Insurance by BogdanovaThe Encyclopedia of Insurance and Risk Science

RESEARCH DOSSIER

India PMFBY Tendering and State Insurer Coordination

Abstract. India's Pradhan Mantri Fasal Bima Yojana (PMFBY) is a large, publicly subsidized agricultural insurance program in which insurance coverage is produced through a sequence of decisions distributed among state governments, the central government, participating insurers, lenders, technical agencies, and farmers. State tendering selects insurers and their premium quotations for notified risks; governments finance the substantial part of premiums beyond capped farmer contributions; yield estimation and the National Crop Insurance Portal mediate claims. The consequences of any one decision depend on the performance of the other institutions. This monograph reconstructs that coordination problem from the Government of India's operational guidelines, subsequent official clarifications and transitional instructions, administrative disclosures, and original academic studies. It distinguishes contracting rules from implemented behavior, farmer applications from distinct farmers, central transfers from state liabilities, actuarial quotations from realized claim ratios, and claims calculation from actual receipt of funds. It derives transparent illustrative bid-ranking and subsidy arithmetic, formulates a season-specific sequence for analyzing delays, and specifies the evidence necessary for causal evaluation. The analytical synthesis finds that a least-quoted tender can provide weak protection if risk information, fiscal timing, or yield-data governance deteriorates; equally, a timely subsidy transfer cannot correct an invalid exposure record or a disputed crop-cutting estimate. The three-year tender arrangement and exceptional Kharif 2026 transitional options establish why rules must always be dated. Digital portals improve transaction observability while leaving the interpretation and quality of their source records central to scientific assessment. The study supplies an institutional and measurement-based account of PMFBY coordination, identifies testable implications and limitations, and proposes a reproducible research agenda without asserting a new theory or original program-wide causal estimate.

Keywords: Pradhan Mantri Fasal Bima Yojana; PMFBY; agricultural insurance; crop insurance; state tendering; insurer selection; bid weighting; actuarial premium; fiscal federalism; premium subsidy; insurance clusters; crop-cutting experiments; area-yield insurance; DigiClaim; claims settlement; basis risk; public-private coordination; India.

Introduction: Research object, scope, and documentary method

PMFBY converts agricultural production risk into a contractually specified financial obligation financed jointly by insured farmers and governments and administered by selected insurance companies. Its institutional distinctive feature is the involvement of state authorities at almost every consequential step: notifying crops and areas, arranging insurer tenders, generating and validating crop yield data, contributing subsidy funds, overseeing grievances, and coordinating claims. The Union government establishes program rules and bears a defined share of premiums. Insurers quote the relevant premium rates, administer enrolled exposures, process qualifying losses, and disburse authorized claims. Farmers supply crop and land information and hold the practical interest in prompt, accurate indemnification. The 2023 PMFBY operational guidelines give this division of responsibilities its most important documentary expression.

The precise research object is the interaction of insurer selection, central-state subsidy commitments, information generation, and settlement performance. A tender outcome is the observed selection of a quoted rate and carrier for a specified cluster and season. A farmer's insured status is a different fact. A confirmed yield loss, an eligible calculated claim, an insurer's payment instruction, and a bank-credit event are further distinct observations. Any inference that collapses these events into one measure risks assigning responsibility to the wrong institution or overstating effective financial protection.

This investigation uses documentary institutional analysis and bounded economic reasoning. Its primary records are the Government of India's PMFBY guidelines, dated public parliamentary statements and administrative releases, and a May 2026 transitional communication. Original empirical work is employed to test plausible mechanisms and examine farmer experience. Documents are treated according to their evidentiary character: a binding program rule is evidence of an obligation, an administrative press statement is evidence of what the reporting authority has disclosed, an observational study is evidence of its studied sample, and a mathematical illustration is a demonstrable implication of explicit assumptions. No nationwide tender-level microdata or matched farmer-claim panel was available to this research, and no new causal parameter is estimated.

Institutional references to “state” encompass the relevant state or union-territory implementing authority where the underlying document so provides. An insurance “company” refers to an empaneled and selected participating insurer, with distinctions among insurers where contracting documents require them. “Farmer applications” are records for enrollment combinations across crops or seasons and are not assumed to count unique individuals. Monetary examples are expressed in Indian rupees and explicitly marked hypothetical. The inquiry examines crop insurance and the PMFBY framework; results that government publications combine with the Restructured Weather Based Crop Insurance Scheme (RWBCIS) are labeled as combined evidence rather than assigned wholly to PMFBY.

Conceptual and Historical Foundations

Program chronology and dated authority

The Pradhan Mantri Fasal Bima Yojana began in 2016 within India's longer history of agricultural risk support. Its core policy choice was to combine farmer premium limits with actuarial pricing and fiscal subsidy. The Union Cabinet's February 2020 reform announcement described revisions intended to improve program implementation, adjust subsidy support, and provide states with more flexibility. Reforms, however, are meaningful only in the season for which they apply. Terms introduced for Kharif 2020 should not automatically be read into Kharif 2016, nor should a 2026 transitional instruction be treated as having governed all prior bids.

