8/
One honest note: the calculator models a constant return, so it gives you the baseline — ₹1 crore at ₹60,000/mo and 10% flat runs out in year 20.
The ordering above I ran separately. Start with the baseline:
https://t.co/16uoAlJbU4
6/
What actually survives a bad start is the withdrawal rate. Nothing else comes close.
Same ₹1 crore, same brutal first 3 years:
₹45,000/mo → year 16
₹40,000/mo → year 19
₹35,000/mo → year 23
₹30,000/mo → all 30 years
3.6% is what held.
7/
Which flips the question you should be asking.
Not "what will my corpus return?" but "what does my corpus survive?"
₹60,000/month for 30 years needs ₹2 crore, not ₹1 crore — if you want it to hold through a bad first 3 years.
The gap is the insurance premium.
5/
Now the part that surprised me.
The standard advice is "keep 2-3 years of expenses in cash so you don't sell in a crash."
I modelled it. A 3-year cash bucket moved depletion from year 16 to year 17.
One year. For ₹16 lakh parked at 6%.
4/
And this isn't only about greedy withdrawal rates.
Drop to ₹45,000/mo — 5.4%, well inside what most people call safe:
Bad years last: survives 30 yrs, ₹3.92 crore left over
Bad years first: dead in year 16
Same returns. ₹3.92 crore, or nothing.
2/
The setup, so you can check it.
Three bad years: -25%, -10%, +2%
The other 27 years: 12.72%
Compounded, both orders average exactly 10% a year. Identical returns, identical average.
Put the three bad years last → year 30.
Put them first → year 11.
3/
Why order matters once you're withdrawing:
In -25% year you still need ₹60,000 a month. So you sell units at the bottom to buy groceries.
Those units are gone. They aren't there for the 12.72% recovery.
A crash while accumulating is paper. While withdrawing it's permanent.
1/
₹1 crore corpus. ₹60,000/month, rising 6% a year for inflation.
I ran the same 30 annual returns twice. Same numbers. Only the order changed.
Order A: money lasts 30 years.
Order B: gone in year 11.
Nothing changed except when the bad years showed up. 🧵
8/
₹1,17,277 → ₹42,340. A 64% cut, and not one extra percent of assumed return.
Run your own four. One goal at a time, then add them up:
https://t.co/LZEuz6yYWr
1/
House in 7 years. Child's education in 12. Retirement in 25. A car in 3.
I ran all four through the target-corpus calculator. Four separate required SIPs:
₹35,585 + ₹32,917 + ₹19,478 + ₹29,297
= ₹1,17,277 a month.
The budget was ₹40,000. 🧵
8/
Step-up SIPs work. Just not for the reason stated, and not at 10% unless your salary cooperates.
Full comparison at 10, 20 and 30 years, plus when flat is the right call:
https://t.co/eYkd5zSBBW
1/
₹10,000/month SIP, 12% return, 20 years.
Flat: ₹98.9 lakh
Step up 10% a year: ₹1.97 crore
Exactly 2x, and both pay the same ₹10,000 in year one.
Everyone stops the thread here. The interesting part is what happens next. 🧵
7/
So match the step-up to the raise you actually expect.
6% step-up, same ₹10,000 start, 20 years: ₹1.46 crore. Still 1.47x flat.
Year-20 SIP: ₹30,256 — against an income that also tripled. The SIP never gets heavier.
That one you'll finish.