We Should Know How Much Transit Components Cost

This piece is part of IFP’s Transit Abundance Playbook, a collection of proposals for reducing American transit construction costs.

Summary

Contracts for large, complex infrastructure projects such as urban rail lines are often lump-sum, obscuring the true cost of work and enabling cost inflation as the project inevitably evolves. Bidders should instead be required to itemize costs in project bids, estimating labor costs by job, material costs by commodity, and systems cost by individual manufactured product. Under the supervision of a new unit within the US Department of Transportation (USDOT), transit agencies should establish a reference point for costs by publishing the expected cost of each component used for transit projects. Publicly itemized costs will reduce conflict over change orders by pricing in each change, whether it is due to market conditions, unexpected delays resulting from geology affecting underground construction, or any other reason. This provides both transit agencies and bidders with greater certainty, reducing costs by an average of 15-20%.

Problem

“Ron Tutor is the greatest change-order artist that I’ve ever seen. He submits the low bid then makes it up on the change orders.”
— Los Angeles Mayor Tom Bradley, 1992

Tutor-Perini was one of the lead contractors on the 1986–1993 Los Angeles Red Line project, the city’s first heavy-rail subway. The cost of the three-mile subway was high for its time even before its 16% cost overrun and totaled $1.45 billion ($3.8 billion in 2025 prices). The subway construction contracts ran over by 35%, and some contracts ran over by 70% or more, many of which were given to Tutor-Perini.¹ Mayor Bradley’s comments on Tutor-Perini pointed to a recurring problem — a 2012 review by the Center for Investigative Reporting found that in the preceding 12 years, this one contractor made $765 million off of change orders on 11 different Bay Area public infrastructure contracts, inflating project costs by 40%.

The civil structures, systems, and tools used in transit projects interact in unexpected ways; no matter how detailed the design phase of an infrastructure project is, mid-stream changes are to be expected. If the project is underground, adjustments may be necessary to account for small-scale geology, unmapped utilities, archeology in old city centers, or equipment failure. Even above-ground construction is unpredictable — unmapped utilities and unanticipated environmental effects, such as noise and street closures, can require last-minute changes.

Contracts get more complex as transit agencies and their contractors try to offload risk onto each other, and their competing priorities often lead to conflict as each side tries to pass the buck. The contractor typically blames another party for the change — one common refrain is that the transit agency did not give the contractor sufficient site access for proper planning; the agency often charges that the contractor failed to perform due diligence and should be liable. This leads to expensive litigation, increasing project costs. Multiple New York, Boston, and Seattle contractors and transit agencies interviewed for the Transit Costs Project have suggested that Tutor-Perini is not an anomaly — a large number of transit construction contractors across the country make their entire profit off of change orders.

To avoid litigation and unforeseen cost increases, American transit agencies have increasingly opted to privatize risk through larger, lump-sum contracts. Under this model, parties negotiate a fixed price for the contract in advance; the transit agency and the contractor each rely on one of a handful of independent cost estimators to arrive at a price, with the project cost breakdown hidden behind non-disclosure agreements (NDAs). Contractors are then responsible for any cost overruns due to project complications.

Despite leaving the contractor responsible for overruns, this model has failed to mitigate change order conflict and merely swept it under the rug. Contracts cannot fully offload transit agencies’ risk, since contractors can walk away and fight any clauses that make them liable for cost overruns in court. Multiple studies suggest that public-sector clients typically give in to political pressure to cut a deal and complete the promised project, regardless of overruns.² The Maryland Purple Line contract is illustrative of this dynamic: the design-build-operate-maintain contract expected the contractor Fluor Enterprises to take on all risk for construction and 36 years of operations. The initial 2016 contract was worth $5.6 billion, including $2 billion in construction costs and $3.6 billion in 36 years of operating costs paid in advance, with opening projected for 2022. Amid delays and cost overruns, Fluor Enterprises walked away in 2020 after having spent half of the construction budget, forcing Maryland to find a new contractor. Project costs grew to $9.3 billion by 2022, and have further climbed to $9.6 billion since.

Interviews with contractors on American urban rail projects routinely surface complaints about risk and uncertainty driving up costs by a minimum of 20%; one interviewee suggested the increase could be as high as 40–50%. In anticipation of inevitable scope creep, delays in dealing with utilities, and geotechnical surprises, contractors inflate their bids — they may not know what will happen, but they know that something will. From the perspective of the contractor, they’re expected to take on risk to the tune of hundreds of millions or even low billions of dollars; one wrong decision can break them. No wonder they bid high.

Solution

Countries such as Italy and Spain achieve lower construction costs by requiring itemized bids on large, complex contracts: bidders are expected to not only name a price, but also substantiate it with detailed breakdowns of labor costs by job, material costs by commodity, and systems cost by individual manufactured product. Simpler fixed-price contracts are only used for smaller, more routine projects, such as fuel purchases or public parking lots, but not for projects like subway tunnel construction. Government agencies maintain public databases with the costs of each item and track market prices, keeping contractors honest and deterring cost inflation. When projects change — and they invariably do, given natural variation in commodity prices, soil conditions, and other factors — an itemized register of inputs ensures that their change orders are priced at their true marginal costs, rather than in opaque lump-sum contracts.

Two studies by Nicholas Ryan and one by Valentin Bolotnyy and Shoshana Vasserman bear this out, showing that itemization improves cost discipline. Their respective studies of Indian power plant construction and Massachusetts road works show that itemization can cut final costs by 15–20%, even if initial project bids are higher. Clearly assigning the cost of each potential change order upfront eliminates the need to litigate or otherwise contest the changes, and guarantees that contractors will not have to eat unforeseen costs. With more certainty and less room for questionable practices like making large profits on change orders, costs fall.

