The Cost of Global Investing: You Pay Going Out. You Pay Coming Back.

The advice arrives without a price

Diversify globally. Own the world, not one country. Sensible advice, and it is everywhere – on every app selling access to US stocks, in every thread about home-country bias.

What never travels with it is the invoice. When cost does come up it arrives as one number: “about 2–3% in forex”. Beat that, the reasoning goes, and you are ahead.

That number is roughly right. It is also only half the bill, because it is charged again on the way back.

Four meters, and three of them run twice

Under the RBI’s Liberalised Remittance Scheme a resident individual can send up to USD 250,000 abroad in a financial year. Getting rupees through that door, and back again, costs four things.

The first two are charged on both legs. That is the part the “2–3%” number quietly leaves out – not that it is understated, but that you pay it twice.

Most of that first line is a markup, and a markup deserves attention because it never appears on a statement. A fee is charged to you; a markup is simply a worse exchange rate, applied at the moment of conversion. Two platforms can both advertise zero commission and still sit a full percentage point apart on the rate they hand you.

Follow ₹2 lakh, all the way there and back

Percentages hide things. Here is one ticket – ₹2 lakh, held five years, at 2.5% to move the money each way and 0.25% brokerage each way.

Note where the bigger bite lands. Moving the money out cost ₹5,000. Moving it home cost ₹8,070 – more, because by then there was more money to charge a percentage on. The exit is the expensive barrier, and it is the one nobody quotes.

The same ₹2 lakh, at the same 12%, in an Indian fund would have left ₹3,48,897. The gap is ₹34,192, and none of it came from the market being worse. It came from the road.

Then the gain is taxed on worse terms

Most people assume a foreign share is taxed like an Indian one. It is taxed like a different asset class, in three specific ways.

Read the last row twice. Sell an Indian stock inside a year and the gain is taxed at 20%. Sell a foreign one inside two years and it is added to your income at your slab rate – 30% plus surcharge and cess at the top. A ten-point gap on the same profit.

Dividends arrive pre-cut: the US withholds 25% on dividends paid to Indian individuals under the India–US treaty. Reclaimable in India as a credit, but only if you file for it. And the holding itself must be declared in Schedule FA of your return every year you own it, dividend or not.

So what does it actually have to beat?

One thing decides it, and it is not the market. How long you hold.

Two things stand out, and they pull in opposite directions.

The first is how violent the short end is. A one-year holding has to out-earn the Indian alternative by nearly seven percentage points a year just to draw level. Not to win – to draw. There is no market view good enough to be reliably worth that.

The second is that it never quite goes away. Stretch the holding to fifteen years and the toll is still costing you around 1.1% a year, every year. That is the part the “it amortises” argument gets wrong. It amortises a long way. It does not amortise to nothing.

The verdict

Nothing here says foreign markets are a poor place to own assets. It says the arithmetic never punished that money for being foreign. It punished it for being brief – 6.86% a year over one year, against 1.13% over fifteen.

Which turns the usual question inside out. Before money leaves the country the thing to establish is not whether foreign markets will beat Indian ones. It is whether you are going far enough down the road for the toll to stop being the story.