This wasn’t just sticker shock. Three bids were opened last week for a large contract to build nine of the 20 stations being planned as part of the city’s elevated rail project, and even the lowest bid — $294.5 million, from Nan Inc. — came in 60 percent higher than anticipated.
This raises alarm bells, not merely that costs are rising in an accelerating construction market but real fears that they’re doing so at a pace that will burn through even the generous contingency fund set aside for such overruns.
There’s enormous urgency for the Honolulu Authority for Rapid Transportation to learn some lessons from the distressing results of this episode and reform the bidding process going forward to save money.
HART Executive Director Dan Grabauskas said it’s too soon to know whether the station construction bids — which also included proposals priced at $312.3 from Nordic PCL and $320.8 million from Hensel Phelps — signal a truly budget-busting trend. He underscored that the project is still within the budget, given the roughly half a billion left in the contingency fund.
However, with about 35 percent of the project still left to contract for $500 million, and given that bids this time exceeded expectations by about $100 million, that cushion doesn’t feel so comfortable right now.
"Clearly our estimates right now are suspect," Don Horner, vice chairman of the HART board, said last week, which is plainly true. And although the delays caused by rail’s courtroom battles have pushed construction into a pricier market, it’s just as clear that’s not the only problem here. Some major reconfigurations, both in the structure of the work ahead and in the timeline, are in order.
Grabauskas said the executive office is poring through the bids to discover what factors are driving the high prices. He declined to go public with any potential cost-saving ideas, beyond saying that the bid likely relates to the size of the project: nine stations in one contract.
Additional factors include the complexity of the work required and the "compressed" schedule — time lost in court is being recouped through an accelerated timeline, he said, a schedule that may be forcing prospective bidders to increase projections of both labor and equipment costs.
The construction calendar was shortened in an effort to make a 2017 deadline to open part of the system, roughly the first 10 miles. However, officials ought to consider pushing that off if it means significant savings can be achieved.
The Ironclad deadline for the start of operation, under the federal full-funding agreement, is Jan. 1, 2020, which is why HART proposed a 2019 end date. That is the deadline that remains paramount. If the "soft" opening allows at least adequate system testing to happen in advance of 2019, HART shouldn’t be wedded to the 2017 date now circled in red on its calendar.
HART needs to look for other ways of setting up the contracts that would encourage more competitive bidding, given the somewhat limited availability of subcontractors for the longer term of a nine-station project. Officials need to figure whether enough savings await to warrant rebidding this project, but at the very least, restructuring the next contracts in line must be explored.
Grabauskas also said the commitment to the federal government is to budget for a $193 million ending balance when the last check is written to pay off the city’s bond in 2023, a surplus he described as a "second contingency." That balance shouldn’t even come into play.
The "red-hot" construction economy that Grabauskas described may yield enhanced tax revenues for the project, but the authority can’t count on that. HART should work to leave some money in the account to launch a system that’s built well and with fiscal support.
If adjusting the financial planning means the trial run will have to start a year or so later, most taxpayers would see that as preferable to landing in a sea of red ink.
