The decision to use a public-private partnership to complete the Honolulu rail project has always been a bit of a gamble, but now the stakes are getting higher, and the city is all in.

When rail officials first publicly pitched the plan to use a public-private partnership or P3 agreement to build the last four miles of rail line through the urban core, the rail authority board of directors was assured there was an “off-ramp” or backup plan ready if the unique and complex P3 plan didn’t work out.

But the Honolulu Authority for Rapid Transportation has now told federal officials the so-called off-ramp is “not feasible at this time, and (it) believes that the P3 procurement will be successful,” according to a new report from Hill International Inc., a consultant for the Federal Transit Administration.

That is a significant shift from just April 18 when HART Executive Director Andrew Robbins assured the rail board his staff was ready with alternatives if the P3 procurement effort failed. Robbins said at the time there were “two or three, I would say, probably more, but two potential off-ramps that we could employ.”

One obvious risk is the possibility that rail construction or operation costs might turn out to be too high under the P3 approach. If that happened, Robbins said at the time, the city could adopt one of the alternative off-ramp approaches and still remain on schedule to open the completed 20-mile rail line by the end of 2025.

Robbins made those comments as HART was on the verge of releasing the second phase of the P3 procurement, a major step down the P3 road. But now Robbins says the “off-ramp” is unnecessary because HART can always revise its financing or design components to cut costs if P3 bids come in too high.

The off-ramp was originally developed as a contingency plan to be used if private companies were not interested in bidding on the proposed P3 contract, Robbins said in a recent interview. Now that companies are actively competing for the contract, he said the backup plan is unnecessary.

Honolulu Mayor Kirk Caldwell said in a written statement that when HART approached the administration with the P3 idea, it “provided assurances that an ‘off-ramp’ would be available if the P3 option fell through.”

“If HART is now saying an ‘off-ramp’ is no longer required, it is incumbent upon HART to ensure a P3 partnership successfully completes the city center build out and the Pearl Highlands transit center as promised, without adding significant costs to the project or financial risk to the city and its taxpayers.”

The proposed P3 contract would be the largest in city history, involving an estimated $1.4 billion in construction work. That piece of the contract would include 4.1 miles of elevated rail line, eight train stations in the urban core, and the Pearl Highlands transit center, along with a 1,600-stall parking garage.

The winning P3 bidder would also maintain and operate the entire 20-mile rail line for 30 years, which may be the juiciest part of the deal.

Operations contract

HART’s “moderate range” estimate for the annual cost of operations and maintenance starts at $137 million in 2026, and increases to $169 million by 2036. That means the contract to operate and maintain the rail system for 30 years may be worth considerably more than $4 billion.

Last week Robbins told the rail board the P3 procurement process is now in its “fourth quarter,” and praised the work by staff and rail consultants in answering more than 600 queries from the bidders, who are not being publicly identified. About 150 additional inquiries from bidders are pending, he said.

HART also announced last week the P3 bidders have all asked for more time to prepare their bids, and the rail authority has agreed. The deadline for the critically important price proposals from the P3 bidders had been set for Dec. 6, but Robbins told the HART board last week that deadline has been extended to Jan. 29.

The rail authority will also delay issuing the notice of award of the P3 contract by a month, until Feb. 23, Robbins said. That marks the third time HART has pushed back that particular milestone, and Robbins said there may be more delays. “It’s possible, and I just want everybody to know that is a possibility,” he told reporters Thursday.

Up to this point, for the $9.2 billion rail project HART has generally used a more straightforward “design-­build” approach in which contractors bid to design and build a segment of the project. But that approach resulted in hugely expensive change orders and claims from contractors that contributed to billions of dollars in rail cost overruns.

Construction of the rail line and stations through the dense city center is one of the one of the most difficult and complex pieces of the project, and the rail authority has attempted to bid out that work before. In December 2015, HART froze the solicitation of bids for the design and construction of that last four-mile segment of rail.

