# CROSS-EXAMINATION — RETAIL CUSTOMER REVIEWER (Stage 2) Reviewer role: consumer-protection analyst / homeowner perspective (adversarial). Stage 1 position: KILL (reviews/retail-customer.md). That verdict stands; nothing in the other eight reports weakens it, and several reports (tax, estate, title, red-team) strengthen it. This file answers the two Stage 2 questions and flags every direct inter-reviewer conflict I can see, per the runbook. Inputs: all nine Stage 1 reports in reviews/; spec at disclosures/equity-flow-invention-disclosure.txt (baseline from Stage 1). Claim tags: SETTLED LAW / LIKELY INTERPRETATION / STATE-DEPENDENT / AMBIGUOUS / SPECULATIVE. -------------------------------------------------------------------------------- ## PART 1 — FINDINGS I DISPUTE I dispute four findings. I do not manufacture disputes: the overwhelming bulk of the panel's findings corroborate mine (the tax report's SAM analysis, the estate report's discharge analysis, the title report's Olson/Roberts recharacterization line, the operator's pause-miscalibration finding 1.4, and the securities report's 6.A marketing table are all consistent with, and stronger than, my Stage 1 COMPREHENSION/TRUST/PREDATORY findings). The disputes below are places where I believe a reviewer is wrong, overstated, or has misread the record — plus two honest self-corrections at the end, because cross-examination runs both ways. ### D1. DISPUTED: securities-consumer-finance, Verdict section 8.1 — ### "the 1.0x no-premium structure genuinely improves the equity ### characterization versus incumbents" and the premise that a marketing ### rewrite salvages the consumer side. (Wrong on both grounds.) - What sec-cf says: the core contract is fix-before-launch partly because (i) incumbents operate "under materially similar characterization risk," (ii) the 1.0x no-premium design "genuinely improves the equity characterization," and (iii) the consumer defects are curable by rewriting both website sections from the disclosure. - Why I dispute (i) and (ii): the tax report (D-1, A-2) and the title report (front matter, citing the CFPB Roberts amicus loss-sharing test) both conclude the opposite on the sponsor's own facts — the 1.0x senior capital return is the single most debt-like feature in the structure, and Equity Flow fails the Reg Z investment-plan loss-sharing test MORE cleanly than the lump-sum incumbents, because the investor never shares depreciation (pause + 1.0x + Shield floor). The characterization risk is not "materially similar" to incumbents; it is strictly worse (tax A-2: the two facts incumbents' opinions rest on — single lump sum, genuine downside — are the two facts this product deleted). Sec-cf's own section 3.1 lists the 1.0x floor as the FIRST factor cutting toward "loan"; its verdict paragraph then credits the same feature as an equity improvement. The report contradicts itself, and its verdict rests on the wrong half. Tag on the underlying characterization point: LIKELY INTERPRETATION (adopting tax D-1 / title 1.2). - Why I dispute (iii), from my own domain: a disclosure rewrite cannot cure this product, because the customer base exists only under misunderstanding. My Stage 1 VALUE section shows that with honest numbers (APR-equivalents of 8-35% in normal markets, exit obligations of 1.4x-6.2x cash received), every HELOC-eligible customer exits the funnel and the residual base is definitionally the population that cannot access alternatives. A page that truthfully discloses the pause trigger, the ratchet, the gross-basis fee, the occupancy covenant, and dollarized exit costs is not a fixed product; it is a confession. "Fix the marketing" assumes the product and the marketing are separable. On this record they are not — the marketing IS the product's distribution mechanism (spec's own investor pitch: the framing "doesn't feel like borrowing or selling"). Sec-cf's kill findings on the pages are right; its inference that the core survives the pages' correction is not. ### D2. DISPUTED: insurance-actuarial, Finding 5.1 (parenthetical) — the ### retail-facing ratchet framing "increases modestly to compensate" is ### "accurate about modestly." (Wrong; misreads the shipped artifact and ### uses the wrong denominator for a consumer.) - Two