Welcome to Artemis’ fourth ILS executive roundtable in Bermuda, in which our participants discussed a variety of opportunities and challenges that the sector is dealing with as it responds to the losses of a second active year for global catastrophes.
It’s a challenging time in the ILS market at the moment, with more loss creep coming through and a lack of fresh capital. So what can the market do to ensure its access to risk stays open, how is Bermuda helping that, and when do we feel things are going to kick off again in the cat bond market?
Adderley: You talk about the capital markets and when they might jump in again. For years and years we’ve all heard this thing about excessive losses – the big events – how big does the event have to be before the market changes? And everyone assumed that if we have a USD150 billion or USD200 billion loss, with interest rates going up, then we have this great change in the marketplace and rates would go up. Now, for three years in a row we’ve had these losses. And to me it’s been compounded and compounded. But every year we’ve kept on saying: “When’s the bubble going to burst? When’s the market going to change? At what point are prices going to go back up?” And if I’m the investor now, I’m thinking: “Well, do I believe in the market? Is it going to change? And what’s going to make it change?” So surely I’m going to wait until June 1 renewals to see what happens, see if they change a lot.
Johansmeyer: Well the problem is if you’re waiting for the norm to change – and I’ve been saying this since they started calling $200 billion events market changing – if you’re waiting for the norm to change you’re just going to grow old and frustrated. This is the market. I once heard it described by an old economist, he said the reinsurance market is a crappy market every year. Occasionally there is an event, you have a good year of rate increases afterwards, and then it over time goes back to normal again. And that’s the reality. We’ve got two years of big losses right now across several parts of the U.S, the Caribbean, and Japan. And we might see a little bump in Japan as we did after 1/4, Florida might clean itself up a little bit thanks to AOB as of the 1/6, but other than that it’s going to go back to room temperature again. I think what you need is more original risk coming into the market. That’s why everyone is talking about cyber right now. Better access to industry loss index triggered risk in Japan is going to make a massive difference. I think we’re going to see a little hiccup in the worldwide index market after everything we saw with informal loss triggers on Jebi. That turned out to trap a lot of collateral, as we’re seeing, and for no good reason. So did that make people think twice before trading those same triggers on worldwide? Yeah absolutely.
Bardon: And funds need to deploy. So after 2017 there was all this hoopla that there were going to be large rate increases at 1/1. There were increases and we were able to deploy within our expected range, but certainly not at the high, and we considered that in determining the amount of capital to accept. Then 2018 was an active catastrophe year as well, meaning that investors that had invested after losing money in 2017 lost again. That’s why I think that there haven’t been as many capital raises this year, because investors are more hesitant than they were before. Many funds promised much larger rate increases than what really materialized. We did have some rate increases on the property side for the last few months and given our fund structure, we’ve still been able to raise capital continuously through 2019.
Mello: I think a problem is also that the reinsurance companies aren’t being able to get the retrocessional cover that they need in terms of product and price, so they’re going to have to drive the inwards reinsurance rates.
Adderley: If everything we’re saying here makes sense then the rates should go up a lot. We just heard: AOB, less capital coming in, people are doubting rate increases, the renewals are coming up and I can’t buy retrocession protection. Therefore if I write the risk, I must really believe in the risk. It must be really good risk for me to write, because I can’t pass it on. So that means it must be priced properly because I’m not passing the risk on. Well, when June 1 renewals come along and you write that risk and the rates haven’t gone up, I lose my sympathy. And that’s what the interesting thing about it is. I mean look at the renewals last year. Investors are tired of this. We’re hoping for things to change and it hasn’t changed. But yet all we’ve heard for four years now is when there are losses, it’s all going to happen and we’re going to get that ice cream sundae now. And it hasn’t happened.
