The Nifty 50 Inclusion Trade Does It Still Work

TITLE: The NIFTY 50 Inclusion Trade: Does It Still Work? | The Long & The Short Ep. 40 CHANNEL: In The Money by Zerodha DATE: 2026-06-24 ---TRANSCRIPT--- Welcome to the long and the short. Have you ever wondered as to what happens to a stock’s price after it gets included in Nifty 50? Well, researchers have been studying this question on the S&P 500 for 40 years now. And what they found for the US market should make us think before we assume the same for Nifty 50. The question is, when a stock gets added to a major index like Nifty 50, is there an edge in trading it? On paper, the logic is simple. Passive have to buy the new entrant. Demand should go up. Price should follow. So, if you buy right before the funds do, you get to ride the wave. In the US, that trade worked until it didn’t. The abnormal return from an S&P 500 addition went down from 7% in 1990s to under 1% in the last decade. The edge got discovered, it got traded, and [music] eventually it got up the way. Quite expected for a market like S&P 500. So, where does that leave Nifty 50? So, I went back through the Nifty 50 inclusions from 2015 to 2025, 33 events, and tracked what happened to those stocks before and after they were added. What I found is a bit of a mixed bag of results, and the more interesting story might actually be what happens before the announcement. We will cover this topic in two parts. This is part one, where we look at inclusions. We’ll do part two on exclusions separately. So, without further ado, let’s get into it. Some disclaimers before I get started with the next section. The examples and ideas shared in this episode are strictly for educational and illustrative purposes only. Nothing discussed here should be construed as a recommendation, investment advice, or a solicitation to trade. Trading in stocks or derivatives involves significant risk and can result in complete loss of capital. The examples and data presented here are meant to explain market behavior, not to suggest that similar outcomes will occur in future. Lastly, as far as the specific backtests are concerned in this episode, I have excluded all merger and demerger stocks that got included in Nifty 50. I have only taken companies that were included in the periodic index re-constitution review. Like I always do, before we get into testing the idea, we need to first understand how Nifty 50 rebalancing works. That is, how [music] do new stocks get in and get out of the index and at what frequency? You see, NSE reviews its constituents twice a year, in March and September. Data is assessed for the 6 months ending Jan 31st and July 31st, and any changes are implemented from the first working day after [music] the March and September F&O expiry with 4 weeks notice given to the market. To be eligible for inclusion, a stock needs to clear a few filters. One, it needs to have been listed for at least 1 month, though in practice any stock that makes into Nifty 50 would usually have years of listing history behind it, but there are exceptions. Next, it needs to be available in the futures and options segment. Then, it needs to pass a liquidity test called impact cost, which is a measure of how much a large trade in the stock would move its price. For Nifty 50, that threshold is 0.5% or less, measured on a portfolio of 10 crore rupees over the past 6 months for 90% of observations. Beyond that, it’s about size. The index is weighted by free float market capitalization, meaning only the shares that are actually available for public trading, not the ones locked up with promoters. The 50 largest eligible companies by this measure make it in. Also note that this process is not purely algorithmic. The final call rests with a committee at NSE, which means there is an element of human discretion involved. Now, when a stock gets announced for inclusion, something specific happens, rather has to happen. Every passive fund or ETF that tracks the Nifty 50 has to buy that stock because they need to hold all 50 constituents in proportion to their index weights. And they have to do it by a specific date. And all of that is known and hence there will be a predictable burst of buying, which immediately raises the question, if you know buying is coming, can you get ahead of it and profit? That’s the hypothesis we tested. We looked at 33 Nifty 50 inclusion events between 2015 and 2025 and tracked what happened to those stocks both before the announcement and after. One last note on the data set, we excluded inclusions that were triggered by corporate actions such as mergers, spin-offs, or delisting and only looked at stocks that came in through the regular semi-annual rebalancing process. So, here’s the hypothesis that we used to structure the test. Now, there are two dates that matter here. The announcement date, when NSE tells the market which stocks are coming in or going out roughly 4 weeks before the change takes effect, and the effective date, the day the stock officially enters the index. We anchored our study to the effective date. That means our T0 is the day when the stock enters the index and our forward returns measured what happened after that [music] point. From there, we measured forward returns at six horizons, 1 week, 2 weeks, [music] 1 month, three months, six months, and 12 months. We also looked backwards, how the stock performed in the one week, two weeks, one month, three months, six months, and 12 months before the effective date. The forward returns tell us whether there’s an edge [music] in buying after inclusion. The backward returns tell us how the stock performed prior to its inclusion in the index. A few things to note about the data. The returns are price returns, dividends are not included. And as I mentioned earlier, we filtered to include only the stocks that came in