The test
One-year returns of gold, silver and equity are noisy. Five years is a more honest horizon for a long-term holder. This post starts a five-year window on every trading day from 30 August 2000 (when our rupee gold and silver series begins) and measures the compound annual return to the same date five years later, for gold and silver in rupees per gram and for the Nifty 50 price index. Windows overlap, so the counts are large but they are not independent observations.
| 5-year window (a year) | Gold | Silver | Nifty 50 |
|---|---|---|---|
| Windows counted | 5,297 | 5,297 | 5,297 |
| Worst | −3.1% | −11.5% | −2.3% |
| Lower quartile | +8.2% | +4.9% | +9.3% |
| Median | +13.7% | +14.0% | +12.4% |
| Upper quartile | +18.9% | +21.4% | +16.9% |
| Best | +31.0% | +41.4% | +44.5% |
| Share of windows below 0% | 7% | 18% | 1% |
| Share of windows below 8% | 24% | 31% | 20% |
| Share of windows at 15% or more | 43% | 46% | 33% |
What the windows show
- Medians: gold +13.7%, silver +14.0%, Nifty 50 +12.4% a year.
- Losing five years: 7% of gold windows, 18% of silver and 1% of the Nifty 50.
- Worst window: gold −3.1% (from 12 December 2012), silver −11.5% (from 25 April 2011), Nifty 50 −2.3% (from 23 March 2015).
- Best window: gold +31.0% (from 8 March 2021), silver +41.4% (from 26 January 2021), Nifty 50 +44.5% (from 29 October 2002).
Silver's range between best and worst is about 1.6 times gold's. That is the cost of its extra swing: it is the metal that sometimes loses money for five years running and sometimes rises 5.7 times in five.
How to use this
- A five-year horizon has not been a guarantee for any of the three. The lower quartile line shows how poor a typical bad stretch was.
- The Nifty 50 figure excludes dividends and the metals pay none, so the comparison understates equities by their dividend yield.
- The start date matters more than the asset in the worst windows: the worst window for each began near a peak.
- For one-year patterns, see gold and silver by calendar month and gold against the Nifty 50 over ten years.
Source: WealthTicker daily gold and silver prices in rupees per gram and NSE Nifty 50 daily closes, 30 August 2000 to 21 October 2026; windows end on or before 21 October 2026.
Frequently asked questions
How often has gold lost money over five years?
In 7% of the 5,297 five-year windows that start on a day from August 2000 and end by 21 October 2026. The worst window returned −3.1% a year, starting 12 December 2012.
Has silver or gold given better five-year returns?
Gold's median five-year return was 13.7% a year against 14.0% for silver, but silver's best window (+41.4% a year) was far above gold's best (+31.0%), and its worst (−11.5%) far below.
How does the Nifty 50 compare over five years?
The median five-year Nifty 50 price return was 12.4% a year, and 1% of windows were negative. Dividends are not included.
This is commentary on published data, not investment advice. WealthTicker is not a SEBI-registered adviser or distributor. Figures are WealthTicker’s own calculations from public data — AMFI’s fund NAVs and disclosures, NSE’s index and FII/DII files, and the fund houses’ monthly portfolios (how we calculate). They are as of the dates stated and can be revised by their source.
Keep reading
Gold and silver: how long each took to regain its peak
Gold took 2,474 days to regain its 2012 peak in rupees; silver took 3,391 days. Today gold is 18.0% and silver 44.9% below their January 2026 highs.
Gold and silver in every Samvat since 2007
Gold rose in 13 of 18 Diwali-to-Diwali years since 2007, silver in 10. Gold's median year was +15.1%, silver's +12.9%, but silver's swings were far wider.
Gold vs Nifty 50: ten years of returns compared
A gold ETF returned 15.7% a year over the ten years to September 2026, against about 11.5% for the Nifty 50 with dividends. Why one year explains most of it.