U.S. Fish and Wildlife Service v. Sierra Club

Oral argument transcript · No. 19-547 · October Term 2020 · argued November 2, 2020 · the petitioner won

Decision: votes, opinion and citations · Official transcript (PDF) · Argument audio at the Supreme Court

Advocates

• Matthew Guarnieri, Assistant to the Solicitor General, for the petitioner
• Sanjay Narayan, Oakland, California, for the respondent

Transcript

From the Court's official transcript; speaker labels as in the transcript, with advocates' full names from its list of appearances.

Chief Justice Roberts: We will hear argument first this morning in Case 19-547, United States Fish and Wildlife Service versus Sierra Club. Mr. Guarnieri. ORAL ARGUMENT OF MATTHEW GUARNIERI ON BEHALF OF THE PETITIONERS

Matthew Guarnieri: Mr. Chief Justice, and may it please the Court: The December 2013 draft biological opinions are privileged, predecisional, deliberative material. They were written by staff at the Services as a recommendation to agency decisionmakers about the position the Services should adopt in the ongoing consultation with the EPA. When the relevant decisionmakers were presented with these drafts, they did not adopt them, they did not sign them, and they did not even transmit them in full to the EPA. Instead, they decided that more work needed to be done before making a final decision. The Ninth Circuit committed two 1 principal errors in concluding, nonetheless, that these drafts are outside the scope of the deliberative process privilege. First, the Ninth Circuit treated the drafts as final, rather than predecisional, because it viewed them as the Services' last word on the version of the EPA rule under consideration in December 2013. But the EPA modified its approach, and the Services never had any occasion to make a final decision about the abandoned version of the EPA rule. In the D.C. Circuit's memorable words, the December 2013 draft opinion "died on the vine" without ever blossoming into a final decision. Second, the Ninth Circuit viewed these drafts as final documents because the drafts largely don't contain red-lining, marginal comments, or other obvious signs of still being in flux. That reasoning is unsound. No one would confuse a law clerk's draft with a final decision by a judge, even if the draft is pristine. The key point is that the decisionmakers at the Services had not yet made 1 up their mind. Their deliberations had not yet come to an end. When the Services did make a final decision in May 2014, they released an 85-page joint opinion explaining their reasoning to the public. Here, Respondent seeks to compel the disclosure of earlier drafts which recommended reasoning that the Services never adopted about a version of the EPA rule that never saw the light of day. The Court should reject Respondent's efforts to pry into those materials.

Chief Justice Roberts: Mr. Guarnieri, before you can decide whether something is predecisional, you have to know what the decision is. And why -- why isn't the decision here EPA's final rule on the cooling water intake structures, and the Services' opinion simply is predecisional from the perspective of that final rule?

Matthew Guarnieri: Mr. Chief Justice, we think the final decision here is the Services' decision in the ongoing consultation. So that -- that is -- the final decision occurred in May of 2014, when the Services exercised their 1 authority under the Endangered Species Act to render a biological opinion with respect to whether EPA's proposed action would cause jeopardy to endangered species.

Chief Justice Roberts: But that doesn't do -- I mean, that -- that itself doesn't rep -- represent any action by the Service with respect to anything other than the EPA decision. I mean, it is predecisional with respect to that decision.

Matthew Guarnieri: That's right, Your Honor, but -- but we think the deliberative process that -- that should be the focus of the Court's attention here is the deliberations that were occurring within the Services about whether or not the EPA's proposed action would cause jeopardy, not --

Chief Justice Roberts: No, I know that's what you think. I'm trying to figure out -- figure out why. I mean, you talk about, within an agency, the different steps in the process, and you say: Well, none of those steps is actually, you know, final and decisional. But, here, all of a sudden you get to the end of the Services' role and it's final and 1 not predecisional, even though it's just part of another ongoing process. In other words, I'm not sure that your position doesn't prove too much.

Matthew Guarnieri: Mr. Chief Justice, we -- we think the statute itself makes clear that there is a decisionmaking process that concludes with the issuance of the final biological opinion. That's in Section 7(b)(2) of -- of the Endangered Species Act. The implementing regulations also make clear that, with respect to the interagency consultation, it concludes with -- that -- that process concludes with the issuance of a final opinion. Now, of course, if -- if the Court were to view the -- the deliberate -- the deliberations here more broadly as encompassing also the EPA's rulemaking, then it's clear that the EPA didn't make a final decision and didn't issue a -- a final rule in that rulemaking until May 2014. So the drafts that are at issue here would also be predecisional with respect to the EPA's final rule.

Chief Justice Roberts: Thank you --

Justice Thomas: Counsel, the --

Chief Justice Roberts: -- counsel. Justice Thomas.

Justice Thomas: Thank you, Mr. Chief Justice. Counsel, the -- I'd like to follow up with the Chief Justice's line. The -- what if there were not a regulation that prohibited the issuance of a final opinion before or while the draft was under review by the requesting agency?

Matthew Guarnieri: Well, you know, certainly, we think it's helpful here that the -- the regulations clearly contemplate that there would be a sharing of drafts in some circumstances between the Services and the action agency. And, here, the Services didn't actually even reach the point of sharing the draft opinion in December 2013. The drafts were never transmitted in full to the EPA at that point. But -- but, counterfactually, if there were no regulation, then still at least on the facts here, it's clear that the Services had never made a final decision in the consultation 1 in December 2013. The declarations that have been submitted by agency officials make clear that the relevant decisionmakers did not adopt these drafts when the drafts were presented to them in December 2013. So we really think that's dispositive here of the fact that the deliberations were ongoing at that point in time.

Justice Thomas: So what if you were right -- I'm trying to figure out, like, if you're right up to the line that there is -- there is no more deliberation, that let's say it's there -- it's a final, final draft and you simply are call -- calling EPA to give a heads-up that you're about to send it in -- in five minutes. Would you make the same argument? Anything short of actually sending it, that's what I'm getting at.

Matthew Guarnieri: Yes, Justice Thomas. Now, of course, we're quite far from that point in this particular case. But I -- I think, in general, the principle that we are advocating here is that until there is a -- until there is 1 actually a final decision, the agency decisionmakers are free to change their mind in a consultation. And so that -- that is really the critical distinction between sort of having a mere final draft that is on the verge of being transmitted and actually having a final agency decision that represents the Services' final opinion in -- in the consultation. And until reaching that point, the deliberations are -- are still ongoing, and -- and the Services -- the decisionmakers at the Services could change their mind about any aspect of the agency's analysis.

Justice Thomas: So anything short of -- of just pressing the send button is -- is -- is non-final?

Matthew Guarnieri: Well, it's not -- to -- to be clear, it's not the transmittal. It's not hitting the send button to send it to the EPA that we think is the critical distinction. It's -- it's the point in time at which the agency decisionmakers actually exercise their authority to issue a biological opinion in the consultation. And -- and they were far short of 1 that here in December of 2013.

