Investments in US toll road PPPs have not proven to be very reliable, as assets across the country struggle through restructurings, downgrades, distressed sales, and even insolvency. Toll road financings have hit the buffers on an almost annual basis. The operator of South Carolina’s Southern Connector filed for Chapter 9 bankruptcy protection in 2010. The South Bay Expressway in California filed for reorganisation in 2011. Transurban started handing the Virginia-based Pocahontas Parkway back to the project’s lenders in 2013.

Last year Alinda Capital used a pre-packaged Chapter 11 proceeding to hand American Roads over to Syncora. Syncora insured the bonds issued by a portfolio that comprises the Detroit-Windsor Tunnel and four Alabaman toll bridges. Today, State Highway 130 in Texas (SH 130) is poised to miss its next debt service payment, after Moody’s Investors Service downgraded the road’s bonds to junk.

These PPPs were the subject of separate financings, but all the deals had one weakness in common: optimistic, if not wildly inaccurate, traffic projections. “Traffic projections simply were just too far off,” says Erich Eisenegger, a partner at McDermott Will & Emery in New York. “Unfortunately, an ultra-aggressive traffic forecast has at times been the central factor for winning a US toll road procurement.”

The trend is not new. An early privately-financed US toll road, the 22.5km Dulles Greenway in Virginia, struggled when it opened, after closing a $338 million bond financing in 1993. “When the Greenway opened to traffic in September 1995, traffic fell short of projected levels, and tolls were reduced,” noted the US Federal Highway Administration. Dulles Greenway’s original equity sponsor, TRIP II, was forced in 1999 to restructure the debt. It subsequently increased speed limits and introduced variable toll rates, but in 2005 it sold the road to Macquarie, for $617.5 million.

Roads to ruin

Several troubled toll road PPPs began operating amid, or immediately after, the 2008 financial collapse – years after traffic studies were completed, notes Chee Mee Hu, a Moody’s managing director in New York. “While things are improving, it is not yet robust,” she adds.

As the financial crisis hit, user growth slowed or dropped, causing revenues to slump even further. One adviser active in the market says: “There was the view that the real estate market would keep developing, the economy would continue to grow and so would car usage. Rather than slowing, car usage actually began to decline, so what was a slowing factor became a negative one.”

But the roads may still have struggled to meet their traffic forecasts even if their local economies had not cratered. “Overstated traffic projections are often not an honest mistake,” says Robert Bain of RBconsult in London. “The way the process is structured is that the public sector makes assessments on promoting an asset on an economic basis, so that incentivises inflated numbers to boost benefit-to-cost ratios. The contract is then often awarded to the highest bidder. This means the private sector bidders have to show strong cashflows to justify high bids in the first place.”

Examples of overbidding include Macquarie and Ferrovial’s $3.85 billion pitch for the Indiana Toll Road (ITR), which fetched twice the value that the state put on the asset. The Chicago Skyway – the first big US toll road monetisation – brought the city of Chicago $1.8 billion. SH 130 cost $1.4 billion to build, 1nterstate 495 in Virginia $2.1 billion, and the LBJ Freeway in Texas (IH 635) $2.6 billion.

High costs or sale valuations require more debt and generous traffic predictions that would reassure lenders that this was sustainable. There is even an argument that sponsors knew that their forecasts were not sustainable and still over-bid, assuming they could sell assets before their financings collapsed. After the financial crisis took hold, the traffic growth – which had been in the low single digits – ceased, and then reversed.

Slump in the road

Traffic projections tend to be linear, such that once a target is missed, a concessionaire will have to recover even more strongly than its unreliable projections had forecast just to make up lost ground. “The key for concessionaries is how to bide their time to catch up, which requires strong sponsor support and liquidity,” Moody’s Hu says.

But strong sponsor support and liquidity were not central to the initial financings, which tended to be overleveraged. “The main reason for the failures is not operational but financial,” Bain says. “The projects had too much debt. The size of the debt is based on traffic projections, so there is an upward pressure from the start.”

The financial collapse made these struggles worse. The decreasing revenue from toll roads happened as their debt costs increase. There are few areas where infrastructure finance mirrors practices in the subprime housing market, but the widespread use of accreting swaps, which backloaded debt repayments, appears in retrospect extremely irresponsible. American Roads’ bond financing featured an accreting swap, as did the Chicago Skyway and ITR acquisitions.

The swaps initially reduced financing costs by deferring repayments to further into the life of the loans. Yet, according to a Fitch report from 2013, near-zero interest rates caused the mark-to-market liability on accreting interest rate swaps to spike well beyond sponsors’ expectations. This increased toll roads’ risk profile and left sponsors with a liquidity gap to fill.

