By Sandeep Handa | Co-Founder, Aspyra Partners
If you own two large-cap funds from two different fund houses, there is a number about your portfolio that nobody has ever shown you.
On average, about 45% of those two funds is the same thing. The same companies, held at similar weights.
To be clear, it's not that one fund is 45% duplicated and the other is fine. Put the two funds together, and close to half of your total money is sitting in the very same companies twice — once through each fund.
I want to be clear that not knowing this is not a failure on your part. There is no statement, no app screen and no rule anywhere that shows you this number. I didn't know it either until we started measuring it on real portfolios.
And this year, for the first time, SEBI did something about overlap.
Just not about yours.
What did SEBI actually do?
Earlier this year the regulator put a number in the rulebook that had never been there before.
Any sectoral or thematic equity scheme must now keep its portfolio overlap with other equity schemes under 50%.
The teeth are real. Overlap gets computed every quarter. Existing schemes have three years to comply, reducing the excess in stages rather than all at the end. A scheme that still can't meet the limit gets merged into another one. And every fund house has to publish its overlap numbers online each month.
I think this is good regulation. For years the industry has sold differently-named funds that quietly hold much the same stocks. Until now, "these two are basically the same fund" was an opinion you could argue about. Now it is a number somebody has to report.
So it would be reasonable to read that and assume the problem is handled.
It isn't. And what follows is not a loophole or a scandal — it is simply what the rule was built to do.
Does the rule cover my portfolio?
No. There are two boundaries, and almost none of the coverage mentioned either.
The first boundary is who the rule binds.
The obligation begins "Mutual Funds shall ensure..." A fund house can only ensure things about its own schemes. It cannot see inside a competitor's portfolio day by day, so it cannot compute overlap against one. And the final penalty is merging two schemes together, which only a single fund house can do.
So the rule governs two funds sold by the same company. It says nothing about two funds sold by two different companies.
Now think about how your own portfolio was actually built.
The portfolios we open up usually hold five or six equity funds, from four or five different fund houses, picked up over a decade. One came from a bank relationship manager. One came up in a conversation. One got added the year the first one went quiet, because stopping felt like giving up and adding felt like doing something.
Every one of those funds can be perfectly compliant with the new rule. And the thing they add up to can still be a single bet.
The second boundary is what gets counted.
When the regulator wrote the rule, it specifically left large-cap schemes out of the comparison. Overlap against a large-cap fund is not counted at all.
Hold that thought. It matters in a minute.
How much overlap is actually out there?
Value Research measured this properly — same category, but funds from different fund houses. Exactly the pairs nobody caps. Their data runs to the end of March this year.
I'll give you the numbers one at a time, because they are easier to take in that way.
Banking funds first. Two banking funds from two different fund houses average 48.4% overlap. The most similar pair reached 55.9%.
Sit with that second figure. Banking funds are sectoral, so they sit squarely inside the new rule. Under one roof they now have a hard 50% ceiling. Under two different roofs, nearly 56% — and none of it is a breach, because nobody is measuring it.
Large-cap funds average 45%.
And remember, large caps were left out of the comparison entirely.
I don't think that was an oversight. I think it was the regulator quietly conceding the arithmetic.
Why does this happen? (Nobody is cheating)
This part matters, because it is easy to read those numbers and assume somebody is being lazy or dishonest. Mostly they are not.
SEBI itself requires a large-cap fund to keep at least 80% of its money in India's hundred biggest listed companies. That is a sensible rule.
But look at what it means. Every large-cap manager in the country is choosing from the same hundred names. Most then narrow further, to the thirty or forty they have real conviction in.
When two managers each pick thirty stocks from the same hundred, they are going to hold a lot of the same stocks. Neither of them did anything wrong. There simply aren't that many to choose from.
Banking is tighter still. India has fewer than 45 listed banks of any real size.
So high overlap in these categories isn't misconduct. It is a consequence of a small starting list. And rather than cap something that can't really be avoided, the regulator stopped counting it — which leaves India's most widely held equity category with no overlap number published anywhere.
If you want confirmation that the size of the starting list explains most of this, look at the other end of the same data. International funds average 4.7% overlap. Thousands of companies to choose from, and two funds end up almost entirely different.
One honest note: Value Research and SEBI measure overlap slightly differently — one is a single-date snapshot, the other a quarterly average. Same picture, not the same arithmetic. And all of it is history, not a forecast.
Am I paying extra for this?
No. And I want to settle this one properly, because it is the most common thing said about overlap and it is simply wrong.
