It can sound like a simple step, to end Level of Service (LOS) metrics in CEQA transportation analysis. The more conceptually elegant Vehicle Miles Traveled (VMT) metric is easy to welcome in the abstract, with its incentives for shared and active transportation, its arguably simpler calculation methods, its potential to realign CEQA analysis with state climate protection law – and most of all, its escape from the addictive spiral of induced demand for broad, free-flowing highways that, under the logic of LOS analysis, always need widening again.

But in early August the Governor's Office of Planning and Research (OPR) published a detailed discussion draft setting out an alternative transportation impacts metric in compliance with last year's SB 743 mandate. And alongside the big-picture discussions of environmentally conscious innovation, the technical arguments began. Last week, with the appearance of a highly critical online comment from the Holland & Knight law firm, the depth of the disagreements involved became clearer.

The state-level policy people who watch this matter aren't so much debating the merits of rail lines or bike lanes. Nor are any of them professing special love for the LOS metric or the internal combustion engine. They're talking about which LOS analyses are still going to be around (a lot), and who may now apply which version of VMT analysis under CEQA, and whether suggested means of reducing VMT represent constructive reforms or potentially "shackling" mandates, and what mix of fresh data with safe assumptions can be used, and whose fault it will be if the assumptions are wrong, and how much this all needs to have to do with the unforgiving math of AB 32 emissions reductions.

Only the 'beginning of the end of LOS'

Amanda Eaken, the sustainable communities deputy director with the Natural Resources Defense Council (NRDC), said "I hope this is the beginning of the end of LOS." But for now the likelihood is that VMT and LOS analyses will both continue to be used, though somewhat differently than before, each still representing important means of negotiating what it means to take up space, and who has power to do so.

VMT analysis is already familiar under the Clean Air Act and, more recently, from greenhouse gas (GHG) emissions analysis and Sustainable Communities Strategies under SB 375. Christopher Calfee, Senior Counsel with OPR, called it "a different way of measuring transportation impacts using tools that people are already familiar with" that would involve "building on that information that's already been developed."

Environmental and transit advocates have their criticisms of the proposed draft but on the whole they like it better than do advocates for developers. For one thing, advocates had not been certain that the proposed new standards would apply statewide – and they do. Or rather, if the guidelines are adopted they'll begin to apply statewide under a phase-in schedule starting with developments within half a mile of transit, and will cover the whole state as of January 1, 2016. NRDC senior attorney David Pettit called it "properly aggressive as to the dates of phase-in." Eaken said state agencies, including Caltrans, were solidly behind the plan.

As Bill Fulton explains, local governments will still be applying LOS standards for reasons unrelated to CEQA. These are likely to include general plan provisions, zoning, safety, municipal Congestion Management Programs and impact fee programs. As he also mentions, analysis within CEQA may use LOS approaches to analyze some impacts including noise and air quality – though under the proposed guidelines "automobile delay" (congestion) would not be treated as an environmental impact in itself. (Eaken said she and representatives of the Council of Infill Builders pushed successfully for the clarification regarding auto delay, and it "is going to change things fundamentally.")

'Radical mitigation recommendations'?

Amanda Monchamp, a partner with the Holland & Knight law firm who works on land use and environmental issues, called the draft "far broader and more complicated than the statute actually requires," with "a huge list of mitigation measures" applicable to infill projects, and with continuing requirements for LOS analysis on air quality, noise and "hot spots".

Since Monchamp's comments, the Holland & Knight firm has posted an aggressively phrased critical commentary on the draft. The comment is primarily authored by partner Jennifer Hernandez, with Monchamp and another attorney as coauthors. It charges that "OPR's VMT proposal goes far beyond CEQA's statutory scope by recommending mitigation measures that delve into socioeconomic issues, undermine regional and local greenhouse gas reduction plans, attempt to erode local agency constitutional land use policy authority, and increase the cost, complexity and litigation uncertainty already inherent in CEQA."

The Holland & Knight comment refers to Subdivision II(d)(6) in a proposed amendment to Appendix F of the guidelines, which lists 14 "potential measures to reduce" a project's VMT. The comment characterizes them as "radical mitigation recommendations". The listed possibilities for reducing VMT include "Improving or increasing access to housing," "Increasing access to common goods and services such as groceries, schools, and daycare, "Incorporating affordable housing into the project," "Improving the jobs/housing fit of a community," and assorted other measures including traffic calming and bicycle parking. The firm's argument attacks each of these suggested items as a possible means of imposing new demands or mandates on developers beyond the scope of existing CEQA requirements.

