Calibrating Environmental Review to the Scope of Municipal Discretion Under the HAA


This is the last in a four part series. SLoGLaw thanks Chris Elmendorf and Tim Duncheon for this timely treatment of the important issue of affordable housing for California, and urban regions around the country--Ed.


The California Environmental Quality Act (CEQA) requires state and local agencies that have discretion to choose among possible options to study environmental effects before making their choice. In theory, this leads to better agency decisions. (A contestable claim – but that’s for another day.) But when other laws require an agency to select a particular option, CEQA doesn’t apply. There’s no reason to write a detailed list of the pros and cons of different options if you know from the start exactly which choice you have to make.


When a developer submits a housing proposal, the Housing Accountability Act (HAA) substantially limits the choices open to the city. So you would think that review under CEQA would be limited accordingly. You would, unfortunately, be wrong – at least as revealed by current practice.


So it is that a proposal to build 500 apartments on a downtown San Francisco parking lot, a block from the subway, in a designated “priority development area” under the region’s climate plan, ended up mired for years in the most extensive and costly form of environmental review required by CEQA: the Environmental Impact Report (EIR).


And why? Because San Francisco’s planning department had concluded, on the basis of a 342-page Initial Study, that a “fair argument” could be made that the Stevenson St. project may have a significant local environmental impact in the form of shadows, wind, or (during construction) noise and air pollution. The Initial Study evaluated the project’s potential impact relative to current environmental conditions nearby. It did not ask whether the project would have a significant marginal impact, relative to any other project of the size that the HAA entitles the developer to build on the site.


If the Stevenson St. project’s marginal impact would be close to nil (as we think likely), then the EIR was an environmentally pointless exercise. Its real function, apparently, was to give local activists and city officials a way to tie up the project until the developer either walked away or paid off TODCO, the politically connected nonprofit that led the charge against it.

The argument of this post is that the scope of CEQA review of housing development projects should be tailored to the scope of municipal discretion. A housing project should require an EIR only if the city exercises discretion to shape the project in some way that generates a significant marginal impact, relative to what the HAA compels the city to approve.


Our approach would not “relieve local governments from complying with” CEQA. But it would require overturning or significantly limiting several judicial precedents that have been incorporated into the official CEQA Guidelines. As such, our proposal poses a stark test of whether the HAA really is a super-statute, one that “sticks in the public culture” and exerts “a broad effect on the law.” If courts and the gubernatorial appointees responsible for the CEQA Guidelines get behind our approach, then the HAA will in fact “meaningfully and effectively curb[] the capability of local governments” to hobble housing development projects. If they do not, there can be little doubt that NIMBY cities will become ever more expert at exploiting CEQA to undermine the HAA.


“Effect” Relative to What?

We begin with an elementary point about causation. It is senseless to try to characterize the environmental effect of a proposed housing project without comparing it to some alternative use of the site. Consider an analogy: What is the effect of a new drug or medical device? The answer depends on what you’re comparing it to. Relative to a placebo, the effect of the new drug may be large. Compared to the best treatment currently in use, the effect of the very same drug could be small or even negative.

The same goes for housing projects. They have effects only when they’re compared to some alternative. Let’s call the point of comparison the reference alternative. What is conventionally labeled “the baseline” in an environmental impacts study is, properly understood, a compound of two things: an alternative use of the site (the “reference alternative") and a projection of environmental conditions in and around the site conditional on that use of it.

CEQA analyses, relying on CEQA caselaw, usually elide this fundamental point. By convention, they purport to measure the “effect” of a project relative to “current environmental conditions” on the site and in its vicinity. This is a misleading point of reference if current environmental conditions would change absent the project. No medical researcher would measure the “effect” of an experimental treatment by comparing the health status several years in the future of elderly patients who received the treatment with their health at the time the treatment was administered. That comparison would obscure the effect of the treatment, because old people tend to decline as they age.

The CEQA analyst’s conceptual mistake about baselines is not a problem in contexts where the permitting agency has authority to deny the project and doing so would maintain current environmental conditions. In such circumstances, the current-environmental-conditions baseline is equivalent to treating the “no-action alternative” as the reference alternative. This is like a placebo reference condition in a drug trial.

But the current-environmental-conditions baseline is nonsensical when the public decisionmaker lacks legal authority to maintain it. This is precisely the situation that cities face when developers propose HAA-protected housing projects. Cities may place discretionary conditions of approval on such projects, but they may not deny the project or reduce its density. Accordingly, the environmental impact of the project should be gauged relative to a reference-alternative project of the scale the city is required by law to approve.


An HAA-Informed Protocol for CEQA Review of Housing Projects

The first step in CEQA review is preparation of the Initial Study, which seeks to determine whether there is a “fair argument” that the proposed project “may” have a significant impact on the environment. If the answer is “Yes,” then the project proponent must pay for an EIR that fully analyzes the potential effects identified in the Initial Study.


