A Toll Policy Primer for Oregon

Oregon doesn’t have tolls on any of its major roads or bridges.  But faced with stagnant gas tax revenues, and with an appetite for huge freeway expansion projects, the Oregon Department of Transportation has committed itself to using tolls to generate billions of dollars in revenue.

And let’s be clear, as economists, we support the idea of congestion pricing:  our urban transportation problem exists primarily because we don’t reflect back to road users the costs of providing the infrastructure they use.  We try to run the state highway system like every day is “Free Ice Cream” at Ben and Jerry’s–people are lined up around the block and we don’t have enough money.

Before the state heads down the toll path, there are a few things it should know about tolling, and the risks and consequences of the approach that’s being pursued by ODOT.  We’ve tried to reduce the key lessons down to this baker’s dozen of essential points.

1.  The Oregon Department of Transportation thinks it has a capacity/congestion problem on Portland area freeways (I-5, I-205, I-84, 217).  In reality this is a peak hour congestion problem (or peak several hours) and results from the fact that roads aren’t priced.  (And also that WA essentially pays people to travel to Oregon to shop). These roads, and all others, have plenty of capacity most hours of the day.

2.  ODOT’s traffic forecasts incorrectly predict that traffic will continue to grow on these roadways, and that congestion will get worse and worse.  (In fact, that isn’t true.  Congestion is limited by capacity, and bottlenecks like the I-5 Columbia River Bridges meter traffic on I-5 (esp. in the AM).  Volumes across the I-5 bridges can’t grow—according to ODOTs own toll consultants, because there isn’t enough capacity.

3.  ODOT is using these exaggerated forecasts of future growth in travel to justify the size of new roadways (10 lanes at Rose Quarter, 12 lanes for the I-5 Bridge and approaches, as well as additional lanes for the Boone Bridge and Abernethy/I-205. ODOT describes these as “bridge” projects (interstate bridge, Abernethy bridge, Boone bridge) but then all of them involve widening miles and miles of freeway on either side of the bridge.  It’s a conscious tactic to mislead. A project that is actually just “replacing” a bridge, like the proposed Burnside Bridge project in Portland, doesn’t raise these issues, because its not really a highway-widening masquerading as a bridge replacement.

4.  ODOT will issue billions of dollars in bonds to pay for all these projects.  And bonds will be issued before the projects are started. ODOT will also put off charging tolls until after construction is started, or until after the projects are completed.  This will require capitalized interest (i.e. borrowing money in excess of the construction cost and using it to pay interest before tolls are collected).  The giant projects will be committed and the debt obligation put in place before there’s any actual experience with tollling.

5.  Traffic volumes on tolled roads will be lower, not just lower than ODOTs excessively optimistic forecasts, but lower than before the projects were built.  On average, we can expect even modest tolls to produce a 20% to 40% reduction in traffic from un-tolled levels.  Those estimates are borne out in the Pacific NW based on the experience of tolling previously un-tolled roads (i.e. Highway 99 Tunnel in Seattle) and on removing tolls from previously tolled facilities (in 2017, BC eliminated tolls on the Port Mann and Golden Ears Bridges).

6.  These widened roadways and bridges will have excess capacity.  The classic example of this happening is the I-65 Ohio River Bridges project in Louisville, KY.  It is nearly identical to the proposed CRC/Interstate Bridge Replacement.  KY & IN paid $1 billion to double the I-65 bridges from 6 lanes to 12, and AFTER they were completed imposed a toll (which works out to $1 each way for regular commuters).  That caused traffic on I-65 to drop by about half from 130,000 ADT to about 70,000.  The twelve lane bridge looks almost empty even at PM peak hours.  Here’s the I-65 bridge on a Monday afternoon at seven minutes past 5 PM.



7.  There’s related issue of toll design.  Congestion pricing focuses on using tolls to manage traffic, i.e. lowering toll rates when traffic volumes are low, and raising them when traffic volumes are high.  In contrast, with a fixed debt obligation under HB 3055 “build/borrow/then toll”) approach, ODOT will likely want to capture revenue from all road users, even in off-peak hours.  This was exactly reflected by the final changes to the CRC tolling strategy in 2013, which doubled off-peak tolls:  the project’s Final Environmental Impact Statement claimed said minimum tolls would be $1.30; the project’s investment grade analysis (IGA) raised them to $2.60.  Originally, peak tolls were more than double off-peak tolls; in the IGA there was just a 65 cent difference, so you end up losing much of the time-shifting incentive. Also:  to show the project was “justified” in terms of demand, and in order to maximize revenue, ODOT has strong incentives NOT to raise peak period toll rates so high that they shift traffic to off-peak periods because they will get less money from motorists.   If we just used tolls to manage demand, rather than finance over-sized construction projects, tolls could be much lower, and off-peak tolls could be set to zero, effectively created a low cost travel alternative when we have abundant capacity.

8.  ODOT will likely have to raise toll rates over time to generate more revenue.  This is already happening in Washington State, where tolls have been increased, and further increases are planned for SR 99, SR 520 and Tacoma Narrows Bridge.   These toll increases have the effect of further depressing traffic growth, and at some point, it’s impossible to get significant additional revenue by raising tolls. Even with the toll increases, there isn’t enough to pay back the bonds.

9.  Toll based projects will need to be bailed out from other sources.  WA is currently making “loans” from its motor vehicle account to bail out the Tacoma Narrows Bridge, and the SR 99 tunnel.  Tolled roadways in CA, FL, and TX have defaulted debt payments, been bailed out, and been forced into bankruptcy.  In OR, under HB 3055, ODOT would be required to divert monies from state taxes and federal grants to make bondholders whole, which would mean foregoing other projects and cutting maintenance and operations.

10.  None of these projects are actually fully funded by tolls.  Tolls usually cover 20-40% of project costs.  For example, toll backed bonds covered less than 10 percent of the cost of Seattle’s SR 99 tunnel  Toll backed bonds were a minority of the revenue for SR 520 as well.  Tolls would have been about a third of the proposed CRC, when its budget was about $3 billion. There’s no guarantee that toll revenue will be sufficient to cover a significant portion of project costs. With the I-205 Abernethy project, ODOT is going ahead with building the project without doing an independent, investment grade analysis that will assure that tolls are sufficient to cover capital costs.

