Ind AS 117 (IFRS 17) — Life Insurance CSM Model

Indian Market | IALM 2012-14 | RBI G-Sec | Monte Carlo Risk Adjustment

Author: Prithiyanga Bintu Mani
Qualification: MSc Actuarial Science — University of Kent (Merit, 2025)
IFoA Exemptions: CM1, CM2, CB1, CB2, CS1, CS2

---

Overview

This project implements a working actuarial model under Ind AS 117 (India's equivalent of IFRS 17) for a simulated portfolio of 1,000 Indian term life insurance policies.

India's insurance regulator IRDAI has mandated Ind AS 117 implementation by FY2027 (April 2027). This model demonstrates the three core building blocks of the General Measurement Model (GMM):

Component	Value	Description
BEL (Best Estimate Liability)	₹ −1.607 crore	EPV of future outflows minus inflows
RA (Risk Adjustment)	₹ +0.997 crore	75th percentile VaR via Monte Carlo
CSM (Contractual Service Margin)	₹ +2.345 crore	Unearned profit released over 20 years
Total Liability	₹ 1.735 crore	BEL + RA + CSM on Day 1 balance sheet

---

Data Sources

Data	Source	URL
Mortality rates	IALM 2012-14 — Institute of Actuaries of India	actuariesindia.org
Discount curve	CCIL Tenorwise G-Sec Yields — FY2025	ccilindia.com
Historical yields	RBI Database on Indian Economy	data.rbi.org.in
Expense benchmarks	IRDAI Annual Report 2024-25	irdai.gov.in
Regulatory standard	Ind AS 117 — MCA Gazette Aug 2024	mca.gov.in

---

Project Structure

IndAS117-CSM-Model/
│
├── IndAS117_Model.py          ← Main Python model (run this)
├── dashboard.html             ← LinkedIn visual dashboard
├── Report_IndAS117.docx       ← Full actuarial report
│
├── outputs/
│   ├── Chart1_IALM_Mortality.png
│   ├── Chart2_Yield_Curve.png
│   ├── Chart3_CSM_Rollforward.png
│   ├── Chart4_MonteCarlo_BEL.png
│   ├── CSM_Rollforward.csv
│   └── Portfolio_Data.csv
│
└── README.md

---

How to Run

1. Install dependencies

pip install numpy matplotlib pandas scipy

2. Run the model

python IndAS117_Model.py

3. Expected runtime

Section	Time
Sections 1–5 (setup, BEL)	~30 seconds
Section 6 (Monte Carlo, 500 scenarios)	~2–3 minutes
Sections 7–9 (CSM, charts, save)	~10 seconds

---

Model Architecture

Section 1 — IALM 2012-14 Mortality Table

Exact qx values from the published IAI PDF loaded into Python dictionaries for age-based lookup (ages 2–115). Female rates derived as 78% of male per IRDAI/LIC actuarial practice (LIC Valuation Basis L-42).

Section 2 — RBI G-Sec Yield Curve

Nine tenor points (1yr to 30yr) interpolated using scipy.interpolate.CubicSpline to produce a smooth discount curve. Under Ind AS 117 Para 36, discount rates must reflect current market rates at each reporting date.

Section 3 — Actuarial Functions

Three core functions:

• get_tpx() — survival probability vector from entry age
• epv_death_benefit() — EPV of claims paid at mid-year of death
• epv_annuity_due() — EPV of premium annuity (start of each year alive)

Section 4 — Portfolio Simulation

1,000 policyholders simulated with:

• Ages 25–55 (uniform random)
• Gender 70% male / 30% female (IRDAI 2024-25 industry mix)
• Sum assured log-normal (mean ₹25L, range ₹5L–₹100L)

Section 5 — Premiums and BEL

Net premium = EPV(claims) / EPV(annuity) — pure equivalence principle
Gross premium = Net / (1 − expense ratio − commission − profit margin)
BEL = EPV(outflows) − EPV(inflows)

Pricing assumptions:

Expense ratio   : 15.6%  (IRDAI Annual Report 2024-25)
Commission rate :  5.0%
Profit margin   :  8.0%  → creates positive CSM at inception

Section 6 — Risk Adjustment (Monte Carlo)

Key methodology advantage of Python/R over Excel:

for yr in range(1, term + 1):
    qx    = get_qx(age + yr - 1, gender)
    death = rng.binomial(1, tpx[yr] * qx)   # stochastic death draw
    if death == 1:
        total_pv += sa * disc(yr - 0.5)
        break

500 scenarios produce a distribution of total claims PV.
Risk Adjustment = 75th percentile − Mean (VaR approach, Ind AS 117 Para 37)

Section 7 — CSM at Inception

FCF  = BEL + RA + acquisition costs − first-year premium
CSM  = −FCF

If CSM < 0 → onerous contract (loss recognised immediately, CSM = 0).

Section 8 — CSM Roll-Forward (Ind AS 117 Para 44)

Opening CSM
+ Interest accretion (locked-in rate = 10-yr G-Sec at inception = 6.53%)
− CSM released (proportional to coverage units delivered)
= Closing CSM

CSM reaches ₹0.000 at year 20 — all profit recognised by end of coverage.

---

Key Results

Balance Sheet at Inception

══════════════════════════════════════════════════════════
  IND AS 117 — INSURANCE CONTRACT LIABILITY AT INCEPTION
══════════════════════════════════════════════════════════
  Best Estimate Liability (BEL)     : ₹   -1.607 crore
  Risk Adjustment (RA) [75th pctile]: ₹    0.997 crore
  Contractual Service Margin (CSM)  : ₹    2.345 crore
──────────────────────────────────────────────────────────
  Total Insurance Contract Liability: ₹    1.735 crore
══════════════════════════════════════════════════════════

CSM Roll-Forward (Selected Years)

Year	Opening	Interest	Released	Closing
1	2.345	0.153	0.131	2.367
5	2.390	0.156	0.166	2.380
10	2.217	0.145	0.223	2.139
15	1.625	0.106	0.295	1.436
20	0.365	0.024	0.389	0.000

All figures in ₹ crore.

---

Regulatory Context

Milestone	Date
Ind AS 117 notified by MCA	August 2024
IRDAI Expert Committee reconstituted	February 2024
First proforma Ind AS 117 submission	December 2025
Second proforma submission	June 2026
Full implementation deadline	April 2027 (FY27)

India joins 140+ countries implementing IFRS 17. The transition affects all 58 Indian insurers, requiring them to move from Indian GAAP to a market-consistent, fulfilment-value basis of reporting.

---

Limitations and Assumptions

• Female mortality derived as 78% of male (separate female IALM table not publicly available)
• Lapse rates set to zero (no policyholder surrenders modelled)
• No expense inflation — expenses assumed level in real terms
• Portfolio is homogeneous (single product, single term)
• RBI G-Sec rates assumed flat over projection period (locked-in at inception per Ind AS 117)

---

License

This project is for educational and portfolio purposes. Data sources are all publicly available.
IALM 2012-14 © Institute of Actuaries of India. RBI data © Reserve Bank of India.

---

Built as part of an actuarial portfolio project demonstrating Ind AS 117 (IFRS 17) implementation skills for the Indian insurance market.