Roundtable: Achieving optimum diversification in ILS investing

Diversification, how important is it and what options does the ILS market offer? This roundtable, titled ‘Achieving optimum diversification: understanding liquidity, volatility and performance risks for uncorrelated yield’, discusses these questions and more.

This roundtable is taken from the recently published report by specialist financial services, pensions and investments publisher Clear Path Analysis, titled ‘Insurance-Linked Securities for Institutional Investors 2014.’ Artemis owner and editor Steve Evans chaired the roundtable and Clear Path Analysis have kindly allowed us to republish it here in full.

Roundtable participants:

– Adam Beatty, Business Development Director, Nephila Advisors.
– Elizabeth Garner, Head of Pensions and Investment, Save the Children.
– Jens Hagendorff, Professor of Finance, The University of Edinburgh.

Steve Evans: What liquidity, volatility and performance risks must a pension scheme understand before considering an insurance-linked securities (ILS) allocation?

Adam Beatty: Understanding volatility is vital. For an asset class such as this, pension schemes must be aware of the complete risk/return distribution; especially in relation to the downside risk. In catastrophe risk, the return distribution is not normally distributed around a peak mean but instead has a left tail bias. This essentially means that there will be some downside years in the tail of the distribution. However, for a large proportion of the time, not much happens with regards to losses with returns being above the long-term mean expected rates.

On the performance risk side, the key is to understand the market dynamics of capital supply, the demand for reinsurance risk and the pricing levels generated. Investors must ask managers whether that level of pricing, supply and demand is sufficient to generate the type of targeted return required but within the downside risk constraints of a particular ILS mandate.

There is some liquidity in catastrophe (CAT) bonds but generally speaking reinsurance risk is a fairly illiquid strategy. A portfolio that’s invested more broadly across different ILS instrument is going to be relatively illiquid; think 90 day redemption terms typically.

Elizabeth Garner: From a defined benefit (DB) pension scheme viewpoint liquidity is not so much of an issue, providing we can of course extract our money at some point. The volatility and how that transcribes onto performance risk, is more of an issue because you could potentially lose all of your money. It’s not so much about the way in which the bond’s liquidity might actually change but rather that your capital could be at risk whilst you are seeking a higher return.

Jens Hagendorff: I’ve spent quite a lot of time looking at whether a risk transfer is occurring and find, that investors who invest in CAT bonds actually assume catastrophe risk. I would also add model risk to the list; how reliable are the models used to determine the expectant losses as a result of a natural catastrophe?

ILS investors assume some catastrophe risk and whilst there are very established models out there for determining the amount of risk assumed, it is possible that these models could still underestimate the risk. This is most prominently seen in areas such as flooding and the like where the models are having to be revised again to reflect the reality that such catastrophes are occurring more frequently. As these models are revised there will be an awful lot of investors that have assumed more risk than they presently realise. So whilst currently a theoretical risk, going forwards, it could have a huge impact on the market.

Steve: As ILS instruments have evolved have you felt threatened by the emergence of any new risks?

Elizabeth: I don’t feel that the instruments have led to an emergence of new risks but rather that the issue remains to be around old risks. We have to consider the Lloyd’s Register type of risks that are now coming into the mainstream with plenty of pension funds taking on this risk. Then there are other risks such as global warming which of course will probably increase the catastrophe risk for all parties invested. I don’t feel threatened by it though but rather feel it’s a case of trying to understand what that risk is in the first place.

Adam: The evolution of the asset class and the instruments traded is not something that we feel threatened by; on the contrary we see it as an exciting part of the market. New capital that has come in from institutional investors has spurred innovation in the ILS marketplace. On the subject of new risks, we are increasingly seeing other managers diversifying their insurance exposures to incorporate more than just natural catastrophe risks. Some are getting involved in the marine, aviation and terrorism markets; although at Nephila we haven’t been pursuing such areas. Generally there is less need for incremental capital in those markets and arguably a lower expected return. This means that diversifying into those areas may not necessarily be in investors’ best interests.

Jens: It’s great that there have been a take-off of investments in this market. Though I do wonder whether the nature of the securitised risks are the same, or whether they are becoming riskier. There is presently a lot of indemnity based triggers in the CAT bond market whereas previously very few insurers actually issued CAT bonds with indemnity based triggers because institutional investors were wary of such arrangements. The slight worry I do have though is around increasingly risky events being securitised because it might lead to investors taking on more risk than they realise.

Steve: Do you feel that the introduction of broader terms and conditions as transactions made to resemble traditional reinsurance and the addition of modelled risk will increase this asset class’ volatility?