The commentary argues the existing proposal should be "tabled," and Monchamp said she would be calling for "a shorter, cleaner, more precise change" consistent with SB 743's direction but more narrowly tailored and more effective in facilitating infill.

The dual use of LOS and VMT analyses concerned Monchamp as a possible cause of extra work for project proponents. She mentioned, for example, the North San Jose area-wide traffic mitigation process, which imposes a clearly defined traffic mitigation fee on projects. She said that process was all LOS-based, whereas, under the new state-level requirements, VMT analysis would have to be added for projects in that area.

Prospects of extra paperwork and uncertainty likewise concerned Mott Smith of Civic Enterprise Associates LLC, a board member with the Council of Infill Builders.

Smith viewed rules on matters like air quality and noise impacts as "back doors for people to reintroduce LOS" into the CEQA environmental process. With requirements such as general plan consistency determinations in the mix, he said builders could end up "right back where we started and we're still looking at level of service."

(Anticipating such discussion, OPR wrote in the draft, "there is nothing in SB 743 that requires analysis of noise or air quality in a transportation section of an environmental document.")

Calfee responded further that LOS analysis for air quality and noise does include a vehicle component but it's not the full LOS traffic study – the analysis does not have to be done "intersection by intersection, roadway segment by roadway segment." He said the reduced analysis for those purposes may rely on data about traffic volumes, speeds, or other trip generation data that are readily available. So he said, "it's not like you'll be doing a full LOS analysis" at least for those purposes.

He said local decisionmakers choosing how to address congestion outside the CEQA context are entitled to a lot of deference if challenged in court – more so than under CEQA – so "This should end up being a good thing for both cities and developers."

But Smith said "I don't have the comfort that... this protects infill builders trying to do the right thing when they've got hostile actors" contesting their projects. He questioned, for example, whether, if he built a project without parking – which ostensibly would reduce VMT – an opponent might point to impacts from cars circling the building in search of parking, hence might "push me into EIR territory" due to the "potential air quality and noise consequences of lowering my VMT."

So in Smith's view, the new approach "pushes the potential impact from one category on the checklist... to other categories," in which case "it doesn't give me the protection that I really need to do smaller-scale... lower-value projects and more incremental projects in infill areas."

(On the issues of noise, air quality, safety and circling for parking, Calfee wrote: "These issues are already being raised in the CEQA context. This proposal does not create those issues.")

Smith said, "If these rules are successful they will make it easy for builders like me to build good incremental, walkable, transit-accessible projects in California cities." He praised the drafters for intending such an effect but asked, "Can they protect me from hostile CEQA petitioners?"

Calfee responded in writing to a somewhat different but related question: whether VMT presents a disadvantage to infill projects by comparison with the prior LOS analysis. He wrote that LOS analysis "puts infill at a disadvantage" compared with lighter-traffic urban peripheries, while "Vehicle miles traveled will not penalize infill development the way that LOS does, because infill developments tend to reduce the amount of driving that is required to get between destinations. The proposed Guidelines go a step further and say that if the project is located near transit, the transportation impact is likely less than significant, which should in most cases eliminate the need for any study."

Thresholds of significance

Many technical responses to the OPR draft have focused on its guidance for local lead agencies in setting thresholds of significance for projects' VMT levels, viewing it as one of the most novel parts of the document.

And under that heading of threshold-setting guidance, areas of argument and uncertainty include how much latitude exists for local lead agencies to choose their own thresholds, and whether the standards should be strict enough to reduce GHG emissions in accordance with statewide AB 32 goals.

The OPR draft states in its "background" narrative, "SB 743 did not authorize OPR to set thresholds, but it did direct OPR to develop Guidelines 'for determining the significance of the transportation impacts of projects[.]'" It argues for viewing the threshold of significance as the regional average VMT, in part because metropolitan planning organizations (MPOs) are already gathering regional VMT data under SB 375. The draft's proposed new Guidelines Sec. 15064.3(b)(1) suggests, "A development project that is not exempt and that results in vehicle miles traveled greater than regional average for the land use type (e.g. residential, employment, commercial) may indicate a significant impact."

