Why AIF returns are not comparable in India, and what performance data does exist

Can I compare AIF returns in India?

Category-level AIF performance is published. SEBI's Master Circular makes benchmarking mandatory, and the appointed agencies, CRISIL and NSE Indices, publish pooled performance by category, sub-category and vintage year on a post-expense, pre-carry, pre-tax basis. Per-fund performance is not published. It sits in the private placement memorandum and in reports to existing investors, neither of which is a public document.

The short version

There is a common claim that India publishes no AIF performance data at all. That is wrong, and a fund manager will tell you so.

What is true is narrower and more useful. Aggregate, category-level performance is published, under a SEBI mandate, on a stated basis. Per-fund performance is not. Everything on this page follows from that one line, and it is the reason this site publishes no per-fund return, NAV or IRR from any source, including a manager's own material.

What is published

Chapter 22 of SEBI's Master Circular for Alternative Investment Funds, issued 3 June 2026, makes performance benchmarking mandatory. It consolidates circular SEBI/HO/IMD/DF6/CIR/P/2020/24 of 5 February 2020, which introduced the regime, and SEBI/HO/IMD/DF6/CIR/P/2020/99 of 12 June 2020, which set the threshold for the industry association that appoints the agencies.

The machinery, in SEBI's own terms:

ProvisionWhat it requires
22.3.1An association representing at least 33% of AIFs notifies one or more Benchmarking Agencies. Every AIF enters an agreement with one.
22.3.3Every scheme past one year from first close reports scheme-wise valuation and cash flow data.
22.3.4Benchmarking is half-yearly, on data as at 30 September and 31 March.
22.3.5The agencies compile the data into industry benchmarks for Categories I, II and III, separately for each year since 2012, and those benchmarks "will be disseminated in a manner that is accessible to the public".
22.4.2Data is due within 45 days of 30 September and within 7 months of 31 March.
22.4.3March data is audited. September data may be unaudited.
22.4.8Performance reporting and benchmarking is carried out on pre-tax net asset value of the scheme.
22.4.9Reported in both rupee and US dollar terms.

The association is the IVCA, which has appointed three agencies: CRISIL from 1 October 2020, NSE Indices from 14 March 2022 and Preqin from 25 May 2022. An AIF may obtain its report from any of them.

Two of the three publish openly.

CRISIL Intelligence publishes category benchmarks for Categories I, II and III plus 17 sub-category benchmarks. For Categories I and II it computes pooled IRR by vintage year, along with DPI, RVPI and TVPI, quartile thresholds and a public market equivalent. For Category III it builds an asset-weighted index. Its methodology document is free to read and states the basis plainly: "returns are calculated on post-expense, pre-carry, and pre-tax basis". Its published tables are stated as at 30 September 2025.

NSE Indices publishes the Nifty AIF Benchmark Report semi-annually and now lists 18 sub-category benchmarks across the three categories. Its page carries the same statement: "Returns are calculated on a post-expenses, pre-carry, and pre-tax basis." Its current highlights are stated as at 30 September 2025. Its press release of 24 January 2025 put the report at sixteen sub-category benchmarks built from data received from more than 700 AIF schemes as at 31 March 2024.

This site does not reproduce the values. They are the agencies' own property, they are restated every six months, and a figure copied out of context and left to age is exactly the failure this page exists to argue against. Read them at the source, and note the as-at date on the table you are reading.

See AIF benchmark for the term itself.

Published to whom

"Published" needs qualifying, because the answer differs by document.

DocumentWho gets it
Industry and sub-category benchmark tablesPublic. SEBI 22.3.5 requires public dissemination, and both agencies carry headline tables free on their websites.
The full Benchmark ReportOn request from the agency, not downloaded. NSE Indices offers a request link; CRISIL supplies its sub-category report on request.
A fund's own performance-versus-benchmark reportThe fund, under 22.3.7. Then any investor or prospect the fund shows a past-performance claim to, under 22.5.1.
Customised Performance ReportsCommissioned by a fund, on consent of the funds whose data is used, as a fee-based service under 22.6.
Scheme-level valuation and cash flow dataThe agency alone. 22.3.2 requires the agreement to cover confidentiality, and IVCA states the AIF and the agency enter a non-disclosure agreement.

So the regime is not secretive and it is not a data-collection failure. It collects a great deal and discloses the aggregate. What it never does is attach a number to a named fund in public.

Why the gap is structural

An AIF is a private placement. Its terms, its strategy and its results are set out in the private placement memorandum, which is filed with SEBI but is not a public document. Nothing in the regime contemplates a public per-fund disclosure, because the product is not offered to the public.

That is a design consequence, not an oversight, and it will not be closed by better data collection. The data is already collected.

Four further reasons make a per-fund league table unsound even if every manager volunteered a number tomorrow.

  • Marks are not exits. Most of a private fund's reported value before its final years is the manager's own valuation of assets nobody has sold. A net asset value for an unlisted portfolio is an estimate produced by an interested party, checked by a valuer, and it becomes a result only on realisation. This is why the benchmarks report DPI, which is cash actually returned, separately from RVPI, which is what is still marked.
  • Vintage decides a great deal. Close-ended funds start in different years and sit at different points on the same curve. Both agencies benchmark by vintage year for this reason, and neither compares a 2019 fund with a 2023 one. A ranked list that ignores vintage is largely measuring age.
  • The J-curve. Early-life private funds routinely show negative returns by construction, because fees are drawn before value is realised.
  • No common calculation. Gross or net of fees, before or after carry, with or without the hurdle, money-weighted or time-weighted. Each choice moves the number materially. The benchmarking regime fixes a basis for itself, which is post-expense, pre-carry and pre-tax on pre-tax NAV. Nothing fixes the basis of a figure a manager puts in a deck.

