Formal Exclusion, Informal Extraction, and the Case for Structural Reforms

Abstract

(Pakistan’s formal agricultural credit system has expanded nearly five-fold in nominal disbursements over the past decade and a half. This growth has widened, rather than narrowed, the gap between institutional reach and distributional equity. Tenant farmers and the landless, who together constitute the majority of Pakistan’s agricultural workforce, received just 0.2 percent of the principal agricultural development bank’s loan portfolio in 2022-23, down from 8 percent in 1996-97. This exclusion is not market-generated. It is the direct product of land-based collateral requirements that systematically make 34 percent of cultivating households ineligible as they do not have registered title. For those who do qualify in principle, cumbersome procedures impose a secondary barrier. A formal agricultural loan requires ten branch visits, twenty days, and unofficial facilitation payments that add an effective surcharge of approximately 12 percent to the nominal markup rate. Sectoral allocation compounds the problem. Livestock, which contributes 60.5 percent of agricultural GDP, receives barely 8 percent of formal credit while non-farm rural enterprises, accounting for 42 percent of rural household incomes, remain structurally underprovided. The vacuum left by formal finance is filled by an informal credit market whose effective interest rates range from 44 to 83 percent per annum. Its defining instrument is the tied loan, in which pre-sowing finance is repaid through compulsory harvest sales at below-market prices, an arrangement opaque enough that most borrowers cannot calculate its true annualised cost. This paper maps the institutional anatomy of formal agricultural credit, documents the scale and mechanics of informal lending, and draws on comparative evidence from Bangladesh, India, Kenya, and Indonesia to advance a structured reform agenda. It argues that collateral reform, procedural simplification, competitive disruption of tied credit markets, digital delivery modelled on India’s Kisan Credit Card, and portfolio reorientation towards livestock and the non-farm rural economy represent the minimum necessary conditions for making agricultural finance genuinely developmental. – Author)

I. Introduction

1.1 The Central Contradiction

Agricultural credit occupies a position in Pakistan’s development discourse that is simultaneously celebrated in policy documents and consistently undermined in institutional practice. The country’s formal agricultural lending apparatus spanning ZTBL and commercial banks mandated by the State Bank of Pakistan, microfinance banks, and provincial cooperative banks has expanded impressively in nominal terms over the past two decades. Yet the great majority of Pakistan’s 8.3 million farm households, as enumerated in the 2010 Agricultural Census, together with the considerably larger population of tenants, sharecroppers, and rural workers who depend on agriculture without owning the land they till, remain beyond the reach of formal finance. They borrow, instead, from the commission agent or the input dealer, who charges rates embedded in commodity transactions opaque enough to conceal their true cost from the borrower.

The paradox is not accidental. It reflects a coherent, if unjust, institutional logic: a formal credit system designed around land as collateral will serve those who have land, and exclude those who do not. Since approximately one-third of Pakistan’s cultivating households are tenants or sharecroppers without registered title, and since the landless agricultural labourer is not a borrower the formal system contemplates at all, the system’s distributional outcomes are a direct expression of its design parameters. Understanding those parameters — and the institutional and regulatory context that has kept them unchanged through successive reform cycles — is the analytical task this paper sets itself.

1.2 Data and Sources

Pakistan’s Agricultural Census of 2010, the most recent comprehensive enumeration of farm structures available, recorded 8.26 million farm households cultivating 51.7 million acres, of which 65 percent held less than 5 acres and 34 percent operated under tenancy arrangements without formal land title. A new digital census, conducted by the Pakistan Bureau of Statistics using Computer-Assisted Personal Interviewing (CAPI) methodology across all four provinces, was launched in 2024; full results were awaited at the time of writing, though preliminary indications suggested that farm household numbers and the proportion under tenancy have both increased since 2010, consistent with the trend of smallholder fragmentation documented by Ali and Najam (2022).

The principal household survey source used in this paper is the Pakistan Rural Household Survey 2012, conducted by the Pakistan Institute of Development Economics, which updated the earlier 2001-02 wave with revised estimates of rural credit market participation and income composition. Financial inclusion data draws on the State Bank of Pakistan’s Financial Inclusion Survey 2021 and the Financial Inclusion Insights survey of 2022-23. Agricultural credit disbursement data comes from the SBP’s Agricultural Credit Statistics series covering 2010-11 to 2022-23, supplemented by ZTBL Annual Reports and MNFSR agricultural statistics.

1.3 Structure of the Paper

The paper is organised across five substantive sections following this introduction. Section II examines the institutional anatomy of formal agricultural credit, its growth, distributional pattern, sectoral composition, and the procedural barriers that constitute a secondary mechanism of exclusion. Section III analyses the informal credit market, its scale, its structure, and the specific instruments through which it extracts surplus from rural producers. Section IV reviews comparative international experience from countries that have made measurable progress on rural financial inclusion. Section V draws inferences from the preceding analysis and advances a structured set of policy prescriptions. Section VI concludes.

II. The Institutional Anatomy of Agricultural Credit

2.1 Volume, Institutions, and the Limits of Aggregate Growth

2.1.1 Expansion of the Formal Credit System

The growth of institutional agricultural credit in Pakistan over the past decade and a half is, measured by disbursement volumes, a considerable achievement. Table 1 presents total agricultural credit disbursements disaggregated by major institutional category from 2010-11 to 2022-23.

