Gold Rally 2026: Should You Book Profits? The Rebalancing Math
Updated: August 2026 | Category: Investment Basics | Read time: 9 min | Applies to: FY 2026-27 (AY 2027-28)
Should you book profits in gold? If you have asked that every time gold printed a fresh high through 2026, you are asking a question that has no general answer — and that is exactly the problem. Whether you sell has very little to do with where gold goes next, and almost everything to do with how far gold has drifted from the weight you originally chose. This post replaces the forecast with arithmetic: how much drift is too much, what the tax bill on selling gold actually looks like in FY 2026-27, and four ways to bring gold back to target while handing over less of the gain.
The reason this matters now is not that gold is expensive. It is that a portfolio which was 10% gold two years ago is no longer 10% gold, and most people have never checked. A rally does not just make you richer. It quietly rewrites your risk.
Why "should I book profits in gold" is the wrong first question
Booking profits is a market call. It requires you to believe gold is near a top. Nobody — including the people writing daily "gold rate today" pages — knows that.
Rebalancing is a risk call. It only requires you to know two numbers: the gold weight you decided on, and the gold weight you have now. If those two numbers have separated far enough, you sell some gold. If they have not, you do nothing. Gold's price appears nowhere in the decision, which is precisely what makes the rule usable.
The distinction is not academic. An investor who "books profits" sells on a hunch, then spends the next year deciding when to buy back and usually never does. An investor who rebalances sells a defined slice, moves it into whatever is now underweight, and is finished in one afternoon.
So the first question is not is gold expensive. It is: what is my gold weight today, and what did I sign up for?
What drift actually does to a ₹50 lakh portfolio
Assume you built a ₹50 lakh portfolio with a conventional split: 60% equity, 25% debt, 10% gold, 5% cash. Now assume a stretch where gold materially outruns both equity and debt — the pattern Indian investors have lived through more than once in the last few years.
Here is what happens to the weights, using round illustrative growth numbers so the mechanism is visible:
| Asset | Start value | Start weight | Illustrative growth | End value | End weight |
|---|---|---|---|---|---|
| Equity | ₹30,00,000 | 60% | +10% | ₹33,00,000 | 53.2% |
| Debt | ₹12,50,000 | 25% | +7% | ₹13,37,500 | 21.6% |
| Gold | ₹5,00,000 | 10% | +45% | ₹7,25,000 | 11.7% |
| Cash | ₹2,50,000 | 5% | +6% | ₹2,65,000 | 4.3% |
| Total | ₹50,00,000 | 100% | — | ₹56,27,500 | 100% |
Two things worth noticing.
First, a 45% move in gold shifted the gold weight by only 1.7 percentage points, from 10% to 11.7%. That is not a rebalancing trigger under any sensible rule. A single strong year in a 10% sleeve rarely is.
Second — and this is the part people miss — equity fell from 60% to 53.2% without losing a rupee. Drift is relative. The asset that "needs attention" after a gold rally is often not gold at all. It is the underweight equity leg, and the fix is to direct new money there rather than to sell anything.
Now repeat the exercise with gold starting at 20% and running up 45% while equity is flat. Gold lands near 26%. That is a trigger. Same rally, completely different answer — because the starting weight was different. This is why "should you book profits in gold" cannot be answered for everyone at once.
The two rebalancing rules — and which one to use
Indian advisors mostly use one of two bands. Pick one and write it down.
- The 5-point absolute band. Rebalance when any asset is more than 5 percentage points away from target. A 10% gold target triggers at 15% or 5%. Simple, but it is far too loose for small sleeves — a 5% gold target would have to triple before it triggers.
- The 25% relative band. Rebalance when an asset is more than a quarter away from its own target, in relative terms. A 10% gold target triggers at 12.5% or 7.5%. A 5% gold target triggers at 6.25%. This scales properly, which is why it suits the small satellite allocations most Indian portfolios give gold.
For a gold sleeve of 5–15%, use the relative band. For your equity-versus-debt split, either works.
A third rule — rebalance on a fixed calendar date, once a year, regardless — beats both for one reason: it survives contact with human behaviour. Band rules require you to keep checking. A calendar rule requires you to look once. If you have never rebalanced at all, start with the calendar version, not the sophisticated one.
Whichever you choose, the rule must be set before the rally, not during it. A rule invented while staring at a number that has already moved is just a forecast wearing a rulebook's clothes.
The tax bill on rebalancing: what selling gold actually costs
This is where Indian portfolios differ from the textbook, and where most rebalancing advice goes quiet. Selling gold is a taxable event, and gold does not get the exemption equity gets.
For FY 2026-27, the broad position by holding vehicle:
| How you hold gold | Long-term after | Long-term tax rate | Short-term treatment |
|---|---|---|---|
| Physical gold, jewellery, coins | 24 months | 12.5% without indexation | Added to income, taxed at slab |
| Gold ETF (listed units) | 12 months | 12.5% without indexation | Added to income, taxed at slab |
| Gold fund-of-fund / gold savings fund | 24 months | 12.5% without indexation | Added to income, taxed at slab |
| Sovereign Gold Bond, held to maturity | — | Exempt on redemption | — |
Three consequences follow, and they change the arithmetic materially.
