Should TriMet consider the MOS for Milwaukie MAX, particularly if funding from Clackamas County is no longer available?
The Portland-Milwaukie MAX project commenced construction last year, albeit concentrated in downtown. Trees have been ripped out along SW Lincoln, work on the bridge has started, and work is also starting on the so-called “Harbor Structure”, a viaduct over Harbor Drive between SW Naito and SW Gibbs. Work on the bulk of the project, beyond the new bridge, doesn’t start until next year.
That hasn’t stopped activists in Clackamas County, who are skeptical of the project, from trying to block the county (and the city of Milwaukie) from contributing funding. Whether these effort a) are legal, and b) will succeed at the ballot box, but my guess is the latter question will be a yes–and the former question is a big I-don’t-know.
County commissioner Paul Savas, who is critical of the project, has suggested that Clackamas County renegotiate the contract (already executed) which commits the county to pay $25 million to the project in exchange for MAX being built to the Oak Grove area. One proposal that he makes–and which has also been previously suggested here as a potential fallback plan– is that instead of constructing the full line to Oak Grove, the region instead construct the MOS to Milwaukie, and stop there.
What is a MOS?
MOS stands for “minimum operating segment”. It is a standard part of a federally-funded project’s Environmental Impact Statement (EIS), which specifies a smaller subset of a project which, if built, would still meet cost-effectiveness goals. Identification of the MOS serves several purposes–it helps a project deal with funding contingencies–such as a participating government agency withdrawing their contribution–and it also helps (in theory) to make sure a boondoggle section isn’t tacked on to the tail end of an otherwise useful project.
According to the Record of Decision, the full version of the project (the Locally Preferred Alternative, or LPA) consists of a two tracked light rail line from Portland to Oak Grove, terminating at Park Avenue, with 11 new stations, 20 new rail cars, several viaducts, and the Big Bridge over the Willamette River. Two stations, at SE Tacoma and Park Avenue would have park-and-rides, with capacities of 800 and 600 spaces respectively.
The MOS consists of the entire full line, except for the last station. The line would instead end at Lake Road station in Milwaukie, rather than ending at the Park Avenue station in Oak Grove. The MOS would eliminate the construction of a viaduct over McLoughlin Bouleavard, and the Park Avenue station. Instead of building a park-and-ride at Park Avenue, it would be built at Lake Road instead (the LPA has no park-and-ride at this stop). Only 16 new railcars would be purchased instead of 20.
The FEIS contains one other option–and this is what TriMet actually intends to build. It’s described as the “phasing option”, and it is essentially the same as the LPA, but with only a surface parking lot (no garage) at Tacoma Street, giving 320 spaces instead of 800, and a smaller structure at Park Avenue, with 355 spaces instead of 600. It also includes potentially fewer railcars, and a few other cost-saving measures (such as fewer switches along the line, where trains can transfer from one track to the other).
There does not appear to be a “phased MOS” option, which combines the cost savings of the phased LPA with the elimination of Park Avenue.
How much are the various options projected to cost? In year-of-expenditure dollars, the figures are as follows (all figures taken from Table 5-1.1 of the FEIS, page 5-4):
- For the LPA to Park Avenue, US$1.55 billion.
- For the LPA Phasing option, $1.49 billion (a savings of about $60 million from the full LPA).
- For the MOS, $1.38 billion (a savings of about $170 million from the full LPA, and $110 million from the phasing option
Projected ridership by 2030 is as follows–and here’s where it gets interesting (all figures taken from secion S.5.1.1 of the FEIS, in the Executive Summary).
- For the LPA to Park Avenue: 25,570 daily trips.
- For the phased approach, 22, 700 trips.
- The MOS: 24,810 trips
Note that the MOS, which is $100 million less expensive, nonetheless produces over 2000 more daily trips than the phased LPA–and the difference between it and the full LPA is only about 760 daily trips.
Hmm. While I’m generally not fond of cost/rider as a metric, for reasons beyond the scope of this article, I’ll do the calculations. For the full LPA, that works out to $60.6k per new rider. The MOS, $55.6k/rider. The phasing option: $65.6k per new rider.
What this tells us is that ridership is dependent on ample park-and-ride spaces. That isn’t terribly surprising–Clackamas County is not terribly dense nor transit-oriented, so many users of the MAX from the county will be driving to a park-and-ride; fewer spaces will naturally depress ridership. It also suggests that the bulk of the riders using Park Avenue will be users of the park-and-ride, and will happily drive an extra mile or so to a different park and ride. But it also tells us that the option that TriMet is planning to build, is apparently the least cost-effective one–there’s more bang for buck to be had by eliminating the Park Avenue station (and the trackwork and viaducts needed to reach it) then there is from the various cost-reduction exercises undertaken to produce the phased approach.
Let’s now turn to the question of paying for the thing. Most of the local funding for the project is secure–right now the only doubtful piece is $25 million from Clackamas County, and $5 million from Milwaukie. Since there is a 1:1 federal match on the local contribution, were this $30 million to vanish, so would another $30 million from Uncle Sam–for a grand total of $60 million. There isn’t sufficient local funding to build the full LPA; the best the region can do with pledged funding sources (including the Clackamas County portion and the federal match thereon) is the phased approach. Without Clackamas County’s share (and match), even the phased approach can’t be afforded.
However, the MOS is still quite affordable even without Clackamas County’s share–it costs $110 million less than the phased LPA. If the region built that, it could do so without a dime of Clackamas County’s money, and still have $50 million left over. Uncle Sam would keep half of the cost under-run, but that would leave $25 million. Which brings up an interesting question: Who would get it?
My suspicion is that the under-run proceeds would either be spent on additional capital improvements (such as the recent solar panels at the current end of the Green Line), or divided up among the contributing governments in rough proportion to their contributions. But given that TriMet has pledged operating dollars to help float a bond to finance MLR, it would be nice to think that the cost savings could be used to offset that.
At the recent #askneil event on Twitter (transcript here), I asked Neil McFarlane about the prospect of building the MOUS:
@engineerscotty: Any thought to only building PMLR MOUS to Milwaukie, w/o Park Avenue, if Clack Co voters block county from contributing its share?
Neil responded thusly:
@talktrimet: @engineerscotty The petition drive in Clack Co is still an evolving process; too early to speculate.
@talktrimet: @engineerscotty Clack Co has committed has to light rail and to the $25M contribution to build the line to Park Ave.
Unfortunately, he didn’t answer the question–something I don’t consider promising.
However, building the MOS might have other advantages:
- Public goodwill. TriMet’s reputation is not in the greatest shape at the moment, with many accusations abounding about the agency’s motives–including a significant number of folk who believe the MLR project is about pork-barrel politics, not transit improvement. Voters in Clackamas County, in particular, are skeptical. While some of this is no doubt motivated by hard-core anti-transit ideology from the usual suspects, quite a few transit supporters are doubting the agency. Building the MOS might be politically useful.
- Greater future flexibility. While I don’t expect MLR to be extended for quite a while, at some point in the future an extension to Clackamas, or Oregon City (via Gladstone), or Lake Oswego might be in the works. By truncating the line at Milwaukie, all of the options remain available; the Park Avenue extension somewhat commits further extension down the 99E corridor–other extensions would then have to branch at Milwaukie.
- A better anchor. An important factor in an efficient rapid transit line is a good “anchor”–a destination at the end of the line that drives ridership. A park-and-ride in downtown Milwaukie, where support for the project has generally been strong, is doubtless a better anchor than one in Oak Grove–a poorer community with low residential density, and many residents who are both suspicious of transit and terrified of gentrification.