The Evidence-Based Revenue Forecasting (EBRF) report has been systematically developed and institutionalized over the past five years under the World Bank–supported project Pakistan Raises Revenue, specifically pursuant to Disbursement Linked Indicator-2 (DLI-2). This reform initiative has been spearheaded by the Directorate General of Revenue Analysis (DGRA) within the Federal Board of Revenue (FBR), with the objective of strengthening analytical capacity, improving transparency in fiscal projections, and embedding a culture of data-driven policymaking within tax administration. Over time, the EBRF framework has evolved from a project-based deliverable into a structured forecasting system grounded in empirical data, macroeconomic linkages, and tax-specific behavioral parameters. The tax-wise forecasting model incorporates historical revenue trends, elasticity estimates, sectoral growth projections, policy adjustments, and compliance dynamics, thereby ensuring that projections are not merely extrapolations of past trends but are anchored in underlying economic fundamentals—Evidence Based Revenue Forecasting Report FY 2026-27
A retailer can remain outside a registration campaign while a salaried employee pays through payroll before receiving wages. A documented business supplies invoices, bank records and returns, yet may still have to pursue the state for repayment of excess tax. This unequal relationship is the central problem behind Pakistan’s recurring revenue disappointments. The latest failure to enlist traders is a test of who can resist the fiscal state and who cannot.
According to the retailer figures reported on 1 November 2026, a fixed scheme attracted 1,750 participants from an estimated 3.6 million potential non-filers, with only 220 fresh filers. The respective proportions are approximately 0.0486 and 0.0061 percent. Most participation did not represent first entry into the filing population. Registration, filing and payment are different outcomes; counting one as another conceals the distance between a campaign and genuine enforcement.
The revenue collection reporting should be precise and transparent. The Federal Board of Revenue (FBR) on its website has stopped showing month wise collection after fiscal year 2024-25! The biannual tax collections, however, for July-December 2025 and January to June 2026 are available. Since July 2026, on has no option but to quote newspapers/media reports—which are not only at variance with each other but also lack any official authenticity.
The organisation, which is forcing taxpayers and withholding agents to comply digitally and move to online payments through e-invoices, has itself such a lethargic and non-transparent data reporting in the age of AI and blockchain technology. FBR has its own 100 percent owned market-waged Pakistan Revenue Automation Limited (PRAL) company with competent Board of Directors, some independent members are from reputed technology companies. Yet its own website is clumsy in design, users-unfriendly and data-reporting-deficient.
First quarter reporting by an English daily placed July–September 2026 net receipts at Rs. 3083.5 billion against the target of Rs. 3053 billion—an excess of Rs. 30.5 billion. There was a mood of jubilation in FBR claiming before Press that aggregate target was exceeded!
After purported record receipt of returns (7.6 million) by October 31, 2026, all news reports are showing a fourth-month (July-October 2026) shortfall ranging from Rs. 129 billion to around Rs. 96 billion.
Different cutoffs and target figures in media require reconciliation and official confirmation by the FBR and/or Ministry of Finance. The weakness in collection after receipt of over 7.5 million returns on the last day of October 2026, despite the earlier achievements (sic) warrants scrutiny on its own terms.
For achieving the target of ongoing fiscal year (2026-2027), FBR’s Evidence Based Revenue Forecasting Report FY 2026-27 observes:
“For FY2026–27, revenue collection—excluding the impact of any new taxation or budgetary measures—is projected at Rs. 14,500 billion. This represents a 11.7 percent increase over the expected collection of Rs. 12,983 billion for FY2025–26, as reported in Table 1. In absolute terms, the projected increment amounts to Rs. 1,517 billion. This increase has been estimated by applying tax-specific buoyancy coefficients to projected macroeconomic indicators, thereby capturing autonomous revenue growth under existing policy parameters. In other words, the forecast isolates the structural responsiveness of the tax system to economic expansion, without incorporating discretionary policy interventions such as rate changes, new levies, or withdrawal of exemptions”.
In view of above, meeting an aggregate cash target on monthly basis is officially of a limited achievement—what happened on October 31, 2026 after substantial increase in new returns needs a thorough probe. Collection can rise through prices, imports, rates, litigation recoveries or the acceleration of payments. None necessarily demonstrates an expansion of the underlying taxpayer population. Equally, a shortfall can reflect economic weakness as well as poor administration. The relevant question is how much recurring revenue comes from previously concealed taxable capacity, after separating these influences.
This distinction is acknowledged by the institution most closely associated with Pakistan’s fiscal targets. The IMF’s 2026 review says that important transformation interventions continue primarily to target existing taxpayers. That admission should shape the debate. It is insufficient to announce more technology when its practical reach repeatedly stops at businesses already supplying the administration with information.
Also Read: Why Civilisations Collapse—V
The political economy is familiar. Organised trading interests can mobilise collectively and negotiate. Individual employees and compliant businesses cannot readily withhold taxes already deducted or collected by intermediaries. Administrative convenience then becomes distributive policy: the state repeatedly approaches the accessible taxpayer. Over time, this rewards resistance and weakens the competitive position of enterprises that disclose their activity.
Such criticism must distinguish prosperous concealment from subsistence trade. A small shop may owe no income tax after allowable expenses and thresholds. A non-filer may already bear withholding taxes. Absence from a return database is an investigative lead, not a calculation of tax evaded. Enforcement becomes both more defensible and more productive when it begins with high turnover, valuable commercial assets and repeated inconsistencies between purchases and declared sales.
