Our Thailand explorations have taken us throughout Kanchanaburi, Phuket, and Bangkok thus far! We have been extremely fortunate to meet with various people from different organizations and initiatives, from the BAAC to the Srinagarind Dam to KU Veterinary Hospital. Recently, we visited an organization that I found particularly interesting, the Hydro and Agro Informatics Institute, otherwise known as the HAII. As a statistics major, I really enjoyed hearing Mr. Surajate Boonya-aroonet’s presentation about how they employ data science and modeling to became more educated about natural hazards, particularly flooding.
The HAII is a governmental organization that collaborates and shares environmental data with 36 other agencies, in addition to local groups. They use data science technologies to face the various types of flooding Thailand deals with, from the most common type, river floods, to the most dramatic form in terms of water per hour, flash flooding. The HAII appeared to be a very collaborative effort among both field experts and data scientists to collect, clean, integrate, store, and analyze massive amounts of data, which are then used to build models for events like storm surges and river flooding. The data collection and implementation of solutions must be completed on a local level, so trust between the HAII and local communities is crucial. For example, the HAII worked throughout the country to establish 935 water level telemetry stations to gather data, and they must maintain contact with regional authorities to ensure the security of these stations. We also learned in the presentation’s info-graphic that the rainfall in 2017 was almost as high as in 2011, but the impact was much less severe as Thailand had learned from its previous experience. For example, the 2011 flooding led to the creation of the National Hydro-informatics and Climate Data Center, compiling more data to enhance education and become better prepared for the next disaster.
As an actuarial student, I could not help but think about how valuable this type of data would be to insurance companies trying to price insurance premiums. Flood insurance is particularly interesting because flood risk is difficult to model and thus price premiums. It is tricky to insure against an event like flooding (versus a tornado or house fire, for instance) because it has catastrophic potential to a geographically concentrated area. Therefore, many countries throughout the world have adopted various strategies in an attempt to address catastrophic risk, like reinsurance, catastrophe bonds, and government-sponsored insurance. For example, after being slammed by an earthquake and tsunami in 2010, Chile made adjustments in terms of catastrophic preparation and insurance strategies that served the country well when another tsunami struck in 2014. They enforced building regulations, developed up-to-date risk maps to ensure fair premiums, and pushed for mandatory flood insurance to improve risk pooling.
According to 100 Resilient Cities and reports from insurance companies, the 2011 Bangkok flood caused an estimated $45 billion in damages, of which only $10 billion was insured. As a developing country, Thailand is not at the same stage as countries like the U.S., Chile, and Japan in regards to flood insurance of private properties. Flood insurance in Thailand exists primarily on a commercial level; many global firms purchase flood insurance for their plants in Thailand, but personal homes and small businesses are not insured. After the 2011 flood, the Thai government offered insurance through a National Catastrophe Insurance Fund at a time when private insurance companies could not, but this program does not seem to be largely successful. However, with accurate flood risk maps and continued government support, it may be possible to make flood insurance more affordable and even give private insurance companies the confidence to invest in this market in the future. I think that the risk maps and catastrophe modeling that the HAII is involved with could result in more actuarially fair premiums and thus a wider interest in purchasing flood insurance on private property. Whether the public or private sector takes the lead on this remains to be seen, as consulting companies such as Aon and Guy Carpenter have been building and improving upon flood models for Thailand since the 2011 flood as well. When it comes to pricing insurance, data is absolutely essential, and the work of the HAII does seem to lend itself to offering flood protection to citizens in its future endeavors.
