On 24 August, 1992, Hurricane Andrew tore into Florida’s populous Miami-Dade County, killing 15, leaving 250,000 homeless and a bill of $30 billion for property damage.
Displaced and determined to never again fall victim to another hurricane, many Floridians moved north to neighbouring Broward County, ballooning its population by about 200,000 in the two years after the storm.
“None of the planning models for growth had projected that kind of influx and that kind of change,” says Joe Borello, I-595 project manager for the Florida Department of Transportation (FDOT).
Not surprisingly, the county’s I-595 expressway, which had opened for traffic just three years earlier, became congested way ahead of FDOT’s forecasts. So in 1994, the state’s transportation department began planning to expand the busy 10.5-mile corridor linking Broward’s other crucial traffic arteries, I-75 in the East and I-95 in the West.
Fast-forward 14 years. In September 2008, as FDOT was getting ready to evaluate final bids on a long-awaited contract for the corridor’s improvements, another catastrophic storm struck the US – this time, of the financial kind. Investment bank Lehman Brothers collapsed into bankruptcy, disrupting all commercial lending across all sectors of the economy, including infrastructure. Suddenly, the widening of the corridor, a $1.68 billion undertaking, looked set to be stuck on the drawing board.
And yet, some four months later, against all odds, the project had gone live. FDOT closed on the deal, handing the corridor’s improvements over to Spanish developer ACS Infrastructure. The successful close on such a large transaction, achieved during the most difficult economic environment since the Great Depression, was a coup for all parties involved. In the US, infrastructure professionals are citing it as the most celebrated toll road deal since the $1.8 billion Chicago Skyway project five years earlier.
How it happened is, essentially, a tale of two storms. The rapid, post-Andrew population explosion heavily influenced FDOT’s policy goals for the corridor improvement and led the state to use a European financing scheme, known as availability payments, for the project. In the risk-averse post-Lehman financing environment, these availability payments, along with hefty credit from the US government’s infrastructure lending programme that underpins the Transport Infrastructure Finance and Innovation Act (TIFIA), made the deal bankable at a time when the bond market was shut. And, as in any crisis situation, weather-related or not, flexibility and quick thinking saved the day.
With the I-595 congested and the county’s population only getting larger due to the post-Andrew influx, FDOT knew right away what any expansion project had to accomplish: “Our main goal was to maximize throughput through the corridor”, says Borello.
Time was also of the essence. At an average throughput of 180,000 vehicles per day, I-595 is probably already the most congested corridor in Florida, says FDOT construction project manager Paul Lampley. However, FDOT uses a pay-as-you-go system for funding projects, so the entire corridor improvement was broken up into 15 individual projects to facilitate its construction as funds became available. This piecemeal approach, combined with a funding shortfall, meant that, after 14 years of planning, it would take the state another 20 years to complete the corridor improvement, Lampley estimates.
A faster way was needed, so FDOT looked to the private sector. “We knew with the funding shortfall that [by] bringing in a concessionaire, we could see some efficiencies in the project, all the way from innovation to just an economy of scale,” Lampley explains.
But combining all the projects into one mega-project for the private sector, while efficient, created a roadblock. To maximise throughput, FDOT proposed adding three managed toll lanes to the corridor, or lanes whose tolls rise and fall with traffic levels. The toll lanes would be constructed alongside the existing free lanes, which would be expanded. The private sector would of course be fine with building toll lanes and collecting tolls later on. But asking it to fix up free lanes as well meant either building the toll lanes first and raising the tolls precipitously to pay for the freeway improvements or just leaving the freeway portion to the public sector all together. A middle ground was needed.
“We didn’t want to have a perverse incentive there to maximize toll revenue at the expense of the toll-free general purpose lanes. So we were looking for a payment structure that would really align both interests: maximize throughput and at the same time bring private innovation”, says Borello.
“The availability payment seemed to be the best match,” he concludes.
To any infrastructure investor based in Europe, availability payments are of course nothing new. They’re simply an alternative way of compensating a private sector concessionaire. In the case of the I-595, rather than collecting tolls, ACS Infrastructure will instead receive annual payments of an inflation-adjusted $65.9 million from FDOT in return for meeting specified levels of service operation and maintenance for the road, while FDOT will handle toll collection. It’s money that ACS is getting for making the infrastructure “available” to the public – hence the name.