problems. First, factually: there is no retail-facing ratchet framing to be "accurate." The ratchet appears nowhere in the homeowner-facing section (my C3; sec-cf 6.A4 confirms). Any "increases modestly" language lives in the spec/investor materials, i.e., in documents the homeowner never sees. Blessing a consumer disclosure that does not exist creates a false impression, in the Stage 5 synthesis, that the consumer-side ratchet problem is a tone problem. - Second, substantively: "modest" is true only per-year in investor present-value terms at a 10% discount rate — the actuary's denominator. In the homeowner's denominator (nominal dollars owed at exit, the only number a household experiences), the ratchet moves the investor's take from 25% to as much as 50% of an appreciation pool that compounds for decades. My Stage 1 table: at a 25-year hold and 5% appreciation, the ratchet-inclusive settlement is $767K against $123K received; the actuary's own Finding 5.1 table shows the year-30 payoff at $554K versus roughly $290K flat-25% equivalent. A term that roughly doubles the largest payment of a consumer's life is not "modest" in any consumer-protection sense, whatever it does to a discounted buyer's PV. The actuary's real finding — the ratchet UNDERCOMPENSATES the investor while still costing the homeowner hundreds of thousands — makes the term worse, not better: it is a lose-lose mechanism whose only function (actuarial's word) is narrative. I ask Stage 5 to carry BOTH halves: expensive to the homeowner in nominal exit dollars AND inadequate to the investor in PV. Those are simultaneously true and jointly damning. ### D3. DISPUTED (partial): estate-bankruptcy, Finding 1.3 — "the ratchet ### monetizes probate delay" / "the investor is economically rewarded by ### probate delay and by objecting to the estate." (Overstated as an ### incentive claim; the harm finding survives on different reasoning.) - The estate report's harm conclusion (heirs' share shrinks during a period when they cannot lawfully convey; unconscionability exposure; the spec never tolls the ratchet at death) is correct and I adopted it in Stage 1 (P7). - But the incentive mechanism is wrong in PV terms. Actuarial 4.2/5.1 demonstrates that even WITH the ratchet, each additional year of hold destroys roughly 8-10% of position present value — +1 point per year of share is worth far less than a year of discounting and reinvestment. A rational holder is not rewarded by probate delay; a rational holder wants the estate OUT as fast as possible, which is exactly the red team's M2 (post-assignment covenant weaponization). This matters for consumers because a fix targeted at the wrong incentive fails: tolling the ratchet at death (estate fix 3) cures the optics of accrual-during-delay but does nothing about the real danger to heirs, which is an assignee accelerating settlement via the 12-month clock, the probate petition, and occupancy-covenant defaults. If Stage 5 adopts estate fix 3 (I support it) it must ALSO adopt red-team M2's assignee-binding remedy limitations, or the heir-protection problem remains fully intact. Tag on the PV-incentive point: not a legal claim; arithmetic per the actuarial model. ### D4. DISPUTED (partial): red-team, Finding F3 — enrollment-value inflation ### "pays the homeowner twice." (Overstates the homeowner's benefit; misses ### that the same mechanism injures honest homeowners more than it enriches ### colluding ones.) - F3 is right that the enrollment appraisal sets both the payment size and the appreciation baseline, and that origination controls are absent. But the claim that inflation pays the colluding homeowner "twice" ignores the pause mechanics that the operator (1.4) and I (T3/T9) established: the pause triggers the moment an AVM prints below ENROLLMENT value, paused months are never made up, and paused months do not reduce the investor's share. An inflated enrollment value therefore sets a hair-trigger on the homeowner's own income: the bigger the inflation, the more of the "bigger checks" get permanently confiscated by pause, in any market that does not immediately outrun the inflation. The scheme is partially self-punishing, which caps its attractiveness below what F3 implies. - The consumer-protection corollary the