The policies of the HAA and the policies of CEQA can be reconciled, to some extent, by asking the threshold HAA question at the outset of the Initial Study: Does the project as proposed comply with applicable, objective general plan, zoning and development standards, as defined in the HAA? If it does, the city may deny or downsize the project only if it violates a written, objective health or safety standard within the meaning of the HAA. So for zoning-compliant projects, the Initial Study should gather information about potential health / safety violations and determine whether a preponderance of the evidence establishes a violation. A conventional CEQA review is in order only if such a violation is established (because the city may deny the project).

For projects that comply with general plan and zoning standards, and that don’t violate health or safety standards, it’s meaningless to conduct an environmental review that benchmarks the project against a no-action alternative or “current environmental conditions” in the vicinity of the site. The city’s discretion is limited to altering the project with conditions of approval that do not reduce its density, and the CEQA baseline should be defined accordingly.


There are two plausible reference alternatives in this circumstance. First, the analysis could proceed using a project-as-proposed benchmark. The reviewer would inventory any discretionary conditions of approval that the city is considering imposing on the project, and then benchmark (1) environmental conditions if the project goes forward with the discretionary condition(s) imposed, against (2) environmental conditions if the project goes forward in the form it was proposed. The difference represents the environmental effect of the city’s exercise of discretion.


To illustrate, if the city were considering a discretionary condition of approval that would require rooftop solar panels, and concerns were raised about glare from the panels, the Initial Study would undertake to determine whether there is a fair argument that the rooftop solar condition may cause a significant environmental impact in the form of glare, relative to the scenario in which the city approves the project in the form it was proposed.

Alternatively, the city could posit a green-reference benchmark, measuring the impact of an HAA-protected project relative to a model “green” project of the same density on the same site. The green-reference alternative might be defined as a project that provides the minimum number of on-site parking spaces; that uses low-energy building materials; and that minimizes impermeable ground cover (insofar as the city has authority to impose such conditions). The key point is that the green reference alternative would be a legally available option, and as such represents an informative benchmark against which to compare the proposed project.


Under either model, it would be the rare HAA-protected project that requires an EIR. Cities do not often impose conditions that reduce environmental amenities in the vicinity of a project, so the project-as-proposed benchmark would yield pro forma negative declarations in most cases. As for the green-reference benchmark, developers who anticipate opposition from neighbors, unions, or other interest groups would likely conform their proposal to the benchmark. If the project as proposed is HAA-protected and uses the green-reference design, then by construction it would have no environmental effects for CEQA purposes.


Does CEQA Allow It?

The idea of tailoring the scope of environmental review to the scope of agency discretion has precedent under statutory analogues to CEQA at the national level and in New York. Review under the National Environmental Policy Act (NEPA) is limited to “effects” that are proximately caused by the agency’s discretionary choices. Thus, in Department of Transportation v. Public Citizen, the U.S. Supreme Court held that an environmental impact study prepared by the Department in connection with the North American Free Trade Agreement need not analyze pollution resulting from an increase in Mexican truck traffic, because the Department had no legal authority to exclude Mexican trucks. To date, no California court has ruled on whether CEQA incorporates the proximate-causation theory of Public Citizen, but California courts do seek guidance from NEPA precedents when tough questions arise under CEQA.

In New York, courts got to a similar place by rejecting the “no-build baseline” in cases where the project proponent may build something as of right. Specifically, if a developer proposes an office or residential building that would require rezoning, on a site where a smaller building is allowed as of right, the effect of the proposed project is analyzed relative to the “as-of-right alternative” rather than the “no-build alternative” or “current environmental conditions.” Because the city lacks authority to deny the smaller project, it would be uninformative to conduct an environmental review using a no-project baseline.


Like the National Environmental Policy Act and New York’s State Environmental Quality and Review Act, CEQA exempts “ministerial” permits from environmental review. Discretion is always the trigger. However, the Court of Appeal has held in several cases that if a city has any discretion to shape a project, the city must analyze and mitigate the impact of project “as a whole” relative to a current-environmental-conditions baseline. Projects whose permitting is “not wholly ministerial and not entirely discretionary but a compound of both” have been treated as entirely discretionary for CEQA purposes. In one case, an EIR was produced using a zoning-complaint project baseline, similar to New York practice, and the California Court of Appeal rejected it out of hand. The court faulted the EIR for not “present[ing] a clear or a complete description of the project's impacts compared with the effects of leaving the land in its existing state.”


This line of cases is rooted in CEQA’s traditional premises: that new construction is bad for the environment, and that CEQA should be construed broadly to give “the fullest possible protection” to the environment. The working assumption is that requiring more environmental review and mitigation is the greener way. But as we’ve seen, the HAA inverts this premise when it comes to housing. The HAA declares new construction of zoning-compliant housing projects to be presumptively good for the environment, and it aims to “meaningfully and effectively curb[] the capability of local governments to deny, reduce the density for, or render infeasible housing development projects.” A reading of CEQA that leaves cities with open-ended discretion to require time-consuming studies and costly mitigation of so-called “impacts” that are not even proximately caused by the city’s exercise of discretion would do pointless violence to the policy of the HAA.


