Barclays Capital and Mediobanca's Eu1.84 billion ($21 billion) refinancing of the company's existing debt had securitization traders and syndicate officials raising cautious, even gleeful eyebrows when news of the deal first broke in November last year.

As a complete recapitalisation of an existing company with a secured loan structure, the deal is the closest the Italian ABS market has come to a whole business securitization.

Just as ambitious was the core element of the intended debt structure ? Eu1.2 billion of wrapped, bullet bonds in euros, dollars and sterling, with maturities stretching up to 35 years.

Perhaps inevitably, the final product had a hint of anticlimax about it.

In response to investor demand, the longer-dated tranche was scaled back to 20 years, and the dollar tranche was removed altogether. Two of the bonds ? over a quarter of the total wrapped issuance ? were privately placed.

But if it was a tough sale, the fact that it closed at all illustrates the versatility of the ABS market.

From a company funded solely with two cumbersome bank loans, Aeroporti di Roma (AdR) emerged with a diversified investor base of bond investors, and a far more flexible debt maturity profile.

Aping whole business

The background to AdR's refinancing is the mountain of bridge and term acquisition debt arranged by Barclays and Mediobanca to finance the privatisation of the airports, when they were bought by the Leonardo consortium in 2000 (see panel on page 42).

The acquisition finance was just the first step towards the company's new financial structure. The debt was initially set to amortise over time with step-up coupons, a structure typical in project financing.

However, the idea was always to replace the debt with a longer, bullet capital markets structure, made possible through whole business securitization techniques.

A monoline wrap would also allow the company to reach the broad universe of triple-A investors. Ambac, which has wrapped $8.6 billion of airport financings in the US and Australia, was the natural choice.

The refinancing would also have to be completed before March 2003. In July 2002 a consortium including Macquarie Airports Group agreed to buy a 44% share share in the company, and the refinancing was a precondition of the share purchase.

At its most basic, Romulus Finance is just a refinancing through an SPV of an existing investment grade loan extended by two banks.

The underlying rating is the same. There is little change to the existing security package, and no structural subordination between the different tranches of debt.

The transaction occupies a marginal space between a securitization, a corporate bond and a project financing.

The closest point of comparison for the deal in the ABS market is the whole business securitization techniques used by several UK water and sewerage companies (WASCs) ? most recently, Anglian Water Services' restructuring via Barclays Capital and Citigroup/SSSB in July 2002.

As in the UK water deals, the SPV is not actually able to hold security over government-owned assets ? although it does hold a charge over the accounts of the company, and certain moveable assets via privilegio speciale.

And, since it is not possible to hold security over future receivables in Italian law, the security interest has to be updated every six months to accommodate new contracts.

The primary credit concern, however, is the company's ability to continue to hold its core revenue-generating asset ? the concession to run the airport until 2044.

As with the recent Anglian deal ? which uses ratios based on post-maintenance capex and regulatory asset value to ensure the WASC can continue to operate efficiently ? the key covenant in the AdR deal is designed to maintain the key asset.

In return for its concession, AdR has to pay a fee based on the flow of traffic. If it failed to pay this, the company would risk losing the concession. Ambac therefore required a conservative test of the company's continuing ability to meet the fee.

Revenues must cover the fee by 2.1 times between 2002 and 2004, and by 2.5 times thereafter. If this test is breached, further indebtedness and dividends are blocked. The deal also has debt/EBITDA and debt service coverage ratios.

Unruly revenues

Unlike a water company, however, an airport is not just a single business ? it provides a commercial locus at which several businesses with different risk profiles converge.

?AdR presents a complex combination of two different cashflows ? regulated aviation revenues, and non-aviation revenues, such as retail,? says Rebecca Pynt, director at Ambac in London. ?It is effectively a complex hybrid of a whole business deal, like a pub securitization, as well as infrastructure and utility characteristics.?

The non-aviation revenues range from managed and franchised retail outlets and duty free shopping to advertising.

These revenues are expected to grow significantly in the next few years as AdR implements various efficiency measures and introduces a more selective product and brand mix in its retail outlets.

Standard & Poor's included growth in non-aviation revenues as a major strength of the transaction ? even though several past whole business deals such as Welcome Break and Roadchef Finance have been downgraded in the last two years after growth assumptions made by the deal's sponsors proved overly optimistic.