Table 1: Institutional Agricultural Credit Disbursement in Pakistan, 2010-11 to 2022-23 (Rs. Billion)

Source: State Bank of Pakistan, Agricultural Credit Statistics, various issues 2010–2023; SBP Annual Reports; ‘Other’ includes rural support programmes, NGO-MFIs, and cooperative societies.

Total disbursements rose from Rs. 103.7 billion in 2010-11 to Rs. 505.4 billion in 2022-23, a nominal increase of just under five times representing substantial real growth even after adjustment for the severe inflationary pressures of 2021-23 when Pakistan’s consumer price index rose by over 60 percent cumulatively. Several structural shifts are visible within this aggregate. ZTBL’s share of total institutional credit declined from approximately 66 percent in 2010-11 to 51 percent in 2022-23, as commercial banks responding to SBP mandatory agricultural credit targets, and microfinance banks entering the sector from 2012 onward, expanded their own portfolios. Microfinance bank agricultural lending reached Rs. 48.6 billion by 2022-23 from a negligible base a decade earlier, a qualitatively new channel with particular relevance for smallholder access.

2.1.2 Coverage: The Gap between Volume and Inclusion

The aggregate disbursement data conceals more than it reveals. The number of distinct farm households reached by formal institutions, estimated by the Pakistan Microfinance Network at approximately 4.2 million in 2022-23, represents barely half the farm households recorded in the 2010 census, and a considerably smaller fraction of the total rural population dependent on agriculture for their livelihoods. As Beck, Demirguc-Kunt and Levine (2007) demonstrated across 52 economies, aggregate expansion of the financial sector tends to benefit those who are already relatively well served unless specific institutional interventions are designed to reach the excluded. Pakistan’s agricultural credit sector illustrates this pattern with consistency, volume growth and distributional inclusion has moved in opposite directions.

2.2 The Distribution of Formal Credit

2.2.1 Land as the Master Key

The single most important determinant of a rural household’s access to formal agricultural credit in Pakistan is not creditworthiness in any economically meaningful sense. Simply, it is the size of registered landholding. Land serves simultaneously as collateral and as the proxy variable through which social standing, administrative literacy, and geographic proximity to branch offices are bundled together into a single, convenient screening criterion. As Kiyotaki and Moore (1997) established in their foundational model of credit cycles, the dual function of land as productive input and as collateral creates a self-reinforcing mechanism: those who hold land obtain credit, invest, generate returns, and extend their creditworthy asset base; those who do not are excluded from the outset and remain in subsistence. Table 2 extends the distributional record of ZTBL lending by landholding category from 2006-07 to 2022-23.

Table 2: ZTBL Agricultural Loans by Size of Landholding, 2006-07 to 2022-23 (Share of Total, %; Total in Rs. Billion)

Source: ZTBL Annual Reports 2006–2023; Agricultural Statistics of Pakistan, MNFSR; SBP Agricultural Credit Data; author’s calculations.

2.2.2 The Exclusion of Tenants and the Landless

The most consequential finding in Table 2 is the near-total disappearance of landless and tenant borrowers from ZTBL’s loan portfolio. In 1996-97, tenants received 8 percent of total ZTBL disbursements, inadequate relative to their demographic weight but constituting at least a visible presence in the institution’s lending operations. By 2005-06, that share had fallen to 0.5 percent. The extended series shows a further contraction to 0.2 percent by 2022-23. In absolute terms, as ZTBL’s total loan book expanded from Rs. 11.6 billion to Rs. 257.1 billion over this period, the volume reaching tenants barely moved. The 2010 Agricultural Census recorded 2.84 million tenant and sharecropper farm households, 34 percent of all cultivating households, who tilled land under various forms of tenancy without holding registered title deeds. Their effective exclusion from formal credit forces them into the informal market at the extractive rates documented in Section III. As Khandker and Faruqee (2003) found, households unable to access formal credit invest significantly less in inputs and technology, earn lower returns per acre, and are considerably more vulnerable to income shocks than comparable households with formal credit access.

The trajectory of tenant exclusion is not a market-generated outcome reflecting differential default risk. It is the result of institutional design. The collateral requirement embedded in ZTBL’s lending rules, combined with the absence of alternative credentialing mechanisms and the slow implementation of movable collateral legislation, has meant that each successive year of operation has deepened rather than corrected the original exclusion. The Secured Transactions Act 2016 created the legal basis for non-land collateral but its practical uptake by agricultural lenders has remained minimal reflecting both limited institutional awareness and the inertia of established lending procedures.

2.2.3 Large Landowners

A second distributional pattern visible in Table 2 concerns landowners holding over 40 acres who constitute approximately 2 percent of rural households but absorb around 5 percent of ZTBL’s disbursements, a disproportionality that has persisted across the entire series. When commercial bank agricultural credit is added to ZTBL’s portfolio, the concentration of formal credit towards larger farms becomes even more pronounced, since commercial banks’ internal credit assessment procedures are even more heavily weighted towards collateral value and balance sheet strength. The explanation for this pattern lies not in the agricultural productivity of large holdings, per-acre yields in Pakistan’s large farm sector are not consistently superior to those of smaller farms, but in the institutional characteristics of formal lending like collateral richness of large landholdings, administrative capacity to navigate documentation requirements, and the established credit relationships that large borrowers maintain with branch managers over time.

