Insurance by BogdanovaThe Encyclopedia of Insurance and Risk Science

RESEARCH DOSSIER

Postharvest Agricultural Production Loss Coverage

Abstract. Postharvest agricultural production loss coverage concerns the contractual transfer of specified physical and quality-related risks arising during harvest completion, in-field drying, immediate handling, and, only where expressly insured, subsequent storage and movement of agricultural products. Its scientific object lies at the intersection of crop insurance contract timing, agricultural engineering, physical loss measurement, causal attribution, loss adjustment, public program design, and actuarial aggregation. This monograph develops a comparative analysis grounded in the United States federal crop insurance framework, India's Pradhan Mantri Fasal Bima Yojana (PMFBY), the European Union's agricultural risk-management framework, and empirical research on food loss measurement. The central finding is that a postharvest loss in agronomic terminology does not itself establish an insured postharvest event. Eligibility depends jointly on the insured subject, covered peril, production stage, spatial location, duration, applicable crop-year documents, proof of causation, and adjustment rules. India supplies an unusually explicit example of limited in-field postharvest cover, whereas the United States offers a different structure in which ordinary crop insurance generally ends at harvest or another specified termination event and carefully regulates claims involving delayed harvest and quality. The analysis explains how survey bias, moisture conventions, grade discounts, salvage, weather verification, storage transfer, and overlapping benefits shape the magnitude and distribution of indemnity. A set of transparent accounting identities and hypothetical examples illustrates how claims can be reconciled without confusing production losses with price fluctuations or generic food-chain waste. The resulting synthesis identifies the evidence required to evaluate coverage effectiveness, settlement accuracy, and producer protection across diverse climatic and institutional settings.

Keywords: Agricultural insurance; postharvest loss; crop insurance; in-field drying; harvest delay; insurance period; moisture adjustment; quality adjustment; indemnity; PMFBY; USDA Risk Management Agency; food loss measurement; loss adjustment; agricultural risk management; basis risk; climate risk.

Introduction

Agricultural output changes legal and economic character while it moves from the living crop to harvested product, field-dried material, farm-stored inventory, transport consignment, and marketed commodity. Physical deterioration can occur at every stage. The producer may experience a material loss at any of them, while the insurance contract may attach to only some of those circumstances. This gap between the agricultural chronology of loss and the legal chronology of coverage defines the central problem of postharvest agricultural production loss insurance.

The Food and Agriculture Organization's 2019 assessment distinguishes losses along the food supply chain from waste at retail and consumption and explains why definitions affect estimates. The FAO Food Loss Index methodology starts with harvest and extends toward, but excludes, retail. Neither classification specifies that an insurer covers every measured loss. An agricultural insurer prices and indemnifies a narrower contractual event, and coverage may depend on a peril occurring before an official end-of-insurance date even when the final quantity is verified later.

For an international comparative study, the decisive questions are therefore substantive. What is the insured crop or production interest? At which instant is a lot considered harvested? Does a named postharvest extension apply to cut-and-spread produce lying on the field? Is rain damage during a delayed harvest insured because a qualifying peril prevented timely harvest? Which deterioration belongs to later storage and handling? How are losses of mass, grade, food value, and sale proceeds separated? How is the same event prevented from generating overlapping claims under multiple benefits?

The 2023 PMFBY Operational Guidelines answer some of these questions with a detailed 14-day, state-notified in-field cover. The USDA 2026 loss adjustment handbook answers others through end-of-period and production-to-count procedures. Both systems connect a peril to an agricultural production unit, but they embody different contractual designs. The EU risk-management legislation sets support conditions for instruments that member states implement through their own choices; it should not be read as a single crop-by-crop EU postharvest insurance policy.

Methodologically, this work uses interpretive examination of primary public policy texts, comparative analysis of insurance mechanisms, and critical synthesis of empirical postharvest research. It builds no newly named theory or proprietary risk model. The mathematical expressions below are accounting identities and pedagogical illustrations. No confidential claims data, original field experiment, or cross-country causal effect estimate is asserted.

Conceptual and Historical Foundations

The insured object and the postharvest boundary

An insurable agricultural interest can be expressed in physical units, monetary value, or a contractually specified index. For production-loss indemnity the central object is an insured quantity or its admissible value subject to defined perils. That object is associated with an insured producer, a crop, an agricultural production unit, a crop year, an area, a guarantee, and a measurement method. In postharvest contexts the same quantity may pass from a biological crop to harvested product while remaining exposed to rainfall, wind, pathogens, mechanical damage, and delay.

Three boundaries must be recorded for any loss investigation. The physical boundary identifies when the commodity was cut, separated, dried, threshed, stored, moved, or sold. The contractual boundary identifies the beginning and end of insurance and any explicit continuation or add-on. The valuation boundary determines the counterfactual quantity or value against which the actual result is measured. These boundaries may occur at different times. For example, a crop may remain unharvested after an end-of-insurance calendar date because a timely harvest was prevented by an earlier insured storm. A claim may be settled after the date without implying that insurance against all future hazards was extended.

The FAO food-loss database guide distinguishes on-farm postharvest operations, transport, storage, and wholesale. This supply-chain terminology helps researchers locate damage but is not a substitute for a policy schedule. The 2025 SDG metadata likewise sets statistical boundaries and explicitly excludes preharvest losses from its principal food-loss measure. Contract insurance may nevertheless cover preharvest loss, in-field postharvest loss, or a defined subset of both. Statistical relevance and insurability therefore require separate assessments.

A claim may involve quantity loss, quality loss, additional mitigation expenditure, a change in market price, a lost contractual premium, or lost future revenue. These categories cannot be assumed interchangeable. Contractual coverage for damaged production is a proposition about a policy and covered cause; a decline in market price by itself requires separate evidence of price or revenue insurance. If damaged grain can still be sold at a discounted price, the monetary reduction and the physical weight lost are different observables. If wet grain is dried, a lower post-drying weight may reflect evaporation as well as damaged or missing usable product.

Distinct purposes of loss measurement and indemnity

Food-security research asks how much edible material or nutritional and economic value was lost along a food chain and where prevention would be productive. Insurance research asks how much of an admissible covered loss occurred to a defined insured interest. The Affognon et al. (2015) meta-analysis documents heterogeneous methods and recurrent confusion between economic and physical losses. The Sheahan and Barrett (2017) review argues that the objectives and opportunity costs of postharvest loss reduction deserve explicit treatment. Together, these studies support a methodological caution: an estimate of aggregate food loss cannot be inserted mechanically into an insurance tariff or used to calculate an individual indemnity.

At the level of a lot, a simple product balance is useful. Let Q_in be measured material entering a stage, Q_out usable material leaving the stage under a specified standard, and Q_diverted the material moved to another documented use. The balance Q_in = Q_out + Q_diverted + Q_discarded + Q_unexplained is a bookkeeping reconciliation under consistent moisture and units; its components require observed weights or defensible conversions. The mere existence of Q_unexplained does not establish an insured peril, and diversion to animal feed may have value even if the crop no longer satisfies its original market grade.

For economic valuation one can distinguish physical quantity loss from quality degradation. With a stable reference price p_ref and counterfactual marketable quantity Q_ref, a descriptive gross-value comparison is V_ref - V_actual, where V_ref = Q_ref × p_ref and V_actual is measured net realizable product value under the chosen valuation rule. The policy might instead use a fixed insurance price, approved price election, production-to-count adjustment, grade discount factor, or sum-insured percentage. An analyst must not replace the contractual payment rule with an unapproved market-value calculation.

Such distinctions matter especially when harvest and storage decisions are endogenous. Farmers may delay sale in expectation of price changes, choose a drying method according to equipment access, or retain stocks longer when market access is difficult. The World Bank's Missing Food report places postharvest grain loss within a larger chain of infrastructure, income, and food-security concerns. A policy that fails to separate an insured sudden event from routine deterioration risks compensating storage choices or market outcomes that were never included in its priced exposure.

International institutional context

The Indian PMFBY framework explicitly includes, subject to notification and eligibility, a brief postharvest exposure for harvested crop left in the field for drying. The United States federal structure predominantly organizes production insurance by planting, growing, harvest, crop-specific insurance period, and quality adjustment. The European Union provides an overarching legal mechanism through which member states may support agricultural insurance premiums or mutual funds. None of these architectures should be flattened into a global standard.

These differences arise from institutional goals. A limited in-field extension can target a short, recognizable vulnerability at harvest without converting a yield policy into open-ended warehouse insurance. A production-to-count mechanism can protect against causes arising during an insured growing period while maintaining a clear termination date. Subsidized insurance assistance can increase access to risk-transfer instruments but must coexist with member-state eligibility, national policy wording, and compensation limits. The current RMA insurance-cycle guidance and the EU Article 76 framework exhibit these distinct levels of regulation.

Historical claims about the chronology of worldwide adoption demand an original dated instrument for each jurisdiction. This dossier consequently makes the narrower defensible assertion that the cited 2023 PMFBY text, 2025/2026 USDA texts, and 2026 EU consolidation demonstrate contemporaneously different designs. It does not infer from these documents which country first invented postharvest loss insurance or rank their welfare performance.

Harvest Stage Definitions

The production-stage and contract-stage clocks

Harvest is a process rather than a universally identical instant. For a grain crop, agronomic maturity, commencement of cutting, threshing, gathering, and movement to storage can be separated by days. Perishables may be picked in batches, cleaned, graded, cooled, packaged, and moved rapidly. Multiple passes over one parcel create multiple production lots with different loss exposure. A universal claim form recording only a field and a date is therefore insufficient when the legal benefit depends on the precise stage of the product.

A defensible temporal record should identify crop and variety, parcel geometry, planting and maturity dates, first and final harvest dates, cut-and-spread or bundling periods, movements between field and storage, ownership at each step, weather observations, and the moment damage became detectable. The record should then be mapped against a crop-year version of the Basic Provisions, Crop Provisions, Special Provisions, endorsements, government notifications, and individual coverage selection. These data have two functions: deciding eligibility and preserving a counterfactual account of what production would have been realized absent a covered event.

