out of putting up an expensive
100 year statue of its founder, provided it was allowed to class
this as capital investment. The RCV financial surplus leads to
overcharging, and to large profits for private companies but it
also distorts the investment programmes of the utilities and in a
way, which is consistent with their unconcern about leakages.

How could the RCV approach have gone so badly adrift? The
basic answer is that the RCV approach calculates consumer
charges on the basis of assumed costs of capital assets valued
at todays prices whereas, in reality, an industry with very long
lived assets, (as water is), operating in an era of even moderate
inflation, will face actual costs which
relate to the lower prices of capital
goods several years ago. The RCV
approach then makes a further
mistake, in attributing this difference
between assumed and actual costs as
a reward primarily to equity holders-
overstating, and over rewarding, the
equity holders contribution to the
finance of the business. This accounts
for the excess profits generated by
companies in England.

Faced with the above pressures,
the challenges to which we must
rise are to resist the move towards
privatisation, whether this comes as
an outright move; to mount an effective
intellectual campaign against the RCV
method and to campaign for much
greater democratisation of the water
industry in Scotland.

While privatisation is the issue, which
is liable to make the headlines, in many
ways the key strategic ground for the
struggle is the need to question the validity of the RCV approach.
If the RCV approach were abandoned, the excess profits to be
earned from a privatised water industry would no longer be
available- and most of the pressure for privatisation would
disappear. Conversely, if the RCV approach is maintained, even
without privatisation, then the Scots will still be overcharged for
water. So an important requirement now is to have a full, and
fully informed, debate on the RCV method.

Given the essentially technical nature of the issues involved in
constructing an effective critique of the RCV method, it will be
essential that those who are arguing on the anti-privatisation
side focus on the detail, not just on emotion: and that they put
sufficient resources into researching the technical aspects of
RCV. The rewards of success, however, will be large. A successful
critique of the RCV method will have important implications not
just for Scotland but also for the international campaign against
the excesses of globalisation.

For more information visit www.cuthbert1.pwp.blueyonder.
co.uk .

__Jim Cuthbert was formerly Chief Statistician at the Scottish
Office. Margaret Cuthbert is an economist.__