 The bias is primarily institutional and procedural. It is produced by collateral requirements, branch-office geography, documentation complexity, and credit assessment methodologies that were designed for large commercial borrowers and have never been systematically retooled for smallholder or tenant lending. Iqbal, Ahmad and Abbas (2003) estimated, in their PIDE study, that credit-constrained small farm households in Pakistan produced output approximately 35 percent below their unconstrained potential, a productivity loss that is simultaneously a household welfare loss and a macroeconomic loss, representing forgone agricultural output that the nation does not produce because its credit system is not designed to serve the majority of its farmers.

2.2.4 Small Farmers

The share of farmers holding up to 5 acres in ZTBL’s portfolio has risen from 35 percent in 1996-97 to approximately 69 percent in 2022-23, a genuine improvement that reflects SBP policy pressure, expanded branch networks, and the gradual simplification of small-loan procedures but small farmers’ average loan sizes remain insufficient for productive investment. Their access to medium and long-term credit for orchard development, land improvement, or equipment purchase remains negligible; and the procedural burden they face, documented in Table 4, is substantially heavier than that encountered by larger borrowers.

2.3 Farm and Non-Farm Credit: A Structural Imbalance

2.3.1 The Non-Farm Economy and its Credit Needs

The majority of Pakistan’s rural poor are not primarily commercial farmers. They are tenants, agricultural labourers, rural artisans, petty traders, and workers in the rural service economy, who substantially derive their incomes from activities beyond crop cultivation. The PRHS 2012 estimated that non-farm income sources accounted for approximately 42 percent of rural household incomes, a share consistent with Lanjouw and Lanjouw’s (2001) cross-country finding that the rural non-farm economy constitutes the primary pathway out of poverty for the landless in developing countries. Formal agricultural credit’s almost exclusive focus on farm lending thus systematically misses the livelihood activities of the rural poor.

Table 3: Agricultural Credit Disbursed by Farm and Non-Farm Purpose — All Institutions Combined, 2010-11 to 2022-23 (Rs. Billion)

Source: State Bank of Pakistan, Agricultural Credit Statistics; author’s calculations.

2.3.2 Trends and Remaining Gaps

Table 3 shows a gradual increase in the share of non-farm credit within total institutional agricultural lending, from 14 percent of disbursements in 2010-11 to 31 percent in 2022-23. This improvement reflects the SBP’s progressive broadening of the definition of eligible agricultural lending to include livestock, horticulture, agro-processing, and rural enterprise. However, 31 percent non-farm share still falls short of the 42 percent contribution of non-farm activities to rural household incomes indicating that formal credit allocation continues to lag the actual structure of rural economic life. Non-farm rural enterprise lending demands more varied assessment skills, carries higher perceived risk, and benefits from fewer institutional supports. Correcting this imbalance requires regulatory incentives, broadened SBP refinance eligibility and portfolio composition targets alongside a deliberate investment in lender capacity to assess rural enterprise creditworthiness, an investment that as Indonesia’s BRI Unit Desa experience demonstrates, can be commercially self-sustaining (Robinson, 2001).

2.3.3 The Livestock Financing Gap

Within farm credit itself, an especially striking misallocation persists. Over 90 percent of formal agricultural loans are directed to seeds, fertilisers, pesticides, and tractors, while livestock which contributes approximately 60.5 percent to agricultural GDP according to the 2022-23 national accounts receives only around 8 percent of formal agricultural credit. This gap is particularly damaging for female-headed and near-landless households for whom cattle and buffaloes constitute the primary productive asset and the most accessible vehicle for income diversification. Correcting the livestock financing gap requires both product innovation, simplified, insurance-linked livestock credit that can be secured against registered animals, and changes to SBP refinance incentives that have historically favoured crop finance over livestock and horticultural lending.

2.4 The Transaction Cost Wall

2.4.1 The Architecture of Procedural Exclusion

For those rural households who in principle qualify for formal agricultural credit, smallholders with registered land titles, the procedural architecture of formal lending institutions imposes costs in time, travel, documentation, and unofficial payment that function as a secondary barrier to access, equivalent in practical effect to a formal exclusion. These costs are not incidental bureaucratic friction arising from under-investment. They are structural features of institutional design that were calibrated for educated, urban-adjacent, administratively literate borrowers and have never been systematically redesigned for the semi-literate smallholder located hours from the nearest branch office.

Table 4: Difficulties Encountered by Growers in Obtaining Formal Agricultural Loans (Updated Indicative Estimates, c. 2020)

Source: Punjab Economic Research Institute (PERI); Pakistan Microfinance Network field surveys; author’s updated estimates. KB = Khushhali Bank; PPCBL = Punjab Provincial Cooperative Bank Ltd.

2.4.2 Quantifying the Burden

The contrast between NRSP, a semi-formal institution designed specifically for rural outreach, and the formal banking institutions is striking. An NRSP borrower obtains credit in approximately 5 days, making 2 visits, incurring negligible documentation costs, and not required to make unofficial payment. A ZTBL borrower navigates a process requiring 20 days and over 10 visits, submits documentation costing Rs. 1,650 to prepare, incurs travel costs approaching Rs. 1,000, and in a substantial proportion of cases makes facilitation payments exceeding Rs. 1,100. For a small farmer seeking a loan of Rs. 30,000 to finance seed and fertiliser for the coming season, these transaction costs amount to an additional charge of approximately 12 percent on top of the nominal markup rate, a charge that is regressive, officially invisible, and directly contradicts the SBP’s stated objectives on financial inclusion.