Watson-Brown: It hasn’t shaken off everybody though. There is a renewed interest from asset managers. That’s the big side that we play in and what we see is those initial conversations with U.S-based asset managers looking to enter this sector. There’s volume coming. You will hear aspirational AUM targets versus where they land, but we are seeing that billion dollar plus commitment, deploying the capital over two to three years, to get into it the sector. So, to Tom’s point about where the market is and where the pricing is, I think they’re seeing it as “yep, this is where we are and we can deploy a strategy that’s going to deliver the risk/return profile that our investors want.”
DeSilva: And when existing ILS investors experience losses it really helps to educate and shape their risk appetite. As new investors enter into the market and start participating in diversified risk types, consistent delivery of data is going to be necessary to give investors the comfort they need to keep participating. As a result, I think the existing and emerging markets that need coverage are the ones that are more likely to find capacity.
Taylor: I think I agree with that, I’ve definitely seen where some investors are testing out smaller deals to build up their confidence in the asset class, and as that confidence grows, they are committing to larger deals.
Ching: That might be true with some of the newer asset classes, but when it comes to cat, which has been very much on the radar of a lot of the investment world for the last 10, 15 years, I think another year of above average loss activity driving negative returns would be tough for them to bear in its current form. Investors will be looking for a market-wide response via open demonstration of further discipline and great efficiencies to gain enhanced returns in the future.
Taylor: The ILS market has clearly arrived at an inflection point due to loss creep and trapped capital over the past few years ; investors are working through issues that developed from the reality of having to deal with recent losses, however, the right conversations are taking place around better reporting of losses, more timely and accurate valuations and the development of a secondary market – this should not be seen as a negative, as the conversations taking place now will strengthen the assets class and as the market moves towards the second half of the year and the upcoming renewals, it should return to decent growth trajectory. To put the recent losses into perspective, one could say that losses become more inevitable as the number of deals in the market grow, for instance, if 10 years ago you had 50 deals in the market and you now have 500 deals in the market, the chances of losses hitting an ILS deal is multiplied by 10 times.
Johansmeyer: What kills me is we had this conversation too after Fort McMurray in Canada. Wildfire can be big. This turned into one of those big cats that nobody expected and now McMurray is so small compared to the most recent ones nobody’s talking about it anymore. I’ll talk until I’m blue in the face about the hailstorms in Istanbul in Summer 2017, because that’s a market that only knows quake, flood and terror. But there’s been hail in Istanbul, wildfire in Canada, and massive wildfire in California. We need to start thinking about $3 billion hail events in the U.S. It’s not every hailstorm but to see one or two of those a year, it’s not so unusual anymore.
Mello: But most people are using catastrophe models for pricing and there is a lot of uncertainty in the models. For instance, the California wildfire models in the market have much uncertainty and this makes it difficult to determine an appropriate rate to charge for this risk. You also need to look at the experience that companies have for the risk and incorporate that into your view of risk.
Ching: Only reasonably straightforward wind losses correlate well with the model as you track back over time. Each Cat loss event in recent memory has a twist or a turn inbuilt which has led to various levels of surprise. I think the last loss that ‘behaved itself’ was Hurricane Charlie – what’s that, 13 years ago?
DeSilva: Most companies are creating post model-modeling tools that help account for vendor model uncertainty. So they’ll get the model data, they’ll apply their own modelling on top of that, and this allows for better risk selection. I think these models assist with consistency and clearer understanding of the risk, but in my opinion it doesn’t replace the pen in the understanding process.
Bardon: Well there’s no transparency around it. How can investors trust your modelling when they have suffered from deterioration on the prior years? If that deterioration is material then there is clearly something wrong either with your valuation process or your modelling, and there’s no transparency. We invest an enormous amount of energy in seeking to publish reliable marks.
Adderley: Why does everyone in the insurance industry want losses? Tell me when losses occurred and you’ve actually made money back for two years now. That means looking at the valuation point. So that means that people must be valuing highly, and saying I have to deploy it, and then suddenly coming in much lower, because they have no choice.
Johansmeyer: The loss thinking is old thinking. You look at the rate on line charts over the past 30 years and we’ve all seen cone, the convergence, right? The post-event price swings after Andrew, then the 9/11 post-event swing got smaller, then Katrina, and so on. So the notion that you get a nice big bump afterward with all this capacity out there, I think those days are gone. I mean the two things we tend to look for as an industry for prices to go up are post-event price increases – you’re right, that’s not happening any more – and expense reduction – which there’s still some room to do, but I’m not an actuary and even I know that you can’t cut past zero. So your growth opportunities by cutting are somewhat limited, this means going out and doing the hard work – bringing in new risk.
Bardon: Yeah, there has to be an increase of demand for insurance or there has to be a decrease of supply, and there’s been a flood of supply. So you can find some niche opportunities where there’s not enough supply or you can increase the demand. That’s where you’re going to see the price increases . You’re going to see it in risks like wildfire. In Puerto Rico there are also small opportunities to get large rates, or there are other markets where the supply is pulling out, like non-standard auto.
Watson-Brown: Do you think the ILS market has the capacity to branch out into new risks and that sort of stuff? Or does it rely on the larger reinsurers to do the R&D, to breach that first?
Bardon: Yes, we do cyber and personal auto and health. And those are just the collateralised re transactions more focused on the non-cat ILS side, but we also have private equity and debt investments across the entire insurance spectrum. And one area that we’ve found very interesting is distribution finance, so we’re investing in agents and brokers as well.
Adderley: When you’re talking about different fund managers or investment managers it’s important to think about who has access to risk that no one else has access to. There’s so many of my clients that come to me as investors who say: “I’m choosing a manager now because of the fees.” Well, buy the cheapest TV and see how long it lasts. Surely it’s got to come down to distribution, to who has the access to more interesting risk.
DeSilva: I think there also needs to be increased transparency in the underlying risk for investors. Investors also need risk analytics tools so they can ask the right questions. A transformation is occurring, and new lines of business will be introduced to the ILS market. So I think that existing lines will need to be packaged and distributed differently then offered to the ILS market.
And what do people think about harmonised practices? Does that erode alpha for managers or is valuation potentially a source of alpha?
Bardon: I think it can give investors some confidence. If you’re not showing deterioration on prior events, if you’re actually marking to a true best estimate, then that gives investors confidence in you. And investors will look at your past results and how those have moved over time, and that’s how they get comfort in your abilities. Not just your actual results but also how your results evolve.
Watson-Brown: They’re starting to increase inquiries on the diligence side as well now from institutional investors. And while there has been a shift in their skill sets in being able to come out and do DD on fund managers and providers in the ILS space, there is still room for improvement both on the due diligence process and managers’ approach to valuation. With valuation of these portfolios, you’ve got to get it right. Because once you lose that investor confidence in your ability to manage the portfolio following loss events, I think it’s very tough to get back. Investors get frustrated with creating new side pockets for continuing deterioration, have fees crystalize or how collateral is managed and the impact on the open share classes in a fund. These operational matters can have a greater negative impact that recognizing a decline in NAV.
Ching: It could be a case of it being a lot of ‘huff and puff’ for a relatively small part of your portfolio as well – whether it’s your legal team or valuation department being burdened – it could lead to frustration if you’ve got to continually keep going back to this smaller asset class that you’ve got within your portfolio. Whilst at this stage you might not be closing down your interest you’re not necessarily going to be growing it until you gain more confidence in it.
Watson-Brown: Does there need to be more a more active market or does there need to be a way of creating a more active market and offloading some of that risk? So once you’ve been hit, how easy is it in the industry to commute contracts for exposures, take what you can and get that capital redeployed?
Adderley: There needs to be more liquidity, more exchanges. It is too difficult to trade the note, the cat bond or the preference share and get rid of it because I want to do something else. I’ll sell it, get out of it before the storm hits the investor and so on.
Bardon: We looked at that last year though, and we tried to be more active in offering to buy LPs out of other funds at a certain price. But those we approached just wanted to sit and hold.
