Prepared: 2026-04-07 Question tested: Is July 2025's $40,768.39 / $6,391.71 = 6.4× "fee-credit vs. 20%-formula" ratio anomalous or representative of a 20-month pattern? Answer: The 6.4× ratio itself is NOT anomalous — it sits at ~P85 of a 21-month distribution ranging 2.22× – 7.72×. What is anomalous about July 2025 is the 8-month suppression, not the number.
Source documents: 21 monthly "Payment Allocation by Client Report" PDFs in output/goldberg_financial_attachments/, filtered to the "RCL as originating attorney" scope. Each report contains two lawyer-summary sections:
Extraction script: /tmp/extract_allocations.py (pdfplumber, regex against the Fee Credit section; three months manually spot-checked against source PDFs — Oct 2023 $112,379.82, Oct 2024 verified, May 2025 $28,526.40 / $64,792.40, Jun 2025 $20,936.48 / $56,028.54 — all match).
Column definitions in the CSV:
| Column | Source | Meaning |
|---|---|---|
Total_Collected_Fees |
First "Total" row, Fees column | Firm-wide fees collected on Litman-originated matters |
Total_All_Incl_Costs |
First "Total" row, Total column | Same, plus hard+soft cost reimbursements |
Fee_Credit_Litman_Fees |
"Richard Litman" row, Fees column | Litman's fee-credit allocation on fees only |
Fee_Credit_Litman_Total |
"Richard Litman" row, Total column | Litman's fee-credit including hard+soft pass-throughs |
Formula_20pct_of_Fees |
Total_Collected_Fees × 0.20 |
The "20% of collected fees" benchmark |
Multiplier_Fees |
Fee_Credit_Litman_Fees ÷ Formula_20pct_of_Fees |
Tests if fee share matches the 20% rule |
Multiplier_Total |
Fee_Credit_Litman_Total ÷ Formula_20pct_of_Fees |
The Finding-#50 ratio (total credit ÷ 20% formula) |
Paid_To_Litman |
Not in source | Allocation is not "paid" — payment trace requires Wells Fargo/BOA records |
| Stat | Value |
|---|---|
| Mean | 1.009 |
| Median | 1.000 |
| Std dev | 0.025 |
| Range | 0.95 – 1.06 |
Interpretation: NGM's internal system applies the 20%-of-fees rule to Litman with mechanical precision. Litman's Fees-only fee-credit is effectively indistinguishable from 20% of firm-wide fees collected on his originating clients, every single month, for 21 straight months. This validates the 20% formula as the contractual/operational baseline.
| Stat | Value |
|---|---|
| Mean | 3.12× |
| Median | 2.86× |
| Std dev | 1.17 |
| Min | 2.22× (Oct 2024) |
| Max | 7.72× (Jun 2024) |
| Jul 2025 (reference) | 6.38× |
Distribution: - 21 of 21 months ≥ 2.22× - 8 of 21 months ≥ 3.0× - 3 of 21 months ≥ 4.0× (Jun 2024, Apr 2024, Apr 2025) - 1 of 21 months ≥ 5.0× (Jun 2024) - Jul 2025's 6.38× is ~P85 — above typical, but below the 7.72× Jun 2024 high
Why the ratio varies so much: The numerator is Fees + Hard costs + Soft costs, but the denominator is 20% × Fees only. In months with large foreign-filing or maintenance-fee hard-cost pass-throughs on Litman-originated patents (KFU, KSU, KNPC, KISR, etc.), the ratio spikes. Jun 2024 in particular shows $144,630 Litman total credit against only $93,691 firm-wide fees — meaning hard/soft costs exceeded fees for that month on Litman matters.