The central documentary starting point is the Operational Guidelines 2023, headed “Effective from Kharif 2023” and circulated in a formal November 6, 2024 communication that explains the earlier preliminary guidance and final detailed wording. The circular's date and the guidelines' effective season have separate meanings. Historical analyses must record both rather than describing the document as a new scheme originating only in November 2024. At the tender level, the active solicitation, clarification, award, notification, and relevant government sanction must also be retained: these determine the actual version bidders and beneficiaries faced.

The 2023 guidelines establish a normally three-year insurer selection arrangement. This length alters the economic interpretation of a bid. Expected weather risk, yield measurement practices, insured acreage, administrative costs, and public finance conditions must be assessed over multiple seasons, while the policy is still implemented through crop-specific seasonal notifications and applications. The multi-year selection principle was tested by an exceptional May 2026 transition advisory for the move beyond the Kharif 2023–Rabi 2025–26 contracting cycle. That advisory discussed extensions or a one-year tender for Kharif 2026 and Rabi 2026–27 pending the new financial-commission period arrangements, with specific state exceptions. It is evidence of a dated administrative bridge, not evidence that the ordinary three-year architecture disappeared.

Institutional actors and units of observation

Six sets of actors participate in the same insured exposure. Farmers provide coverage requests and crop information; banks and other channels may transmit enrollments and premium debits; state or union-territory governments define notified crops and insurance units, conduct procurement, coordinate yield studies, and finance their subsidy share; the Union government provides national scheme rules, central subsidy support, and digital systems; insurers accept the notified insurance obligation subject to the contract and claims procedures; agricultural and technical agencies provide yield, land, and weather evidence. The March 2025 parliamentary response expressly locates important operational decisions with the relevant states or their joint committees.

For scientific purposes the narrowest relevant observational unit is typically a farmer–crop–insurance-unit–season application with insured sum and enrollment status, linked where appropriate to a cluster tender and insurer. The contracting unit is larger: a notified group of districts and crop risks over a selected period. The claim-calculation unit may be an insurance-unit average yield for the notified crop and season, while localized losses can require plot-level or event-specific procedures. The subsidy unit may be an invoiced eligible premium tranche. The grievance unit is a specific contested application or settlement. These denominators are related through identifiers and dates; none is interchangeable with another.

A district-level average tender rate may conceal extreme differences among crops and insurance units. A state-level claim aggregate may conceal unpaid valid claims, obsolete bank details, invalid applications, different insurer assignments, or disputed yields. It is therefore appropriate to build analysis from the most specific documented units upward, preserving the ability to reconstruct every aggregation.

Insurance protection as an operational sequence

Protection develops through a series of necessary operations. A government notifies an eligible crop, insurance area, and season. An insurer is selected and quoted terms are accepted. The farmer is correctly enrolled and the premium and identity are validated. Public subsidy contributions become payable under applicable rules. Loss information is produced and checked. The indemnity, if triggered, is computed, authorized, transferred, and actually received. At every interface a distinct failure mode exists.

This sequence yields three different concepts of coverage. Nominal coverage records a valid policy or application. Contractual coverage denotes the risks and sums insured that meet eligibility and trigger conditions. Effective protection includes the amount and timing of money actually delivered to a farmer after an eligible event. A state can report growing enrollment alongside worsening payout delays; a disbursement system can improve while a poor yield proxy continues to miss individual losses. The December 2025 enrollment-validation disclosure confirms that the application-quality stage is substantive, with insurable interest, land status, crop identity, and recorded area among the qualifying checks.

PMFBY's basic area-yield structure means an adverse loss at a single farm need not trigger an area-wide payment. Conversely, an insurance-unit shortfall can create an eligible payment for individual insured farms whose own losses differ from the average. That arrangement creates basis risk at the farm level. It also concentrates administrative importance on the definition of the insured geography and the production of reliable observed yields. Protection cannot be interpreted solely from the premium percentage printed on the farmer's receipt.

Tender and Cluster Design

Public procurement and multi-season insurer selection

State governments organize competitive processes among insurers that meet the applicable empanelment and tender conditions. Under the 2023 tendering rules, the three-year selection expectation, state-level crop and geographic notifications, procedural timelines, premium-bid scrutiny, and cluster-level selection are distinct requirements. The guideline's procurement provisions contemplate early tendering relative to the crop season and publication of necessary yield histories so that bidders can assess expected loss costs. Tendering late relative to sowing compresses enrollment, data verification, and notification windows.

The economically relevant price is the actuarial or tendered premium rate charged for the insured risk, not the capped fraction paid by the farmer. Government support makes participation more affordable to beneficiaries but does not remove the need to price accumulated weather exposure, catastrophe correlation, operational expense, and uncertainty. A tender outcome documents the winning quoted schedule under the official award method. It does not by itself demonstrate adequacy of the insurer's risk capital, fairness of the geographic distribution, future claims-paying speed, or effectiveness of supervision.

Evaluation must examine the procurement documents actually applied in the state. Some tender calculations use area weights drawn from preceding seasons and prescribed assumptions for newly notified crops. Source precision matters: the 2023 guidelines contain method-specific provisions on cluster quotations, bid evaluation, ceiling-rate handling, and risk disclosure. An independent researcher should reconstruct the awarded bid from every offered crop-level quotation and official weight, rather than replacing the actual rule with a generic arithmetic average.

