WASHINGTON — As industry participants know all too well, revenue forecasting, an inexact science at best, can either make or break the toll road and public-private partnership projects that are sprouting up all over the country.
One need look no further than the 8.8- mile Pocahontas Parkway near Richmond, Va., to see how an inaccurate feasibility study wreaked financial havoc upon two private companies and raised questions about the viability of the toll road project.
When Australia-based investment giant Transurban signed a 99-year lease to operate the toll road in 2006, it was taking over a P3 in dire financial straits. The nonprofit Pocahontas Parkway Association, created for the sole purpose of issuing bonds to fund the road's construction, was in danger of being unable to meet its debt requirements.
Virginia kicked the project's original private sector partner, Fluor Daniel/Morrison Knudsen, to the curb and hooked up with Transurban, which defeased the $400 million debt and took on a $150 million Transportation Infrastructure Finance and Innovation Act loan as part of its new long-term lease.
The revenue estimates were the original partner's undoing, according to the National Council for Public-Private Partnerships.
"The project's first contract lacked proper planning and revenue estimates," the NCPPP says on its website. "Traffic along the Pocahontas Parkway was severely overestimated by Fluor Daniel/Morrison Knudsen and led to devastating revenue shortfalls. At the time that Transurban took over operations, average daily traffic was generated only 60% of projected traffic in the adopted business model."
Transurban did not reduce its projections for toll revenue and average daily trips for the parkway until June. Its new projections led Fitch Ratings to conclude, in an analysis, that "these reduced forecasts are likely to impair the ability" of Transurban to pay debt service on the bonds and repay the federal TIFIA loans used to fund the road's development.
Transurban announced in early August that it had devalued the road by $138.1 million because of the reduced revenue forecasts, and had suffered a 50% profit loss for fiscal 2012 as a result.
If Transurban cannot pay debt service on the bonds, it will be hurt even further because it is responsible for the debt, according to its contract with the state. Under Virginia's earlier contract with Fluor Daniel/Morrison Knudsen, any unpaid debt would have become an obligation of the state.
The story of a traffic and revenue, or T&R, study seemingly viewed through rose-colored glasses has become a familiar one to industry leaders, financial advisors and public advocates keeping an eye on infrastructure finance trends.
Neil Gray, director of government affairs at the International Bridge, Tunnel and Turnpike Association, said the issue is attracting more and more attention.
"A recurring theme over the last couple of years has been why are T&R studies always so wrong?" he said.
Christopher Mwalwanda, vice president of toll finance and technology at CDM Smith, a Massachussetts-based infrastructure firm with global experience, characterized the revenue estimating process as very complex and multi-layered, and affected by any number of variables.
Before a project ever gets a revenue forecast that can be used to obtain a credit rating, it will have to go through environmental and other evaluations. The feasibility study typically takes anywhere from six months to a year to conduct after that, depending on circumstances, according to Mwalwanda.
"You're trying to collect an amalgam of data and tools," he said. "A lot of that comes down to the historical data that exists, and the accuracy of that data," he said.
He added that it is the traffic and revenue consultant's role to have a "comprehensive understanding" of these data gaps and inaccuracies, and to supplement them where necessary.
Mwalwanda said he uses four major factors when evaluating a project's revenue potential. The first is the demand for travel along the route of the new road. Another is the expected growth of the area in coming years.
A forecast also has to take into account the availability of non-tolled alternatives for drivers traveling the route of the proposed tollway, and the willingness of drivers to pay for the convenience of a less heavily trafficked or more direct route.
"A lot of it does have to do with people's willingness to pay," he said.
Gray agreed, saying firms conducting feasibility studies often conduct polls of local drivers to gauge how receptive they might be to ponying up some money to use the road. He said they will ask drivers, for example, "If you could shave 20 minutes off your commute, what's it worth to you?"
From the project operator's standpoint, the rates and demand represent a balancing act.
In a presentation prepared for the Georgia Department of Transportation in 2010, infrastructure firm HNTB said state officials should be aware of the relationship between "traffic management" and "revenue maximization" scenarios. While the first aims to maximize traffic flow, the second typically uses a higher toll structure to maximize the project's financial potential.
"There is a high degree of sensitivity and compromise between the two scenarios," HNTB said. Generally, lower toll rates means higher usage of tolled facilities and thus, reduced operational benefits." Conversely, higher toll rates might send drivers looking for alternate routes.
The National Federation of Municipal Analysts' guidance on feasibility studies, termed "expert work products," or EWP, stresses both the importance of the projections to investors and the need for them to understand the extent to which they were reviewed.
