Islamic finance is a US$5.98 trillion industry whose contract law includes instruments purpose-built for agriculture — the salam forward sale for pre-harvest working capital, and output-sharing cultivation partnerships (muzaraʿa, musaqa) — yet agricultural assets constitute a negligible share of industry balance sheets, and practice is dominated by markup-based structures that economically replicate debt. This paper proposes and evidences an explanation grounded in information economics rather than jurisprudence: the verification gap. Risk-sharing contract forms concentrate commodity-price, production, and effort-observability risks on the financier; where farmer identity, land linkage, crop condition, and delivery integrity cannot be verified at feasible cost, rational intermediaries substitute toward collateral- and invoice-based structures, independently of doctrinal preference. Comparative evidence is assembled from four settings: Sudan, where bank-intermediated salam operated at national scale and contracted from 8.4% to 3.7% of Islamic bank financing (1997–1999) under unhedged price risk; Malaysia, where a state agricultural bank completed whole-institution conversion to Shariah-structured operations while raising its mandated agricultural financing share from 69% to 82%; Pakistan, where a benevolent-loan microfinance institution reports 6.8 million loans with a 99.9% repayment rate secured on community-based accountability; and the multilateral level, where the Islamic Development Bank Group's US$20.6 billion agricultural portfolio indicates that capital availability is not the binding constraint. The paper then specifies a design framework in which parcel-level satellite monitoring, digital identity and registry infrastructure, mandatory crop-loan insurance, and electronic warehouse receipts jointly close the verification gap, and derives four testable propositions for Pakistan — where a constitutional amendment requires all new financing to be non-interest-based from 1 January 2028, including an agricultural credit programme exceeding Rs 3 trillion annually. The framework implies a sequenced product ladder in which risk-sharing modes activate as borrower-level information accumulates, offering a research and pilot agenda ahead of the 2028 deadline.
Keywords: agricultural credit; Islamic finance; information asymmetry; remote sensing; contract design; financial inclusion; Pakistan
JEL codes: Q14 · G21 · G23 · O33 · K12
Suggested citation: AgroHub R&D (2026). "The Verification Gap in Islamic Agricultural Finance: Comparative Evidence and a Technology-Enabled Design Framework for Pakistan's 2028 Interest-Free Transition." AgroHub R&D Working Paper AH-RD-WP-2026-02, v2.0. Lahore: AgroHub Global (Pvt) Ltd.
Two facts motivate this paper. First, under Pakistan's 26th Constitutional Amendment (2024), which operationalises the Federal Shariat Court's 2022 judgment on interest-based finance, all new financing in Pakistan must be contracted on a non-interest basis from 1 January 2028, with existing conventional agreements running to maturity.[1],[2] This applies to an agricultural credit programme that disbursed Rs 2.16 trillion in the first nine months of FY2025-26 against a Rs 3.06 trillion indicative target.[3] Second, the global Islamic finance industry — US$5.98 trillion in assets in 2024, growing at 21% annually — holds agricultural assets in shares too small for its principal statistical reports to disaggregate, with Islamic microfinance and takaful (mutual insurance) at roughly 2% of assets each.[7]
The conjunction is a policy problem of unusual precision: within eighteen months, one of the world's ten largest agricultural economies must convert its entire flow of new farm credit to a contractual family in which the international industry has demonstrated almost no scaled agricultural capability. Understanding why that capability failed to develop — despite the availability, within the same contractual tradition, of instruments specifically designed for agricultural cash flows — is therefore of immediate practical consequence, and of broader interest to the economics of contract choice under asymmetric information.
The paper's argument can be stated in one sentence: the historically limited scale of risk-sharing agricultural finance is better explained by the cost of verification than by the content of the contracts. Where the financier cannot observe the borrower's identity-to-land linkage, the crop's existence and condition, or the delivery chain, contract forms that concentrate crop-linked risks on the financier are rationally abandoned in favour of collateralised or markup-based structures — an adverse-selection and monitoring-cost account consistent with the broader credit-rationing literature on smallholder finance.[20],[21] The contribution is threefold: (i) a conceptual framework ordering the classical contract set by information intensity (Section 3); (ii) a structured synthesis of comparative evidence from Sudan, Malaysia, Pakistan, and the multilateral level (Section 4); and (iii) a design framework with four testable propositions, specifying how parcel-level remote sensing, digital registries, mandatory crop insurance, and electronic warehouse receipts jointly alter the feasible contract set (Section 5), with Pakistan's 2028 transition as the natural experimental setting (Section 6).