Watson-Brown: I think we found it the same. We had inquiries from vulture funds and they thought this would be a good opportunity to get in; normally they wouldn’t look at these cat books due to the short-tail. But again, firms need to be willing to open their books to allow diligence on the contracts; not many were interested in undertaking the process; or when some early price discussion would take place, the response was a quick ‘no’.
DeSilva: And the other thing is that decisions are often being made with outdated information, because NAV calculation and reporting to investors is taking such a long time now due to inefficiencies. In our view, shortening the window of time to calculate the NAVs will allow investors better planning and participation on risks being offered by sponsors.
Taylor: I’ve been involved in quite a few conversations about secondary markets lately, and at the moment there is no strong solution, however, we are working to get there in due course.
Johansmeyer: Yeah, and the commutation discussion comes in when liquidity stops. So let’s say you’ve got an instrument where there’s no additional information flowing and there would be a stimulus to trade – it looks like it’s blown. So do you sit on the loss or do you try to commute out and watch it deteriorate? I don’t think liquidity is a solution for the commutation discussion. I think liquidity is necessary, there’s a role for it, and as a market we’ve been desperate for that for decades, and over the past five years we’ve really seen movement toward it. At the same time, you’ve got to look at an instrument and think is it time to try to get out of this to get capital deployed? Or should I wait it out and hopefully not take the loss? And the good news is that in one of those two choices you’ll be a genius and in one of them you’ll be an idiot, it just depends on where the loss ultimately winds up.
Mello: Except that it comes down to the reinsured deciding on whether they want to release those assets back to the reinsurer. The buffer loss factor tables can allow upwards of 2 or 3 years to hold collateral. If the reinsured is very conservative in their ultimate loss values, they may hold the collateral until they are confident there will be no other shifts in that value. This affects the ILS markets that may be relying on capital being released.
Rachel, you mentioned buying positions off other funds potentially. Do you think if those options were available we’d be forced to see more transparency coming through, so it can stimulate better valuation practices?
Bardon: We found that most people weren’t interested. They thought it was an interesting concept, for us to come in and either offer to purchase out their LPs or some other format of essentially what was dead cat. But I think they just wanted to sit and hold. Obviously the other funds had more insight into what their true valuation was, as opposed to us trying to estimate it ourselves. But yeah if there were more transparency around it I think it could create a better market. Because then your views wouldn’t be so divergent on what the potential results are.
Ching: We were involved in some of those discussions about selling positions, and it didn’t even come down to what’s the price. It was just there was seemingly no interest at that particular time. It’s not to say it might be of interest in the future.
Bardon: I think it’s just it was such a small amount for some of these investors that it wasn’t worth their time.
So shall we move on from pricing, and maybe talk a little bit about regulation and structures and what might be coming next? It’s one way you bring new risk into the market, by having new structures and new ways to deliver that risk to investors. Brad, are you seeing anything in this area?
Adderley: Well, obviously the BMA came out last year with this idea of the sandbox. And that’s this idea of creating a fledgling, innovative structure which perhaps has not been done before. You sit in the sandbox for a short period of time and at some point you mature enough to become a properly licenced insurance company. So I think what’s interesting about the concept of course is that it’s a bit of a black box. Because the thought is something’s new, so how do you regulate it? What do you give them? Is the application process more difficult or easier? So if it’s more difficult, why go through the sandbox? But how can it not be more difficult, because it’s doing something new. I also think you are going to see in Bermuda more classifications of insurance companies and updated guidance notes – no question. I think the BMA in its business plan or annual report talked about new classifications. So there’s definitely thought about what’s going on. And the question is simply: Will they have a new classification for the collateralised marketplace to do commercial reinsurance? Because obviously special purpose insurers are meant to be a one-off vehicle. And a lot of people treat them as basically commercial reinsurers on a collateralised basis. And the question is: Is that the right view to have of them? But saying that, I think the largest ILS players are all moving towards a rated re vehicle. So that means they’re basically saying “I don’t want to be an SPI anymore. I do want to have more regulation and more capital requirements and become a Class 3A.” And then they’re scared of ratings at the same time. And then potentially in the future the company collateralises risk not using trust accounts or LoCs, because they write the business off their balance sheet and save on administration costs. And then potentially retain risk on a normal basis.