Justice Thomas: Thank you.

Chief Justice Roberts: Justice Breyer.

Justice Breyer: Well, I'm not -- one, I'd like any comment you have about the test. I mean, the object is will this document, in fact, diminish agency decisionmaking quality by -- by discouraging the staff and others from debating if it's going to become public? And you might have a better choice or not, I'm curious, about the -- the words that have been used are predecisional and deliberative. Hmm, maybe. Okay. So that's one in the back of my mind. Anything you want to say? And the other thing, at least in some of these documents, and we'll have to look through the record, it -- it seems to have reached a final stage. I mean, people say, when you make these final changes, which they made to the draft, I can e-mail the Assistant Director, and we have an autopen with his signature we can use to send it out, and they'll send it to the EPA. And, in fact, that's what they normally 1 do. I mean, normally what they do is they send these things over before they're absolutely final. The EPA makes changes. And then it never appears. And that happened apparently thousands of times, and only twice did they actually publish it. So -- so those -- those -- those are -- do you get what I'm driving at? And, if so, I'd appreciate your -- your thoughts.

Matthew Guarnieri: Sure. Well, with respect to your second set of questions, Justice Breyer, the e-mails that you're referring to are e-mails about finalizing the draft for transmittal to the EPA. And so it's abundantly clear in those e-mails, even in the e-mail suggesting that the transmittal letter could be signed by autopen, it's very clear that the Services understood themselves to still be working on draft biological opinions, and they had committed previously to sharing a draft with the EPA. So even if they had transmitted that draft or transmitted those documents, they -- they understood themselves to be transmitting a 1 draft that did not yet represent a final decision. With respect to Your Honor's first question, the lower courts have understood this Court's precedent on the deliberative process privilege to require that a document be both predecisional and deliberative in order to qualify for the privileges. We're not taking issue with that. We think, if the Court were to apply that framework here, these documents would certainly satisfy it.

Justice Breyer: Okay. Thank you.

Matthew Guarnieri: Justice Breyer, if I may also, your question alluded to the fact that there are relatively few jeopardy opinions. That -- that's certainly true. There are some -- there's an empirical study cited in the amicus briefs suggesting that jeopardy opinions are relatively rare. But we don't really see any particular problem with that. And I think that could be simply a sign that the consultation process required by the Endangered Species Act is working as intended. Federal agencies are 1 incorporating these standards into proposed agency actions and are --

Chief Justice Roberts: Justice Alito.

Matthew Guarnieri: -- applying law and avoiding actions that would cause jeopardy.

Chief Justice Roberts: Justice Alito.

Justice Alito: Are there examples of sit -- situations in which a jeopardy biological opinion has been issued, but the action agency has then gone ahead with the action in the face of that?

Matthew Guarnieri: Justice Alito, I'm -- I'm not aware of such a -- such a -- such an example. There -- there are a handful of examples -- so, to -- to step back for a second, the Endangered Species Act -- as -- as the Court discussed in Bennett against Spear, the Endangered Species Act at least theoretically permits an action agency to decide that, notwithstanding the biological opinion, the action agency has decided that its agency's action would not cause jeopardy. I -- I don't know that that's ever occurred. There are a handful of examples. The -- the statute also permits an action agency to 1 obtain an exemption from a -- a -- a -- a cabinet-level committee, and that has occurred on -- on a handful of occasions.

Justice Alito: Well, if it -- if it almost never occurs, then something that is labeled a draft biological opinion may really be tantamount to the Services' final word on the subject, unless it can be persuaded by the action agency to change its opinion or the action agency makes an adjustment in what it was previously proposing to do. Isn't that the case?

Matthew Guarnieri: Well, I -- I think the -- the draft aspect of the opinion -- I mean, from our perspective, the key point here is that the agency decisionmakers at the Services had not yet actually made up their mind about whether the version of the rule that was under consideration in December of 2013 would cause jeopardy.

Justice Alito: Okay. Well, maybe that's -- maybe that's true, but I'm interested in where your argument goes, where we should draw the line. Do you want us to draw a line between 1 those draft biological opinions that do not reflect -- that reflect all of the deliberation that the Service intends to conduct internally, but -- and those that do not reflect all of the deliberation that the Service wants to conduct internally, or does your argument logically lead to the conclusion that no draft biological opinion can ever be final?

Matthew Guarnieri: Well, I -- I think it's the latter, Justice Alito. No draft biological opinion -- by -- by definition, if it is a draft opinion, then it is predecisional and deliberative because the agency has not yet made up its mind. The Services have not yet made up their mind in the ongoing consultation. And -- and there are good reasons for that, and one of them is that, you know, as the regulations contemplate, the Services will often share these draft opinions with the action agency, and that process sort of -- the give and take between the Services and the -- the agency that is consulting the Services can be helpful to refine the -- the draft opinion and to -- for the Services to better understand the proposed agency --

Justice Alito: Thank --

Matthew Guarnieri: -- action.

Justice Alito: -- thank you, counsel. My time has expired.

Chief Justice Roberts: Justice Sotomayor.

Justice Sotomayor: I am following up a little bit on Justice Alito's questioning. In Bennett, we held that biological opinions, while technically advisory, have "a powerful coercive effect" on the action of the agency. Why is it that a draft jeopardy opinion doesn't have the same coercive effect? As I think the Ninth Circuit pointed out, what was at issue was the November rule that the EPA was proposing, and the draft that was sent to the EPA made them change their mind. They did something completely different. With respect to the decision relating to that November action, the draft opinion did exactly what a final opinion is intended to do. So I -- I understand your basic argument that it wasn't clear the agency's final decisionmaker had accepted that that was the jeopardy opinion they were going to get, but I 1 go back to Justice Thomas's question. If that agency head was about to sign it and said, ah, I'll -- I'm just going to send it to them and tell them I'll sign it on -- on Monday, I don't want to go into the office on Sunday, it would be your argument that that wouldn't be a final opinion worthy of disclosure?

Matthew Guarnieri: Well, Justice Sotomayor, to -- to -- to sort of take those questions in -- in sequence, I think it's abundantly clear that in Bennett against Spear, the Court was discussing final biological opinions, as evidenced by the Court's focus on the legal force and effect of the incidental take statement that is issued when the agency -- when the Services render a jeopardy -- excuse me, a no jeopardy opinion. So the -- the -- the Court's focus there was on the fact that the final biological opinion does have real force and effect. It has legal consequences. None of those consequences attach to a draft biological opinion. The statute and the regulation attach no legal consequences whatsoever to a draft biological 1 opinion of the kind that are at issue here.

Justice Sotomayor: Counsel, I have one question I want to get to. In the Ninth Circuit, you agreed that a remand would be appropriate to determine whether the documents contain segregable factual information. Do you think that if we were to rule in your favor, we would still have to remand for that to happen?