The instruments at issue are codified in the Shariah standards of AAOIFI (the Accounting and Auditing Organization for Islamic Financial Institutions) and, for Pakistan, in the State Bank of Pakistan's Guidelines on Islamic Financing for Agriculture (2008), which enumerate them for domestic agricultural use.[12],[13] Exhibit 1 summarises the set. Historically, the salam forward sale is notable as an early codified exception to the general prohibition on selling goods not yet in existence: classical commercial law permitted pre-harvest sale subject to strict specification of commodity, quantity, quality, and delivery — a regulatory response to pre-existing agrarian financing practice, and thus, from an institutional-economics standpoint, one of the oldest documented standardisations of an agricultural forward contract.[6],[13]
| Contract form | Economic structure | Agricultural application | Financier's principal exposures | Information the financier must verify |
|---|---|---|---|---|
| Murabaha (cost-plus sale) | Asset purchase and resale at disclosed markup, deferred payment | Input supply (seed, fertiliser, feed) | Counterparty default | Invoice; borrower identity |
| Ijara (lease) | Lease of productive asset | Machinery, cold storage, irrigation | Residual value; utilisation | Asset condition; borrower identity |
| Istisnaʿ (commissioned production) | Staged payment for constructed asset | On-farm infrastructure | Completion risk | Construction milestones |
| Qard hasan (benevolent loan) | Principal-only loan | Distress smoothing; graduation lending | Repayment | Community standing (see §4.3) |
| Salam / parallel salam (forward sale) | Full prepayment for specified future delivery | Pre-harvest working capital | Commodity price; crop failure; delivery default | Crop existence & condition through season; delivery capacity |
| Musharaka (equity partnership) | Profit by agreement, loss by capital share | Livestock, orchards, processing | Business performance | Operations; accounts; effort |
| Muzaraʿa / musaqa (output-sharing cultivation) | Inputs or land from one party, cultivation from the other; output shared | Tenant-farmer and orchard partnerships | Production; effort observability | Continuous cultivation effort and crop state |
The empirical puzzle follows directly. Industry practice concentrates overwhelmingly at the top of Exhibit 1 — murabaha and functionally similar markup structures — a pattern extensively documented and critiqued in the specialist literature as "debt-dominance."[21] The forms in bold, despite matching agricultural cash-flow timing by construction (payment at planting, settlement at harvest, or settlement in output shares), are marginal everywhere. Notably, the one product-design regularity established in the secular smallholder-finance literature — that in-kind, harvest-synchronised credit achieves the highest observed repayment performance (approximately 98% at One Acre Fund across 180,000+ borrowers)[15] — describes the bold rows' cash-flow geometry almost exactly. The contracts that theory and secular evidence jointly favour are the contracts practice avoids. Section 3 offers an explanation that requires no reference to doctrinal preference at all.
Consider a financier choosing between contract forms for a smallholder portfolio. Under markup structures, the financier's information requirement is an invoice and an identity; under salam, the financier holds a claim whose value depends on an unobserved crop for an entire season; under output-sharing partnerships, returns depend additionally on unobservable cultivation effort — a classic moral-hazard setting. Let v denote the per-borrower cost of verifying the state variables in Exhibit 1's final column. For smallholdings, conventional verification is manual: field visits for existence and condition; local knowledge for identity and tenure; physical inspection for delivery. Estimates from Pakistani practice place per-visit costs in the thousands of rupees against loan sizes in the tens of thousands, so v is large relative to expected contract margin precisely in the segment where the risk-sharing forms are most relevant.
The prediction is immediate: as long as v remains high, intermediaries substitute toward low-verification contract forms regardless of institutional mission — the observed debt-dominance — and the substitution is more severe for smaller holdings, compounding exclusion. Conversely, any technology that reduces v discontinuously (near-zero marginal cost per additional parcel, as with satellite monitoring over free public imagery) shifts the feasible set toward the risk-sharing forms, first for observably verifiable states (crop existence, condition, area) and progressively for behavioural states as transaction histories accumulate. Four propositions operationalise the framework:
Method. The evidence base is a structured comparative synthesis of published sources — regulatory documents, institutional disclosures, and peer-reviewed studies — across four settings selected on outcome variation: a system where risk-sharing agricultural finance operated and contracted (Sudan); a whole-institution conversion (Malaysia); a scaled benevolent-loan system (Pakistan); and the multilateral capital layer (IsDB Group). No primary fieldwork was conducted; the design is observational and the usual case-selection caveats apply (§6).
Sudan's fully Islamised banking system constitutes the longest-running natural experiment in bank-intermediated salam. Bank of Khartoum's documented mechanics — full prepayment to farmers at spot terms for specified future delivery, with resale on delivery — financed real agricultural seasons at portfolio scale.[16] The critical observation is the trajectory: salam's share of Islamic bank financing in Sudan fell from 8.4% to 3.7% between 1997 and 1999, a contraction attributed in the literature to market volatility and price uncertainty rendering the contract the "least favoured" among Sudanese institutions; Bank of Khartoum subsequently institutionalised a ±33% price-movement band beyond which contract terms are considered inequitable to one party.[6],[16] Interpretation: the contract cleared the market at origination — farmers accepted it, banks offered it — and failed at the risk-management layer, exactly where the verification-gap framework locates the binding constraint (P2). Sudan is thus evidence that the form is operable, and simultaneously a specification of the apparatus (offtake commitment, price bands, in-season observation, insurance) that any revival must attach.