red team missed: the bigger victim of enrollment-value aggressiveness is the HONEST homeowner. An investor-selected, merely optimistic origination appraisal (no collusion required) puts every honest enrollee at the pause boundary from day one — my T3 harm. So the control F3 prescribes (origination AVM cross-checks, rotation, retro QC) protects the fund; the symmetric control — an enrollment-value haircut or pause buffer protecting the homeowner from the investor's own appraisal — protects the customer, and neither exists. Red-team F12/M11 gestures at this asymmetry; F3 as written points the fix only at the fund's loss. Both directions must appear in Stage 5. ### Findings I considered disputing and do not: - Operator 1.1 (origination loses money): outside my domain; arithmetic looks sound; no dispute — but see FX2 below on its proposed fix. - Investor B6 / sec-cf 3.1 (homeowner implied cost ~11-12% in the base case): consistent with my independent model (12.8% at 5% HPA, year-7); no dispute. - Actuarial verdict that the core contract's escrow recovery is "genuinely robust in sale exits even at -45%": not actually in tension with my T4 — the same waterfall fact viewed from opposite sides. The investor recovers BECAUSE the homeowner absorbs the entire decline first. Both statements are true; Stage 5 should present them as one fact, not two. ### Self-corrections (concessions extracted by the panel): - SC1. My T4 worked example (-20% at year 3, $54K capital return) inherits the spec Section 20 inconsistency that investor A10 exposed: on a monotonically declining path the quarterly pause should have stopped disbursements within months, so both the spec's downside example and my T4 arithmetic overstate capital deployed (and thus the exit repayment) in down markets. My finding's direction is unchanged and arguably worsens: the homeowner in a down market loses the income EARLY (pause) and still absorbs 100% of the decline — the harm shifts from "repay a large advance" to "the promised paycheck stopped in year one and the home fell anyway." But the numbers in T4 should be re-run in Stage 4 with pause-consistent deployment. Sec-cf 6.A2, which also uses spec Section 20 at face value, has the same inherited defect. - SC2. My P4 ("the class waiver removes the only economically viable remedy") is overstated for California: red-team M5 is right that McGill v. Citibank makes public injunctive relief non-waivable (SETTLED LAW in CA; federal survival per Blair — LIKELY INTERPRETATION). P4 stands for the other 49 states and for damages claims everywhere; I adopt M5 as a refinement. -------------------------------------------------------------------------------- ## PART 2 — PROPOSED FIXES FROM OTHER REVIEWERS THAT CREATE NEW PROBLEMS ## IN MY DOMAIN (consumer protection / homeowner experience) ### FX1. Title fix 1 / Estate fix 1: "Pick the lien" — record a true ### performance deed of trust or mortgage. (Creates the largest single new ### consumer harm proposed by any reviewer.) Both reviewers are right that the current "equitable interest, not a lien" posture is self-destructive FOR THE INVESTOR. But observe what the fix does to the homeowner: - (a) It arms every downstream assignee with a recognized security instrument in the family home. Combined with Section 35 free assignability and red-team M2 (holders whose PV decays 8-10% per year of homeowner tenure), the fix converts "you're never forced to sell" from a contested contract question into a foreclosable instrument — in deed-of-trust states, potentially a NONJUDICIAL power-of-sale remedy in the hands of whichever yield buyer holds the paper in year 18. Mechanics of nonjudicial foreclosure under a deed of trust: SETTLED LAW; that a recorded performance DOT here would carry or acquire such remedies absent express waiver: LIKELY INTERPRETATION and drafting-dependent. - (b) It deletes the homeowner's one accidental protection. Estate 4.1/4.4 establishes (SETTLED LAW) that as currently drafted the interest is a dischargeable claim and the debtor keeps the home and future appreciation; with a true lien, the interest rides through discharge (Long v. Bullard; Caulkett) and defeats the homestead exemption (522(c)(2)). The estate reviewer notes the irony; I am stating the consumer