How Hartford Went Astray
Flawed Process Hurt Project Cost Estimates
D'Avion Martin, an eighth grader, cashes out eighth grader Samantha Walz in the newly renovated cafeteria at Hartford Middle School in White River Junction, Vt., on April 10, 2014.
Valley News - Sarah Priestap
Students dressed as potato heads exit the newly-renovated gymnasium at Hartford Middle School in White River Junction, Vt., after a presentation as a part of the Spring Carnival on April 10, 2014.
Valley News - Sarah Priestap
Town Manager Hunter Rieseberg, standing in front of the old library that was damaged during Tropical Storm Irene, outlines the plans for the new facility. (Valley News - Sarah Priestap)
Hartford superintendent Tom DeBalsi listens to the discussion during the joint town and school meeting in White River Junction, Vt., on March 29, 2014. (Valley News - Geoff Hansen)
Hartford — Town and school officials committed more than $370,000 to a local architect to design a bundle of recreation projects that voters approved through a joint town-school bond of $8.9 million in the spring of 2013.
So having allocated that kind of money, how did Hartford come up so short in its cost estimates of the various components of the project?
Reconstructing the precise method used by town and school officials to estimate project costs is all but impossible because the joint committee that did the work failed to take minutes at its meetings, despite a state law that requires all “public bodies” to do so. In general, though, many of those who were involved in the planning attribute their off-target estimates to timing: Almost all of the money spent on design was expended after the bond was approved, making the estimates used to write the bond proposal little more than rough estimates.
A lack of funds prior to the bond vote combined with a tight time line forced them to deviate from the standard practice of undertaking the design work before forecasting project costs, officials say.
“We didn’t pay for engineering studies,” said School Board member Lori Dickerson who served on the four-person joint committee that put together the cost estimates. “And you get what you pay for.”
The consequences were serious. Hartford school officials ended up asking voters for an additional $3 million last month — a request voters overwhelming rejected.
The bond that won approval at the 2013 Town Meeting allotted $5.74 million to the town and $3.25 million to the school district.
The town planned to spend its portion on creating several playing fields at the Maxfield property along Route 5 for $3.1 million, renovating the aging Wendell A. Barwood Arena for $2.5 million and constructing a new pump house near the Sherman Manning pool on the high school campus for $140,000.
On the town side, the Maxfield property and pump house projects appear to be on schedule to be completed for the targeted amounts. Not so for the Barwood arena, however: Bids on that project left the town with a $405,000 to $1.2 million s hortfall, forcing officials to scale back some of the work and cover the remainder, in part, with money shifted from the town’s operating budget.
Cost estimates for the school projects proved to be even more off target.
Work at the middle school — renovating its gymnasium and cafeteria — ended up costing $1.5 million, not the $900,000 originally estimated. The middle school overrun of $600,000 ate up money that was supposed to be spent on a new $1.55 million field house and a $800,000 track and all-weather turf field.
But even if money hadn’t been shifted to cover the middle school project, school officials would have come up short of what they needed for the track and turf field and field house. That’s because project planners hadn’t budgeted $1.7 million for site preparation work.
Now, school officials have $4.5 million of work to do on the track and turf field and field house and only $1.5 million to do it. In the meantime, $600,000 in state-of-the-art fitness equipment obtained with a grant sits in storage.
“Routinely, the numbers do not match the estimates,” Town Manager Hunter Rieseberg said about estimating the cost of major projects. “I think what is different about this is they are usually within 5 or 10 percent; they aren’t usually (this far) off.”
Ideally, Rieseberg said, once a community comes up with a project and decides to move forward with it, designs are made. When a final design is completed, the project then goes out to bid. When the bids are received and show what the project will cost, a bond request is compiled and put out to voters.
This process ensures accurate estimates because bidders are committing themselves to “do the project for that amount of money,” Rieseberg said.
Such a process, however, can take “a year to 18 months,” a time frame the joint committee didn’t have to meet deadlines for a vote in spring 2013 , Rieseberg said.
School Superintendent Tom DeBalsi’s conception of what process should be routinely followed to gauge project costs was similar to Rieseberg’s. The school superintendent said that project designs should be in hand before costs are estimated, although not necessarily put out to bid.
In any case, neither approach was followed for several projects under the recreation bond.
“Time really dictates what method you use,” Rieseberg said.
The committee that put the bond proposal together formed around September 2012 after the town and school district decided to look at jointly tackling recreational needs. The committee, comprised of two School Board and two Selectboard members, was assigned several responsibilities: e xploring whether the town and school district could enter into a joint bond; deciding what projects would be included; and estimating the projects’ costs, said committee member Ken Parker, a Hartford selectman who owns an insurance agency.
The committee calculated the cost estimates used in the bond proposal from information gathered by local architects, engineers and athletic groups that would benefit from the recreation projects. There’s conflicting accounts of exactly how those estimates were proposed, though.
Parker maintains that Rieseberg and DeBalsi brought the numbers forward based on information gathered by the local architects, engineers and athletic groups, as the committee “relied on people who had their fingers on the pulse of what these projects might cost.”
