Pakistan’s rulers repeatedly discover that the Federal Board of Revenue (FBR) is failing. They announce transformation, hire advisers, promise technology and demand higher collections. The institution then explains how it should be transformed. The citizen pays for the exercise, finances the tax machinery and waits for a refund. When disappointment returns, another reform package arrives.
Mir Taqi Mir supplied the diagnosis long before the first consultancy contract: “Mir kya saada hain, beemaar hue jis ke sabab / Usi attaar ke launde se dawa lete hain.” How innocent to seek the cure from the source of the affliction. The tragedy is not that government lacks advice. It keeps returning to arrangements that protect the distribution of power responsible for the failure.
The developments surrounding the prime minister’s 19 October 2026 review sharpen that indictment. According to The Express Tribune, the government has abandoned the proposed new tax agency, requested a staffing restructuring report and decided to accelerate pending legislation enabling redundancy. The threatened officers’ strike was withdrawn. These are reported decisions and developments, not evidence that a new statutory framework has already taken effect.1
The question is whether this retreat from institutional reconstruction will leave Pakistan with a smaller version of the same problem.
Punishment is not an institutional design
The News reports the premier’s call for faster faceless administration and action against corruption. It also reports two suspensions and two recalls to headquarters for investigation concerning an unauthorised statement and a proposed work stoppage.2 The distinction matters: action over a statement is not a finding of corruption, and suspension is not conviction.
Government must investigate misconduct and protect public services. Officers cannot claim proprietary rights over an institution funded by citizens. Neither can an executive demonstrate reform merely by displaying its power to punish employees.
Corruption, deficient performance, redundancy and disagreement about policy are different matters. Each requires evidence and an appropriate procedure. Conflating them produces obedience more readily than competence.
The Inland Revenue Service Officers Association’s supplied statement, internally dated 16 October, opposed fragmentation, questioned consultancy spending and defended specialist expertise. It also pledged to defend cadre identity and announced the strike call.3 Its concerns deserve scrutiny; its institutional preferences deserve no automatic endorsement. Specialists can possess indispensable knowledge while also defending arrangements that need to change.
Government and officers should face the same test: demonstrate how a proposed structure improves lawful collection, lowers compliance costs, protects information and provides remedies. A dispute over who controls the machinery must not displace the question of whom it serves.
Rightsizing without arbitrary power
The parliamentary record makes the personnel issue more serious than a routine reshuffle. The National Assembly committee report dated 23 July 2025 contains proposed sections 22A and 22B on redundancy and restructuring.
The text provides compulsory acceptance of severance and a seven-day representation over its calculation or amount; it purports to make the committee’s decision unchallengeable in courts. Syed Rafiullah’s dissent raises concerns about arbitrary selection and access to justice.4
This is a published committee proposal, not proof of the final law now applicable. Parliament should debate its current text openly. A savings target cannot justify choosing employees through undisclosed criteria or making an administrative decision effectively immune from independent scrutiny. Legal protection should prevent retaliation without becoming permanent protection for incompetence.
There is also an institutional danger. A tax officer who fears removal for displeasing superiors may become less willing to resist an unlawful collection demand. The system can consequently produce more inflated assessments while congratulating itself on discipline. Honest officers need protection for refusing illegality. Taxpayers need protection from officers who obey illegal instructions. These are complementary safeguards.
Proper restructuring begins with functions: which activities are necessary, which are duplicated, which require specialist skills, and which can be automated? Staffing follows that analysis. Starting with a desired number of departures and inventing an organisational justification afterwards reverses the process. Publishing functional reviews, selection criteria, redeployment options and transition costs would help distinguish reform from a purge.
A smaller estimated gap is not recovered revenue
Officials reportedly told the review that the estimated tax gap had fallen from 6% to 5% of gross domestic product. They also cited 957 contracted auditors and a fourfold expansion of audit capacity.2 These are claims requiring published supporting evidence. Recruitment is an input. Sustainable compliance is an outcome.
A movement between two percentages cannot establish what caused it. Were the estimates calculated for comparable periods, using consistent coverage and assumptions? Did economic activity change? Were tax-law changes incorporated? Were outstanding refunds and accelerated collections treated consistently? Which sectors improved, and how much additional revenue survived adjustment, refund and appeal?
