In July, Typhoon Bavi and Typhoon Maysak impacted multiple regions across China, forcing the relocation of more than 1.7 million people in Zhejiang, breaching reservoirs and triggering severe flooding in Guangxi, with subsequent storms increasing rainfall in northern China by nearly 30%. As the tragic Miyun floods that claimed the lives of 31 residents in an elderly care home last year and the 2024 Typhoon Yagi that cost billions of RMB in damage to Hainan’s fisheries remain vivid in public memory, this year’s disasters serve as another reminder that climate risks are becoming increasingly severe, frequent, and erratic. Floods alone are estimated to cost China roughly 1% of GDP each year, with indirect losses through supply-chain disruptions and economic dislocation adding significantly to the total burden. While much attention has focused on disaster prevention and emergency response, an equally important question receives far less attention: how should these growing expenses be financed?

China’s strategy for natural catastrophes can be characterized by a top-down, government-led “Whole-Nation System” focused on building robust institutional capacity across disaster prevention, early warning, emergency response, and relief. Its financing model is similarly state-led, with public finance dominating both disaster prevention and relief. From 2011 to 2024, roughly 77% of the investment for water conservancy projects came directly from central and local governments. The central government also sets aside natural disaster relief funds as part of its fiscal budget each year. In 2026, it allocated 20 billion RMB, of which 1.19 billion RMB had been released during this flood season for immediate relief. This was supplemented by another 14.5 billion RMB mobilized through sector-specific channels such as agriculture and transport to fund corresponding reconstruction. By contrast, private disaster financing remains limited. Natural Catastrophe (NatCat) insurance has generally compensated less than 10% of China’s disaster losses in recent years, compared with a global average of around 40%.

However, public spending is increasingly struggling to keep pace with the country’s growing disaster prevention and relief needs, especially as climate hazards increase in frequency and intensity. Despite substantial investment in flood protection, aging reservoirs, dams, and underground pipelines continue to hold back disaster resilience particularly in rural areas. Infrastructure designed according to historical risk levels also becomes inadequate as extreme events exceed previous assumptions. Moreover, while disaster governance is directed from the top, much of its implementation and financing remains decentralized. Local governments shoulder the frontline responsibility for emergency response and much locally oriented preparedness, whereas the central government finances co-fund projects with wider strategic significance and provides additional support when disasters overwhelm local capacity. This arrangement leaves fiscally constrained localities particularly exposed. Against the backdrop of rising local-government debt, localities are not only less able to absorb immediate losses, but also less capable to maintain infrastructure and invest in preparedness for future emergencies. These pressures make it necessary to look beyond direct public spending for additional ways to finance and manage disaster risk. 

Granted, public finance remains the primary source of disaster prevention and relief worldwide as many resilience projects provide public benefits without generating clear revenue streams. Private capital can play a role where adaptation projects produce income—for example through water tariffs, electricity sales, or other operating revenues—and public funds can help make such projects investable through interest subsidies, guarantees, and risk-compensation facilities. However, the scope for private capital to directly finance essential public infrastructure remains inherently limited. Therefore, the objective should be to make public finance work more effectively rather than to substitute for it. Private capital’s comparative advantage lies in risk sharing through insurance. Insurance can pool, price, and transfer the residual losses that neither protective infrastructure nor public budgets can fully absorb, while also creating incentives for risk reduction before disasters occur. By converting volatile disaster liabilities into predictable premiums and timely payouts, it can reduce fiscal uncertainty and help ensure that responding to one disaster does not weaken preparedness for the next.

This approach is not entirely new to China. Over the past decade, a number of local governments have experimented with publicly-funded, commercially-underwritten catastrophe insurance to mobilize market mechanisms that amplify the impact of fiscal expenditure. These pilots provide early evidence of both the potential and limitations of regional insurance coverage. Ningbo offers perhaps the clearest example. When the city launched its public NatCat insurance program in 2014, the municipal government paid 38 million RMB in annual premiums to cover approximately 10 million residents, alongside a catastrophe fund and a separately financed risk reserve. In 2015, a series of disasters generated 78 million RMB in payouts—almost twice the government’s premium expenditure—demonstrating how predictable ex-ante spending could mobilize substantially greater compensation after a disaster.