Matthew Guarnieri: Yes, Your Honor. That -- that would be appropriate. Under -- under Section 552(b), if a document qualifies for one of the exemptions set forth in subsection (b), then and only then would an agency determine whether, notwithstanding the fact that the document qualifies for an exception, there are portions of it that could be segregated and released. So that -- that did occur after the Ninth Circuit decision. That did occur with respect to three documents that the court of appeals found to qualify for Exemption 5. If this Court were to sustain our assertion of Exemption 5 here, then the same kind of analysis would follow on remand.

Justice Sotomayor: Thank you, counsel.

Chief Justice Roberts: Justice Kagan.

Justice Kagan: Mr. Guarnieri, could I focus on the part of your argument which is that the relevant decisionmakers were still working on the draft and give you a hypothetical, which is that there was -- the draft came to the relevant decisionmaker, and he looked at it and he realized that it was going to cause a big problem for the EPA, and so he said: You know, I'm not going to send this over. I'm going to give the EPA head a call and just tell him everything in it, and that will ensure that there's nothing FOIA-able that -- in this document. What -- what would your answer to that be? You know, he has the document, he's not working on it, but -- but he doesn't want to make it FOIA-able. And can he end-run this in that way?

Matthew Guarnieri: Justice Kagan, there is a body of existing law in the lower courts addressing circumstances in which an agency has implicitly made a final decision. Some of those 1 principles might -- might be brought to bear on the hypothetical that Your Honor's posing. But -- but, here, there's really nothing in the record to suggest that the agencies had implicitly made a final decision, even if they had not memorialized that by, for example, signing and publicly issuing a biological opinion. And -- and, in fact, here, the evidence is really all to the contrary. There are declarations from agency officials. Including for the Fish and Wildlife Service, there's a declaration from Assistant Director Frazer, who was himself the agency decisionmaker for Fish and Wildlife, and he says that he was presented with these draft declarations and he determined not to make a decision at that time because he felt that more work was needed in the consultation.

Justice Kagan: Yeah, it -- it -- it -- it's a very general statement. Do you have any sense of what more work needed to be done? Because one way to understand what happened here is that everybody really responded and acted as if there were a completed draft opinion. You 1 know, there was sending the reasonable alternatives over. The AP -- the EPA starts talking to the Service about how to change its rule. The EPA does change its rule. It was as if -- you know, everything that happened was as if there had been a final draft opinion that was sent to the EPA.

Matthew Guarnieri: Well, Justice Kagan, among other things, Assistant Director Frazer said that -- and this is at page 58 of the Joint Appendix -- that elements -- key elements of the EPA's rule were still being deliberated within the EPA. So I think the declaration reflects that there was a -- a -- a significant degree of fluidity here both as to the -- the -- what the Services planned to say in the consultation and -- and also what the EPA's final rule would look like. Now it's difficult to describe exactly the conversations that occurred from December onwards without reviewing the substance of the agency's privileged discussions, but the declaration, I do think, makes clear that, you know, there -- there were a number of moving 1 parts here. It wasn't simply EPA revising its rule in response to the draft biological opinion.

Justice Kagan: Thank you.

Chief Justice Roberts: Justice Gorsuch.

Justice Gorsuch: Good morning, counsel. Let's say EPA had decided at the end of it to just withdraw its rule and -- and give up, at least for the time being, maybe come back to it in five or 10 years. Would that be a final decision? And, if so, would that have made the last draft that the Service gave to EPA discoverable or not in your view?

Matthew Guarnieri: Justice Gorsuch, as I was trying to articulate earlier, we think the -- the key question here is whether the Services had made a final decision. And so, if the Services rendered a final biological opinion finding jeopardy in the consultation and that caused the EPA not to proceed with its proposed action, then, certainly, there the Services' final opinion would itself not be privileged and -- and -- 1 and, actually, as a matter of agency practice --

Justice Gorsuch: All right. I -- I --

Matthew Guarnieri: -- serve --

Justice Gorsuch: -- I got that, counsel. I guess what I'm ask -- so would the EPA decision itself not to proceed, it died on the vine, but would that be nonetheless final and itself discoverable?

Matthew Guarnieri: Would -- would the EPA's decision be final and discoverable?

Justice Gorsuch: That's the question now.

Matthew Guarnieri: I -- I suppose it would depend. I mean, if the EPA memorializes its decision not to proceed in the rulemaking in some sort of agency document, then its explanation of -- of why it had chosen not to proceed would not -- hypothetically would not be predecisional or deliberative and, therefore, would be FOIA-able.

Justice Gorsuch: But, if they just decide in its last draft, you know, we're not -- we just -- it's too hard, we can't do it, we give up, internally, but it doesn't -- it 1 doesn't publish anything, would that be final?

Matthew Guarnieri: No, I -- I don't think that would be final. And, in fact, that would be reminiscent of the situation that this Court contemplated in Footnote 18 of its decision in Sears, which is that, you know, you're -- you're going to have privileged internal deliberations that do not result in any final decision because, you know, you could have a degree of kind of agency brainstorming that doesn't ultimately lead to any final agency action. And -- and, in that circumstance, the agency's deliberations are privileged, even though they do not culminate in any specific final decision.

Justice Gorsuch: And what if -- what if -- what if, alternatively, the leadership of the Service had, you know, signed that last draft and sent it over to EPA? EPA didn't -- what -- whoever -- whatever happened at EPA happened, but the -- the Services signed something. EPA, though, you know, ultimately maybe decided not to do anything. Would the Service document be discoverable?

Matthew Guarnieri: If -- if the Services' document remained in draft form and had never actually been issued as a final biological --

Justice Gorsuch: No, I'm supposing now it was signed by the leadership of the Service and sent over to EPA.

Matthew Guarnieri: Yeah, in -- in that circumstance --

Justice Gorsuch: But EPA decided to do nothing. Its -- its regulation died on the vine.

Matthew Guarnieri: I -- I see. I see. Yes. That -- that would be the -- when the agencies reach a final decision in a consultation, they release their opinion, their biological opinion to the public, and --

Justice Gorsuch: Oh, no, they didn't. They sent it over to EPA because EPA's was still in draft form, but the Service decided they'd come to a final view on the draft of EPA and they signed it. Then what?

Matthew Guarnieri: Then -- then that -- that would be a decision that would -- we could not withhold under the deliberative process privilege because the Services had -- have 1 reached a final decision. They have exercised their authority under the Endangered Species Act. And there was a --

Justice Gorsuch: Counsel, thank you. My -- my -- my time is over. Thank you.

Chief Justice Roberts: Justice Kavanaugh.

Justice Kavanaugh: Thank you, Chief Justice. And good morning, Mr. Guarnieri. Just to follow up on the Chief's questions and Justice Gorsuch's questions, it's possible, I think, in your view, that even a final agency or -- opinion or memo could still be predecisional as part of a broader deliberative process, and a draft opinion, which we have in this case, in your view, is even more obviously predecisional. Is that correct in terms of a general statement?