Calfee said the proposed guidelines only gave suggestions as to what may or may not be a significant impact. He returned repeatedly to the established legal principle that "Under CEQA it's always the lead agency that determines the level of significance."

On the other hand, a consultant who works in the area suggested the choice to define even an optional threshold standard directly in the Guidelines at the state level would implicitly burden local lead agencies to justify any different choices they might make.

Asked how opposition to VMT metrics might appear, Pettit suggested "a local jurisdiction that hates this idea may fool around with the threshold of significance" so projects would qualify easily – but he warned, local officials raising the threshold too high could expect "litigation by some pesky enviro group, maybe us."

But Calfee said, "That is something that's baked into CEQA... It's always up to the lead agency to determine what's significant... That's not something that this proposal is causing."

Calfee emphasized repeatedly that lead agencies have power to choose their own VMT models. In an interview he noted the proposed guidelines mentioned out applicable case law supporting such power: "That should give relief to lead agencies in picking the model that's right for them." He wrote separately: "The proposed Guidelines expressly note lead agency's discretion in the choice of methodology, the use of professional judgement in adjust model inputs and outputs, and apply the rule of reason.  These will each make lead agency decisions more defensible."

Are regional average VMT thresholds fair? To whom?

Eaken's strongest comment on the OPR draft was that significance thresholds based on regional average VMTs would not do enough to meet statewide climate change goals for 2030. She asked, "Is average good enough?"

What qualifies as "average" is part of that question – and a reason why, to a developer, a regional average might not always seem locally fair.

Eaken brought up the possibility of unfair leniency under a "regional average" standard in the very large region covered by SB 375 programs of the Southern California Association of Governments (SCAG) – which states it serves "more than 18 million residents" across Southern California. She suggested it might be too easy for a project that fell below the VMT average for such a large region to qualify as having "no significance".

The consultant (who asked not to be named) made a converse point: with relatively dense Los Angeles bringing down overall average VMTs for the SCAG region, some projects in Riverside County might be almost automatically over the regional average.

Calfee, for his part, said actually the city of Riverside was fairly densely populated so downtown "you can be fairly comfortable that you have a lower VMT," while projects in outlying areas could reduce VMT by means such as adding a commercial component to a residential subdivision.

To the objection that Riverside still wasn't Santa Monica, he said the standard wasn't based on Santa Monica, but on a regionwide average, and it was still up to lead agencies to set the level of significance.

The consultant, however, questioned whether the surveys and methodology yet existed to be sure of VMT levels in the Riverside area as divided up by type of use (office, housing, etc.).

On the more general matter of SB 375 consistency, Calfee pointed to current "tension" between the CEQA mandate for LOS-based traffic studies and SB 375 calls for reductions in automobile use: "What we've done in these proposed guidelines is attempt to alleviate that tension." He said the proposal was intended to allow lead agencies to use models and data in their analyses that are also used under SB 375, "so that should lead to fewer conflicts, not more."

(The proposal's Subdivision 15064.3(b)(1) reads in part, "Land use plans that are either consistent with a sustainable communities strategy, or that achieve at least an equivalent reduction in vehicle miles traveled as projected to result from implementation of a sustainable communities strategy, generally may be considered to have a less than significant impact.")

Monchamp argued SB 375's "sustainable communities" requirements were only "just starting to come into bloom" and "the schemes aren't harmonized"; that SB 375 needed to play out without adding new complexity.

She said standards worked out under SB 375 in Sustainable Communities Strategies might actually come out different from regional averages for CEQA purposes, requiring double calculation. She argued the new VMT guidelines – especially in the (b)(1) area discussing thresholds – should be handled as a cross-reference to regional averages the MPOs have already calculated under SB 375.

The Holland & Knight comment states as an aspect of inconsistency between the OPR draft and existing SB 375 standards that SB 375 metrics in the Bay Area find new market-rate infill development tends to command higher prices from richer residents, so that "affluence rather than location has led to higher VMT levels." It complains, "OPR's proposal undermines the SB 375 planning framework with a VMT regime... that has social justice and planning implications that extend beyond CEQA's existing framework." It adds: "This proposal raises many litigation issues for those unhappy with the land use decisions" made in local Sustainable Communities Strategies and other local planning documents, "and allows project opponents to happily use CEQA to seek different projects or a different project location."

Are the models 'plug and play'?

