How Global Gold Prices and the Rupee Affect Gold in India

The gold price an Indian buyer pays is three things multiplied together and several more added on. Knowing which one moved explains most of what looks mysterious about it.

Why does the gold price in India differ from the global price?

Meera has jewellery to buy in the next couple of years and has started paying attention to the price. She read that gold had a weak stretch globally, went to check what that meant for her, and found the rupee price had barely moved. On another occasion the opposite happened: nothing much in the international headlines, and the rate at her jeweller had climbed.

Neither of these is a mistake in the reporting. The gold price in India is not the world gold price in different units. It is the world price passed through an exchange rate and then a stack of local costs, and each of those layers moves on its own.

Once the layers are separated the behaviour stops being mysterious, and — more usefully — Meera can tell which layer is responsible for what she is looking at.

What makes up the gold price you pay in India?

The first layer is the international price, quoted per troy ounce in dollars. This is the number the headlines mean, it is set by global trading, and it is the one thing about gold that is genuinely a single world price.

The second layer is the exchange rate. Meera pays in rupees, so the dollar price has to be converted, and the conversion is not a formality — it is a second variable, moving mostly for its own reasons rather than gold's (measured further down this page), not a quantity guaranteed to be independent of it. This is the layer that explains her observation directly: if the international price eases while the rupee weakens by a similar amount, the two cancel and the rupee price sits still. If the international price holds and the rupee weakens, the rupee price rises with no gold story behind it at all.

The third layer converts the unit, from troy ounces to the grams she will actually buy, which is arithmetic and changes nothing.

The fourth layer is everything local: the duties and taxes that apply on import and sale, and the premium the market charges over the landed cost, which varies with local demand and supply. These move too, sometimes sharply and sometimes by policy rather than by market.

The fifth layer applies only to Meera, because she is buying jewellery rather than metal: making charges, which are a cost of fabrication rather than of gold, and which she will not recover when she sells. That layer is examined in choosing the form of gold for a goal.

How does the rupee affect gold prices?

Most explanations stop at "the rupee matters". It is worth going one step further, because the direction of the effect surprises people.

A weakening rupee raises the rupee price of gold, all else equal. That means an Indian holder of gold has, embedded in the holding, a position that gains when the rupee falls — which is also the circumstance in which imported things generally become more expensive for her household. This is part of why gold has the reputation it has in India, and it is a real mechanism rather than a sentiment.

It is worth being careful about how far that goes. The relationship holds for the currency component specifically. It does not mean gold protects against Indian inflation generally, which is a different and much less reliable claim — domestic prices can rise for reasons that have nothing to do with the exchange rate, and gold's international price can fall at the same time and swamp the currency effect. A currency hedge is not an inflation hedge, and gold is more clearly the first than the second.

Does gold earn anything while you hold it?

The other structural fact worth having, because it separates gold from most of what else Meera might own.

A deposit pays interest. A bond pays a coupon. A share is a claim on a business that earns money and may distribute some of it. Each of those has an internal source of return that exists whether or not anybody wants to buy the asset from you.

Gold has none. It sits there. The entire return, over any period, is the difference between what Meera paid and what a future buyer will pay, less what it cost to hold and to transact. There is no income accruing in the background to cushion a flat decade.

This does not make gold a bad thing to own, and it is not an argument against it. It does mean the case for holding it has to be made on a different basis from the case for holding a productive asset, which is the subject of whether gold belongs in a long-term portfolio. And it means that a long stretch of a flat gold price is not a period of modest returns — it is a period of no return at all, minus costs.

How do you compare gold's returns fairly?

Because of the layers, gold comparisons go wrong easily, and usually in gold's favour.

The dates and the currency have to match on both sides. A gold return quoted in dollars against an Indian equity return quoted in rupees is not a comparison; it is two different questions. Costs have to be on both sides too — the spread between buying and selling price, storage or fund expenses, making charges where relevant, and the tax treatment, which differs by the form the gold is held in.

And the alternative's income has to be counted. Comparing gold's price change against a share index's price change omits dividends from one side only, which understates the alternative by a margin that compounds. This site's equity figures use a total return series precisely to avoid that error.

What can the gold price record tell you, and what not?

Over the 16.7 years covered by a gold ETF's NAV (physical metal held in Indian vaults — read here only for its return, never as a retail price) and the USD/INR exchange rate, the rupee gold price has compounded at +12.89% a year, while the rupee itself weakened against the dollar at +3.29% a year over its own longer record.

Those two rates alone cannot say how much of the rupee price's rise came from the metal and how much from the currency — that needs the international, US-dollar gold price as a third series, set against the same window. Splitting a compounding return into pieces has to be done with log returns, not by adding percentage changes directly: adding them manufactures a distorted leftover, because a return that compounds is not additive the way a log return is.

Done properly, the international price rose +294.46% in dollar terms over the gold ETF's own span while the rupee weakened +106.74% against the dollar, compounding to the ETF's own +656.02% rupee return. Splitting that return between the two: the metal's own dollar rally accounts for about 68% of it, the currency for about 36% — gold's own price is the larger driver of what Meera has actually paid, not the rupee, which runs against the shorthand that gold's rupee gains are mostly a currency story.

The rest — a residual of about -7.29% cumulatively over the period — is what neither series explains: India's import duty, GST and the ETF's own tracking and expense drag, bundled together because none of them is held here as its own series over time. Its size and direction are consistent with a cut to import duty at some point in the window, but which change, when, and by how much would need checking against the actual customs notification before this page named a date and a rate as fact — it does not hold that confirmation, so it states the residual without naming a cause for it. Splitting the residual further needs a duty-rate history, which this page does not have.

-16.36%+81.69%+179.74%+277.79%+375.83%International gold priceRupee vs dollarWedge (duty, GST, drag)4 January 201011 September 2026
How much of the rupee gold ETF's own cumulative return, as of every date since 2010, would be explained by the international dollar gold price alone, by the rupee's slide against the dollar alone, and by the leftover wedge (India's import duty, GST and the ETF's own tracking and expense drag) — computed with log returns, which add up exactly, rather than a method that manufactures a distorted residual on a compounding series.
Show these numbers as a table
YearInternational gold priceRupee vs dollarWedge (duty, GST, drag)
2010+27.15%-3.19%-1.33%
2012+49.84%+18.29%-1.00%
2014+5.92%+36.92%+2.66%
2016+2.89%+46.76%+2.62%
2018+14.37%+51.06%-2.00%
2020+69.55%+58.00%-1.32%
2022+63.39%+78.96%-3.31%
2024+136.29%+85.33%-11.57%
2026+294.46%+106.74%-7.29%

This page also cannot say how large making charges typically run, or how wide the buy-sell spread is. What it can do is tell her which questions to ask her jeweller, and that the answers are checkable.

What to take away

The rupee gold price is the international price converted at the exchange rate, with local duties, taxes and premium added, and making charges on top if the purchase is jewellery. Each layer moves independently, so a global headline does not describe what an Indian buyer faces.

The currency layer means gold carries a position that gains when the rupee weakens, which is a currency hedge and should not be over-read as an inflation hedge. Gold generates no income while held, so its whole return is the price a later buyer pays less the costs of holding and transacting. Compare it against alternatives in the same currency, over the same dates, with costs and the alternative's income on both sides.

Disclaimer

Educational content only. This is not personalised financial, investment or tax advice. Figures quoted are historical or illustrative and are not forecasts. Consult a qualified professional before acting on anything you read here.

Also from Manr.ai