Matthew Guarnieri: I -- I -- I do think that's correct. And -- and that's why, in our briefing, we -- we refer to these in places as drafts of drafts, because, here, the EPA -- the December 2013 draft biological opinion had not even reached the point of being transmitted as 1 drafts to the EPA.

Justice Kavanaugh: To follow up on something that Justice Kagan asked, does the motive of the agency official with respect to FOIA play a role in determining how a court should assess whether it's obtainable under FOIA?

Matthew Guarnieri: It -- it has not traditionally been a part of the analysis that this Court has engaged in for the deliberative process privilege. As I -- as I said to Justice Kagan, I think there are existing doctrines that can address any concerns along those lines, including doctrines under which an agency may be determined to have implicitly reached a final decision, even though a document might be notionally labeled a recommendation or a draft. But -- but those are simply inapplicable here. There's really no basis to infer that the Services -- that the decisionmakers at the Services had made a final decision in December of 2013.

Justice Kavanaugh: I think there's a concern lurking in this case that executive 1 branch officials might just stamp drafts on everything and, therefore, evade FOIA. Can you respond to that concern?

Matthew Guarnieri: Sure. You know, I -- I -- I take the point, Justice Kavanaugh, but we're just very, very far from that here. Here, we are in the molten core of the deliberative process privilege, where it's -- it's clear from the record that the agency -- the decisionmakers at the Services did not adopt the draft opinions when they were first entered.

Justice Kavanaugh: Well, if I could interrupt, I -- I understand that point as to this case, but how we frame the rule or the principle will matter. And how exactly would you have us frame the principle of law that governs here?

Matthew Guarnieri: Well, we -- we accept that the -- the lower court's formulation that a document must be both predecisional and deliberative is -- is a sort of -- accurately captures the substance of this Court's case law. With respect specifically to whether or not a document is labeled "draft," we do think that is important. I mean, it has a 1 significant meaning within the executive branch when a document is labeled "draft." It's a signal to other parties that the document has not yet been finalized. But, you know, of course, to address the concern that -- that Your Honor has mentioned, the labeling of a document as "draft" wouldn't necessarily be dispositive. It would be an important factor. But a court could also look to other factors to ascertain whether a document was still, in fact, in draft format.

Justice Kavanaugh: Okay. Thank you.

Chief Justice Roberts: Justice Barrett.

Justice Barrett: I want to pick up on the thread that Justice Kavanaugh was just exploring with you. You said that if a government official simply stamps "draft" on it and sent it over and, as Justice Kavanaugh is positing, did so in order to avoid FOIA disclosure requirements, you said that a court might look at other factors to determine whether it's still final. What other factors would a court consider?

Matthew Guarnieri: I -- I think a court might look to the -- the -- the treatment of the document within the agency's process. And there are circumstances in which the lower courts have found that the labeling of a document as "draft" might be considered pretextual, if you will, in light of other evidence about the processes that generated the document or the consequences that were attached to the document within the agency's administrative process. But, you know, here, those factors tip decisively in our favor. We have the declarations making clear that the agency decisionmakers didn't reach a final decision. We know that they didn't publicly issue the December 2013 draft, even though a final biological opinion was publicly issued. And we know that they had committed in advance to share a draft with the EPA and they didn't even reach that point because they determined that more work needed to be done with the -- with the draft opinions that were presented to them in December 2013. So those are --

Justice Barrett: I mean, that's --

Matthew Guarnieri: -- all kinds of 1 peripheral considerations that a court might take into account in determining that the label of a document as "draft" is -- is, in fact, accurate.

Justice Barrett: That's a pretty fact-intensive determination then. So it's not your position that we should adopt some sort of bright line saying, listen, it's not over until it's over, it's not until it's actually issued in the sense of being final, maybe even in the Bennett versus Spear sense of the word? You're not asking for a rule that's that bright?

Matthew Guarnieri: Well, we -- we think that those considerations for this particular scheme, the line is very easy to draw because it's so clear that the Endangered Species Act and its implementing regulations set up a process in which the deliberations conclude with the issuance of a final biological opinion. And I think the Court could dispose of this case on that ground alone. But, if there are concerns that, you know, disposing of the case on those grounds might lead to evasions or pretextual -- the -- the pretextual use of the label "draft" in the 1 future, then I -- I was just trying to give the Court some comfort that there are other considerations that could also be brought to bear to -- to make sure that that isn't occurring.

Justice Barrett: So your first order of preference would be the kind of formalistic line that I was just describing, and then your backup argument would be, if the Court was uncomfortable about the possibility of avoiding FOIA obligations by, say, the stamping "draft" on the top, that we go with the more kind of multi-factor fact-specific test, you know, maybe to see was the agency holding this out as a final opinion?

Matthew Guarnieri: Yes, Your Honor. Yes. I -- I think that that captures the -- the -- the way that we think the case ought to be resolved. The -- the deliberative process privilege here ought to extend to all of the deliberations that precede the issuance of the actual final biological opinion.

Justice Barrett: Thank you.

Chief Justice Roberts: Mr. Guarnieri, do you want to take a minute to wrap up?

Matthew Guarnieri: Thank you, Mr. Chief Justice. I -- I would just emphasize again, as I've tried to several times this morning, that -- that there can be no real dispute on the facts of this case, that the decisionmakers at the Services did not make a final decision in December 2013. The -- the record is clear that they did not make a final decision in the ongoing consultation until May of 2014. And when they made that final decision, they -- they released an 85-page joint opinion with -- with several hundred pages of appendices explaining to the public the reasoning that led them to issue a no jeopardy opinion in this particular consultation. Until that point in time, the agency decisionmakers were free to change their mind. And the deliberative process within the agency had not yet come to an end. Thank you.

Chief Justice Roberts: Thank you, counsel. Mr. Narayan. 1 ORAL ARGUMENT OF SANJAY NARAYAN ON BEHALF OF THE RESPONDENT

Sanjay Narayan: Thank you, Mr. Chief Justice, and may it please the Court: I'd like to begin with the standard that follows from Sears and which should resolve this case. If a document explains the decision made by the agency with appreciable legal consequences, Exemption 5 does not apply. The opinions here explain a decision made by the Services, that EPA's proposed regulation jeopardized protected species. And Bennett holds that jeopardy decision has legal consequence. Because the Services' conclusions invariably get deference, the Services' jeopardy determination made it very likely that EPA's regulation would be overturned unless EPA adopted additional protections. None of that turns on whether the Services label their opinions "draft" or "final." Indeed, as the amicus briefs explain, the Services almost never exercise their jeopardy authority through a final jeopardy opinion. 1 The jeopardy decision reached earlier in the process achieved the same legal consequence. It forecloses the agency's proposed action and requires adoption of a more protective alternative. Petitioners have expressed concern as to the workability of the standard insofar as it allows courts to look past labels like "draft" or "non-binding." But the lower courts have been doing just that for 40 years with no unusual difficulty. Those cases, like this one, follow the standard FOIA practice. They're resolved on summary judgment based on the governing statutes and regulations, together with the agency's record and declarations, and, if necessary, examination of the documents themselves. The Endangered Species Act may call the Services' review a consultation, but, in reality, the statute gives the Services decisive gatekeeping authority over other agencies' actions. At stake here is whether the public has access to the reasons underlying the Services' exercise of that statutory authority.