through the regular semi-annual rebalancing, not the ones which came through corporate action events. In total, that gives us 33 inclusion events between 2015 and 2025. Let’s start with the forward returns, which means what happened to the stock after it entered the index. We buy on the close of effective date and measure the return over the following period. That’s it. Here’s what the data shows us across all six horizons. The first thing that jumps out is the two-week window. A win rate of 78.8% meaning nearly four out of the five inclusion stocks were positive two weeks after entering the index. The mean return is 2.21% and the median is at 3.07%. When mean and median are both positive and close to each other, it tells you the result isn’t being distorted by a few big outliers. That’s a statistically clean signal, I would say. The one-week window, as you can see, is weaker. 54.5% win rate, barely above a coin flip. Two weeks is clearly the sweet spot. Why two weeks? Well, think back to what we discussed earlier. When a stock enters the index, every passive fund and ETF tracking the Nifty 50 has to buy it. That buying doesn’t happen in one day. Maybe it gets absorbed over roughly 2 weeks. That’s how it looks like. Once that demand is fulfilled, the index dynamics take over and the stock is just another constituent being held in proportion to its weight. The buying pressure is gone. Now, let’s look at what happens after 2 weeks. At 3 months, the mean is still positive at 3.19%, but the median has flipped to negative at 0.07%. Now, that’s a signal to pay attention to because median tells you what the typical stock did and the typical stock was flat to slightly negative. In other words, a few big winners are pulling the average up. At 6 months, the median is down to minus 2.28%. The average included stock was actually losing money at that point of time. And the standard deviation tells the rest of the story. At 2 weeks, it’s 5.6% by 3 months, it’s nearly 15%. By 12 months, it’s almost 30%. The outcomes are becoming increasingly dispersed. Some stocks do very well, many don’t. So, what does all of this tell us? Yes, there is an inclusion effect, but it’s very short-lived. The 2-week window is where it shows up most cleanly. Beyond that, you’re just holding an index constituent stock and nothing much. So, we just saw that the 2-week window shows the clearest signal. Using that as a reference, what does the returns look like if you traded every inclusion over the past 10 years? Starting with 1 lakh in 2015, the strategy ends at 1.22 lakhs by 2025. A total return of 22.25% with a max drawdown of 14.3% giving a return to max drawdown of 1.56. Do note, we are compounding the capital here. Example, if 1 lakh becomes 1.1 lakh after the first trade, and then there are two inclusions in the next, we split 1.1 lakh equally and invest in both stocks and so on. Now, before you react and sneer at the 22.5% number, this is not a buy and hold strategy. You are invested just for 2 weeks at a time, roughly twice a year. Your capital sits idle for most of the year, so comparing this to an annual return benchmark isn’t logically right. What you’re really doing here is every time a stock enters the Nifty 50, you are buying it for 2 weeks and asking if the returns generated are worth the risk. And the answer, on an average, is yes, but with some important caveats. Look at the equity curve. It is not smooth. The worst drawdown of 14.3% hit on the very second trade in May 2015. The strategy then spent a long time recovering before making new highs. And the gains are not evenly distributed. Certain inclusion cycles contributed far more than others. The return to max drawdown of 1.56 is a reasonable number for a simple, event-driven strategy, not spectacular, but not so weak, either. It tells you the return you earned was about 1 and 1/2 times the pain you had to absorb to get it. Here’s my take on it. The edge is real, but quite modest. It works more often than not in the 2-week window, but the absolute return over 10 years is limited simply because the strategy is active only for a few weeks each year. So far, we’ve spoken about forward returns, that is, what happens after inclusion. Now, let’s flip that question. Instead of asking what happened after a stock entered the index, let’s ask what happened before. The backward return is simply the return the stock had already delivered in the period leading up to its inclusion date. As you can see, the numbers here are striking, and they tell a very different story compared to the forward returns. Go back 12 months before inclusion and the mean return is 58%. The median is 43% and the win rate is 93.9% meaning every single stock that entered Nifty 50 had already been positive in the year before it was added. Go back 6 months and you get a mean return of 19.94% median of 9.68% and the win rate of 81.8%. 3 months, you have a mean of 7.36% a median of 3.2 6% as you can see the pattern is clear. The further you go back, the stronger the performance. By the time the committee already announced the inclusion, the heavy lifting was already done. There’s one detail worth pointing out though. In the very short term, one week before inclusion, the mean return is actually slightly negative at minus 1.6%. The stock dips a little right before it officially enters. Possibly some profit taking or maybe just noise in this small sample set. So as you can see, there seems to be a bigger edge in predicting the future inclusions and holding on to them till T0. But before we get to that, is there a logical fallacy in this pre-inclusion returns test? Can you guess what? The back with returns we just looked at are great on paper. Stocks are up 58% on average in the 12 months before they enter the index. A win rate of 93.9%. It’s tempting to read that and think if I can