11.  The issuance of toll-backed bonds creates strong financial incentives for ODOT to maximize traffic on tolled roadways.  Once bonds are issued, measures that decrease VMT on tolled roads reduce the  stream of revenue to repay bondholders and force ODOT to take money from other projects and/or reduce maintenance expenditures.  If you applied a “climate lens” to this policy, you would see that this is exactly the reverse of the incentives that should apply to ODOT.  In short, if we don’t increase traffic on widened tolled Portland area freeways, we’ll be forced to cut back spending on everything else ODOT spends money on.  From a climate perspective, that’s simply perverse. Toll based capacity expansions are inherently in conflict with out stated climate objectives.  They create the financial necessity of maintaining or increasing VMT, and with it, GHG, in order to avoid cutting other projects.

12.  The fact that fewer people use these roads than today if they’re tolled even modestly means people will drive on these expanded roadways only if someone else pays their cost.  The value provided by the roads to the people who would actually drive on them isn’t so high that they would pay the costs.  Spending money on these roads is a value-destroying proposition:  It costs the people who bear the costs to spend more on the roadway than its services are worth to those who use it.  Even with tolls, users aren’t being asked to pay anything approaching the actual cost of the capacity (see #9).  In a very real sense, we’re using scarce public resources to pay people to drive more.

13.  Viewed in the context of climate and generational change, this policy is tragic and perverse.  It builds more roads now and sends the bill for these roadways to future generations.  They will not only have to repay the financial debt, they’ll also have to bear the environmental burdens from worsening climate chaos.  And the more future generations figure out ways to reduce travel and GHG emissions, the larger will be the debt that has to be repaid by slashing other transportation projects.

ODOT’s safety lie is back, bigger than ever

Oregon DOT is using phony claims about safety to sell a $1.45 billion freeway widening project

People are regularly being killed on ODOT roadways and the agency claims that it lacks the resources to fix these problems

Meanwhile, it proposes to spend billions of dollars widening freeways where virtually no one is killed or injured and labels this a “safety” project.

A wider I-5 freeway will do nothing to improve road safety in Portland.

On October 4, Sarah Pliner  was killed at the corner of SE 26th and Powell Boulevard, when she and her bike were crushed  by a truck.  Pliner was a chef at a Southeast Portland restaurant; she was cycling to work. This was no accident:  this intersection is one of the deadliest in Portland, having killed or maimed many over the past several years.  Powell Boulevard (aka US 26) is owned and managed by the Oregon Department of Transportation.  Pliner’s tragic death is just the latest example of the carnage on Oregon roads:  in the past year, pedestrian deaths in Oregon are up 61 percent.

Sarah Pliner (Bike Portland)

Powell is one of the multi-lane arterial highways ODOT operates that are the deadliest roads in the Portland metropolitan area.  Recently, ODOT has agreed to transfer another of its deadly arterials (a portion 82nd Avenue, aka Oregon highway 213) to the City of Portland, so that the City can implement a traffic calming and safety strategy.  That’s been in the works for years, but the hang up has been getting enough money to finance the improvements before the city can take over management.  Several years ago the city and state estimated it would cost $31 million to address the safety issues on this part of Powell Boulevard, but little has been done.  ODOT routinely pleads poverty when asked to fix these killer roads.

ODOT’s main priority is widening area freeways.  They’re proposing to spend as much as $1.45 billion widening about a 1.5 mile stretch of Interstate 5 in Portland’s Rose Quarter.  And they’re cynically and deceitfully packaging this spending as a “safety” project.  We’ve call out this lie time and again, but ODOT still repeats a false claim that this is the worst crash location on the Interstate.

That isn’t true—other stretches of Portland area interstate have higher crash rates, as shown by ODOT’s own data—it’s an intentionally misleading and calculatedly cynical talking point, crafted by the agency’s PR team.  And it’s beside the point.  Interstate freeways are relatively very safe, with few fatalities and serious injuries.  What “crashes” do happen are overwhelmingly minor fender benders, bad for a car’s bodywork, and sometimes a source of traffic delays, but not something that routinely kills and maims Portlanders.  ODOT loves to pretend it cares about safety by ginning up “crash” numbers, while ignoring serious injuries and fatalities. Meanwhile, it studiously ignores the deaths that occur on the arterial roads in operates, like Powell Boulevard, that regularly kill people like Sarah Pliner.

But that doesn’t stop ODOT from claiming that the Rose Quarter project is really all about safety.  Just weeks before Pliner was killed on ODOT’s Powell Boulevard, the agency sought legislative approval to apply for federal funding for the Rose Quarter, labeling it an important safety project.  The request for federal funds comes after the agency has allowed the project’s budget to more than triple from $450 million to as much as $1.45 billion—meaning that the Rose Quarter will consume a billion dollars that could have been used to fix roads like Powell.  Here’s what Kris Strickler said in a letter to legislators.

This Rose Quarter section of I-5 . . . has the highest crash rate of any section of interstate within Oregon. The project adds new ramp-to-ramp connections (also known as auxiliary lanes that connect one entrance ramp to the next exit ramp) and adds full shoulders to I-5 to improve traffic flow and reduce the frequency of crashes. Also included in the project is the construction of a highway cover over I-5 that will create new community spaces and enhance safety and connections for people walking, rolling, biking, riding transit, and driving on local streets. [emphasis added]

The claim that the “cover” over I-5 will enhance safety for pedestrians and cyclists is spurious and misleading.  By bringing even more cars into the Rose Quarter, the freeway widening increases the threats to vulnerable road users.  And as we’ve noted at City Observatory, a key part of the project’s design is widening the radius of curvature at many intersections, increasing the turning speed of traffic and lengthening the distance pedestrians must travel to cross the street, both of which are guaranteed to make the area more dangerous, not safer.  The entire purpose of this project is more and faster cars, something that we know will hurt safety, not help it.

ODOT’s own statistics from its Environmental Assessment make it clear that there haven’t been any fatalities, or even any serious injuries on I-5 in the five year period for which it reported data.  Their data also show that most crashes happen during the mid-day (9am to 4pm) not during peak hours.

ODOT is currently in the process of updating its Environmental Assessment to sell its now $1.45 billion project.  It hasn’t bothered to update any of the crash or safety data, and is relying on information that is at least seven years old to bolster its rationalization that this is somehow a “safety” project.

Claims that widening the freeway will make the region safer are simply wrong.

