Greenville Southern Connector up for concession offers - not-for-profit ailing
By Peter Samuel
The Greenville Southern Connector (GSC) is on offer to patient investors. As a not-for-profit (NFP) tollroad financed with conventional fixed interest bonds it is inexorably going broke. Last summer the Connector 2000 Association (C2A) was approached by an investor group (identify still undisclosed) with a proposal to take over the ailing non-profit, says general manager Peter Femia. However after extended discussion they decided in about Nov 2006 to close down the negotiations with the unsolicited bidder and go for competitive bids and a open selection process- now under way.
The C2A operates the GSC tollroad under a 50 year license from South Carolina DOT.
SCDOT and the board of the C2A have to both sign off on any concession, but they are both agreed on trying for one.
The association is in financial trouble and headed for default on its debt.
Requests for qualifications for a concession bidder are due Nov 14 with the following timetable:
2007-10-17 Meeting held for interested parties (see list attendees below)
2007-11-14 Qualifications deadline
2007-11-16 Confidentiality agreements signed with qualifieds
2008-02-15 Conclude concession agreement
2008-03-15 Receive bids on concession agreement
Toll rates studies by URS - required under the terms of the bond covenants - have reported there is no realistic prospect that traffic and revenues can be boosted sufficiently to make the tollroad financially viable under its present not-for-profit capital structure.
The road is now in its seventh year. Traffic is around 16k transactions/day still under half of the WSA forecast (34k) on which the business plan was built. Traffic has growing solidly but from a very low base.
Toll revenue is slightly further behind forecast since truck traffic paying higher per vehicle tolls has been especially scarce.
Revenues should, according to the original forecast, be of the order of $14m/yr now, but they are about $5.4m - 38% of forecast.
Operating expenses ($2.5m) are easily covered by tolls but the operating surplus of $2.9m is being swamped by interest costs ($16.6m) and other smaller obligations such that the latest year shows a "decrease in net assets" (a loss, we called it in more candid days) of around $20m that has to be covered by running down reserve funds. The C2A has debt now of around $290m.
A toll increase made in early 2005 showed quite high responsiveness of traffic to toll rates - a price elasticity of demand estimated by URS at -0.44, meaning that a 100% increase in tolls would reduce traffic by 44% and increase revenues only 12%. URS reckon that toll rates can be pushed up about 50% above current levels before going over the hump and actually reducing revenue.
The C2A plans to increase tolls by about that in two stages.
Cars presently pay 7.8c/km (12.5c/mile) cash and 5.8c/km (9.4c/mile) for the length of the road. 5-axle tractor trailers pay just over three times the car rate - $3.00 cash and $2.25 transponder at each mainline plaza.
There is no vehicle classification at the ramps where all vehicles pay the same 50c tolls.
Tolls are planned to rise to $1.50 for cars cash by 2010 or 11.7c/km (18.8c/mile) at the mainline plazas. This is said to be close to estimates of revenue maximizing toll rates (63% higher than now).
Southern connector is "south of south"
The major east-west longdistance highway through the area, I-85, was built south of Greenville as it existed in the 1950s but by now the developed area has spread south of it. The Greenville Southern Connector is designated I-185. Along with a free section of I-185, and I-385 the Connector forms something of a belt route around the south of the Greenville area.
The road was justified on two grounds:
- relieving congestion on parallel I-85 and providing a quicker connection between Atlanta and Columbia
- providing improved connections for local traffic on the developing southern fringe of the area
The first full calendar year 2002 saw an average daily traffic of 10.9k, rising to 14.1k in 2004. With the toll increase beginning 2005 there was a slight (3.5%) decline in traffic to 13.6k that year. 2006 traffic was back up to 14.1k.
The first nine months of 2007 show an 8.5% increase in traffic over 2006 to 16.2k/day. Revenues are running at around $5.4m/yr.
Trucks constitute less than 4% of the traffic and 8% of revenue.
Overestimation of traffic
Peter Femia the Association general manager says he thinks the poor state of the economy was a major factor in the undershoot. After years of growth manufacturing in Greenville took a major hit just as the tollroad was going into construction. Factories closed. New development almost dried up. But the economy has rebounded since.
Traffic also appears to have been overestimated by WSA because they made mistakes in estimating travel times on the tollroad vs I-85 and other routes. In the planning stages WSA had estimated significant time savings on several typical routes.
URS however reports based on 2006 measurements of actual trip times:
"The Southern Connector offers little distance saving for these movements, and in most cases the distance is longer using the Southern Connector. A time and distance comparison was made between taking I-85 directly or taking the Southern Connector as a bypass of Greenville. The route including the Southern Connector (GSC) was longer and took more time. This displays the lack of competitive ability that the Southern Connector currently has with I-85 and the resulting low share of traffic ther Southern Connector carries in this corridor." (p16)
The Connector (GSC) is simply too indirect a route to be competitive with I-85 for through traffic. It heads too far south. Traffic from Atlanta heading for I-385 south to Columbia is taken much too far in (to MP42) on I-85 before they can join the GSC. And then they head a bit back west of south for several miles before getting back to their eaterly course, providing no distance savings over sticking on I-85 until I-385 (at MP51).
I-85 is mostly 3+3 lanes through the Greenville area and, except when there are incidents, it flows pretty freely at 60 to 70mph (96 to 112km/hr). It has average daily point flows of up to 96k/day vs the Connector's 7.4k highest point flow. One disadvantage of the free route Atlanta-Columbia is the I-85/385 interchange. Its single lane ramps often get clogged up, making the Connector a time saver.
I-85's six lanes can give pretty good service in the range 100k to 120k/day and the Connector doesn't have any guarantees against 4th laning of I-85 in ten years or so. Only if congestion grows on I-85, and it is not 4th laned, will the Connector become an attractive alternative for through traffic during congested times.