The FAO on-farm postharvest stage definitions are an appropriate measurement vocabulary. They can be linked to parcel and lot observations without pretending that they decide contractual liability. A transition from cut grain lying on the field to grain removed into an on-farm store can be especially decisive where a benefit covers only in-field drying. Another distinction is between a claim filed after harvest and a loss that first arose after harvest. The former may be recognized under a preharvest insured cause; the latter ordinarily requires its own valid coverage.

The United States crop insurance boundary

For ordinary federal crop insurance, the Common Crop Insurance Policy Basic Provisions 26-BR establish the policy architecture, and crop-specific provisions control exceptions and details. As reflected in RMA's insurance cycle description, damage is reported, inspected, and settled using crop and production evidence. The usual insurance-period rule includes harvest as a potential terminating event, along with other specified endpoints. It follows that a freshly harvested lot entering ordinary commercial storage should never be assumed automatically protected against any later storage peril under the growing-crop policy.

The nuanced case is delayed harvest. The USDA January 2026 Loss Adjustment Manual addresses an insured crop that could not be harvested on time because of an insurable cause during the insurance period. Its end-of-period guidance permits consideration of subsequent unavoidable damage linked to that earlier cause when the insured complied with notice and attempted to harvest when reasonably possible. It also directs assessors to identify windows of opportunity and assess avoidable losses. The calendar date for the end of the insurance period remains unchanged. Additional time to harvest is a procedure for settling the earlier insured loss, rather than a blanket calendar extension of coverage.

The same manual explicitly distinguishes an insured impediment from a producer's decision to wait for lower moisture to avoid drying expense. Where the mature crop could have been harvested by the end date and the producer chose to wait solely for moisture reduction, the delayed-harvest treatment described above is unavailable. This distinction is foundational to risk engineering and claim causality. A wet field that physically prevents entry is evidence of a different causal chain from grain held deliberately because the producer prefers a lower drying bill.

Under the RMA obligations and expectations guidance, crop-specific provisions and producer responsibilities remain material to every claim. A study comparing late harvest outcomes across crop years must confirm the effective edition rather than extrapolate a current general page to historical policies. It must also distinguish the time for reporting or settling a loss from the period during which new peril occurrence can be covered.

India and limited in-field postharvest cover

India supplies a contrasting direct postharvest benefit. Under section 21.6 of the PMFBY Operational Guidelines 2023, an optional state- or union-territory-notified cover concerns specified crops harvested and retained for drying in cut-and-spread or small bundled form in the field. Its outer temporal limit is 14 days from harvest. The enumerated hazards include hailstorm, cyclone, cyclonic rain, and unseasonal rain, with a specified rainfall trigger for the latter and verification by joint survey. Crop notification, harvesting window, premium status, parcel details, and qualifying form of field retention are integral eligibility conditions.

The state's selection matters. The central instrument requires the relevant crops and harvest windows to be named in official notifications and bid documents; the mere existence of section 21.6 does not mean every insured crop everywhere receives the add-on. The guidelines require timely notification, generally within 72 hours, and provide for app-based reporting, geolocation, images, and joint field assessment. Thus an operational claim depends on an interlocking documentary chain: central guideline; state selection and seasonal notification; insured application and premium; weather event; field condition; survey; calculation; and payment.

The Indian government's 2026 public explanation continues to describe a 14-day cut-and-spread postharvest protection and explains the boundary at removal of the crop from the field. The 2023 text itself is more precise for an applied historical or current claim because it separates the general entitlement architecture from the necessary state notification. A country-comparison table based only on whether an agricultural program mentions postharvest damage would overstate effective access.

Two logical consequences follow. First, a crop that has been harvested but moved directly to a warehouse may face an economically significant postharvest peril without satisfying this named in-field benefit. Second, a crop that has not actually been harvested may remain governed by the standing-crop benefit instead. The transition must be evidenced by production and location data, not inferred from the farmer's reporting label.

Ownership, location, storage, and evidence

Storage ownership and physical custody should be separately recorded. A grower may own produce kept in a cooperative facility, a buyer may take title at the farm gate, or a contractor may handle drying without a transfer of ownership. The exposed party, financial interest, and locus of risk can change on different dates. Insurance contracts aimed at stored goods, transit, business interruption, or goods in custody can be relevant, but those policies require independent terms; their existence cannot be presumed from crop insurance participation.

A measurement ledger should assign each lot a unique ID and record the initiating plot, harvested quantity and measurement basis, moisture percentage, grade, timestamp, storage location, and transfers. Repeated weighing under different moisture content must be corrected to a common reference before attributing a mass difference to physical disappearance. Warehouse receipts and sales invoices are useful corroboration, while photographs show condition and location but normally cannot establish entire lot weight or grade without sampling.

