When Governments Change, Reforms Stall: Pakistan’s Policy Continuity Problem

Abstract

Pakistan has no shortage of reform agendas. Its harder problem is keeping long-term reforms alive across political transitions without insulating them from democratic change. Development visions, civil-service reform and local-government restructuring illustrate a recurring tension between the time required for institutional reform and the shorter political cycles in which governments operate. Pakistan already has a sophisticated system for monitoring development projects, but project monitoring does not fully preserve the policy memory of reforms that span ministries, programmes and governments. This policy brief argues for a shift from protecting individual policies to protecting policy memory. It proposes four practical measures: Policy Transition Notes for major long-term reforms; a digital National Reform Register; explicit Continue, Modify, Terminate decisions by incoming governments; and stronger parliamentary reporting on reform continuity. None would bind successor governments to inherited agendas. Instead, they would ensure that when governments change policy, they do so with a clear record of what was attempted, what worked, what failed and what remains unfinished.

Keywords: Policy Continuity; Public-Sector Reform; Political Transitions; Institutional Memory; Governance; Pakistan

The Reform Cycle Pakistan Keeps Repeating

Pakistan does not suffer from a shortage of reform plans. National development visions have promised structural transformation. Civil-service commissions have revisited familiar questions of specialisation, performance, tenure and accountability. Local-government systems have repeatedly been redesigned. New governments arrive with new priorities, new programmes and often new language for problems the state has been trying to solve for years.

Some of that is democracy working as intended. Elections should matter, and a new government must be able to reject a predecessor's priorities, redesign programmes or abandon policies that no longer make sense. The problem begins when political change also wipes the institutional slate clean. Pakistan's own Planning Ministry now acknowledges this risk. In explaining URAAN Pakistan, the government has described earlier initiatives, including Vision 2010 and Vision 2025, as having been derailed by political instability, policy reversals and a lack of continuity. URAAN has consequently been presented as an attempt to create broader ownership, measurable outcomes and an implementation mechanism less vulnerable to political disruption^[1].

That diagnosis deserves attention beyond URAAN itself. The underlying problem is structural: long-term reform runs on policy time, while governments run on political time. Civil-service reform, digital transformation, decentralisation or education reform cannot reasonably be completed within the political life of one minister—or sometimes one government. They depend on accumulated knowledge: what was tried, what encountered resistance, which assumptions proved wrong, which indicators moved, what money has already been spent and what commitments remain outstanding.

When that knowledge does not travel with the reform, the next government may inherit a programme but not its memory. Pakistan's policy-continuity problem is therefore not simply that governments change policies. It is that the state lacks a standardised mechanism for ensuring that the history of a major reform survives the government that launched it.

What Discontinuity Looks Like

Civil-service reform offers perhaps the clearest example. Pakistan has been diagnosing weaknesses in its bureaucracy for decades. A recent evidence review by the Institute of Development Studies found a substantial body of literature on the structure and problems of Pakistan's civil service, but considerably less evidence on the actual implementation status of reforms that had already been agreed.^[2] Andrew Wilder reached a related conclusion more than a decade earlier. He noted that Pakistan had been studied repeatedly by administrative reform committees and commissions, yet argued that the central obstacle was not simply lack of knowledge about what needed to change, but the politics surrounding reform and the limited constituency available to sustain it.^[3]

The issue remains current. On 19 September 2026, the Ministry of Planning, Development and Special Initiatives convened political leaders, policy experts and other stakeholders for a multi-party consultation on civil-service reform. The stated purpose was telling: to build broad-based ownership of a long-term reform agenda. The proposed framework itself followed approximately fifteen meetings of the Prime Minister's Committee on Civil Service Reforms.^[4] The consultation is encouraging, but it also raises the question at the heart of this brief: what happens to those fifteen meetings; the evidence considered, compromises reached, options rejected and implementation logic developed; if political ownership changes three years from now?