The practical implication is straightforward. A small farmer who calculates, even intuitively, that accessing a ZTBL loan will require ten separate trips to a distant branch, an outlay of over Rs. 3,000 in preparation and travel before the first rupee is received, and unofficial payments of unpredictable amount, will rationally prefer the commission agent across the road who lends immediately, without documents, and without declared interest, recovering his costs instead through the harvest settlement. This rational preference for the informal lender, replicated across millions of households, is not a cultural disposition towards informality. It is a predictable response to the comparative transaction costs that formal and informal lenders impose.

2.4.3 Digital Finance as a Structural Remedy

India’s Kisan Credit Card programme, significantly revamped in 2020, provides pre-approved revolving credit to farmers through a digitised card system with minimal documentation, a single-visit application process in most states, and loan sanctioning within 14 days. By March 2023, NABARD reported 74 million active accounts, making it the largest agricultural credit delivery programme in the world by borrower count. The KCC’s operational success rests on the prior digitisation of land records which eliminates documentation costs. Pakistan’s ZTBL digital agriculture finance pilot, launched in selected districts from 2021, reduced processing times from 22 days to 7 and visits from 10.5 to 3, confirming that the barriers to procedural simplification are institutional and managerial rather than technical. Scaling this pilot nationally, underpinned by the land record digitisation programmes underway in all four provinces, would deliver a structural improvement in smallholder access at modest additional cost.

III. The Informal Credit Market

3.1 Scale and Structure

3.1.1 Dominance of the Informal Sector

The formal credit sector’s systematic exclusion of tenants, smallholders, and the landless has not produced a credit vacuum. It has produced an informal credit market of enormous scale, deep community penetration, and structurally extractive character. The PRHS 2012 confirmed that approximately 78 percent of cultivator households participated in the credit market, with informal sources accounting for two-thirds of all credit flows by value, a proportion consistent with the 2001-02 trends suggesting that more than two decade of formal sector expansion has not meaningfully altered the informal sector’s dominance. The SBP’s 2021 Financial Inclusion Survey found that 79 percent of rural borrowing households relied exclusively on informal sources and among tenant and landless households, the proportion was higher.

3.1.2 The Spectrum of Informal Lenders

Informal agricultural credit in Pakistan spans a range of providers whose terms and exploitative character vary considerably. At the relatively benign end, credit from friends and relatives, typically interest-free and directed at consumption smoothing, represents a social insurance function. It is an expression of community solidarity that supplements household income during lean periods without imposing permanent financial obligations. At the damaging end are the professional intermediaries, commission agents, arthis, input dealers, shopkeepers, and moneylenders. Their lending is profit-motivated, exploits the information advantages they hold over formal lenders, and is structured in ways that obscure its true cost from the borrower. As Akram and Li (2017) document, informal agricultural credit markets in Pakistan display the characteristics of bilateral monopoly where the borrower faces few alternatives and the lender faces limited competition, enabling sustained pricing well above competitive levels.

3.2 Tied Loans

3.2.1 How Tied Loans Work

The defining instrument of informal agricultural lending in Pakistan is the tied loan, a credit arrangement in which the receipt of cash or inputs is conditioned on the delivery of future produce to the lender at a predetermined, below-market price. This arrangement is ubiquitous in Pakistan’s major agricultural commodity belts, the cotton tracts of central Punjab and upper Sindh, the wheat fields of Punjab and Khyber Pakhtunkhwa, the rice districts of Sheikhupura and Gujranwala, and the sugarcane zones of southern Punjab and interior Sindh. The farmer who approaches a commission agent for pre-sowing finance does not simply receive credit he simultaneously pre-sells his harvest at a discount, embedding the interest charge in a commodity transaction. The opacity of this mechanism is its most pernicious feature. Surveyed farmers frequently report that their informal loans carry no interest. The charge is not expressed as a percentage rate. It appears as a price differential between what the market offers for their produce and what the commission agent pays.

The farmer cannot easily compute the annualised cost of this differential because it requires knowledge of the market price at harvest time, the quantity to be delivered, and the period over which the loan is outstanding, information that, at the time of borrowing, the farmer typically does not have and the commission agent does. This informational asymmetry was first rigorously documented by Aleem (1990) for the Chambhar credit market in Sindh, where effective borrowing rates averaged 78.5 percent per annum in 1980-81 against formal deposit rates of 10 percent, a differential that, notwithstanding four decades of formal sector expansion, remains structurally similar today.

3.2.2 Quantifying the Cost of Tied Loans

Table 5: Effective Cost of Tied Informal Loans by Major Crop (Updated Estimates, 2022-23)

Source: Author’s estimates based on commodity prices from MNFSR; PERI field survey data; SBP commodity price data. Effective annualised rate calculated over the loan period for each crop.

Table 5 puts specific figures on this exploitation using commodity prices. A cotton farmer borrowing Rs. 95,000 and delivering 125 maunds at the commission agent’s price of Rs. 8,150 per 40 kg against market price of Rs. 8,500 incurs a commodity-embedded interest cost equivalent to an annualised rate of 46 percent. The wheat farmer pays 48 percent per annum, the rice farmer 44 percent, and the sugarcane grower 50 percent. These rates are five to eight times the nominal markup rates of 6 to 9 percent charged by formal institutions to those who can access them. The differential cannot be explained by a risk premium for informal lending. Pakistan’s microfinance sector, where recovery rates above 95 percent are routine, demonstrates that the rural poor are reliable repayers when lent to through appropriate mechanisms. The differential is a monopoly rent, extracted from borrowers whose access to alternatives has been systematically foreclosed.

