Why a Company's Effective Tax Rate Changes

A company's effective tax rate drifts from the statutory rate whenever something makes taxable profit differ from book profit, or changes the tax charged on it. The usual causes are past losses being used, income taxed on concessional terms, expenses not allowed, and timing differences unwinding. Some of those causes come back next year and some never do, and the difference is why the rate is worth reading rather than noting.

Here is what sits underneath that. The statutory rateThe rate written into the tax law for a class of taxpayer in a given year. The statutory rate is a rule, not a measurement, and it is applied to a profit computed under the tax rules rather than the profit shown in the accounts. is charged on a profit computed under the tax rules. The effective rate is measured against the profit shown in the accounts. Taxable profit and book profit are two different profits, arrived at by two different sets of rules for two different purposes, and once that is accepted the gap between the two rates stops being a puzzle and becomes a list. Every item on the list has a name, a size in rupees, and a lifespan.

The effective tax rate itself, deferred tax and the carrying forward of a loss are each covered separately. Put together, they raise a question none of them answers alone: why did this particular rate come out at this particular number this year, and will it still be there next year?

Why does the effective rate differ from the statutory rate at all?

Because the two rates are not measured against the same thing. A household's tax as a share of what it earns has almost nothing to do with the rate printed in the slab table. The slab rate is applied only after deductions, exemptions and set-offs have reshaped the income it lands on. A company works the same way, only with more items and a formal document at the end. The rate in the law is applied to taxable profit. The rate a reader computes is measured against book profit. Nobody is being clever and nothing has gone wrong.

The statutory rate is charged on a profit the accounts never report, so the effective rate is the wrong rate applied to the wrong profit producing exactly the right answer. Watch it on a case. Anjani Stationers, an invented notebook printer, reported profit before tax of Rs 38,00,000 in year two. The tax computation for the same year starts from that figure and then does three things to it: it adds back Rs 2,00,000 of expenses that were charged in the accounts but are not allowed as deductions, it takes off Rs 7,20,000 because the depreciation allowed on the delivery van for tax is faster than the depreciation charged in the accounts, and it takes off Rs 8,00,000 of the accumulated losses that Chitra Binding, acquired at the start of year two, brought into the group. The profit left after those three adjustments is Rs 24,80,000, and tax is computed on that figure, not on Rs 38,00,000.

Three adjustments and a rate applied to what is left is the whole of the mechanism. The rest is a matter of organising the adjustments by direction and by lifespan. The document that lists the adjustments in public is called a tax reconciliationA short statement that starts at one figure, lists each named difference with its amount, and arrives at a second figure, so a reader can see not just that two numbers differ but exactly what made them differ., and it is the only place a reader outside the business gets to see the list.

Which causes push the rate down?

Four kinds, and they have almost nothing in common except direction. Losses from earlier years being set against this year's profit is the first, and it is the one on the case. Income taxed on concessional terms is the second, whether that is a lower rate on a particular class of income or a deduction offered for a particular kind of spending. A tax holidayA period during which a qualifying activity, location or class of taxpayer is charged little or no tax, granted for a stated number of years and usually attached to conditions that must keep being met. on a qualifying activity is the third. An over-provision from an earlier year being written back when the assessment finally closes is the fourth, and it is the quietest of them because it looks like nothing at all in the accounts.

Every one of the downward causes is finite or conditional. None of them is a property of the business the way a gross margin is, and that is the single most useful thing to know about them. Think about a household that pays no tax one year because a large medical deduction was available. Nothing about the household's earning changed. A one-time item sat in the middle of the computation and then it was gone. The finance version behaves identically. Anjani Stationers' rate fell this year because Rs 8,00,000 of Chitra Binding's accumulated losses were utilisedSet against profit in the tax computation so that the profit is reduced by that amount. A balance that has been utilised is spent: it can only be used once and then it is no longer available. in the group's tax computation, saving Rs 2,00,000 at the illustrative 25 per cent. A loss being utilised is a balance being spent, not a business becoming more efficient.

