The challenge of pricing cat bonds

By Fabio Salim

Cat bonds can be a great option to investors and issuers as a concept. However, they might not be such a great idea due some challenges related to the implementation and pricing of such assets.

As discussed in the case, the way the cat bonds are designed are great for investors due correlation and diversification benefits. To the issuers, it allows to transfer part of the risks of an infrequent, but massive loss to a third party that currently does not play a role in the system. This benefits ultimately benefit the pricing to policy holders. As discussed in the case and in the classroom, these benefits, among others, include capital access, credit risk management and transparency

However, there are challenges in its implementation as an effective financial instrument that allows for efficient risk transfer. In this particular case, the problems go beyond the usual drawbacks of new types of instruments, such as uncertainty about how they will be regulated, market with limited liquidity, costly securitization/underwriting, and lack of information about the actual mechanics of the asset.

What is particular challenging in this case to make cat bond an effective financial instrument is how to price it properly. In a very simple way, a financial instrument allows risk transfer from an issuer to an investor.

What is important here is to have a price that sets an appropriate Sharpe ratio (units of excess return / units of risk) for the future performance of the asset, compared to the opportunity cost to investors. Risk here is estimated through past price volatility, usually a somewhat reliable indicator of future variability. However, the nature of cat bonds makes this risk assessment harder. And if the price cannot be set correctly, cat bonds may not be effective financial assets in terms of risk transfer (to issuers) and return generation (to investors).

Below are the three main features that make pricing cat bonds a difficult task:

• Catastrophic events are infrequent and represent big losses. Prediction the occurrence of such “fat tail” events is hard because it is necessary to take very long time series (given the low frequency), and the actual losses and price variations need to be estimated without a great level of confidence. In the case it is mentioned that this data needed to be estimated using a series of assumptions.

• Given the low frequency of the events, as thus or lack of experience in dealing with them, estimating the actual losses of an event might be also very imprecise, and it is hard to make extrapolations using benchmarks, given the uniqueness of the location and features (i.e. earthquake scale, wind speed, and so on) of each event. The case also mention the uncertainty involved in this assessment.

• Predicting future probability and impact of natural catastrophes in a context of climate change (including the magnitude and timing of such changes) is challenging given the lack of credible projection on how the world will actually change.

Even though there might still be a market for this instrument, its functioning might be limited by the correct assessment of price, what can cause a mismatch in terms of risk expectations from investors and issuers, and cause the price of the instrument to be at a level that is simply not a good deal for the issuer (given the “extra” discount that investors will include). Or they can create additional financial systematic risk that mispriced assets carry in fat tail events (big losses), such as was the case with asset backed securities during the 2007 financial crises.


About macomberjohnd

HBS Finance faculty interested in sustainability in the built environment including devices, structures, townships, and cities.

One Response to “The challenge of pricing cat bonds”

  1. To your first concern around pricing of risk, I could imagine incorporating data like sea surface temperature to help create real-time pricing of hurricane cat bonds. NOAA does a lot of data and modeling around this and may make predictions on if we’ll have a bad hurricane season or not based on several factors. Of course, it is then even more difficult to determine if there will be a storm making landfall in a densely populated area, but there’s also uncertainty in corporate management that affects stock and bond prices so I don’t see why this should be so different.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: