The Discovery Objections Series · Part 4 of 5 · Overview · July 2026
The privacy objection has one real interest behind it. Give that interest everything it asks for, and the objection collapses into what it was protecting all along — the insurer’s own ledger.
The first three parts of this series answered the insurer’s objections by contesting them. Relevance and its fishing costume (Part 1), overbreadth and proportionality (Part 2), burden in its real and fictional faces (Part 3) — each was a fight the claimant won by pushing back. Part 4 is different, because the privacy objection is different. It is the only objection in the catalogue asserted on behalf of someone other than the insurer: the nonparty insureds whose claim files, medical histories, and identifying information sit inside the discovery the bias case requires. Those people are real. Their interests are real. And that is precisely why the winning move here is not to fight the objection but to concede it — to hand the privacy interest everything it can legitimately ask for, and let the concession dissolve the objection down to its residue.
Because when the concessions are made, what remains is not a privacy interest at all. What remains is an insurer using nonparties’ moral capital to shield its own payment ledgers, its retention data, its expert-selection protocols. Strip the sympathetic facts away by giving them away, and the objection has nothing left to stand on.
The privacy objection is beaten by the requesting party who protects privacy more carefully than the insurer does.
The instrument is a set of concessions, offered affirmatively and early — before the motion rather than in response to it. Four moves, each of which surrenders something the claimant never needed and gains something the claimant cannot otherwise get.
| One: stipulate the protective order | Propose it first, in a form generous to the nonparty, and load it with the two terms that do most of the work. The first governs the public record: produced materials are used only in the litigation, and anything filed is filed with personal identifiers redacted rather than under seal — no sealing motions, no shadow docket, just the identifying information blacked out on the public page. The second governs contact: neither party reaches out to any nonparty insured whose information appears in the production, unless that identity surfaces independently or a later order authorizes it. That anti-contact promise answers the oldest privacy anxiety in this area — the fear that discovery is a solicitation drive in disguise. Concede it away, and the fear has no target. |
|---|---|
| Two: concede unredacted production | The instinct is to offer redaction as the privacy answer. But mass redaction of claim files is exactly where the cost and burden of production live — thousands of pages, every name reviewed by hand — and it is the burden the insurer then invokes to resist the whole request. The protective order is what protects the data; redaction is what makes producing it expensive. So the concession runs the other way: produce the files unredacted, under the order, with the order doing the protecting. Reserve redaction for the genuinely sensitive core — discrete medical conditions, an individual’s personal income — where the marginal privacy interest justifies the marginal cost. For everything else, conceding unredacted production removes the insurer’s best burden argument at the same moment it removes its best privacy argument. |
| Three: concede that you don’t need the names | The pattern proof this framework runs on is aggregate — counts, rates, totals, dispositions, the contents of files with identities protected. It does not require knowing who any other insured is. Say so, in the requests themselves: outside the class-action setting, there is generally no need to obtain personal identifying information at the outset of discovery. Identity discovery has its place, but its place is targeted and late — sought only if a specific, recognized ground crystallizes, which in practice it rarely does. The front-loaded demand carries a hidden cost: courts read a sweep for thousands of names less as discovery than as a settlement lever. Conceding that you neither need nor seek the names removes that suspicion and the objection’s only sympathetic fact together. |
| Four: hold privilege to its narrow empire | Privilege is the one objection in this series that is absolute — no balancing, no machinery. If the material is privileged, it is gone. Which is exactly why its boundaries matter, and why the concession here is a demand rather than a surrender: concede that genuinely privileged legal advice is off the table, and insist in return that the empire is small. Business records are not legal communications — a compensation ledger does not become privileged because a lawyer later found it awkward. Privilege is claimed document by document, on a log the requesting party can test; the blanket assertion without a log is how privilege is forfeited, not preserved. And in ERISA cases the fiduciary exception inverts the analysis: an insurer administering a plan acts for the beneficiaries, and its pre-decisional communications about the claim — including the selection, compensation, and evaluation of the experts used to decide it — belong presumptively to the very claimant they were generated to judge. |
The logic is the same one that runs under the whole series, turned inside out. Sort the requests and two piles appear. One seeks materials with real nonparties behind them — other insureds’ claim files, the reputational-metrics discovery. The other seeks the insurer’s own institutional information — its payment ledgers, its 1099 issuance, its selection and supervision protocols — where no third party stands behind the objection at all. The concession package is built for the first pile, and it is devastating to the second. Once the claimant has offered a generous protective order, promised no contact, waived the names, and accepted redaction for the sensitive core, the privacy interest in the first pile is fully accommodated — and the insurer’s continued resistance can only be about the second pile, where there was never a person to protect. The concessions do not weaken the case. They isolate the insurer’s real objection and expose it: a corporation, holding a ledger, standing where a person is supposed to be.