The same care is needed when discussing refunds. The Karachi Tax Bar Association’s October representations addressed delays in repayment and due process in recovering advance tax. These complaints identify a serious administrative problem without establishing that every collection is unlawful. Publication of refund payments alone cannot answer it: payments are a flow, while unresolved claims form a stock. Both must be reported, together with ageing, rejection reasons and successful appeals.
A legitimate advance payment is part of a lawful collection system. Pressure for payment before its legal due date, or beyond properly determined liability, is different. A receipt credited to a future period improves today’s cash position without adding a taxpayer. Every performance dashboard should identify the tax period, statutory due date and basis of payment. Unsupported allegations of a specific amount should give way to a demand for transaction level audit evidence.
For firms, delayed refunds can function economically as compulsory lending to government. Working capital that should finance inventories, wages or exports remains unavailable. That cost belongs in any evaluation of tax administration. An agency should not receive credit for a collection increase while the cost imposed on compliant enterprise disappears from the account.
What, then, are the reform partners doing? Their activities are real and their roles differ. Pakistan Raises Revenue began as a $400 million World Bank programme; the Bank subsequently approved another $70 million. Its project paper connects support to FBR’s transformation plan and identifies McKinsey’s technical assistance in designing activities. These facts establish a chain of responsibility for design and evaluation. They do not establish a consultancy fee or prove that the entire financing envelope was spent wastefully.
The ADB package comprises a $300 million policy loan and a guarantee of up to $500 million. A guarantee is not expenditure on FBR software or advisers. REMIT’s stated work covers institutional capacity and compliance, while its HMRC partnership includes data and international taxation.
Criticism should examine the attributable results of each intervention, including any successor arrangements, rather than collapse different institutions and financial instruments into one accusation. A common evaluation framework is possible.
Publish actual expenditure, procurement terms, deliverables and ownership of systems. Measure active new taxpayers, sustained payments, independently confirmed sales coverage, audit recoveries surviving appeal, refund turnaround and compliance costs. Compare participating sectors with credible baselines adjusted for economic conditions. An invoice platform is successful when it improves truthful reporting at reasonable cost, not merely when a contract is delivered.
Institutional responsibility cannot be outsourced. Ministers decide whether influential groups face consistent enforcement; legislators determine the tax base and safeguards; FBR applies the law. Advisers cannot supply political commitment, but they can refuse to label an unevaluated installation transformative. Donors should demand publication of adverse findings as firmly as they demand implementation milestones.
The case for a simpler retail system also deserves accurate arithmetic. Joint work published in 2020 advocated a low rate and effective point of sale connectivity. The 2024 proposal used a $300 billion turnover assumption and combined four per cent sales tax with one per cent income tax. Its worked example calculated income tax on sales inclusive of sales tax, producing an effective combined rate of 5.04 per cent. Four percent alone is a different proposition.
Using the earlier $300 billion assumption, and scaling it by the ratio between the PBS historical FY2023 GDP figure of $338.9 billion and the October 2026 revised FY2026 figure of $450.4 billion, gives approximately $398.7 billion. At four per cent, the arithmetic yield is $15.95 billion. At 5.04 per cent, it is $20.09 billion. This is a fresh, reproducible sensitivity calculation, but proportional growth with nominal dollar GDP is an assumption, not a retail survey. The historical base and fiscal year alignment also require validation.
There is a demanding cross check. The Economic Survey put household consumption at 82.64 per cent of GDP in FY2026; the contemporaneous GDP estimate was $452.1 billion. Together they imply household consumption of about $373.62 billion.
Even applying four percent to that entire amount gives $14.94 billion, before excluding services, imputed consumption, protected necessities and other nonretail components. This uses the internally consistent survey vintage rather than attaching its consumption share to the later revision as though both were newly measured.
Retail purchases by institutions and genuinely unrecorded activity can alter the comparison. But the gap must be evidenced; it cannot be filled by adding an arbitrary informal economy percentage. Official national accounts already estimate some informal activity. Nor can wholesale turnover and the resale of the same goods be accumulated repeatedly to manufacture a final retail base.
There is another distinction between a tax potential and additional revenue. Consumers already pay taxes embedded upstream; traders may already pay withholding and income taxes. A replacement regime must deduct receipts it displaces, honour credits and refunds, and allow for exemptions and less than complete compliance. A four percent addition on top of existing sales tax would defeat the stated purpose of lowering the burden. A turnover income tax also needs a margin test: one per cent of sales absorbs one fifth of a five per cent profit margin.
The institutional alternative should begin with a time limited retail demonstration in selected large markets. Match supplier sales and imports to retailers, resolve duplicate identities, and invite corrections before assessment. Give small firms a simple return and accessible payment channels. Record cash sales as well as digital receipts. Protect data, publish audit criteria and provide an effective appeal against erroneous matching. Neither artificial intelligence nor geotagging can substitute for lawful proof.
Policy should be separated from the collection hierarchy so that revenue pressure cannot continually rewrite the rules. An independent tribunal and enforceable taxpayer rights should make correction practical.
Federal and provincial coordination must respect their respective tax assignments; a single interface need not mean extinguishing provincial authority. The retail experiment should be evaluated against the broader VAT architecture, not installed as another overlapping permanent regime.
Pakistan’s choice is not between tolerating evasion and intensifying pressure on every shop. It is whether public authority will distinguish taxable capacity, apply clear rules and account for its own performance.
The powerful bazaar deserves credible enforcement; small traders deserve simplicity; compliant taxpayers deserve their money back when it is due. Reform financing should purchase that institutional change. Until it does, transformation remains an assertion that the collection accounts alone cannot validate.