Jens: It’s pretty transparent as it is but there are certain types of indemnities and risks that previously few investors would have selected whereas now, there’s so much demand for CAT bonds that investors are less cautious. One of the really odd things about the CAT bond market was that until very recently the returns were very good, largely because insurance companies were very cautious in the types of risk that they securitised, and investors were getting a very good deal, but now this seems to have ceased.

Steve: Within catastrophe risk why should pension funds look to diversify the ILS instruments they use?

Adam: This is a very interesting question and can be approached in two ways.

If you asked a computer allocation model to analyse this sector it would say that since this asset class is very diversifying with very low correlation to other strategies and normally a small allocation for the pension scheme, then it actually makes sense to focus on the best paying risk in the asset class and accept a lot of volatility. The computer is likely to say, do not diversify much and stay concentrated on the best paying risk.

However, it’s fair to say that a number of investment committees and trustee boards won’t be entirely comfortable with that type of risk/return profile because ILS is still viewed as a relatively esoteric, alternative asset class. A number of investors want to instead take the uncorrelated return stream and seek to reduce the volatility. Perhaps by replicating some kind of fixed income type return but without too much downside risk. We see both approaches and have investors on our platform that are seeking to follow both of those different philosophies.

Elizabeth: Adam is absolutely right. As an investor you may want it to give you a higher return in the short-term but are concerned about a total loss for that bit of capital. Consequently it’s important to spread your risk and look for many small little investments within the bond that you’re holding in order to ensure that there is less risk of losing that capital. To assess this, particularly if we’re talking about lay trustees on a pension trustee board, you need the experts.

Jens: Both Elizabeth and Adam pointed out quite rightly that the key issue here is the risk of a total loss which we have with very few other types of investment. Therefore, diversification within the ILS asset class is pretty important. The trouble though with diversification in ILS securities is that a lot of the instruments have fairly similar risks: U.S. based risks and wind based risks. Within that asset class, certainly for CAT bonds, it’s not very easy to diversify. Some of my industry contacts tell me that there are some German primary insurers keen to issue CAT bond risk to flooding for instance, but it still seems to be very much in the pipeline.

Adam: As Jens correctly points out, the CAT bond market is very concentrated around U.S. hurricane risk and so any element of possible diversification has some inherent limitations. It is valuable to diversify across instrument types and to be able to transact across the broader reinsurance market. By doing this a manager is able to help create a much larger investable universe. The more positions you have to choose from the easier it becomes to control downside tail risk for a given level of return.

Steve: I come across a lot of investors now actively seeking non diversification, in that they are investing a small amount of their portfolios in, most predominately, U.S. wind risk. What do you think of that as a strategy and will it always be a small part of the market?

Adam: It’s a very valid strategy and mathematically is what portfolio theory tells you to do when investing a small allocation in an asset class inherently uncorrelated to the rest of the portfolio. Certainly many investors have that mindset but it’s not for all. There’s presently a battle with human nature whereby people prefer to spread their risks a little in order to control their downside tail risk; even though as a strategy the former is perfectly valid.

Jens: As a strategy I understand its validity. We have to bear in mind though that we’re looking for low levels of volatility linked to CAT bonds; this is why a lot of these mean variance portfolio optimisation approaches say it’s a good strategy. That said, what a lot of investors have to bear in mind though is that even with a low average volatility there are incidences, because of peak risk, where losses can be very large indeed.

Steve: Is there currently enough supply in the marketplace to account for different risk appetites, tactical and strategic investment objectives?

Elizabeth: It’s really an observation but a lot of people I talk to are now either in ILS or considering ILS. Going back 5 years ago this wasn’t the case. I don’t quite know what’s caused this change and if you are a sceptic then you might say it’s because the insurance market is trying to spread risk away from themselves. Whether that makes it a good product or not, I put the question out there as to what’s actually driving this change. Overall though I haven’t heard people report that they’ve been unable to buy bonds or that they consider the price to be too expensive.

Adam: The reason we’re seeing institutional investors becoming increasingly interested in this asset class is because of the fundamental attraction of an uncorrelated, investable asset class. In addition to the diversification there is also a positive expected return over time. These investors are bringing efficient capital to the market with a lower cost of capital than some of the incumbent reinsurance capital. This in itself is attracting reinsurance risk supply to managers like ourselves who are allocating this efficient capital on behalf of investors.