Chief Justice Roberts: Counsel, 1 government decisionmaking often involves several different layers, you know, the issues addressed by the section and then it turns over to the bureau, then it goes to the division, and, eventually, say, to the final decisionmaker. What -- what if that decisionmaker, looking at all this, says, you know, I think -- I think the bureau got it right; I don't think the division did much at all; I like what they did? Does that mean -- in other words: And that's why I'm making the decision I am, because I think the bureau analysis was right. Does that mean that the bureau analysis is disclosable because it is the one that had operative effect?

Sanjay Narayan: If a decisionmaker on behalf of the agency adopts that bureau's view as a basis of the agency's decision, then, yes, I think that is the basis of a decision that is actually adopted.

Chief Justice Roberts: Well, but that -- that's certainly predecisional. I mean, it goes up to the division and then only then to the -- the agency director. I mean, the -- the agency did not adopt the final recommendation 1 from the division but, rather, something that certainly was predecisional.

Sanjay Narayan: I'm sorry. We're talking about the decision that the director chooses as the basis of the agency's decision?

Chief Justice Roberts: Yeah. In its -- in this chain of -- of responsibility that leads up to him, he picks one in the middle. He says, that's what is going to affect -- that -- that's what I'm going with. Is that -- even though it's predecisional in the sense that there were several other layers before it got to him.

Sanjay Narayan: I think it is under Sears because Sears says that if an agency chooses to adopt a document that was predecisional, but it makes it the basis of its actual decision, then, you know, that doesn't raise concern for two reasons. One is that that's the decision of the agency. Any criticism is going to go to the agency, its decision. And the second is that, in general -- again, this is -- this is Sears, not me -- but, when that happens, the lower decisionmaker's not 1 embarrassed. They -- they generally tend to like the fact the agency has taken their decision and the decisionmaker has chosen it as his or her own decision.

Chief Justice Roberts: Well, counsel, your -- the operative effects test seems sort of tailor-made for the facts here, but it doesn't seem to be very helpful in most cases. In most cases, you can't pick a particular item in the decisional process and say this is the one that drove the decision. So how would your effect -- your test work in the typical case?

Sanjay Narayan: If -- if there is no statement of basis there at all -- you know, FOIA doesn't require an agency to write one up, so, you know, one feature here is that the regulations do require the Services to have a statement of basis that is their opinion, their jeopardy opinion, available if the action agency asks for it. Therefore --

Chief Justice Roberts: Thank you, counsel. Justice Thomas.

Justice Thomas: Yes, I'd like to pick 1 up a bit on what the Chief -- on the Chief's line of questioning. How far back would you go in the process? We asked the government how close to the line of actually sending the proposal out or the rule out would he -- would he -- can he come to before it ceases being a draft. I'd like to ask you, how far back in the process can we go before it is not discoverable and it's a part of the deliberative process, as opposed to something that is subject to FOIA?

Sanjay Narayan: Well, I mean, in this case, I think the important thing is that by all indications of the record, this analysis that it's a jeopardy opinion was complete and reached a conclusion. I don't think it goes much further past that. That is, if the analysis is not complete and they haven't -- and they're still working on it, then that is legitimately deliberative and predecisional and is not disclosed.

Justice Thomas: So how do you determine that? The -- the government says as 1 long as they still -- they have not said it's final, it's still a part of the deliberative process. Why don't we take them at their word?

Sanjay Narayan: Well, I mean, I think the reason not to take them at their word is that that then hands control of disclosure to the government based on how they choose to characterize documents. In this case, I really do think it is the record as a whole, and there are four elements in particular that I think deserve attention, keeping in mind that the burden is on the Services here. I mean, the first is that we know the Services had a decision to make. EPA gave them their regulation so they could decide: Does it pass muster under the Endangered Species Act? And the second is that when the time came to make that decision, under the schedule that the agencies agreed to at Joint Appendix 91, the Services conveyed the conclusion that the regulation caused jeopardy and that the next steps were reasonable and prudent alternatives. And we know that there were no further 1 deliberations, either contemplated or that occurred, as to the viability of the proposed regulation. And in EPA's final rule, it said these changes were the result of the Services' consultation in order to avoid jeopardy. I mean, all of that lends no support to the Services' claim that their analysis was somehow not yet done or incomplete or -- or inconclusive.

Justice Thomas: So what's at stake here? EPA's first rule, it doesn't -- it's gone now. They've got a different rule. So what's at stake? Why do you need -- what information are you trying to get about a rule that's no longer in place?

Sanjay Narayan: Well, in this rule, what they've said is that they are going to make permit-by-permit determinations as to what is required to avoid jeopardy and protect the species. And what the Sierra Club's interested is -- in is knowing that those future decisions are consistent with the basis by which they have made these changes so that if, for example, it's turtles at a certain kind of plant, that future 1 permits protect turtle -- turtles at -- at those plants. And industry has a different set of concerns, which is simply knowing that when the agencies exercise this authority, they're doing it on sound grounds.

Justice Thomas: Thank you.

Chief Justice Roberts: Justice Breyer.

Justice Breyer: Well, this is complicated because I think there are several documents here. So I'm going to ask you just to check me. And don't say I'm right if I'm not right, okay?

Sanjay Narayan: Okay.

Justice Breyer: Okay. First, we are talking about the -- the animals, the Fish and Wildlife Department and the Marine Fisheries, okay? What they're supposed to do is they write a document called a biological opinion, is that right?

Sanjay Narayan: Yes.

Justice Breyer: Okay. In the history of this Act, that document, once they publish it, will force EPA to change, basically, and that document has been actually issued never. I 1 mean, let me not exaggerate. In 7,000 cases in seven years, they issued one exactly twice. Now the reason is there is a different document called a draft biological opinion, and what happens when they write that draft is they send it to EPA and they negotiate, and EPA eventually ends up probably doing what they finally agree to do. So we're talking about that draft biological opinion, and it has two things about it: One, we're going to negotiate this with EPA; and, two, private people, who are nothing to do with the EPA, can get a hold under FOIA of that document. Is that right?

