12 Million vs 7.5 Million: The 'Stop-Filer' Truth Buried in FBR's Filer Database

In FBR's numbers one figure returns again and again: nearly 12 million people filed returns. It is presented at every budget speech and every press release as proof of 'expansion of the tax base.' But the same dataset carries a second number that rarely gets a headline — of these, only about 7.5 million are filing returns continuously. Where did the other 4.5 million go?

I have spent years from 2026 building franchise and board-level memos, and one thing I learned there — when an institution begins using the same metric for two different purposes, the problem is not in the metric, it is in the definition. FBR's case is exactly that. This is not a story of tax evasion; it is a story of database classification.

According to FBR, nearly 12 million people have filed returns for one-time transactions such as property and vehicles. These are people who registered themselves to complete a single transaction, filed a return, and then dropped out of the system. FBR keeps issuing notices to them, but many never learn that their status has become 'stop-filer' — until they reopen their IRIS account. Notices go by email, and for a person who opens IRIS once a year, an email notice is like shouting in a closed room.

One clear figure needs to be recorded here — registered filers and active filers are two different things, and presenting them merged into one number creates statistical illusion. The 12 million is a historical pool containing every person who ever did a one-time transaction. The 7.5 million is the active base that remains in the system continuously. Until FBR reports these two separately, any claim about the 'tax base' will remain incomplete.

The Finance Act 2026 spoke of restrictions on economic transactions — meaning if you are not in filer status, you may face obstacles in buying or selling property or vehicles. On the surface this measure looks logical: connect tax compliance to transactions, so that people cannot ignore notices. But the original text itself acknowledges that this restriction remained compromised by political-economy constraints.

The problem is that a person with no income also became a filer once to buy a motorcycle — why should he now be penalised with an annual return? This question is not administrative, it is policy. Two sources in the original report raised the demand that the filing obligation be removed for non-income earners. The demand is reasonable, but FBR's database cannot yet make this distinction.

For years I have also seen this same pattern in Indian cricket data. When you apply one metric to the wrong population, the metric starts producing noise. In cricket we call it 'small sample' — here I will call it 'wrong sample.' FBR's 12 million pool is a sample that includes every one-time person; treating it as a 'care base' is like evaluating a Test batter from T20 numbers.

FBR perhaps has an opportunity it has not recognised — cleaning the database. If the administration tracked 'transactions' instead of 'persons,' a non-income earner would not need to file returns repeatedly. But for that, FBR must first accept that its 'stop-filer' pool is not an achievement statistic but a classification error.

Data that says 'how many people filed' does not say 'how many people have income' — and policy should be built on the latter question.

On the future question: if transaction-based restrictions are implemented and FBR's communication system remains the same email-centric one, compliance will not genuinely improve — rather, the tendency to avoid formal transactions will grow. The real signal will be when FBR begins reporting 'continuous filers' and 'one-time filers' separately. Until that day, the 12 million figure is a price, and the database's disarray is its actual value.