In the United States, the 2026 RMA loss adjustment handbook has detailed procedures for production stored on the farm, commercially stored production, delivery records, and delayed measurement under qualifying conditions. The existence of a rule allowing a later measurement does not imply that subsequent warehouse spoilage is insured. It may simply preserve a defensible estimate of the crop's actual production and admissible preexisting damage.

Specified Postharvest Perils

Weather mechanisms and causal chains

The main weather mechanisms during in-field drying include precipitation wetting harvested material, sustained humidity preventing safe drying, hail physically damaging exposed produce, wind scattering or contaminating bundles, and floodwater making material unsaleable. Each mechanism can produce combinations of missing quantity, fungal deterioration, altered moisture, lower grade, and higher handling costs. Their claim relevance is determined by the policy's named peril list and exclusions, the recorded exposure stage, the event date, and evidence that the damage was caused by an insured event.

For the PMFBY add-on, the 2023 specification of postharvest perils names the covered events and requires joint confirmation. Unseasonal rainfall has a stated benchmark rather than an unlimited subjective definition. That weather threshold is part of an eligibility assessment, while the assessed farm loss remains linked to observed physical damage. A weather reading near a plot and a field inspection answer different questions. The station series can support event occurrence; the field investigation estimates damage to the insured crop.

A scientifically adequate causal narrative follows the sequence: crop condition before event; insured hazard; physical mechanism; immediate observed change; mitigation opportunity; residual deterioration; and final admissible adjustment. It should also test competing explanations such as pest damage predating the event, poor threshing, contamination during handling, preexisting moisture problems, deliberate sale delay, or an uninsured later storage incident. An insured rain event does not convert every subsequent decline in lot value into an automatically attributable insured loss.

The Affognon et al. review and Delgado, Schuster, and Torero (2021) both underscore the need to identify precisely where loss happens. This empirical challenge bears directly on indemnity: when an investigation lacks stage-specific baseline observations, losses from harvesting, in-field drying, farm storage, and distribution may be pooled even though the policy covers only one of them.

Drying, moisture, quality, and marketability

Agricultural drying reduces moisture mass and changes storage stability. It is a technical production operation as well as a possible insured exposure. A recorded drop in weight between wet and dry samples should therefore be interpreted through a moisture-normalization rule. In simplified dry-matter accounting, if initial measured wet weight is W1 and moisture fraction on a wet basis is m1, the dry material is D = W1(1 - m1). If the same dry material is later observed at moisture fraction m2, the corresponding weight is W2 = D/(1 - m2). This equality assumes constant dry material and no damage or foreign-material correction; any residual requires investigation. It is an accounting conversion, not an insurance-specific loss formula.

Consider 100 metric tons measured at 20% wet-basis moisture. The implied dry material is 80 tons; at 14% moisture, an unchanged dry-matter lot weighs approximately 93.02 tons. The apparent 6.98-ton reduction is expected from removing water under this assumption. Treating it as missing insured crop would overstate physical loss. Conversely, a reduction in merchantable grade with little change in dry mass can produce a genuine economic loss that may be addressed through approved quality adjustment if the cause and terms qualify.

The RMA 2026 quality-adjustment guidance distinguishes authorized moisture adjustments, quality discount factors, damaged production value, and conditioning. Its procedures caution against double allowance for the same damage factor and specify that drying charges are not simply eligible conditioning charges. A farm's actual drying expenditure is therefore not automatically paid merely because the crop needed drying after a storm. The policy must authorize any applicable adjustment or additional benefit.

Grade changes demand representative sampling, chain of custody, laboratory or approved grading results, and a clear mapping between the observed deficiency and the covered peril. A discount on a sales receipt may reflect actual grain damage, regional market congestion, handling expenses, or ordinary commercial bargaining. Only appropriate evidence can separate these channels. Likewise, a commodity deemed unsuitable for one buyer might retain feed or industrial value, which affects salvage and potential indemnity calculations.

Risk differentiation and prevention

Postharvest exposure is shaped by crop physiology, harvest timing, labor and machinery, local weather, soil trafficability, drying facilities, storage ventilation, transport access, and the producer's ability to move material rapidly. These characteristics can differ sharply across parcels and households within the same district. A short coverage window can lower the duration of insurer exposure while concentrating expected loss in adverse weather episodes when many farms simultaneously leave harvested material outside.

Loss-prevention measures include correctly timed harvest, rapid collection when appropriate, drying infrastructure, drainage, moisture monitoring, representative grading, secure covered storage, and prompt documentation. Their costs and technical feasibility vary with the crop and location. The Sheahan and Barrett review cautions against assuming that every measured postharvest loss should be prevented regardless of its economic cost. The economically relevant comparison is between avoided expected loss and the full costs of storage, handling, technology, and risk transfer, as well as distributional effects for constrained producers.