A regional average, as the Subdivision (b)(1) guideline draft itself accepts, can be calculated a number of different ways: the subdivision suggests it be done "per capita, per employee, per trip, per person-trip or other appropriate measure." Further questions on how to calculate regional average data include how to count particular trips that fit multiple categories. The many different VMT modeling data sets linked in the draft's amended Appendix F further suggest a diversity of approaches. So does the Holland & Knight comment in its discussion of litigation potential surrounding the possibility of disputes over VMT calculation.

Smith expressed uncertainty whether the models were truly "plug and play," in the sense of whether "I can figure out on my smartphone whether or not my project is above or below" the applicable level of significance – though, again, he suggested "that'll be relatively easy to answer" and the hard part would be "how vulnerable is the methodology itself to one-off attacks by bad actors?" (i.e., by objectors to projects.)

Calfee's comments noted that disputes are common over the methodology of LOS traffic studies. He wrote, "OPR has identified approximately two dozen models that could be used to estimate VMT, many of which have been routinely used to estimate greenhouse gas emissions. To minimize disputes about methodology, the proposed Guidelines would specifically acknowledge the lead agency's discretion to pick its model and adjust model inputs and outputs as needed. The draft Guidelines would also expressly apply a rule of reason to the analysis."

Calfee wrote that for LOS analysis "depending on the scope of the project, the study can take months and cost tens of thousands of dollars," whereas "VMT can be estimated using free models within a couple of hours."

Responding to a prior similar comment from OPR, Monchamp said, "It's far more complicated and takes much longer." She said the draft did not define VMT, while she viewed the guidance on methodology as consisting mainly of links to studies. She said it "is not giving much direction at all as to how to do this," and hashing it out "is going to be a messy and cumbersome process."

Critics' opinions diverged again on the types of projects that Subdivision (b)(1) said "generally may be considered" to lack significant impact: projects within half a mile of good transit, or projects expected to reduce local VMT overall." Eaken said a project near transit could still be high-VMT so such projects should be reviewed for impacts. Even on a transportation project, Eaken said unless it was "crystal clear" that a project wouldn't raise VMT, as with a bike lane, projects should still be analyzed. For example, she said, a new carpool lane might add capacity to a road, hence couldn't be presumed to lack impact.

Taking the opposite view, Monchamp described projects outside the 0.5-mile radius as "shackled" with many more mitigation measures.

Calfee wrote, under the heading of factors likely to reduce litigation, "Projects near transit generally will not be subject to doing a study, so there will be fewer issues to litigate."

Smith, saying he'd borrowed the point from Monchamp's colleague Jennifer Hernandez, argued it might work better to drop VMT analysis from some projects entirely, as "by definition good, and we're not going to analyze the traffic impacts, period." Which he said wasn't to argue against local authorities doing their own traffic analyses, only against allowing traffic analysis to push a pro forma project toward a required EIR.

Monchamp, having noted that the proposed Subdivision (b)(1) language does not really grant exemptions, said "Oh, exemptions would be lovely." She said so would clarifications: "We're not against the VMT approach. I don't think you're going to get anyone on our side of the world to say VMT is bad, it's just VMT has to be done more clearly and with more certainty."

Cumulative impact

David Pettit of NRDC asked how cumulative impacts would be handled. Calfee, in an interview clarified by written comments, responded that under current law, consistency with a plan that deals with a cumulative problem can support a presumption that a contribution is less than significant. He wrote: "If you are asking 'does my project have a cumulative VMT impact?' you might look at whether the project is consistent with the regional sustainable communities strategy to help address cumulative impacts."

Pettit had asked how fairness would be maintained if one of several similar projects in a popular infill area became the first to exceed local average VMT for its building type.

Calfee responded, "When you look at the data and the maps, that's not how it appears to play out." He noted some regional governments have been developing maps of VMT levels for areas within the region, in part through past work under SB 226, which also refers to regional averages. Recalling the lead agency's power to use professional judgment in adjusting models, he wrote: "If the model gives you a result that does not make sense, the proposed Guidelines would recognize that it is appropriate to make adjustments."

Comment on the OPR draft is due October 10, 2014.


  • The August 6 proposal from OPR is at: via
  • Holland & Knight's critical comment:
  • Latham & Watkins:
  • Fehr & Peers:
  • Streetsblog LA initial news analysis:
  • CityLab preview before draft release: