Beyond the 5.2 Trillion RMB Cover: The Reality of Agricultural Insurance Claims

Yi Fujin: The 155 billion refers to premium income, which comprises premiums paid by farming households plus subsidies from governments at all levels, totalling 155 billion RMB. The 5.2 trillion RMB in risk coverage is a theoretical figure based on the assumption that all insured assets suffer total loss and claims are settled up to their actual policy limits. This represents 60% of the country’s agricultural GDP, which is already exceptionally high. However, 5.2 trillion is not liquid funds; it would rely on premiums accumulated over many years to be paid out all at once. It is largely a theoretical number, and in practice, it is highly unlikely that such a volume of claims would ever need to be settled simultaneously. What we should really focus on is the 155 billion RMB figure. The scale of agricultural insurance premiums has now reached the largest in the world. Of this 155 billion, roughly 80% comes from government funding, with the remaining 20% or so contributed by farmers’ premium payments.

Yi Fujin: The structure of agricultural production in China has a very distinctive feature: “大国小农” (the structural reality of China’s vast territory managed by fragmented smallholder farms). Currently, smallholder farmers account for roughly 97% of agricultural producers in the country. This fundamental reality means that agricultural insurance cannot achieve 100% precision in practice. Coupled with the inherent complexity of farming itself, agricultural insurance is also unlikely to provide the kind of door-to-door, individualised service that motor or property insurance typically offers.

During our fieldwork in Henan last year, we came across a fairly typical township comprising over 20,000 mu (approx. 0.165 acres) of arable land and more than 5,000 farming households. Insurance companies are already severely stretched for personnel at the grassroots level. Typically, a single local insurance agent is assigned to cover two to three townships of this size, and in some regions, even more. In other words, one individual might need to oversee 40,000 to 50,000 mu of land and tens of thousands of households. The challenge is further compounded by the fact that agricultural production is highly time-sensitive. Consequently, the state imposes equally strict deadlines on agricultural insurance, requiring the entire claims process—from on-site inspection and loss assessment to the disbursement of funds directly to households—to be completed within 14 days.

The issue, however, is that claims surveys, loss assessments, verifications, and disbursements across tens of thousands of mu of farmland must be completed within just over ten days, creating immense pressure. It is hardly surprising that, in practice, faced with a vast number of smallholder farmers, it is virtually impossible for insurance companies to conduct door-to-door, plot-by-plot damage assessments. This stands as the greatest obstacle to implementing agricultural insurance under the national reality of “a large country with smallholder farming”.

Consequently, some regions now employ an alternative approach, issuing a “collective insurance policy” to the village collective. When smallholder farms in a village suffer damage, the insurer will randomly select three to five plots within the village within 14 days. Based on these samples, they assess the overall scale and severity of the disaster across the village, then determine the village’s total payout amount according to the average loss level. Once the total figure is confirmed, village committee officials—who are effectively auxiliary claims staff hired by the insurer—distribute the funds among individual households based on their actual circumstances, thereby completing the claims settlement for each smallholder.

This gives rise to two possible outcomes.

The first scenario is that, although the current model is nominally termed “yield insurance” or “full-cost insurance”, it actually operates on the principle of regional yield insurance, with coverage based on the village’s average yield. For instance, a farmer’s plot on a mudflat might be completely washed away by flooding, while the village as a whole remains unaffected. Under the random sampling method, that specific plot may not be selected, or its loss might be judged to have no material impact on the village’s overall yield.

The second possibility stems from land fragmentation following rural land reform. Many farmers’ arable land is no longer contiguous; a household with ten mu might have it split into four or five parcels, with fragmentation being even more pronounced in mountainous areas. If only one of those plots is affected, the insurance contract may not recognise it as a qualifying disaster under its terms, thereby failing to trigger even the most basic payout. Yet farmers often reject this logic, arguing that if their specific plot is damaged, they should receive compensation regardless of the loss’s scale. This widespread dissatisfaction with agricultural insurance stems from farmers’ misunderstandings of its fundamental principles, compounded by inadequate publicity and explanation.

Consequently, whether due to the operational mechanisms or the inherent characteristics of agricultural insurance itself, it inevitably leads to so-called misallocated or missed claims.

For large-scale operators, the threshold is generally at least 50 mu. In certain regions, such as the northeast, this can reach 100 or 300 mu. Almost every province mandates that insurers issue separate policies and contracts for these larger holdings. Following a disaster, insurance adjusters must visit these farms specifically to assess the losses. By comparison, the accuracy of agricultural insurance claims for this group is significantly higher.

◉In 2025, Shaanxi suffered a severe drought. Wheat on the river terraces south of the Wei River turned yellow and withered, with flag leaves completely dried out; the soil around the wheat roots felt scorching to the touch during the day. Photo: Kong Lingyu

Yi Fujin: There is another reason why farmers feel that payouts fall far short of their expectations: payouts from agricultural insurance (specifically full-cost or materialised cost insurance) are actually limited to covering the cost of factor inputs.

Unlike industrial production, where all inputs are deployed upfront, the total cost of agricultural production is invested gradually throughout the growing season. For example, at the seedling stage, only about 50% of the total cost may have been expended. By the jointing stage, cumulative investment typically reaches around 70%, rising to roughly 90% by flowering, with the full cost only committed once the crop reaches maturity and harvest. As a result, claim payouts depend entirely on which stage of production a crop is in when the disaster occurs.

To put it plainly, suppose a disaster strikes during the wheat seedling stage—for example, an unseasonable spring frost after the Lunar New Year that kills the young crops. In such a scenario, insurers typically do not pay out the policy’s maximum coverage limit. Instead, they calculate the claim based on the inputs already committed at the seedling stage. On a full-cost basis, this early phase might account for just 50%. Therefore, even if the contract specifies a coverage level of 1,000 RMB per mu, a crop freeze at this stage might yield a maximum payout of only 50%, or 500 RMB per mu, because subsequent costs for the jointing and flowering stages have yet to be incurred. Conversely, if a disaster hits at harvest time—such as crop lodging or persistent heavy rain causing total yield loss—the payout could reach 100%. Agricultural insurance involves numerous technical nuances that are rarely spelled out clearly in the policy wording. Yet in practice, claims are settled precisely along these lines.

Yi Fujin: In the industry, this practice is known as a “premium rebate”. In years when farmers are unaffected by disasters, insurers will rebate the premium to maintain uptake for the following year; should a disaster strike, claims are still settled through the standard process.

How should we view the premium rebate issue, and what lies behind it? Payouts for smallholder farmers rely on sampling-based loss assessments, which are inherently imprecise. Meanwhile, the premiums are priced too high. One could even say that such policies are, in effect, ‘defective’. To keep farmers willing to purchase them, insurers have to give them something for a product that isn’t worth its asking price. That is why, in practice, a portion of the premium is returned to farmers in years when no disasters strike.

However, this is by no means beneficial for agricultural insurance. The fundamental purpose of agricultural insurance is to accumulate a surplus during non-disaster years, reserving funds to cover payouts when genuine disasters occur. In current practice, funds that ought to be retained as surplus are being paid out. This leaves insurers with no liquidity when a major catastrophe strikes, resulting in what is known as the problem of ‘overpaying for minor disasters and underpaying for major ones’. This is precisely why the 2024 No. 1 Central Document introduced the ‘dual precision’ policy—focusing on precise enrolment and precise claims. This has emerged as the most pressing practical challenge facing agricultural insurance today.

◉Widespread lodging in rice paddies following flood submersion, with mud and water clinging to the panicles and stalks. Photo: Yvonne

Yi Fujin: There are considerable differences among insurance companies. Take Zhongyuan Agricultural Insurance, Henan’s largest insurer, for example. It operates its own branches at the grassroots level, often referred to as county-level sub-branches. For minor disasters, as long as the payout amount is not particularly large, the county-level or prefecture-level sub-branches can process and settle the claims directly. However, for particularly severe disasters, especially widespread incidents like those seen last year, a county branch, or even a prefecture-level sub-branch, or indeed the provincial company, would simply be unable to cover the payouts.

Because only 40 RMB in premiums is collected per mu of land, while that 40 RMB is intended to cover a 200 RMB payout should a farmer suffer a disaster in the same year. In other words, the premium collected is insufficient to cover the 200 RMB payout amount. Under these circumstances, claims during major disasters must be funded from risk reserves accumulated in previous years (that is, risk surplus).

Typically, insurance companies set aside a portion of premiums as a “catastrophe reserve fund” in normal years. They also purchase agricultural reinsurance from reinsurance companies. If a local branch faces claims it cannot cover, payouts can be drawn from the catastrophe reserve fund, or the risk can be spread through reinsurance companies. The major challenge now is that China still lacks a comprehensive catastrophe risk response framework.

In an ideal scenario, for a company such as PICC (People’s Insurance Company of China), if a regional branch were to exceed its payout limits, headquarters could theoretically mobilise resources across different areas to coordinate the response. Local agricultural insurers, however, such as Zhongyuan Agricultural Insurance in Henan or Guoyuan Agricultural Insurance in Anhui, are far more constrained. When a major disaster strikes locally, payout pressures become extremely high, making it difficult to disperse risk on a national scale. This can trigger a situation akin to a bank run, where the insurer cannot immediately cover such massive payouts, potentially leaving it under severe financial strain.

In countries such as the United States, Canada, Japan, and even India, central governments typically establish emergency funds. These can draw on contingency reserves already built into the fiscal budget, and can also issue public bonds to absorb the financial shock of catastrophic events. China’s institutional framework in this area remains comparatively underdeveloped. At the national level, the China Agriculture Reinsurance Company Ltd was founded in 2020. The system currently operates through a 20% ceded reinsurance arrangement: for instance, if an insurer collects 1 billion RMB in premiums within a province, it must cede 20% of that amount—200 million RMB—along with the corresponding risk to the reinsurer. Through this mechanism, the national reinsurance company and local insurers form a risk-sharing framework.

But this does not address the root of the problem. The risk of payouts exceeding reserves is widespread – for systemic catastrophes on the scale seen in Henan, Anhui, Shanxi and Shandong, a reinsurance mechanism simply isn’t enough. Under these circumstances, no commercial insurer could operate sustainably. Yet our country’s institutional framework in this area still lags behind.

◉Tractors become bogged down in the mud after rainfall, unable to move. Operating heavy machinery on cultivated land in wet conditions further compacts the soil, degrading the farmland. Photo: Xiaoliushu Farm

Yi Fujin: It does sound profitable, and indeed it is. The overall risk associated with China’s bulk agricultural products—namely, staple crops—is relatively manageable. Even as disasters intensify year on year, national food security has steadily improved, with consecutive years of rising yields and bumper harvests. This means that, in broad terms, crop losses remain controllable. Given this backdrop, the use of collective insurance policies and consolidated claims processing for smallholder farms has significantly reduced administrative costs for insurers. As a result, over a considerable period, the average profit margin for agricultural insurance hovered around 9 per cent.

The issue is not that profit margins are too high, nor that insurers ought not to make this money, but rather what this money actually represents. On paper, it naturally appears as profit. However, considering the substantial year-on-year fluctuations in the frequency and severity of disasters inherent to agricultural insurance, the reality is quite different. Take the United States, for instance, where agricultural insurance is the most advanced and has the longest history. In extreme years, loss ratios can reach 160%, 170%, or even 200%. This would be unimaginable in the motor insurance sector, yet it is a defining feature of agricultural insurance. When a catastrophic event strikes, such as in 2022 when regions south of the Yangtze River experienced virtually no rainfall for well over half the year, a loss ratio exceeding 200% would be entirely normal if claims were settled based on actual losses.

Consequently, the surplus funds recorded on the books of many insurance companies can be viewed not only as profit but also as a risk reserve fund intended to cover future catastrophic events.

However, a catastrophe risk protection framework has yet to be established, and the government remains significantly behind in developing relevant policies. These funds also essentially reflect the benefits insurers have reaped from partnering with the state. Through this cooperation, the government absorbs the costs, while the profits flow to the insurance companies. When this accumulated surplus is recorded as book profit for insurers, it should in future be pooled and managed at national and provincial levels, structured as catastrophe reserve funds or management funds. Corresponding adjustments will also need to be made to insurers’ accounting practices going forward.

◉In 2023, the Guanzhong region of Shaanxi experienced ten consecutive days of “烂场雨” (persistent rain during harvest that ruins crops), causing partial moulding of the wheat. Source: 绿我农场

Yi Fujin: The profit margin might not be that low, but overall it remains relatively high. That said, a combination of factors has contributed to this situation.

First, as climate change drives an increase in extreme weather events, the payout pressure on insurance companies has undeniably been mounting.

Second, the government has also observed that insurers’ average profit margins remain notably high. Since these are public funds, local authorities are putting substantial money into premium subsidies. In regions where the mechanics of insurance are less well understood, there is often a requirement to spend allocated budgets in full—for example, if fifty million RMB is earmarked, it must be disbursed by year’s end. This suits local governments, as the funds yield immediate results within the fiscal year and are paid out to farmers. Yet, when major disasters actually strike, this approach creates serious vulnerabilities.

Third, farmers are becoming more assertive in defending their rights, which has placed considerable pressure on insurers to maintain strict operational compliance. That said, moral hazard is certainly a factor. During our field research in Henan and Anhui last year, we encountered this issue firsthand: some farmers felt that payouts of eighty or ninety RMB were insufficient and demanded claims of at least five hundred or six hundred RMB, simply because the policy limit was a thousand RMB. Yet, given the insurers’ actual payout capacity, meeting those demands is fundamentally unfeasible.

Fourth, under increasingly stringent regulation in recent years, insurers have also been compelled to maximise claim payouts wherever possible.

These four factors are collectively driving the current decline in profit margins. Is a fall in profitability a positive or a negative development? From a personal standpoint, I wouldn’t say it’s necessarily a good thing. As the major disaster risk management fund has yet to be established, insurers’ capacity to meet claims will be severely compromised if a major catastrophe strikes. So, while a drop in profit margins is undoubtedly welcome from a farmer’s perspective, it is also a worrying signal.

Yi Fujin: I don’t have the complete figures, but from what I’ve gathered indirectly, the levels aren’t particularly high and may even be declining. In some countries, such as Japan, risk capital reserves have built up to 1.6 times premium income to cope with extreme disasters. We are still far from reaching that standard in China.

◉In July 2025, torrential rains struck Beijing once again, engulfing all 450 mu of land at Sohu Farm, located beside the Qingshui River in Beizhuang Town, Miyun District. The floodwaters nearly erased every trace of the farm, leaving behind only collapsed road foundations, waterlogged mud, and scattered construction debris. Photo: Zhen Rui

Yi Fujin: Indeed, from 1982 to 2004, the sector was operated on a commercial basis. A purely market-driven approach has failed. This is not an isolated case in China; globally, no government has managed to establish agricultural insurance purely through market forces, largely because agricultural risk is exceptionally high. On the supply side, the costs of providing agricultural insurance are also prohibitively high. On occasion, I’ve even half-joked that, although run by commercial entities, agricultural insurance’s geographical coverage is second only to that of the national gov ernment. It essentially requires having a dedicated insurance representative in every single township. The combined costs of staffing, transportation, and technical support are enormous. These high costs simply mean that agricultural insurance is not a profitable venture. Overall, attempting to commercialise agricultural insurance without government backing—whether financial or administrative—has yielded no successful examples anywhere in the world. In practical terms, it is simply unworkable.

Yi Fujin: There is an element of speculation on the part of the farmers here. The reason insurance is failing to fulfil its intended role as a risk safeguard—largely due to the current strain on local public finances—lies in the specific logic governing fiscal subsidies in China.

Under the current subsidy framework, once insurers collect 20 per cent of the premium from farmers, the county-level government must provide a matching contribution of 5 to 10 per cent. Only with this matching funding in place does the process move to the prefectural level for further subsidies, then up to the provincial level, and finally to the central Ministry of Finance. Central funds will only be released once it is confirmed that subsidies from the three tiers below are all in place. This gives rise to the so-called outstanding premiums issue. It is particularly pronounced in the central and western regions, where local budgets are stretched thin and authorities simply cannot find the money. Right from the outset of the application process, even if farmers have paid their share, if the local government cannot contribute its portion, the entire insurance chain breaks down. The policy is never officially issued, which remains a major problem today. With local finances under such strain, if the process does not even reach the stage of formalising the contract, how can payouts even be considered?

Recent figures indicate that the central government has allocated a budget of around 50 to 60 billion RMB. Yet in practice, since 2024, the disbursement rate for central subsidies has already fallen below 100 per cent. The core issue is that local authorities are unable to provide the initial matching funds to cover farmers’ premiums; without this, central allocations cannot be released. This remains one of the most pressing challenges at present.

That said, local authorities remain indispensable from an implementation standpoint. Without the backing of township governments or village collectives, smallholder farmers would simply not participate in agricultural insurance programmes.

◉In early August 2023, Wuchang in Heilongjiang experienced rare torrential rain and flooding, submerging roughly 40% of the area’s 2.5 million mu of farmland (approx. 167,000 hectares). The section pictured was originally a rice paddy, now completely underwater. Photo: Yvonne

Yi Fujin: When facing minor flooding or heavy, concentrated rainfall, insurers can certainly respond by supplying farmers with pumps, drought-resistant varieties, and similar measures. But once an extreme disaster strikes, these measures are inevitably ineffective.

Logically, I have no objection to that viewpoint; pre-disaster prevention is undoubtedly important and can be highly effective against certain types of events. However, when it comes to more severe disasters, such preventive measures offer very little practical value.

The most fundamental question, then, is what the actual role of agricultural insurance is meant to be.

Were it to be positioned primarily as a preventive measure, current financial regulations would not permit it anyway. From discussions with colleagues in the industry, I have learned that only 20 per cent of the premium pool is allocated for administrative and operational expenses. Insurers can only draw a small portion from this 20 per cent for preventive initiatives; they are strictly prohibited from using the remaining 80 per cent for such purposes.

That said, attempting to replace more effective interventions—such as investing in rural water infrastructure, advancing agricultural technologies, or developing high-standard farmland—with agricultural insurance would clearly be a case of misplaced priorities. Therefore, clarifying the specific role and boundaries of agricultural insurance within China’s broader agricultural policy framework is absolutely crucial.

I have encountered a wide range of perspectives on this, but the prevailing view among academics, government officials, and insurance professionals is that state subsidies for agricultural insurance premiums are fundamentally designed to support the agricultural sector and help farmers manage risk. On the surface, that reasoning holds up perfectly well.

Yet there is a deeper issue to consider. When viewed from a broader perspective—particularly regarding staple crop production—reliable grain supply, or food security, functions as a public good and the cornerstone of social stability and economic development. Everyone benefits from it. In typical harvest years, consumers have ample access to food at stable prices and reap the rewards of a secure supply chain. However, in years of poor harvests, the state intervenes through macroeconomic policy: releasing reserves to stabilise prices and importing grain to balance domestic supply. Under this system, consumers are shielded from significant financial loss. Farmers, by contrast, bear the brunt of these shortfalls as a byproduct of the national food security strategy, without receiving any corresponding risk premium or even partial compensation.

In other words, while society as a whole reaps the benefits of a secure food supply in pursuit of this strategic goal, we have yet to pay the corresponding risk premium for the very possibility of crop shortfalls.

So from this perspective, the government’s premium subsidy for agricultural insurance is not a subsidy, but rather the risk premium that the government ought to pay to ensure food security.

Viewed through China’s millennia-long history, the adage “cheap grain hurts farmers” rings true — increased yields bring no benefit to farmers because prices fall; yet when yields drop, farmers still suffer losses despite the state’s effective grain price controls. When considering both scenarios together, farmers effectively bear all the losses, while society shares in the market benefits, which is fundamentally inequitable. In other words, benefits and risks are entirely misaligned. Therefore, the government should take measures to help farmers manage risk, so that low risk and low returns can be properly matched. From the perspectives of risk management and social equity, the government is absolutely justified in making this payment.

▼

Click the image to read related articles

▼