consequence: the fix transfers the entire bankruptcy downside from the fund to the household — the exact population (equity-rich, cash-poor) the funnel selects for financial stress. - (c) It creates a worst-of-both-worlds interim: lien-grade enforcement with no loan-grade protections. If the sponsor records a mortgage-shaped instrument while continuing to litigate "not credit," the homeowner gets foreclosure exposure without TILA disclosures, rescission, servicing rules, loss-mitigation rules, or the 1639c(e) arbitration ban — unless and until a regulator wins the recharacterization fight on the homeowner's behalf. - Condition for acceptability in my domain: the fix is tolerable ONLY as a package — recorded instrument containing an express, assignee-binding waiver of foreclosure/power-of-sale (title 4.4's covenant-running fix), voluntary adoption of a full Reg Z-equivalent disclosure stack, independent HECM-style counseling, and the estate reviewer's homeowner-side credit protections. A bare "record the lien" instruction in Stage 5 would make the product MORE dangerous to homeowners than the defective original. ### FX2. Operator Finding 1.1 fix: reprice the origination fee (percent of ### home value, like incumbents) and/or raise the eligibility floor. - Repricing to 2-3% of home value multiplies the consumer fee roughly 4-6x ($3,150 to $12,500-$18,750 on the $625K base home) — a fee that is currently DISCLOSED NOWHERE on the homeowner page (my T6; sec-cf 6.A6) and repaid on a gross basis. Enlarging an undisclosed charge deepens the hidden-charge pattern, and if the product is recharacterized as a dwelling-secured loan, a larger fee walks directly into HOEPA points-and-fees triggers and state fee caps (sec-cf 3.2, 4 — SETTLED LAW contingent on recharacterization; my application: LIKELY INTERPRETATION). - The alternative (raise the eligibility floor to exclude modest homes) is a wealth screen: it excludes precisely the moderate-value homeowners for whom the product's cash flow is most material, and a value-floor screen layered on the "demographic factors" lock-in score aggravates the fair-lending/FHA 3605 exposure sec-cf 3.4 and ops 2.3 identified (STATE-DEPENDENT / LIKELY INTERPRETATION per those reports). - Acceptable only with: full dollar disclosure of the fee in the offer flow, capital-return basis = net cash actually received (ops 1.2's fix — which I endorse without reservation), and an APR-equivalent that includes the fee. ### FX3. Operator Finding 5.2 fix: a contractual "regulatory-event" clause ### (force-majeure-style suspension with tolling) in the homeowner agreement. - This installs a SECOND investor-invocable off-switch on the homeowner's income, alongside the pause. My T9/P2 and sec-cf 3.3 established that the first off-switch is the product's most abusive mechanic; the fix adds one triggered by an event (regulator action) that the homeowner cannot observe, predict, or dispute, and that correlates with exactly the states trying to protect them. It also guts the Section 31 forfeiture remedy — the homeowner's only counterparty-credit protection — in the very scenario (reclassification amid fund stress) where red-team F1/M1 shows that protection is already disarmable. The operator half-admits the disclosure problem; I am making it a hard objection. - Consumer-coherent alternative: regulatory events should trigger HOMEOWNER options, not investor suspension — homeowner-elective termination with appreciation-share forfeiture, or conversion of the undisbursed commitment to an escrowed/LC-backed lump payout (pairs with estate fix 6). If Stage 5 adopts a regulatory-event clause in the investor's favor, the marketing can no longer say "binding obligation" anywhere, and the pitch collapses honestly. ### FX4. Investor verdict fix path: "a settlement-dated instrument (defined ### maturity)"; actuarial 3.2's rated-wrapper prerequisite (rated notes need ### a legal final / settlement deadline). - A maturity date is a forced settlement date. It converts the product's one genuinely pro-consumer structural virtue — no forced exit — into a balloon event: sell, refinance at market rates (destroying the rate lock the product exists to preserve), or face enforcement, on a date fixed decades earlier. For the aging-in-place