Which one you are in right now
Unless you actively opted out, you are in the new regime. The Income-tax Act 2025 makes it the default at s.202. The old regime is now the one you have to ask for, and asking only pays if your deductions clear a break-even figure you can work out yourself.
The practical test takes a minute. Look at the declaration your employer collected at the start of the year, the regime named on the tax computation attached to your payslip, or the field near the top of your last return. If you cannot remember making a choice, then no choice was made and the default applied — which is the correct answer for most people and the wrong assumption in most articles.
One complication is worth knowing about before you go looking, because it explains why the section numbers you find will not agree with each other. Two Acts are live at the same time. Income earned in the year to 31 March 2026 is still governed by the Income-tax Act 1961, so a return filed for that year carries the section numbers everybody grew up with. Salary being taxed now runs under the Income-tax Act 2025, which repealed and replaced the 1961 Act on 1 April 2026.
The rates barely moved. The addresses all did.
| What people still call it | Where it lives now | What it covers |
|---|---|---|
| Section 80C | s.123 read with Schedule XV | The list of qualifying investments and payments |
| Section 80D | s.126 | Health insurance premium |
| Section 80E | s.129 | Interest on an education loan |
| Section 111A | s.196 | Short-term capital gains on listed equity |
| Section 112A | s.198 | Long-term capital gains on listed equity |
| Section 115BAC | s.202 | The regime rules, and which one is the default |
An adviser, a payroll portal or an article still saying “claim it under 80C” is not necessarily wrong about the substance. It is citing a repealed statute, which matters the moment you go to check the wording for yourself.
The whole decision is one subtraction
Strip away the commentary and the two regimes differ in exactly one structural way. The default regime charges less on the same taxable income. The old regime lets you make the taxable income smaller. Everything else follows from that.
So the question is not which regime is better. It is whether the amount you can shave off your income is large enough to beat the discount you gave up to shave it. That is a subtraction and a division, and you can do both in about ten minutes with the current rate table in front of you.
Four numbers, in this order.
| The number | Where it comes from | |
|---|---|---|
| A | Your tax under the default regime | Current new-regime rate table, applied to your income |
| B | Your tax under the old regime, claiming nothing | Same income, old-regime rate table |
| G | B − A — the cost of standing in the old regime empty-handed | Subtraction |
| m | The rate at which your top rupees are taxed under the old table | The old-regime rate table |
| D* | G ÷ m — the deductions you need before the old regime breaks even | Division |
The logic underneath it is short. Every rupee you deduct comes off the top of your income, so it saves you m rupees of tax. To recover a gap of G you therefore need G divided by m rupees of deductions. Below that figure the default regime wins; above it, the old one does.
Two housekeeping points decide whether that division is honest. Measure both sides in the same units. Compute A and B on the same footing — both with cess and any surcharge included, or both without — and if A and B carry cess then m has to carry it too, or the numerator and the denominator are describing different things and D* comes out wrong.
The second point costs people more. Anything a regime gives you automatically belongs inside A and B, not in your deduction list. The standard deduction on salary is the usual example: you receive whichever version your regime allows whether or not you opt out, so it is already priced into both bills and counting it again as a reason to leave the default double-counts it. D* is a target for the optional, evidence-backed deductions only.
Now the refinement that almost nobody writes down, and the reason to treat D* as a floor rather than an answer. Deductions eat your top slice first. Claim enough of them and you drop below a slab threshold, at which point every further rupee of deduction saves you less than m did — so the true break-even is somewhat higher than G ÷ m. Use the formula to find out whether you are anywhere near the line. If the answer comes out close, compute both bills properly with your actual deductions in place, because the approximation is exactly where it is least reliable.
The useful consequence is that the break-even is a figure about you, not about the regimes. G grows with income, and m changes as you cross thresholds, so a colleague on a different salary has a different D* and their conclusion carries no information about yours.