Cluster construction and heterogeneous exposure

The official cluster design provisions describe grouping districts and selecting an insurer for a cluster, generally involving several districts. Combining heterogeneous risk areas can enlarge a premium pool and spread some local loss variation. At the same time, monsoon disturbances, extreme rainfall, drought, pest outbreaks, and price-linked farm behavior can create highly correlated losses. A nominally broad pool need not diversify a regional shock. Clustering therefore involves a tradeoff among risk composition, tender contestability, insurer service capacity, and the geographical granularity of yield data.

Risk heterogeneity matters also for the state's choice of aggregation. Suppose two districts have similar insured hectares but one grows a high-value crop with a much greater sum insured per hectare. Area-weighted evaluation and premium-weighted economic exposure answer different questions. The first may be prescribed for selecting a bid; the second is necessary for measuring the resulting insurer liability. Their divergence should be displayed rather than silently reconciled. If the mix of farms changes after bidding, the observed realized portfolio can differ materially from the portfolio that generated the winning score.

A strong cluster study should document each included district, notified crop, insurance-unit level, historical sum insured, insured area, farmer application counts, realized yield variation, frequency of observed loss payments, and weather covariates. It should also distinguish the theoretical cluster portfolio from the applications validated at the enrollment deadline. The latter is the insurer's operationally accepted exposure, subject to scheme-specific adjustments.

Bid aggregation and a reproducible illustration

A hypothetical example demonstrates how historical weights affect a least-cost ranking. Suppose a tender score uses two crop-rate quotations weighted 80 percent and 20 percent. Insurer A quotes 8 percent for the first crop and 18 percent for the second; its illustrative weighted bid is 0.80 × 8 + 0.20 × 18 = 10.0 percent. Insurer B quotes 10 percent and 11 percent; its score is 0.80 × 10 + 0.20 × 11 = 10.2 percent. Under these assumed weights and identical eligibility, A ranks below B. The arithmetic illustrates weighted selection, not the exact result of any actual PMFBY procurement.

If the exposure mix for analytical purposes changes to 40 percent and 60 percent, A's comparable score becomes 14.0 percent, whereas B's becomes 10.6 percent; the ranking reverses. Realized exposure after award may therefore differ from historic bid weights. The official procurement decision remains governed by its published methodology. The research inference is that bid rank, expected insurer portfolio loss, and government subsidy outlay should be recomputed under explicit and distinct weight vectors. A rate that appears low at award need not be low for the crop mix eventually insured.

The example makes three points about causation. First, rank reversal is a property of the declared weights; it is not evidence of manipulation. Second, a winner's cost advantage may reflect a different assessment of yield risk, claims frequency, capital requirements, or administrative efficiency. Third, without the full bid schedule and exposure vector, selecting the lowest cluster premium reveals little about the adequacy of any individual crop quotation. Evaluation therefore requires bid-level records, award documents, and subsequent crop-specific experience.

Contract continuity, renegotiation, and the Kharif 2026 transition

Multi-season arrangements can reduce repeated procurement costs and strengthen an insurer's incentive to invest in staff, farmer outreach, and yield-data coordination. They also commit an administration to a carrier through changing weather risks and fiscal conditions. In a well-specified contract, responsibilities for amendment, crop entry and exit, revised sums insured, dispute resolution, and performance monitoring must be clear before the next season. The guideline limits on post-tender changes protect the comparability of competing bids and constrain opportunistic renegotiation.

The May 26, 2026 advisory is a concrete institutional stress test. It recognized a transition after the normal contracting block and communicated practical interim procurement options for the forthcoming seasons, including extensions under existing arrangements or a shorter fresh selection in specified circumstances. The key point is temporal: a state that opted for continuation and one that held an exceptional short tender are not two observations of identical contracting treatment. They may differ in insurer familiarity, historical claims, risk experience, and the constraints imposed by the administrative transition.

A credible evaluation of Kharif 2026 must recover the actual state selection notices, the terms offered to incumbent insurers, the identity and coverage of clusters carried forward, and the date of each final decision. The 2026 advisory should not be backdated into the 2023 guideline as a standing rule. Nor can an observed short-term contract be classified as unlawful merely because its duration differs from the normal selection principle; the relevant authorization and transitional context determine its legal character.

Incentives, contestability, and information asymmetry

Tendered premiums reflect insurers' estimates, which may vary because of information access and modeling assumptions. The state often controls past yield data and notifications; the insurer may hold private claims experience; financial authorities hold details of prior subsidy arrears. Differences in these information sets can widen rates or reduce participation. The mandated sharing of yield histories is therefore economically consequential. Bid competition functions better when all eligible competitors can verify comparable exposure and prior claims information.

The classic concern that a low quote may be followed by later administrative underperformance is plausible but requires evidence. A winning insurer may have correctly priced risk through better diversification or superior operational capacity. Delayed claims may originate in state yield disputes, subsidy shortages, bank errors, or insurer conduct. Distinguishing these explanations requires the timestamped transfer history and the claim-level decision record. Treating every delayed case as insurer opportunism is as unsound as treating every tender winner as a quality success.