Which causes push the rate up?

Fewer kinds, and they are blunter. Expenses charged in the accounts that the tax rules do not allow as deductions are the main one and the one on the case: the accounts take the cost off profit, the tax computation puts it back on, and the business is taxed on money it has genuinely spent. Penalties and fines settled during the year usually sit here too. Income taxed at a higher rate than the general one is the second kind. The third is a previously recognised tax asset being written down once the profits it depended on are no longer expected. Nothing in the trading has changed at all, and for that reason it is the one that most often surprises a reader.

An expense not allowed for tax is a permanent gap that costs real money every single year the spending happens. Of every item on a reconciliation it is the least glamorous and the most durable. Anjani Stationers charged Rs 2,00,000 of such expenses in year two. At the illustrative 25 per cent that adds Rs 50,000 to the tax bill, and the Rs 50,000 does not come back in a later year, get netted off, or reverse. A gap of that kind is a permanent differenceA difference between the accounts and the tax computation that is never made up in any later year. A disallowed expense stays disallowed, so the effect on tax is real and final rather than a matter of timing., and the word permanent is doing all the work in it. If Anjani Stationers spends the same way next year, the same Rs 50,000 is added again.

Which of these come back next year, and which never do?

The question separates a reader who has noticed the rate from a reader who has understood it. Answering it means sorting each cause twice: once by direction, and once by whether it will still be around in twelve months. Two questions, four boxes. A cause placed in a box has been described in every way that matters for next year, and next year is the only thing a reader is actually trying to work out. One item refuses to sit in the grid at all. A temporary differenceA gap between the accounts and the tax computation that later closes by itself. Both eventually reach the same total for the same item, and only the year each one lands in differs. A temporary difference is also called a timing difference. unwinding moves the cash tax and the deferred tax by the same amount in opposite directions, so the charge and the effective rate never feel it.

A cause that pulls the rate down and will not be there next year is borrowed, and treating a borrowed rate as a normal one is the costliest mistake in this subject. Anjani Stationers' loss relief sits in exactly that box. Rs 14,00,000 of accumulated losses came in with Chitra Binding, Rs 8,00,000 of them were used in year two, and Rs 6,00,000 remain. The relief is real, the saving of Rs 2,00,000 is real, and it is a balance running down. The disallowed expenses sit in the opposite box: they push the rate up and they will be there every year the same spending happens. Two live causes, two very different lifespans, pulling in two directions, and the 21.1 per cent is what is left when they are netted.

How is a tax reconciliation actually read?

Top to bottom, and it takes about a minute. The statement starts with profit before tax, applies the statutory rate to it to get the tax the year would have cost if nothing were different, then lists one row for each named reason the actual charge differs, and ends at the charge that appears in the profit ladder. Most companies publish it in the notes behind the statements. Some present it in rupees, some in percentage points, and a few give both columns side by side. The shape is the same either way.

The reconciliation is the only document in a set of accounts that names the causes and sizes them. Every other route to the same answer is inference, and this one is not. A reader without it can see that the rate is 21.1 per cent and can guess why. A reader with it can say that Rs 2,00,000 of the gap is loss relief with a finite balance behind it and Rs 50,000 of it is a permanent add-back, and can then do something useful with both facts. Sizing the causes is the difference between noticing a number and reading it. If a business publishes no reconciliation, the honest thing for a reader to say is that the causes are unknown, not to invent a plausible reason.

Why is the rate 21.1 per cent this year, and what would move it next year?

Two causes, pulling opposite ways, with the downward one three times the size of the upward one. The whole arithmetic sets out as rows, short enough to hold in mind.