What a deduction is actually worth
Here is the misunderstanding that produces most of the bad decisions in this area, and it is arithmetic rather than law.
A deduction does not reduce your tax by its own size. It reduces your taxable income by its own size. Take the s.123 limit — ₹1.5 lakh under the old regime, on the investments and payments listed in Schedule XV. Claiming it in full does not put ₹1.5 lakh back in your pocket. It cuts your bill by that figure multiplied by the rate on your top rupees. If that rate is a fifth, you kept a fifth of it; if it is a third, a third. The rest was never yours to keep.
Two consequences follow, and they run in opposite directions.
The first is that the same deduction is worth more to a higher earner, because their top rupees are taxed at a higher rate. This is not unfairness so much as a definition, but it does mean the old regime becomes arithmetically reachable at higher incomes and stays out of reach at lower ones, for the same rupee value of deductions.
The second is sharper. Once your bill is already nil, a deduction is worth exactly nothing. The rebate mechanism takes the tax payable to zero below a stated income; below that line, an extra rupee of deduction reduces a number that was already zero. Every year, people at that income level buy an insurance policy or a tax-saving fund in March to reduce a liability they did not have. The purchase may be defensible for its own sake. As a tax move it returns nothing at all.
Which items belong to which regime
This is where most explainers quietly mislead, by listing deductions without saying that the majority of readers cannot use them.
Under the default new regime there is no s.123, no s.126, no s.129, and no deduction for interest on the house you live in. Those four are old-regime items. Naming the regime alongside the deduction is not pedantry — it is the difference between a plan that works and a plan built on a column of numbers that do not apply to you.
| What you pay | Old regime | Default new regime |
|---|---|---|
| Schedule XV investments and payments (s.123) | Up to ₹1.5 lakh | Not available |
| Health insurance premium (s.126) | ₹25,000, raised to ₹50,000 where the insured is a senior citizen — separate limits for yourself and for your parents | Not available |
| Interest on an education loan (s.129) | Available | Not available |
| Interest on the house you live in | Available | Not available |
| Short-term gains on listed equity (s.196) | 20% | 20% — unchanged |
| Long-term gains on listed equity (s.198) | 12.5% above ₹1.25 lakh a year | Identical — unchanged |
House rent allowance belongs in the old-regime column, and for a salaried tenant it is often the single item that decides the whole question. The exclusion is explicit rather than incidental — s.202 names the Schedule III entry carrying the HRA exemption among the things the default regime is computed without. The same goes for the separate relief available to someone who pays rent and receives no allowance: it sits in Chapter VIII, which the default regime also does without. Either way, rent stops being a tax saving the moment you are in the default regime. See how the HRA exemption is actually computed, because it is the least of three quantities and not simply the rent you pay.
Two things this table still leaves out because they are genuinely unresolved: the retirement contributions an employer routes on your behalf, and the standard deduction on salary. Both are worth real money and both need checking against the Act as it now stands.
What the table does show is the shape of the reader for whom the old regime is even arithmetically possible: somebody already paying a home loan on the house they live in, already paying health insurance premiums, and already contributing to whatever on the Schedule XV list they contribute to. Those outgoings exist whether or not there is a tax code. For everybody else, reaching D* requires buying something.
What the choice does not touch
A regime election is a decision about how your income is taxed. Several things people assume are in scope are not.
Capital gains on listed equity are the clearest case. Short-term gains under s.196 and long-term gains under s.198 carry their own flat rates, and those rates do not consult your regime. Selling equity does not become cheaper by opting out and it does not become dearer by staying in. What can differ is the layer sitting on top: surcharge and cess are computed on the assembled bill rather than on the slab table, and whether their rates and thresholds are identical under the two regimes is worth confirming before you rely on it for a large gain. The base rate is not in question. The topping-up might be.
