“Extra ₹25,000 a month from the new job. Loan or mutual fund?” That one-line question, from a client in Surat in early 2025, is probably the single most common one I get — and the internet's standard answer (“equity returns beat loan rates, always invest”) is lazier than the question deserves. So let's do what I did for him: run his actual numbers both ways, and then talk about the part the spreadsheet can't see.
His position: ₹42L outstanding on a home loan at 8.6%, 14 years left, EMI around ₹43,000. New-regime taxpayer, so no Section 24(b) interest deduction softening the loan — his 8.6% is a true 8.6%. That detail matters more than people realise: under the old regime, the deduction could pull the effective cost of the loan down toward 7%, and the whole comparison tilts. On the new regime, the loan costs exactly what it says.
Route one: prepay ₹25,000 a month
Paying ₹68,000 a month instead of ₹43,000 collapses the loan from 14 years to just under 7. Total interest paid drops from about ₹30.4L to roughly ₹13.6L — a saving of nearly ₹17L, guaranteed, tax-free, and immune to whatever the market does. Prepaying a loan is the only investment I know with a certain 8.6% post-tax return. That sentence deserves more respect than it gets.
Route two: invest ₹25,000 a month
Same ₹25,000 into an equity index SIP for the full 14 years. I use 11% as the working assumption — not the 14-15% that fund marketing implies, because after 12.5% long-term capital gains tax and at least one ugly multi-year stretch, 11% post-tax is what I'm willing to plan around. At 11%, the SIP grows to about ₹99L. The same money compounding at the loan's 8.6% is worth about ₹81L. The investing route wins by roughly ₹18L over 14 years — while the loan runs its full course in the background.
So the spreadsheet has a clear answer. ₹18L is not a rounding error. If the question were purely arithmetic, we'd be done in two paragraphs.
Why I still didn't say “invest all of it”
Because that ₹18L edge rests on a 2.4% spread — 11% versus 8.6% — held for 14 straight years, through every correction, job scare, and family emergency in between. The 8.6% side is guaranteed. The 11% side requires him to keep the SIP running in the exact months when his portfolio is down 25% and prepaying the loan suddenly feels like the only sane thing on earth. I've watched enough people stop SIPs in March 2020 to price that risk honestly: the investor who was supposed to earn 11% and actually behaves like one is rarer than the spreadsheet assumes. A 2.4% spread is thin enough that the person matters more than the plan.
There's also the sleep line. Some people carry a ₹42L liability lightly. Others feel it every single month, and for them the loan isn't an 8.6% number, it's a weight — and being debt-free at 46 instead of 53 is worth more than ₹18L of theoretical difference. That's not financial illiteracy. That's a preference, and a plan that ignores it will be abandoned.
What we actually did
Split it — but not 50/50 as a cop-out. ₹15,000 into the equity SIP, ₹10,000 into prepayment, reviewed annually. My reasoning: he'd already proven he could hold equity through 2020 without flinching, which earned the larger share for investing; the prepayment slice still cuts about 3.5 years and ₹8L of interest off the loan, which keeps the debt-free date visible enough to feel real. If his job situation ever turns shaky, the split flips — a smaller loan is worth more than a bigger portfolio when income is uncertain. And if his loan rate had been 9.5% instead of 8.6%? I'd have told him to prepay most of it, full stop. The spread decides, and at a 1.5% spread the guaranteed side wins on my desk every time. Working out where your own numbers land is a standard part of what we do on the loans side.
This is a general account of a real, anonymised case for educational purposes and isn't a substitute for advice on your own loan terms, tax regime, and risk profile, which change the math materially.