3.2.3 Input Credit: A Parallel Channel of Extraction

The costs of informal credit extend beyond tied produce loans. Agricultural inputs, fertilisers, pesticides, weedicides, certified seeds, are routinely purchased on credit from dealers and shopkeepers, with a price premium embedded in the credit transaction that constitutes an additional, largely invisible interest charge.

Table 6: Implied Interest Rates on Agricultural Input Credit from Informal Dealers — Indicative Estimates, 2022-23

Source: PERI; MNFSR commodity price data; field surveys. Rates are indicative; actual rates vary by district and dealer.

Table 6 shows implied annualised rates ranging from 72 percent for DAP credit to 83 percent for weedicides. The farmer purchasing urea on three months’ credit pays the equivalent of 78 percent per annum. Certified seed credit runs at 76 percent. For a smallholder or tenant who finances both his working capital loan and his inputs through informal channels, as the majority do, the cumulative interest burden is severe. A farmer carrying Rs. 150,000 in total informal credit exposure at a weighted average effective rate of 65 percent per annum surrenders approximately Rs. 97,500 annually to informal lenders. Against median net farm incomes estimated at Rs. 80,000 to 130,000 per annum for bottom-quintile rural households in the PRHS 2012, this implies that a substantial portion of net farm income is transferred to the intermediary class simply as the price of accessing working capital.

3.3 Market Distortion, Hoarding, and Broader Social Costs

3.3.1 Commodity Market Distortion

The social costs of tied loan arrangements extend beyond the individual borrowing household to the structure of commodity markets and the food security of the wider population. The commission agent who has acquired wheat, rice, or cotton at below-market prices from hundreds of indebted producers holds a commodity inventory that he can deploy strategically. By withholding supply during periods of peak demand, or releasing it in calculated quantities to influence prevailing prices, he extracts rents from consumers as well as producers. The recurring wheat and flour crises that have periodically afflicted Pakistan’s urban markets, most acutely in 2019-20 and again in 2022-23, are not fully explicable by production shortfalls alone. The market power of the intermediary class, rooted in their tied credit relationships with producers, is a structural contributor to supply-side manipulation that drives consumer prices upward while producer prices are held down.

De Janvry and Sadoulet (2020) identify the control of commodity marketing by intermediary classes with market power as a central mechanism through which the gains of agricultural growth are redirected away from producers. When farmers cannot sell in competitive markets because they are geographically isolated, informationally disadvantaged, or already committed to delivering their harvest to the lender, the price-discovery function of the market is distorted, and the surplus generated by improved productivity accrues to the intermediary rather than the cultivator. Pakistan’s arthis system is precisely reflective of this mechanism operating at national scale.

3.3.2 The Debt Cycle and Land Alienation

At the household level, the cycle of indebtedness under informal lending is self-reinforcing across seasons and years. The farmer who surrenders 46 to 50 percent of his gross commodity income to the commission agent retains insufficient surplus to finance the following season’s inputs independently, compelling him to borrow again at the same terms. Over time, persistent debt service obligations accumulate to the point where they cannot be met from seasonal income, driving smallholders to sell or mortgage their land, often the only significant productive asset they hold, to satisfy accumulated obligations. This land alienation process, documented in successive PERI field surveys, is both a cause and a consequence of deepening rural inequality. As smallholders lose their land, they join the ranks of the landless tenant class that the formal credit system already excludes further concentrating land in the hands of those with the resources to absorb it.

IV. International Comparative Evidence: Lessons for Pakistan

4.1 What the International Evidence Shows

Pakistan’s agricultural credit challenge is not exceptional in kind. Comparable economies across Asia and Africa have confronted the same structural barriers, collateral constraints, information asymmetry, high transaction costs, and informal market exploitation. They have, in varying degrees and through different institutional routes, devised solutions that have improved credit access for the rural poor. Four country experiences offer particularly instructive models.

4.2 Bangladesh

The Grameen Bank and BRAC in Bangladesh established the foundational empirical proof that the rural poor, when served through appropriately designed mechanisms, are reliable borrowers whose repayment performance exceeds that of conventional formal sector clients. Grameen’s group lending model, in which five-member groups jointly guarantee each other’s obligations substituting social capital for physical collateral, achieved recovery rates above 97 percent across a portfolio that reached 9.4 million borrowers by 2023. The model addresses both adverse selection (group members screen each other reducing the lender’s information deficit) and moral hazard (social pressure within the group incentivises repayment even when the lender cannot monitor individual behaviour) issues. The deliberate prioritisation of women, who constitute 97 percent of Grameen’s clients, improved repayment rates and produced household welfare effects as female borrowers were more likely to invest loan proceeds in children’s health and education (Khandker, 2005). Pakistan’s NRSP and Akhuwat have drawn on this model with comparable operational results (NRSP recovery rates above 95 percent and Akhuwat above 99 percent), but neither has achieved the national scale that Bangladesh’s institutions reached two decades ago pointing to a gap in the policy and regulatory environment rather than in the model’s applicability.