The International Monetary Fund’s revenue-gap framework distinguishes non-compliance under existing rules from the effects of policy choices. Its methodological work also cautions that estimated gaps are not exact measures of immediately collectible additional revenue.5 That distinction should govern the debate. Revenue forgone through a legislative concession cannot simply be recovered by ordering an auditor to work harder.
Nor should an audit programme be judged by demands issued. A large demand that collapses on appeal is not a public achievement. Publish additional tax sustained, cash collected, refunds paid, assessment reversals, time spent and the taxpayer’s compliance cost. A performance incentive tied only to collections can reward behaviour that damages both business and the credibility of the state.
The accounts behind the applause
Advance payments from government entities require similarly exact scrutiny. Parliament should seek a reconciliation identifying the payer, liability, period, funding source and eventual adjustment. A profitable public enterprise paying tax legally due is one thing. Treasury-funded payments accelerated to embellish a collection target are another. Without that distinction, an accusation of fiscal manipulation is too broad and an official claim of achievement too easy.
Where money released from the Federal Consolidated Fund finances premature payments entering divisible-pool tax receipts, government is rearranging public cash and its distribution. The consolidated public sector has not necessarily acquired additional resources from outside itself. At entity level, tax credits or refund claims may accumulate. A timing benefit today can become a revenue hole tomorrow.
The divisible pool describes a constitutional revenue-sharing arrangement; it should not be imagined as an unrelated bank account into which money mysteriously disappears. The objection is to presenting fiscal circulation as new tax effort. The amounts and financing chain must be established before attaching a quantified allegation to particular entities.
Refund suppression creates a parallel illusion. A business forced to finance the Treasury may borrow to pay wages or postpone investment. The collection ledger looks stronger while productive capacity weakens. Gross receipts, net receipts and unpaid refund liabilities should therefore be disclosed together. The state cannot demand accurate accounts from citizens while celebrating its own incomplete picture.
What Zaidi gets right and where he stops short
Shabbar Zaidi’s “PRAL in, FBR out” usefully challenges bureaucratic control over Pakistan Revenue Automation Limited (PRAL). His proposal separates policy, technology and residual administration; his second part argues for an autonomous digital institution and integrated transaction information.6, 7 Specialist recruitment, better data and fewer unnecessary encounters with officials are sensible objectives.
His wider framing is insufficient. Administration cannot be detached from the policies it enforces. High transaction deductions, minimum taxes unrelated to profitability, favoured exemptions and frequent changes are choices about distributing burdens. An excellent computer system can execute those choices efficiently while making the economy less fair. Removing discretion at the counter does not remove privilege written into legislation.
The second part also acknowledges capability deficiencies in both FBR and PRAL. That makes the transition question more pressing: which institution will procure, build, own and maintain the systems, and with what independently tested capacity? Corporate form cannot supply missing engineering expertise. Nor does relocating headquarters to Karachi create independence. Authority, appointments, funding, procurement and legal safeguards determine independence more than geography.
Banks can provide valuable information under a lawful framework. But a bank credit is not automatically taxable income: it may represent borrowing, a transfer between accounts, repayment or capital. Matching data generates questions; it does not answer every question about liability. An effective digital administration needs traceable evidence and accessible correction, not merely a powerful ability to issue notices.
Brazil deserves a more accurate reading
Zaidi’s Brazilian comparison supports a specialist technology partner working alongside a revenue authority. It does not establish that the technical partner should inherit every sovereign function. Brazil’s Public Digital Accounting System, known by its Portuguese abbreviation SPED, also reflects cooperation involving federal and state authorities and other participants, as the Federal Data Processing Service (SERPRO) describes.8 Integration matters as much as organisational separation.
His assertion that conventional annual corporate filings became obsolete needs qualification. Brazil’s official guidance explains that Escrituração Contábil Fiscal (ECF), its fiscal bookkeeping obligation, replaced an earlier corporate declaration from the 2014 calendar year. ECF still has an annual submission deadline, generally the last working day of July following the relevant year.9 Changing the form of reporting is not abolishing periodic corporate reporting.
