The headline
Over the ten years to 30 September 2026, gold has done better than Indian equities.
| Gold ETF | Nifty 50, with dividends (est.) | |
|---|---|---|
| 1 year | 26.85% | −6.91% |
| 3 years (a year) | 35.12% | 6.17% |
| 5 years (a year) | 25.08% | 6.48% |
| 7 years (a year) | 20.49% | 11.60% |
| 10 years (a year) | 15.71% | 11.53% |
| ₹1 lakh over 10 years became | ₹4.30 lakh | ₹2.98 lakh |
The gold figures are for Nippon India ETF Gold BeES, the oldest gold ETF in India, using its market price, so its expense ratio is already deducted. The Nifty figures are the price index plus NSE's published dividend yield, our estimate of what an index fund earns before costs.
On those numbers, gold won every period. That is an unusual result, and it rests on one unusual year.
Year by year
| Year | Gold ETF | Nifty 50 (est., with dividends) | Ahead |
|---|---|---|---|
| 2017 | 2.43% | 30.11% | Nifty |
| 2018 | 6.67% | 4.39% | Gold |
| 2019 | 22.95% | 13.42% | Gold |
| 2020 | 27.00% | 16.55% | Gold |
| 2021 | −5.19% | 25.48% | Nifty |
| 2022 | 12.80% | 5.67% | Gold |
| 2023 | 14.68% | 21.68% | Nifty |
| 2024 | 19.44% | 10.16% | Gold |
| 2025 | 71.81% | 11.96% | Gold |
| 2026 to 30 Sep | 10.43% | −12.61% | Gold |
Gold's 71.81% rise in 2025 is the story. Take it out, measuring from 30 September 2016 to 31 December 2024, and the result reverses: gold returned about 10.4% a year, the Nifty about 14.4%. Before 2025, equities had the better decade.
Much of gold's rupee return also came from the rupee itself. Gold is priced in dollars, so a weaker rupee lifts its rupee price. Our gold and silver price post estimates that about half of gold's rise in the past year was the rupee falling against the dollar.
The other half of the comparison: risk
Returns are only half of the comparison. The other half is how each behaved when things went wrong.
- The Nifty's worst fall in these ten years was 38.4%, from 14 January to 23 March 2020. On 1 October 2026 it was 14.8% below its January 2026 peak.
- Gold's worst fall was 24.4%, from its peak on 29 January 2026 to 23 March 2026. On 1 October 2026 it was still 17.1% below that peak.
Gold is not a low-risk asset. It went nowhere for long stretches in earlier decades, and its sharpest fall in this record came only this year. What it has done is move differently from equities. In 2020 and again in 2026, gold rose while the Nifty fell. That is the case for holding some of each.
What a mix did
We tested the simplest version: half in the gold ETF, half in the Nifty 50, rebalanced back to 50:50 on the last day of each year. Over the ten years to 30 September 2026 it returned 14.18% a year, more than the Nifty alone and close to gold alone, while every year one half cushioned the other.
That figure leans on gold's exceptional 2025, so treat it as an illustration of rebalancing, not as a forecast. Our guide on debt and gold as shock absorbers covers how much to hold, and rebalancing with new money instead of selling shows how to keep a mix on target without triggering tax.
Should you switch to gold now?
The past ten years answer a different question. Three cautions apply:
- Gold has no earnings. An equity index grows with company profits and pays dividends. Gold's price depends on what the next buyer pays, which makes its long-run return harder to anchor.
- Big years tend to be followed by quiet ones. Gold's 2011 peak in dollar terms was not passed for years afterwards. Nothing guarantees a repeat, but buying after a 72% year is buying at a high starting price.
- The Nifty is cheaper than it has been for years. Its P/E is near the bottom of its range since 2021; see the Nifty P/E page.
If you hold gold, a fund or ETF avoids making charges and purity questions. Gold funds and ETFs and the SGB vs gold ETF calculator cover the options, and gold fund returns compares the funds.
Notes on the data
- Gold: Nippon India ETF Gold BeES closing prices. Two days of obviously wrong prices around its December 2019 unit split are excluded from the drawdown figures.
- An ETF tracks the domestic gold price, which includes import duty, so its return can differ from the international spot price converted to rupees. Over the past year the ETF rose more than spot.
- Nifty 50: price index from NSE, with our estimate of reinvested dividends.
Sources. Index data from NSE Indices; ETF prices as traded on NSE; calculations by WealthTicker.
This article is for education, not investment advice. Past returns do not predict future ones.
Frequently asked questions
Has gold beaten the Nifty 50 over 10 years?
Yes, as of 30 September 2026. Nippon India ETF Gold BeES returned 15.71% a year over ten years against an estimated 11.53% for the Nifty 50 with dividends. ₹1 lakh became about ₹4.30 lakh in gold and about ₹2.98 lakh in the Nifty.
Was gold ahead of the Nifty before 2025?
No. From 30 September 2016 to the end of 2024, the gold ETF returned about 10.4% a year and the Nifty 50 about 14.4% with dividends. Gold's 71.8% rise in 2025 is what put it in front.
How much gold should be in a portfolio?
There is no single right figure. Many planners suggest 5–15% of a portfolio, held to cushion equity falls rather than to chase returns. In our data a 50:50 mix of gold and the Nifty, rebalanced every December, returned 14.18% a year over ten years.
This is commentary on published data, not investment advice. WealthTicker is not a SEBI-registered adviser or distributor. Figures are as of the dates stated and can be revised by their source.
Keep reading
How to build an all-weather portfolio from scratch
Equity, debt and gold rarely fall together. Nine years of Indian data on Nifty 50, gold and debt funds, and how to build a portfolio that survives any year.
Active vs passive investing: which wins in the long run?
To 1 October 2026, most active large-cap funds beat Nifty 50 index funds over 5 years, but the lead shrank over 10. What the data shows, and what it doesn't.
Asset allocation by age: a starting point for Indians
How much in equity, debt and gold at 30, 45 and 60? A rule-of-thumb split, why age is a weak guide, and a worked rebalancing example in rupees.