Ambac, conversely, had to assume zero growth in its base case scenario.

The more stable, regulated revenues are linked to inflation and based on a price cap minus an efficiency charge, with a periodical review every five years from 2004.

?We assumed zero growth in our base case for the regulated revenues,? says Pynt at Ambac. ?But there is a significant possibility of growth with EU harmonisation, since some of AdR's regulated revenues are approximately 40% lower than the European average.?

Despite the myriad cashflows, the main bet for investors is the continued use of the airport.

?The two revenue types experience different levels of volatility, but both sources of revenue are ultimately a function of the number of passengers using the terminals,? says Roderick Gadsby, associate director in the corporate securitization group at Barclays in London.

A key credit factor, therefore, is the continuing importance of Fiumicino and Ciampino as airports, and airlines' willingness and ability to use them.

A large percentage of the airports' revenues stem from one airline ? Alitalia. When Alitalia moved its intercontinental flights to Malpensa in Milan in 1999, the volume of passengers dropped from26.1 million to 24.7 million.

Despite the volatile nature of the airline industry, the airports themselves, and particularly Fiumicino and Ciampino, fulfil a vital infrastructure role, which supports the deal.

?The traffic at Fiumicino and Ciampino airports is largely origin/destination in nature, rather than hubs, which makes them less reliant on the credit of a single carrier,? says Barclays' Gadsby. ?If a carrier were to go bankrupt, another carrier would step in to fill the point-to-point route served by the bankrupt carrier.?

Agencies show their teeth

One risk Ambac and the leads could not pre-empt, however, was a triple-notch downgrade of Aeroporti's debt in the run-up to marketing the deal.

In July 2002 Moody's affirmed the AdR secured bank loan facility with a negative outlook.

But in October the agency placed the debt on review for downgrade, and the following month slashed it from A3 to Baa3.

The agency cited the prospect of slower than anticipated growth and the possibility of military action in the Middle East or further terrorist activity after the atrocities of September 11, 2001.

The most immediate effect of the downgrade was one of cost. Not in the existing debt ? it was structured without rating triggers or ?ratchets' ? but because the change in rating on the secured loan would have to be factored into the wrap.

?With the downgrade from Moody's late last year, the cost of the wrap went up, and some of the interest benefits of the deal were inevitably eroded,? says Gadsby at Barclays.

?However, the monoline premium is structured with pricing ratchets which will reduce the premium costs to the issuer if the rating of AdR improves.?

The downgrade also called for a last minute revision in the deal's covenants, and a far stricter business model for AdR.

?We recognised that the Moody's downgrade was due to external factors and were still very comfortable with the business profile of the company,? says Rebecca Pynt at Ambac. ?Nonetheless, we introduced a BBB- trigger for no further indebtedness to preserve the credit rating of the company.?

The new covenants mean that AdR is prohibited at launch from assuming any further debt, unless it is upgraded by Moody's.

Braving the markets

Shortly after the downgrade the leads began a preliminary marketing phase for the bonds before year end. Marketing proper began in the New Year with 55 one-on-one investor meetings over more than a month of roadshows.

A further obstacle to the deal was the controversy about monoline insurers' exposure to CDOs, which flared up at the end of 2002 when hedge fund manager Gotham Partners published a report on MBIA questioning the monoline's triple-A rating.

The investor reaction to that report had been enough on its own to push another wrapped deal, due to have been launched in 2002, into the New Year.

The week before Romulus Finance's launch, Moody's published another report, detailing the potential impact of CDO losses on monoline insurers, potentially further weakening investor appetite for wrapped risk.

?There is a lot of emotional opinion about monolines and some investors are still not yet ready or currently able to buy into the monoline story,? said Richard Mann, head of ABS syndicate at Barclays Capital in London. ?So we needed to do a significant amount of investor education with Ambac and the borrower on the roadshow.?

Throughout the marketing period the tranche sizes and maturities were kept fluid to best tap the different possible investor bases.

Preliminary guidance suggested three classes of bonds, as well as Eu150 million of bank facilities and a Eu450 million loan. All the debt tranches rank pari passu.