"The existence of an EWP as part of a bond issue may result in significant investor reliance upon experts and their EWPs, the NFMA guidance states. "Investors and the municipal securities market in general expect parties to a financing to review EWPs that are presented with other offering documents. If this is not the case, it should be stated."
The NFMA said one of its goals is to foster "professionalism of the parties that produce expert work products."
Mwalwanda said revenue forecasters already hold themselves to high standards. "There are best practices that are inherent among the firms that have been doing this for a long time," he said.
Some of those practices are laid out in a six-year-old National Cooperative Highway Research Program report examining the practice of toll road revenue projections. It examined how projections of toll road revenue compared to the actual revenue obtained for 26 toll road projects between 1986 and 2004.
The report showed that only one project generated between 90% and 110% of the revenues projected within the first two years of operation. But none of the projects were on target with their projected revenues in each of the first five years of operation.
A few roads in the review, such as the GA 400 in Atlanta, were actually chronic overperformers.
Some, including the Pocahontas Parkway and the South Carolina Greenville Connector, performed at less than half of anticipated levels. The Connector project defaulted on more than $200 million of bonds in 2010.
Firms that produce feasibility studies are quick to point out the many variables inherent in a forecast, and concede that the projected numbers could consequently be off by a significant amount.
A study produced by Jacobs Engineering Group Inc. for the Rhode Island Turnpike and Bridge Authority cited such variables as new road construction, economic developments, and the adequate maintenance of the project for years to come.
The firms conducting the studies, toll industry members and credit analysts are all aware of how frequently feasibility studies miss the mark, but most of those projections still show potential issuers what they were hoping to see.
"The problem is exacerbated by the 'confidential' or 'proprietary' nature of the forecasts and methods that are developed for toll roads, and also by 'optimism bias' on the part of the sponsor, local elected officials or other advocates of the proposed toll road," the NCHR report states.
Gray agreed, saying, "Historically, you don't get a lot of negative answers."
This can lead to a stark contrast when it comes time to sell or reevaluate bonds.
A Standard & Poor's analyst who scored the BBB-plus Dulles Toll Road in Northern Virginia said his agency is inclined to just analyze existing historical debt without relying on a third-party revenue estimate.
Fitch analysts announced earlier this year that because of the evidence of high volatility in the earnings of managed lanes — an increasingly popular type of toll project allowing cars to pay a toll to use traditional "carpool lanes" — it will use "conservative growth estimates" for such projects in the future.
Robert Doty, an officer of a municipal advisory firm and president of his own consulting firm, AGFS, said investors shouldn't place too much emphasis on expert-produced revenue studies. Too many investors look only at the bottom line, according to Doty, and don't take into account the assumptions used in such forward-looking analyses.
"It is important for issuers and investors to understand among other things that, at times, there are no published professional standards applied by the experts and that the experts use assumptions given to them by others without reaching their own professional judgments whether the assumptions are reasonable." Doty said.
Mwalwanda said traffic and revenue estimates should be viewed as a range of possible outcomes and not a concrete forecast.
Meanwhile, critics of toll roads fear that, for some forecasters, the desire to produce favorable revenue estimates will outweigh the necessity to produce accurate estimates.
The NFMA guidance recommends that experts producing studies disclose recent work done for the issuer of that project and any personal connections to the issuer. These are potential conflicts of interest that tolling watchdogs have said can lead to inflated revenue estimates from experts hoping to please clients and score future contracts.
Those recommendations are followed by many major revenue forecasting firms, including CDM Smith.
Nearly all transportation policy professionals agree that tolled roads and other revenue-producing P3s will continue to grow in popularity in the coming years.
State departments of transportation are continuing to test the waters with new projects and new efforts to make tolling palatable in areas where few toll roads existed before. Revenue forecasters will continue to play a big role in the process.
Though forecasters do not pretend their best efforts produce perfect results, they say they do the best they can.
Issuers, investors, elected officials and the public are, in Gray's analysis, "kind of stuck with this."
"The investing community needs to have something," said Brad Guilmino, HNTB's chief financial consultant, who said revenue forecasts are "a means to an end."
"You need to have some kind of baseline for predicting your revenue and paying down your debt," he said
Ultimately, Guilmino said, the take-away from recent flare-ups over inaccurate forecasts is that studies will no longer be able to rely on growth generated by nonexistent infrastructure, such as a new bridge or railway line.
Rating agencies are cracking down on that practice, he said, and the kind of problems that have plagued some underperforming projects might become less common.
"They are definitely getting more scrutiny," Guilmino said.