In the near term, however, any effort to use the HAA to put a limiting gloss on misbegotten CEQA-baseline precedents would be complicated by the fact that those precedents are now codified in the official CEQA Guidelines. The Guidelines stipulate that “the physical environmental conditions in the vicinity of the project, as they exist at the time the notice of preparation [of the EIR] is published . . . will normally constitute the baseline physical conditions by which a lead agency determines whether an impact is significant.” This “existing conditions baseline” “shall not include hypothetical conditions, such as those that might be allowed, but have never actually occurred, under existing permits or plans.”


The only exception that the Guidelines presently recognize is that an agency may use a “projected future conditions ... baseline ... if it demonstrates ... that use of existing conditions would be either misleading or without informative value to decision-makers and the public.” This exception codifies a practice that developed around very long-term projects, such as railways. Neither the Guidelines nor any published case approves the use of a “future-conditions baseline” where the future in question is a build-out of the project site under an alternative development scenario. Then again, neither the Guidelines nor any published case has considered the implications of the HAA for CEQA baselines or causation.


Though it wasn’t written for the HAA problem, the Guidelines’ narrow allowance for “future conditions” baselines at least recognizes that circumstances may arise where the conventional baseline is inappropriate. And the crux of our argument is that it is misleading and uninformative – and a colossal waste of resources, and a serious threat to the environmental and housing policies of the HAA – to require developers to engage in a multi-year analysis of putative environmental “effects” that are the byproduct of a nondiscretionary statutory mandate, not the discretionary choices of the local permitting authority itself.


The Governor’s Role

Courts are conservative creatures. It’s not in their nature to upend long-established precedents just because those precedents are at odds with another statute the Legislature has declared to be super, but which also preserves the statute that spawned the bad precedents.


But courts don’t implement CEQA by themselves. CEQA authorizes the Governor’s Office of Planning and Research and the Natural Resources Agency to issue implementing guidelines. The CEQA Guidelines codify judicial precedent, but they also embody policy choices, and the courts give measured deference to such choices. If environmental review is to be reshaped by an HAA-informed theory of causation, the Guidelines are the best tool at hand.


The Guidelines are a good tool for this purpose not only because making policy and changing direction is, by tradition, more squarely in the agency wheelhouse than the judicial wheelhouse, but also because of politics. Through his appointments and directives, the Governor can shape the Guidelines. And, presently, the Governor is better positioned than any other state-level actor to navigate the politically treacherous waters of CEQA reform.


Though it was a noble environmentalism that made CEQA super in the 1970s, the continued strength of CEQA today has much to do with the constellation of interest groups – first and foremost the building-trades unions – that have mastered the art of using CEQA to extract costly concessions from developers. In expensive housing markets, the threat of CEQA litigation and delay can be used to make developers sign project-labor and “community benefit” agreements with influential unions and nonprofits. The building trades wield a lot of power in Sacramento, and in recent years they’ve derailed every legislative proposal for CEQA reform or streamlining except those that require qualifying projects to use union labor. Not even a trivial bill that would let churches build affordable housing without CEQA review could escape Labor’s grip.


But Governor Newsom is riding high. He was elected by a twenty-four point margin. He defeated the recall attempt by the same margin. The California Republican Party is all but dead, and the odds that the Governor will face a strong Democratic challenger when he’s up for reelection in 2022 are remote. A tussle with the building trades wouldn’t derail his career.


Of course, no Governor can single-handedly make the HAA “stick” in a manner that limits abusive use of CEQA. If there were a legislative consensus that project-labor agreements are more important than housing production, the Legislature could quickly abrogate any reformist CEQA Guidelines and then override a gubernatorial veto. But it’s a fair hope that no such veto-proof consensus exists. The Republican minority is no fan of CEQA, and Democratic legislators are loathe to override their co-partisan Governor. Moreover, politically vulnerable legislators, who wouldn’t dare cast a roll call vote against the trades, may acquiesce in the appointment of pro-housing committee chairs, who in turn could block any bill that would reverse the Governor’s reform of the CEQA Guidelines. It’s also possible that a transparent, public debate about CEQA abuse – a debate that would probably accompany any legislative effort to roll back the reformed Guidelines – might itself subtly alter the politics of CEQA reform, in a way that gives the HAA the upper hand.


Although super-statutes on Eskridge and Ferejohn’s telling embody great normative principles, it appears that CEQA’s continued potency owes much to a small number of rent-seeking interest groups that depend on it. The generational clash between the HAA and CEQA is about power as much as principle.