The IRDAI has published a consultation paper on the Public Insurance Registry. The media commentary I have read has framed the registry as India's UPI moment for insurance. But much of the coverage misses an important proposal. The paper proposes an Insurance Risk Score (IRS): "A unified, consent-based risk score synthesizing insurance history (IIB), credit (Credit Information Companies (CICs)), and other permitted external data sources -- as a decision-support input at underwriting."
The proposal appears on pages 29 and 31, and the term gets a glossary entry as well. Page 30 lists credit history among the external information insurers want for "protection cover and surety bonds", but the paper never defines protection cover (does it include term life?). The paper references a range of insurance products, including life and general, but does not specify which of them the risk score would apply to. IRDAI should specify the products covered.
ALSO READ: Reading Between The Lines: Illusion Of Protection? Why High Premiums Leave Indians Underinsured
The paper does not prescribe a formula that links the score to a premium. It places the score inside underwriting, and user stories 19 and 21 discuss pricing alongside it. The score could therefore influence whether you are accepted, what you pay and on what terms. IRDAI has invited comments on the paper until Sept. 30. The registry is an important piece of public infrastructure and much of what it proposes is overdue. But the risk score deserves scrutiny before the paper becomes regulation.
Can credit data predict mortality?
One of the first things that came to my mind on reading about the risk score: can credit data predict mortality, and therefore inform underwriting decisions? TransUnion built a credit-based mortality score called TrueRisk Life, which ranks people from 1 to 100 (100 being highest risk).
Munich Re, one of the world's largest reinsurers, reviewed it against a sample of 16 million people in the US and found that the scores predicted mortality in the general population. Reinsurance Group of America (RGA), another major life reinsurer, tracked 50 million Americans with credit records from 1999 to 2020, covering six million deaths.
People in the worst-scoring band died at 4.48 times the rate of those in the best. The consultation paper makes a case for better insurance data. It records that life insurers work from mortality trends "more than 6 to 7 years old" and add "conservative safety margins" to compensate for that. Better data could reduce the cost of cover for some buyers.
Whether credit data has predictive value for a particular Indian insurance product still needs to be validated. Even if it is, two questions remain: is its use fair and consistent with the registry's objective of inclusion, and can the person being scored inspect the result and challenge its consequences?
ALSO READ: Reading Between The Lines: FY26 Premium Data Shows The Insurance Industry's Shifting Dynamics
How can an insurance customer access the score or dispute it?
Paragraph 10.3(b) contemplates customers requesting access to their own information. Its examples include a consolidated view of policies and a claims history shared with a new insurer. The paper also commits PIR to compliance with the Digital Personal Data Protection Act, 2023, whose section 11 provides for access to a summary of personal data and processing activities, subject to conditions. Section 11 is not yet in force.
Under the staggered commencement notified in November 2025, sections 3 to 17 of the Act take effect on May 14, 2027. These provisions do not clearly settle whether a person can obtain a score computed about them, the inputs behind it and an explanation of how it was used. Access to a summary of processing does not necessarily tell you how an insurance decision was reached.
The proposal leaves some practical questions open: how would someone inspect their score, dispute it and get a response within a defined period? Paragraph 10.9 holds insurers responsible for the accuracy of source records. PIR is responsible for "the linkages, indicators, signals and other outputs it generates or holds". Paragraph 11.4(e) adds that PIR can correct only errors in its own computations or linkages.
Paragraph 10.6 routes individual requests to "the concerned Insurer for necessary action". An insurer could serve as the front door and escalate to PIR. The paper does not describe that handoff or explain how the responsible institution must resolve a dispute about a derived output, or by when.
A complaint could be about three separate things. The underlying data could be wrong; the linkages or computation could be wrong; or the calculation could be correct, but the insurer's decision is open to challenge. The paper explicitly allocates responsibility for data and computation errors. It does not explain how someone challenges an adverse decision based on an accurately computed score.
ALSO READ | LIC Launches Two New Plans Under Jeevan Sathi Brand — Check Details
What existing systems give the person being scored
India's credit bureaus operate under the Credit Information Companies (Regulation) Act, 2005, supervised by the Reserve Bank. Individuals can obtain one free credit report from each bureau every year. A customer can raise a complaint or rectification request. If the information is not corrected within 30 days, RBI rules stipulate Rs 100 to be paid to the customer for every day of delay by the bureau or the lender depending on where the delay occurred.
The United States has something the Indian credit framework lacks. Under section 615(a) of the Fair Credit Reporting Act, an insurer that denies cover, raises a charge or imposes other adverse terms based on a consumer report must tell the applicant so, identify the agency that supplied the report, and explain to the customer their right to a free copy within 60 days and to dispute it.
Access is not left to the applicant to seek out once a year. It is triggered the moment an underwriting decision based on a report goes against them. US rules also reach the score itself, not only the report behind it. Neither requirement appears in the PIR paper. It should require notice at the point of an adverse decision, access to the underlying information, and disclosure of the score and the factors behind it.
ALSO READ: IRDAI Member Urges Insurers To Cut Costs And Fix Mis-selling For Affordability
Four safeguards that need to be considered
Disclosure when the score is used against you. If an insurer relies on the score in declining an application, increasing the premium or imposing less favourable terms, the applicant should be told. They should receive the score, its date, the material factors behind it and a route to review. A dispute route that reaches the responsible institution. Errors in PIR's outputs should reach PIR.
Challenges to how an insurer used the score should reach the insurer, with a clear escalation route. Individuals should have one entry point, clear handoffs and deadlines. A rule for missing or insufficient credit histories. Some people will have no credit file at all. Others will have a file too thin to support a reliable score. The paper does not specify how either situation would be treated.
The RGA study I referred to earlier looked at people with thin credit histories. It found they usually scored badly on TrueRisk Life, which means the model treats them as higher mortality risks. RGA also found that the score sorted risk accurately within that group. Either way, it raises a question the paper does not answer: would a thin credit file mark an Indian applicant as a worse risk?
Munich Re saw the same pattern among younger applicants, who have short credit records simply because they are young. The rules should establish their treatment before a model goes into use. An alternative underwriting route for people who decline. The paper promises that consent is revocable. But withdrawal of consent cannot undo an existing underwriting decision based on the individual's score. It also does not say what happens if someone declines in the first place.
Indian life insurance underwriting already works without a credit score, through medical tests, income proof and disclosures. That is an alternative route already available. The regulations should confirm it stays available and should bar a refusal from being treated as an adverse signal. These safeguards are easier to write into the regulation now than to add once the score is in use.
(IRDAI has invited comments on its paper. Comments can be filed on the web form at iib.gov.in/pir, or emailed to pirfeedback@iib.gov.in.)
Ashok Hegde is the founder of Gyansurance.com, a term insurance education platform.
Disclaimer: The views expressed in this article are solely those of the author and do not necessarily reflect the opinion of NDTV Profit or its affiliates. Readers are advised to conduct their own research or consult a qualified professional before making any investment or business decisions. NDTV Profit does not guarantee the accuracy, completeness, or reliability of the information presented in this article.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.