Answers

UPDATE 6/11: Eliminated duplicate chart; added links below to source documents

In early April the NCDOT announced they had selected Cintra, a Spanish company, as the contractor for the I-77 toll lane project. (As it turns out, the Cintra bid was the only one received.) Under the terms of the contract, NCDOT will donate the remaining public right-of-way  and grant Cintra an exclusive concession to design, build and operate toll lanes along the Lake Norman stretch of I-77 for the next 50 years. The NCDOT noted Cintra’s proposal had been subject to an extensive review of over 300 pass/fail criteria, and the Cintra package met all of them.

Despite this apparent rigor and detail, the public has been kept largely in the dark about even the most basic aspects of this project: How will drivers access the toll lanes? Where will they access them? How much are the tolls? I attempted to find out by filing a FOIA request, but weeks after filing and receiving nothing but stonewalling, I finally went to the media. Allison Latos at WSOC made a few phone calls and was able to shake loose more information in a week than I’ve been able to in two months. Thanks, Allison!

The NCDOT forwarded a series of traffic and revenue (T&R) memoranda from Stantec, an infrastructure consultancy. NCDOT also provided electronic copies of two memorandums to the NC legislature summarizing the I-77 tolling project.

Below is my summary.

Toll Rates

Stantec predicts a one-way morning trip from Mooresville to Charlotte will initially cost $9.05.  The return commute will cost $11.75, so a one day toll lane commute will cost over $20.  By the year 2035, the tolls will more than double to $20.60 and $21.63 for the morning and evening commutes, respectively, so the price of a commute twenty years from now will cost over $40 per day. (These are all in 2011 dollars.)

The toll for the worst stretch of SB congestion- Catawba to Hambright- is $2.09. This stretch routinely takes over twenty minutes to travel some mornings, and a $2 toll seems like a bargain. That is, if you are able to exit the toll lane once you’ve bypassed the worst of the traffic. I’m doubtful this will be the case; drivers entering the toll lane at Huntersville or Cornelius will probably be forced travel (and pay tolls) all the way to downtown, as the ingress/egress discussion explains.

Ingress/Egress

The contract requires at least one ingress/egress point between:

• Exit 13 (I-85) and Exit 18 (W.T. Harris)
• Hambright Road (approx mm 20) and Exit 23 (Gilead)
• Exit 25 (Sam Furr) and Westmoreland
• Exit 31 (Langtree) and Exit 33 (Williamson) Note: Davidson is bypassed

While it looks like a southbound driver may enter around Westmoreland and exit five miles later, the contract gives Cintra tremendous latitude in determining access points. The contract states that Cintra “at its discretion may propose alternate ingress or egress points other than those identified below ….”  So Cintra may propose (and why would the NCDOT object?) any access points they choose.  The financial model assumes traffic in the toll lanes will stay in the toll lane. It follows logically that access points will be selected to ensure this is the case.

Toll Revenues

In the first full year of tolling, revenues are expected to gross approximately $20 million. (In comparison the Triangle Expressway, North Carolina’s first toll road, had first year revenues of $13 million.)  This would make the I-77 toll lanes one of the highest-grossing in the country. In twenty years, the toll revenues are predicted to grow to $136 million, and by the contract’s end, a staggering $704 million dollars per year. (All numbers are in nominal dollars.) The total toll revenue over the 50 year life of the contract is predicted to be  $13 billion. This does not include penalties, fines and associated fees which can typically add another 25- 30%, bringing the grand total to over $16 billion.

In contrast, adding a single lane in either direction just where it is needed would cost $80- 100 million.  Thus, by resorting to tolls we will pay to widen the road one hundred and thirty times over.

These numbers lead to two conclusions.  First, if they are correct, the I-77 toll lane project will be an economic catastrophe, siphoning billions of dollars out of the local economy.  Indeed, in 2035 tolls will cost every person in the region an average of over $500 per year.

However, I do not believe these numbers are even close to reality.  The toll lanes serve one of the smallest metropolitan areas yet are expected to be among the highest grossing in the country. The predicted toll revenues have thus far been historically unattainable in all but two metropolitan areas (LA and DC). If the bond underwriters and NCDOT used this model for their ratings and approval, they have put hundreds of millions of taxpayer dollars at risk- profligate risk. Unfortunately this appears to be the case; the memo containing these figures is titled “Fitch Equity T&R Case”.  (T&R – Traffic and Revenue.) Fitch was the ratings agency hired to review the project’s creditworthiness, and they gave it a BBB- rating, meaning the project is of “investment grade” and therefore carries “low to moderate credit risk.”

