Nifty 50 Rolling Returns

Nifty 50 Rolling Returns by Holding Period

The same history, summarised. Notice what changes as the holding period lengthens: the average barely moves, but the spread between the best and worst outcome collapses.

Average, minimum and maximum annualised return of the Nifty 50 Total Return Index at each holding period, with the share of windows that ended in a loss.
Holding Period	Windows	Average CAGR	Worst	Best	Windows Negative
1 Year	314	+15.7%	-51.7% Nov 2008	+97.3% Apr 2004	24.8%
3 Year	290	+15.1%	-12.6% Mar 2003	+59.5% Apr 2006	6.6%
5 Year	266	+15.4%	+0.2% Oct 2012	+46.7% Oct 2007	0.0%
10 Year	206	+14.1%	+6.4% Mar 2020	+22.1% Apr 2013	0.0%

Dates under the worst and best figures are the month in which that window ended.

What happens after the Nifty goes nowhere for two years?

Since 30 Jun 1999 there have been 18 two-year stretches in which the Nifty 50 TRI finished within ±2% of where it started — two full years of nothing. 16 of them have since run their course. The last 2 are still open — that is where we are now.

Every 24-month window in which the Nifty 50 Total Return Index returned within plus or minus 2%, and the index's cumulative and annualised return over the one, three and five years that followed each window.
Window Start	Window End	2-Yr Return	Next 1 Yr	Next 3 Yrs	Next 5 Yrs	3-Yr CAGR	5-Yr CAGR
Jun 2007	Jun 2009	+1.4%	+25.3%	+27.5%	+88.2%	+8.4%	+13.5%
Sep 2009	Sep 2011	-0.6%	+16.9%	+67.2%	+85.1%	+18.7%	+13.1%
Nov 2009	Nov 2011	-1.9%	+23.3%	+84.4%	+81.0%	+22.6%	+12.6%
Apr 2010	Apr 2012	+1.7%	+14.5%	+61.9%	+88.7%	+17.4%	+13.5%
May 2010	May 2012	-0.9%	+23.2%	+77.5%	+107.5%	+21.1%	+15.7%
Jun 2010	Jun 2012	+1.7%	+12.0%	+64.2%	+91.5%	+18.0%	+13.9%
Jul 2010	Jul 2012	-0.2%	+11.2%	+69.1%	+104.8%	+19.1%	+15.4%
Aug 2010	Aug 2012	-0.2%	+5.3%	+56.9%	+100.4%	+16.2%	+14.9%
Dec 2010	Dec 2012	-1.4%	+8.1%	+39.3%	+89.4%	+11.7%	+13.6%
Mar 2011	Mar 2013	-0.2%	+19.5%	+41.1%	+89.5%	+12.2%	+13.6%
Nov 2014	Nov 2016	-1.9%	+25.9%	+52.5%	+119.7%	+15.1%	+17.0%
Dec 2014	Dec 2016	+1.2%	+30.3%	+54.7%	+125.6%	+15.6%	+17.7%
Jan 2015	Jan 2017	-0.4%	+30.5%	+45.4%	+115.6%	+13.3%	+16.6%
Jun 2018	Jun 2020	-1.5%	+54.6%	+93.2%	+163.1%	+24.5%	+21.3%
Jul 2018	Jul 2020	-0.2%	+44.2%	+84.8%	+137.5%	+22.7%	+18.9%
Aug 2018	Aug 2020	-0.2%	+52.2%	+75.3%	+127.9%	+20.6%	+17.9%
Jun 2024	Jun 2026 → we are here	+1.7%	?	?	?	?	?
Jul 2024	Jul 2026 → we are here	+0.1%	?	?	?	?	?
Average of the 16 completed windows			+24.8%	+62.2%	+107.2%	+17.3%	+15.6%

“Next 1 / 3 / 5 Yrs” are cumulative total returns measured from the window’s end. The two CAGR columns annualise the same three- and five-year figures. A ? means that future has not happened yet. Shading is scaled within each column, because a 25% gain over one year and a 25% gain over five are not the same event.

Before you read too much into that table

That average row is the kind of number that gets screenshotted, so here is what is wrong with it.