Malaysia's state agricultural bank, Agrobank, completed conversion to a full-fledged Islamic bank on 1 July 2015 after a staged 2012–2015 programme covering products, systems, staff, and the existing book — during which its mandated agricultural financing share rose from 69% (2012) to 82% (2015).[17] The institution has since accessed Islamic capital markets (a RM500 million SDG-linked sukuk)[18] and, with the agriculture ministry, introduced Malaysia's first Shariah-structured paddy insurance.[19] Interpretation: conversion of an agricultural lender on a fixed timetable is operationally documented — directly relevant to Pakistani institutions facing the 2028 deadline — and capital-markets funding can be wired to agricultural mandates within the same contractual family. Malaysia does not, however, evidence scaled risk-sharing origination; its portfolio remains predominantly in the low-verification forms, consistent with the framework.
Akhuwat, a Pakistani microfinance institution lending on the benevolent-loan (qard hasan) form, reports 6.8 million loans disbursed, approximately US$1.52 billion, and a 99.9% repayment rate across 800+ branches, with agricultural products in the PKR 20,000–80,000 range; its documented mechanism substitutes community-based accountability — public disbursement in communal venues and locally embedded follow-up — for physical collateral.[14] A randomised evaluation of its interest-free microcredit finds positive borrower returns.[20] Interpretation: within Pakistan's institutional environment specifically, socially embedded accountability generates repayment performance exceeding most collateralised benchmarks (P4). The scored-variable formulation matters: the claim defended here is not that social mechanisms replace underwriting, but that they carry predictive information a data-driven underwriting system should price — a hypothesis testable with standard discrimination metrics on pilot portfolios.
The Islamic Development Bank Group's cumulative agriculture and food-security financing stands at US$20.6 billion across 1,538 operations[8]; its US$10.54 billion Food Security Response Program (2022) had approved US$8.0 billion by late 2024 — over 90% of plan — including US$4.5 billion in trade finance through ITFC.[9] Pakistan's domestic Islamic banking system held Rs 14.47 trillion in assets by end-2025 (22.9% of banking assets; 38.1% of financing), with the largest Islamic bank reporting Rs 84.5 billion in after-tax profit.[10],[2] Interpretation: at both multilateral and domestic levels, deployable capital and profitability are demonstrably present. The scarce factor is verifiable origination — pipelines of identified farmers, verified parcels, and monitorable crops — which is an infrastructure problem, not a funding problem.
Each risk that historically drove intermediaries out of the bold rows of Exhibit 1 now has an operating or near-operational countermeasure in Pakistan. Exhibit 2 states the mapping; none of its rows is speculative technology.
| Risk blocking the contract form | Documented consequence | Countermeasure and status |
|---|---|---|
| Crop state unobservable until delivery | Forward-contract abandonment | Parcel-level Sentinel-2 optical + Sentinel-1 radar time series; the EU applies this population-wide to verify ~€55B/yr in area-based payments[22],[23] (operational) |
| Commodity-price exposure on financier inventory | Sudan's 8.4%→3.7% contraction[6] | Parallel salam / pre-committed offtake; disclosed price bands (±33% precedent[16]); marketplace price discovery (design) |
| Identity and land-claim verification cost | Collateral-only lending; smallholder exclusion | National biometric ID (NADRA) e-KYC; digitised land records (LRMIS); GPS boundary capture verified against imagery (in build) |
| Multiple pledging of one crop | Fraud losses; instrument distrust | Central registration with uniqueness enforcement — the function underpinning Brazil's US$-scale rural product note market[24] (design) |
| Catastrophic weather loss on one party | Contract-fairness objections; exit | Mandatory crop-loan insurance (CLIS+) on production loans for major crops[4]; takaful-integrated structures analysed in recent literature[21] (operational) |
| Delivery and storage integrity | Quality disputes at settlement | SBP-regulated electronic warehouse receipts (Naymat CMC) as the delivery leg[5] (operational, early adoption) |
| Cultivation-effort observability (output-sharing forms) | Partnership forms abandoned | Vegetation-index trajectories against the parcel's own multi-year baseline: effort becomes a measurable time series (operational method; product design) |
The framework implies a sequencing rule: offer each contract form only when the information it requires exists. Operationally, this couples product eligibility to a staged credit-scoring regime (developed in companion paper AH-RD-WP-2026-01) in which behavioural signals carry zero weight until observed. Exhibit 3 states the ladder. Its analytic significance is that it reframes the debt-dominance critique constructively: markup structures are not the terminal state of Islamic agricultural finance but its data-collection phase, with risk-sharing forms activating on accumulated evidence (P3).
