When Fonden—Mexico’s natural disaster insurance fund—was created in 1996, initially it was born as a budget account only. This was not the best idea, because in the end the money all ultimately returned to the treasury, and funds could not be retained the following year to capitalize the fund. That’s why Fonden changed in 1999 to become a trust, which works much better.
For us, the biggest challenge at the start was collecting the data to feed into the risk model we would use, because Fonden didn’t have any information about public infrastructure, schools, hospitals, roads, bridges, and so on. It was not an expensive endeavor compared to the cost of reconstruction after disasters, though, and surprisingly we found that we already had a lot of data housed with our federal entities.
The best example was in schools. The Ministry of Education had developed over two years a big project to obtain all key information about schools, including geographical location, construction characteristics, how many students, etc. We had had a similar experience with other ministries, which was hugely helpful, so in the end we spent just two years on this project and it cost us only about $2-2.5 million.
Asian countries looking to prepare for a risk management strategy similar to that of Mexico should obtain information on roads, bridges, schools, hospitals, and hydraulic infrastructure. One particular area to pay attention to is mapping the homes of the poor, because these people are most vulnerable after a catastrophe.
In 2010, Fonden created its own risk model called R-Fonden (R stands for “riesgo,” risk in Spanish). The model determines and measures the disaster risk in order to establish the different levels of retention and transfer. It allowed us to figure out what kind of instruments we could use to offset the risks. One of these was the catastrophe insurance program established a year later, and the other catastrophe bonds, which provide liquidity for emergency losses rather than for reconstruction.
Alongside catastrophe bonds based on a single natural hazard, we have also done multi-catastrophe bonds, which provide a crucial benefit to investors in the from of diversification. Mexico is exposed to hurricanes coming from the Atlantic as well as the Pacific oceans, and there is no correlation between the two. The risk is different for earthquakes given the country’s particular location in between three major tectonic plates: North America, Pacific, and Cocos. The risk is a very specifically located one.
A multi-catastrophe bond may well be useful for Asia, which is threatened by a variety of risks including volcano eruptions, tsunamis, cyclones, droughts, and floods.
We are currently looking to extend coverage through multi-catastrophe bonds. We are consistently refining what we do, not least following Hurricane Odile in 2014. In the preliminary report, the National Hurricane Center (NHC) determined that the level of pressure would trigger the bond. However, when the NHC provided the last and final report, it found that was not the case, so we are looking to see if we need changes in terms of the different levels of trigger. This is a more complex scheme, of course, but one our main goals is to try to better calibrate the trigger points for the next occasion.
As well as emergency risk, Fonden also promotes prevention, so if a natural disaster affects infrastructure we can provide extra money to rebuild better. It’s better to have prevention as part of the overall fund so that the many entities involved in infrastructure, such as local governments and many others, don’t plan different independent projects. We need to connect all the activities on prevention and mitigation, given that risk financing is, in the end, the overall goal.
For Asia, or others looking to do what Fonden has done, information is key. We need to know the risks that could affect infrastructure, and Asian governments should make the best possible use of available data to get the process started. The important part is obtaining the information, you can always refine the process later on.