Insurance introduces a further interaction. If payout depends only on a broad weather declaration while damage prevention is unobserved, incentives and claims verification differ from a program requiring physical field inspection. In PMFBY, eligible weather events are accompanied by farm-level evidence and conditional notification. In RMA delayed-harvest cases, efforts to harvest when possible are specifically relevant to distinguishing avoidable damage. These provisions mitigate ambiguity but impose documentation and administrative costs. Their empirical success cannot be established from wording alone.

Spatial dependence and actuarial implications

Weather events can damage many harvested lots in a region simultaneously. A pricing study must therefore go beyond average farm-level frequency. Let L_i denote admissible loss for insured unit i. Aggregate loss is L_port = sum_i L_i; its variance includes both individual variances and pairwise covariances. Severe regional rainfall during harvest may generate substantial positive covariance. The number of farms sharing a district weather trigger is not itself a measure of insured loss: losses still depend on crop notification, actual in-field exposure, damage severity, and the timing of each harvest.

A plausible exposure measure for a limited in-field cover is the time-varying insured quantity left in an eligible condition, rather than total seasonal output alone. If E_i(t) is eligible exposed production and lambda_i(t) is a peril-event intensity, an illustrative conditional loss expectation integrates hazard, exposed amount, and severity across the contractually valid window. Every component requires validation. The expected-loss expression must also account for subdistrict weather dependence, differing crop calendars, and claims administration. It is a conventional actuarial formulation, not a newly proposed pricing model.

Actuarial calibration is complicated by sparse claim counts, changing reporting channels, subsidized participation, and event clustering. A rate inferred from past paid claims may understate physical exposure when eligible farmers missed the notice deadline or when administrative capacity delayed surveys. Conversely, a count of all postharvest food losses may greatly overstate insured severity. Product pricing therefore demands matched denominators: insured plots and eligible exposure days for a coverage-specific loss ratio; relevant insured sums for monetary severity; and consistently defined event and settlement records.

Loss Settlement

The chronology of a defensible claim

A complete loss-adjustment chronology begins before the peril. It documents the insured interest and policy version, notified crop or insured unit, limits and applicable deductible, premium status, production condition, harvest activity, and potentially affected quantity. At event discovery the producer should preserve available evidence, make timely notice through the prescribed channels, and avoid destroying material required for inspection unless policy and safety obligations dictate otherwise. The insurer then verifies peril, eligibility, causation, exposure, admissible loss, salvage, and payment.

This is a general evidentiary sequence rather than advice to any individual claimant. The controlling obligations come from the actual contract and jurisdiction. The USDA insurance cycle emphasizes notice, inspection, production estimates, and relevant records. India's PMFBY 2023 guidelines require named reporting channels, 72-hour intimation, farm or plot details, and joint assessment for the postharvest add-on. Neither set of requirements should be transplanted mechanically to another country's program.

The most consequential decisions often concern classification. If crop damage originated while the crop was standing, it may belong to a standing-crop claim even when final evidence emerges at harvest. If a harvested lot remains eligible in the field, a specific short extension may apply. If material has entered storage, separate insurance may need to be tested. A well-kept timeline prevents a claim from being accepted or rejected solely because an adjuster used the general descriptive phrase "postharvest loss."

Eligible quantity, quality, and payment

For analytical reconciliation, define Q_base as counterfactual insurable crop quantity, Q_count as production-to-count under the applicable policy, and Q_short as their nonnegative difference, subject to guarantees, appraisals, unit-level adjustments, and contract terms. A simple quantity-based gross loss measure is max(0, Q_base - Q_count); actual indemnity further depends on the guaranteed quantity, elected price, liability limit, deductible, share, and authorized quality rules. The analyst must never equate the gross difference with a payable claim until all conditions are applied.

Quality-adjusted production can sometimes be expressed using a contractually approved quality factor a between zero and one, giving Q_adj = a × Q_measured. That notation is an illustration of multiplication, not a representation of every RMA crop formula. As the 2026 RMA handbook shows, eligibility for quality adjustment, how discounts are determined, and when unsold or stored production is valued all depend on the relevant provisions. Other crops, contract forms, and countries can apply different calculations entirely.

Double counting can arise in several ways: lowering the payable quantity and then also paying the full grade discount; paying a drying charge already reflected in a net value measure; treating a weight difference caused by drying as lost dry matter; or reimbursing the same crop value under two overlapping programs without offset. The settlement record should reconcile each physical change and monetary component to one approved basis of payment.

Indian farm-level claim design

Section 21.6.6.2 of the 2023 PMFBY guidelines sets out a farm-level postharvest calculation as sum insured multiplied by the proportion of insured area affected and by estimated crop damage within that affected area. In symbols, for a qualifying claim, B = SI × A × D, with A and D expressed as proportions. The actual payment also remains subject to notification, eligibility, outstanding balance of sum insured, survey rules, and interaction with other benefits.

A hypothetical arithmetic illustration makes the calculation transparent. If an eligible insured plot has a sum insured of INR 50,000, 60% of its insured area is affected, and the estimated qualifying damage within the affected area is 40%, the illustrative component is INR 50,000 × 0.60 × 0.40 = INR 12,000 before any applicable remaining-limit and overlapping-benefit checks. These numbers are invented only to illustrate the formula; they are not a measured PMFBY claim or reported official case.