Contestability can be measured without asserting improper conduct: the number of eligible bidders per cluster, bids received, the difference between first and second quoted rates, exclusions at qualification, repeat incumbency, retendering frequency, and award concentration. Further analysis may test whether thin bidding correlates with information gaps, high catastrophe risk, past arrears, or small service populations. Such relationships would be observational until a credible identification design addresses selection.

Central-State Subsidy Flows

Contract premium and farmer contribution

PMFBY separates the full actuarial or tendered premium from the share paid by the farmer. The Government of India's March 10, 2026 subsidy statement confirms maximum farmer contributions of 2 percent of the sum insured for Kharif foodgrain and oilseed crops, 1.5 percent for Rabi counterparts, and 5 percent for relevant annual commercial or horticultural crops. The remainder of the eligible actuarial premium is publicly financed under the scheme's sharing arrangements. Actual eligibility, crop notification, and coverage conditions remain essential to a correct calculation.

Consider a hypothetical Kharif insurance amount of ₹100,000 with an eligible tender rate of 12 percent. The total premium is ₹12,000. With the farmer paying 2 percent of the insured sum, the farmer's contribution is ₹2,000. The subsidy to be financed by government is ₹10,000. In an ordinary state where the eligible public subsidy is split equally, the central and state obligations are each ₹5,000. In a qualifying jurisdiction operating under a 90:10 public subsidy split, the illustrative obligations are ₹9,000 and ₹1,000. No administrative rate, tax treatment, liability cap, or special contract term has been added to this example.

This arithmetic establishes a major distinction: lowering the nominal farmer premium does not eliminate the economic cost of protection. It reallocates that cost to public budgets and makes continued coverage dependent in part on fiscal execution. It also shows why the total tendered premium and a farmer premium receipt cannot be substituted for one another in statistical comparison.

Intergovernmental financing

The normal central-state subsidy sharing rule allocates the eligible public portion 50:50, with specified northeastern and Himalayan jurisdictions qualifying for a 90:10 central-to-state division in the periods identified by government. The 2026 statement specifies the later start date for the Himalayan arrangement. Analytical tables must retain the state, crop season, recognized eligible cost, and actual legal ratio at that date. A single India-wide subsidy percentage erases meaningful intergovernmental variation.

The 2023 funding and payment provisions prescribe financial interfaces involving insurers, state governments, the Union government, and the National Crop Insurance Portal. They provide for premium-subsidy demands and specified advance and subsequent releases, with central-share treatment distinguished from state-share disbursement. The legal obligation to contribute, the amount demanded, the tranche sanctioned, the amount transferred, the insurer's bank receipt, and the final account reconciliation are different records. Evaluating fiscal cooperation requires each one.

A government reporting that it has sanctioned funds may still leave a period before the insurer has usable cash. Conversely, a season's central share can arrive while the state's residual contribution remains delayed. A claim-level study must identify the sequence and distinguish claims whose eligibility and yield data are final from those awaiting official inputs. The February 2026 official account of delays identifies state-funding, bank information, and yield disputes as different sources of disruption; their coexistence is exactly why gross delay duration alone cannot identify a responsible party.

Premium liquidity and insurer performance

Insurance liabilities can become payable before all corresponding subsidy cash has been received. The insurer is therefore exposed to a liquidity and working-capital problem distinct from the ultimate expected cost of claims. During a geographically correlated shortfall, many claims may mature together. If a substantial portion of public subsidy remains unpaid, the cost of bridging finance can affect the insurer's financial position and operational incentives. The direct causal size of this effect cannot be inferred from a press release or a cross-sectional claim ratio.

The 2026 parliamentary response on DigiClaim and penalties describes delinking the central share of premium subsidy from state subsidy releases so that corresponding proportional claims can proceed. It also reports penalties for delayed payments by insurers and later state-delay measures. These reforms indicate an administrative attempt to reduce dependence across transfer stages. Their effectiveness must be measured from season-specific payment timestamps and affected claim amounts, rather than assumed from the adoption of a digital module.

Consider two otherwise equivalent claims for ₹20,000, both calculated on the same day. In one case the necessary official data and funding have arrived and payment is credited 12 days later. In another, a valid claim is calculated but a subsidy tranche is outstanding for 45 days, followed by a 12-day disbursement period. Farmer experience differs by 45 days; the responsibility and contractual timeliness assessment depend on the applicable rule's starting event, funding condition, and penalty provision. A research table should show each interval separately and should never reclassify a payment made within a prescribed post-calculation period as timely relative to the farmer's actual harvest loss without explaining the difference.

Fiscal risk and alternative transfer arrangements

The March 2026 government description also identifies alternative risk transfer options, including the 80:110 and 60:130 cup-and-cap arrangements and a profit-and-loss-sharing arrangement. These are not interchangeable with ordinary full-risk transfer: the distribution of very favorable and very adverse claims experience between public authorities and insurers may differ from the standard contract. The actual chosen model is a fundamental component of the tender observation.

A state may prefer one contractual arrangement because it changes premium requirements, predictable budgetary exposure, or retained catastrophic risk. The correct evaluation is therefore joint: compare expected fiscal cost, downside exposure, payment obligations, incentives, insurer participation, and realized claims under the selected model. Describing an insurer's high earned-premium-to-paid-claims spread as “profit” without recognizing alternative risk-sharing terms, acquisition and administration expenses, reinsurance costs, reserved but unpaid claims, taxes, and timing effects confuses an accounting snapshot with contract economics.

