By Don McIntosh
Ryan Nielsen read the announcement in the Daily Journal of Commerce, a construction industry newspaper: Nonunion Todd Construction had just won another contract with Portland Public Schools, a $4.6 million deal to build a new career training center at Roosevelt High School. The work is part of a decade-long series of property-tax-funded school modernization projects.
Nielsen, a strategic researcher for the District Council of Laborers, has been keeping tabs on Todd Construction because it’s been beating union contractors competing for public construction projects.
Construction is a competitive business, and union officers aren’t the type to gripe and moan when their allied contractors lose out after fair and transparent processes.
What seemed strange to Nielsen about the DJC announcement was that Todd’s bid was the highest of those listed, $280,000 higher than the bid submitted by union-signatory Swinerton Builders.
Nielsen submitted a public records request Nov. 18 for documents detailing how the district scored bids it received. Two days later, they arrived.
Nielsen studied them closely. Of the three firms selected as finalists, not only was Todd the high bidder, but PPS evaluators had also given it the lowest rating. And yet it won the contract anyway.
Based on the documents PPS released, Nielsen was able to reconstruct how that happened. When he presented his findings at a Nov. 24 Zoom meeting of the Columbia Pacific Building Trades Council, union leaders with decades of experience in construction were open mouthed with astonishment, calling it a “jumbled mess,” “a Frankenstein process,” and “categorically alarming.”
“How is this possibly legal?” asked one business manager. “What happened to lowest responsible bidder?”
Based on close scrutiny of the documents, Nielsen thinks he understands how PPS awarded the contract to the highest bidding and lowest rated of the three finalists. It was a multi-stage process, with problematic features at every step of the way.
It starts with three evaluators, identified only as Evaluator A, B, and C. In the first tier, there are 10 firms competing, and they’re given points by each evaluator based on experience, project understanding and delivery, and certified business development strategy—as outlined in each bidder’s written proposal.
Already, there are problems. The evaluators vary quite a bit in the points they award, even on criteria you’d think would be objective, like how much experience a contractor has. And Evaluator C is brutal. Rating experience, A and B each give Skyward Construction 20 points and A.C. Schommer & Sons 15 points; C gives each of those firms 7.5 points.
Next, points are awarded for price, according to some objective formula, so that all three evaluators agree, with the lowest bidder getting the highest points. The points for price are added to the points for the written proposals. Based on that, the bidders are ranked, and the top three are selected as finalists: union-signatory firms P&C Construction and Swinerton Builders, and nonunion Todd Construction.
Now, in the second tier of the process, finalists are invited to give a presentation, which the three evaluators score based on “key personnel experience,” “risk management,” and “collaborative approach.” But again evaluators are all over the map. Each of the three evaluators thinks a different one of the three finalists is the best, and there’s no consensus. There’s also a huge spread in the points that individual evaluators award. Evaluator A’s scores are pretty close together for the three firms. But Evaluator C’s scores vary widely, with the best-rated getting almost twice the points as the worst.
Finally, the first tier and second tier points are added. P&C is the highest rated, and Todd is the lowest. But Todd wins. Because instead of basing the final award on which contractor got the most points, the points are converted again to rankings, and rankings from the first tier are added to rankings from the second tier.
Nielsen says that’s deeply problematic.
“It introduces so much randomness that the district’s procurement department might as well pick the winning contractor out of a hat.”
When you combine rankings from a group of 10 bidders and rankings from a group of 3 bidders, a slightly less favorable ranking from a single evaluator in the first round can kill a bidder’s chances. Swinerton, for example, was hurt in the final calculation because Evaluator C rated it fourth in the first tier, below a bidder that was eliminated. PPS third graders know that you can’t place fourth out of three contestants, but that’s exactly the kind of result generated in the final stage of PPS’ construction procurement process.
To find out independently how PPS ranks construction firm bids—and see if Nielsen’s reconstruction got it right—the Labor Press reached out to PPS Bond Communications Manager David Mayne and Marina Cresswell, senior director of PPS’ Office of School Modernization. The two failed to respond to two emails and three phone messages over a 12-day period.
Leaving aside the dismay union representatives felt on hearing the strangeness of PPS’ bid award system, they might not want to make an issue of it if it wasn’t for concerns about Todd Construction’s record of wage theft. Last year Todd paid $60,355 to settle charges that it had underpaid 38 workers on a school construction project in Albany, Oregon. Now Todd Construction is a party to three prevailing wage cases currently being investigated by the Oregon Bureau of Labor and Industries—two at Portland Public Schools, and one at North Marion School District. [Two of the three investigations are of violations by subcontractors.]
Building trades union leaders plan to take their concerns about the process to elected members of the school board.
FUN WITH MATH AND LOGIC
Based on the documents PPS released, here’s how Nielsen thinks the district’s process produced the strange result [click the link to see the documents]:
- STEP ONE: Evaluators A, B, and C rate written proposals from 10 bidders, awarding points based on experience, project understanding and delivery, and certified business development strategy. When the scores are totaled, P&C Construction and LCG Pence tie for best rating, followed closely by Todd Construction, with Swinerton Builders a somewhat distant fourth.
- STEP TWO: Those written proposal points are combined with points based on bid price to produce a total score (referred to in the spreadsheet as “total offeror points.”) The top three scorers are Todd Construction, P&C Construction, and LCG Pence. Todd is now in the lead because its bid price was substantially lower. But total points aren’t how finalists are chosen. Instead, each bidder’s price scores and written evaluation scores are combined for each evaluator, and that number is used to create a ranking from 1 to 10 for each evaluator. The three ranking numbers are added for each bidder, and the lowest three are selected as the finalists. Now the top three finalists are Todd Construction, P&C Construction, and Swinerton Builders.
- STEP THREE: The three finalists are invited to give a presentation, and the three evaluators rate each on key personnel experience, risk management, and collaborative approach. P&C here is the clear leader, followed by Swinerton a distant second and Todd a distant third.
- STEP FOUR: But again, the total points for the second stage evaluation are in effect tossed out, because again each evaluator’s choices are reduced to rankings. Evaluators B and C rated P&C the highest of the three for its presentation, while Evaluator A rated Todd the best.
- STEP FIVE: When the actual numerical scores for written proposals, bid price, and presentation are totaled, P&C is the clear winner, followed by Swinerton, with Todd a close third. But again, the points don’t matter, because they’re converted to rankings, and the rankings from the first phase (with 10 contestants) are combined with rankings from the second (with three contestants.) Voilà: Todd, lowest of the three finalists in points, wins a contract to build a career training center for $4.6 million.