Electronic invoices can support rapid validation while annual accounts reconcile profits, adjustments and liabilities. The two functions can coexist. Pakistan should learn how to connect them, rather than promise that transaction capture eliminates every retrospective obligation. The practical test is whether information already supplied is reused and genuinely redundant reporting removed.
Even the reference to Brazil’s tax authority as “the Lion” needs historical care. Receita Federal traces the symbol to advertising prepared in late 1979 for the 1980 income-tax campaign.10 It was not created as an accolade for modern auditing algorithms. This small correction illustrates a larger requirement: international models should be examined through their actual institutions and records, not persuasive mythology.
Brazil strengthens the argument for capable technology, shared standards and accountable administration. It does not resolve Pakistan’s federal distribution of powers, judicial remedies or relationship between taxation and social entitlement.
A federal agency that also serves citizens
This writer with Huzaima Bukhari have repeatedly proposed integrated national collection, including in “Case for ‘NTA’” in Business Recorder on 27 November 2015. Our books examine the failed reform cycle and offer alternatives.11, 12 The purpose has never been another acronym. It is to reduce fragmentation while placing collection within a coherent framework of representative government and enforceable rights.
A National Revenue Agency should emerge from agreement between the federation and provinces, supported by the legislation needed for its functions. Legislative power to impose a tax, administrative responsibility to collect it and entitlement to its proceeds are distinct. Shared administration need not transfer provincial taxing rights to Islamabad. Provincial participation must be substantive, including governance, data access and transparent settlement of receipts.
Canada offers a stronger federal analogy than an isolated software company. Its Canada Revenue Agency (CRA) administers federal taxes and most provincial income taxes. Quebec is an exception for personal income tax; Quebec and Alberta administer their own corporate income taxes. Customs operations belong to a separate border agency.13, 14 Canada provides lessons in negotiated cooperation, not a claim that one institution collects every tax.
Its governance is instructive too. Eleven of the CRA’s fifteen board members are nominated by provinces and territories. The board oversees management but has no authority over administering or enforcing tax legislation.15 Shared governance does not mean that business representatives should influence individual assessments. Pakistan should build similarly clear boundaries between institutional oversight and decisions about particular taxpayers.
The missing half of our debate is benefit delivery. CRA administers the Canada child benefit and provincial programmes; tax information helps determine entitlements.16 Pakistan should progressively connect its revenue administration with legislated income support and refundable credits, coordinating existing institutions rather than abruptly dismantling them. A citizen with no tax payable should still have a reason to engage with the system.
Digital inclusion matters here. People without reliable connectivity, formal addresses or sophisticated advice must have assisted access. Errors should be corrected without depriving a household of subsistence. Personal information should have specified lawful uses, restricted access and enforceable protections. A national database must not become an unrestricted instrument of surveillance or political pressure.
Justice must be part of the machinery
A stronger collector without stronger remedies is an invitation to abuse. The proposed National Tax Court should consolidate specialist tax adjudication through the necessary constitutional and legislative arrangements, with appointments and administration independent of the collector. Provincial jurisdiction must be settled by lawful agreement, not assumed away.17
Its task would extend beyond clearing a backlog. It should provide accessible hearings, reasoned decisions, consistent interpretation, timely refund relief and proportionate costs where parties abuse proceedings. Recovery safeguards should prevent a disputed demand from destroying a viable enterprise before meaningful review. Constitutional supervision must remain available within the applicable judicial framework.
Technology can assist that court through electronic records, scheduling and publication. It cannot replace the duty to hear a party or explain a decision. A citizen must know whether an error arose in source data, software, legal interpretation or official conduct—and which institution must correct it. Faceless administration must never become responsibility without an identifiable bearer.
A programme measured by results
The immediate programme should begin with an independently checked baseline covering refunds, disputes, collection quality and compliance costs. Publish the expenditure and outcomes of reform projects, including what consultants delivered and what systems remain in use. Commission an open comparison of rebuilding FBR and creating a federalised agency, with transition costs and risks stated honestly. The possibility that a new body might inherit old defects is a design challenge, not a reason to preserve those defects indefinitely.
During transition, pilot common registration, shared data standards and coordinated audits. Protect essential collection and taxpayer services. Set public milestones for expanding successful arrangements, supported by trained staff and independent evaluation. Policy reform should proceed alongside this work: wider bases, lower predictable rates, protection of subsistence and removal of unjustified privilege. Claims about revenue gains must rest on evidence rather than promises that every rate cut finances itself.
Pakistan needs political accountability as much as administrative discipline. Ministers and legislators choose exemptions, approve expenditure and tolerate evasions. They cannot place every failure at the collector’s door while withholding the decisions needed to change outcomes. Equally, officials cannot invoke expertise to escape scrutiny of their own performance.
The real choice is between a revenue state that serves citizens and a collection apparatus that treats them as a source of emergency cash. A National Revenue Agency and National Tax Court should anchor that change. Pruning staff, moving offices and buying software may support it; none can substitute for it. Mir’s patient kept returning to the same dispensary. Pakistan cannot afford to keep paying for the same failed prescription.
Endnotes
- Shahbaz Rana, “PM asks FBR to plug tax gap amid pruning plans”, The Express Tribune, 20 October 2026. Reporting of the 19 October decisions; proposed changes are distinguished from enacted law.
- Javed Aziz Khan and Mehtab Haider, “No room for corruption, underperformance in FBR: PM”, The News, 20 October 2026. Official performance claims and reported disciplinary action.
- Inland Revenue Service Officers Association, press release internally dated 16 October 2026, supplied as IRSOA Press Release 17.10.2026.pdf, pp. 1–2. The statement announced a strike; subsequent reporting records withdrawal of the threat.
- National Assembly of Pakistan, Standing Committee on Cabinet Secretariat, report on the Civil Servants (Amendment) Bill, 2024, dated 23 July 2025. Annex A, proposed sections 22A–22B; Annex B, dissent of Syed Rafiullah. This is a committee-stage text.
- International Monetary Fund, Estimating the Corporate Income Tax Gap: The RA-GAP Methodology, Technical Notes and Manuals 18/02, August 2018, section II. Distinguishes compliance and policy gaps and cautions against equating static estimates with obtainable additional revenue.
- Syed Shabbar Zaidi, “PRAL in, FBR out”, Business Recorder, 16 October 2026.
- Syed Shabbar Zaidi, Part II of the PRAL and FBR argument, supplied text. The discussion of its proposals refers to that text.
- SERPRO, “Sped proporciona arrecadação moderna e simplificada”, 5 January 2009. Describes cooperation among Receita Federal, state finance authorities, businesses and other institutions.
- Government of Brazil, “Entregar Escrituração Contábil Fiscal”, updated 29 July 2026. Explains ECF, its replacement of DIPJ and the general annual submission deadline.
- Receita Federal, “1968 a 1981 — Começa a Era da Secretaria da Receita Federal”, institutional history. Traces the lion symbol to preparations for the 1980 income-tax campaign.
- Huzaima Bukhari and Dr Ikramul Haq, “Case for ‘NTA’”, Business Recorder, 27 November 2015.
- Huzaima Bukhari and Dr Ikramul Haq, Towards Broad, Flat, Low-rate and Predictable Taxes, third revised edition, PRIME Institute, November 2024. Publicly available copy.
- Canada Revenue Agency, “Provincial and territorial tax and credits for individuals”; CRA financial statements also identify Quebec and Alberta as corporate-income-tax exceptions. Corporate tax exceptions: CRA financial statements.
- Canada Revenue Agency, “Structure and operational framework”. Distinguishes CRA from Canada Border Services Agency customs operations.
- Canada Revenue Agency, “Board of Management”. Provincial and territorial nominations and limits on the board’s authority over tax-law administration and enforcement.
- Canada Revenue Agency, “Alberta tax information for 2025”, benefits section. Describes delivery of the Alberta child and family benefit using CRA information and its relationship with the Canada child benefit.
- Huzaima Bukhari, Dr Ikramul Haq and Abdul Rauf Shakoori, “Tax cases backlog”, Business Recorder, 2026. Earlier case for independent specialist adjudication.