The structure fit I-595 perfectly because it transferred the revenue risk to the state of Florida and allowed the concessionaire to focus instead on delivering the requested services for both the toll and non-toll lanes in exchange for IOUs from the state. And this could be done while the concessionaire still made good money on the deal: ACS will realise an internal rate of return of about 11 percent on the project, says Alvaro Muelas, chief executive officer of I-595 Express, the concession company formed by ACS to manage the deal.
Importantly, availability payments also lowered the risk to the concessionaire. With traffic risk, “no one owes you any money unless they drive on the road. And if they don’t drive on the road, you’re out of luck”, explains George Miller, a partner at law firm Mayer Brown who represented the club of 12 banks on the transaction. With availability payments, at least the state has an obligation to pay. The only wild card is that “there’s no legal obligation unless and until the legislature appropriates the payments that are due that year”, or what’s called appropriation risk. But given Florida’s solid track record of honoring payments, the appropriation risk was perceived by the banks to be lower than traffic risk, Miller says.
As events turned out, this could not have been more crucial to getting the deal done.
On 24 October 2008, as FDOT unveiled ACS as its preferred bidder for the 35-year concession, credit markets were frozen. With the markets in free-fall after the Lehman collapse, no one was willing to buy, much less underwrite debt without steep premiums. So ACS’ original plan to raise debt for the project by issuing private activity bonds – municipal bonds used to fund privately-backed projects – “just didn’t make commercial sense in comparison with what was available in the bank market in terms of pricing”, explains Miller. In December, ACS went looking for bank debt instead.
Banks, of course, were more wary than ever about taking on risk. So not having to price debt against several decades of traffic risk was a big plus. “Market perception of revenue risk really changed at end of 2008 and I don’t know if we could have really achieved close at that point without the revenue stream being balanced against the department’s AA rating,” says Jeffrey Parker of Jeffrey A. Parker & Associates, FDOT’s financial advisor on the project.
But capital was scarce, so just because the project had lower risk associated with it didn’t guarantee that banks were going to step up. For the numbers to work, the project needed to raise approximately $1.68 billion of financing, of which only about $200 million would be equity. That left $1.4 billion of debt. Pre-credit crisis, sure – that was possible. But not post-Lehman. Someone needed to fill the gap.
Luckily, FDOT had applied and was approved for a federal loan for the project from the US government’s TIFIA programme. Authorized by the Transportation Infrastructure Finance and Innovation Act in 1998, this programme makes super-cheap, super long-lived debt available to qualifying infrastructure projects, both publicly and privately-financed. With TIFIA, FDOT was able to garner $678 million of the $1.4 billion required.
“If TIFIA was not in the picture and, assuming that there was enough money available from senior loans, it would have resulted in an availability payment that would have been much, much higher,” says Paolo Curiel, vice president at Taylor-DeJongh, financial advisor to the TIFIA credit programme office. For Florida, that would make the project unaffordable, says Curiel. But if Flordia didn’t pay up, it “would have killed the returns of ACS”, he says. In short: the deal would have fallen apart.
Then came another lucky break: TIFIA credit rates moved counter-cyclically while ACS and its financial advisor, Macquarie Capital, were negotiating the terms of the senior debt. So as senior debt became more expensive, TIFIA became less expensive, stabilizing the overall cost of capital enough to get 12 banks to lend about $780 million to the deal.
Among the banks were a host of European lenders, including Spain’s Santander, Belgium’s Dexia and France’s SG and Calyon.
“If you look at the banks that participated in the deal, there’s a lot that you normally don’t see lending in the US,” says Mike Parker of Jeffery Parker & Associates. “ACS was able to call on those relationships and make sure it got those loans.”
No matter what their role, though, all parties involved in the transaction agree that flexibility and quick thinking helped save the day. FDOT was flexible on sharing some credit risk with ACS; TIFIA was willing to make some exceptions to its lending policies; ACS was able to switch from the bond market to the bank market.
Now, thanks to their quick thinking and flexibility, Broward County’s 1.7 million residents are on their way toward a less congested future. About 800 jobs have been created, FDOT estimates, and many more are to come. Pre-construction preparation has already begun and sound barriers are being erected, says I-595’s Muelas. Most importantly: rather than being completed in 20 years or so, the new corridor is scheduled for completion in 2014.
“Financial close was a challenge. In the end, it could be done because of the interest of the parties, all working together and developing a true partnership,” says Muelas.