holder the maturity date is a scheduled displacement event; for heirs it stacks on the 12-month probate clock. It also makes the product functionally indistinguishable from a term reverse mortgage under the states' functional definitions sec-cf flagged (STATE-DEPENDENT), while still carrying none of HECM's counseling, age-gating, or spousal-deferral protections. If duration is unfundable without a maturity (the investor and actuarial reports jointly suggest it is), that is evidence the product shouldn't exist — not license to bolt a balloon onto retirees. ### FX5. Investor verdict fix ("a priced spread or premium restoring ### loss-absorption") and actuarial fix (iv) ("a small fixed-dollar accrual ### senior in the waterfall"). - Any time-based accrual charged to the homeowner is interest by function (tax D-3 — LIKELY INTERPRETATION) and a growing balance by arithmetic — the one thing the consumer marketing swears cannot exist, and the negative- amortization optics that made reverse mortgages a complaint category. It destroys the only genuinely cheap cells in my Stage 1 cost table (flat-market exits at ~0-4% APR-equivalent), making the product worse than a HELOC in ALL states of the world rather than most. At that point no honestly-informed customer exists at any literacy level. Both proposing reviewers concede the accrual re-detonates debt characterization; from my domain it also detonates the last shred of the value proposition. This fix is a kill confirmation wearing a fix costume. ### FX6. Sec-cf gating item (i): launch footprint excluding CT, MD, WA ### (title adds IL, ME; kill-equivalent TX). - As compliance strategy this is coherent. As consumer protection it is regulatory arbitrage: the product would be sold ONLY where legislatures have not yet acted — i.e., deliberately concentrated on the least-protected consumers, while the trend line (title 1.2: CT, MD, IL, ME enacted; WA, MA, PA moving) guarantees mid-book reclassification somewhere. Ops 5.2 shows what mid-book reclassification does to customers: their income lands on a three-missed-payment fuse. The exclusion strategy doesn't remove that risk; it schedules it, and shifts it onto homeowners in the "allowed" states who were told they bought a stable 7-to-30-year arrangement. Any Stage 5 adoption of a footprint strategy must pair it with the FX3 homeowner-elective protections and disclosure that the product's legality is contested in the customer's own state (which, again, is a confession no funnel survives). ### FX7. Sec-cf gating item (vii): arbitration clause "redrafted with ### severability for 1639c(e)/MLA contingencies." - A better-drafted arbitration clause is a worse consumer outcome. The current clause is defective in the consumer's favor: void outright on recharacterization (1639c(e) — SETTLED LAW contingent), holed in California (McGill — SETTLED LAW in CA), and unconscionability-vulnerable (sec-cf 3.5). "Fixing" it means engineering the class waiver to survive those attacks — entrenching my P4 harm (no economically viable remedy for systematic, small-dollar-per-customer failure modes like wrongful pauses and miscalculated settlements) more durably. The consumer-protective redraft is deletion of the class waiver and the punitive-damages bar, not their fortification. Stage 5 should not score clause-hardening as a consumer fix; it is a defense fix billed to the customer. ### FX8. Sec-cf gating item (viii): "fee and covenant disclosure added to all ### consumer materials" — insufficient as stated, and hazardous if treated ### as curative. - Adding fee and covenant text does not cure comprehension for THIS product, because the load-bearing terms are quantitatively incomprehensible in narrative form: a contingent percentage of an unknowable future value, a share that escalates with time, an income stream switchable by the counterparty's model, and a settlement computed off an appraisal the customer pays to dispute. Under the net-impression standard, accurate-but-buried or accurate-but-unusable disclosure does not fix a deceptive impression (FTC/CFPB net-impression and buried-disclosure doctrine — framework SETTLED, application LIKELY INTERPRETATION). Worse, a compliant-LOOKING disclosure stack functions as a safe-harbor shield for the seller