18 February 2013News

The lie of the land: A roundtable review of 2012

Have events such as Hurricane Irene and Superstorm Sandy changed perceptions within the industry regarding US East Coast wind risk?

Marty Becker: I’m not convinced Superstorm Sandy changed industry perceptions, although as with all events the industry will learn from it. The models aren’t particularly good at storm surge, or business interruption exposure—and there’s been a lot of that. Probably, as virtually every company has said, it’s an earnings, rather than a balance sheet event. It is far from a market-changing event, but it is meaningful and we’ll learn from it.

Charles Cooper: I agree. Previously, when Northeast risk was talked about it was discussed in the context of peril, and there’s been a lot of debate on perils, whether you could have a category 3, 4 or 5 hurricane heading to the Northeast. That is a bit irrelevant now because a $25 billion event resulted from a relatively low-level wind event and so the peril is certainly interesting. The more important thing is the shift on the exposure side. The concentration of values on the Eastern seaboard makes it really unique. You also have aging infrastructure, much older building stock, less stringent building codes. When you consider the exposure presented by the Northeast it’s quite unique, and Sandy has opened people’s eyes to the concentration of values in the area.

Kean Driscoll: From our perspective, the probability of a category 1 occurring is one every ten years. However, we saw two consecutive years—1954 and 1955—that faced category one or greater storms, so from a frequency perspective this isn’t outside expectations. With respect to storm surge, we think the slosh model generally performed well, and we calibrated our US storm surge expectations from that.

I concur with Marty and Charles—there are lessons to be learned. From my perspective the most important takeaway is hurricane deductibles, particularly at lower wind speeds. This is an issue that is rather politicised so we’ll look to price that differently in the marketplace going forward. We’ll take a much more conservative view as to whether we incorporate hurricane deductibles for category 1 or 2 events.

There’s going to be clear winners and losers from our perspective in flood underwriting and how they’ve performed, so application of deductibles, sub-limiting, attention to policy form, and how that translates into a loss in the reinsurance market will all be important. Overall the impact of these events on the re/insurance market makes them atypical. It’s probably around 65 percent commercial, whereas with a large windstorm event it is typically far more residential.

Also, with a $20 billion-plus event with a 65 percent commercial concentration there will be some significant net retentions for the big national accounts—the 16 or 17 companies that make up that pool. That’s $12.5 billion of net retentions, so distribution of loss through the marketplace will be atypical.


The other key takeaway is that there are some regional and superregional companies that are pretty deep into their programmes. If I was sitting on the board of one of those companies and saw a category one or tropical storm producing pretty similar losses to the top of the programme, I’d be asking some pretty hard questions such as ‘do we buy enough protection?’. We haven’t seen that demand flow through yet, but I think the concentration of values, the scope of losses, and all from a relatively modest event in terms of wind speed, but with a huge geographic footprint, probably dictates some serious questions about whether companies are buying sufficient protection.

Is there an expectation that insurers will buy more coverage in response to Sandy?

James Few: It might well happen. One of the positive things that may come out of Sandy is that the event might encourage people to question whether they bought enough cover. While Irene and Sandy were not the biggest events that could have happened, they could well change some thinking at management level about how much limit is purchased.

Returning to flood underwriting, it is going to serve as a test and differentiator for the quality of underwriting within the industry. On the personal lines side, the US government’s National Flood Insurance Program (NFIP) was already $17 billion in debt prior to Sandy. Sandy might well add another $10 billion of debt to that programme. This raises the question as to how much political will there is to continuerunning a scheme that clearly is underfunded. There are moves to offer private solutions for flood as a peril and the Reinsurance Association of America for example has recently helped make the case for legislation that will allow the NFIP to buy private reinsurance. That is a step in the right direction towards actuarially sound rates.