There is one large exception, and it is the sort of thing that makes a difference to actual portfolios. Gains taxed at slab rates do move with the regime. Units of a debt fund bought on or after 1 April 2023 fall under s.76 — a scheme holding more than 65% in debt and money-market instruments is a Specified Mutual Fund, and it is taxed at slab rates, with the gain always treated as short-term. “Slab rates” means your slab rates, so the same redemption produces a different bill depending on which rate table you are standing on. Interest income behaves the same way.
So the regime choice is silent on your equity gains and loud on your interest and debt fund gains. That asymmetry is worth carrying into the calculation, because it means the answer depends not only on your salary and your deductions but on the composition of what else you earn.
Running your own number, in the right order
The order matters more than the arithmetic, because the common error is not a miscalculation. It is counting the wrong things as deductions.
- List only what you already pay. Loan interest, insurance premiums, provident fund coming out of your salary, education loan interest, school fees — and rent, with a caveat. Rent becomes a tax saving through the house rent allowance in your salary, so a tenant whose pay packet carries no such allowance is in a different position from one whose does — and neither route survives in the default regime, so rent belongs on this list only if you are weighing the old one. The test for everything else on the list is the same: would this outgoing exist next year if the tax code did not?
- Compute A and B — your bill under each rate table with no deductions claimed at all — and take G as the difference.
- Read m off the old-regime table at your income level, and divide. That is D*, the floor.
- Compare D* against the list from step one. Not against the list you could have if you bought things. Against what you already hold.
- If, and only if, step four clears the line, compute both bills properly with the actual deductions applied, because the D* shortcut is an approximation and the exact answer is what you will be filing.
Step one is doing the real work. Run the comparison with a wishlist in the deduction column and the old regime will look better than it is, because you have credited yourself with savings you have not made and costs you have not yet paid. Run it with what is already leaving your bank account and the number is honest.
Two further points about timing. This is an annual computation, not a settled position: the year a home loan is repaid, a child leaves school or a salary jumps, G and your deduction stack both move and the answer can flip without you doing anything. And the choice is not equally reversible for everyone — a taxpayer with business or professional income faces a restriction on moving back and forth that a purely salaried taxpayer does not. The specifics of that restriction under the 2025 Act are the one item here worth confirming from the Act itself before you opt out, rather than after.
None of this tells you which regime to elect. It tells you what the election costs, which is the only part an article can honestly supply.
Three mistakes worth naming
Each of these is specific, each is common, and each is recognisable in a real March.
- Buying the deduction. A deduction is a discount, not a return. Spending a rupee to save m of it only makes sense if you wanted the thing for the remaining part — and the products sold hardest in March are frequently the ones that would not survive being assessed on their own merits. The clean test is to price the purchase with the tax benefit deleted and see whether you would still make it. Bundled savings-and-cover policies are where this goes wrong most reliably; the arithmetic of separating the two is in term insurance and what it actually buys.
- Taking the lock-in without the deduction. An equity-linked savings scheme — the tax-saving mutual fund category — carries a statutory lock-in of three years. That lock-in is a property of the scheme category. It does not lift because the deduction turned out to be unavailable to you in the default regime. Buying one out of habit, in a regime that gives you nothing for it, is paying a liquidity cost for a benefit you cannot claim.
- Deciding once and never again. The break-even moves every year, in both directions. A household that correctly chose the old regime while a home loan was running does not automatically stay correct once the loan is closed and the largest deduction on its list disappears.
And the trade-off underneath the whole decision, stated plainly. The old regime buys a lower bill at the price of committed, documented, maintained spending — money that has to keep going to particular places, with proof, year after year. The default regime buys freedom over where your money goes, at the price of paying more when your fixed obligations already happened to be the tax-favoured kind. Neither is free. Which cost you would rather carry is a question about your circumstances, and it is not one this article can answer for you.
The purchase behind the deduction
Notice that the tax question, once you have run it, usually reduces to a different question entirely: is the thing you were about to buy in March worth owning?
