Inflation on Transit Projects

Representative Alignment, which this estimate refers to (Sound Transit)

Graham Johnson's KIRO report on ST3 cost escalation was notable for its literate discussion of inflation adjustment:

Sound Transit says the estimate in ST3 was $5.8 billion in 2014 dollars, which the agency considers equivalent to $6.8 billion in 2018 dollars. The newest estimate is $7.5 billion in 2018 dollars.

First of all, good for both ST and Johnson that they took the care to compute and report this. Although the real increases here are indeed a story, revising an estimate from 2014 to 2018 dollars is no news at all. That context is usually sorely lacking in stories about increasing costs.

Our comment thread had a spirited discussion as to what inflation measure the estimate used. ST's Scott Thompson verified for STB that they used "Construction Cost Index for construction estimates, Right of Way index for real estate, and the Consumer Price Index for soft costs." This seems reasonable enough. But there are at least three different ways one can deflate rising costs, and they serve different purposes.

The CPI provides the most straightforward definition of 2014 and 2018 dollars, and therefore is the way to communicate inflation in a concise and technically accurate way. However, it doesn't communicate much else, except that construction costs are rising faster than other things.

If you're interested in assessing Sound Transit's project management, then the measure they provided is the right one. If the 2014 estimate were 100% designed and they'd stuck to that plan, the estimate would have increased accordingly. It makes it clear that the remaining $700m is some combination of unforeseeable costs, deliberate lowballing, and scope creep. Your opinion on which is probably based on your prior prejudices about Sound Transit.

Those interested in checking on project delivery, rather than assessing blame, would most benefit from hearing how expenses have changed with respect to revenues. A strong economy escalates both costs and income and in theory that might even out. The extent that they don't is the extent that it will affect project delivery.

While the agency has come a long way by clearly separating out construction inflation from other problems, the current formulation leaves questions. In particular, it's not clear what the $700m overrun entails, or what the implications are for project delivery. Perhaps that will come out in the coming months.