Over the past several years at City Observatory, we’ve dug deep into the claim that a wider freeway will somehow be safer.  A range of studies, some prepared by ODOT, confirm that the I-5 Rose Quarter project will do essentially nothing to improve traffic safety in the Portland region:

Metro’s Regional Transportation Plan  officially categorizes the purpose of the  the Rose Quarter widening as “reducing minor or non-injury crashes.”

After ODOT widened a nearby stretch of I-5, the crash rate went up, not down.

The “ISATe” methodology ODOT used to claim that crashes would go down as a result of widening doesn’t apply to freeways with ramp-meters—which I-5 has.

After years of badgering ODOT to stop making blatantly false claims that the Rose Quarter was “the #1 crash location in Oregon”—we got them to drop that claim from the project’s website.

ODOT narrowed its proposed freeway widening at the South end of the project to avoid encroaching on the Vera Katz Esplanade, with 11 foot lanes and narrow shoulders, yet its own analysis showed this would make only a trivial difference to the crash rate.

As Canadian planner Brent Toderian says, “The truth about a city’s aspirations isn’t found in its vision. It’s found in its budget.”  ODOT may claim to care about safety, but is only real interest is wider freeways, and more and faster car traffic—something that’s been repeatedly shown to lead to even more crashes, injuries and deaths, not fewer.

Two out of three candidates for Oregon Governor are climate denialists

The Republican and Independent candidates for Oregon Governor are happy to spout a convenient myth that we can fight climate change by widening highways.

That myth has been completely disproven:  wider roads encourage more driving and more greenhouse gases

Advocating for more and wider roads is climate change denial

Oregon has been one of the nation’s leaders in recognizing the dangers posed by climate change.  The state enacted a first-in-the-nation goal of cutting greenhouse gas emissions by 75 percent from 1990 levels by 2050.  While we’ve been frequent critics of the efficacy of some of the steps in the state strategy, each of Oregon’s last several Governors have made it clear they support strategies to address climate change.

In the November 2022 election, voters will choose between three competing candidates:  Democrat Tina Kotek, Republican Christine Drazan and Independent Betsy Johnson.  (Incumbent Governor Kate Brown is term-limited).  All are former state legislators.  Recently Oregon Public Broadcasting interviewed each of the candidates on their position on climate change.

Left to Right: Tina Kotek, Betsy Johnson, Christine Drazan:  One will Oregon’s next Governor (OPB).

To their credit, OPB focused the discussion on transportation emissions, which are the largest source of greenhouse gases in the state, and which have been increasing in recent years.


I support widening our highways, by building more lanes. I believe we can both reduce traffic times and reduce emissions from idling engines.


Of course I would increase highway capacity when necessary to more efficiently move vehicles through our highways. The more we maintain efficient flow, the less emissions will be released.

Tina Kotek, at least, didn’t repeat the “wider highways mean less idling and will reduce emissions” canard.  She also reiterated her support for Governor Kate Brown’s Executive Order directing state agencies to plan for reductions in greenhouse gas emissions (something the other two candidates say they would rescind).

As we’ve pointed out before at City Observatory, the claim that wider roads, faster traffic and less idling means lower greenhouse gas emissions has been scientifically disproved.  And some of the best science on the subject comes from Oregon, specifically Portland State University.

This myth has been repeatedly debunked

The science on this question is unequivocal and uncontested:  making traffic move faster in urban settings, whether by expanding capacity or “operational improvements” to increase traffic flow tends to induce more traffic.  The higher volume of traffic causes congestion to quickly return to previous levels, and the combination of more vehicle miles of travel and no long term change in congestion, means that greenhouse gas emissions actually go up.  The literature refers to this as “the fundamental law of road congestion.”  The most definitive study of the question was completed right here in Oregon, by transportation researchers Alex Bigazzi and Miguel Figliozzi, in a paper published by the Transportation Research Board,. Their work showed that increasing capacity on congested roads to allow traffic to move faster and more smoothly actually increases total emissions.

For those who are interested, here’s a link to the video of Alex Bigazzi presenting the findings of his research at Portland State University in 2011.  The takeaway:  greenhouse gases are tied to how much and how far we drive, not to the speed of traffic or the time spent idling.  There’s no relationship between overall traffic congestion and carbon emissions.

It’s great that all three candidates for Oregon’s Governor are willing to acknowledge the scientific fact of climate change.  But the science doesn’t stop there:  repeating the myth that speeding traffic or reducing idling will lower greenhouse gas emissions is simply flat-earth thinking and climate denial.

Falsely claiming that reduced idling will lower greenhouse gases is a popular lie among highway advocates.  Current Oregon DOT director Kris Strickler made this a centerpiece of his confirmation testimony in 2019.  But repeating this lie doesn’t make it true.  What it does do is further delay taking action that would really address the climate crisis.

Risky Bridges: Deja vu all over again

Needed:  An independent review of technical mistakes that could cost billions

The proposed multi-billion dollar Interstate Bridge Replacement is shaping up a repeat of the Columbia River Crossing (CRC) fiasco because the two states haven’t done anything to independently verify the work of their staff.

Oregon DOT and WSDOT are repeating all the key mistakes that caused the Columbia River Crossing (CRC) to fail a decade ago:

  • Designing an oversized project
  • Kicking the can down the road on hard financial decisions
  • Ignoring engineering and regulatory warning signs
  • Not developing a plan to break the project into affordable phases
  • Rebuilding too many closely spaced interchanges.
  • Not getting Coast Guard approval of bridge height until after spending tens of millions designing a bridge

Critically, the Interstate Bridge Replacement project is not being independently reviewed to determine whether its engineering design, traffic plans, travel projections, revenue forecasts and budget are reasonable.  In the case of the CRC, a series of outside experts were called in, and spotted problems that were created or ignored by state DOT staff.  Project officials for the IBR project are making the same errors, but haven’t been subjected to any real scrutiny from disinterested, outside experts.

In the case of the Columbia River Crossing, four different times, outside experts were called in to independently examine the work of the Oregon and Washington transportation departments:

  • 2010:  Independent Review Panel
  • 2011:  Bridge Review Panel
  • 2010:  Bain Traffic & Revenue Forecast Review
  • 2013:  CDM Smith Investment Grade Analysis

Every time, they found costly errors that could have potentially doomed the project that needed to be fixed.  The two states spent millions of dollars on these independent reviews ($1.2 million for the two independent review panels, and another $1.5 million for independent traffic and toll revenue projections).  These expenditures were money well-spent because they avoided even costlier mistakes.  (We detail each of these reports below).