One adviser suggests that US toll roads were the subject of an asset bubble. Lenders could put in place elaborate financings, such as accreting swaps but the problem was those swaps were designed to withstand growth that was slower than predicted rather than a contraction. “When debt is provided even on a marginal growth basis for an asset but that asset is not growing, it is no longer a downside but a loss,” the adviser says.

Many lenders ended up taking over troubled projects. Even so, some advisers active in the market believe that the roads’ fundamentals will allow them to be restructured. Some have already been successfully refinanced or purchased, including South Bay, which the San Diego Association of Governments bought.

Alternatives to real toll deals are becoming more common. Government may decide that it wants to retain toll revenues and traffic risk, but transfer construction and life-cycle risk to the private sector. “States are now looking towards availability-based PPPs,” Mee Hu says. “Under the demand risk concession model, they handed over the asset to the private sector for a large up-front payment in return for a long term lease. There have been some problems in these concessions, however, and the states are keen to have more control over the projects while passing off up-front costs and construction risk to the private sector. Shadow tolls are also another option.”

The first availability-based toll road was the Interstate-595 Corridor in Florida in 2009. The project company, owned by ACS, and subsequently also TIAA-CREF, receives a maximum escalating availability payment, starting at $65.9 million per year from 2014. The model has also appears on Goethals Bridge between New York and New Jersey, the Ohio River Bridges East End Crossing, Presidio Parkway in California and Port of Miami Tunnel. These concessions offer sponsors milestone payments, final acceptance payment and inflation-linked availability payments.

Eisenegger also questions the viability of huge upfront payments by private operators, which makes the need for more accurate traffic forecasts, and hence long-term asset valuations, more acute. “The availability payment mechanism helps mitigate this, as well as asking for annual performance and revenue guarantees from the operator in exchange for an annual profit sharing mechanism,” he says.

Availability payments, however, limit the amount of new capital that government can raise. “New income is generated by real tolls, whereas availability payments involve recycling the same public-sector funds,” Bain says. “It is not a funding solution but a financing one.”

Public support

The US government’s TIFIA programme, first authorised in the Transportation Infrastructure Finance and Innovation Act of 1998, involves the US Department of Transport (USDOT) helping finance transport projects procured using newer methods – with direct loans, guarantees and standby lines of credit. TIFIA has approved more than 30 projects, including the I-595, Port of Miami Tunnel, Presidio Parkway and Goethals.

TIFIA facilities are used to support a wider debt package and loans can run for tenors up to 35 years, often cheaper than commercial debt. The East End Crossing was most notable for not using a TIFIA loan, so frequently has USDOT been a US PPP lender. While TIFIA loans are not perfect instruments, TIFIA can absorb some traffic risk or provide revenue bonds. “If I were a US toll road investor, I would look to TIFIA funding being in the financing mix as a big reason why I’d be more comfortable with the investment,” Eisenegger says.

One adviser suggests that it is too soon to predict how struggling PPPs will affect TIFIA, but notes that it did step up on the South Bay’s Chapter 11. TIFIA and the lenders took 100% of the restructured debt and all of the equity. “When the project was then sold to the San Diego Association of Governments TIFIA had a recovery,” he adds. “I think TIFIA will be able to make full or partial recoveries, as it has a long-term outlook and expects to be repaid over decades.”

Perhaps more importantly, Moody’s Hu warns that even though the amount that TIFIA can lend has increased 10-fold, its funding falls under the Moving Ahead for Progress in the 21st Century Act of 2012 (MAP-21), which expires this year. And the proposed shortfall facing the US Highway Trust Fund is creating concern over the availability of public funding for projects.

“A major factor is the legislation that provides for US funding of transportation through the Highway Trust Fund is due to expire in September,” Hu says. “If a replacement bill is not enacted in a timely manner, and with adequate revenue sources, states will have to rely on patchwork solutions, including PPP.”

The anxiety over public funding comes as Moody’s upgrades its outlook for US toll roads from negative to stable, based on 2014 traffic growth of around 1.5%, compared to a 3% decrease in 2009. Moody’s expects revenues to increase off the back of toll increases, which went up by 11.2% in 2012, after inflation-indexed or multi-year toll increases to help make revenue growth more predictable.

Every PPP, of course, has its own narrative. Public disquiet over private sector acquisitions of public assets has receded a little. But the volume of asset monetisations has collapsed. The US is far ahead of other markets in involving institutional capital in infrastructure, and sponsors and government will have to work hard to maintain their faith in the asset class.

“In addition to optimising risk allocation to support the long term viability of projects, the US infrastructure market needs to ensure that the traditional infrastructure investors – the pension funds and insurance companies – continue to be key players in our projects, looking for long term steady returns to match long term liabilities,” Eisenegger says.