More on BACA and data centers

Considering the arguments in a recent op-ed arguing Prop 45 does not cover data centers

A key issue in the debate over the Cal Chamber of Commerce CEQA initiative (BACA, or Proposition 45 on the ballot) is whether it includes data centers as the types of projects that are eligible for the CEQA exemptions and vested rights provisions in the initiative.  I’ve written about why they may well fall within the scope of BACA.

There’s an op-ed to the contrary (available here).  It’s well written, and if Proposition 45 passes, I expect this to be the arguments taken by opponents if a data center project tries to use Proposition 45.  But it is only one set of arguments – and, contrary to the op-ed, I really don’t think the answer to the question of whether Proposition 45 covers data centers has a “clear” answer.

To make the point, I’ll channel the arguments likely made by a project proponent for a data center in a world in which Proposition 45 passes, and that proponent wants to use Proposition 45.  As a reminder, the key question is whether data centers are “essential broadband Internet access projects” that are covered by Proposition 45.

The op-ed relies heavily on claims that the language in Proposition 45’s definition of “essential broadband Internet access projects” parallels language in another section of California law (in an entirely different code, Government Code Section 53167) that defines what “Public Broadband Services” are for purposes of public provision of those services.  The op-ed then argues that this definition is distinct from “edge providers” (also defined in Section 53167) which provide the kinds of services that data centers provide.  Finally, the op-ed notes that this distinction is the same as the one used by the FCC in making regulatory decisions.

That is true.  But at the same time, there is no cross-reference in the initiative to Government Code Section 53167, to federal law, or FCC decisions.  Nor is there any such reference in the ballot materials, the relevant legislative history.  So the argument has to be that a court, interpreting this provision, would choose to rely heavily on the textual similarities, without explicit cross-references, to resolve ambiguities in the meaning of “essential broadband Internet access projects”.  It’s possible a court would do that, but it is far from guaranteed.  (Courts often focus within the statute – in this case Proposition 45 itself – to resolve ambiguities, see e.g., People v. Valencia, 3 Cal.5^th 347, 364-65 (2017), though they do look elsewhere within the specific code too, see e.g., People v. Anderson, 28 Cal.4^th 767, 787-88 (2002).  This is asking a court to look outside the specific code.)  As I noted in my prior posts, the data center industry has heavily pitched that their services are essential to use of the Internet by retail consumers, and has sought to blur the line (when it is advantageous for them to do so.)

The op-ed’s second primary claim is that the provision of Proposition 45 that calls for the initiative to be broadly interpreted to “to afford the fullest possible weight to the interest of, and the approval and realization of, essential projects” (Proposed Section 21029(a)) can’t expand the scope of what counts as an essential project, and therefore won’t affect a court’s determination of what is an essential project.  Again, a plausible position for a court to take, but not the only one.  It is also possible that a court, faced with a perceived ambiguity about what is an essential project, will see Proposed Section 21029(a) as requiring it to be generous in its determination of what falls within the scope of an essential project – since doing so would advance the goals of improving “massmarket retail service by wire service, wireless service, or radio to customers in this state.”  In other words, if a court interprets the definition of essential projects in the initiative as purposive – facilitation of certain kinds of activities to achieve certain goals – it well then see its task as broadly interpreting what kinds of projects achieve those goals, in order to achieve the “interest of, and the approval and realization of” essential projects”.

There’s another, more fundamental problem with the op-ed’s argument.  Even if data centers do not directly fall within the scope of the definition of “essential broadband Internet access projects”, they may well fall within the scope of “ancillary projects” for “essential broadband Internet access projects”, and thus are covered by the initiative as well.

The initiative requires exemption from current CEQA requirements, and broad vested rights, not just for “essential projects” but also “ancillary projects” for those “essential projects,” which are defined as “all related and ancillary public, private, and utility infrastructure and public service facilities required by a utility or public agency, or included in an essential project application as part of the “whole of the project,”  (Proposed Section 21028(m)(1)(B).  Could data centers be included as “part of the ‘whole of the project’” for an “essential broadband Internet access projects”, even on the narrower definition argued in the op-ed?

Certainly.  Data centers already are co-located with the routing and other key infrastructure for Internet connectivity – often because doing so reduces latency and costs.  (See this example of “Internet exchanges”, or this example of “network peering” that pairs data center infrastructure with co-location with network service providers.)  So a data center proponent could pair the hosting of “edge providers” with Internet service provider operators, argue that the pairing is required to make connections run better, quicker, and make the project cheaper, and then argue the “whole of the project” must be treated as an essential project covered by Proposition 45.  And then the statutory interpretation provision requiring broad application of the Proposition would kick in, arguably requiring a court to defer to the proponent’s determination of what is ancillary to the essential project, in order to achieve the “interest of, and the approval and realization of” essential projects”.  The op-ed’s efforts to rebut this are not particularly convincing, claiming that this would “erase the underlying definition of the essential project itself.”  But that is not the case – in this scenario, there is an essential project (by the op-ed’s own standard), and there are other components that the proponent plausibly claims make the essential project cheaper and better, and without which the essential project would not be built (and thus are part of the “whole of the project”).  That is precisely how the initiative functions!

Again, unlike the op-ed, I’m not going to argue (and I haven’t argued before) that the answer is clear here.  It’s uncertain.  I think it’s likely that a court would follow the reasoning above, in part because of the initiative requiring it to be interpreted broadly – especially if proponents strategically start connecting data centers with Internet service provision.  But it’s not certain.

That’s still a problem though.  Because if I’m correct and the initiative is interpreted broadly to include data centers, there isn’t much ability to change that.  The initiative requires a 2/3 legislative majority to amend it.  That would be difficult, especially given the lobbying power of the companies building data centers.

Finally, the one last thing that bugs me in all this:  Why not just explicitly exclude data centers from the initiative?  The proponents are smart, strategic actors – they knew to amend their initiative to exclude the Delta water project, because they knew what a political football it was.  Data centers have become particularly unpopular lately, but it was an issue back in the winter.  A simple one line stating that data centers are not covered by the initiative would have obviated all of this debate.  And even more than that, the proponents could have added such a disclaimer to the ballot language for the initiative, which is the most important legislative history for the initiative – that language was finalized over the summer, when it was clear that the inclusion of data centers in the initiative would be a campaign issue.  But again, no such statement is in the ballot language.  That alone weighs heavily with me in understanding how the initiative will be interpreted.

BACA, CEQA, data centers