Rieseberg, however, says he played no role in the process. “I don’t know who they are; I don’t know how they decided on the numbers; I don’t know who they consulted to arrive at the numbers.”
DeBalsi said in an email that “the committee worked on (coming up with the estimates) together as I recall.”
Three committee members interviewed — Eric Michaels, one of the two School Board members, didn’t respond to requests for comment — said that the process used to come up with estimates for each component of the project varied and partly explains why some were more accurate than others. Here’s how they differed:
∎ The estimate that missed by the widest mark was for the track and turf field at the high school — projected at $800,000, but now believed to be closer to $3 million.
The school district later discovered that the project entailed far more work than the committee factored in.
Dickerson, a financial manager at Dartmouth’s Geisel School of Medicine, said athletic groups, such as Friends of Hartford Track, were consulted about how much the project would cost. That figure was crosschecked with towns who completed similar projects.
Although the committee was well aware that the Hartford High site was different than the schools used to develop the estimates, Dickerson said, members nonetheless neglected to factor in such variables as site layout and soil composition.
Hartford officials have since discovered that site preparation alone will cost $1.7 million.
“We never in a million years thought that moving dirt was going to cost that much money,” Dickerson said. “That was mind blowing to me. ”
∎ To project the cost of overhauling the Hartford Memorial Middle School gymnasium and cafeteria and to build a field house at the high school, the committee enlisted an area firm to provide estimates. However, that company was not working off of full design plans, but rather did on-site assessments to come up with an estimate for both projects, Dickerson said.
Notes taken by Dickerson at the committee meetings, which were provided to the Valley News, show ed that the firm took the square-footage that would undergo construction and multiplied it by $194 per square foot as a way to help come up with both estimates.
Committee members now believe that the absence of specific design plans was the cause of their off-target estimate on the middle school. (The field house estimate appears to be on target.)
“Someone in the business took a crack at what it would cost,” committee member Simon Dennis, a selectman who directs a White River Junction-based nonprofit, said of the process followed.
“We didn’t pay him,” Dickerson said of the firm that did the estimates for the middle school and field house.
Dickerson’s notes list an individual’s first name near the cost estimates for the projects, but she said she couldn’t remember his full name. “He was very clear that he said these are a rough idea” (of what it would cost).
DeBalsi, who was “there to help (the committee) in any way I could,” said the 67 percent cost overrun at the middle school was a combination of the estimates not being “perfect,” and “more work had to be completed than anticipated.”
More asbestos needed to be removed than was anticipated, DeBalsi said by way of offering an example of one of dozens of change orders that were issued during the middle school renovation.
∎ David Laurin, a licensed architect with an office in White River Junction, did the cost estimate for the Wendell A. Barwood Arena renovation. Laurin came up with his numbers based on the arena’s design plans, and later bounced his figures off of “a couple different contractors,” Laurin said.
Parker said design plans were available for the arena project, but not for some of the other projects, because Laurin has been involved with past attempts to renovate the arena.
Working off designs that were specific to the arena provided a measure of accuracy that was missing from the track and field, middle school and field house projects, Dennis said.
The arena was the only recreation projection for which Laurin’s estimating work occurred before the bonding was secured. He said he did the best he could when coming up with a project estimate.
“We looked hard and we made decisions about the way the structure was going to be built. We made decisions about the finishes, we consulted with building professionals,” Laurin said.
Once he came up with an estimate, Laurin said, he met with the committee, and decisions were made to make part of the project optional to limit the possibility of cost overruns.
For example, an addition on the east side of the arena building was made an “add alternate,” which Rieseberg described as a “wish list.”
If officials had decided to build the east addition, the shortfall in funds over the bonded amount would have totaled nearly $1.2 million. That includes more than $700,000 for the east addition, more than $200,000 for utility line upgrades, and a nearly $300,000 overage for the core renovation project.
Parker said the estimates for the arena were calculated a y ear before going out for the bond. By the time the project went out for bid, the market had changed, and the 12 contractors who competed for the project had less incentive to underbid each other.
“We couldn’t gauge with WABA where the market was at the time when we went out to bid,” Parker said. “We didn’t have a good sense of how aggressive contractors were going to be.”
As far as the other projects under the bond, Parker said he felt the market impacted “all of the projects ... that were estimated with some degree of clarity.”
∎ The estimate for the Maxfield athletic complex off Route 5 ended up being spot-on. There’s a reason for that, say those involved: The town had already solicited bids, so the committee knew exactly how much to bond for — $3.1 million. (The project’s total cost is $ 4.1 million, but the town sold 17 acres of land from the donated property for $1 million to help cover expenses.)
“We had ample time to roll out that project and that enabled us to get the information in advance,” Rieseberg said.
In addition, money from the Maxfield property land sale covered full design costs for the project, Parker said.
“(The) money was there to work with,” he said.
∎ Estimating the cost of the Sherman Manning Pool and pump house renovations was fairly straightforward: It was determined by using the equipment estimate the town received from its pool equipment supplier and estimating the cost of a small shed to be built around the equipment, Rieseberg said.