The merits of itemized contracting have been evident in Italy’s implementation of the practice. The 1990s Mani pulite (“Clean hands”) investigations of the Tangentopoli scandal targeted rampant political corruption, including accusations of bribery in securing metro and other public infrastructure contracts; half of the parliament was indicted, and emerging political parties embraced itemization to prevent bribery. Italian legislation established reforms that centered transparency and pressured agencies to publish open data. These reforms have enabled NGOs, citizen-activists, and researchers to observe government activity, detect waste, fraud, and abuse, and suggest improvements.

Today, Italy’s itemized bidding process is the gold standard. The Italian government prepares detailed itemized costs for each work category, commodity, and manufactured product, publishing region-specific cost guidance to account for wage and other cost differences. Contractors are required to base their bids on these costs, but given the latitude to deviate based on spot changes in market conditions, such as rising wages. They need to justify any departures, but the process is manageable and there is no litigation risk. As a result of this system, Italian metro construction costs fell between the 1980s and the 2000s; Milan and smaller Northern Italian cities have been able to build for about $300 million per mile, and even the most complex construction amid historical ruins in the middle of Rome runs $800 million per mile — a fraction of the cost of subway projects in Manhattan or Downtown San Francisco.³

Itemized bidding has been similarly effective in Spain, long hailed as a leader in efficient, low-cost infrastructure construction. The Madrid Metro has built entirely underground extensions for only about $170 million per mile — less than one-tenth of the per-mile cost of New York’s Second Avenue Subway. Madrid Metro CEO Manuel Melis Maynar cited itemization as a key intervention that ensured smooth and on-budget project execution. In effect, this system prices in future changes in costs and limits both absolute costs and cost overruns.

Efficiencies can often arise from catching innocent technical mistakes; in one Boston example, the activist group TransitMatters found that an element in the city’s end-of-day shutdown of its subway system delayed trains and connecting buses so much that fixing it saved $3-4 million a year. But while the US enables some amount of transparency through open data on transit vehicle schedules, information on megaproject procurement is lacking.

American good government reforms date to the Progressive Era, when the tools of corruption were different. Those reforms enabled low-cost subways in the 1910s and ‘20s, generations before the current global engineering and consulting oligopolies formed. Modern reforms must account for how procurement has evolved. 

The US has already embraced itemized costs for road construction. Granted, road projects are usually simpler to itemize than mass transit and can rely on many more real-world projects for calibration; with the exception of urban tunnel construction, American costs for routine road projects are generally more reasonable than transit costs, despite cost increases since the 1960s. 

Implementing itemization for US transit projects would not require much additional effort. Transit agencies and bidders already produce detailed estimates for project costs — bidders need to approximate project costs when compiling their proposals, and agencies estimate project costs as part of their budgeting processes. Yet today, agencies obscure these itemizations through NDAs and trade secret agreements; the lack of transparency diminishes their efficacy as an accountability mechanism. 

The federal government can catalyze itemization by developing reference cost data. Local transit agencies are often understaffed and lack the expertise needed to properly itemize urban rail projects, and per-agency databases would require duplicative effort. The US Department of Transportation (USDOT) should lead a federal data collection project and publish national itemized costs. This data should be updated regularly and account for regional multipliers (such as at the level of a metropolitan statistical area or metropolitan division).

Implementing itemized bidding would require a comprehensive database of materials and individual systems. USDOT already publishes estimates in its standardized cost category (SCC) database, detailing costs for major station, tunnel, and viaduct elements per mile, broken up by height of viaduct, tunnel size, and similar variables, but the data is coarse and therefore insufficient. The database uses historical American urban transit costs and is maintained part-time by a single planner.⁴ 

A comprehensive database would require market research into commodity prices, capital goods prices, local freight transportation costs, and local white-collar wages. It could rely on both public infrastructure and private construction records — in most cases, concrete is concrete and a pipe is a pipe regardless of whether they’re used in a subway tunnel or a new building. Some prices are readily available, since blue-collar labor costs are itemized under union agreements and prevailing wage laws. Gathering this data will require an upfront investment in expert staff analysts, but there is already significant private-sector capacity, which could be temporarily retained in the initial database’s development. After that point, a smaller, permanent staff could track price changes and maintain the database. 

Given plausible savings of 15-20% on every project (as high as tens or even hundreds of millions of dollars, as the Boston example and many others demonstrate), the capacity necessary for this program would quickly pay for itself. Agencies and contractors should use the database as a first step in planning and only make modifications based on spot prices or unusual local conditions, rather than build the cost estimates from scratch every time. USDOT could learn from Italian, Spanish, Nordic, and other peers for implementation guidance. 

An itemization database should be paired with complementary action from transit agencies — something the federal government is well-positioned to enforce. Because federal funding is essential for most transit projects, conditioning Capital Investment Grants on a commitment to itemized contracts and change orders would force most projects to adopt them. Such a requirement should not require transit agencies to follow the exact pricing schedule USDOT produces, which could calcify and hinder transit procurement. It would simply require transparent unit pricing, with the federal benchmarks providing a reference for agency staff and a check against excessive unit costs if one transit project sponsor is out of step with the norm. As a preliminary step, transit agencies could be prioritized for CIG funds if they follow best practices for itemized bids and change orders for major infrastructure projects. 

As transit agencies begin to use the federal database for cost estimation, they will build the in-house capacity necessary to forgo independent estimators. This will help agencies increasingly select qualified and responsible bids, more effectively delivering projects — and much-needed transparency — to the public.

Further reading

• Valentin Bolotnyy and Shoshana Vasserman, “Scaling Auctions as Insurance: A Case Study in Infrastructure Procurement,” Econometrica, 2023.
• Leah Brooks and Zachary Liscow, “Infrastructure Costs,” American Economic Journal: Applied Economics, 2023.