The operational design links an objective weather-related eligibility dimension to individual crop-damage assessment. In principle, this combination may avoid the mismatch of a wholly area-based index with individual postharvest exposure. It also creates costs of timely site visits, survey consistency, farmer notification, and disputes over whether crop material was eligible and remained in the required field condition. The 2023 joint-assessment provisions prescribe institutional participation and digital evidence collection, but observed quality of implementation requires administrative microdata.

The 2023 central guidelines and the 2026 official PMFBY summary should be used together: the latter supports the continued public description of the benefit, while the former specifies the detailed cited procedures for its edition. A current state claim analysis must additionally obtain the actual state-notified crop, tender, season, district, and window. Without that record, program-level recognition of the feature does not establish legal entitlement for a particular crop or location.

United States appraisals and delayed harvest

The USDA 2026 loss adjustment standards make the counterfactual harvest opportunity central when damage continues after the normal end-of-period date. Evidence includes the initial insured cause, harvest equipment, surrounding producers' completion, conditions of the field, crop variety, normal growing season, and actual windows in which reasonable harvest was possible. A mature crop left in the field simply to avoid drying cost is treated differently from a crop a qualifying event prevented the producer from harvesting.

A claim may be determined by appraised unharvested production if harvest will not occur or by recorded harvested production when the insurer allows time for a delayed harvest. Such a procedural allowance does not move the official calendar endpoint of coverage. It seeks to measure the loss stemming from a peril in the original covered period while taking the producer's mitigation conduct into account. If a later independent uninsured event contributes damage, attribution must be separated to the extent supported by evidence.

The RMA handbook also specifies procedures for measuring farm-stored production and, in some circumstances, allowing delayed measurement. These procedures serve the reliability of production-to-count and quality adjustment. A scientifically valid comparison should never categorize all delayed measurement cases as insured warehouse losses. Administrative timestamps for measurement and payment are not substitutes for event occurrence and causal timestamps.

Overlapping public and private instruments

Postharvest claims can intersect with wider seasonal yield insurance, farm disaster assistance, storage or property insurance, credit arrangements, and mutual funds. Coordination rules are therefore material. The 2023 PMFBY operational text includes benefit settlement and account reconciliation processes across claim components, and the EU CAP Article 76 includes anti-overcompensation requirements for combined public or private risk-management support. These are distinct legal architectures, yet both identify compensation overlap as a potential source of error.

A cross-program study should maintain a benefit ledger containing event, insured unit, measured loss category, deductible, policy share, provisional payment, final payment, offsets, and recoveries. It should also distinguish an ex gratia government transfer from contractual indemnity. Financial assistance and insurance can produce different incentives, legal rights, timing, and fiscal accounts even when both respond to the same storm.

Administrative effectiveness cannot be inferred from formal payment deadlines alone. A rigorous evaluation requires actual dates of loss occurrence, report, survey assignment, field visit, claim determination, insurer receipt of relevant funding, payment instruction, and beneficiary receipt. The Indian government's 2026 explanation of PMFBY and the scheme's detailed procedures make clear that institution-level coordination is an empirical variable. Here the 2026 press note is used for its program-level description, not to certify a particular crop year's settlement performance.

Evidence Validity and Sensitivity

Primary-source hierarchy and rule versions

The evidentiary hierarchy for this subject begins with the exact insurance contract and legal instruments applicable to a particular insured crop year and location. These include general policy conditions, crop-specific and special provisions, state notifications, attached endorsements, official operational handbooks, official claim forms, and insurer loss-adjustment records. General institutional web pages aid interpretation and discovery but do not replace the binding instrument. Empirical scientific papers inform exposure and measurement, but do not establish a legal entitlement absent contractual text.

This monograph checked India's 2023 PMFBY rules, USDA's 2026 claim-adjustment instructions, and USDA's 26-BR policy text as dated primary materials. The date annotations matter. The 26-BR policy's summary specifies crop-year applicability conditioned on contract-change dates; changes for 2027 and later periods must be handled on their own terms. The consolidated EU Article 76 text expresses the legal framework as of its cited consolidation and does not establish the terms of every member-state scheme.

A source retrieved in 2026 should not be cited as though it governed a 2014 harvest. A working paper published in 2014 should not be presented as a current estimate of every nation's farm loss. Similarly, a site-wide explanatory FAQ may lag a program amendment or leave out state-level variation. The evidence ledger appended to this package identifies what was checked and which claims each source actually supports.

Observational, measurement, and selection problems

The empirical literature demonstrates why postharvest loss evidence is difficult to combine. Kaminski and Christiaensen's 2014 World Bank study uses farmer survey data in Malawi, Tanzania, and Uganda; the often-cited reported maize-loss range depends on the covered population and survey measure, while the distribution of losses is strongly heterogeneous. Such figures should be interpreted as estimates of farm postharvest outcomes in surveyed settings. They are not estimates of insured postharvest claims in India, the United States, or the European Union.