Say you’re looking at buying a used car.  While the owner assures you its it good shape, you’d definitely want to check things out.  You’d be well-advised to spend a few bucks and get an independent mechanic to look it over, and you’d probably spend a few bucks getting a “CarFax” report to see the vehicles history. Same thing about buying a house:  you’d want to have a thorough inspection by an impartial expert.

Oregon and Washington leaders would be well-served by taking similarly prudent steps to check out the validity of the work being done for the Interstate Bridge Replacement project.  The history of the project clearly shows why:  The failed Columbia River Crossing collapsed in significant part because of errors and sloppy work done by the two state departments of transportation.  A decade ago, reviews by independent experts hired by the two states show that the traffic and financial projects were flawed, the schedule was unreliable, the chosen bridge design was “unbuildable;” plus the initial design for the bridge was too low to qualify for Coast Guard approval.  Independent experts also found that the project was making overly optimistic financial assumptions, failed to create a reasonable contingency plan (including phasing the project), and was perpetuating traffic problems (and driving up costs) by not removing one or more interchanges.

Before Oregon and Washington move forward with the latest version of the CRC, now called the “Interstate Bridge Replacement,” (IBR) this project, which current estimates say could cost as much as $5 billion (and which past history has shown to be a significant understatement), they would be wise to hire some independent experts to check out the quality of the work done.  So far, decision-makers are being asked to simply trust the two agencies, something that led to the epic failure of the Columbia River Crossing a decade ago.  As we pointed out, ODOT pre-construction cost estimates for major highway projects have routinely been way too low, with the typical project ending up costing more than twice as much as its initial estimate.

The proposed IBR would be more expensive, more complex, and more financially risky than any other project ODOT has ever undertaken.  The likelihood of errors is high, and the necessity for quality control checks on ODOT and WSDOT is critical.  And recall, these are agencies that have repeatedly made false claims about key project issues, for example, falsely saying that if the two state’s didn’t move forward with the project they’d have to repay the federal government the $140 million spent planning the failed Columbia River Crossing.

1.  Independent Review Panel findings:  “unbuildable,” “not accurate”, “problematic”, “seriously suspect”

In 2010, Governors Ted Kulongoski and Christine Gregoire appointed an  Independent Review Panel (IRP) to audit every aspect of the Columbia River Crossing project.  The panel spent months studying the project, meeting with project staff, carefully studied the “open web” bridge structure the two DOTs designed, and in their report declared it “unbuildable” and directed that a new design be selected.  The Panel of experts from around the country looked at every aspect of the project’s design, management, and financing had issued a 317-page report

The Independent Review Panel warned that the project finances were tenuous and uncertain, just as they are with today’s IBR.  The panel of national experts warned:

“As currently envisioned development of the CRC is counting on full funding from multiple sources, including tolling which will be new to the community and unproven in its revenue generating potential. Failure to achieve one or more major sources of funding can make the entire project unmanageable or unaffordable in the present.”

The IRP had harsh criticism of the sketchy and inconsistent project budget and schedule.  Their report flagged numerous problems, saying the budget and schedule had:

  • “significant risk”
  • “not accurate enough”
  • “the reliability of the final outputs for cost and schedule are seriously suspect”
  • “the credibility of the cost basis is . . . problematic”

2.  Bridge Review Panel:  A totally new bridge design;

One direct outcome of the 2010 IRP was a determination that the proposed “open-web” design for the river crossing was “unbuildable.”  That led the two governors to appoint another panel, the bridge review panel, to come up with an alternative design.  That panel, also chaired by Tom Warne, issued its 146-page report in 2011.  

The Bridge Review Panel described themselves and their work as follows:

This 16 member panel was comprised of national and international bridge experts, plus key representatives from federal, state and local partner transportation agencies. The mission of the BRP was to examine the current design and potential bridge types given current project constraints and including scenarios where constraints are relaxed or modified. Issues such as meeting current environmental project commitments, sound technical and engineering approaches, aesthetic statements and cost effectiveness were also key considerations.

The panel’s report concluded that any of three different bridge designs could work, including both a cable-stayed and tied-arch designs, which would be considerably taller than the design selected for the IBR.  They determined that these taller designs had no insurmountable conflicts with aviation at Pearson field.

In all, Oregon and Washington spent nearly $1.2 million on consultant services specifically for the two panels.  This doesn’t include the costs of staff time for the two state transportation departments, or the time of other consultants already hired for other tasks, who provided information to the panels.

Independent Review Panel and Bridge Review Panel Expenses
Consultant Amount Description (per CRC)
John Clark         210,003.56 Participated on Bridge Expert Review Panel
Tom Warne         184,745.20 Led Independent Review Panel & Bridge Review Panel
Public Knowledge         141,921.40 Governors Expert Review Panel Administrator
Pegasus Global Holdings          99,439.44 Participated on CRC Independent Review Panel
Cascadia Law Group           85,825.52 Participated on CRC Independent Review Panel
Lenhardt, Andra & Partner           82,643.64 Participated on Bridge Review Panel
ERF           79,711.36 Participated on CRC Independent Review Panel
Aecom Technical Services           68,547.57 Participated on CRC Independent Review Panel
TY Lin International           58,367.04 Participated on Bridge Review Panel; CEVP
URS           47,191.48 Participated on Bridge Review Panel
Ralls Newman           45,522.99 Participated on Bridge Review Panel
Stephan Thoman Consulting           41,121.30 Participated on Bridge Review Panel
Mary Lou Ralls           26,012.50 Participated on CRC Independent Review Panel
Michael Meyer           16,983.50 Governors Expert Review Panel Member
Total      1,188,036.50
Source:  Columbia River Crossing

3. The Bain Report:  Flawed traffic projections

Accurate traffic projections are crucial for designing the correct size for the bridge and approaches, and for correctly estimating potential revenue from tolling.  The Oregon and Washington transportation departments have poor track records in traffic projections. Washington’s state treasurer raised alarms about CRC toll financing after revenues for the newly built Tacoma Narrows toll bridge came in well under WSDOT projections.  In 2010, concerns about the inadequacy of ODOT and WSDOT’s CRC travel projections led Oregon State Treasurer Ted Wheeler to hire international toll finance expert Robert Bain to review their work.  Bain’s review found:

  • Traffic and revenue analyses prepare for the CRC were “not suitable” for credit analysis
  • CRC traffic projections were “confusing” and “outdated”
  • Authors of the traffic projections failed to examine historical data or verify their models against actual trends
  • Diversion estimates to I-205 were “worrying.”
  • Overall, the CRC appears to have overestimated traffic.