However, the program also exposed the structural limits of local experimentation. The PICC-led insurance consortium incurred an estimated RMB 56 million underwriting loss that year, and a subsequent joint analysis described the program as having faced a risk of insolvency. Although the scheme continued, fiscal concerns about long-term ex-ante commitment reportedly led local officials to reject an informal proposal to extend a later contract from three to five years. Coverage has also remained limited by the premiums the municipal government can afford. Beyond Ningbo, numerous localities have implemented their own regional NatCat insurance pilot or replicated other province’s models. However, the vulnerability of regional experimentation remains, as the fiscal benefits of insurance can quickly reach their limits when severe regional disasters simultaneously increase claims, strain insurers, and weaken the government’s ability to finance future coverage.

These limitations point to the need for a nationally coordinated system. One option would be to establish a national catastrophe-risk facility that pools exposure across regions and provides a common layer of reinsurance above local schemes. Local governments could continue purchasing basic catastrophe coverage from commercial insurers, but with premium support from a dedicated central budget allocation, calibrated to both local disaster exposure and fiscal capacity. This would allow poorer, high-risk regions to participate without shifting the entire cost to the central government. Insurers could in turn transfer part of their exposure to the national pool, which could accumulate reserves over time and access broader reinsurance and capital markets for exceptionally large losses. Given limited public awareness and weak demand for voluntary catastrophe insurance, government-purchased coverage could provide a basic floor of protection, while households and businesses remain free to purchase additional commercial coverage. Such a framework would preserve local responsibility while spreading risks and financing capacity well beyond the jurisdiction where a disaster occurs.

Such a nationally coordinated model can find institutional precedence in China’s agricultural insurance system. China’s agricultural insurance is the largest in the world and operates through strong policy-based subsidies where around 80% of the total premium is subsidized by central and local governments. At the same time, a national agricultural reinsurance mechanism helps diversify risks and absorb catastrophic losses. Although agricultural and catastrophe risks differ in their frequency, concentration, and insurability, this experience demonstrates China’s ability to build a layered insurance system combining central support with differentiated local implementation. Moreover, it highlights the importance of a strong policy anchor in advancing insurance schemes, as agricultural insurance is tied to food security and rural revitalization which occupy the front and center of China’s national policy. By contrast, comprehensive catastrophe insurance, despite being promoted in the National Climate Change Adaptation Strategy 2035, remains largely a collection of local schemes and is still treated as a supplement to disaster relief rather than an integral component of national fiscal-risk management.

Granted, China’s NatCat insurance still faces fundamental challenges such as the lack of centralized and publicly available insurance loss data that are critical for insurance modelling and policy design. Moreover, the constant position changes of local cadres reduces the incentives for local governments to commit to insurance programs whose benefits might not be reflected during their tenure. However, these features further demonstrate the importance of central engagement. As climate risks intensify and policymakers increasingly recognize insurance as a tool of financial governance rather than merely post-disaster compensation, a stronger central role deserves consideration.

A nationally coordinated system would not eliminate public liabilities or difficult trade-offs. Central support would still involve transfers between regions, correlated disasters could overwhelm the insurance pool, and poorly designed coverage might weaken local incentives for prevention. These risks call for differentiated coverage, targeted assistance for fiscally constrained regions, and continued local responsibility for risk reduction. The objective is not to make disasters inherently less costly, but to shift part of the fiscal burden from discretionary post-disaster transfers toward predictable arrangements established in advance, while mobilizing insurers’ and the capital market’s risk-bearing capacity. Elevating catastrophe insurance from fragmented local experimentation to national fiscal-risk management would allow public finance to remain the foundation of resilience without bearing every disaster alone.