The Affognon et al. (2015) synthesis finds important variation in assessment techniques and loss definitions. Delgado, Schuster, and Torero (2021) compare alternative quantity and quality measurement methods across selected value chains and report that aggregate self-reports tend to understate losses relative to more detailed measurement. These results jointly imply that claims studies should use verified crop weights and documented grade/condition when feasible and quantify how survey-only methods differ.

Four forms of selection require particular attention. Enrollment selection determines who buys or receives eligible insurance. Crop and area notification selection determines which producers can use a special benefit. Reporting selection determines whether an eligible loss enters the claim system, especially under short notice requirements. Settlement selection determines which claims are recorded as closed or paid by the observation date. An administrative dataset of paid claims therefore measures an end product of multiple filters, not the full physical occurrence distribution.

There is also left-truncation and right-censoring. Preexisting defects may be omitted if sampling begins after a rain event; later payment values may remain unknown when records are extracted. Moreover, an insured farmer may have stronger incentives to document rainfall and damage than an uninsured farmer. Comparing losses across those groups without accounting for enrollment, crop type, storage behavior, and exposure duration can produce misleading differences.

Comparability and counterfactual identification

The appropriate population for a contract-specific research question is the set of insured eligible crop units exposed during the actual insured window. A food-system question may instead use all agricultural units or all quantity entering a specified chain. Mixing those denominators corrupts both claim rates and effectiveness calculations. A postharvest insurance penetration ratio based on total crop output can be useful descriptively, but it is not an actuarial frequency measure if only harvested lots temporarily left in a qualifying state are at risk.

To identify an insurance impact on productive decisions or welfare, compare outcomes under a credible counterfactual. For example, the outcome of interest might be investment in temporary drying capacity, probability of bringing crop to market, producer income volatility, time to settlement, or damage reduction. A comparison of paid claimants with nonclaimants cannot estimate policy effectiveness because receiving a payment is itself selected on adverse weather and damage. Randomized rollout, carefully validated discontinuities, staggered adoption, or credible matched panel designs may become relevant when actual administrative and farm datasets are available.

A narrower descriptive study can still be powerful. It can report loss-notification rates by notified crop and district, inspection times, agreement between satellite weather evidence and field observations, average quality discounts for the same crop grade, and proportions of reopened claims. Such statistics require consistent contract versions, recorded data provenance, and explicit handling of missing records. They should not be presented as causal effects.

Sensitivity analysis and documentation standards

Every quantitative comparison should expose sensitivity to at least six assumptions: the exact start and end of eligible exposure; the chosen moisture standard; the estimated unaffected quantity and quality; the assignment of mixed causal mechanisms; the spatial weather threshold and observation network; and salvage or alternative-use valuation. Additional tests should vary reporting completeness, farmer selection, claim closure censoring, inflation adjustments, and crop-year policy changes.

An illustrative claim audit can recalculate a lot under different but explicitly documented moisture assumptions. A 100-ton lot at 20% moisture converted to 14% moisture has a theoretical dry-material-conserving weight near 93.02 tons. Changing the assumed moisture fractions changes the reconciliation and therefore the unexplained residual. The analyst must disclose sampling uncertainty, whether contaminants or broken kernels were measured, and whether the claimed loss pertains to total mass or marketable dry matter. None of these assumptions should silently alter the insurer's own authorized rules.

Weather reanalysis and station observations can help reconstruct a regional event, but a model grid's rainfall value may diverge from field exposure. For a threshold-based qualification test, reporting the nearest weather station, missing data, temporal aggregation, benchmark period, and sensitivity to alternate stations is necessary. Physical damage evidence should be treated separately. A threshold above the specified amount is neither sufficient proof that a particular farmer lost production nor proof that every drying lot was eligible.

Internal QA of this manuscript confirms the logical separation of contractual coverage, agronomic damage, and statistical measurement. It does not claim an independent peer review or an original calibration to insurer microdata. Public-source limitations include restricted access to full insurer claim files, individual PMFBY state notifications, crop-specific RMA schedules for every crop, and proprietary storage insurance wording. Those constraints limit statements about actual market coverage and claims experience, while leaving the cited comparative legal and methodological propositions testable.

Scientific Synthesis and Research Agenda

Comparative findings and design implications

The comparative record supports six substantive findings. First, postharvest is a stage label in agronomy and statistics, while insurance requires a specified contractual attachment. Second, a specialized time-limited in-field benefit, as demonstrated by PMFBY, can recognize harvested-crop vulnerability without promising open-ended inventory coverage. Third, under the USDA framework, post-period evidence and delayed-harvest adjustment may address damage rooted in an insured earlier cause without modifying the policy's calendar endpoint. Fourth, quantity and quality damage require different evidence and may produce different monetary results. Fifth, weather monitoring and field assessment complement one another because event occurrence and individual damage are distinct facts. Sixth, overlapping benefits and incomplete administrative records can distort any assessment of financial protection.