The wrapped bonds would comprise two euro-denominated tranches ? a seven-year floater and 10-year fixed-rate tranche, a dollar floater, and a sterling fixed-rate tranche with maturities ranging from 15 to 35 years to match the life of the concession.

In the end, the leads offered four tranches.

The fixed-rate tranche was sized at Eu500 million with maturity in February 2013, in line with expectations, and priced in line with pre-launch guidance at 75bp over swaps. It pays a coupon of 4.94%.

The bond was sold to institutional investors across Europe in ticket sizes of Eu25 million, with investors from Switzerland, France, Germany, the Netherlands, the UK and Spain.

?We set out to create a liquid benchmark product in 10-year fixed rate euros,? said Mann at Barclays. ?It helped that we were offering a simple bond structure that was widely marketable to both corporate and ABS investors, as well as an attractive spread of 75bp for triple-A risk.?

The dollar floater was scrapped altogether, and two euro bonds were privately placed with two capital markets investors.

The bonds, worth Eu200 million and Eu175 million with maturity in February 2015, were priced at 90bp over three month Euribor.

?We offered the euros in a variety of maturities throughout the marketing?, said Mann at Barclays. ?But we received substantial reverse enquiry for 12-year floaters, and finally chose a narrow, private distribution.?

The sterling tranche was truncated to just a £215 million 20-year bond, priced at 110bp over the 2021 Gilt, wider than the early guidance of 90bp over.

It was sold to around 20 accounts, primarily UK investors, as well as a multi-currency asset manager.

?It was quite a tough sale, but then it was also a tough market,? said one of the co-lead managers. ?It took a bit of time to be placed, given the unusual structure and collateral.

?Many investors are full on Ambac paper, or afraid of insurance companies in general,? said the banker.

?But investors look at the primary risk, and if you're comfortable with that risk, the deal is a great way to diversify at very attractive spreads.?

Bond market versus bank debt

The lynchpin of many whole business securitizations, including Romulus Finance, is that capital markets debt is cheaper than bank debt, and that longer maturities are cheaper than shorter ones.

The structure of Romulus Finance precludes the third ingredient for whole business deals ? the possibility of issuing debt at a higher rating than would be possible in other markets .

In fact, the shadow rating for the wrapped debt issued through Romulus Finance is the same as the Romulus class B notes, as well as AdR's new bank facilities and secured bank debt ? Baa3/A-. The Romulus debt also ranks pari passu with the rest of the new capital structure, so there is no benefit from seniority of payment rights.

The real purpose of the deal is simply to restructure the existing loan into cheaper, more manageable debt by refinancing through a capital markets SPV.

That in itself must have been an attractive enough proposition for Aeroporti di Roma, recently leveraged to the hilt and facing looming interest payments with step-up coupons.

The different maturities on the Romulus bonds, as well as the wrap, and the impact of the downgrade, make pricing comparisons difficult between the bank and capital markets debt.

The one direct point of comparison is the Eu65 million ?B' tranche. The bond was placed privately and priced at 125bp over three-month Euribor with scheduled maturity in February 2010 ? breaking even with the existing bank debt's coupon of 125bp over Libor.

That does not take account of the step-up provisions of the old bank debt. The coupon on the bank debt steps up by 15bp in 2004, climbing to 190bp over Libor by term in 2017.

?The previous bank loan had a project finance structure with step-up margins to encourage refinancing, so the bond issue is comparatively cheaper over time,? says Rod Gadsby, associate director in the corporate securitization group at Barclays Capital in London.

The slim gains on the interest coupons are not as relevant, however, as the bonds' respective maturities.

The weighted average maturity of the new debt structure ? which includes term bank facilities maturing in five to seven years ? is a mere 2.25 years longer than the existing debt.

But the new Eu1.2 billion debt structure establishes a flexible corporate debt profile, with maturities staggered over 13 years from 2010 to 2023. The bonds are designed to be continually refinanced like those of a normal corporate.

Refinancing risk is handled by cash trapping mechanisms that block dividends to shareholders a year before maturity if no refinancing package is in place.

?The key benefits of the deal are to allow AdR to extend the average life of its debt,?said Gadsby. ?And in particular to put in place a more flexible and equity-friendly bullet maturity profile.? n

This article appears courtesy of Structured Finance International magazine: go to www.ew-sfi.com