just identify the next inclusion early enough, I can ride that wave and retire, if not in Maldives, at least in Goa. But before you get too excited, there’s something important to acknowledge. Those returns aren’t surprising. They are almost guaranteed by the flawed construction of the test. [music] You see, as I had shared earlier, the eligibility criteria for Nifty 50 is free float market capitalization. A stock gets added because it has grown large enough, which means the price has to go up for it to become eligible for inclusion. The backward return and the selection criteria are essentially measuring the same thing. This is called sample truncation bias. We only observe stocks that cleared the inclusion threshold, so we can’t see the counterfactual of what didn’t get in. So, the strong backward returns aren’t a discovery about what happens before inclusion. They are a feature of the stocks that end up in our data set to begin with. So, the question isn’t why did these stocks go up before inclusion? The question is, can I identify which stocks are going to be included before the announcement, and is there still an edge left after accounting for that? And that’s a harder problem to solve, but not an impossible one. For Nifty 50, the methodology is public and rule-based. The index is built from the top 50 stocks by free float market cap within the Nifty 500 universe, which means the Nifty Next [music] 50 is essentially the waiting room for stocks. So, stocks within Nifty 50, which are at the top, are your natural candidates for the next inclusion cycle. The date cutoff is January 31st and July 31st. So, 6 months before each cutoff, [music] you can start tracking which of the Nifty Next 50 stocks are closing the gap on the bottom of the Nifty 50 from a free float market cap standpoint. But, there are real challenges. First, you can only identify the pool of candidates, but not always the exact stock or the exact cycle. [music] A stock can hover on the boundary for multiple review periods before finally crossing over. Second, [music] the committee retains discretion. The methodology is a strong guide, but it’s not a guarantee. Human judgment can and does intervene. Third and the most important one is that you’re not alone in doing this. All of this information is noble from public data and other participants are also watching the same rankings. The more efficiently this information gets priced in ahead of the announcement, the smaller the edge that remains by the time you act. By the way, there is one AMC in India which has come up with a large cap fund and in their fact sheet, they had mentioned something similar to what we did here to beat the index. It’s in fact quite an interesting move, but [music] can you guess the fund house? Let me know in the comments. Back to the test. Now, these are real challenges and I totally agree, but for the curious and the patient, it’s still a worthwhile experiment. I think the potential edge is still large enough to justify the effort. Let’s now put both pre and post inclusion stories side by side and observe them. As you can see, the contrast is stark. In the 12 months after, it’s 4.81%. 6 months before is 19.94%. 6 months after is 1.55%. 3 months before is 7.36% and 3 months after is 3.19%. At every horizon beyond 2 weeks, the pre-inclusion return beats the post-inclusion returns. If you look at the medium, the story gets even clearer. Medium and long-term performance, especially the 3 to 12-month period, is dramatically stronger before inclusion. At 3 months, the pre-inclusion median is at plus 3.26%. Post-inclusion median comes down to minus 0.07%. At 6 months, 9.68% on pre-inclusion and minus 2.28% after. At 12 months, 43.45% before and 4.07% after. The median strips out all the outliers. It tells you how the typical included stock actually did. And as you can see, the typical included stock after entering the index delivered very little over the medium to long term. This brings [music] us to the central conclusion of this episode. If you are approaching this from a pre-inclusion angle, the data suggests the real edge lies in predicting which stocks are likely to enter the index 6 to 12 months ahead. That is a harder problem, but as we discussed, not an impossible one. The Nifty Next 50 gives you a structured starting point. The real work is in identifying the right candidates early enough. However, if you are approaching this from a post-inclusion angle, the data is clear. The edge exists, but it’s narrow. 2 weeks from the effective date is where it shows up most consistently. A win rate of 78.8% a positive mean and median and a relatively contained dispersion. Beyond 2 weeks, you are no longer trading an index event, you’re just holding a stock. A few more things worth calling out on the test design before we close. We have deliberately kept this simple. No complex filters, no optimization, just a clean look at what happened to these stocks across six time horizons before and after inclusion. One important limitation is that the returns in this study are raw price returns and not benchmark adjusted. We haven’t compared them against what the Nifty 50 itself returned over the same periods. If you want to properly isolate the inclusion effect to see whether these stocks actually outperform the index, not just whether they went up, you would want to compute the excess returns over the benchmark. And that’s the natural next step if you would like to take this analysis further and build something more rigorous around it. So yes, that brings me to the end of this episode on index inclusions. I hope you found this useful. We will study the exclusion in the next episode. If you have any questions, feel free to drop them in the comments. I’ll be happy to respond. Till then, take care and trade safe. I’ll be back soon with the next one.