While agricultural insurance has existed in China for over 60 years, it is only since 2007 when the government started to subsidise and standardise coverage, that the industry saw a rapid growth from USD 100 million premium (2006) to USD 7.1 billion (2017). Within a short time, China has become the second largest agricultural insurance market after the USA with the largest livestock and forestry insurance portfolios. Over 25 insurers are authorised to provide government-subsidised agricultural insurance and include national general insurers (e.g., PCIC, Ping An, China United) and regional insurers that focus on one or several (e.g., Anhua, Goyuan, Sunlight).

Anhua was among the first insurers to use outputs of crop models to assess possible maximum losses to its crop insurance portfolio and to deploy drones to support damage assessment for field crops. Leading Asia Risk Centre, a provider of agricultural risk models that is now integrated into Risk Management Solutions (RMS), I was invited to speak at the 2nd International Agricultural Insurance Conference (2013) organised by Anhua (picture). My presentation was on extreme events in agriculture in China and how catastrophe risk models can be used for risk management, reinsurance optimisation and solvency reporting in agriculture. The presentation later led to an interview with Insurance Times

The conference took place in Baotou (Inner Mongolia) and included an excursion to the grassland areas and sand dunes as well as a demonstration of how Anhua uses drones for loss assessment of field crops. I was surprised how accurately areal photography can identify damage ratios of drought impact on corn and supports ground-based loss adjustment. I remember sitting in front of large screens in Anhua’s office that showed high-resolution images from the drones and geo-coded farm boundaries. In the meantime, most Chinese insurers are using unmanned aerial vehicles to support loss adjustments and furthermore, crops drones are tested in other countries to support damage assessment.