4.  The CDM Smith Investment Grade Analysis:  FEIS Toll Traffic & Revenue Analysis Wrong

In 2013, two years after the issuance of the Final Environmental Impact Statement and the Record of Decision, the Columbia River Crossing finally got the results of the Investment Grade Analysis (IGA) prepared by its consultants, CDM Smith. The Oregon and Washington Departments of Transportation paid CDM Smith more than $1.5 million to develop their traffic modeling for the Investment Grade Analysis.  The results were dramatically different than portrayed in the FEIS, and confirmed the flaws that the Bain report identified in the earlier modeling.  The CDM Smith report said tolls would have to be at least twice as high (a minimum of $2.60, rather than $1.35) and that the level of traffic that could be expected on the new widened I-5 bridge would be perpetually lower than that volume carried on the old I-5 bridge, because tolls would reduce and divert traffic. In short, the investment grade analysis confirmed what critics had been saying all along:  that a tolled bridge would need no more capacity than the existing structure.

That history should be powerful proof to current decision-makers that they should insist on seeing an investment grade analysis before deciding on the size of the “replacement” bridge.  But project manager Greg Johnson obstinately told the Metro Council in January 2022 that the investment grade analysis would not be used to size the bridge.

. . . the question regarding the investment grade traffic study. That’s one that we’re going to have our folks look deeply into as far as the timing, but I do want to want to correct a misnomer. That investment grade traffic study is not to size the bridge. What sizes the bridge is the data that we take from the regional models that are a part of Metro and RTC . . .

Reflect for a moment what that means:  Johnson is saying he’ll disregard objective expert third-party information about how much money (and traffic) a tolled bridge will generate in deciding how big the bridge should be.  But economics and practical experience tells us a tolled bridge will have dramatically less traffic than the current structure.  Louisville, Kentucky’s tolled I-65 bridges, identical in many respects to the IBR, resulted in a 50 percent decline in traffic—and a huge revenue shortfall.  The IGA prepared for the Columbia River Crossing by ODOT’s own consultants, CDM Smith, said a tolled I-5 bridge would carry only about half as many vehicles when finished as did ODOT’s less sophisticated (and frankly, biased) models.

Coast Guard Rejection of the low fixed spans

Even with two independent external reviews that considered engineering, and a much trumpeted “Cost Estimate Validation Process” designed to catch and prevent risks, the project failed to adequately address a key issue:  navigation clearance.  A crucial element of any river crossing on a navigable waterway is allowing sufficient room for shipping traffic, a determination that is made by law,by the US Coast Guard.  The current I-5 bridges have a 178 foot river clearance under their lift span.  Then, as now, the state transportation departments are Ignoring or downplaying the Coast Guard’s sweeping authority to regulate bridge heights.

A decade ago, with the CRC, ODOT and WSDOT willfully ignored early advice from the Coast Guard that a 95-foot navigation clearance would be insufficient.  As early as 2006, the Coast Guard signaled it would need 115 or 125 feet of navigation clearance; the CRC project decided on its own that 95 feet ought to be enough.  The two state DOTs attempted to bludgeon their way to Coast Guard approval, but since the USCG has clear and independent statutory authority to regulate all structures over navigable waterways, it held firm and in 2011, reached its own determination that the CRC would have to clear at least 116 feet.  That led to a year of delays and tens of millions in additional costs to re-engineer the bridge to have a higher clearance.  Importantly, this was not a risk that was identified or provided for in the projects schedule or cost management system, showing a clear failure to manage risks on this large project.

The IBR seems hell-bent on repeating this blunder once again.  A Coast Guard preliminary determination has found that a new bridge over the Columbia needs to have a navigation clearance of 178 feet.  Despite the Coast Guard’s ruling, the project is proceeding with its proposal for a 116 foot navigation clearance, and steadfastly refusing to look at alternatives, like a moveable bridge span or a tunnel, that would enable a lower and far less expensive and disruptive crossing.  WSDOT and ODOT would like to pretend that the preliminary determination doesn’t really mean anything, but under the agreement between the US DOT and the Coast Guard, alternatives that don’t meet the preliminary determination are supposed to be excluded from further NEPA review.  When the two state DOTs disregard the 178-foot clearance determination, interagency agreement says they are “proceeding at their own risk.”

Deja vu all over again

The same errors that doomed the CRC are being repeated now by the Oregon and Washington transportation departments.  They’ve designed their bridge with a 116 foot clearance, assuming that this will meet approval by the Coast Guard.  But their USCG-bridge permit expired years ago, and they will need to apply for a new one, and go through an entirely new permitting process, which will likely end up mandating an even taller bridge—one that the project hasn’t considered.

Even the IBR’s proposed 116′ high bridge poses major and as yet unanswered questions.  To reach that height, the bridge will require extremely steep approaches on the Oregon and Washington sides of the river.  In Oregon, the roadway grade exceeds the design standard for Interstate freeways, and will require an exception.  The steep bridge grades have led one local engineer to argue that the bridge will be particularly dangerous in icy weather.  The project calls for rebuilding every one of the seven closely spaced interchanges that cause congestion, contrary to federal design standards and the recommendations of the bridge review panel.  Unlike with the CRC, there hasn’t been any independent review of this design.

The project has yet to produce a definitive financial plan.  The project hasn’t developed any contingency plans if one or more of the project revenue sources doesn’t materialize.  It hasn’t prepare a plan for project phasing.  In fact, the selected high bridge design may be difficult or impossible to phase, because the extreme height of the proposed new river crossing  will make it impossible to access the new structure from existing approach ramps.

The project has no plans to undertake an independent, investment grade analysis of the project until 2025.  Just as before, the project mades optimistic assumptions about toll revenues—its current traffic forecast uses minimum tolls of $1.35—only half of what the 2012 CDM Smith Study said would be necessary to provide a $1.3 billion tolling contribution to the project’s finance plan.

Now, as before, the project is proposing to rebuild every single interchange in the project area, even though outside experts (and their own problem statement) show that’s a substandard design approach that leads to traffic problems and needlessly increases the cost of the project.