For a fiscal study, separate cash expenditure from recognized subsidy liability and distinguish crop years from financial years. Accrued subsidy obligations may exist at the reporting date even when payment has not cleared; claims may be paid after a season closes; retrospective premium adjustments can occur. The exact payment and accounting definitions should accompany every aggregate. This requirement is particularly important for comparison across different states, whose budget cycles and subsidy execution capacities vary.

Claims Performance

Area-yield indemnity and basis risk

PMFBY's principal crop-yield component operates through notified insurance units, insured crops, and an observed area yield relative to a threshold yield computed according to applicable guidelines. The yield and indemnity sections of the guidelines determine the official aggregation and settlement conditions. Under the simplified standard proportional area-yield formula, an admissible shortfall produces a claim equal to sum insured multiplied by (threshold yield minus actual yield) divided by threshold yield, if actual yield is below threshold yield. Exact contractual conditions, alternative perils, and season-specific instructions must always be applied before any real payment is calculated.

For illustration, suppose an eligible sum insured of ₹80,000, a threshold yield of 2,500 kilograms per hectare, and an actual insurance-unit yield of 1,875 kilograms per hectare. The observed shortfall is 25 percent of the threshold and the simple indemnity is ₹20,000. This example assumes all coverage requirements are fulfilled and makes no assertion about a real Indian crop or district. If actual insurance-unit yield were 2,600 kilograms, the area-yield formula would generate zero despite the possibility of an individual farmer having suffered a loss.

This difference between insurance-unit yield and farm-specific production is basis risk. A larger insurance unit can aggregate diverse local conditions and reduce sampling costs while increasing the mismatch for certain farmers; a smaller unit may improve local representativeness but require more credible experiments, field work, and quality assurance. Comparisons across states are misleading unless the insured geographic unit, notified crop, sampling design, and crop-cutting protocol are held visible.

Data production, crop-cutting experiments, and disputes

Yield measurement is not a passive statistical input. States arrange crop-cutting experiments, maintain historical series, and provide data used in determining actual yield. The crop-cutting requirements define sampling, notification, data transmission, and insurer participation requirements. The 2026 government's description of digital yield systems lists crop-cutting applications, remote-sensing-assisted estimation, insurer observation, and associated portal integration. These mechanisms can improve monitoring and auditability, but a digital record remains dependent on field sampling quality and the correctness of the associated crop and land data.

A yield dispute should be analyzed as an event with its own causal sequence: the stated estimate, quality objection, requested remeasurement or review, documentary support, decision, modified yield if any, and dates of subsequent indemnity calculation. Inference that an insurer “refused to pay” is incomplete if the claim had not yet become computable due to a formally unresolved data issue. The converse also applies: a data dispute cannot indefinitely explain failure to pay a portion of a claim whose contractual calculation has already been accepted. Administrative responsibilities must be resolved against the actual timeline.

A credible quality system asks whether the crop-cutting sample represents the insurance unit, whether the historical yield baseline is comparable after boundary or crop-calendar changes, whether missing experiments are imputed consistently, and whether extreme-event observations receive independent review. Remote sensing can improve coverage, but a modeled vegetation signal is not mechanically identical to farmer yield. Validation requires matched field observations and uncertainty estimates over multiple seasons, with particular attention to rare, severe events.

Loss reporting and settlement workflows

Loss reporting varies by coverage component. Widespread yield shortfall, localized events, prevented sowing or planting where applicable, and specified post-harvest situations can trigger different notification and proof requirements. Claims for different perils should not be merged into a single count without evidence that their definitions and settlement routes are compatible. The official claims and assessment procedures provide the authoritative contractual starting point for these distinctions.

The administrative journey runs from confirmed enrollment to event reporting or yield-data availability, calculation of admissible amount, authorization, disbursement instruction, banking transfer, and final receipt. Failure at an earlier stage can make a later-stage timeliness indicator appear misleading. The March 2025 government statement lists incorrect or delayed banking proposals, yield discrepancies, delayed state subsidy funding, and insufficient insurer staff among historical causes of complaints. These causes imply different remedies and different evidentiary records.

An effective claim-performance dataset must distinguish a claim recorded as “settled” in the insurer's system from one credited to a farmer's verified account. Failed transfers, rejections, duplicate applications, partial payouts, reopened disputes, and reversals must be visible. Dates should be stored as events rather than overwritten status fields. The final farmer-facing metric should measure elapsed time from a relevant loss or yield-finalization milestone to receipt, with additional mandatory contractual-clock metrics reported separately.

Digital coordination and delay measurement

The February 2026 DigiClaim description and the more detailed later 2026 parliamentary account identify the National Crop Insurance Portal's integration with payment and accounting systems. DigiClaim was operationalized for claims from Kharif 2022, with further digital payment integration and penalty processing introduced in later seasons. The 2026 parliamentary account specifies an automatic 12 percent insurer-delay penalty from Kharif 2024 and additional state-delay obligations and escrow arrangements from Kharif 2025. Each feature must be evaluated as of the crop season to which it applies.

A time-series comparison that codes the entire year 2024 as “with penalty” could misclassify Rabi 2023–24 or administrative events associated with preceding contracts. Coding Kharif 2024 and later separately, with actual implementation evidence, is more defensible. Likewise, the existence of a mandatory escrow arrangement from a specified season is evidence of a rule; demonstrating funded escrow balances and effective application requires bank, state, and insurance records.

A useful timeliness analysis reports the distribution—not only the average—of days in each stage: enrollment close to validated application; yield-data submission to accepted yield; accepted yield to calculated claim; calculation to authorized transfer; transfer instruction to successful credit. At a minimum, provide medians, upper quantiles, amount-weighted durations, and the proportion of claims still unpaid at a fixed follow-up horizon. Right-censoring must be acknowledged when recent crop seasons have not finished their settlement cycle.

Farmer experience and distributional consequences

Aggregate payout statistics reveal part of the system; they do not establish that every eligible farmer was compensated correctly. Kaur, Raj, Singh, and Chattu's 2021 empirical PMFBY analysis describes enrollment, awareness, and operational barriers in the available Indian evidence. Its findings provide hypotheses about implementation and perception but should not be generalized as a randomized evaluation of tender structures. Ghosh's 2019 uptake and willingness-to-pay study demonstrates the research value of eliciting farmers' preferences over program attributes, including timeliness; stated preference, however, is not equivalent to observed willingness or ex post claims experience.

Matthan's 2025 ethnographic study of crop claims describes farmers' experiences of claiming and collective negotiation in a situated regional setting. Such field evidence is particularly valuable because it shows why individual claim eligibility and perceived entitlement can diverge from administrative categories. It cannot by itself identify the nationwide quantitative effect of three-year tenders or calculate the average causal impact of the subsidy-sharing ratio. Documentary and qualitative evidence must be read together without erasing their different observational scopes.

Distributional evaluation should stratify by land-tenure position, crop, insured sum, region, remoteness, access to banks and digital registration, and timing of realized losses. The official application-quality requirements list ownership or qualifying tenancy evidence among eligibility checks, making record availability an important potential selection variable. A lower enrollment rate for tenants may reflect barriers to proof, but the size and mechanism of any disparity require linked household and administrative data.

Evidence Validity and Sensitivity

Source hierarchy and date discipline

The research distinguishes four evidence classes. First are rules and formally issued advisories defining duties for a specified season. Second are official data releases reporting program execution, usually with stated as-of dates and scope. Third are peer-reviewed or independent studies with explicit designs and populations. Fourth are reproducible hypothetical calculations used to clarify institutional implications. When these classes conflict, the first task is to determine whether they describe the same period, denominator, and legal object.

The formal 2023 guidelines and the exceptional Kharif 2026 advisory demonstrate the importance of time specificity. The March 2026 premium announcement supplies contemporaneous national parameters, but does not replace every state tender document. The 2026 parliamentary response is informative about program administrative reforms; it does not independently prove that all insurers met prescribed service standards. The institutional sources make a coherent account of obligations possible while leaving empirical implementation as a separate research problem.

The existence of negative documentary evidence also matters. A March 2025 parliamentary answer on audit states that no specific Comptroller and Auditor General audit of PMFBY was conducted during 2023. It would be improper to invent findings from a purported 2023 CAG report on this program or to infer from that statement that no audits of related matters have ever occurred. Precise negative statements can prevent false bibliographic attribution.

Interpretation of administrative totals

An official September 2026 scheme progress note reports more than 92.46 crore farmer applications insured since inception, more than 26.33 crore applications receiving claims, and claims exceeding ₹2.06 lakh crore, with implementation in 25 states and union territories during Kharif 2026. These are aggregate program-reporting categories at a particular disclosure date. They are useful for documenting reach and fiscal scale, while their “applications” denominator is not the number of distinct farmers. Repeated enrollment by crop and season can cause the same farmer to appear more than once.

Another December 2025 subsidy-and-claims disclosure reports central subsidy and paid claims over specified 2022–23 through 2024–25 reporting periods using combined PMFBY/RWBCIS data. Dividing those claims by the central subsidy figure would produce neither a program loss ratio nor an insurer profit margin. The denominator omits state subsidy, farmer contributions, and potentially other earned-premium items; the numerator combines paid claims with different development lags. A valid claims ratio needs earned premium and incurred claims allocated to the same insured portfolio and accounting period.

The 2025 parliamentary disclosure of pending claims shows why a publication date is insufficient without the reporting cutoff. Claims for a recent season may be incurred but unreported, reported but disputed, approved but unpaid, or settled after the snapshot. Comparing two released totals without maturity adjustment can make improving settlement activity appear to be increasing losses. Conversely, an artificially low recent paid-claims figure can be mistaken for extraordinary underwriting profitability.

An evidence table should therefore record for every statistic: official instrument, publication and cutoff date, whether PMFBY or both PMFBY and RWBCIS are covered, state and season range, application or distinct-person unit, claim development basis, gross or net premiums, rupee scale, and cash versus accrual status. That provenance is part of the scientific result because it defines which comparisons are admissible.