The "6.4× anomaly" framing is imprecise. The accurate finding is:
| Aggregate | Amount |
|---|---|
| Firm-wide fees collected on Litman matters | $8,607,871.79 |
| Firm-wide all-in (fees+hard+soft) on Litman matters | $11,666,261.86 |
| Litman Fee-Credit — Fees only (20% share) | $1,731,898.18 |
| Litman Fee-Credit — Total (incl. cost pass-throughs) | $4,790,288.25 |
| 20% formula check (8,607,871.79 × 0.20) | $1,721,574.36 |
The "fees-only" fee-credit of $1,731,898 matches the 20% formula within 0.6% — confirming this figure is the defensible 21-month baseline for Litman's contractual share of fees actually collected (per NGM's own internal books).
4 missing months from the Oct 2023 – Jul 2025 window: - Jul 2023, Aug 2023, Sep 2023 — precede our series (Q3 2023 — likely covered by a quarterly "3Q2023 Payment Allocation Listing" analogous to the 3Q/4Q 2022 PDFs in the corpus; demand in discovery). - Jul 2025, Aug 2025, Sep 2025 — Jul 2025 is already recovered (Finding #50). Aug & Sep 2025 are known suppressed.
The 21-month Litman Fee-Credit Total averages $228,109/month (median $134,657; range $42,081–$570,424). Using 12 months of 2024 as a closer baseline: monthly mean ≈ $260,894, median $130,234.
Conservative projection (median-based):
| Month | Projected Fee_Credit_Litman_Total | Projected Fees-only |
|---|---|---|
| Jul 2023 | $130,000 | $29,000 |
| Aug 2023 | $130,000 | $29,000 |
| Sep 2023 | $130,000 | $29,000 |
| Jul 2025 | $40,768 (actual) | $6,392 (actual) |
| Aug 2025 | $50,000–$150,000 | $10,000–$30,000 |
| Sep 2025 | $50,000–$150,000 | $10,000–$30,000 |
Note on Jul 2025 deviation from projection: Jul 2025's actual $40,768 total credit is BELOW the 21-month median ($134,657). One interpretation: the mid-2025 collapse (Apr–Jun 2025 fee-credit totals were also all under $107,000) reflects a genuine collections slowdown — OR active diversion of receipts away from the Litman-originating classification. The July report confirms the trajectory is not aberrant; the 6.4× multiplier is symptomatic of cost pass-throughs exceeding fees in a low-collection month, not of inflation.
Adding Q3 2023 (separate pre-series gap) at ~$90K: Oct 2023 – Sep 2025 full 24-month span ≈ $1.85M – $1.91M in Litman's 20%-of-fees share alone (before hard/soft pass-throughs, and before the $1.98M KFU-only unallocated universe from Finding #51).
Critical caveat: These Payment Allocation Reports show what NGM credited to Litman internally, NOT what NGM paid Litman. The actual-payment column (Paid_To_Litman) is empty in the source documents. Tracing actual payments requires Wells Fargo / Bank of America trust-account records (Account 003926278751, Finding #57).
Conservative inference: If Litman's Q4 2025 owed balance is $246,628 against $1.23M collected fees (per Finding #39, Connell Foley production), that's a ~20% ratio — consistent with the 20% formula. Scaling to the 21-month $1,731,898 credited: if a similar proportion went unpaid (~20% unpaid rate applied by analogy), underpayment would be in the range of $340K–$520K across the 21 months — consistent with the $424K–$928K defensible-anchor range already on the books (Finding #49).
This analysis does NOT independently establish underpayment — it establishes that the 20% rule was applied with mechanical precision and that the 21-month Litman Fees-only fee-credit base is $1.73M, which can be cross-checked against actual payment records in discovery.
goldberg_financial_attachments/ — it was recovered separately and its figures are cited from the Finding #50 memo, not re-extracted here.| File | Path |
|---|---|
| CSV time series | output/MONTHLY_FEE_CREDIT_TIMESERIES_2023-10_TO_2025-06.csv |
| Analysis memo | output/MONTHLY_FEE_CREDIT_ANALYSIS_MEMO.md |
| Bar chart | output/MONTHLY_FEE_CREDIT_CHART.png |