Guide · Costs and taxes · Updated 2026-09-18

Presumptive taxation for traders: 44AD covers one of your two businesses, not both

Most Indian writing on this question answers it in one sentence, and the sentence is wrong. It says that a trader whose turnover is below the limit may declare six percent and avoid an audit. That is true of one half of a typical trading year and false of the other half, and the half it is false about is the one most retail traders have the most transactions in.

The error is not a technicality. It produces a return that puts speculative income on a form that cannot carry it, understates a book that had to be computed on actual results, and starts a five-year clock the filer did not know was running. This page works the distinction, the thresholds, the arithmetic of the deemed figure, and the consequence of changing your mind.

Two businesses, one PAN, and only one of them qualifies

The Income-tax Act does not recognise a category called trading. It recognises business income, and within business income it separates speculative from non-speculative. That separation is done by section 43(5), which defines a speculative transaction as a transaction in which a contract is settled otherwise than by the actual delivery of the commodity or scrip.

An intraday equity trade is settled by difference. No shares reach the demat account, no delivery obligation arises, and the position is closed against itself before the session ends. It is inside the definition. A delivery trade is outside it because delivery occurs. The test is mechanical and has nothing to do with intent, holding period, analysis quality or the size of the position.

Derivatives sit outside the definition for a different reason. They are also settled without delivery in the ordinary case, which would place them inside section 43(5), except that a proviso to the section removes an eligible derivative transaction carried out on a recognised stock exchange from the definition. The carve-out is what makes exchange-traded futures and options ordinary business rather than speculation for this purpose.

Section 44AD(6) then closes the question. The presumptive scheme does not apply to a person carrying on a speculative business. There is no turnover test to fail and no election to make. The intraday book is outside the scheme by its own nature.

The set-off rows matter as much as the 44AD row. A speculative loss can only be set off against speculative income, so an intraday loss cannot be used to reduce a derivatives profit. A filer who treats the year as one business discovers this at assessment rather than at filing.

The threshold, and the condition that moves it

Section 44AD(1) sets the limit at two crore rupees of turnover. A proviso inserted by the Finance Act 2023 raises it to three crore where the aggregate of amounts received in cash does not exceed five percent of total turnover, with a corresponding condition on payments.

For a trading business this condition is close to automatic. Funds move from a bank account to a broker and back. Settlement is electronic by construction. The cash proportion in an ordinary retail trading business is zero, which means the operative threshold is the higher one rather than the lower one that most summaries quote.

The deemed profit is a floor, not a rate

The scheme deems income at eight percent of turnover, reduced to six percent to the extent turnover is received through an account payee cheque, bank draft, electronic clearing system or other prescribed electronic mode. Both figures are minimums. The Act permits a higher declaration and expects one where the actual profit is higher.

This is the point at which the scheme stops being an unambiguous convenience. Six percent of turnover is not six percent of capital, and for a derivatives book the two numbers are not close. Turnover under the scheme is not contract value, but it is still a figure that scales with activity rather than with the account. A trader running a modest account at high frequency can generate a turnover figure whose six percent exceeds the profit actually made.

The third and fourth rows are the ones to sit with. The scheme is described as a relief, and in those rows it is a charge on income that was never earned. The convenience of skipping books and an audit is bought with tax on a profit the account did not produce. Whether that trade is worth making is an arithmetic question specific to the year, and it is answerable before the return is filed rather than after.

The five-year consequence, and why the decision is not annual

Section 44AD(4) is the provision that changes this from a filing choice into a multi-year commitment. Where an eligible assessee declares profit under the scheme for a previous year and does not declare under it for any of the five assessment years that follow, the assessee is not eligible to claim the scheme for five assessment years after the year of non-declaration.

Section 44AB(e) then attaches the consequence. Where 44AD(4) applies and total income exceeds the maximum amount not chargeable to tax, accounts must be audited. The audit arrives not because turnover grew but because the scheme was left.