T ime Crunch
Both the town manager and school superintendent attributed much of the problem to the rushed process forced on the committee. Its work started in early autumn 2012 — a year after Tropical Storm Irene caused substantial damage in Hartford . To meet legal requirements for adequate warning to have a bond vote at the 2013 Town Meeting, the bond article had to be ready by the first week of January, DeBalsi said.
“We did the best we could under the time we had,” Rieseberg said. “We didn’t have the ideal amount of time to fully vet any of these projects. And that is not untypical.”
Rieseberg said it is not uncommon to use a “construction company’s and estimating firms” numbers to cost out a project, and it is in fact an “acceptable practice.”
“It simply didn’t work so well this time,” he said.
Dennis said he felt pressure to move forward with a bond vote out of recognition that the cost of materials and construction inevitably climb as time elapses.
Dickerson said that the 2013 Town Meeting season seemed the opportune time to propose the bond because market conditions for undertaking such projects were favorable, or seemed to be.
“You strike when the iron is hot,” she said.
Parker, though, said he felt the committee had enough time to get done what it needed to.
“I think the time was right to put it to the community before we engaged in any further expense,” he said.
So Far Off
Parker said he thought the numbers for town projects were off because the committee didn’t or couldn’t accurately gauge the contracting market. He blamed the school’s miscalculations on inadequate research.
“There wasn’t an adequate degree of inquiry and examination of what the projects entailed,” Parker said of the school projects. “Hunter, for us, has a good handle of how you go about soliciting bids and getting pricing done and managing construction projects.
“I don’t think the school side has a group of people who are experienced in doing that,” Parker said. “I am not sure that they came to the table prepared enough.”
Acknowledging that the committee’s work “was a joint effort,” he said, “I am not pointing fingers.”
Dickerson was less sure about what went wrong.
“I really can’t say,” she said. “We had very good people who were putting their best efforts forward. Do I blame anybody? No; we are talking about honest citizens that are thinking the best for our town.”
In terms of doing due diligence by spending money on engineering studies before coming up with estimates, both Dickerson and Parker said there was no money to be spent.
DeBalsi doesn’t recall any discussion about the need to undertake such studies.
He did say, though, “There was no pot of money to spend to get real accurate numbers.”
Even if there had been money available to spend on full design studies, Dickerson said, doing more precise estimates might have created different problems.
“If we put the money toward it, people would have come back and said, ‘You spent $60,000, and you didn’t have the approval to begin with,’ ” she said. “We didn’t have that kind of money. We didn’t even know if it was going to pass.”
Dennis said he would rather not spend money on design plans beforehand, even if it produces more accurate estimates.
“I think it is irresponsible to invest hundreds of thousands of dollars before bonds have been passed,” he said. “You want to go into the bond process asking for the money with the ability to walk away from it.”
Full architecture and engineering studies can cost “somewhere in the vicinity of roughly 6 to 15 percent of a project’s actual construction cost,” Rieseberg said.
Might the committee have tried too hard to keep total spending low as a way to get the votes needed to pass the bond?
Each person interviewed denied that was the case, saying the committee made a good-faith effort and did the best it could with the resources at hand.
Resolving contradictions in how committee members recall events is impossible because of the failure to keep minutes of meetings. Although Dickerson kept notes, which she shared with the Valley News, they shed little light on how the committee arrived at its estimates .
Vermont Secretary of State Jim Condos declined to comment on the case specifically but wrote in an email that minutes generally are required of all “public bodies,” which state law defines as “any board, council or commission of the state or one or more of its political subdivisions, any board, council or commission of any agency, authority or instrumentality of the state or one or more of its political subdivisions, or any committee of any of the foregoing boards, councils or commissions.”
Condos wrote that the minutes “shall cover all topics and motions that arise at the meeting and give a true indication of the business of the meeting.”
Parker and Dickerson both said it didn’t cross their minds to record and accept official minutes into the record after each meeting.
“I don’t have an answer,” Parker said. “... Maybe it was just something we didn’t think about doing.”
“I guess ... we never even though t about taking minutes. We were tasked with a job. We had a very short time line. It was just not a thought,” Dickerson said.
Resident Tim Fariel, a former School Board member who often attends public meetings in Hartford, said that official boards should learn a lesson from what has happened to the recreation bond.
“If you are going to do bond work, you probably should allocate money on the front end to do your research,” he said. “It reduces the uncertainty.”
“I would be more insistent that if we are going to do a capital project, that we spend the money that is necessary to do a proper evaluation (beforehand),” Parker said. “If you are going to commit to something, you make a commitment to put the money up to gauge what needs to be done.”
Dickerson had a slightly different view about the steps that should be taken before bond money is in hand.
If a joint bond is undertaken again, she said, a committee should do a study first that would allow residents to weigh in on various aspects.
One question, she said, should entail, “If we wanted to do this and there was a cost of (doing) preliminary work, would you be willing to pay for that?” Dickerson said. “Let’s get the pulse of the community first.”
The superintendent, too, said the process used for future projects must be different.
“Even if time is of the essence ...” DeBalsi said. “I would never fall into this situation again.”
Jordan Cuddemi can be reached at email@example.com or 603-727-3248.