Johansmeyer: It gives you more flexibility for deploying capacity, because everyone is awash in capacity and needs to find more ways to do it. Which means the ILS sector is moving to the past to find more ways to deploy. But if it’s forcing you to move away from your original competitive advantage, then what are you trying to accomplish? This is again where I see the solution isn’t to find more ways to write Florida or wildfire or Japanese quake. The solution really is to go out and find those really new classes of business that are going to come in and change the market. The problem is getting in front of the buyers and showing them that buying this protection is a good idea.
DeSilva: What is the BMA doing that’s going to differentiate the Bermuda market? Other jurisdictions like Hong Kong and Singapore are gaining traction into the ILS space but are they just playing catch-up? Are they currently re-creating Bermuda’s model at a time when Bermuda is looking to evolved from the existing model? In Guernsey ILS funds do not require the full collateral within the first 30 days. What is Bermuda’s position on this and does that 30 day allowance propose a huge risk if losses occur during that time?
Adderley: Well Singapore is interesting. They have been successful, or have tried to grow their business, partly because there’s access to the Asian market but partly because they cover a large portion of the transaction costs. But the view has been that Singapore regulations have not been conducive to cat bond listing. With Guernsey, let’s take this idea of the 30-day window funding. We all know when it comes to renewals it is very hard to get a trust account in place, it’s the reality. And then you need to open up the trust account. The contract may not even be signed yet or drafted. So that means that coming into January 1 there’s risk, no contract, no collateral, no trust account. So the window is a great idea, except now if I’m an onshore cedant or an onshore regulator and know there is a contract in place which says: “I don’t need to be funded for 30 days,” am I going to give you reinsurance credit now for taking that contract and running forward and not get funded for 30 days? Well my job is to see both sides. So then I say “is it a really good idea to do business when there is no reinsurance contract. And what happens if you lose money in the first five days? Are you really going to fund?” I have an issue with the platform when you’ve got someone who’s promised to give you the money and your name is on the line and it’s not collateralised. So people think of Guernsey and go “this is revolutionary” and the people who say it to me haven’t actually thought of the whole circle and what all the issues are. And they’re missing the point that there is risk to the marketplace, the risk to the ILS marketplace, the risk to the reputation of Guernsey, is worse in my mind than actually trying to be that competitive to be 30 days.
Ching: As far as regulation though and deal certainty, I think it’s a market wide concern and not just ILS. The acid test would be a 02:00, 1st January earthquake scenario – a time when contracts are still largely uncertain. It would be quite interesting to have a regulator – whether that be in the UK or in Bermuda – to put that scenario out there and build framework to ensure trades are concluded in a timely fashion. Due to the fact trading often goes up to the last minute, trust agreements are sometimes not in place until February or March on a January placement.
Mello: Each reinsurance contract is required to go through a negotiation between both parties that can cause delays in the signing of the contracts. For ILS markets that are collateralized and not rated, you may also have a fronting party. This contract may also cause delays as it too is required to be agreed by parties. So, there are timing inefficiencies in the process that need to be considered.
Adderley: It’s also when you date the contracts. There’s a big difference to me if you date a contract for February 1st which is effective from January 1st, and if you date the contract for January 1st and you sign it February 1st. All I care about if I’m advising you is that you fully fund/collateralise your liability – even if your liability is not documented yet – and that if the BMA came and did an on-site visit you can defend yourself.
Mello: The process may become more efficient if contracts are designed and implemented quickly in the process. This may reduce the amount of time required for contracts negotiation following the agreement of the deal.
So is this another reason why people are moving to 3As?
Adderley: I think that people are moving to 3As for lots of reasons. However, they don’t realise there are complexities that go into a 3A which means they have to make sure they’re acting like a proper, mature insurance company. And some people think “oh I’m just going to upsize, it’s easy.” But in real life there’s a big difference between an SPI and a 3A. From a regulatory point of view, to structures, to capital reduction, to filings, to public disclosure. But if you’re going to get a rating then you don’t care, because you’re going to comply with the rating requirements which are similar to a Class 3A anyway.