Therefore, there is potential for more market opportunity for flood in the private sector. Clearly, it has to be acceptable for the market to charge a fair rate—the market doesn’t want to be $27 billion in debt. There will need to be a lot more detail on flood mapping and original underwriting details—flood is a very locallyspecific peril. It’s difficult to model for lots of reasons, but one is that you can have two risks quite close together geographically that are completely different flood risks. Are the flood zone maps that underpin a lot of primary underwriting accurate enough? Could that be improved?

Konrad Rentrup: Currently we don’t see a general increase in demand. There is always the exception that cedants buy additional cover due to an increase of the underlying exposure. Some clients might try to add flood coverage into their programme, but that’s it. Going forward, this event is a wake-up call for some companies, especially for some primary insurers. They will need to look closely at their original policies and will evaluate the question of whether hurricane deductibles are adequate or if it would be more prudent to implement a wind deductible. The cover of flood exposure may yet prompt a political discussion in the US as so much of the loss will end up with the American taxpayer. Considering the financial situation facing the US budget, there is a definite need for close, intensive discussion regarding how the flood risk can be transferred into the private market. Insurers and reinsurers are willing to take on this risk if they are allowed to underwrite it and price it appropriately.

The expected losses caused by Superstorm Sandy underline the tremendous values located in hurricane-prone areas in the US. The observed migration of the US population to the US coastline is a continuing challenge especially for the insurance and reinsurance industry.

Cooper: One of the fundamental problems with the personal lines of insurance provided to homeowners in particular, is that the product is not easily understood by the consumer. This issue of deductibles, wind versus hurricane, is very opaque. That is compounded by wind versus flood, and most consumers don’t understand the nuances. We as an industry have done a pretty poor job of selling and explaining our product. Yes, you have to balance that with the economic reality and charge accordingly, but when our product is sold based on calculations using one methodology and then post-event that’s changed extracontractually, that causes problems for us as an industry.

Few: Another thing that may come out of Sandy, is that before the event there was some suggestion that rates had peaked for peak-zone US accounts. There were still increases on the primary side, both property and casualty were in single digits, but there was some fear in the marketthat those increases would lose some steam. That is less likely to happen now. Sandy will give more force to the need for rates to remain firm.

"Reinsurers could well benefit from increased rates in the primary market flowing through into reinsurance, or at least the reversing of a trend of increased retentions."

There was also a trend—and this has been around for years— of increasing retentions by primary carriers, and as Kean said, the primary market retained a considerable amount of the losses from both Irene and Sandy. That will likely act as a reminder of the value of reinsurance. Reinsurers could well benefit from increased rates in the primary market flowing through into reinsurance, or at least the reversing of a trend of increased retentions.

Do urban flood risks present unique challenges for the industry?

Driscoll: I wouldn’t disagree with that. If you get a loss in a nonurban area or suburban area it’s primarily going to affect residential programmes and the products themselves have a fair level of homogeneity. It’s also easier to calibrate the size of the loss. When you bring in an urban area, you involve meaningful commercial values, you bring in complexity, business interruption and concentration. From my perspective, business interruption in the context of this loss is a little bit overblown. It is rather different to the Thai loss.

It’s not a large manufacturing area—certainly there were some manufacturing losses—but it’s primarily white-collar firms. Contingency plans at the company, state, federal and municipality levels were quite effective and went a long way towards mitigatinglosses. Nevertheless, business interruption still adds complexity, and can create vast differences in commercial underwriting philosophy. If an urban area is impacted by an event, you increase the likelihood of flood, and when there is flooding it all too often highlights some firms’ inability to effectively manage risk.

Few: Business interruption clearly has the potential to really hurt the industry, and if you look at the scenario in Japan, everyone worried about the contingent business interruption (CBI) coming from manufacturing losses. It wasn’t as significant as it could have been due to the global economy and the fact that there were other places, such as Thailand that could take up the slack. Then we faced a flood in Thailand and people started thinking: that’s now two high-tech manufacturing locations that have been affected one after the other. There was considerable concern about CBI at that time.