Sanjay Narayan: Yes, except I think that what is being negotiated is not the jeopardy conclusion but what happens next; that is --

Justice Breyer: Yeah. All right. What happens next, but there is a document called draft biological thing, and things happen as a result of that, and it's pretty clear that private people can get ahold of it. Indeed, there's a reg to that effect. Is that right?

Sanjay Narayan: That's right if there's an applicant involved in the process.

Justice Breyer: An applicant can get ahold of it. So I thought that that was, once you write that draft biological opinion, you've got something that's final enough that somebody can get it under FOIA. Now the question here is: Well, what about the draft leading up to that draft? And that's what we're trying to get or not get. Is that right?

Sanjay Narayan: No, Your Honor. I think our position is that this document is that draft. I mean, EPA didn't ask for it.

Justice Breyer: Yeah, yeah, yeah, yeah, but it doesn't say it. And so the government says: Well, this is just a draft of the draft, or maybe the government means, no, you can't get ahold of the draft, in which case you can never get ahold of anything because they never use anything beyond the draft. Is that right?

Sanjay Narayan: That's how I understand the government's position. But, to be clear, you know, all they said is that EPA didn't ask for it, so it was never formally transmitted.

Justice Breyer: All right.

Sanjay Narayan: They did send over a portion, and this does appear to be the statement of basis for the jeopardy conclusion, as we --

Justice Breyer: So your view is I looked through the nine -- or the -- the documents that are supposed to be turned over or not, I read this record, and I ask myself: Is this, in effect, the draft biological opinion, or is this a document that is part of the debate within the agency that will lead up to the draft biological opinion that will serve as a basis for the discussion with EPA? Have I got that right?

Sanjay Narayan: Yes, along with the rest of the record, right.

Justice Breyer: Okay. Okay. Now I know what to do. That's extremely helpful. Thank you.

Chief Justice Roberts: Justice Alito.

Justice Alito: Well, I really don't know what to do. We face a conundrum. One possibility is for us to say that, if it's a draft, it's -- it's privileged. The other is to try to draw a 1 distinction between different kinds of drafts. So let me ask you this: Suppose that the Services went through a three-step process in issuing a biological opinion, and the draft produced at Step 1 turns out to be what the Service will ultimately issue 90 percent of the time. What emerges from Step 2 is what it will issue 97 percent of the time. And what emerges from Step 3 is the final product. Now at what point do you think a document would become non-privileged?

Sanjay Narayan: So I think the answer is going to depend on the particular process -- decisionmaking process that actually occurs because, you know, there aren't -- they've given themselves a lot of flexibility. So, in a case like this one, where there was no further deliberation that seemed to have been contemplated or that occurred, then, you know, it is the -- the 90 percent draft because, you know, the 10 percent simply doesn't seem to be at issue in this case. EPA chose not to interrogate their -- their determination here, as agencies normally do.

Justice Alito: Well, you know, with 1 respect, counsel, I don't know how satisfactory that answer is because, if a Service is determined not to have this released under FOIA, all they need to do at every step is simply to say: This is what we think up to this point, but, of course, this isn't our final word, we're open to hearing other information about this so that it's made explicitly non-final, subject to further internal deliberations, until it's finally issued.

Sanjay Narayan: Well, no, I mean, the fact that they could change their mind in their own discretion is clearly not enough. I mean, Sears said that at Footnote 25. Grumman does not stand for that proposition. So, if -- if all they say is, well, in our discretion, we have some room to move, I mean, that's even true about final biological opinions. I think, you know, what I'm saying is that if you have a situation in which, you know, they have a -- a process set up where they say, well, we're giving you a draft and we want your opinion and we intend to respond to it, you know, if a real deliberative process is set up within the rulemaking, then that's a different 1 kind of draft.

Justice Alito: Well, so help me. Could you just say as succinctly and precisely as possible what you think the test is we should apply in distinguishing among opinions that are labeled drafts?

Sanjay Narayan: If the draft opinion reflects a decision made by the agency with appreciable legal consequences, then it needs to be released.

Justice Alito: Well, I mean, if it was -- if Step 1 was followed 50 percent of the time, that would have appreciable legal consequences, wouldn't it?

Sanjay Narayan: I -- I think it would. I just want to be clear that, you know, 50 percent -- I mean, if it is a -- a discretionary reconsideration process and, you know, the Services have all the power to control what happens in that other 50 percent, then, yes, that's --

Justice Alito: All right. Thank you.

Chief Justice Roberts: Justice Sotomayor.

Justice Sotomayor: Counsel, I'm -- I 1 guess I'm getting bogged down in the details, but I do want to a little bit. In this case, what is clear is that all we know is that portions of the draft jeopardy opinion went to the EPA. And I know that -- or at least my law clerk has looked through the record and not been able to find an answer as to what portions. But how can we say that there was a final draft jeopardy opinion that was signed off if the EPA never saw it? And if they never saw it and were working in a collaborative process thereafter to change their rule, how could -- how can it help you to look at that draft when the EPA, in following whatever it's doing now, was never informed by the draft?

Sanjay Narayan: So, in answer to, I think, the first part of your question, if you look at 402.14(g)(5), what it says is that when the Services reach this stage of the process, when they make a jeopardy conclusion as to the regulation they have been given, they need to have essentially a statement of basis available to the action agency if the action agency wants to submit comments. 1 And, you know, here, EPA, again, chose not to submit comments. It didn't want to interrogate the Services' opinion apparently. But, you know, that is still a statement of basis that provided -- that explains the Services' decision. And, you know, that EPA didn't ask for the thing to be formally transmitted, I think, doesn't really affect that -- that document's purpose or the operative effect of the decision itself. As to why we want to see it even if EPA hasn't, the problem is that these jeopardy decisions don't just affect the action agency. You know, ultimately, they affect the public and regulated communities -- the regulated community and everybody else. So, for our purposes, you know, that -- the Services going down the line as they exercise their supervisory -- supervisory authority behave in a way that is consistent with the conclusions they reached here remains important.

Justice Sotomayor: All right. Could you articulate for me your rule again? We know there has to be some collaborative process. 1 There has to be some collaborative process within the Services and one -- and then with the EPA. At what point is that draft biological opinion -- articulate the rule that you want us to write -- final in your judgment or subject to disclosure?

Sanjay Narayan: I mean, when the draft opinion provides the basis for a decision made by the Services, that is, the agency as a whole, that has appreciable legal consequences, then that opinion needs to be disclosed.

Justice Sotomayor: Thank you, counsel.

Chief Justice Roberts: Justice Kagan.

Justice Kagan: Mr. Narayan, assume I agree with you for the moment that what we might call a final draft opinion is FOIA-able, you know, the one that triggers the back and forth between the Service and the agency, but the government here says: Well, this was not such a final draft opinion. It -- you know, it characterizes it as a draft of a draft. And they point to the various decisionmakers' declarations. 1 So what evidence do you have that the government is wrong to say that?