4.3 India

India’s Kisan Credit Card (KCC) programme demonstrates the transformative impact of procedural simplification on agricultural credit access at scale. Launched in 1998 and significantly revamped in 2020 to incorporate digital delivery, the KCC provides pre-approved revolving credit to farmers through a digitised card linked to their bank accounts. The application requires a single branch visit in most states. Documentation is drawn automatically from digitised land and identity records and credit limits are set on the basis of landholding and crop patterns rather than requiring fresh assessment each season. As of 2023, there were 74 million active KCC accounts, a coverage rate impossible through conventional branch-based lending. The KCC’s success rests on three institutional prerequisites and those are digitised land records, interoperability across lending institutions, and a regulatory framework that treats agricultural credit as a priority sector with associated accountability mechanisms. Pakistan has made some progress on each of said three elements. Land record digitisation is underway in all four provinces, SBP maintains agricultural credit targets for commercial banks, and ZTBL’s digital pilot has demonstrated local feasibility but has not yet assembled these elements into a coherent national architecture of the kind India has built.

4.4 Kenya

Kenya’s M-Shwari, built on the M-Pesa mobile money infrastructure, launched in 2012 demonstrates that credit can be extended profitably to previously unbanked borrowers using behavioural data i.e. mobile money transaction history as the primary credit scoring input in place of collateral or formal income documentation. By 2022, M-Shwari had disbursed over KSh 1.3 trillion in micro-loans. Its agricultural adaptation, M-Kilimo, extended input credit to smallholder farmers with repayment structured around harvest cycles and credit assessment supported by satellite-derived crop monitoring. The critical innovation is the substitution of transactional data for physical assets as the basis of creditworthiness. A borrower’s pattern of mobile money use, frequency, regularity, quantum, reveals financial behaviour in ways that a land title does not and that a collateral-focused lending system cannot capture. Pakistan’s mobile money sector, Easypaisa and JazzCash together exceeding 45 million registered accounts by 2023 with transaction volumes growing rapidly, provides the data infrastructure for a comparable credit scoring approach. The regulatory and product design work required to convert this infrastructure into accessible agricultural credit products for smallholders is the outstanding institutional challenge.

4.5 Indonesia

Bank Rakyat Indonesia’s Unit Desa network demonstrates that inclusive rural banking can be commercially self-sustaining without permanent government subsidy provided that transaction costs are managed through local presence rather than urban branch procedures. BRI’s Unit Desa system places sub-branch banking units in rural villages enabling loan officers to develop the community-specific information that reduces adverse selection to manageable levels and that no central credit scoring model can replicate. By 2022, BRI’s micro and small loan portfolio exceeded USD 50 billion served through over 9,000 rural offices. Robinson (2001) documented that Unit Desa operations consistently earned positive financial returns directly disproving the received wisdom that lending to rural smallholders is inherently loss-making. Pakistan’s ZTBL, with approximately 450 branches serving a rural population of over 130 million, is vastly under-present relative to what the BRI model would imply. Expanding ZTBL’s sub-branch and agent banking footprint into remote rural areas using mobile credit officers and digital transaction infrastructure to reduce the fixed cost of each additional point of service would bring the informational and accessibility advantages of the Unit Desa model within reach.

These four cases converge on a consistent set of principles. Physical collateral is not a necessary condition for creditworthy lending to the poor. Social capital (Bangladesh), behavioural data (Kenya), and community knowledge (Indonesia) are all effective substitutes. Procedural simplicity is not a convenience but a precondition for inclusion. The Indian experience shows that a 60 percent reduction in transaction costs is achievable through digitisation and produces a proportionate increase in coverage. And the commercial viability of inclusive rural finance is not an article of faith. It is an established empirical finding demonstrated at scale in multiple country contexts. None of this is unknown to Pakistan’s policymakers. The constraint is not information but implementation.

V. Inferences and Policy Prescriptions

5.1 Summary Inferences

The data and analysis in the preceding sections support five inferences of direct policy relevance. First, formal agricultural credit has expanded in volume but contracted in reach. The share of tenants and landless borrowers in ZTBL’s portfolio has declined from 8 percent in 1996-97 to 0.2 percent in 2022-23, a period during which total ZTBL disbursements grew by a factor of 22. Disbursement volume is not a proxy for developmental impact. Second, the structural bias of formal credit towards larger farms is primarily the product of institutional design, collateral requirements, documentation procedures, branch network geography, rather than any deliberate decision by any single actor. It persists because reforming it requires sustained institutional effort and involves short-term costs that lending institutions, measured on loan recovery ratios, have limited incentive to accept voluntarily.

Third, informal credit is not a safety net but a poverty trap. Effective annual rates of 44 to 83 percent on informal agricultural loans do not supplement the rural poor’s incomes. They confiscate a substantial portion of them. Fourth, the transaction costs of formal lending constitute a secondary exclusion mechanism as consequential as the collateral requirement itself. The 12 percent effective surcharge imposed on ZTBL borrowers through procedural friction is regressive and inconsistent with the SBP’s financial inclusion mandate. Fifth, the non-farm rural economy accounting for 42 percent of rural household incomes but receiving only 31 percent of formal agricultural credit is consistently underserved with disproportionate consequences for the landless and near-landless households whose livelihoods depend primarily on non-farm activities.

5.2 Collateral Reform

5.2.1 Operationalising Movable Collateral

The Secured Transactions Act 2016 established the legal framework for movable collateral, livestock, stored crops, equipment, receivables, but its implementation by agricultural lending institutions has been minimal. A functional Movable Collateral Registry, accessible to district-level bank branches and integrated into ZTBL’s loan origination system combined with standardised loan products built around movable security, would immediately expand the eligible borrower universe without requiring legislative change. The technical prerequisites, a centralised registry, standardised security agreement templates, and clear enforcement procedures, are well within administrative capacity. The constraint is institutional prioritisation.

