A Surcharge Levied on the Relief Itself: FBR Asked to Fix an IRIS Computation Error

A club ground on the edge of Sharjah's industrial belt. Friday, seven in the morning. Dew still on the grass. Expatriate Bengalis, Pakistanis, Malayalis, Sri Lankans — four nationalities staring at the same scoreboard. The phone stream running eight seconds behind. The scorer sits with pencil in hand, and I notice: the runs are not adding up.

I had not gone to watch a match that day. I had gone to reconcile a total.

I have been reading scorecards for thirty-five years, and the first task is always the same: check whether the column adds up. Because the biggest lie in sport is never born at the wicket. It is born beneath the column, where nobody looks again.

The stadium fell silent, and I began to hear the game.

Now the real story. This week a file landed on my desk. The label read: cricket, Asia. Inside were thirty-three information points. Not one sentence contained a bat, a ball, a pitch, a format, a ranking, or the ICC.

The file concerns Pakistani income-tax administration. An alleged computation defect in IRIS, the Federal Board of Revenue's online return-filing system, which is partially clawing back a parliamentary tax concession on National Savings profit.

The label is wrong. The file is not disposable. Because the machine at work here is one I recognise: a scoring system that counts not only runs, but tax.

Context

Let me lay out what the file holds.

Three National Savings products sit at the centre: Behbood Savings Certificates, the Pensioners' Benefit Account, and the Shuhada Family Welfare Account. Under clause (6) of Part III of the Second Schedule to the Income Tax Ordinance, 2026, profit on these schemes enjoys a tax concession. Section 4AB is the surcharge provision — an additional levy computed as a percentage of income tax.

The claim has been raised by Waheed Shahzad Butt, Chairman of the Public Interest Litigation Committee of the Lahore Tax Bar Association (LTBA-PILC). He has written to the FBR Chairman, with copies to the Members (IR-Legal), (IR-Operations), (IR-Policy) and the Director General IT. The stated filing deadline is October 31, 2026 — a date that has not been verified and that sits implausibly far away.

One thing must be said at the outset. This report has a single source: the claimant. No FBR response appears in the file. No court ruling. No neutral institutional comment. Every figure below is his computation, and I have checked it for exactly one property — whether it reconciles.

And why a man who writes about sport is writing this tax story has an answer buried in the diaspora. Dubai to Karachi, Sharjah to Lahore — Gulf expatriates send money home, and much of that money rests in precisely these certificates. They file returns remotely, from phones, late at night, in the lunch break. To them IRIS is not an abstract government application; it is the one window through which they settle accounts with home. If the window is warped, the road back feels warped too.

In 2026 I stood in an empty stadium in Berlin, Union against Bayern. That day I learned that absence has an architecture — the ball's echo, the faint hum of VAR, the shouts of players. There is an absence here too: the relief is visible on the form but invisible in the base of computation. And the invisible number is always the loudest.

Core analysis

Now my actual work. My profession runs on people; my habit runs on numbers.

Take taxable income of 11,447,109 rupees. The Behbood and similar portion is 2,484,000. Normal tax is 4,241,199.

Now the Second Schedule reduction: 920,332 minus 124,200 leaves a reduction of 796,132. In IRIS this relief is applied through code 9309.

The surcharge, however, has been charged at 424,120 — ten per cent of the tax. The problem is that this ten per cent sits on 4,241,199, not on the reduced figure. The relief was granted, and then erased from the surcharge base.

After the reduction, tax stands at 3,445,067. Ten per cent of that is roughly 344,507. So the excess collected is 424,120 minus 344,507 — 79,613 rupees.

Now notice the cleanest piece of evidence: 79,613 is exactly ten per cent of 796,132. The overcharge is precisely one-tenth of the relief itself — the signature of an order-of-computation error, not a discretionary assessment.

I ran four numbers through four checks. Reduction: 920,332 minus 124,200 equals 796,132, consistent. Surcharge: ten per cent of 4,241,199 is about 424,120, consistent. Correct surcharge: ten per cent of 3,445,067 is about 344,507, consistent. Excess: 424,120 minus 344,507 equals 79,613, consistent. All four reconcile.

When every column on a scorecard reconciles, you can be certain of at least one thing: the person who did the arithmetic did not make a mistake. The mistake belongs to the system that forced him to do it.

The legal question is this: section 4AB speaks of surcharge on "income tax imposed under Division I of Part I of the First Schedule." Does tax reduced under the Second Schedule fall inside that definition? The entire dispute hangs on that one phrase — and a software patch cannot settle it; only statutory interpretation can.

The machine cannot simply be swapped out, because nobody broke the machine. The language of section 4AB was drafted in one era; the Second Schedule relief was drafted in another logic. When the two are forced into a single form, the system must choose an order. That choice is now quietly manufacturing a tax policy Parliament never passed.

And there is a second layer. If IRIS displays a non-nil payable balance, the filer faces two roads. Pay — and reserve the right to a refund later. Or dispute — and absorb the time, the counsel, the hearings. Both roads impose a cost beyond the tax itself. A wrong number on a form becomes a notice, a default, and a long fatigue.

The contrarian angle

Here is where everyone searches for the error at the wrong address.

Some discuss the percentage of relief. Some discuss the rate. Some discuss the surcharge threshold — taxable income above ten million rupees. Nobody asks the plainest question: which number is the calculation running on?

My experience says the most dangerous error in scoring is never in the runs. It is in the sequence. Who comes first, who comes after — change that one thing and the whole result changes, and nobody notices, because the number still looks like a number.

The second point is more uncomfortable. The claimant argues this is not one taxpayer's misfortune. If the surcharge base is hard-coded alongside code 9309, the same defect returns in every affected return of the filing window. He rejects manual rectification as inadequate and wants a systemic fix plus a public notice. There is a clear signal in that: the legislature granted a relief, the executive is taking part of it back — and that gap is the real political event, not the tax figure.

But here is my second doubt. The arithmetic reconciles; that does not make the claim true. Every figure originates from the same source. There is no neutral party, no FBR statement, no judicial observation. I have watched scorers misadd runs many times while the card still looked handsome. Looking handsome and being correct are not the same.

Third, the scale. Filers whose taxable income exceeds ten million rupees and who hold these products — nobody has put a number on that population. It may be small; it may be large. Guesswork does not survive publication.

Fourth, the deadline. If decisions are taken on October 31, 2026 and the date later proves wrong, all the urgency becomes meaningless. An old rule of this trade: you cannot write a story about time until you have verified the date.

Takeaway

So what is the story? 79,613 rupees? No.

The story is that a parliament granted a relief, and a piece of software quietly took back a tenth of it. Nobody noticed, because the number on the form looked legitimate. No penalty, no notice, no commentator shouting.

New media taught me speed; old stadiums taught me to wait for meaning. The waiting here is for the FBR's next statement. If the IRIS logic changes, the defect was real. If a public notice appears, that is a form of admission — one notice means a confession across many files at once. And if nothing comes, a tenth of the relief will be silently collected on every affected return, and expatriate trust in IRIS will thin, increment by increment.

I still believe systems can correct themselves. The only victory in this story will be the moment an affected filer sees the number finally sit right on his return. It will not appear on any scoreboard. It will not sit in any trophy cabinet. It is still the real result.

The ball's echo became a character in the story. Only this time the ball was not a ball. It was a code — 9309.