Sanjay Narayan: Well, I think the measure of those sorts of simply conclusory statements, right, like this was deliberative, is the fact that the government has offered to support their conclusion.

Justice Kagan: Well, it's not just conclusory statements. I mean, you have the Service head saying, I -- I -- I -- I thought that more work needed to be done on it, I was not ready to sign off on this.

Sanjay Narayan: That's correct. But the only work that they've described is coming up with an alternative that they would approve, right? So, yes, more work needs to be done. But then, in this particular process, there appears to have been no work either contemplated or that occurred as to the permissibility of the proposed regulation. And the only reason they give for changing their jeopardy conclusion is that EPA agreed to add these additional measures, which, again, EPA itself ascribed to the Services' jeopardy determination through the consultation. 1 So, you know, again, I think it's fair to say that more work needs to be done, but, if the only work that needs to be done is to essentially follow through on the consequences of the jeopardy determination, then I think that's not enough.

Justice Kagan: Well, suppose that -- suppose that the -- the -- the -- the Service head got a memo from a staffer saying this is -- this is a bad idea, there's going to be all kinds of jeopardy, and the Service head really had not -- did -- did not look at it very closely, you know, hadn't decided whether he was ready to sign off on it, but he did realize that there were some issues here, and he calls up the EPA and he says: Look, I -- I haven't gotten all the way through this, I haven't made a final decision yet, but I -- I -- I think that there might be a problem here, and I want to get you to talk to my guys and to try to work this out informally. Would -- would you say that there's a FOIA-able document there?

Sanjay Narayan: No, I wouldn't. I mean, our point here is that the Services haven't made 1 that showing in this record. And the other point is that they are required to have a statement of basis available under 402.14(g)(5) when they reach this stage, that is, when they make a jeopardy conclusion. So --

Justice Kagan: Well, I guess why isn't my hypothetical essentially this case? If we -- if we treat declarations as -- as serious and as worthy of, you know, being -- you know, we should respect them unless we see something to the contrary, that basically the head of the Service looked at this and said: I don't know if I'm ready to sign off on this. I think maybe more work would need to be done to put this in final form. I think that there's probably a -- a -- an issue here. I want to get everybody to start talking about it.

Sanjay Narayan: Well, I think the -- the reason the lower court didn't reach that factual conclusion is that by all indications, this document was ready to go to EPA if EPA asked for it. All that happened is that EPA didn't ask for it.

Justice Kagan: Thank you.

Chief Justice Roberts: Justice Gorsuch.

Justice Gorsuch: Good morning, counsel. I -- I think we all understand the problem of -- of the government effectively stamping everything "draft" and -- and -- and the concerns that attach to that. But I just wanted to explore the concerns on the other side of the coin too, and that is, you know, without adequate room to kind of back down privately, the government sometimes winds up making worse decisions rather than better ones. And, here, it does seem like that because of the back and forth privately, thanks to the Services' intervention, EPA came up with a -- a rule that might be better from your perspective. To what -- what -- how do we balance that concern and allow agencies sufficient room to maneuver privately to avoid having, you know, to embarrass themselves later and allow them to save face to get to better policy results?

Sanjay Narayan: You know, if an agency puts those facts into the record, then I think, 1 fair enough, but the key point is that if the agency is going to back down, it has to back down from its decision, right? It can't make the decision and then say, well, we're not --

Justice Gorsuch: I guess I'm asking you to abstract up a level with me, counsel, and say: You know, I think you'd agree here that EPA got to a better result thanks to the Services' informal interventions, right?

Sanjay Narayan: Yes, we'd agree.

Justice Gorsuch: Okay. And -- and -- and so there's got to be some room for that kind of private negotiation, don't you think?

Sanjay Narayan: Yes. To be clear, we're not complaining about the Services' making EPA make its regulation more protective. And an important fact here is that the Services really do have the authority.

Justice Gorsuch: Well, I guess I'm more -- I'm asking don't -- do you -- are you at all concerned that a more invasive rule might deter this kind of productive back-and-forth discussion? And how do we -- how do we balance that concern?

Sanjay Narayan: No, Your Honor. I mean, 1 we're not concerned, I think, for a couple reasons. One is that the biological opinion really is a mostly science -- scientific studies and facts, you know, so it's not the sort of thing that lends itself to the sorts of embarrassment. Those things are normally subject to peer review, right? So -- and then, you know, in this sort of back and forth, I mean, what's important is that you have a position -- you have one party who has authority, effective authority, and one who's acting like a subordinate, and that's the action agency. And in that circumstance, I think, you know, yes, there is some balancing, but it is really important to know why the Services are saying what they're saying, at least when they effectively foreclose a -- a proposed regulation, that from EPA's purposes, that they had said: We've -- we've reached the end of the line for the Endangered Species Act for us.

Justice Gorsuch: Thank you, counsel.

Sanjay Narayan: All we need to know from you is, you know, are we good?

Chief Justice Roberts: Justice 1 Kavanaugh.

Justice Kavanaugh: Thank you Mr. Chief Justice. And good morning, Mr. Narayan. I wanted to pick up first on Justice Thomas and Breyer and Alito and -- and others have talked about. What is the agency decision? I would -- I would have thought that the way to do this is to start by figuring out what is the decision, capital D decision, and then everything that led up to that decision is -- is predecisional. And that would be a pretty simple formula. And, obviously, there would be case -- questions in some cases about what the decision is. Here, theoretically, you could argue EPA's decision, but the government acknowledges that the Services' opinions are the decision. What's wrong with that framework?

Sanjay Narayan: I don't think anything is wrong with it so long as we recognize that when EPA gives a proposed regulation to the Services and asks does it pass muster under the Endangered Species Act, that is a capital D decision.

Justice Kavanaugh: And the second question I wanted to ask is the need for clear rules in the FOIA context. So the -- the need for that, I think, is multi-pronged. First, the agency officials who are engaged in deliberations need to be able to speak with candor, as Justice Gorsuch was just saying. Second, FOIA officers, who are rampant throughout the executive branch, spend an enormous amount of time and resources on FOIA, could use clarity. And then district court judges in the District of Columbia and elsewhere, if you talk to, would lament the lack of clarity and clear rules in -- in FOIA cases. So that raises the concern that the effects-based test or looking at the effects of the memo could become so fact-intensive and could really blur the longstanding predecisional principle of the deliberative process privilege because lots of drafts have -- have real effects within the executive branch. Can you respond to all that?

