Civil Beat Analysis: IMG’s Rail Operating Cost Off By Miles
Editor’s Note: After three months of work, a state consultant in early December released a report reviewing Honolulu’s financial plan for its proposed $5.5 billion rail project. Civil Beat is taking a closer look at the two analyses and will evaluate the key areas in which they differ: construction revenue, construction costs, operating revenue and operating costs.
When it comes to long-range financial projections, there’s only one thing we can be sure of: They’re all wrong. But finding out who’s closest requires a look at the methodology and assumptions each employed in coming to their conclusions.
We’ve already explored the city’s optimistic projections for tax revenue to pay for construction, the state consultant’s $227 million error in comparing construction costs and arbitrary federal funding estimates. This story, the fourth in our series, explores how mileage and hours will impact the cost of the transit system.
Civil Beat’s conclusion on the financial future of rail in Honolulu will follow. A discussion of the topic has already begun.
It costs money to run a transit system. Lots of money.
The state consultant hired to analyze the Honolulu rail project’s finances says the city’s plan for operation, maintenance and major equipment replacement underestimated those costs to the tune of some $500 million over the next 20 years.
That sum, the consultant said, would be added to the $4.7 billion operating subsidy that Honolulu taxpayers will need to chip in to support public transit — the vast majority of which is for TheBus. The $5-billion-plus price tag is on top of the $5-billion-plus cost to construct the rail project.
But the consultant, , made a number of questionable assumptions to reach its $500 million conclusion. Civil Beat found its numbers don’t appear to add up.
IMG’s analysis fails to account appropriately for two big factors — hours of operation and miles traveled. Together they’re used to determine average speed. The result is an overstatement of potential cost overruns.
For every hour a bus is in operation, a driver needs to be paid. And for every mile traveled, the city needs to buy fuel.1
The reason the consultant’s estimate is off base is complicated.
After reviewing data from 2009 and 2010 — years for which there are actual results that can be compared to the projections the city produced for its financial plan — IMG decided the city didn’t use the commonly-accepted federal definition of vehicle hours.2 The city disputes that allegation, but its figures for the past two years are off just the same.
The city projected the average bus speed at 14.8 miles per hour for 2009 and 2010, but the ‘s National Transit Database shows it was actually 13.2 miles per hour.
Source: Infrastructure Management Group — Enlarge Image
IMG calculated the impact of the “correction” — fixing the city’s model to fit recent history — at $311 million for bus operations between now and 2030. And, because “it appears logical to assume that, if the bus operating cost methodology had this issue, so does the rail operating cost calculation,” it tacked on another $29 million for the train. Together, that’s a $340 million increase.3
An analysis of the disagreement shows that while there may well be some increase in costs, the data doesn’t support a hike as big as the one IMG is projecting. The reason: IMG holds the city’s mileage projections steady and corrects the speed discrepancy solely by increasing the number of hours.
But in reality, speed can be lowered not only by increasing the number of operating hours but also by decreasing the number of miles traveled. That’s what actually happened in 2009 and 2010 — the city’s projections underestimated hours by more than 5 percent, but they also overestimated miles by 6 percent.4
If that trend were to continue — if the city’s bus fleet travels fewer miles than expected — it would actually mean a savings of millions of dollars in operating costs versus the city’s model. It’s not clear if such savings would totally cancel out increased costs due to more service hours or merely mitigate them.
Operating costs are in part a function of transit ridership. IMG has said the city’s projections for ridership are far too high but left the city’s numbers alone because it could not come up with its own model. (Civil Beat touched on ridership projections briefly in a previous story.)
Regardless of who turns out to be right, IMG’s projected $340 million increase in operations and maintenance costs seems to be disproportionate.
If ridership ends up far lower than the city thought it would be, operating costs will drop. And if ridership even approaches the city’s projections, it would negate part of the consultant’s rationale for an increase in vehicle hours, again holding down any cost increase.
The bus speeds discrepancy accounts for the biggest difference between the city’s projections and IMG’s projections for post-construction transit costs, but it’s not the only piece.
The actual difference in operations and maintenance is $484 million. About $144 million of that is due to a two-year inflationary adjustment that IMG added because it says the project is already behind schedule without a federal Record of Decision.
And then there’s ongoing capital expenditures, like purchasing new buses and rail cars after the system is up and running.
Rehabilitation and Replacement
IMG says that the city will need to pick up the tab for an extra $111 million in ongoing capital expenditures because it declined to include the cost of doing major repairs or making replacement purchases in the first decade of running the train.
The city’s financial plan estimates that it will spend no money on rail rehab and replacement until 2026, when it will spend $1 million, then $5 million in 2027, $11 million in 2028, $18 million in 2029 and $13 million in 2030 for a total of $49 million.5 Another $75 million will be spent between 2024 and 2025 for nine more railcars to supplement those purchased during construction.
Source: — Enlarge Image
The bulk of ongoing capital expenditures — $1.044 billion of $1.382 billion between now and 2030 — will go to acquiring buses to connect to the train.
IMG said that the city “highly underestimated” renewal and replacement costs for rail. It said those costs would start sooner than 2026. Its projections for ongoing capital expenditures over the next 20 years included $131 million for rail rehabilitation and replacement, up 167 percent from the city’s $49 million figure.6
In its response to IMG’s report released Thursday, the city said rail vehicles will have a 25-year life. Honolulu will still have a “very young fleet” given that the majority of the cars will be delivered after 2015, so its estimate of spending no money on rehab and replacement until 2026 is “reasonable,” the city said.