while leaving the asymmetry intact — disclosure as liability laundering. The only disclosure that would actually inform is the one my Stage 1 verdict specified: dollarized best/base/worst exit tables at 3/7/15/25-year holds, an APR-equivalent, and a side-by-side matched-cash-flow HELOC comparison — and the panel's shared conclusion is that a product disclosed that way does not sell. That is not a reason to soften the disclosure; it is the product verdict. ### Fixes from other reviewers I affirmatively endorse (no conflict): ops 1.2 (net-basis capital return); ops 1.4 mechanics PLUS red-team M1's stronger terms (independent pause governance, hysteresis, platform-paid disputes, paused months extend the term, pause-abuse = investor default with forfeiture teeth) — ops's version alone still permits permanent confiscation of paused months (my T9) and is insufficient without M1; estate fixes 4 (incapacity and medical occupancy carve-out), 5 (all-spouse joinder), 6 (escrow/LC-backed committed capital with springing release); title fixes 3-4 (subordination and payoff machinery with deadlines and penalties); red-team M2 (remedy limitations running with the instrument, binding assignees, permanent servicer-of-record) and M7 (recording hygiene, payment-change authentication). -------------------------------------------------------------------------------- ## DIRECT CONFLICTS (all pairs I can see, including those not involving me) DC1. Title (verdict fix 1) and Estate (fix 1) say: record a true security instrument / performance deed of trust — everything downstream "snaps into settled law." Tax (D-1, P-6, verdict) and Sec-cf (3.1, 3.2) say: lien/secured posture is the confession that detonates equity characterization, TILA, usury, and the arbitration clause; Estate's own kill-condition concedes the product cannot be simultaneously tax-clean, regulation-exempt, and bankruptcy-survivable. These fixes are incompatible: the product needs the lien to survive courts and needs the not-lien to survive regulators, and no proposed fix delivers both. (Red-team verdict pt 1 makes the same meta-point; recorded here as a formal conflict for Stage 5.) DC2. Sec-cf (verdict 8.1) says the 1.0x no-premium structure "genuinely improves the equity characterization versus incumbents." Tax (D-1, A-2) and Title (front matter, Roberts loss-sharing test) say the 1.0x floor is the single most debt-like feature and fails the loss-sharing test more cleanly than incumbents. These cannot both be right; sec-cf's own section 3.1 sides with tax/title against its own verdict paragraph. DC3. Estate (1.3) says the ratchet economically rewards the investor for probate delay. Actuarial (4.2, 5.1) and Red-team (M2) say every year of delay destroys 8-10% of holder PV even with the ratchet, so holders are incentivized to force exit, not prolong it. Both cannot be right about the holder's incentive; the fixes differ accordingly (toll-the-ratchet vs bind-the-assignee) and Stage 5 must adopt both harms but the correct incentive model (actuarial's). DC4. Investor (verdict fix path: priced spread/premium restoring loss-absorption) and Actuarial (fix iv: fixed-dollar senior accrual) propose compensation mechanisms that Tax (D-3, verdict minimum redesign: eliminate the ratchet / any time-based accrual; genuine participation in depreciation) says are definitionally interest and must be removed. The economics fixes and the tax fix are mutually exclusive; the design is trapped between its tax posture and its return requirements (investor's own words). DC5. Actuarial (fix ii) says: re-underwrite honestly as a 20-30-year hold-to-organic-exit vehicle. Investor (C1, C4, B5) says: no LP base exists for an illiquid ~7%-net 15-20+ year first-time vehicle — the honest version is unfundable. One reviewer's fix is the other reviewer's proof of kill; they cannot both stand as a launch path. DC6. Ops (1.4 fixes) and Red-team (M1) require softening the pause (buffer, hysteresis, term extension, pause-abuse forfeiture, platform-paid disputes) — i.e., more capital deployed into falling markets and real penalties on the investor. Investor (C3, C5, AS4) establishes the pause IS the investor's only capital protection in a structure with zero spread or loss-absorption anywhere. The