The rate fell by 3.9 percentage points, and Rs 2,00,000 of the movement came from a balance with Rs 6,00,000 left in it. Most of this year's fall was borrowed rather than earned. The two steps in the drawing below make the shape of the year obvious. The bar starts at Rs 9,50,000. The disallowed expenses push it out to Rs 10,00,000. On its own that step would mean an effective rate of 26.3 per cent, above the statutory one. The loss relief then pulls it back to Rs 8,00,000. Without the relief this business is a slightly above-statutory taxpayer, not a below-statutory one, and the reconciliation is the only thing that would have shown it.

A rate read on its own says almost nothing. A rate read in a row says nearly everything, and four years side by side settle it. Year one carried tax of Rs 12,00,000 on profit before tax of Rs 50,00,000, a rate of 24.0 per cent. Chitra Binding was acquired only at the start of year two, so no loss relief was available before then. Year two is the 21.1 per cent worked above. Years three and four are projections on one stated assumption: that profit before tax holds at Rs 38,00,000 and the same Rs 2,00,000 of expenses keeps being disallowed.

Once the loss relief is exhausted the only cause left standing pushes upwards, so the rate does not return to 25 per cent but goes past it, to 26.3 per cent. The overshoot is the part almost everyone gets wrong. The permanent add-back never goes away, so a reader who assumes the rate reverts to the statutory rate has still understated tax by Rs 50,000 a year. In year three the remaining Rs 6,00,000 of losses gives relief of Rs 1,50,000 and the charge is Rs 8,50,000, a rate of 22.4 per cent. In year four there is nothing left to relieve, the charge is Rs 10,00,000, and the rate is 26.3 per cent. Three years, three different rates, one unchanged business.

Four settings of the switches above carry the whole lesson. Switch every cause off and the rate is 25.0 per cent. No other setting makes the effective rate and the statutory rate agree. Switch off the loss relief alone and the charge rises to Rs 10,00,000, a rate of 26.3 per cent. Switch on the invented concessional deduction alongside the two real causes and the charge falls to Rs 6,50,000, a rate of 17.1 per cent. Switch on the timing difference and something else happens entirely. The charge stays at Rs 8,00,000, the rate stays at 21.1 per cent, and the cash tax rate alone moves, from 16.3 per cent to 18.9 per cent. Three of the four causes move the rate. The fourth moves only the timing of the payment, and that distinction is the one most readers of a tax line never make.

How does an analyst turn a reconciliation into a forecast rate?

Nobody reads a reconciliation for its own sake. An analyst building a three-year model, a credit officer sizing a facility, and an owner deciding what to draw all need one number: the rate this business will actually bear in a normal year. Nobody uses the reported rate for that, and nobody uses the statutory rate either. Each of them builds a third rate, and the reconciliation is the only input.

A practitioner keeps only the causes that will still be there next year, drops the ones that will not, and calls what is left the normalised rate. For Anjani Stationers the normalised rate is 26.3 per cent, not the 21.1 per cent it reported. The procedure is three steps and it takes a few minutes. Read the reconciliation and list every named cause. Mark each one as recurring or finite, using the two-question grid above: a permanent add-back is recurring, a loss balance is finite, a one-offAn item that belongs to a single year and is not expected to happen again, so a reader who is describing a normal year should take it out rather than carry it forward. settlement is finite, a concessional treatment is recurring only while its conditions hold. Then rebuild the rate with the finite ones removed. The result is what the business would bear with nothing borrowed, and it is the figure that belongs in a forecast.

Notice that the practitioner ends up holding three different rates for one year and is perfectly comfortable with it. The reported rate describes what happened. The normalisedRestated so that items belonging to one year only are taken out, leaving what a typical year would look like. A normalised rate is an analytical adjustment made by the reader, not a figure any company publishes. rate describes what a normal year would look like. The cash rate describes what leaves the bank. Each of the three answers something the other two cannot, and asking which of them is the real rate is the wrong question.

Anjani Stationers Private Limited and Chitra Binding Works Private Limited are invented.
Educational material. Not advice on any investment, tax, budget or market position.