PFI: pretty feeble initiative so far

  • 01 Mar 1998
Email a colleague
Request a PDF

The role played by the UK debt capital markets in financing the Private Finance Initiative (PFI) has failed to live up to expectations.

Many reasons are cited -- including the inertia of the previous Conservative government, the competitive financing available in the bank loan market, the small size of many of the individual transactions and the lack of a dedicated investor base.

But, slowly, the pace is picking up. A number of recent healthcare and road initiatives have used bond issues, while the financing of some major UK projects that are unrelated to the PFI has shown some pointers as to how future PFI operations might be financed.

And bankers say investors are starting to show enthusiasm for the sector, believing it to bridge a gap between traditional highly rated sterling bonds and the new high-yield sector at a time when institutions are increasingly keen to put their growing credit analysis skills to work.

To describe a market as suffering from "a gigantic constipation problem" is hardly the most poetic turn of phrase. But this observation from one London-based banker on the impact on the capital markets of the UK's Private Finance Initiative (PFI) is an accurate summation of the progress which has been made to date.

There appears to be plenty of cash which could find its way into infrastructural projects constructed under the PFI, but precious few viable capital market opportunities for it to be channelled into.

When projects do come along, complain some bankers, they do so in a similar manner to London's notorious buses -- in groups of three.

Originally launched towards the end of 1992 under Britain's former Conservative government, the PFI scheme had started life hoping that some 600 projects worth £14bn could be funded from private sources by the end of 1998/99.

By October 1997, the PFI had inched past its halfway post, with projects worth just over £7.5bn having been signed. This figure, however, is slightly misleading as almost half of the total (£3bn) had been accounted for by a single project in the form of the Channel Tunnel Rail Link (CTRL), the immediate future of which is in any event shrouded by controversy and uncertainty.

It is also misleading as far as the capital market is concerned because not more than a fraction of the total finance raised for these projects has been generated via bond issues -- in spite of great hopes when the scheme was originally initiated that it would provide for a steady flow of project bonds which would be readily lapped up by UK institutions in search of higher yields and longer maturities.

There seem to be have been any number of problems which have restricted the development of PFI's recourse to the UK capital market over the last two to three years, although four stand out.

First, there was politics. While the old UK Conservative government was reportedly very good at making all the right noises about PFI and the valuable role the capital market could play in fuelling its growth, bankers say the administration which was bundled out of power last May was hopeless when it came to implementation.

"PFI under the Conservatives was frankly a disaster," says one project financier uncompromisingly. "It was all talk and it dragged on and on. The only reason there was any advance on prisons was because it was an opportunity for union bashing."

Second, there was the growing competitive threat of the bank financing market to contend with, which has confronted ever-dwindling revenues in the traditional loan maturities and which, as a result, has been edging further and further along the maturity curve and into the territory of the capital market.

"On a number of occasions where we've made bond market proposals which we were fairly confident about, we were beaten by the banks," says one investment banker.

"Up until October last year we found the banks becoming more and more aggressive and showing a willingness to go out as far as 20 or 23 years which would have been inconceivable a few years ago."

Problem number three has been the size of individual PFI projects, although some bankers report that this has been less of a restrictive factor than many had anticipated.

"I think a lot of people have been surprised about the extent to which the bond market has been able to accommodate smaller and smaller deals," says Guy Pigache, a director at Charterhouse Bank and an active equity investor in PFI projects.

"I had always thought the lower limit for bond financing would be £100m, but we're now seeing bonds being done for amounts of £70m and even less."

Be that as it may, investment bankers agree that the capital market is a non-starter as far as hundreds of would-be projects valued at up to £50m are concerned.

The answer, say bankers, would be to bundle projects together -- in the form of a quintet of schools, for example -- to achieve the sort of critical mass which the capital market demands.

"We've been talking to various people about the grouping together of projects for a very long time because there's an awfully large number of tiny projects out there which need to be financed," says one banker in London. "But we've encountered huge legal, accountancy and bureaucratic problems."

A fourth problem with promoting PFI within the context of the capital market has been the perceived blurring of debt and equity which has apparently come about in the minds of many investors in the UK contemplating the potential of the PFI market as an investment vehicle.

"When we started talking to institutional investors about PFI we could not find which desk would take responsibility for it," says a frustrated London banker.

"The equity desks told us they were only interested in quoted companies and venture capital, neither of which is applicable to PFI. The debt desks told us they were looking at Gilts and supranationals and rated corporate bonds, which is only partially applicable to PFI."

He adds: "When we asked them why they didn't have a PFI desk they told us that they could not allocate the resources because they did not know whether or not enough bond or equity issues would be coming along to justify them. So we had a Catch 22 situation: no desks because there were no bonds; no bonds because there were no desks."

Perhaps it is not surprising that there should be confusion as to which desk of an investor a PFI proposal should land on, given that throughout the life of a PFI project it will probably need a blend of different financing techniques depending on the phase of the project at which the funding is being raised.

This is explained by Pigache at Charterhouse, which is actively looking to take higher risk, early stage equity investments in PFI projects in much the same way as the other dedicated PFI equity players -- which include Innisfree, Barclays and 3i -- are looking to do.

"PFI projects have two phases: build then operate," he says. "Typically the build phase will take around three years and the operate phase could last 27 years. The first phase is the high risk period, at the end of which the project moves to resemble a low risk, public utility."

Adds Pigache: "Effectively in any PFI project you start by setting up a company, in which the founder shareholders are the construction and facilities management companies. But these companies generally don't want to be significant equity holders. They want to preserve their cash for other purposes and to find other third party providers of equity who will invest alongside them and share the risks, which is where institutions such as Charterhouse and Innisfree step in."

Self-evidently, the two different phases offer investors two very different kinds of risk/reward profile. The holy grail, within this perspective, is for initial providers of equity to find exit routes after the high risk period.

"At the end of the first phase we believe other institutions such as pension funds will be attracted by the long term stable returns from PFI projects," says Pigache. To date, he adds, the PFI market has been too embryonic to put this theory to the test.

In the meantime, PFI equity is a scarce commodity with only a handful of professional investors as players.

"The main pension funds are a pretty cautious bunch of people," says Pigache, "and, with the exception of Hermes and AMP, they haven't exactly dived head first into PFI."

One area where progress has clearly been made since May's UK election in terms of funding PFI projects through the capital market is the healthcare sector.

Last October, NatWest Markets and Schroders launched a £78.5m 30 year bond to underpin the financing of an £85m hospital in the north-western town of Carlisle.

Priced at 65bp over the 2021 Gilt, this was the first bond financing for a healthcare PFI project, which explains why the advisers to Health Management Carlisle (HMC) also opted to wrap the issue with a guarantee from MBIA Assurance, marking the first occasion on which MBIA had provided a guarantee for a sterling issue.

A spokesman at Greenwich NatWest's project finance department says that the HMC bond was placed with about 25 UK investors which included the classic long term institutions keen on 30 year risk.

"Because of the MBIA wrap," he says, "as far as the investors were concerned it was triple-A risk with a yield pick-up. But that is not to say that the investors didn't look at the project carefully beforehand. In fact, we insist that they do look at projects such as these because there is no doubt that in the future we will be offering these same investors unwrapped PFI deals."

The hospital projects which have followed the Carlisle initiative, however, have relied largely on bank financing. The most important of these has been the £214m Norfolk & Norwich Hospital project, backed predominantly by a £195m loan underwritten by Barclays Capital, ABN Amro, Midland, Bank of Scotland and Société Générale.

Progress has also been made in the financing of road projects via the capital market, with one of the most recent PFI projects being financed through a classic combination of loans, bonds and equity.

Financing for the upgrading of the A74/M74 road linking England and Scotland to motorway standard by Autolink Concessionaires (M6) was packaged through a £230m syndicated loan co-arranged by IBJ, ING and Bank of Scotland, an EIB loan of £69m, subordinated debt of £20m and equity of £16m.

The remaining £125m came in the form of a bond issue led by Morgan Stanley, with credit enhancement provided by Financial Security Assurance (FSA), which was priced at 62bp over Gilts, close to the bottom of its pricing range.

"It was a triple-A long dated structure which made it very appealing to local insurance companies and pension funds," says Ewan Labrom, head of Eurosterling sales at Morgan Stanley in London. "It was a very secure structure in its own right but the wrap gave investors added comfort."

The most strikingly successful project bond to have been launched in the UK over the last year, however, had nothing to do with PFI. This was a £286m 25 year facility to finance the construction of Enron's 790MW combined cycle gas turbine power station at Sutton Bridge in Lincolnshire, 140km north of London.

While bond financing had been used to support projects in the UK in the past -- most notably in the cases of the £198m bond for Kilroot power station in Northern Ireland in 1994 and the £400m transaction for First Hydro in 1996 -- a number of features made the Sutton Bridge financing unique.

First, unlike the Kilroot or First Hydro projects which were both post-construction bonds, the Sutton Bridge deal represented the first occasion on which the capital market had shown a willingness to take on pre-construction risk.

"Sutton Bridge was a complex project bond," says Sue Dale of the asset finance group at Barclays Capital in London. "It was the first bond to finance a greenfield power station and took on true merchant risk."

Second, the Sutton Bridge transaction was the first project bond to have been offered in two tranches aimed simultaneously at investors on either side of the Atlantic.

Arranged by an appropriately Anglo-American duo of BZW and Merrill Lynch, the bond -- rated Baa3 by Moody's and BBB by Standard & Poor's -- was split into a £195m sterling bond with an 8.625% coupon and a $150m Rule 144A tranche offering a 7.97% coupon, both of which are due in 2022.

The two-tranche aspect was the feature of the issue which plainly generated most excitement on the part of both its arrangers, although some competitors seem to have some reservations.

"I'm not sure if the Sutton Bridge deal demonstrated the depth of the UK market for project bonds or exactly the opposite," says one. "Ask Barclays or Merrills if the entire bond could have been sold in a single sterling tranche. I suspect the answer will be no."

"No" is not quite the response which springs from either of the lead managers to this question, with both insisting that the dual tranche structure had more to do with pricing and distribution than necessity in terms of raising the required amount.

"I think there is a very good chance that the whole deal could have been done in one currency," says Anthony Barklam, director of global debt capital markets at Merrill Lynch in London. "The more appropriate question would be to ask whether it could have been done at the same spread in a single tranche."

Structuring the bond with sterling and dollar tranches, he argues, created a degree of price tension between buyers on either side of the Atlantic which would not otherwise have been there.

"One of the great advantages of having a dual tranche structure was that everybody had complete confidence from the start that the deal was going to be a sell-out," Barklam adds.

"So if you're asking whether or not we saw enough demand to have generated an additional £100m without the dollar tranche the answer is probably yes, although the pricing may have needed to be wider."

The dual tranche structure, Barklam adds, also supported the bonds in the aftermarket. "By not trying to squeeze the whole deal into one market we probably ensured better secondary market performance in both tranches," he says.

"Both performed well in the secondary market precisely because we did not try to push the edges of the envelope in trying to get too big a size."

The proof of the pudding here, he adds, is in a brief comparison between the secondary market behaviour of Sutton Bridge and the only plausible outstanding benchmark in the market, First Hydro.

The First Hydro deal had been increased, in response to weight of demand, from £300m to £400m. Sutton Bridge was priced to offer a 10bp pick-up over First Hydro -- at 135bp over the Gilt compared to 125bp over for the outstanding First Hydro deal.

By the early weeks of 1998, says Barclays' Barklam, First Hydro had widened out to trade at a premium to Sutton Bridge. "The lesson we learned is that it is very important not to try to over-sell a transaction especially at the long end of the sterling market," he says.

Another new development in the placement of the Sutton Bridge bonds, adds Dale at Barclays, points towards an increased fusion between the bond and the bank markets.

"As well as the classic long term institutions we saw a number of the banks coming in and buying the bonds," she says. "Typically these were banks which missed out on structuring a loan because of the borrower's preference for a bond, but which had analysed the project carefully and liked what they saw."

The Sutton Bridge bond was not wrapped by any form of monoline insurance policy, which contrasts with some of the earlier project bonds arising from the PFI -- in particular the first deal of its kind, which was the £165m 25 year issue for Road Management Consolidated (RMC), which was led by SBC Warburg and Lehman Brothers in March 1996.