So the taxpayer stands to be on the hook for hundreds of millions… unless there is another sleight of hand yet to be played.  As it turns out, there is.

About these projections

Traffic and revenue projections for toll lanes have been notoriously optimistic.  One study of 26 toll roads in the U.S. showed that on average revenue projections were more than double actual revenues for the first five years of operation.[1] Stantec, the firm who made the I-77 projections, also did the projections for the Capitol Expressway. They predicted revenues of $54.7 million in the first year of operations. Actual revenues were less than a third of this.[2]

Given this history and the economics surrounding the I-77 case, how could a ratings agency in good conscience believe the I-77 project carries “low to moderate” risk?

Because in order to secure the required “investment grade” bond rating, the NCDOT agreed to kick in up to $12 million per year anytime toll revenues fall short. (This provision is referred to by the innocuous-sounding term “Developer Ratio Adjustment Mechanism, or DRAM. The total taxpayer liability under the DRAM is capped at $75 million.)

For months I have been saying the toll lane financials do not make sense and that no ethical bond rating agency would ever rate the project as sound. What I failed to envision is a scenario where the taxpayer directly subsidizes toll revenues for the private company. That is what is happening here.

The public bears the risk while profits are privatized.

The original maximum public contribution was capped at a total of $170 million.  When the lone bid came in requiring a taxpayer contribution of “only” $88 million, the NCDOT and some local officials made quite a point of telling the public the “good” news. They never mentioned another $75 million in taxpayer-funded subsidies.

However, in their summary to the NC Legislature, the NCDOT did note that “during the 50 year contract period, (Cintra) will … bear substantial revenue risk.”

Travel Time & Congestion

The report estimates a morning trip from Mooresville to Charlotte takes about 39 minutes today.  By the year 2035 that time will nearly double, to 1 hour and 12 minutes.  The most congested section is from Catawba to Hambright, a five mile stretch where the average speed today is 25mph. By 2035 traffic is expected to crawl at 13 mph.

Northbound in the afternoon doesn’t fare much better, where the average speed in 2035 is a pedestrian 15mph and the trip home will take you 1 hour and 9 minutes. Congestion mitigation is, of course, no concern to the private company- in fact quite the opposite. The more congestion, the more they can charge for tolls. The consultants admit that operating the toll lanes to maximize revenue is “a condition where their utilization would be well below capacity, thus by definition not providing the maximum congestion relief in the corridor.”

Yet in their official report to the legislature and governor, the NCDOT resorted to a half-truth saying managed lanes will help “ease congestion on the free lanes for other drivers.” They failed to mention  average commute times
are expected to nearly double in less than twenty years.

Capital Structure

The total project cost is estimated at $655 million. Of this the taxpayer will directly contribute $88 million. $315 million comes from government-backed loans from two sources- TIFIA loans and private activity bonds, or PABs. These loans are guaranteed by the taxpayer, meaning if the project fails the taxpayer is liable for a bailout. The private company is also providing an equity infusion of $234 million, and $16 million in deferred interest payments will be added to the outstanding debt.

The average cost of the financing, called the weighted average cost of capital, or WACC, is 7.4%. This is significantly higher than interest rates on other collateral-backed loans. For instance, 30 year mortgages on private homes are less than 4%. Average interest rates on municipal bonds are around 4%.

This financing method is therefore very expensive. Assuming a 30 year term, the total interest paid on the project at the WACC interest rate is $832 million. If the project was instead financed through municipal bonds, total interest paid would be $403 million. If the project was funded through normal allocation means, the interest cost would be zero. Granted, this would be a few years down the road, but there is no definitive answer on when funding would be available because the NCDOT has never ranked an alternate project.

Summary

• Initial tolls are expected to be $9 one way, escalating to $20 one way by 2035
• Total toll payments are projected to be $13 billion over the life of the contract
• Travel times on the free lanes are expected to nearly double in less than twenty years
• The taxpayer will subsidize any shortfall in toll revenues up to $75 million

-Kurt

[1] Wilbur Smith’s Traffic and Revenue Forecasts: Plenty of Room for Error, Reston Citizen’s Association, January 27, 2012

[2] “495 Express Lanes went to financing in 2010 with boom-time projections….”,Toll Roads News, October 11, 2013

Source documents:
I77HOTTechmemos     I-77 JLTO Report 4-25-14     I-77 JLCGO Report 4-25-14