Same investor, three products, one answer missing

AIFPMSSIF
Per-fund performance disclosureTo investors, privatelyMonthly, publiclyPublished NAV
Mandated byNo public per-fund mandateAPMIMutual fund regime
Common format across managersNoYesYes
Can a prospective investor compare two managers?NoYesYes
What is publicCategory benchmarksEach strategy's recordEach scheme's NAV

Three products sold to the same investor by the same distributors. Two can be ranked. One cannot. The difference is the disclosure rule and nothing about the underlying investment activity. See AIF vs PMS and AIF vs SIF.

What a "top 10 AIFs" list actually is

Several Indian sites publish ranked tables of best-performing AIFs, with a return figure against each named fund. Some run a whole comparison product on it.

Every one of those numbers came from the manager. No independent source can verify it, no regulator requires it in that form, and no two rows are guaranteed to be calculated the same way. Most of them are unrealised, which means the figure is the manager's estimate of what a portfolio would be worth if it were sold, and it has not been sold. A "since inception" percentage on an unrealised portfolio is an opinion about the future stated in the grammar of a result.

Ask three questions of any such list.

  1. Where did the numbers come from? If the answer is the funds themselves, it is a collection of marketing claims arranged in a table.
  2. What are they calculated on? If the list cannot state one basis that holds across every row, the rows are not comparable and the ranking means nothing.
  3. Which managers declined to appear? If nobody knows, the sample selected itself, and the funds that would have looked bad had no reason to send a number.

A list that cannot answer these is not a ranking of Indian AIFs. It is a ranking of which managers were willing to send a figure.

What you can demand

This is the concrete part, and it is stronger than most investors realise.

Chapter 22.5.1. In the PPM, and in any marketing or promotional or other material where past performance of the AIF is mentioned, the report of performance versus benchmark provided by the benchmarking agency for that AIF or scheme shall also be provided.

Chapter 22.5.2. In any reporting to existing investors, if performance is compared to any benchmark, a copy of that report shall be provided too.

So a return figure shown to you in a deck, a brochure or a distributor's email is supposed to arrive with the agency's report attached. Ask for it by name. Ask which of the three agencies produced it, and as at which half-year date. If what comes back is a Performance Report rather than a Benchmark Report, note the difference: under 22.6.2(e) the first is a customised, fee-based product commissioned by the fund against a peer set it helped define, and the second is the standard report generated under SEBI's mandate.

Also worth knowing: 22.5.3 requires an applicant claiming a track record from overseas funds to give the agencies its Indian investment data at registration, so an offshore record cannot be carried into an Indian registration unbenchmarked.

So how do you actually evaluate a manager

If ranking is unavailable, the work moves to things you can verify.

  • Read the PPM properly. It is the document the regime is built around, and it is the one place the fund's own terms are stated.
  • Ask for the benchmark report, as above, and read the basis and the date on it before the number.
  • Separate realised from marked. DPI is cash back. RVPI is an estimate. A manager who leads with TVPI and cannot break it apart is telling you less than they appear to.
  • Check the register. Registration, category and date are facts, and they are in the directory. Registration is a fact about a filing, not an endorsement.
  • Understand the fee stack, at AIF fees and costs, because it is deducted whether or not the strategy works.
  • Read the risks, at AIF risks.

The full framework is at how to choose an AIF.

What this site publishes instead

Registration facts from SEBI's register, in the directory. Sponsor families, so funds run by the same house are visible as a group. SEBI's own quarterly industry statistics, which measure commitments raised, funds raised and investments made, and are activity measures rather than performance. And the regulation, with each amendment named and dated.

Sources

  • Master Circular for Alternative Investment Funds (AIFs), reference HO/19/34/11(6)2025-AFD-POD1/I/12928/2026, issued 3 June 2026, last updated 30 July 2026. Chapter 22, Performance Benchmarking of AIFs, paragraphs 22.3 to 22.6. It supersedes the Master Circular of 7 May 2024, in which the same material sat at Chapter 16.
  • SEBI circular SEBI/HO/IMD/DF6/CIR/P/2020/24 dated 5 February 2020, "Disclosure Standards for Alternative Investment Funds (AIFs)", cited at footnote 83 of Chapter 22 as the origin of the benchmarking mandate.
  • SEBI circular SEBI/HO/IMD/DF6/CIR/P/2020/99 dated 12 June 2020, cited at footnote 84, on the association threshold for notifying benchmarking agencies.
  • SEBI circulars SEBI/HO/AFD/PoD/CIR/2023/97 dated 21 June 2023 and SEBI/HO/AFD/PoD-1/P/CIR/2024/123 dated 19 September 2024, on timely reporting of audited valuations to benchmarking agencies, cited at footnote 85.
  • IVCA, Performance Benchmarking of AIFs, for the appointment of CRISIL, NSE Indices and Preqin and their launch dates.
  • CRISIL AIF Benchmarks and its methodology.
  • NSE Indices, AIF Benchmark Report, and its press release of 24 January 2025.

Checked against source on 25 August 2026.

Related

Information only. Not investment advice.

Checked against source on 25 August 2026. This page is information, not legal, tax or investment advice.

Glossary · Fund directory · SEBI quarterly data