Diminishing returns of the federal development budget

There is a number buried in the technical annexures of every PSDP document, waiting for someone to do the arithmetic. Yet, despite its significance, the Planning Commission rarely leads with it. That number is Pakistan's Incremental Capital Output Ratio (ICOR), and what it reveals is a silent indictment of how development spending is managed.

The ICOR refers to how many rupees of capital must an economy deploy to generate one additional rupee of GDP. A low ratio signals productive, efficiently deployed investment. As it climbs, the same developmental output costs enhance progressively.

During Pakistan's growth years in the mid-2010s, when GDP expanded at 5-6% annually, the effective ICOR was roughly 3 to 4. Pakistan's ICOR has since drifted to between 5 and 7. To generate one additional rupee of GDP that once required roughly Rs3 to 4 of capital formation, Pakistan now requires Rs5 to 7.

Pakistan's GDP stands at an estimated Rs120 trillion in FY2025-26. Against this base, the federal PSDP of Rs1 trillion, which was further revised down to Rs837 billion mid-year, represents barely 0.6% of GDP. At an ICOR of 5 to 7, that spending generates at most Rs120-167 billion of additional GDP annually, equivalent to 0.10 to 0.14 percentage points of growth. A country targeting 4-5% annual growth cannot reach that destination on a public investment engine this small and this inefficient.

The Rs1 trillion original PSDP allocation sounds substantial until held against the throw-forward: over Rs10 trillion in costs already committed to projects at various stages of implementation. Pakistan has written cheques it cannot cash for a decade or more.

Project data makes this concrete. The Hyderabad-Sukkur Motorway stands at Rs399 billion on its second cost revision. Dasu Hydropower Stage-I carries an approved cost of Rs510.9 billion, far beyond its 2019 ECNEC sanction, after repeated supplementary approvals as rupee depreciation inflated imported plant and civil works. Diamer Basha Dam's land acquisition component alone is on a second revision at Rs174.7 billion. Mohmand Dam, approved in 2018 at Rs309.6 billion, still has Rs173.8 billion outstanding.

Capital efficiency does not deteriorate for a single reason. Four forces compound each other. First, portfolio fragmentation: a fixed annual budget divided across over a thousand schemes means most projects receive a fraction of what they need. A road financeable over three years drags on for a decade; during that time, every input becomes more expensive and the output is eventually delivered at multiples of the original estimate.

Second, politically motivated project selection. The IMF's 2025 Governance and Corruption Diagnostic flagged inclusion of schemes without adequate technical appraisal. Projects entering through patronage channels rarely pass cost-benefit tests; they draw allocations for years without delivering the returns that justified the original commitment.

Third, mid-year budget cuts: PSDP allocations fell from 2.5% of GDP in FY2017-18 to 0.6% in FY2025-26, and a further Rs173 billion was cut mid-year, forcing agencies to suspend procurement and renegotiate contracts across hundreds of projects. Fourth, procurement delays and land acquisition bottlenecks treated as surprises each time, despite being entirely predictable.

Public infrastructure investment, when it works, raises the marginal productivity of private capital. It lowers the cost of doing business, attracts complementary private investment, and generates an output multiplier of 1.2 to 1.8 under conditions of efficient execution. Under current conditions, those multipliers do not apply. The IMF's 2024 assessment placed Pakistan's public investment efficiency gap at 38%, meaning Pakistan captures less than two-thirds of the output its capital spending should generate.

Even if every rupee were deployed at Pakistan's best-performing ICOR of 3 to 4, Rs1 trillion could add at most 0.25 to 0.33 percentage points of GDP growth. At the current ICOR of 5 to 7, the contribution shrinks to 0.10 to 0.14 percentage points. The remaining four-plus percentage points of growth ambition must come from private investment, exports and productivity gains that functional public infrastructure should be catalysing. A dysfunctional PSDP does not simply underperform; it suppresses the private investment multiplier that makes public capital worthwhile.

The Uraan Pakistan framework, which guides PSDP priorities across export competitiveness, energy transition, agriculture and digital infrastructure, is analytically sound. Its weakness is structural: the PSDP machinery is fragmented, revision-prone and built on annual rather than multi-year appropriations.

What needs to change for PSDP 2026-27

The proposed PSDP 2026-27 of Rs1.126 trillion comes with only Rs165 billion of uncommitted headroom. The reforms required to make that headroom count already form part of the Planning Commission guidelines. What has been lacking is the institutional architecture and the administrative and political will to enforce them.

A completion-first rule reserving at least 70% of annual PSDP for projects completable within three years, with a hard cap on throw-forward at no more than five times the annual allocation. This is the single most direct lever for portfolio rationalisation.

Performance-linked fund releases tied to verified physical progress, not merely financial utilisation. Ministries unable to demonstrate on-ground construction milestones do not draw the next tranche. This realigns incentives immediately.

Mandatory economic return thresholds requiring a minimum 12% Economic Internal Rate of Return for all major project approvals at each revision, with no exemptions for politically sponsored schemes.

Statutory sunset clauses: any project not commenced within 24 months of approval, or not completed within 150% of its original timeline, is automatically de-listed. Re-entry requires full re-appraisal, removing the incentive to keep dormant projects alive indefinitely.

A medium-term capital budget shifting from annual appropriations to rolling three-to-five-year commitment ceilings, with future-year costs provisioned at the point of project approval. Annual budgeting for multi-year projects is a structural absurdity that guarantees fragmentation and ICOR deterioration.

Public transparency: a real-time dashboard showing project-level expenditure, physical completion and cost revision history. When revisions are visible and searchable, the political cost of approving them rises.

To conclude, Pakistan has a GDP of Rs120 trillion, a federal development budget trimmed to Rs837 billion mid-year, and an ICOR of 5 to 7: nearly double the ratio associated with high-growth developing economies. That combination means public investment is contributing barely one-tenth of a percentage point to annual GDP growth. Ten trillion rupees in committed but unfinished spending, over 1,000 active schemes, and a 14-year average project life are structural failure to be corrected, and PSDP 2026-27 is the immediate opportunity. Bringing the ICOR down from its current range toward the 3 to 4 would, at unchanged budget levels, more than double the growth contribution of public investment. The reforms are known. What is required now is the resolve to make them stick.