Sanjay Narayan: Yes. I mean, so I'll 1 start with the agencies because it -- I think it is important here that the Services had themselves developed workable standards via the memos we cited at page 55 of our draft by which they do distinguish between drafts that have decisional weight and those drafts that really are deliberative. So there's no suggestion or at least there's no evidence that there's any lack of clarity that has -- is -- is impeding their activities in this context. As to the district courts, you know, the district courts have not -- I mean, yes, they are perhaps fact-intensive judgments but really no more so than those required under other elements of the APA. And it is always possible for an agency to submit declarations that don't speak to the standards and then say: Well, these standards are unworkable. For an agency to put in facts like, you know, in fact, there are elements of the biological analysis that we -- with which we did not agree or -- you know, all of those things are -- are in their possession. And the reason FOIA places the burden on the agency is because 1 they're the only ones that have it, right? So, again, in general, these cases have been resolved in really typical FOIA fashion. I mean, you look at the regulations and the statute, you look at their declarations in the record, and if all of that isn't clear, then there's the option of in camera review.

Justice Kavanaugh: Okay. Thank you. That's helpful.

Chief Justice Roberts: Justice Barrett.

Justice Barrett: Counsel, I have a question following up on Justice Breyer when he gave you the hierarchy of documents that might be at stake here. I want to be sure that I understand the consequences that flow from each. So, you know, in your conversation with Justice Breyer, you identified the biological opinion, which is almost never issued; the draft biological opinion, which is available by regulation; and then the draft of a draft, which the government says this was. Is it true that the draft biological opinion, that second one in the hierarchy, is always FOIA-able and that there's no controversy 1 about that?

Sanjay Narayan: Oh, no, no, there -- there is controversy about that.

Justice Barrett: Okay.

Sanjay Narayan: I mean, it is -- it is available if there's an applicant involved because, in that case, it's -- it's not a purely interagency document, and so, under Klamath, those are -- are produced.

Justice Barrett: Okay.

Sanjay Narayan: I mean, one of the controversies here is if -- is that kind of draft document available or not?

Justice Barrett: Okay. Thank you. I wanted to clarify that. My next question has to do with the -- what you characterized as the legal effect of this document. Why was it a legal effect as opposed to simply a practical effect when having this document caused the EPA to abandon the 2013 regulation and then move on to the 2014 regulation?

Sanjay Narayan: Well, I think Bennett really suggests three reasons why, under the Endangered Species Act, there are legal 1 consequences and not just practical influence. I mean, the first is that there's a statutory prohibition in Section 7, right, no actions are allowed that create jeopardy. The second is that the -- the -- the Services have a mandatory statutory role in enforcing that prohibition. So it's not a matter of -- of just asking for advice from somebody. They have to be involved.

Justice Barrett: But I'm sorry, counsel, let me just interrupt for one second. I think that's true if you have a final biological opinion. But, in this case, would you say that if EPA simply got -- I'm sorry -- if EPA simply got wind of what the Services were thinking and said: Oh, well, it doesn't look like this is going to be on a -- a good track for us with respect to jeopardy, and so abandoned it, that seems to be a practical consequence, and that might be the same kind of consequence that flows from a draft opinion, as opposed to a biological opinion, which does have force in the scheme.

Sanjay Narayan: No, I -- I agree with you that in that scenario, it -- it would be 1 privileged. I think the point here is that the Services reached a conclusion, conveyed the conclusion to EPA, and EPA responded exactly as the Services expect them to respond. When they say jeopardy, EPA then moves to change its regulation.

Justice Barrett: But how can a draft opinion give rise to that legal consequence?

Sanjay Narayan: Well, what Bennett says is that what's important is that when the Services reach a conclusion, the action agency knows that that conclusion is based on an administrative record that is going to get deference. So -- so, as long as EPA knows the Services have reached this conclusion, none of what Bennett describes really depends upon whether the -- the analysis in the record is -- is currently labeled "draft" or "final." What they need to know is the analysis is there and they really are going to have no chance of contesting it in court.

Justice Barrett: Thank you, counsel.

Chief Justice Roberts: A minute to wrap up, counsel.

Sanjay Narayan: The problem with the Services' standard is that it boils down to it's privileged if we say it's privileged. And the upshot here would be to deprive the public of access to the reasons underlying the Services' jeopardy decision, which, again, virtually never appear in final opinions. Those decisions are enormously consequential not just to the action agency or to the Sierra Club but to the regulated parties who ultimately have to comply with the measures the Services demand. The importance of looking to those legal consequences, rather than just labels, is that it tracks FOIA's core concern, making sure the public knows how agencies are actually using the authority Congress gave them. Thank you very much.

Chief Justice Roberts: Thank you, counsel. Mr. Guarnieri, three minutes for rebuttal. REBUTTAL ARGUMENT OF MATTHEW GUARNIERI ON BEHALF OF THE PETITIONERS

Matthew Guarnieri: Thank you, Mr. Chief 1 Justice. In our view, the general rule here should be a clear bright-line test that, until a biological opinion is signed and formally issued, there is no final decision in an interagency consultation. The deliberative process privilege really requires that degree of certainty. Any exception should -- should be rare. In -- in -- in that respect, it's really no different -- the biological opinion here is really no different than a judge's or a court's opinion, which is not actually final until it's adopted by the judge and -- and issued as an official opinion. Now, turning to Respondent's alternative, first, Respondent really this morning has made no effort to defend the actual reasoning of the court of appeals. Respondent does not defend the kind of last version rationale that the court of appeals employed to conclude that the December 2013 draft biological opinions were -- were not predecisional and deliberative. Respondent's alternative instead is 1 this appreciable legal consequences test, which I don't think works for a number of reasons. First of all, as -- as -- as Justice Barrett's questioning illustrated, there -- there is no appreciable legal consequence to a draft biological opinion. Legal consequences attach only to the final biological opinion. Second, as Justice Kagan's hypothetical illustrated, the appreciable legal consequences test that Respondent has proposed here really proves too much because the same -- the same consequences that Respondent is relying on here could have flowed from, for example, an informal recommendation made by a subordinate staff member at one of the Services. You know, we use -- we use this example in our reply brief that, if a junior staffer at one of the Services had sent an e-mail to the EPA at the very outset of the consultation saying: Well, in my view, my supervisors might make a jeopardy determination here unless you change the following things about your proposed rule, no one would confuse that with a final decision. Respondent's -- Respondent's only real 1 answer to those problems is to insist that in this particular consultation, the Services, in fact, made a decision in December 2013. And that contention is just at odds with the record of the proceedings here. It's clear that the agency decisionmakers at the Service never adopted the December 2013 draft as an official agency position. They never signed them. They never publicly issued those documents. And they never even transmitted them in full to the EPA. What Respondent is seeking to obtain here is not an explanation of the decision the agencies actually made. As Respondent confirmed, Respondent would write these drafts in order to impeach the decision that the agencies made in future challenges to the application of the EPA's rule to particular permits. So Respondent is seeking to mount a sort of collateral attack against the agency's decision, not to understand the basis for that decision. The Court should reject that effort. Thank you.

Chief Justice Roberts: Thank you, 1 counsel. The case is submitted. (Whereupon, at 11:10 a.m., the case was submitted.)