Punjab's experience with local government illustrates the problem differently. The province's official legal repository records major local-government frameworks in 2013, 2019, 2021, 2022 and 2025.^[5] The 2019 legislation went further than redesigning the system: it dissolved local governments elected under the 2013 framework before completion of their five-year term. In 2021, the Supreme Court declared the premature dissolution unconstitutional and ordered the local governments restored.^[6]

The point is not that local-government legislation should never change. Different elected governments can legitimately hold different views about decentralisation. The lesson is narrower: repeated institutional restructuring carries transition costs of its own. Structures change, responsibilities move and officials adapt to new rules while the underlying policy challenge; effective local governance; remains. This is why policy continuity should not be confused with keeping a particular law, programme or government agenda permanently in place.

Protect Memory, Not Permanence

A policy-continuity framework that makes inherited policies difficult to change would be bad democratic design. An incoming government may have campaigned precisely on changing them. It may possess new evidence. Fiscal circumstances may have shifted. An intervention may simply have failed. The state therefore does not need to preserve every policy. It needs to preserve enough information to allow the next government to make an informed choice about it.

The distinction is simple:

Policy permanence asks whether the old policy survives.

Policy memory asks whether the state remembers enough to understand what it is changing.

This matters because Pakistan already demonstrates that continuity across a political transition is possible under certain conditions. The 2023–24 International Monetary Fund Stand-By Arrangement crossed an outgoing elected government, a caretaker administration and a newly elected government. The IMF's final review recorded that the programme remained broadly on track through the transition and that the coalition government taking office in March 2024 committed to continue policies consistent with the arrangement.^[7]⁷ This did not mean that every reform moved smoothly or that political ownership was identical across administrations. It showed something more limited but useful: the reform framework, benchmarks, responsibilities and unfinished actions remained visible while governments changed.

This is not an argument for reproducing IMF conditionality across domestic policymaking. It demonstrates something narrower: reform continuity becomes easier when commitments, evidence, responsibilities and unfinished actions are explicitly recorded. Domestic reform should not need an external programme to achieve that basic level of institutional memory.

Pakistan Already Tracks Projects Well

The answer is not another authority. Pakistan already has substantial monitoring machinery within the government. The Planning Commission's PC-I to PC-V system follows development interventions from design through implementation, completion and post-completion assessment. The current evaluation architecture includes completion evaluation, mid-term review, digital monitoring and post-completion assessment. PC-V is designed to assess operational performance and sustainability annually for five years after project closure.^[8]⁸ This is more sophisticated than the familiar claim that Pakistan simply “does not monitor implementation.”

But there is an important distinction between project memory and policy memory. A reform may comprise multiple development projects, legislative changes, administrative decisions, budget allocations and programmes spread across different institutions. Knowing whether individual projects met physical and financial targets does not necessarily tell an incoming government why the larger reform took its present shape, which alternatives were rejected, where political or administrative resistance emerged, or what decisions remain unresolved.

Pakistan therefore does not need to replace its existing monitoring architecture. It needs to connect that architecture to political transition.

What Other Governments Do Differently

This is not a uniquely Pakistani problem. Recent OECD work on centres of government emphasises the importance of permanent administrative capacity in bridging the political-administrative interface, stewarding long-term priorities and coordinating policy across government. It also highlights the role of information, processes and institutional capability in maintaining effective government when political leadership and priorities change.^[9]

New Zealand provides a particularly useful mechanism. After ministers are formally appointed, departmental chief executives must provide Briefings for Incoming Ministers. These concise documents cover the agency, major outstanding policy issues, current programmes, forthcoming legislation and significant decisions likely to require attention in the first months of a new administration. Draft briefings are usually prepared before the outcome of the general election is known and are adjusted once the incoming government's priorities become clear.^[10]

That detail matters because the briefing is not designed to defend the outgoing government's agenda. The permanent administration prepares the institutional record before knowing who will inherit it. Pakistan need not copy New Zealand's model: its political, administrative and federal structures are different. But the underlying principle travels well: an incoming government should inherit an intelligible account of the state it is being asked to govern.