"There is demand for contingent business interruption coverage, but the questions is whether that demand is willing to meet the needs of those who will provide such coverage?"

Sandy is largely an office-type business interruption scenario; mainly just extra expense. The costs involved largely concern relocating staff but the potential for business interruption is clearly obvious to any of us as underwriters and the industry needs to think much harder about collecting original data, particularly for CBI exposures, and also to think again about the size of limits that the industry is prepared to put out there. As an industry we’re not charging enough for this type of risk. The fact that it hasn’t happened to the extent it could, doesn’t mean that it can’t. That’s a burden we’re carrying.

Rentrup: There is demand for CBI coverage, but the question is whether that demand is willing to meet the needs of those who will provide such coverage? The price expectations of buyers and sellers are far apart. The high degree of uncertainty is reflected in the price insurers and reinsurers are charging for cover, and even after the events (Thailand floods, Sandy, Japan earthquake) we have not really seen any great uptake of such coverage.

Few: There have been a few examples of CBI losses over time. If you recall, Ericsson made a significant claim following a Philips plant failure in 2000. There have been examples of hi-tech manufacturing claims for relatively small parts of a supply chain resulting in an incredible concentration on one supplier or one small area of suppliers, and that’s a risk the industry might want to think more about. Despite the sheer scale of the loss in the Northeast, it hasn’t led to the kind of business interruption losses we might have expected in manufacturing locations.

Cooper: In this instance it was financial services versus manufacturing. The other issue is that it occurred in downtown Manhattan post-9/11, which has some pretty robust contingency plans.

Rentrup: Hurricane Sandy also proved a great surprise to the marine market. Losses resulting out of this premium-wise relatively small market are estimated to be around $2.5 billion.

Driscoll: Much like flood, there are going to be clear winners and losers in that space and there’s likely to be a pretty rude awakening for some. It’s a possibility that it will prove a slight marine retrocessional loss. Marine retrocessional capacity is incredibly scarce right now, and because it’s a small market, losses will have a compound effect.

Few: Sandy is probably the biggest marine loss ever. Previously it was Piper Alpha at around $2 billion, but this is probably going to reach $3 billion for the industry. It’s cargo, yachts, pleasure craft. Some interesting stories have come out of Sandy regarding wind and flood risk management, such as cargo containers that were stacked horizontally to avoid wind risk, resulting in still greater flood damage. A lot of the containers are being shipped to their final destination and the damage won’t be clear until they’re opened, which is one of the reasons why the marine loss is creeping.

Another Bermuda player has indicated that it intends to establish a fully collateralised reinsurance vehicle. What are people’s appetites like for such forms and how do they see that appetite developing?

Driscoll: We are an active participant in the managed capital space through our Alpha Cat brand, and that involvement can come in a variety of forms—traditional managed funds and sidecars among others. For us there are compelling reasons to participate in that space. First and foremost it allows us to offer more products and solutions to our partners. We offer not only increased capacity, but diversity of credit and, coupled with a consistent approach and a single point of entry, this represents an important differentiator that has received a positive response.

We like the space. It’s clearly a double-edged sword in terms of the amount of capital that’s being drawn into the space and I’m not quite sure how that ultimately plays out, but we see capital come in and behave responsibly and we see capital come in and behave irresponsibly. It’s probably no different than traditional rated reinsurance. But it’s a reality that we’re going to have to deal with—it is probably 15 percent of the total global catastrophe market right now, and certainly a much greater percentage than that in certain segments.

Becker: It’s 60 percent in the retro market. It’s clearly a phenomenon that’s real and here to stay for some time and as a property catastrophe writer you probably are disadvantaging yourself if you don’t find some way to participate.

Cooper: I agree. The one point that Kean brought up that I would debate is that collateralised reinsurance is a supply-driven phenomenon rather than a demand-driven issue. I would argue that the value of a collateralised product low down in the programme is debatable, and when you offset that with the additional costs of administrating a trust and issues such as that, I’m not sure it’s as valuable from a buyer’s perspective as it would be at the top of the programme, where people would really be worried about such risks.

Nevertheless, the capital markets are decidedly interested in the space. As a traditional player you’ve got two choices: you can just ignore it and hope it goes away—which I don’t think is going to happen—or you figure out how to embrace it, because otherwise you run the risk of eating from a much smaller pie in the future.

Few: We’ve been in this space since 2009 through a managed fund called Iris Re, but the world is changing—the cat world, specifically— and we all need to recognise that it’s probably going to change permanently, rather than in any one cycle-driven type scenario. It is not just due to an extended period of low investment returns that alternative capital is interested in cat. Alternative capital has become comfortable with the fact that this is a non-correlating asset class, buyers are more comfortable with basis risk and there is now a range of products to choose from.

"There may be some fall-off in capacity if interest rates rise and alternative asset classes become more attractive, but the point about diversification has been rammed home since 2008."

There may be some fall-off in capacity if interest rates rise and alternative asset classes become more attractive, but the point about diversification has been rammed home since 2008 and although there have been a few losses, capital is still available. Companies need to think more about this new, permanent reality. Some of the capital is happy with a lower return, so if that capital is coming to the market looking for a different return profile than we’ve been used to servicing, then obviously we need to consider the competitive implications.

Rentrup: It is capital that is going to stay. Yes, if interest rates rise maybe there will be less capital available in this segment, but it’s not going to disappear. It is diversification for the capital markets. Currently there is plenty of capacity in the market and as a traditional underwriter you do not really need further additions, but going forward collateral markets and ILS products can also add value. Currently any new capital that enters the industry will arrive via these kinds of instruments. I do not foresee that new capital will come to Bermuda or other jurisdictions by establishing a multi-billion reinsurance operation.

Driscoll: A point to maybe think about is where this capital comes in, even within the managed space. We all have some interest in the third party space and it’s largely with a closely aligned managed capital strategy, which is atypical with what we’re seeing in the broader space, where there are a lot of independents cropping up. However, what those investors are missing is a true understanding of what it takes to effectively deploy capital. They think it’s plug and play, you license a cat model and you’re good to go. But that is not sustainable, so I expect over the long term that more of the capital will flow to traditional reinsurerswith a strategy within the managed capital space. There’s enough room to go around for both the independents and the allied strategy, but the alliance strategy will, over time, win out on a greater percentage of agreements. That’s a better strategy in the long run because it puts capital in fewer hands and allows for more control over the deployment of capital and more control in terms of price and volatility in the market.

Cooper: I tend to agree with that, because one of the things that we as traditional reinsurers bring is expertise. We’ve been writing the business for years and we get that the model is a tool and it helps you underwrite, but that at the same time you have to think outside the box.

Few: Traditional reinsurers also have a long track record of being trading partners, which I think a lot of buyers care about. We’re less heavy in the cat world than some Bermuda players, but we look to build relationships across all lines and build a trading relationship that is mutually beneficial. It is not quite the same in-and-out type philosophy of a pure cat player. There is a part of the market that appreciates that and enjoys trading on that basis, but there is some value to the longer-term view of building business relationships. Traditional players can bring a lot to the party.

Becker: I would also say you’ve got much better visibility on deal flow. Traditional players are in the market every day, in a variety of places, and you’ve got much better insight into what the markets are doing.

Is an allied strategy the means to get involved in the alternative space without damaging your traditional offering?

Driscoll: We strongly believe that it is the right approach. Other traditional reinsurers have entered the space in a non-allied strategyso just as there’s a diverse approach to targeting traditional rated reinsurance in this room and globally, you’re going to see the same thing in the third party capital space. Our view is that we think over time more of the money will end up in the allied space.