There is a doctrinal reason the concessions land, not just a rhetorical one. The governing principle in this territory is that privacy is conditional, not categorical — a procedure to be satisfied, not a wall. The foundational authority, a California Supreme Court decision nearly fifty years old, held that confidential third-party records may be disclosed once the affected individuals receive notice and an opportunity to object, with the court free to fashion protections tailored to the interests at stake. The insurance privacy statute insurers cite most contains its own exceptions — a facially valid judicial order or subpoena is itself compliance — so that a court order compelling production, under a protective order, satisfies the statute by its own terms. And the reported cases converge on a single corollary: an unsupported privacy objection fails against available protective machinery. The insurer that will not accept a generous protective order must explain, with particularized evidence, what specific harm the order would not cure — and in the reported fights, it almost never can.
Every part of this series names a tell — a move to watch for in the insurer’s brief. Part 4’s is the misstatement of authority by omission: the privacy statute quoted for its prohibition and stopped there, the exceptions left off the page. A statute cited without its exceptions is not authority; it is scenery. And the omission is not confined to insurers’ briefs — this part’s case study is built around a court order that quoted the statute “in relevant part” and left out the very exception that decides contested motions. The claimant who arrives with the concession package also arrives with the whole statute, exceptions included, and hands the court the text the insurer hoped it would not read.
Concede what the objection legitimately protects, and you keep what the case actually needs. Offer the protective order first, and load it with the terms that answer the real fears — identifiers redacted in public filings rather than sealed, and no contact with insureds absent independent discovery or a later order. Take the documents unredacted under that order, because redaction is the cost and the order is the protection; save redaction for the sensitive core alone. Waive the names you never needed, and let the insurer explain why it is still fighting. Quote the whole statute. And hold privilege to logs, not blankets. What survives all of that is protecting someone real — and the concession package was built to protect them and produce the evidence, which has been the point since the privacy fight began.
Free edition · The argument
Why the privacy objection is the only one asserted for someone other than the insurer, why that makes concession the winning move, and how the four-part package — protective order, unredacted production, waived names, narrowed privilege — isolates the corporation standing where a person is supposed to be.
Paid edition · The full framework
The conditional-privacy analysis at all three levels, the statute’s exceptions with every authority, the two protective-order terms drafted for filing, the unredacted-production concession with its burden logic, the never-request discipline with the class-action carve-out, and the privilege module with the fiduciary-exception architecture.
See the Part 4 case study → Get the checklist → The bias-evaluation service →
Subscribe to follow the series →
The burden objections this part follows are in The Two Faces of Burden — and the Trial That Never Happens; the pattern evidence these concessions protect is mapped in The Files the Insurer Hopes You Won’t Ask For; and the discipline for the one discovery you should almost never ask for is in The One Discovery You Should Almost Never Ask For.
This page summarizes Part 4 of the Discovery Objections Series as published on Expert Bias Report. Holdings derive from the project’s primary reading of the conditional-privacy line, California Insurance Code section 791.13 and its exceptions, and Dobro v. Allstate Ins. Co., No. 16cv1197-AJB (BLM), 2016 WL 4595149 (S.D. Cal. Sept. 2, 2016). Source texts in sources_md_conversions/; Part 4 doctrinal synthesis in the project vault. Educational and informational only; not legal advice.