The Connector's license agreement with the state limits only the construction of a completely new expressway standard roadway. It doesn't limit what improvements can be made on I-5 and I-385.
Better link to I-85 west
A better link road - see dashed link on map above - between I-85 at MP39 and the Connector at about the interchange with SC20 might attract a lot of Atlanta-Columbia traffic and make the GSC more competitive for other through traffic. But the traffic to be got from such a link may well not justify the cost, and it may face other obstacles.
That leaves the Connector as a roadway serving local traffic and dependent on development on either side of its immediate vicinity. Most of the current users are commuters and people using it for daily local business, URS finds.
When we visited in 2001 a couple of people there said the road was "too southern" and "too far out" to serve local trips efficiently. "You make better time the old way." they said.
The road should have been built a couple of miles north of where it is, they said. Scarce early traffic numbers reflected that sentiment.
That was six years ago.
Development is now filling in around the tollroad, Peter Femia tells us, but this happened less quickly than was assumed in the WSA forecasts in the early years at least. So WSA forecasts of local use turned out to be optimistic too.
Greenville proper only has a population of 400k but with Spartansburg and Anderson forms a consolidated metro area of about 1.2m making it a substantial concentration. Present forecasts are for population growth about average for the US. Greenville proper was just under 300k in 1990, grew 27% in the 1990s to 379k in 2000 and has grown 7.3% to 407k in 2006. It is projected to grow 10.8% to 451k by 2015.
Economic prospects good
North and South Carolina are generally pro-growth and business friendly in their politics. They have a moderate climate and good crossroads location. For several decades these states have attracted people from the misgoverned and overtaxed northeast - places like Massachusetts, New Jersey, New York and Pennsylvania.
Greenville has a classy attractive central area. Land prices and the cost of living are lower than many other places, so the area should do well economically, and with that traffic should grow.
BACKGROUND: The project was developed by Interwest Management run then by Bob Faris a former federal highway administrator turned highway developer sold on the then popular 6320 nor-for-profit fianncing and ownership model. He teamed with WSA, Thrift Bros, Florence & Hutchinson, Mesirow, and Lehman.
They cooperated with the state to form the not-for-profit Connector 2000 Association (C2A) which was to float the bonds and pay for construction of the road and the developer fees in return for the right to toll the road for the 50 year license or concession period. C2A was awarded that concession in Feb 1998.
Design-build cost was $191m, project cost about $240m.The road was built on budget and many months before the contracted date of completion. It opened to traffic Feb 27 2001. Tolling began Mar 14 2001.
The tollroad is 26km (16 miles) long, 2+2 travel lanes and has expressway to expressway interchanges at each end and four intermediate interchanges.
There are two mainline toll plazas with 4 toll lanes each direction. Only one of the four toll lanes, the far left lane (lane 4) is transponder-only and posted for 45mph (72km/hr), a kind of compromise open road tolling.
Lanes 1 and 2 are cash lanes with toll booths. Lane 3 is coin machine. All lanes accept transponder payment.
There are two pairs of ramp plazas of one lane each equipped with coin machine and transponder reader.
The South Carolina electronic toll system called PamettoPass or PalPass uses Mark IV equipmenty identical to E-ZPass but they are not members of the IAG. The transponders are issued by a contracted operator for the Hilton Head Cross Island Parkway on the Atlantic coast.
Tolls rose Jan 2005 from 75c to $1,00 for cars paying cash at the mainline plazas. The ramp plaza tolls remain at 50c for both cash and transponder. The transponder toll went from 60c to 75c at the mainline plazas.
Engineers report of small defects in road
An engineers report - contained with the RFQ documents - shows the road to be in generally good condition, though several bridge transition slabs have sunk and need fixes. At the mainline toll plazas it seems to us the coin machines make little sense with dollar+ tolls. At some point they'll want to make lanes 3 and 4 into a pair of full highway speed toll lanes posted at 65mph (105km/hr). But until traffic volumes are considerably higher that kind of open road tolling isn't urgent.
SCDOT and the C2A are naturally hoping to do a deal under which a concessionaire pays off the C2A debt in full. However the talk among finance people is that the project will need to be "seriously recapitalized" -- a fancy way of saying the lenders will have to agree to take some losses so the new operator is saddled with considerably less debt than the C2A.
The license agreement under which the GSC operates places no restrictions on tolls that can be charged, and obviously from the high price elasticity of demand competition from I-85 limits that. Toll increases to profit maximizing levels, URS estimates, could increase revenues by double digit amounts annually in the next four years - to $9.2m/yr in 2011 from an estimated $5.5m in 2007.
After that it seems likely revenue opportunities will grow much more slowly. URS is working a full traffic and revenue projection for the next 30 years. This should be complete by mid-November Peter Femia told us, and will be published then
for general info see http://www.southernconnector.com
for the RFQ details and background see http://www.southernconnector.com/pdfs/SCTR_RFQ2.pdf
OPINION: NFP 6320 model dead as a dodo
The not-for-profit, tax code 6320 model, so much the rage among advocates of 'alternative' and 'innovative' financing in the 1990s looks to be dead as a dodo. It has an almost 100% failure rate. First the Pocahontas Parkway, now the GSC. That's a worse record even than the 1990s short term Mexican concessions, some of which survived. In both cases the incentives are all wrong. The developers get their money on turnover of the road, so they have no reason to care whether it is a financially sound operation or not when it opens. They've made their money. They're outta there. Someone else's problem.
Only the long term concession, and the longer the better, forces the investors to focus on ownership and operational viability, and creating and maintaining an asset with value.
TOLLROADSnews 2007-10-26 ADDITIONS 2007-10-27