ZenLearnRead the report
ZenLearn Research · July 2026

LAP Underwriting Failure: The Branch-Level Judgment Crisis

256 adjudicated Indian court cases, 2013–2026: what branch officers chose not to do — and three pre-sanction steps that would have stopped most of it.

Read the report
256
Adjudicated Indian court cases reviewed
2013–2026
84%
Directly implicate the sanctioning officer
of 256 cases
Rs 50
Cost of the CERSAI lookup that was skipped
per query

The situation

Regulators and courts did not punish what branch officers did not know. They punished what branch officers did not do.

Across 256 adjudicated loan-against-property (LAP) cases spanning 2013 to 2026, 86% generated Central Bureau of Investigation or Economic Offences Wing criminal proceedings, and 84% directly implicated the sanctioning officer or credit committee member. The collateral underpin of the LAP product — the property title and its independent valuation — was bypassed routinely, not accidentally.

A forged title certificate passes every credit bureau check, every capital adequacy rule, and every AML transaction-monitoring flag. A multiple-mortgaged property already pledged to three other lenders appears clean on every credit score. Only a review of the origination file itself — with a CERSAI lookup printout, an independent panel-valuer appointment letter, and an encumbrance certificate from the bank's own legal panel — catches it. Three pre-sanction steps. Each one an existing documented protocol. The 256 cases in this corpus are cases where that protocol was skipped.

Key findings

What the origination file already showed.

Finding 01
Staff collusion is the primary failure mode — present in 84% of 256 adjudicated cases.

The existence of best-in-class lenders with gross NPA rates of 0.52–2.15% on LAP books (Bajaj Housing Finance, Piramal Finance, Cholamandalam, IIFL — all primary-verified FY25) in the same regulatory environment demonstrates that the failure pattern is management-controllable, not product-intrinsic.

Finding 02
Three pre-sanction steps were absent or suppressed in the majority of cases.

Title defects or encumbrances that an encumbrance certificate would have surfaced appear in 36% of cases. Multiple mortgages that a CERSAI lookup would have flagged appear in 9%. These are not judgment calls — they are documented steps, and their absence is measurable.

Finding 03
9% of cases involve a property already registered under CERSAI at disbursement.

A mandatory CERSAI lookup costing under Rs 50 would have prevented the advance in each of these 23 cases. This gap sits outside the scope of credit bureau checks, risk-weight rules, and post-disbursement EWS frameworks. The RBI Fraud Risk Management Directions 2024 (NBFC) mandate a post-sanction title audit — but not a pre-sanction CERSAI lookup.

Finding 04
2022 is the inflection year: case count rose 4.6× year-on-year and stayed elevated.

The corpus's 17-year time series shows 2022 as a structural shift, not a single enforcement event. The 2023 count matched 2022's; 2025 reached 3.4× the 2024 level. The risk is accelerating, not historical.

Finding 05
Prevention is finite and cheap: three steps, one protocol.

Independent title search, CERSAI lookup, and independent panel-valuer appointment would have disrupted the collusion mechanism in the majority of cases — because each step requires a second party whose output cannot be fabricated without that party's cooperation.

Pattern frequency · 256 adjudicated LAP cases · ZenLearn Research

Staff collusion drives 84% of the corpus — and it has a checkable pre-sanction signature.

Each bar shows how many of the 256 adjudicated cases exhibit this failure pattern. Patterns overlap: a single case can show multiple failure modes. The three right-hand bars map directly to the three pre-sanction protocol steps.

Staff collusion / criminal referral221 cases (86%)Title defect / encumbrance (PD)92 cases (36%)DRT valuation shortfall82 cases (32%)End-use diversion61 cases (24%)Multiple mortgage / CERSAI gap23 cases (9%)

Staff collusion (215 cases) and CBI/EOW criminal referral (221 cases) are the same failure viewed from two angles — institutional failure and its legal consequence. Shown combined here.

The question not asked: a cooperative bank LAP cluster

Four separate court proceedings arose from a single cooperative bank's origination fraud — each a distinct LAP file, each involving a different borrower, each sanctioned by the same branch-level process. The legal opinion in each case was sourced from the same panel; the valuations came from the same empanelled firm; none of the four files carried a CERSAI lookup printout dated within 30 days of sanction. The question not asked, on any of the four files, was the same: who appointed the valuer, and when was the CERSAI search run?

What's in the report

9 sections, every case cited to a public document.

01
Methodology and Case Volume Trend
17-year time series: how the enforcement record has evolved, 2010–2026
02
Executive Summary and Control Architecture
The three structural findings and why existing controls do not cover the origination gap
03
International Comparators
How FNMA (April 2025) and FCA MCOB 11 closed the same gap — and the 2005–2015 US subprime precedent
04
Sectoral Cuts (PSBs, HFCs, Cooperative, Private)
Pattern differences across four institutional cohorts
04
Pattern and Stakes Cut
Five failure patterns ranked by frequency; the overlap between collusion and criminal proceedings
05
Vignettes
Named cases: the question not asked in each
06
Cost of Inaction and Scenario Model
Conservative / Base / Severe provisioning at three book sizes
06
Board Briefing, Leading Indicators, and Roadmap
Five board-reportable KPIs with owner and cadence; 30/90/180/365-day implementation plan
06
Exposure Self-Assessment and Early Warning Indicators
Nine structural conditions that elevate risk; nine corpus-derived warning signals

The next step is in your own book.

Request the four-page Decision Brief — includes the organizational exposure self-assessment and the 30/90/180-day protocol roadmap.

Read full report
Prepared by
ZenLearn Research · ZenLearn Platform
256 adjudicated court cases, 2013–2026
Primary-verified across four institutional cohorts
RBI FRMD NBFC 2024, FNMA April 2025, FCA MCOB 11 cross-referenced
Contact: research@zenlearn.ai · zenlearn.ai/research
Primary sources

256 court-adjudicated LAP cases, Indian court records 2013–2026 (SARFAESI, CrPC, IPC proceedings). Lender benchmarks: FY25 annual reports, primary-verified. Regulatory sources: RBI Fraud Risk Management Directions 2024 (NBFC), RBI FSR June 2025, FNMA April 2025, FCA MCOB 11. All ZenLearn-derived tallies explicitly qualified as self-derived. Full reference list in the report.

Related research
Housing Loan Underwriting: The Signals Before Sanction
112 court-documented housing-loan failures — the same three pre-sanction signals in 64% of cases.
Read
SME Underwriting: The 13 Signals Banks Miss Before a Default
50 verified MSME court cases fused into a 13-signal EWS codebook.
Read
© 2026 ZenLearn Research · Judgment Intelligence
HomeJudgmentPrivacyTerms