consumer-protective pause and the investor-viable pause are incompatible without new loss-absorbing capital that DC4 shows cannot be added without becoming interest. DC7. Ops (1.1 fix) says: raise/reprice the origination fee to cover origination. Sec-cf (3.2, 4 fee caps) and Retail (T6/P3) say the existing fee is already an undisclosed hidden charge with HOEPA/points-and-fees exposure on recharacterization. A bigger fee fixes the P&L by worsening the consumer-law posture; the two requirements pull in opposite directions. DC8. Ops (5.2 fix) says: put a regulatory-event suspension clause in the homeowner contract. Sec-cf (3.3, 6.A1) says the "binding obligation / payments pause, not you" framing is already deceptive-and-abusive territory, and an additional investor-invocable suspension right deepens exactly that exposure. The operational fix and the UDAAP posture are incompatible as drafted (ops itself flags the tension; recorded here as a formal conflict). DC9. Retail (P4) and Sec-cf (3.5) say the arbitration clause and class waiver are generally enforceable under the FAA (SETTLED LAW), leaving no viable consumer remedy. Red-team (M5) says California public injunctive relief is non-waivable under McGill (SETTLED LAW in CA). Not fully compatible as originally written: P4's "removes the only economically viable remedy" is overstated for California. Resolution: I adopt M5 as a refinement (see SC2); Stage 5 should carry the reconciled version. DC10. Investor (A10) says the spec Section 20 downside example is inconsistent with the pause mechanism by ~10x on deployed capital (payments should have paused within months on a declining path). Sec-cf (6.A2) and Retail (T4) use Section 20's 36-month deployment at face value in their deception/harm arithmetic. Both cannot be right about the downside cash flows. Resolution: A10 is correct; my T4 and sec-cf's 6.A2 dollar figures need pause-consistent re-computation in Stage 4 (direction of both findings unchanged — see SC1). -------------------------------------------------------------------------------- ## TAG INDEX FOR NEW LEGAL/TAX CLAIMS MADE IN THIS FILE 1. Nonjudicial power-of-sale mechanics under a recorded deed of trust (FX1(a)): SETTLED LAW as mechanics; attachment to a recorded Equity Flow security instrument absent express waiver: LIKELY INTERPRETATION. 2. Lien ride-through of discharge and homestead-exemption defeat if a true consensual lien is recorded (FX1(b)): SETTLED LAW (adopting estate 4.1/4.4 — Long v. Bullard; Caulkett; 522(c)(2)). 3. Larger origination fee worsening HOEPA points-and-fees and state fee-cap exposure on recharacterization (FX2): SETTLED LAW contingent on recharacterization; application here LIKELY INTERPRETATION. 4. Functional reverse-mortgage capture of a maturity-dated monthly-advance product under state functional definitions (FX4): STATE-DEPENDENT. 5. Time-based accrual charged to the homeowner characterized as interest (FX5): LIKELY INTERPRETATION (adopting tax D-3). 6. McGill non-waivability of public injunctive relief (SC2, DC9): SETTLED LAW in California; federal survival (Blair): LIKELY INTERPRETATION. 7. Net-impression / buried-disclosure doctrine defeating disclosure-as-cure (FX8): framework SETTLED (FTC/CFPB deception standards); application to this product LIKELY INTERPRETATION. -------------------------------------------------------------------------------- ## CLOSING POSITION Nothing in cross-examination moves my verdict off KILL. The disputes above are corrections of reasoning inside a panel that is otherwise convergent: the fix-before-launch verdicts each depend on a repair that another reviewer's findings render impossible (DC1, DC4, DC5, DC6, DC7), and the repairs that ARE jointly feasible (disclosure, joinder, pause hysteresis, payoff machinery) are precisely the ones that, once implemented honestly, eliminate the customer base. From the homeowner's chair, the most important Stage 5 instruction from this file is FX1: if the panel's most-agreed fix — "record a real lien" — is adopted without its full consumer-protection package, the diligence process will have made the product more dangerous to the people it is sold to than the broken original. A fix that arms the counterparty is not a fix. END OF CROSS-EXAMINATION — RETAIL CUSTOMER