| Evidence tier | Information available | Contract forms offered | Design rationale |
|---|---|---|---|
| Tier 0 — Verified + satellite | Identity, parcel, crop verified; multi-year parcel history (retroactive from archive) | In-kind murabaha; qard hasan graduation tier; shared-machinery ijara | Asset-side exposure only; harvest-synchronised; generates first repayment observations |
| Tier 1 — + platform activity | + one season of observed transaction conduct | Monitored salam (small ticket), insured, offtake-committed | Crop risk observable in season; price risk pre-committed |
| Tier 2 — + repayment history | + completed financing cycles | Full salam working capital; istisnaʿ infrastructure | Counterparty risk priced on observed behaviour |
| Tier 3 — outcome-fitted | Portfolio-calibrated default probabilities | Muzaraʿa/musaqa partnerships; diminishing musharaka; portfolio aggregation into asset-backed certificates (sukuk) | Genuine risk-sharing viable once both parties' risks are statistically priced |
The framework's anchor instrument — specified at design level in the companion paper and summarised here — is a registered salam certificate: a standardised pre-harvest forward contract recording verified identity, parcel, crop, quantity, grade, and delivery terms; registered centrally with uniqueness enforcement; insured under CLIS+/takaful; price-managed through parallel-salam offtake with disclosed bands; monitored by satellite through the season with graduated alerts; and settled through the electronic warehouse receipt system. Structurally, the instrument is the institutional analogue of Brazil's Cédula de Produto Rural — a registered claim on a future crop that anchors a private agricultural funding market exceeding BRL 180 billion in registered stock[24] — re-derived within the AAOIFI contractual family, for which the standards already provide certificate (sukuk) templates over salam and cultivation-partnership assets.[13] Portfolio aggregation into such certificates is the designed bridge between Pakistani farm-gate origination and the multilateral and Gulf institutional capital documented in §4.4.
For financial institutions, the framework converts the 2028 requirement from a compliance exercise into a portfolio-construction sequence: begin with low-verification forms instrumented for data capture; graduate borrowers by evidence tier; hold risk-sharing forms until their information preconditions are met. Malaysia's conversion arc (§4.2), compressed to Pakistan's timetable, is the operational template. For regulators, the analysis suggests the highest-leverage interventions are informational public goods — farmer registry standards, registry interoperability with the warehouse-receipt system, and disclosure standards for monitored forward contracts — rather than further credit targets. For researchers, propositions P1–P4 are stated to be falsifiable with pilot-portfolio data; Pakistan's deadline creates, in effect, a scheduled natural experiment in contract substitution at national scale.
Limitations. Four are material. First, the evidence is a synthesis of published secondary sources of varying rigour — institutional self-reports (Akhuwat's repayment rate; IsDB programme figures) are not independently audited here. Second, case selection is purposive, not exhaustive; Indonesia, Sudanese microfinance beyond Bank of Khartoum, and Gulf agricultural investment vehicles are untreated. Third, the Sudanese contraction data are dated (1997–1999) and their generalisation to contemporary conditions is an assumption of the framework, not a demonstrated fact. Fourth, the author's commercial interest (see Disclosure) makes the design framework of Section 5 a proposal to be tested, not a finding; the paper's falsifiable content is deliberately concentrated in P1–P4.
Research agenda. Priority designs: (i) a randomised or staggered-rollout pilot comparing monitored-salam against in-kind murabaha portfolios on default, yield, and farmer surplus (tests P1–P3); (ii) incorporation of community-accountability variables into scored underwriting with out-of-sample discrimination testing (P4); (iii) calibration studies of satellite-derived crop-state indices against realised delivery on Pakistani smallholdings, extending the EU's population-scale verification evidence[22] to salam settlement conditions.
The scarcity of agriculture on Islamic finance balance sheets has often been discussed as a product-design or commitment failure. The evidence assembled here supports a more tractable diagnosis: the contract forms best matched to agricultural cash flows impose verification requirements that were, until very recently, prohibitively costly for precisely the smallholder populations those forms would most benefit. Sudan demonstrates the forms operate and specifies what breaks them; Malaysia demonstrates institutional conversion on a deadline; Pakistan's Akhuwat demonstrates that locally embedded accountability is bankable information; the multilateral portfolio demonstrates that capital waits on origination, not the reverse. Free-archive satellite monitoring, biometric identity, digitised land records, mandatory crop insurance, and electronic warehouse receipts now close the verification gap asset by asset. Pakistan's 1 January 2028 requirement gives this closure a timetable and a market. Whether the risk-sharing contract set scales when its information preconditions are finally met is now an empirical question — one this paper has tried to state precisely enough to be answered.