In many ways, the re-named IBR project is a scene-for-scene remake of the disaster film that was the Columbia River Crossing.  A key difference to date is that its controversial and questionable engineering, traffic forecasting and financial decisions simply haven’t been vetted by outside experts, as was done with the CRC.  The rush to move forward to a decision to select a “locally preferred alternative” without getting this kind of professional advice magnifies the risks that like its predecessor, the IBR project will also collapse when one or more of these unexamined risks strikes.

Eyes wide shut

The proposed IBR project is a big and risky endeavor.  What’s lacking is any independent verification of the assertions made by the project staff.  Last time around, with the Columbia River Crossing, state leaders took the prudent steps of asking a few basic questions before moving forward with the project.  They hired independent engineers and experts to assess the project design, budget, schedule and phasing.  They hired an international toll bond expert to study its traffic projections.  They conducted an investment grade analysis.  The federal government hired a “project management oversight consultant” to ride shotgun on the project.  With the IBR, none of these safety steps have been taken.

Even the Legislature has been complicit in this failure to put in place basic safeguards and oversight.  In 2017, as part of its major transportation funding legislation, the Legislature created a “Megaprojects Task Force” and directed it to study and report on the state’s process for selecting and managing large projects.

SECTION 121. (1) The Task Force on Mega Transportation Projects is established. For the purposes of this section, a “mega transportation project” includes transportation projects, as defined in ORS 367.010, that cost at least $360 million to complete, that attract a high level of public attention or political interest because of substantial direct and indirect impacts on the community or environment or that require a high level of attention to manage the project successfully.  . . . (11) The task force shall submit a report in the manner provided by ORS 192.245, and may include recommendations for legislation, to the Joint Committee on Transportation established under section 26 of this 2017 Act no later than September 15, 2018.

The legislation set a September 2018 deadline for the Task Force to file its report, but the task force met only twice (after its deadline), never filed any report, and sunsetted, at the end of 2018.  It turns out that even the Legislature, which is expected to make up the shortfalls and pay for the overages when ODOT makes a mistake, isn’t willing to try and learn from past experience.  In the case of the $5 billion (and probably much more) Interstate Bridge Project, that could be a very expensive outcome.


Bain Report, Bain_CRC_Report_July4

Independent Review Panel Report IRP_Report_July30

Bridge Review Panel Report

CDM Smith ReportCDM_SMith_2013

Coast Guard Bridge Permit

Editor’s Note:  An earlier version of this commentary had incorrectly attributed appointment of the Independent Review Panel; it was appointed by Governor Ted Kulongoski.


The Week Observed, October 28, 2022

What City Observatory did this week

A toll policy primer for Oregon.  The Oregon Department of Transportation is proposing to finance billions in future road expansions with tolling.  While we’re enamored of road pricing as a way to better manage our transportation system, the movement to raise moeny with tolls, and in particular by borrowing against the promise of future toll revenues raises significant risks.  This policy primer sketches out the nature of the underlying transportation problem we’re trying to fix (underpriced roads are over-used, and chronically under-funded), and describes the risks from ODOT’s proposed “borrow and build first, toll later” strategy, which promises to saddle the state with billions of dollars in debt for roadways that are unlikely to be used when we ask drivers to pay for them through tolls. Louisville, Kentucky stands as a cautionary tale:  the state spent a billion dollars widening I-65 over the Ohio River, but after it imposed modest tolls traffic fell to half of pre-construction levels, saddling the state with enormous debt for a road that is spectacularly under-used:


The “borrow, build and toll” scheme runs this risk of putting the state in a position where it has to encourage more driving (contrary to its stated climate goals) in order to pay off construction debt.  State policy makers would be well-advised to fully understand the costs and risks of tolling before they pin the state’s transportation future to this perilous source of funding.

Must read

A reform agenda for US DOT.  We’re a little bit late to this particular party, but the Center for American Progress produced a terrific report in 2020 identifying some key opportunities to overhaul transportation policy in the US.  This comprehensive report addresses transportation finance, climate change, and equity, and is a useful framework for those working to improve transportation at a state and federal level.

. . . current policy and program structures need deep reforms to ensure that federal investments are equitable, sustainable, and targeted to communities facing the greatest need. Simply adding more money to the status quo will not help the United States meet its global climate commitments or redress the harms caused by discriminatory project selection and exclusionary labor practices.

You get what you pay for and housing affordability:  Writing at Market Urbanism, Salem Furth has a nuanced take on an oft-overlooked part of the housing affordability puzzle.  A big part of our problem is an overall shortage of housing relative to demand.  But the problem isn’t just one of aggregate supply.  There’s a mismatch between the houses that are available and the characteristics of housing we most value and want to buy.   Housing comes bundled with lots of characteristics and amenities, some of which we value highly, and some not so much.  One of the effects of zoning regulations is to force people to pay for some things they don’t value highly:  Minimum lot sizes require people to “buy” a minimum amount of land in order to be able to have a place to live, and in places where land is expensive, that may make housing to costly for many to afford.  Exclusive single family zoning in some neighborhoods means you have to buy a single family home to live in that neighborhood, even though you might prefer a townhouse or apartment (and be able to avoid lawn maintenance).  More subtly, zoning and building requirements can create mismatches between the kind or size of housing people want, value or can afford and what’s available.  So the more we create flexibility to have different kinds of housing, the more likely we are to have better “matches” between supply and demand and reasonable price points.

Developers push big hgihway package in Sacramento.  California public finance has been hamstrung by super-majority voting requirements for many tax increases, but the state Supreme Court recently carved out a provision allowing citizen-initiated taxes to be adopted with a simple majority.  In Sacramento, a developer backed group has seized on this provision to package up a $8.5 billion road widening package that would facilitate further sprawl in and around California’s capital.  There’s a lot not to like about this proposal:  it would not only undercut the region’s and state’s climate strategy, but it is also funded from the sales tax, which means that there’s no connection between how much you travel or drive, and who pays for the road expansion.  The Sacramento Bee has an in-depth story on the political maneuvering going on.

New Knowledge

Mapping activity centers in the nation’s metropolitan areas.  The spatial structure of metropolitan economies is an under-studied topic.  Our friends at the Brookings INstitution’s Metro Center have a new publication that sheds an interesting light on the clustering of economic activity within metro areas.  They use a wealth of data on the location of businesses and institutions within cities to identify what they call “activity centers.”  Their approach identifies five different kinds of assets or amenities (tourism, community, tourism, consumption, institutional, and economic. They then gather data on the concentrations of those assets in different location and identify three differ kinds of centers.  Places with a concentration of just one asset are called mono-centers.  Places with some concentration of at least two different assets are secondary centers, and the places with relatively large concentrations of two or more kinds of assets are defined as “primary centers.”