Johansmeyer: Well it all comes down to the supply of capital that defines the mechanism by which capital gets deployed. If you’ve got so much capital out there that you need to deploy as quickly and as vigilantly as possible, then you’re going to choose the path to get you there. Cat bonds aren’t remotely efficient, but they’re not intended to be. They’re intended to protect against the downside in every conceivable way, and I’m stunned during the cat bond process by how many documents just come through the work group. But that’s what’s required up for a truly buttoned-up process.
Let’s look forwards a bit and discuss the upcoming and end of year renewals. What does everyone think the shape of the ILS market will look like at that point in time? Or is it all down to whether we have another hurricane?
Johansmeyer: We need a year off. Let’s get through 2017 and 2018 first.
Bardon: Yeah, there are still deteriorations happening within the last two years. Many valuations teams have been very busy lately.
Adderley: Well the needle hasn’t moved yet, so if you really believe pricing is going to change you need to have another major cat event.
Bardon: And it has to be a shocking event. You can’t just have a large hurricane. It needs to be multiple large hurricanes that shock the industry.
And what about the timing of renewals? Do you think companies that are buying coverage off-cycle are going to have a strategic advantage?
Johansmeyer: Now this is interesting, where you’ve got timing of purchase becoming a strategic differentiator. Because at the end of the day you’re looking at basically what I’m paying for protection versus what I’m going to sell of protection. And this turns into a profound advantage and can disjoint the market. I don’t see that as problematic, I see that as capitalism just working the way it should.
DeSilva: ILS is also ripe for off-cycle and shorter term contracts. There is an opportunity to really change the dynamics of contract term lengths and create more stimulation by having contracts renew more frequently.
Johansmeyer: The problem is buyers just don’t want to pay for all those risks in one place. So investors might want something inherently diversified and the buyers will say “oh absolutely, you can throw cyber, terror and marine in there, but I don’t want to pay for it.”
Bardon: Well that’s how we’ve had so much interest from our investor side. We had a good year last year compared to some of the other funds which were negative. We had a pretty decent year because of diversification across sectors. Cat is nearly a third of our main fund and only about two thirds of that has been regular-way cat. The rest has been corporate cat, on which we had strong returns. So our investors are allocating to us to make those allocation decisions across the different classes.
Ching: Continuing to build across diversified classes such as cyber, mortgage, life and insurtech and encouraging the infrastructure and regulation in Bermuda to manage these risks is, I believe, the winning formula.
Johansmeyer: And these things generally have low correlation, just like cat. And we’re seeing every day that cyber is less correlated than people realise. The notion that a big breach will bring a market to its knees? No. It hasn’t even brought a single company to its knees. I mean Equifax bounced back pretty fast. Merck, not a big deal. So we’re seeing cyber is low-correlation.
Bardon: A lot of the cat contracts have some exposure to cyber risk, so a pure cat contract could include some silent cyber.
Johansmeyer: And sometimes it’s not even all that silent, as we saw with NotPetya. We’re watching LockerGoga unfold right now, which is turning into possible an affirmative cyber cat. It’s really interesting too. You mentioned silent, well NotPetya was 85% silent, and with LockerGoga we’ve already got affirmatives across three of the companies disclosed as having been affected. So you could actually see this as an affirmative cyber cat, which would mean something totally different from NotPetya, which was massive and unexpected. Now we’re seeing something that’s massive and presumably expected. And the implications for managing your capital in the wake of that are going to be significant, especially given that they’re hitting heavy industrials.
Taylor: And what’s more, I would say that the BMA’s regulatory sandbox and its newly proposed fully collateralized reinsurer class can combine with Bermuda’s strength in the ILS market to make it an excellent test bed for an ILS cyber deal; certainly the interest in a cyber ILS deal has existed for some time and Bermuda is creating a perfect environment for this type of transaction.