The first finding changes how coverage inventories should be constructed. Researchers should catalog policy clauses, not merely country names and program names. A robust inventory identifies crops, eligible lots, peril list, duration, state or regional selection, reporting windows, basis of assessment, limits, and exclusions. A checkbox labeled "postharvest coverage: yes" has low analytical value without those dimensions. The 2023 PMFBY guidelines and USDA 2026 claim-adjustment handbook provide concrete examples of why.

The second finding places drying infrastructure at the center of both risk control and social access. Producers with rapid collection capacity may carry fewer exposed days; smallholders with limited equipment may leave material on fields longer. A nominally uniform 14-day time limit may therefore protect different fractions of their actual vulnerable production. This is a plausible mechanism that requires direct empirical testing; no distributional effect is quantified here.

The third finding clarifies legal time. A loss resulting from an insured event can be discovered, quantified, and paid later; the later date is not itself evidence of fresh risk attachment. This distinction matters whenever a crop is delayed on the field or measured after storage. Researchers and adjusters should preserve original event dates alongside adjustment dates and should avoid coding late-paid claims as new postharvest-peril claims.

The fourth finding requires data on both dry-matter balance and grade. A moisture-based weight decline may be a normal transformation; a quality discount may be economic damage with conserved mass; and an actual amount of missing crop may coexist with both. A claim audit that applies one unexamined percentage to all three phenomena cannot demonstrate accurate settlement.

The fifth and sixth findings concern institutional capacity. Weather data and app-based images can make a short-window program more auditable, but access to technology, surveyor density, and verification speed may influence who is paid. Public support can expand access while also creating coordination tasks among farmers, insurers, government bodies, and reinsurance partners. The EU risk-management article adds the policy-level requirement of avoiding overcompensation across interventions.

Research priorities and testable propositions

The immediate research agenda is an international, crop-year-specific contract inventory of postharvest risk. The unit of observation should be a legally identifiable cover, not a general reference to agricultural insurance. Records should contain original policy wording, effective dates, named hazards, harvesting states, storage transitions, conditions, and actual publication provenance. Comparing those records could determine which products cover in-field exposure, stored inventory, transit, quality damage, or combinations, without conflating them.

A second priority is linked physical and administrative evidence: representative plot measurements before and after events, moisture-normalized lot weights, independently verified grade tests, weather observations, insurance eligibility, notice timestamps, loss-adjuster records, and settled indemnities. A third is measurement research, directly comparing farmer recall, transaction-based quantity reconciliation, physical sampling, and remote sensing under the same crop and period. The Delgado et al. methodological comparison supplies a demonstrated reason for such multi-method work.

A fourth priority is risk-pricing reliability under changing harvest patterns. Climatic variability, seasonal precipitation concentration, and the adoption of mechanical harvest or drying technologies can change exposed quantities and durations. Relevant actuarial tests should examine whether a past crop-year claim rate predicts a later insured exposure, whether weather-event clustering is correctly represented, and whether disaster-year reporting and delays cause apparent changes in severity. These are standard validation tasks rather than the announcement of a new risk doctrine.

A fifth priority concerns access and fairness. Short notice periods may impose varying practical burdens on farmers with limited connectivity or administrative assistance. Empirical studies could estimate differences in eligible loss reporting and settlement conditional on event severity and crop exposure, rather than inferring unfairness or effectiveness merely from program design. State notification rules in India and crop-specific policy boundaries in the United States should be tested as institutional determinants of real coverage.

The final priority is prevention interaction. Research should measure whether insurance changes collection speed, field-drying exposure, storage adoption, and product salvage, while controlling for infrastructure and weather. The Sheahan and Barrett evidence review supports comparing interventions by economic benefit and opportunity cost. Insurance and prevention can work together when premiums, deductibles, proof requirements, and extension services reflect actual producer constraints.

Conclusion

Postharvest agricultural production loss coverage is defined by the conjunction of insured interest, production stage, covered peril, specified insurance period or extension, causal proof, and admissible settlement method. Postharvest damage is a broad agricultural reality; postharvest insurance protection is a precisely delimited contractual relation. Their intersection must be proven for each claim.

The original sources examined here demonstrate real variation in that relation. India's PMFBY provides a state-notified, time-limited in-field drying benefit for named hazards and an individual loss-assessment mechanism. The United States federal crop insurance framework directs attention to the insurance-period endpoint, insured causes preventing timely harvest, reasonable mitigation, quality adjustment, and production-to-count records. The European Union supplies a legal framework for supporting agricultural risk-management instruments while relying on implementing choices and controls against overcompensation.

The postharvest empirical literature establishes a second decisive principle: physical quantity, quality deterioration, market value, and insurance payout must be measured and interpreted separately. Contract-specific exposure denominators, dated legal sources, actual lot evidence, credible weather records, and transparent accounting are prerequisites to defensible estimation. This analysis supplies a documented comparative foundation and specific research priorities. It neither asserts a new authorial theory nor attributes causal results to data that were not collected.

References