This classification system enables Brookings to map the size and location of activity centers in more than 100 of the nation’s largest metropolitan areas.   These maps illustrate the concentration of activity in particular locations.  Here is the map for Portland, Oregon:


The largest concentration of activity centers is in an around the city’s downtown (in the center of the map). There are a smattering of other centers in the remainder of the region, with most suburban centers being “mono-centers” (blue).

The Brookings research strategy parallels City Observatory research in developing the “storefront index.”  Our index looks at the concentration of common kinds of retail activity, and identifies clusters of storefronts located 100 meters or less from one another.  Here we’ve overlaid the storefront index on the Brookings Activity Centers for the Portland Metropolitan area.

On this map, the red dots represents clusters of storefronts.  The black circle in the center of the map is a 3-mile radius around the center of the region’s central business district.

The definition of activity centers is just the starting point.  The full report looks at the connections between activity centers and employment concentrations, productivity, walkability and vehicle miles traveled.  Primary activity centers tend to be among the most disproportionately economically productive locations, tend to be more walkable and people living in or near these centers are less likely to travel by automobile.  In addition, as the report points out, because

. . . low-income and racial minority groups tend to live in areas with higher accessibility to activity centers, efforts to focus development in these areas have strong potential to benefit these groups.

We need to know much more about the fabric and dynamics of urban spatial structure.  This Brookings report helps us visualize how important this perspective can be.

Tracy Hadden Loh, DW Rowlands, Adie Tomer, Joseph Kane,and Jennifer Vey, Mapping America’s Activity Centers”. the Building Blocks of Prosperous, Equitable and Sustainable Regions.  Brookings Institution, October, 2022.

In the News

The Brown Political Review cited our critique of Bruce Katz’s book The New Localism.

The Week Observed, October 21, 2022

What City Observatory did this week

Using phony safety claims to sell a billion dollar freeway widening.  This past week, Sarah Pliner, a promising young Portland chef was killed when she and her bike were crushed by a turning truck at SE Powell Boulevard and 26th Avenue.

Sarah Pliner (Bike Portland)

This intersection is an Oregon Department of Transportation roadway, which has been long identified as one of the most dangerous in the state.  OregonDOT has dragged its feet to do anything to improve safety, instead prioritizing the fast movement of cars and trucks.  While its done nothing here at a facility thats repeatedly killed and maimed local residents, its peddling is $1.45 billion plan to widen Interstate 5 as a “safety” project–notwithstanding that the only crashes on that facility for years have been largely minor fender-benders.  At City Observatory, we’ve repeatedly shown that the Rose Quarter project would do nothing to address the real safety problems in Portland.  It’s tragic that another person had to die because Oregon DOT continues to use a cynical and self-serving definition of safety to avoid spending money to address its roads that kill.

Must read

AASHTO:  Your highway department is a climate denier.  The Federal Highway Administration has released draft regulations calling on state highway agencies to set goals and report their progress in reducing greenhouse gas emissions from transportation.  The American Association of STate Highway and Transportation Officials (AASHTO) is violently opposed to these regulations.  Kevin DeGood of the Center for American Progress has a close, critical review of AASHTO’s comments.  The highway builders argue that they’re really powerless to do anything about greenhouse gas emissions, that it would be prohibitively costly to even monitor their impacts, and that FHWA lacks the authority to do anything.  DeGood comprehensively disposes of all these arguments:

. . .  DOTs think of GHG reductions almost exclusively in terms of EV adoption. Of course, this is ridiculous. DOTs are principally responsible for land use. Low-density sprawl doesn’t happen without highways. Full stop.

State DOTs claim that federal guidance isn’t needed because they’ve been great stewards of federal largesse and have improved the performance and safety of the highway system.  DeGood points out that that simply isn’t true:  when it comes to safety, highway deaths have increased sharply in the past decade from 34,000 per year to 43,000.  AASHTO is all about rationalizing federal handouts, and the organization is congenitally opposed to any accountability.

Mortgage rates soar; housing market crumbles. Since the beginning of the year, the interest rate charged on a 30-year, fixed rate mortgage has more than doubled, from about 3 percent to nearly 7 percent. This surge is a product of the Federal Reserve’s inflation-fighting policy. The 30 year mortgage rate is now higher than its been any time in the past two decades and the low interest policy that prevailed in the wake of the “Great Recession” is now over.  What its likely to do is choke off investment in new housing. Higher interest rates mean that homebuyers are less able to afford homes, and that it’s harder for investors to profitably build new housing.

Higher mortgage rates dampen the incentives for both buyers and seller:  buyers face higher costs for obtaining mortgages for their home purchases, and prospective sellers with mortgages also face the prospect that the mortgage interest rate on their next home may be much higher than the mortgage rate on their existing loan; this is likely to throttle “trade up” home purchasing for many current homeowners.  The housing market has benefitted from a decade of 30-year fixed mortgage rates at or below four percent.  Those days appear to be over.

New Knowledge

Cycling:  Cultural, not geographic factors predominate.  Cycling is more difficult and less comfortable in cold or wet weather and in hilly locations.  But this study shows that across the US and within metropolitan areas, these climate and geographic factors play a very minor role in explaining variations in cycling rates.  Instead, a range of demographic and cultural factors seem to be more strongly correlated with cycling.

University of Hawai’i economist Justin Tyndall used US census data to look at the correlations between bicycle commute-to-work mode share and a range of topographic, climatic, and social factors across the nation’s metropolitan areas.  A quick visual summary of this work shows few strong correlations between cycle mode share and temperature, precipitation or snowfall.  It turns out that hilliness actually has a slight positive correlation with cycle commuting (more people bicycle to work in hilly places, on average).  And cultural factors (proxied by presidential vote shares) seem to have a far stronger relationship to cycling. Metro’s that voted in larger number for Donald Trump have lower rates of cycle commuting.

What these data suggest is that cycling is less about weather or topography than it is about demographics and cultural attitudes towards cycling. As Tyndall concludes:

The role of geography in cycling uptake is frequently discussed in relation to the construction of bicycle infrastructure such as bike lanes. Opponents of bicycle infrastructure often point to hills or unsuitable weather as evidence that cycling can not be locally popular. The findings in this study have a potentially important lesson for policy: climatic and topographical endowments are unimportant to the general uptake of cycling. The exogenous cause of spatial heterogeneity in cycling appears to be related to local demographic and cultural idiosyncrasies.