Four Practical Reforms

1. Introduce Policy Transition Notes

For every designated major, multi-year federal reform, the responsible ministry or division should prepare a Policy Transition Note (PTN) when a government changes. This should not become another 100-page report. Five to ten pages should be enough to record the reform's objective; original evidence and rationale; implementation status; expenditure and outstanding commitments; results to date; unresolved bottlenecks; major legal or international obligations; stakeholder commitments; risks; and decisions requiring attention.

Crucially, the note should be prepared by the permanent administration rather than the outgoing minister's political office. The question it answers is not, “Why should the new government keep our policy?” It is, “What does the new government need to know before deciding what to do with it?”

2. Build a National Reform Register

The Planning Commission should maintain a digital National Reform Register covering designated long-term federal reforms. It need not contain every government initiative. Inclusion could be limited to reforms that cross ministries, span several years, involve significant public expenditure or require substantial legislative and administrative change.

For each reform, the register should identify its objectives, responsible institutions, milestones, key indicators, major evaluations and current status. Much of this information already exists somewhere in the government. The problem is that “somewhere” is not a continuity mechanism. A searchable register would give the state a stable institutional reference point even when ministries, ministers or programme names change.

3. Require an Explicit Choice: Continue, Modify or Terminate

An incoming government should not be forced to continue inherited reforms. It should, however, be required to decide what it intends to do with major ones. Within an initial governing period; 100 days could provide a useful benchmark, the government could classify each designated inherited reform as Continue, Modify or Terminate.

For substantial modification or termination, a short public explanation should record the rationale. This creates accountability without creating policy lock-in. More importantly, it turns discontinuity from something that can happen through neglect into an explicit policy decision.

4. Give Parliament a Continuity Lens

Policy continuity should not be an internal executive exercise. A concise annual report to Parliament could identify major multi-year reforms, implementation progress and significant changes in direction during the preceding year. Relevant standing committees could then ask not only how much money was spent, but why a reform was redesigned, whether previous evaluation evidence was considered and what will happen to unfinished commitments.

That matters because continuity is ultimately political. It should be scrutinised politically as well as managed administratively.

A Better Way to Change Course

Pakistan will continue to change governments, and it should continue to change policies when voters, evidence or circumstances demand it. The objective is not to engineer consensus around every reform, nor is it to give unelected officials control over political priorities. It is to stop treating institutional memory as collateral damage of political transition.

The current civil-service reform process offers an opportunity to test a different approach. Its multi-party consultation recognises that reforms extending beyond one government's tenure need broader ownership. Pakistan can take the next step by ensuring that, whether political consensus survives or not, the evidence and implementation history do.

A change of government should permit a change of policy, not a loss of policy memory.

Policy Transition Notes, a National Reform Register, explicit Continue–Modify–Terminate decisions and parliamentary continuity reporting are modest interventions. None requires a new ministry, and none prevents a future government from reversing course. What they do is make sure that when Pakistan starts again, it does so by choice, not because the state has forgotten where it was going.

