What is the Nifty 50 TRI?
NSE Indices publishes the Nifty 50 in two main versions. The price index, the familiar Nifty level, moves only with the prices of its 50 stocks. The Total Return Index adds back what those stocks pay out. In NSE Indices' words, dividend payouts "are indexed on ex-date", and the indexed dividends are "then reinvested in the index to give TR Index".
Because the dividends keep compounding inside it, the TRI pulls further ahead of the price index every year. NSE Indices publishes its daily values in the historical data reports on niftyindices.com. Most broker apps and market tickers show only the price index.
How much do the dividends add?
I worked out both indices' calendar-year returns from NSE Indices' daily closes. In each of the nine years from 2017 to 2025 the TRI came out ahead, by 1.24 to 1.62 points.
| Year | Nifty 50 | Nifty 50 TRI | Gap (points) |
|---|---|---|---|
| 2025 | +10.51% | +11.88% | 1.37 |
| 2024 | +8.80% | +10.09% | 1.29 |
| 2023 | +20.03% | +21.30% | 1.27 |
| 2022 | +4.33% | +5.69% | 1.36 |
| 2021 | +24.12% | +25.59% | 1.47 |
| 2020 | +14.90% | +16.14% | 1.24 |
| 2019 | +12.02% | +13.48% | 1.46 |
| 2018 | +3.15% | +4.65% | 1.50 |
| 2017 | +28.65% | +30.27% | 1.62 |
Over ten years the points compound. ₹1 lakh in the price index on 14 October 2016 would have been ₹2.68 lakh on 15 October 2026, a return of 10.3% a year. In the TRI it would have been ₹3.02 lakh, or 11.7% a year. The ₹34,000 or so between them is dividends, reinvested.
Why the price index flatters your portfolio
The dividends your companies pay land in your bank account, and a full XIRR or time-weighted return counts them. The price index counts none, so it hands you the index's dividend yield before a single stock has been picked.
For a portfolio that runs close to the index, a point and a bit a year is the difference between "I beat the market" and "I roughly matched it". The benchmark is the check on the whole process: it is how you find out whether the hours you put into research each quarter beat an index fund that needs none. A benchmark that flatters you by a point a year can hide the answer for years.
The fix is consistency. If your return counts dividends, compare it with the TRI. If you are comparing price moves only, say a stock's price chart against the index's, the price index is fine on both sides.
Mutual funds lost this head start in 2018. SEBI's circular of 4 January 2018 required every scheme's performance to be "benchmarked to the Total Return variant of the Index" from 1 February 2018, because the price variant leaves out dividends. A portfolio of your own deserves the same yardstick.
My own portfolio against both
Over my tradebooks' window, 18 November 2021 to 15 October 2026 (4.9 years), the Nifty 50 made 5.40% a year and the Nifty 50 TRI 6.64%. My time-weighted return was 11.16%.
Against the price index, my picks beat the market by 5.76 points a year. Against the TRI, by 4.52. Both say the stocks did better than the index over those 4.9 years. But 1.24 points a year of the lead I would have claimed came from leaving the index's dividends out while my own number counted mine.
One caution about these numbers. The TRI reinvests every dividend with no tax and no costs, and my 11.16% is worked the same way: dividends before tax, trades before brokerage, STT and stamp duty. So the comparison is fair, and both sit a little above what an investor keeps after tax and costs.
The comparison uses the time-weighted return on purpose. My XIRR, 10.32%, mixes in when I added money, a choice the index never had to make. XIRR vs CAGR explains the difference, and what XIRR means covers what a single XIRR can and can't tell you.
When the price index is enough
Over a few weeks or months the gap is a fraction of a point, so for short windows the price index won't mislead you much.
For reading the market itself (a correction, a rally, a level) the price index is what everyone quotes, and that is fine.