Justin Tyndall, “Cycling Mode Choice Amongst US Commuters: The Role of Climate and Topography,” 2022, Urban Studies, 59(1)https://www.justintyndall.com/uploads/2/8/5/5/28559839/tyndall_cycling.pdf

The Week Observed, October 14, 2022

What City Observatory did this week

Two of the three candidates for Oregon Governor are Climate Deniers. Oregon will elect a new Governor next month, and two of the three candidates for the job insist on repeating the discredited myth that greenhouse gas emissions can be reduced by widening freeways so that people don’t spend as much time idling in traffic.  As we’ve demonstrated time and again at City Observatory, wider roads lead to more driving and increased emissions.  Pretending that wider roads will reduce greenhouse gas emissions is climate denial, something that independent Betsy Johnson and Republican Christine Drazen are shamelessly repeating. Falsely claiming that reduced idling will lower greenhouse gases is a popular lie among highway advocates.  Current Oregon DOT director Kris Strickler made this a centerpiece of his confirmation testimony in 2019.  But repeating this lie doesn’t make it true.  What it does do is further delay taking action that would really address the climate crisis.

Must read

What’s behind regional variation in housing production.  Nationally, we have a significant shortage of housing, but in reality, there isn’t a single national housing market.  Housing is in many ways a distinctly local set of markets.  Why do some markets seem to produce a lot of new housing, and others not so much?

The Urban Institute’s Yonah Freemark offers a typology and analysis of housing production in US metro areas.  Freemark’s unit of analysis is the municipality, and he classifies cities within a Metro area based on their housing costs relative to the regional average–a good indicator of demand and amenities.  Overall, few housing units get built in the lowest value cities and the highest value cities, but for very different reasons.  In low value cities, there’s little demand; in high value cities, local policies often greatly constrain new development.

Freemark’s work points up a consistent pattern in which high value, high demand communities consistently tend to allow less housing to be built than other municipalities in the same region:

. . . of the nation’s most-in-demand municipalities—those where housing values are at least 30 percent higher than their respective metropolitan areas—less than a third added more housing than their encompassing region, despite plentiful developer demand to build there. In contrast, more than 40 percent of such jurisdictions added new housing at 50 percent or less of the rates of their respective metropolitan areas—and many actually lost housing units.


It’s time to reform streets, not just highways.  While there’s considerable push-back against urban freeway expansion projects around the nation, Washington transportation advocate Anna Zieverts says we also need to pay attention to how we use streets.  Writing at Next City, Zieverts, who works for Disability Rights Washington, argues that we priortize car movement on local streets, and that multi-lane arterials, limited pedestrian crossings, and car-oriented signal timing all make streets difficult or dangerous for non-car road users.  While replacing freeways with boulevards may lower their impact, even boulevards are chiefly designed to move cars.  For city streets, we ought to reverse the priority, putting pedestrians and bikes first, and slowing car traffic.

What if slow speed streets were the default for our cities? Yes, it would take longer to travel far distances by car in a city, but that may be one of the more equitable ways we can incentivize people to choose other modes, especially if roads designed for slower car speeds were paired with investments in faster bus-only lanes.

Maryland highway expansion sued by Sierra Club and NRDC.  Environmental and community groups are challenging plans to widen I-95 and the Capital Beltway outside Washington DC, alleging, in part that the project’s environmental impact was concealed by faulty traffic modeling. The most obvious error in the traffic modeling was that it wildly exaggerated traffic levels, congestion and pollution in the “No Build” scenario, assuming a lemming-like behavior of commuters in the face of ever increasing delays.  The plaintiffs complaint argues:

Defendants relied on traffic modeling performed by MDOT to predict that
adding toll lanes will reduce traffic congestion on the Beltway and I-270. MDOT’s
model, however, counterfactually assumes that drivers will continue to pile onto these highways no matter how far the backup and no matter how long the delay. In reality, drivers often re-route when faced with extreme congestion. Defendants did not explain how they could reasonably rely upon the model’s outputs given this serious limitation. MDOT also refused to release its underlying modeling files, further thwarting public scrutiny of Defendants’ justification for the toll lanes project.

By exaggerating traffic congestion and pollution in the No-Build, the EIS makes the build scenario look relatively more benign than it actually is.  And all of the modeling is closely guarded secret: a black box that’s not subject to any external review, and one that conveniently always produces the answers the highway department’s want.

New Knowledge

Work from home is shredding office values.  In the wake of the Covid pandemic, there has been a persistence of remote working, particularly among knowledge work occupations.  According to a new study from three economists at New York University and Columbia University, the work from home trend is having a dramatic impact on the value of the nation’s commercial offices.

Spending more time working at home means spending less time working in the office.  With fewer people in the office, there may be less demand for office space in the future.  This study looks at recent trends in office occupancy, their correlation with work from home, and its impact on commercial leasing.  Most office space is leased for a period of several years, and in the short term (i.e. since the advent of the pandemic) there have been limited opportunities for firms to adjust how much space they lease.  But as leases come up for renewal, and as firms seek (or don’t seek) new office space, we would expect the declining demand for office space to begin to be reflected in the market.

The study shows that there’s been a sharp slowdown in new leasing activity, and in general a shift to shorter-term leases (more leases for fewer than three years; fewer leases for more than seven years).  Prior to the pandemic, the national market included about 250 million square feet of new office leases per six months; after that leasing rate has fallen to less than 100 million square feet.

But while new leasing activity has declined sharply–and not rebounded as the pandemic has eased, the path of prices has been different.  Average rents for new leases did in fact fall during 2020, but have rebounded since then.  As the authors point out, some of this rebound is due in part to a composition effect, a result of a larger share of leases being signed for the highest quality office spaces; adjusting for those effects (the solid line) shows a much smaller rebound.

Looking forward, a decreased demand for office space could significantly effect both commercial office development and city finances.  Based on their estimated trajectory for persistent work-at-home, the authors estimate that commercial office values in New York, for example, could decline almost 40 percent in the long run.  Because the city relies heavily on property taxes tied to values, that could result in a significant fiscal impact.

Arpit Gupta, Vrinda Mittal, & Stijn Van Nieuwerburgh, Work from Home and the Office Real Estate Apocalypse, Working Paper 30526. http://www.nber.org/papers/w30526 National Bureau of Economic Research