The practical shape of this is a trader who opts in during a good year, has a loss the following year, and wants to declare the loss so it can be carried forward. Declaring the loss means not declaring under the scheme. That single decision starts the ineligibility window and, if income in those years crosses the basic exemption, brings an audit obligation for each of them.

None of that makes the exit wrong. Preserving a genuine loss for set-off against eight subsequent years of derivatives income can be worth considerably more than the cost of an audit. The point is that the comparison has to be made deliberately, in the year of entry, with the exit priced in. It is not a decision that can be revisited cheaply each year.

The audit question, decided in the right order

Audit and the presumptive scheme are frequently discussed as if one simply avoids the other. They are separate tests in section 44AB, and the order in which they are applied changes the answer.

The turnover test is 44AB(a). Accounts must be audited where total sales, turnover or gross receipts in business exceed one crore rupees, raised to ten crore where cash receipts and cash payments are each five percent or less of their respective totals. Both limbs of that condition have to hold, not one. For a broker-settled trading business both ordinarily do, which puts the operative figure at ten crore.

The scheme-exit test is 44AB(e), described above. It is not a turnover test at all. It applies where section 44AD(4) has been triggered and total income exceeds the maximum amount not chargeable to tax, and it can therefore require an audit at a turnover far below any of the figures in 44AB(a).

The last row is the one that surprises people. A trader with a few lakh of turnover who used the scheme two years ago and declared a loss on actual results last year can be inside 44AB(e) while being nowhere near the 44AB(a) figure. The audit follows from the filing history, not from the size of the business.

The order that answers the question cleanly is: establish whether 44AD(4) has been triggered by anything in the preceding five years, then test turnover against 44AB(a), then test income against the exemption limit. Starting with turnover, which is what most people do, produces a confident answer to the wrong test.

What a mixed year looks like on the return

Take a year with both books open. Intraday equity produces a loss. Derivatives produce a profit and turnover sits under the threshold. The filer would like to declare six percent on the derivatives book and be done.

That is not available in the form it is usually imagined. The presumptive return form cannot carry speculative business income, and the intraday book is speculative business income whether it shows a profit or a loss. The return is ITR-3. Within it, the derivatives business may still be presented on a presumptive basis, while the speculative business is computed on actual results and its loss carried forward for four assessment years against future speculative income only.

The result is a return that is neither the simple presumptive filing nor the full audited computation, and understanding which parts of it are which is most of the work. The order that makes it tractable is: classify each book first, apply the scheme only to the eligible one, compute the other on actuals, then test the audit thresholds against what remains.

Where this goes wrong in practice

The failures cluster in four places, and all four are decided before the return is opened.

Treating the year as one business. The single most common error. It produces a wrong form, a wrong set-off, and a presumptive claim over income that is statutorily excluded from it.

Computing turnover on contract value. Turnover for a derivatives business is not the notional value of the contracts traded. Using contract value inflates the figure by orders of magnitude, pushes a filer past a threshold they never actually crossed, and manufactures an audit requirement out of arithmetic.

Reading the deemed percentage as a ceiling. Where actual profit exceeds the deemed figure, the higher amount is what should be declared. The scheme simplifies the computation. It does not cap the liability.

Entering without pricing the exit. The five-year provision is invisible in the year of entry and expensive in the year of departure. A trader whose income is volatile enough to produce a loss year, which is most of them, is exactly the person for whom the exit matters most.

What the scheme is actually for

Section 44AD was written for a small business whose records are thin and whose margins are reasonably stable, to spare it a computation that would cost more than the tax it determines. A derivatives book with a steady modest turnover and no appetite for bookkeeping fits that description.

A trading account does not fit it when turnover scales with activity rather than with capital, when losses need to be preserved, or when the year contains a speculative book the scheme cannot reach. Those are the ordinary conditions of active trading rather than the exceptions, which is why the correct answer to whether a trader should use the scheme is usually a computation rather than a preference.

The arithmetic that decides it is the same arithmetic that should have been running before the trades were placed: what the activity actually produced, against what it cost to produce. A filing method cannot repair a year. It can only describe one accurately or inaccurately.