Notes

1. Ministry of Planning, Development and Special Initiatives, Government of Pakistan, URAAN Pakistan: National Economic Transformation Plan 2024–29 (Islamabad: Government of Pakistan, 2025), particularly the discussion of Vision 2010 and Vision 2025; Ministry of Planning, Development and Special Initiatives, “Minister Highlights Policy Continuity and Implementation Framework under URAAN Pakistan,” Government of Pakistan.
2. Iffat Idris, Civil Service Reform Attempts in Pakistan, K4DD Rapid Evidence Review 116 (Brighton: Institute of Development Studies, 2024), doi:10.19088/K4DD.2025.043.
3. Andrew Wilder, “The Politics of Civil Service Reform in Pakistan,” Journal of International Affairs 63, no. 1 (Fall/Winter 2009): 19–37.
4. Ministry of Planning, Development and Special Initiatives, Government of Pakistan, “Federal Minister Ahsan Iqbal Leads Multi-Party Consultation on Civil Service Reforms,” September 19, 2026.
5. Local Government and Community Development Department, Government of the Punjab, “Governing Laws,” listing the Punjab Local Government Act 2013, Punjab Local Government Act 2019, Punjab Local Government Ordinance 2021, Punjab Local Government Act 2022 and Punjab Local Government Act 2025.
6. Asad Ali Khan and Others v. Province of Punjab through Secretary Government of Punjab and Others, PLD 2021 Supreme Court 770, Constitution Petitions Nos. 48 of 2019 and 7 of 2020, decided March 25, 2021.
7. International Monetary Fund, Middle East and Central Asia Department, Pakistan: Second and Final Review Under the Stand-by Arrangement—Press Release; Staff Report; and Statement by the Executive Director for Pakistan, IMF Country Report No. 2024/105 (Washington, DC: International Monetary Fund, 2024), doi:10.5089/9798400275586.002.
8. Ministry of Planning, Development and Special Initiatives, Government of Pakistan, “Evaluation,” Implementation, Monitoring and Evaluation System; Ministry of Planning, Development and Special Initiatives, “Impact Assessment,” noting annual PC-V post-completion assessment for five years following project completion.
9. OECD, Steering from the Centre of Government in Times of Complexity: Compendium of Practices (Paris: OECD Publishing, 2024), doi:10.1787/69b1f129-en.
10. New Zealand Public Service Commission, Te Kawa Mataaho, “Government Processes Before, During and After an Election,” General Election 2026 Guidance, section “Briefings for Incoming Ministers,” accessed September 21, 2026.

^[1] Ministry of Planning, Development and Special Initiatives, Government of Pakistan, URAAN Pakistan: National Economic Transformation Plan 2024–29 (Islamabad: Government of Pakistan, 2025), particularly the discussion of Vision 2010 and Vision 2025; Ministry of Planning, Development and Special Initiatives, “Minister Highlights Policy Continuity and Implementation Framework under URAAN Pakistan,” Government of Pakistan.

^[2] Iffat Idris, Civil Service Reform Attempts in Pakistan, K4DD Rapid Evidence Review 116 (Brighton: Institute of Development Studies, 2024), doi:10.19088/K4DD.2025.043.

^[3] Andrew Wilder, “The Politics of Civil Service Reform in Pakistan,” Journal of International Affairs 63, no. 1 (Fall/Winter 2009): 19–37.

^[4] Ministry of Planning, Development and Special Initiatives, Government of Pakistan, “Federal Minister Ahsan Iqbal Leads Multi-Party Consultation on Civil Service Reforms,” September 19, 2026.

^[5] Local Government and Community Development Department, Government of the Punjab, “Governing Laws,” listing the Punjab Local Government Act 2013, Punjab Local Government Act 2019, Punjab Local Government Ordinance 2021, Punjab Local Government Act 2022 and Punjab Local Government Act 2025.

^[6] Asad Ali Khan and Others v. Province of Punjab through Secretary Government of Punjab and Others, PLD 2021 Supreme Court 770, Constitution Petitions Nos. 48 of 2019 and 7 of 2020, decided March 25, 2021.

^[7] International Monetary Fund, Middle East and Central Asia Department, Pakistan: Second and Final Review Under the Stand-by Arrangement—Press Release; Staff Report; and Statement by the Executive Director for Pakistan, IMF Country Report No. 2024/105 (Washington, DC: International Monetary Fund, 2024), doi:10.5089/9798400275586.002.

^[8] Ministry of Planning, Development and Special Initiatives, Government of Pakistan, “Evaluation,” Implementation, Monitoring and Evaluation System; Ministry of Planning, Development and Special Initiatives, “Impact Assessment,” noting annual PC-V post-completion assessment for five years following project completion.

^[9] OECD, Steering from the Centre of Government in Times of Complexity: Compendium of Practices (Paris: OECD Publishing, 2024), doi:10.1787/69b1f129-en.

^[10] New Zealand Public Service Commission, Te Kawa Mataaho, “Government Processes Before, During and After an Election,” General Election 2026 Guidance, section “Briefings for Incoming Ministers,” accessed September 21, 2026.