Last fall the rail board authorized Robbins to pursue the new P3 approach to procurement for the city center portion of the rail line, a decision board member Terrence Lee described at the time as a worthwhile “gamble.”

However, Lee also expressed concerns that if HART goes through the P3 process and the bids are not acceptable, “have we lost time — critical time — to pursue the only option we have, which is (design- build)? Have we lost critical time such that we’ve raised the cost of the project?” Delays and contractors’ delay claims have also been major contributors to rail’s cost overruns.

Shifting risk

P3 supporters say it will shift risk to the P3 contractor because the winning bidder would be responsible for dealing with change orders and other unanticipated expenses. Caldwell has said the P3 approach also offers an incentive to developers to reduce the front-end construction costs of the project in the hope of winning the contract for the 30 years of payments to maintain and operate the rail system.

A consultant hired by HART estimated in a study last year that the city will save $46 million in construction costs and $300 million in maintenance and operating costs by using the P3 approach.

However, the FTA is watching to determine how much the public-private partnership will actually cost, and will not release $744 million in promised federal funding until it sees the actual P3 price tag, Robbins said. The most recent delays in the P3 procurement may delay the FTA release of the federal funding, which HART had predicted would happen in February, he said.

Caldwell reminded the HART board a year ago that HART and the state have never used the P3 method to develop a project before, and described the board as “pioneers” of the public-private partnership approach. Caldwell supported the P3 strategy as a way to keep the last phase of rail on schedule and on budget, but he also raised some concerns about it.

Caldwell wanted the specifications for the P3 agreement to spell out that the city cannot be required to absorb any more cost overruns, and told the HART board last year the P3 procurement package should limit the city’s annual maintenance and operation payments to no more than the amounts projected in the rail financial plan.

At the time Caldwell addressed the rail board, those payments were expected to be about $127 million a year starting in 2025, with increases in later years for inflation. Last fall HART adjusted that estimate of annual maintenance and operations costs upward in its rail recovery plan to $137 million a year starting in 2026, the first year that the entire 20-mile rail line is supposed to be open.

HART and the city provided the P3 bidders in July with “affordability limits” — meaning the upper limits of what the city is willing to spend on construction, maintenance and operations costs — but those P3 price caps have not been made public.

HART and the city also refuse to disclose the other detailed specifications incorporated into the P3 procurement package, saying state procurement law does not allow them to do so.

Immediate feedback

Robbins told the HART board a year ago that there would be “pretty immediate feedback” if the P3 bidders could not meet the city’s affordability limits. If that happened, “we have looked at the concept of an off-ramp as well so we don’t lose time and we can move into, perhaps, an alternate project delivery,” Robbins said at the time.

If the P3 strategy didn’t work, the backup plan was to award a more conventional design-build contract to one of the competitors that had been seeking the P3 contract, according to the Hill reports.

More recently, Robbins told the Star-Advertiser in an interview that the term “off-ramp” spooked potential P3 bidders somewhat because it might imply the city planned to cancel the project. That isn’t what the city meant at all, he said, because “we’re not here to cancel the (city center guideway). It’s the last part of the project to construct.”

If the P3 bids are too expensive, Robbins said the city will break down the bids to see where the problem might be. If the cost of financing is the problem — the P3 bidders are required to finance the rail construction up front — then the city will re-evaluate the mix of public and private financing, he said.

If the problem is with the actual cost of designing and building the last segment of rail, the city would look to design proposals from the bidders to try to reduce those costs.

Likewise, if the problem with the P3 bids is the projected cost of the operations and maintenance of rail, the city would “look at the scope and look at the service, and presumably make some adjustments to those requirements to make it fit within the budget,” Robbins said.

“It’s like any other project whether it’s P3 or not P3, there’s a project and there’s only so much money, and you’ve got to make it work, and that’s what we’re working hard to do,” he said.

Star-Advertiser reporter Gordon Pang contributed to this report.