Coming back to the question, is there sufficient supply for different risk appetites? The answer is yes; products in the market range from mid-single digit target returns up to 25% or 30% target returns. This means there is a pretty broad risk and return framework available to investors and that the market is able to accommodate all of sorts of risk/return profiles. It’s probably fair to say, most of the institutional money that has come into the sector over the last 2 or 3 years is looking for 5% to 15% returns maximum which the market has been able to accommodate reasonably well to date.

Steve: In 2014, so far, the CAT bond market has seen transactions with on average premiums of between 2% and 4%, obviously a lot lower than recent years. Do you feel that investors will increasingly look to private transactions and traditional reinsurance contracts to boost their returns if the CAT bonds can’t provide the return?

Adam: It makes sense to look beyond the CAT bond market when constructing a portfolio in this sector anyway. A larger investable universe should mean a given return target can be reached with less downside tail risk. Investors will possibly start to look further at the market’s lower returning sectors and come to the conclusion that they will need to be invested in more than just CAT bonds to create
an attractive portfolio with the more remote risk instruments included.

Steve: Pension schemes are not involved in the high-level risk modelling behind ILS. Is this a reason to hold back from allocating or is it unrealistic to expect asset owners to have the detailed knowledge of a funds complex make-up?

Elizabeth: In my previous role at the Atkins Pension Plan, we were more concerned about being satisfied that there was sufficient diversification to make it an acceptable asset class for us to invest in. At Save the Children, there’s much more emphasis on being able to see what’s happening and with CAT bonds it is felt that there could be a conflict between people dealing with the different areas of catastrophe.

Jens: It’s important to bear in mind that the risk modelling for ILS products is done by external bodies and therefore, pension schemes don’t need to understand the detailed complexity behind it. On top of that I would argue that because of funds invested in ILS there is now an intermediary solution out there with an expert who understands a great deal more about the risk modelling and individual risks linked to those securities. I understand that the fees for those funds are pretty high, somewhat in the region of hedge funds. However, as a number of pension funds already invest in hedge funds, the fees shouldn’t be too much of a deterrent.

Adam: It hasn’t held investors back from allocating to this sector and one of the issues with the asset class is that there isn’t really an investable beta for investors to access. Pension schemes may decide it makes sense to use a manager for multiple reasons including for their expertise in risk evaluation and market access. In addition, they get value from the managers’ relationship with brokers, counterparties and broker dealers in order to access the risk.

We make a large amount of portfolio detail available to our investors to allow them to drill down and ask specific questions about portfolio strategy.

It’s also worth mentioning that some institutional investors do actually licence the models, or parts of them, so that they can compare certain CAT bonds or look at a CAT bond portfolio or to compare the risk/return curves being shown to them by multiple managers. There is certainly some sophistication and familiarity with the models amongst the investor base.

Steve: As the ILS market expands and broadens out to include more traditional reinsurance in private transactions, it seems to me that the skillset of the manager in selecting risk will become even more important. Do you feel we’ll see a move as people explore who the best managers are and will this exploration result in capital flow from one to another as investors choose their preferred managers? Secondly do you think managers are going to have to increase their skillsets and the size of their teams?

Elizabeth: They will have to prove to investors that there is skill and science behind what they’re doing. I’m sure that those running the asset classes will be nudged to do so as demand increases.

Adam: It’s absolutely necessary for managers to demonstrate that they are adequately resourced with the right skillset to execute the portfolios proposed to investors. Investors are interested in the depth of staffing and resources that managers have at their disposal and I would expect this to become more important as the asset class grows. There’s already fairly robust competition between managers but in terms of capital moving from one to another I’m not sure about that. Investors have a reasonably broad choice of managers in the sector and already ask the right sort of questions of those managers before they commit their capital.

Jens: ILS is just one of a number of new securities that a lot of asset managers are looking at. This increased interest is as a result of the current, relatively low equity returns and other fixed income securities. Generally skillsets are widening in the industry. I’m working with a few Edinburgh fund managers who are looking closely at areas such as infrastructure projects, but remain very aware that it’s not area where their primary expertise is. That said, it’s an area where they are willing to invest in those skillsets in order to keep up with the times. To conclude ILS is just one of a number of possibilities where investor skillsets will have to widen.

Steve: Thank you all.

Transcript end.

Read our other articles and transcripts taken from this report:

– Institutional investor appetite for insurance linked assets remains strong.

– Roundtable: What are the challenges of evolving insurance-linked securities structures?

– Roundtable: What is the future for insurance-linked securities?

The report from Clear Path Analysis is available to download today.

Visit the Clear Path Analysis website to register to download a full copy of the report ‘Insurance-Linked Securities for Institutional Investors 2014‘ including all of the interviews and roundtables.