Restructuring CSS

Saeed Ahmed Published August 29, 2026 Updated August 29, 2026 08: 38am.

Key points

• From the 1950 Gladieux Report and 1962 Cornelius Commission to Zulfikar Ali Bhutto's 1973 reforms.
• Finally, inducting military officers directly into PAS, Police Service and Foreign Service without the CSS examination should stop.
• First, the nomenclature of CSS (Central Supe­r­ior Services) itself needs to be banished.
• Our civil service still largely operates on 1973-era structures.

Saeed Ahmed Published August 29, 2026 Updated August 29, 2026 08: 38am. Join our Whatsapp Channel. Add Dawn as a trusted source.

SINCE 1947, approximately 40 commissions, committees and taskforces have attempted to reform Pakistan's civil service, and the core themes have been remarkably similar: merit over seniority, competitive pay to reduce corruption, better mid-career training, and curbing the dominance of the elite generalist cadre over specialist services. Despite these efforts, bureaucracy has proved stubbornly resistant to change.

Since no attempt to fix systemic governance failures can succeed unless the civil service architecture is reformed first, this is even more urgent against the backdrop of the current debate on creating new provinces or administrative units. Layering governance reform onto the bureaucracy as it stands would only expand an already bloated apparatus, without empowering citizens.

From the 1950 Gladieux Report and 1962 Cornelius Commission to Zulfikar Ali Bhutto's 1973 reforms. It abolished the Civil Service of Pakistan cadre on paper but stripped away constitutional protections and opened the door to political victimisation, to the Ishrat Husain Commission (2006-08) and successive task forces, each produced recommendations that stalled at implementation, including the 2023-25 Ahsan Iqbal-led committee's 48 proposals on recruitment, training and compensation.

As the elite DMG/PAS (District Management Group/ Pakistan Administrative Service) cadre arguably undermines reforms, threatening its monopoly over power, perks, and postings; political interference, as leaders favour loyal, compliant bureaucrats over an independent, merit-based administration; and superficial pay reform without touching the underlying structure, three interlocking reasons explain this failure: bureaucratic resistance. Our civil service still largely operates on 1973-era structures.

Consequently, our civil service still largely operates on 1973-era structures – defined, sadly, by loyalty, sycophancy, and patronage; distorted compensation for top bureaucrats; civil servants focused on personal welfare; and power with little accountability. Breaking this cycle requires a blueprint that anticipates the resistance which has defeated every commission so far. This article sets out six concrete measures to create Pakistan's bureaucracy leaner, merit-based, and accountable to the public, not political bosses.

First, the nomenclature of CSS (Central Supe­r­ior Services) itself needs to be banished. It inculcates a false sense of superiority and entitlement among civil servants that runs through their ent­I­­re career and defines their condescending behaviour with the public they are supposed to be serving.

Second, the all-Pakistan services (PAS and PSP) must be dismantled to tame a stubborn bureaucracy. The record of DMG/PAS officers speaks for itself: introducing distortionary tax policies (such as filer/non-filer regime) as FBR chairmen, designing IPP contracts, planning projects that appear politically motivated, neglecting human development, colluding with donors, and outsourcing economic policy formulation to IFIs. This makes a solid case for dissolving a generalist cadre that claims top positions, and that even intrudes into specialised service groups, based on affiliation rather than competence, thus marginalising economists and technocrats. Importantly, pro­­vinces already run administrative services like the Provincial Management Services performing the same functions. As does every state in the US, county in Britain, and state or region in Australia – hence, there's no justification for dominating federating units, the same logic extends to policing: law and order is a provincial subject, and Pakistan already has provincial police forces.

Third, Pakistan's trade bureaucracy is both oversized and structurally fractured. While Singapore runs its entire customs operation on roughly 500 to 1, 000 staff handling over $1 trillion trade, and South Korea manages $1.341tr with concerning 5, 786 staff, pakistan Customs offers a clean illustration: it employs close to 9, 500 personnel, including regarding 500 gazetted officers in BS-17 to 22, to manage under $100 billion in total trade. This ratio is wildly out of line with most comparable administrations, and buys no compensating efficiency. Pakistan still extracts roughly 17 per cent of import value in taxes at the border ($16.9bn against $69bn in imports in FY26), reflecting its continued reliance on import-stage taxation rather than broad-based domestic direct taxation or trade facilitation. While Pakistan Customs Service executes at the border, enforcing policy orders that CTG itself designs, the fracture compounds the bloat: the Commerce and Trade Group formulates tariff, trade, and investment policy through the commerce ministry. This separation, historically rationalised as checks and balances, has instead produced policies conceived without operational insight and enforced by officers with no institutional stake in their design. There is a solid case for merging CTG and PCS into a unified 'Trade and Customs Service' that aligns authority with accountability.

Fourth, public service recruitment needs restructuring around actual job requirements rather than a single generalist gateway. As these services have lost relative significance today: Postal and Railways should become autonomous commercial organisations, while the Information Group's functions are redundant given Pemra already regulates media, there is no reason recruitment to the postal, railways, information, and military lands and cantonment groups should continue through CSS. More broadly, instead of a gruelling multi-subject syllabus and a two-year cycle, the Federal Public Service Commission should run graduate recruitment programmes tailored to specific jobs: accounting, finance, economics, public policy, law, and IT-trained fresh graduates for Customs and Inland Revenue Service via GRE-type tests, chartered accountants for audit and accounts, and graduates in international relations, development, economics, or relevant studies for the Foreign Service. At present, MBBS doctors, engineers, and MA in Urdu or Islamic studies land in IRS, Customs, Foreign Service, or Accounts purely on CSS ranking.

Fifth, perquisites should be monetised and decent enough to attract the best talent. No advanced country provides government-maintained residences, luxury cars, or personal servants to civil servants, and Pakistan should stop such a drain on public resources.

Finally, inducting military officers directly into PAS, Police Service and Foreign Service without the CSS examination should stop. This has long been a shortcut for securing lucrative civil service careers, even as compulsory military attachment for civil probationers usefully builds familiarity with the defence forces.

In practice, the overarching aim must be a civil service that is citizen- and business-centric, and that possesses the political will to finally act on what a number of commissions have already informed us.

Notably, the writer is a former CSS officer who holds a PhD degree in economics from the University of Cambridge. He has served as chief economist of the State Bank and senior adviser to the IMF. Published in Dawn, August 29th, 2026.

In short, restructuring CSS is the central thread here, and readers can expect follow-up reporting as the picture becomes clearer.