5.2.2 Warehouse Receipt Financing

Warehouse receipt financing deserves particular emphasis as a mechanism that addresses both the collateral barrier and the tied loan problem simultaneously. Under this model, a farmer who deposits grain at a certified warehouse receives a negotiable receipt that can be pledged to a bank as collateral. The farmer retains ownership of his produce, can sell it when prices are favourable rather than at harvest when they are seasonally depressed, and accesses formal credit at institutional markup rates rather than informal rates. India’s Negotiable Warehouse Receipt programme and comparable systems in Kenya and Tanzania demonstrate the model’s operational viability. PASSCO and provincial food departments already hold the physical infrastructure, storage facilities, quality grading capacity, and distribution networks that a warehouse receipt programme would require.  Legal and quality certification framework can be be built around this existing base.

5.2.3 Group Lending for the Landless

For tenant farmers and the landless, households for whom even movable collateral is negligible, group-based lending on the NRSP and Akhuwat model remains the most operationally proven pathway to formal credit access. Both institutions have demonstrated recovery performance that any commercial bank would consider outstanding. The impediment to their scaling is not the model’s viability but the funding environment. Both operate at below-market markup rates that require donor or government subsidy support to maintain, and neither has yet developed the commercially self-sustaining operating model like that of BRI’s Unit Desa. The institutional design challenge is to combine group-based risk management with market-linked pricing and digital delivery cost reduction that could, over time, reduce the subsidy requirement to manageable levels.

5.3 Procedural Simplification and Digital Delivery

5.3.1 Transformation not Incremental

Reducing the procedural burden of formal agricultural lending requires a clearly specified target, not a general exhortation to simplify. A reasonable minimum standard would require that no borrower be asked to make more than two branch visits to complete a loan application and that documentation be pre-populated from digitised land records and NADRA identities databases and require only borrower verification rather than fresh submission. And that loan disbursement would be by direct transfer to a mobile money account eliminating the need for a collection visit. Against the current ZTBL standard of over 10 visits, 22-day processing, and Rs. 1,650 in documentation costs, these targets represent a transformation rather than an incremental improvement which ZTBL’s own pilot has demonstrated that they are achievable.

5.3.2 The Land Records Foundation

The Punjab Land Records Authority’s ongoing digitisation of property records, together with parallel programmes in Sindh, KP, and Balochistan, provides the foundational data infrastructure for simplified agricultural credit assessment. When a loan officer can verify landholding from a secure digital record in seconds, the documentation requirement that drives borrowers away from formal institutions becomes unnecessary. Accelerating land record digitisation  particularly in districts with large tenant populations and high informal lending rates,  and mandating its integration into ZTBL’s and commercial banks’ agricultural credit workflows would deliver returns in financial inclusion that justify the administrative investment.

5.3.3 Eliminating Unofficial Payments

The widespread practice of illegal facilitation payments, documented at Rs. 1,100 per loan transaction at ZTBL in Table 4, cannot be addressed by exhortation alone. Branch-level accountability, through periodic independent audits of loan processing time and cost, combined with anonymous borrower feedback mechanisms and clearly publicised grievance procedures, can create the monitoring environment in which informal payment demands become operationally risky for those who make them. The SBP has the supervisory authority to mandate such systems and to make compliance a rated criterion in ZTBL’s annual regulatory assessment.

5.4 Disrupting the Tied Loan System through Competitive Alternatives

5.4.1 Why Prohibition Has Not Worked

Legislative attempts to restrict the operations of commission agents and input dealers have consistently failed to reduce their dominance over agricultural credit because they addressed the symptom rather than the cause. The arthis fill the genuine economic functions, credit provision, commodity handling, quality grading, local market intelligence, that formal institutions have not replicated at the farm gate. Statutory restriction of tied loan arrangements without a credible formal alternative simply removes the credit source without replacing it. The effective strategy is competitive displacement, making formal credit accessible enough that the borrower has a genuine alternative, and making commodity marketing competitive enough that the commission agent cannot exploit a monopoly position in the produce market.

5.4.2 Farmer Producer Organisations

Farmer Producer Organisations (FPOs) can aggregate smallholders into collective entities large enough to negotiate with buyers, access formal credit, and invest in shared post-harvest infrastructure. The FPO model directly addresses the bilateral monopoly of the commission agent by creating countervailing bargaining power on the supply side of agricultural commodity transactions. Pakistan’s experience with farmer cooperatives has been mixed largely because cooperative governance has in the past been susceptible to capture by larger farmers within the community. The FPO model’s emphasis on member governance, commercial discipline, and separation from government management offers a more robust design that merits serious pilot investment in major crop-growing districts.

5.4.3 Electronic Commodity Markets and Price Transparency

Agricultural price information systems, publicly available, real-time commodity price data disseminated through mobile SMS, radio, and digital platforms, are a low-cost intervention that measurably improves farmers’ negotiating position in produce markets. When a farmer knows the day’s mandi price before entering into a sale with the commission agent, the information asymmetry that enables below-market price settlement is reduced. Pakistan’s federal and provincial agriculture departments already collect commodity price data. The investment required to make this data accessible to rural producers in real time is modest relative to its potential impact on the terms of trade in agricultural commodity markets.