Holding six funds does not cost more than holding three. A fund's expense ratio is a percentage of the money sitting in that fund. It is not a charge per fund. Split the same amount across more schemes and the rate you pay does not change. Anyone telling you otherwise hasn't done the sum.
The problem isn't how much you are paying. It is what you are getting for it.
When a large part of your second fund is the same companies at similar weights as your first, you are paying an active manager to pick stocks you already owned.
What does it do to my risk?
Let me be careful here, because this is where writing about overlap tends to overreach.
Overlap is not the same thing as everything falling together.
If the market drops on a bad week, your two large-cap funds will both go red even if they held completely different stocks. That is just the market. No amount of fund-shuffling protects you from it, and anyone promising otherwise is selling something.
What overlap does is sit on top of that. Those two funds share a lot of the same names. So your real exposure to a handful of companies is larger than the number of funds in your app suggests.
You think you have spread your money across two positions. A good part of it is one position, sized bigger than you would have chosen if anyone had shown you the number.
That is the useful thing about overlap, honestly. It is the part of hidden concentration you can actually measure.
And when we rebuild a portfolio around how its funds move together, rather than which ones look best on a returns table, the change shows up where people feel it. Across the eleven portfolios we have run through our backtest engine, the average time spent climbing back from a fall went from 403 days to 141 days.
Modelled on historical data, shown for illustration only. Past performance is not indicative of future returns.
If you take one number from this whole article, take that one. Four hundred days of opening your app and seeing red, against a hundred and forty.
Three things worth remembering
-
The new rule stops one fund house from selling you the same portfolio twice. It does not stop two fund houses from doing it, and it does not count large caps at all.
-
High overlap in large-cap and banking funds is not anybody cheating. The list of companies to choose from is short, so everyone ends up holding much the same ones.
-
Compliance is measured one fund at a time. Risk is experienced across your whole portfolio. Nobody is paid to reconcile the two.
What should I actually check?
Do this yourself. You do not need me for it.
Every fund house now publishes its overlap numbers online each month. Full holdings for every scheme sit on AMFI's site. And public tools will compare two of your funds stock by stock, in a few minutes, at no charge.
Go and look. Then run through these:
-
Do you hold two or more equity funds in the same category, from different fund houses?
-
Has anyone ever shown you one number for how much of your total portfolio is duplicated — rather than a factsheet per fund?
-
Did you add a second fund because the first had a bad year, and then keep both?
-
Is most of your money in large-cap, banking or a popular theme — the categories with the shortest lists to choose from?
-
Have you assumed that because each fund you own is SEBI-compliant, the combination of them must be too?
That last one is the quiet mistake. It is an entirely reasonable assumption to make. It is just not how any of this is regulated.
If you want your own number
A Portfolio X-Ray works out the overlap across everything you hold — every fund, every fund house, down to the stock level. Then it does the part no public tool does. It shows you what that duplication has done to how far your portfolio falls, and how long it takes to come back.
Not which funds are good or bad. Whether the thing underneath them can get you through a bad year.
Twenty minutes, on a screen share. No sales pitch, just diagnostics.
aspyrapartners.com/portfolio-xray
Sources and disclosures
Regulation: SEBI Circular No. HO/24/13/15(2)2026-IMD-RAC4/I/5764/2026, "Categorization and Rationalization of Mutual Fund Schemes," February 26, 2026 — clause 2.6.3.5 (the 50% sectoral/thematic overlap cap and the large-cap exclusion), 2.6.3.4 (Value and Contra), 2.6.3.6 (quarterly computation on the average of daily values), 2.6.3.7 and 2.6.3.8 (three-year window, staged glide path, mandatory merger), 2.6.8 (monthly category-wise overlap disclosure on AMC websites). The circular does not use the words "same fund house"; that reading follows from the obligation resting on "Mutual Funds shall ensure," from the impossibility of computing daily overlap against a competitor's undisclosed portfolio, and from the remedy being a merger.
Overlap data and the starting-list explanation: Value Research, "You chose 2 funds. You may have bought the same stocks twice," May 4, 2026 — portfolio data as of March 31, 2026, Value Research categories, weighted stock-level overlap between same-category pairs drawn from different fund houses.
Aspyra backtest figures are internal, modelled and illustrative, across eleven portfolios.
Mutual Fund investments are subject to market risks, read all scheme related documents carefully.
Aspyra Partners LLP is an AMFI-registered Mutual Fund Distributor. ARN-343632. This article is educational content and does not constitute a recommendation to buy, sell, or hold any specific scheme. Backtest and model results shown are hypothetical and for illustration only. Past performance is not indicative of future returns.