That is not a tax question and it does not have a tax answer. A tax-saving fund is a mutual fund scheme with a lock-in attached, and it is assessable the same way any other scheme is — against its benchmark, over full holding periods rather than convenient ones, and on how far it fell in the worst stretch you would have had to sit through. FNOTrader's Mutual Funds app runs on the full AMFI NAV history — around 34 million NAV rows — and reports rolling-return distributions and maximum drawdown alongside the headline figure, which is the comparison a lock-in deserves before it starts.
Where the tax bill sits inside the rest of a household's finances, and how a fixed annual obligation like a premium should be planned for rather than found in March, is the subject of financial planning rather than of the tax code.
FNOTrader is not a tax adviser, an accountant or a SEBI-registered investment adviser, and nothing above is a recommendation to elect either regime or to buy any product. The section numbers here are the Income-tax Act 2025 as it stands; the rates and thresholds you need for the arithmetic change with each Finance Act, so take them from the current table rather than from any article.
Common questions
Which tax regime am I in by default?
The new regime. The Income-tax Act 2025 makes it the default at s.202, so it applies unless you actively opted out. The old regime did not disappear, but it is now the one you have to ask for rather than the one you fall into.
How do I tell which regime is being applied to my salary?
Check the declaration your employer collected at the start of the year, the regime named on the tax computation attached to your payslip, or the field near the top of your last return. If you cannot recall making a choice, no choice was made and the default applied.
How much in deductions do I need before the old regime is worth it?
Compute your tax both ways with no deductions claimed at all. The gap between the two bills, divided by the rate at which your top rupees are taxed under the old table, is roughly the deduction total you need. Treat it as a floor — deductions come off your top slice first, so once they carry you below a slab threshold each further rupee saves less.
Is Section 80C still available?
Section 80C itself no longer exists — the Income-tax Act 1961 was repealed on 1 April 2026. The equivalent is s.123 of the Income-tax Act 2025, read with Schedule XV, capped at ₹1.5 lakh. It is available under the old regime only. Under the default new regime there is no such deduction at all.
Can I switch between the two regimes every year?
A salaried taxpayer generally makes the choice for each year. Where there is business or professional income the switching rule is more restrictive. The exact mechanic under the 2025 Act, including the form and the deadline, is worth confirming from the Act or with whoever prepares your return before you opt out — it is the one part of this decision that may not be freely reversible.
Does the regime change how my capital gains are taxed?
Not for listed equity. Short-term gains under s.196 and long-term gains under s.198 carry flat rates that do not depend on your regime, though surcharge and cess sit on top of the assembled bill and are worth checking separately on a large gain. Gains taxed at slab rates are different: units of a debt fund bought on or after 1 April 2023 fall under s.76 and are taxed at slab rates, with the gain always treated as short-term, so the same redemption produces a different bill depending on which rate table you are on.
Is it worth buying an ELSS or an insurance policy to save tax under the new regime?
Under the default new regime there is no s.123 deduction, so a purchase made to claim one returns nothing on the tax side. The lock-in still applies — an equity-linked savings scheme carries a statutory lock-in of three years whether or not you can claim anything for it. The purchase has to justify itself on its own merits.
Which regime suits someone paying rent, a home loan and insurance premiums?
That is the profile where the old regime is arithmetically reachable, because those outgoings exist regardless of the tax code and several of them are old-regime deductions. Whether it actually wins still depends on the numbers: work out the gap between the two bills with nothing claimed, divide by your marginal rate under the old table, and compare the result against what you already pay.
Does a deduction reduce my tax by the full amount I invest?
No. It reduces your taxable income by that amount, so it reduces your tax by that amount multiplied by the rate on your top rupees. And if a rebate has already taken your bill to nil, a further deduction reduces a number that was already zero — which is why a March purchase made purely to save tax sometimes saves nothing.
Continue reading
More in Tax Planning · App: Mutual Funds · Definitions: glossary · Free tools: calculators · All: every article