While welcomed by some observers as a much needed building block towards the creation of a functioning bond market supporting the growth of PFI in the UK, the RMC deal -- priced at 80bp over the Gilt -- was also lambasted by some as defeating the purpose of 'true' project finance.

This was because the comfort of the monoline guarantee simply made them triple-A rated bonds with an attractive pick-up over Gilts which did nothing to educate the UK investor base about project risk.

Looking forward, project bankers appear to agree that the benefits (or otherwise) of incorporating the insurance wrap around future projects will continue to be a lively topic of debate.

"Whether or not deals should be enhanced is an interesting subject," says Robert Rees, a director at Barclays Capital. "There is no doubt that in some cases having the enhancement helps to get deals off the ground which may not otherwise have been done. But clearly there is also a case for arguing that as institutional investors show an increasing appetite for yield demand will grow for unenhanced deals."

Clearly there are pros and cons to either structure. An enhanced deal, say bankers, will almost certainly carry the benefit of being appreciably less volatile in the secondary market than an unwrapped issue, simply by virtue of its much higher credit rating.

Another clear advantage for issuers looking at the enhanced route, adds one banker, is the reduction in the amount of administrative work involved.

"One thing which is very noticeable if you look at the offering circular for the Carlisle health project bond which is credit enhanced," says one banker, "is that there is very little information on the project, whereas the documentation required for unenhanced issues is enormous."

Anyone who has picked up the offering circular for the Sutton Bridge bond would immediately sympathise: in addition to a main text running to 171 pages, there are eight appendices totalling a further 208.

An additional advantage for investors buying unenhanced deals, say bankers, is that while they may be rated as triple-A bonds, they typically pay a hefty premium to run-of-the-mill top rated bonds.

As Dale at Barclays explains: "Even with a triple-A rating investors do need to do their homework on any individual project. This is because, while the risk of repayment is triple-A, the risk of default remains the underlying project. If the project defaults the enhancer can step in and make the choice of whether to continue to make the payments over the life of the bond or simply to repay the bondholders."

This, she adds, is one of the reasons why a triple-A enhanced bond will typically trade at a spread to Gilts of between 70bp and 80bp, while conventional bonds with a similar maturity and identical rating may trade as low as 30bp over.

"Project bonds play an important role in filling a gap in the market between the new high yield bonds and conventional highly rated corporate bonds," says Dale.

"They offer the advantage of being long term issues which cater to the demands of UK institutions, which enjoy solid backing but which at the same time offer a very reasonable yield pick-up."

Be that as it may, several bankers argue that on balance institutional investors are likely to show a growing preference over the medium term for unenhanced bonds -- given the imminence of the launch of the euro and the increased emphasis which will be placed on the analysis of credit as a result. EW

  • 01 Mar 1998

All International Bonds

Rank Lead Manager Amount $m No of issues Share %
  • Last updated
  • 24 Oct 2016
1 JPMorgan 317,793.98 1355 8.72%
2 Citi 301,114.13 1092 8.26%
3 Barclays 259,580.63 846 7.12%
4 Bank of America Merrill Lynch 258,842.43 934 7.10%
5 HSBC 224,273.23 905 6.15%

Bookrunners of All Syndicated Loans EMEA

Rank Lead Manager Amount $m No of issues Share %
  • Last updated
  • 18 Oct 2016
1 JPMorgan 29,669.98 55 6.95%
2 UniCredit 28,692.62 136 6.73%
3 BNP Paribas 28,431.90 139 6.66%
4 HSBC 22,935.49 112 5.38%
5 ING 18,645.88 118 4.37%

Bookrunners of all EMEA ECM Issuance

Rank Lead Manager Amount $m No of issues Share %
  • Last updated
  • 18 Oct 2016
1 JPMorgan 14,593.71 79 10.38%
2 Goldman Sachs 11,713.19 63 8.33%
3 Morgan Stanley 9,435.23 48 6.71%
4 Bank of America Merrill Lynch 9,019.27 40 6.41%
5 UBS 8,763.73 42 6.23%