Comparative and causal research designs

The strongest descriptive design links state-season tender awards and crop-level rate schedules to insurance-unit exposures, yield observations, subsidy invoices, receipt timestamps, claim calculations, and disbursements. It can reveal the concentration of premiums, the distribution of rate differences, and the administrative location of delays. To estimate causal effects, however, one must distinguish deliberate procurement choices from the factors that led states to choose those procurement options.

A difference-in-differences comparison between states that changed bidding arrangements and those that continued a prior arrangement would be credible only if comparable trends were supported before the change, season definitions aligned, and no concurrent reform differentially altered outcomes. The May 2026 transitional options create potential variation but do not create random treatment: states may select arrangements based on insurer availability, weather forecasts, financial conditions, or prior performance. Event-study diagnostics and documented selection mechanisms would be mandatory.

At the claims level, staggered implementation of digital transfer and penalty measures could support carefully defined comparisons, provided their actual start dates are known and parallel changes in yield estimation, payment regulations, or state budgets are modeled. Insurer outcomes can be compared within similar clusters only when crop risk, insured sums, historical loss distributions, and selection into bidding are controlled. Claims rejected because of invalid land records should be separately measured from valid claims awaiting insurer payment.

Field data can complement administrative measures. A matched farmer survey can test whether reported receipt dates agree with system timestamps, whether missing payments were contested, and whether the insured crop corresponded to the cultivated crop. Qualitative observation, as in the regional claims ethnography, identifies institutional processes that administrative totals may not capture. No single design is sufficient for every causal question; triangulation should retain the scope of each source.

Measurement sensitivity, selection, and missingness

Three sensitivities are central. First, tender-rate rankings can change when historical insured-area weights are replaced by realized insured-sum weights or by catastrophe-adjusted loss exposure. Second, apparent claim timeliness changes with the clock: loss event to payment, crop-cutting completion to payment, final yield confirmation to payment, and calculated claim to bank credit are distinct outcomes. Third, aggregate payment ratios change when the denominator is applications, policies in force, insured hectares, distinct farmers, or only contracts with verified crop losses.

To illustrate the second problem, imagine 100 finally paid claims. Fifty are credited within 10 days of calculation and fifty within 70 days; the median around the split does not adequately describe the late half. If 25 additional valid calculated claims remain unpaid at the observation cutoff, excluding them improves the reported average by construction. Survival analysis with right-censored observations and matched claim amounts provides a more defensible characterization. This numerical example is hypothetical and serves solely to make the choice of denominator explicit.

Application rejection is a selection process. The 2025 validation statement documents conditions involving identity, insurable interest, land area, and crop matching, with specified verification windows. Researchers observing only paid claims omit invalidated applications and eligible persons who never enrolled. Their experience cannot be used as an unbiased estimate of protection for the farming population. An auditable measurement design tracks the full population funnel: potential eligible exposure, submitted applications, validated applications, policies accepted, covered losses, accepted claims, instructions sent, and money credited.

Uncertainty in yield measurement warrants explicit sensitivity bounds. Results should be recomputed with plausible yield-estimation error, alternative historical baselines, missing experiment treatments, cluster reweighting, and different definitions of the claim settlement date. The variation of these results is a direct measure of evidentiary robustness; it should not be presented as a new theory.

Competing explanations and falsifiable claims

Several frequent claims demand additional evidence. “Low tender prices necessarily cause slow payouts” would require identification of comparable insurers and clusters and an account of state subsidy and yield disputes. “All claim delays originate with insurers” contradicts the multi-actor possibilities acknowledged in the official account of claims complaints. “Digital transfer systems guarantee payment” confuses system deployment with completed account credit. “High total claims demonstrate effective coverage” excludes the possibility of extreme losses, delayed claims, and uneven distribution. “A rising number of farmer applications proves universal take-up” fails to adjust for repeated crop-season enrollment.

The alternative hypotheses can be tested with specific falsification strategies. If a tender winner's low rate directly causes underpayment, the relationship should persist after controlling for state release dates, verified claims, exposure and extreme-weather loss. If subsidy timing is decisive, payment hazards should change following receipt of an outstanding tranche among otherwise comparable accepted claims. If yield-data delay is decisive, variation in claim-calculation dates should be related to submission and validation milestones rather than solely to premium disbursement dates. These are research implications, not findings already established by the present documents.

Scientific Synthesis and Research Agenda

Integrated findings

The first finding is documentary: insurer tendering is a distributed procurement process conducted by state authorities within central operational rules, not an independent price-setting exercise detached from crop notifications, past yield data, and state responsibilities. The 2023 guideline's multi-year selection architecture is observable in the contractual record. The 2026 interim advisory confirms that a legal analysis must identify the specific crop season and any exceptional authorization.

The second finding is analytical: the public funding split transforms the insurer's actuarial quote into a shared fiscal commitment. Farmer contribution caps change the incidence of premium expense, while state and central subsidy transfers determine when much of the contract price becomes available. The simplified premium arithmetic proves the scale of this division for declared assumptions; it does not reveal the actual statewide subsidy arrears or the causal effect of funding timing.

The third finding concerns information: the crop insurance contract pays according to notified and verified risk terms. In the area-yield component, the insurance-unit threshold and observed yield jointly govern an indemnity, so field production evidence is part of the economic performance of insurance. Better e-payment technology does not substitute for a credible measurement of the insured event. This conclusion follows from the contractual sequence and is consistent with the government's account of digital reforms.

The fourth finding concerns evaluation: enrollment, premium, claims, and cash-transfer data have different denominators and maturity profiles. Official scale indicators show substantial activity but cannot establish causal program welfare gains on their own. The original empirical PMFBY study and claims ethnography strengthen the case for studying administrative processes alongside farmer experience, within the scope of their designs.

Minimum reproducible data architecture

A scientifically adequate comparative data package should preserve original tender IDs, source file hashes, contract periods, state and cluster identifiers, crop-specific offered and awarded premium rates, evaluation weights, count of eligible bidders, selected insurer identity, and amendments. It should link notifications to season-specific insurance units, baseline yields, observed yields, and crop-cutting metadata. A separate financing ledger should hold farmer receipts, central demand and transfer records, state obligations and receipts, and escrow balances where applicable. The claims ledger should store calculation inputs, authorization decisions, dispute codes, penalty events, banking instructions, reversals, and final credit dates.

All records should retain the original event timestamp and the reporting cutoff. Revisions should append auditable events rather than erase prior values. Cross-scheme totals should carry a PMFBY/RWBCIS scope flag, and identifiers should make it possible to count distinct insured people without exposing their private information. Aggregation should be reproducible from claim-level or clearly documented equivalent source data. Documents unavailable to investigators should be recorded as evidence gaps rather than filled by inference.

A pragmatic first-stage study can sample a small number of states with contrasting tenders and administrative histories. Its primary outputs would be a documented contract inventory, state-season subsidy reconciliation, claim delay decomposition, and independent farmer record verification. A second stage can test causal changes only after showing comparable pretreatment trends and credible variation in policy adoption. This order protects scientific validity without treating a descriptive result as a failed causal result.

Governance and evaluation priorities

Three institutional priorities follow from the record. Tender evaluation should preserve enough risk and data documentation for independent reconstruction of the winning calculation. Subsidy governance should publish reconciled, season-specific payment timestamps and outstanding balances so that delayed public funding can be distinguished from insurer processing delays. Claim governance should preserve yield-estimation provenance and complete payment-event histories so that contracted timeliness and farmer-experienced timeliness can both be measured.

Regulators and administrators can use such information to identify problems before they become recurring grievances. More complete disclosure can also improve insurer competition by reducing differences in access to historical risk and payment data. These are evidence-based evaluation recommendations derived from the coordination structure; the present research does not estimate their costs or claim that every proposed disclosure can be implemented without legal, technical, or privacy constraints.

The government has already introduced instruments directed at several of these interfaces, including the DigiClaim payment framework and the later funding and penalty measures. The research challenge is to establish whether their adoption changes observed disbursement quality under independently verified datasets. An implementation announcement and an observed improvement are separate scientific propositions.

International relevance

The Indian case illuminates a broader question for publicly supported agricultural insurance: how procurement, premium subsidies, loss measurement, and payment systems jointly produce enforceable financial protection. Comparable issues arise where provincial governments define covered farm exposures, public authorities subsidize premiums, and private or public carriers administer claims. The precise distribution of powers, risk-sharing instruments, and insurance-unit design differs by jurisdiction. International comparison should therefore proceed through functional dimensions—who chooses the insurer, who funds the premium, who measures the loss, and who releases payment—rather than equating nominally similar schemes.

India also demonstrates the importance of scale and administrative heterogeneity. National-level enrollment and claims aggregates can provide a system-wide frame, yet the operative insurance promise is implemented at the intersection of a particular crop, geographic unit, state tender, farmer application, and season. A coherent scientific account must be able to move between those levels without converting an aggregate correlation into an individual claim or a national reform announcement into a state-specific fact.

Conclusion

PMFBY tendering and state-insurer coordination form an interconnected institutional process. Competitive selection determines who supplies insurance and at what quoted rate; public subsidy rules allocate much of the corresponding financing between governments; field and administrative data determine which insured losses satisfy the contractual trigger; and payment systems determine whether authorized funds reach farmers. The scientific performance of the program therefore depends on the alignment of contract design, fiscal execution, yield-data quality, and settlement administration.

The dated legal sources establish a normally three-year contracting architecture under the 2023 guidelines, with documented transitional arrangements for Kharif 2026. The official 2026 subsidy statement establishes capped farmer contributions and differentiated intergovernmental sharing. Later administrative disclosures document digital claim reforms, penalties, funding controls, and continuing sources of delay. These sources support a detailed institutional account. They do not establish a nationwide causal estimate of tender competition, insurer profitability, farmer welfare, or the success of any specific digital reform.

The original analytical contribution of this monograph lies in its integrated reconstruction of the state–insurer–Union sequence, its explicit separation of contractual and observed outcomes, and its reproducible demonstrations of bid-composition and subsidy effects. Its strongest substantive conclusion is that a crop insurance tender must be evaluated through the complete path from risk notification and insurer selection to verified loss and successful farmer credit. Effective protection is a measurable outcome of that path, with each stage requiring its own evidence.

References