VI. Conclusion

Pakistan’s agricultural credit system has a clear and measurable failure at its core. A formal sector that has grown nearly five-fold in nominal disbursements now reaches fewer than half of farm households. An informal sector has filled the gap, but at effective annual rates of 44 to 83 percent. These are not market outcomes. They are the direct consequences of institutional design.  The data leaves little room for ambiguity. Landless tenants, who constitute 34 percent of cultivating households, received 0.2 percent of ZTBL’s portfolio in 2022-23. Small farmers who do qualify must make over ten branch visits. Documentation costs and unofficial facilitation payments add an effective surcharge of 12 percent to the nominal markup rate. Livestock, which contributes 60.5 percent of agricultural GDP, receives barely 8 percent of formal credit. The non-farm rural economy, representing 42 percent of household incomes, receives 31 percent. Each figure describes a gap between institutional reach and developmental purpose.  The reform agenda advanced in Section V is neither speculative nor novel. Its components, movable collateral, warehouse receipt financing, procedural simplification, digital delivery, group lending at scale, and portfolio reorientation towards livestock and the non-farm economy, are drawn from what comparable countries have already implemented. The international evidence is conclusive. Bangladesh has shown that the rural poor repay reliably when the lending model is designed for them. India has shown that transaction costs can be reduced by 60 percent through digital simplification. Kenya has shown that mobile transaction history is a viable substitute for physical collateral. Indonesia has shown that inclusive rural banking can earn positive financial returns without permanent subsidy. These are not theoretical propositions. They are operational realities, tested at scale across different institutional contexts.  The rural poor of Pakistan are not a charity case. The tenant in Sindh, the smallholder in southern Punjab, the female livestock keeper in Khyber Pakhtunkhwa, each is productive, resourceful, and, as the record of well-designed programmes consistently demonstrates, a reliable borrower. The constraint has never been on their side. It has always been on the side of institutions that were not designed to serve them. Redesigning those institutions is now a policy choice, not a technical impossibility.

References

Akram, W. and Li, Z. (2017) ‘Formal versus Informal Credit: Do Sources of Rural Credit Matter for Poverty Alleviation in Pakistan?’ Pakistan Journal of Commerce and Social Sciences, 11(1), pp. 18–36.

Aleem, I. (1990) ‘Imperfect Information, Screening, and the Costs of Informal Lending: A Study of a Rural Credit Market in Pakistan.’ The World Bank Economic Review, 4(3), pp. 329–349.

Ali, S. and Najam, U. (2022) ‘Land Concentration and Credit Market Exclusion in Rural Pakistan.’ The Pakistan Development Review, 61(2), pp. 91–112.

Beck, T., Demirguc-Kunt, A. and Levine, R. (2007) ‘Finance, Inequality, and the Poor.’ Journal of Economic Growth, 12(1), pp. 27–49.

De Janvry, A. and Sadoulet, E. (2020) Development Economics: Theory and Practice. 2nd edn. New York: Routledge.

Financial Inclusion Insights (2023) Pakistan Wave 7 Report. Washington DC: Intermedia.

Hussain, A. and Iqbal, M. (2020) ‘Political Economy of Agricultural Credit Policy in Pakistan: Who Benefits and Who Pays?’ The Lahore Journal of Economics, 25(2), pp. 45–78.

IMF (2023) Pakistan: Staff Report for the 2023 Article IV Consultation. Washington DC: International Monetary Fund.

Iqbal, M., Ahmad, M. and Abbas, K. (2003) The Impact of Institutional Credit on Agricultural Production in Pakistan. PIDE Working Paper No. 2003-01. Islamabad: Pakistan Institute of Development Economics.

Khandker, S.R. (2005) ‘Microfinance and Poverty: Evidence Using Panel Data from Bangladesh.’ The World Bank Economic Review, 19(2), pp. 263–286.

Khandker, S.R. and Faruqee, R.R. (2003) ‘The Impact of Farm Credit in Pakistan.’ Agricultural Economics, 28(2), pp. 197–213.

Kiyotaki, N. and Moore, J. (1997) ‘Credit Cycles.’ Journal of Political Economy, 105(2), pp. 211–248.

Lanjouw, P. and Lanjouw, J.O. (2001) ‘The Rural Non-Farm Economy: Issues and Evidence from Developing Countries.’ Agricultural Economics, 26(1), pp. 1–23.

Ministry of National Food Security and Research (MNFSR) — Agricultural Statistics of Pakistan, various issues. Islamabad: MNFSR.

NABARD (2023) Annual Report 2022-23. Mumbai: National Bank for Agriculture and Rural Development.

Pakistan Bureau of Statistics (PBS) (2010) Pakistan Agricultural Census 2010. Lahore: Agriculture Census Organisation, Statistics Division, Government of Pakistan.

Pakistan Institute of Development Economics (PIDE) (2012) Pakistan Rural Household Survey 2012. Islamabad: PIDE.

Punjab Economic Research Institute (PERI) (2006) Agricultural Credit and Tied Loan Arrangements in Punjab. Lahore: PERI. Updated field estimates 2022.

Robinson, M.S. (2001) The Microfinance Revolution: Sustainable Finance for the Poor. Washington DC: World Bank.

State Bank of Pakistan (SBP) — Agricultural Credit Statistics, various issues 2010–2023. Karachi: SBP.

State Bank of Pakistan (SBP) (2021) Financial Inclusion Survey 2021. Karachi: SBP.

World Bank (2006) World Development Report 2006: Equity and Development. Washington DC: World Bank.

World Bank (2022) Pakistan: Rural Economy and Agricultural Livelihoods Review. Washington DC: World Bank. ZTBL — Annual Reports 2006–2023. Islamabad: Zarai Taraqiati Bank Limited.