The ₹1.25 lakh exemption does not apply to gold. That exemption sits in Section 112A and covers listed equity and equity mutual funds only. Gold gains are taxed from the first rupee. An investor who assumes a ₹1.25 lakh cushion on a gold sale is under-provisioning their tax.
On the ₹2.25 lakh gain in the table above, a full trim is not what you owe tax on. Rebalancing from 11.7% back to 10% means selling roughly ₹95,000 of gold, not ₹7.25 lakh. If the embedded gain in that slice is about ₹30,000, long-term tax at 12.5% is roughly ₹3,750 plus cess. That is the real cost of the discipline — not large, and worth paying. Confusing "tax on the sale value" with "tax on the realised gain in the slice you sold" is what makes people avoid rebalancing entirely.
Holding period is a lever, not a footnote. Trimming a gold ETF at month eleven means the entire gain is added to your slab income. At ₹15 lakh-plus income under the new regime, that is the difference between 12.5% and 30%. If your band triggered at month ten, waiting eight weeks is usually the highest-return decision available to you. The same trap shows up across asset classes — we covered its ITR-side consequences in capital gains tax on shares and mutual funds.
One more thing that catches salaried investors: a realised gain creates an advance tax obligation in the quarter you realise it, and your employer's TDS will not cover it. If you rebalance in September, the September instalment moves. See advance tax for salaried taxpayers with capital gains for the instalment mechanics.
Four ways to rebalance while paying less tax
You rarely have to choose between "hold a distorted portfolio" and "pay full tax". There are cheaper paths.
- Rebalance with new money first. If you are still investing monthly, redirect SIPs away from gold and into whatever is underweight until the weights converge. No sale, no tax, no exit load. This alone fixes most drift under 3 percentage points, and it is the only method that costs nothing.
- Sell the highest-cost lots, not the oldest. Indian mutual fund redemptions are FIFO by default, but you can control which units you redeem by holding tranches in separate folios. Redeeming units with a higher acquisition cost realises a smaller gain for the same rupee amount trimmed.
- Pair the gold gain with a booked equity loss. A long-term capital loss can be set off against any long-term capital gain, including gold. If you are sitting on an underwater equity position you no longer want, selling it in the same financial year reduces the net taxable gain. Our walkthrough of LTCG harvesting and the ₹1.25 lakh exemption covers how to sequence this without tripping the wash-sale-style scrutiny that follows same-day buybacks.
- Split the trim across two financial years. If the trim is large and your gains are already high this year, take half in March and half in April. Two smaller advance-tax hits, and the second one funds itself from a later quarter's cash flow.
Note what is missing from this list: switching between gold vehicles to "reset" the cost base. A switch from a gold fund to a gold ETF is a redemption plus a fresh purchase. It is fully taxable and buys you nothing.
What would have to be true for you to hold a higher gold weight?
There is a legitimate answer to "hold, don't trim" — but it has to be a portfolio reason, not a price reason.
Gold earns its place because it behaves differently from Indian equity when equity is under stress, and because it carries currency exposure: a weakening rupee lifts the rupee price of gold even when the dollar price is flat. If your income, your job, your property and your equity are all rupee-denominated and India-concentrated — which describes most salaried investors here — that second property is doing real work.
So a higher target weight is defensible if: you are within a few years of a large rupee liability you cannot postpone, you have no other foreign-currency asset, or your equity allocation is unusually concentrated in domestic cyclicals. In those cases, raise the target to 15%, write it down, and then rebalance to the new target.
What is not defensible is leaving the target at 10%, watching the weight go to 18%, and calling it conviction. That is not an allocation. That is a position you forgot to manage.
If you are still deciding how to hold the metal rather than how much, start with gold ETF vs SGB vs physical gold — the vehicle choice affects both your tax rate and your ability to trim in small slices, which is exactly what rebalancing needs.
Your decision framework
Today (15 minutes). Open your consolidated account statement and your bank locker inventory. Add up every form of gold you own — ETFs, gold funds, SGBs, jewellery you would actually sell, digital gold. Divide by your total investable net worth. Write the percentage down. Most people are surprised in one direction or the other.
This week. Write down your target gold weight and your band. If you have never set one, use 10% with a 25% relative band, which means you act outside 7.5%–12.5%. Put the date of your annual review in your calendar now.
This month. If you are outside the band on the high side, work down this order: first redirect new contributions, then trim only what is long-term, then pair the gain with any long-term loss you were going to book anyway. Compute the advance-tax impact for the quarter before you place the sell order, not after. If you are inside the band, do nothing — and stop reading gold price updates until your review date.
The bottom line
Should you book profits in gold? Only if your gold weight has moved outside a band you set in advance. The rally is not the trigger; the drift is. And when the drift is real, the cost of correcting it — roughly 12.5% of the gain in the slice you sell, with no ₹1.25 lakh cushion — is small enough that "the tax" is not a reason to leave a portfolio mis-weighted for another year.
Set the target, set the band, check once a year. That single habit will do more for your returns than any correct call on where gold goes next.