The rows are not independent. They overlap heavily. Nine of them start within ten months of each other in 2009–2011 and describe substantially the same two years of market history. Counted properly, 18 rows collapse to roughly five distinct episodes: 2007–09, 2009–13, 2014–17, 2018–20, and the one we are in now.
Five episodes is not a sample. You would not draw a conclusion from five coin flips, and an average taken across five overlapping market cycles is not more reliable for being written to one decimal place.
Every one resolved inside a rising market. The Nifty compounded strongly across this entire 27-year period. Any window you pick out of it inherits that drift. A flat two years inside a long bull run is a different animal from a flat two years inside a decade-long stagnation — India has not had one of those in this data, but other markets have.
Nothing here is a forecast. A flat stretch tells you that prices have not kept up with whatever earnings did in the meantime, which is a statement about valuation, not about timing. It does not tell you the wait is over.

Nifty TRI vs the price index — and why this page uses TRI

The “Nifty 50” quoted on the news is a price index. It tracks the share prices of its 50 constituents and nothing else. When a company pays a dividend, its share price drops by roughly the dividend on the ex-date, and the price index dutifully records that drop — but it never records the cash that left the company and landed in shareholders’ hands. Over one day that is a rounding error. Over 27 years it is not.

The Total Return Index (TRI) fixes this by assuming every dividend is immediately reinvested back into the index. Indian large caps have yielded somewhere around 1% to 1.5% a year over this period, so the TRI compounds roughly that much faster than the price index — a gap that widens into a substantial difference across a ten-year holding period.

Every figure on this page is computed on the TRI. That is the honest comparison for anyone holding an index fund or ETF, because those funds receive the dividends too. It also means the returns here will look higher than a chart drawn from the headline Nifty number — not because the maths is generous, but because the headline number leaves money out.

Frequently Asked Questions

What is a rolling return?

A rolling return measures the return you would have earned over a fixed holding period, starting from every possible entry point rather than from one convenient date. A 5-year rolling return series answers "if I had invested at any month end in the last 27 years and held for exactly five years, what would I have annualised?" — once for each of those months. It removes the single biggest distortion in a point-to-point return, which is that the answer depends almost entirely on which start date someone picked.

What is the average 5-year return of the Nifty 50?

Across all 266 five-year windows in the Nifty 50 Total Return Index since 30 Jun 1999, the average annualised return is 15.4%. The worst of those windows still returned 0.2% a year and the best 46.7%. These are total-return figures, so they include dividends. They are a record of what happened, not a forecast.

What is the average 10-year return of the Nifty 50?

The average annualised return across every 10-year window is 14.1%, and the range is far tighter than at shorter horizons: the worst 10-year stretch on record returned 6.4% a year and the best 22.1%. No 10-year window in this series ended below zero. That is a statement about one index over one 27-year run of Indian history, not a guarantee about the next ten years.

What is the Nifty 50 Total Return Index (TRI)?

The Nifty 50 you see quoted on the news is a price index: it tracks share prices only and drops the dividends companies pay out. The Total Return Index assumes every dividend is reinvested back into the index on its ex-date. Because Indian large caps have yielded roughly 1% to 1.5% a year, the TRI compounds meaningfully faster than the price index over long horizons. Every figure on this page uses the TRI, which is the right comparison for an index fund or ETF that also reinvests its dividends.

Does a flat market predict good returns afterwards?

Historically, on average, yes — and honestly, the sample is too small to call it a rule. In 16 completed cases where the Nifty TRI went essentially nowhere for two years, the following year averaged +24.8% and the following five years averaged +107.2% in total. But those windows overlap heavily and cluster into only about five genuinely independent episodes, and every one of them resolved inside a market that was rising over the long run. A flat stretch tells you the index is cheaper than it was; it does not tell you when that changes.

How often is this page updated?

The underlying Nifty 50 TRI series is refreshed daily after the close, and the rolling windows on this page are recomputed from it. Windows are sampled at month ends, so the tables and the chart gain a new point once a month rather than once a day.

Rolling Annualised Return by Holding Period