Zurich has proved to be a popular centre for companies looking to build out their European operations. What are your thoughts on other global centres and what do you think are the prerequisites, from both a jurisdictional and market perspective, for the creation of a successful regional hub for insurance and reinsurance?

Becker: The answer is different for insurance and reinsurance. Insurance by nature is a much more local product, so more geographic gathering points are typically useful in insurance. Reinsurance is much more portable. Zurich has become popular for continental Europe—we’re there along with a number of other people. Singapore is growing—more and more people seem to want to do Asia out of Singapore. But Lloyd’s was very ambitious in wanting to set up geographic gathering points and now they’re sort of questioning whether that’s the right strategy. They find that a lot of that business will make its way to London anyway. It’s not as though these centres don’t have attached expenses. The real question is: how much incremental business do you feel you’re going to receive on the reinsurance side by having a local presence?

Cooper: One of the difficulties is that a lot of business that makes it out of those regional homes is the business where capacity is needed, and that tends to be better priced business than the smaller programmes from a reinsurance standpoint. Any time you have a regional hub set up, you run the risk of running business because you need to write business, and that is clearly what we’ve seen a lot of companies struggle with in Latin America. Singapore has had mixed results. A lot of these markets are relatively immature and it represents an expensive R&D project for some people.

Few: I may be slightly more pro-local market than some others although I agree with the general concept that local expansion needs to be matched by quality and certainly you’ve got to look at the expense issue.

"Any time you have a regional hub set up, you run the risk of running business because you need to write business, and that is clearly what we've sen a lot of companies struggle with in Latin America."

For our part we’re attracted to reinsurance business which is quite local and perhaps less volatile, with greater focus on personal lines and homogeneity. That kind of business around the world is increasingly being closed locally. What you’ll see in London or Bermuda is that those hubs tend to see larger cat-driven placements, more complex business, more complicated commercial risk. Companies will still come to London and Bermuda, but increasingly personal lines and smaller commercial business is being placed locally, and if you’re not there you simply don’t get it. You can now easily place a billion dollar facultative placement in the Singapore market without London participating. That was not happening in the early 2000s, but now it is not a problem. If you are looking at the overall portfolio you are trying to build, local business is attractive enough to have some hubs. To Marty’s point, it isn’t the same as insurance. You don’t need to put an office in every town. For Asia it is Hong Kong and Singapore, and increasingly a presence in China would make sense. For Europe it is pretty much Zurich, while for Latin America it is Miami. These kinds of places are seeing business which is no longer automatically being placed internationally.

So you’ve got to balance: what is the value of that business and what is the cost of getting it? There are challenges. It’s not easy to go into the emerging markets and pick up business. Data standards, modelling standards and legal systems are all different, and long-term relationships are often far more important than they are in Bermuda or London. You need time and longevity—that all adds to cost and risk. But if you want to pick up international business, if you want to improve your distribution, some kind of local presence is increasingly necessary.

Rentrup: I would completely agree with that. The big accounts will always end up in centres like Bermuda or in London. But many of the smaller accounts are regional, where personal contact is very important and worldwide capacity is not needed, and these will be placed in the local markets or regional insurance/reinsurance hubs. With the economic growth in the emerging markets their insurance and reinsurance needs grow and providers of these products who are closer to the market will have the first opportunity to meet the demand.

Few: Another thing that strikes me is that reinsurance is not generally a growing market. Increased retentions, the economic backdrop, other headwinds—in general reinsurance is not growing significantly. But there are parts of the world where GDP growth is exciting—a growing middle class, more need for casualty products, more property value that needs protection, more specialty opportunities through lines such as cargo. That kind of growth is available in certain parts of the world: proper growth, material growth. But it’s from a low base. When you’re struggling to find growth as a reinsurance market in general, surely it’s worth looking at emerging markets to see what’s available. Agriculture is another interesting area. Governments have a vested interest in feeding their voters. Potentially that’s a marketplace that might become increasingly interesting to national insurers and reinsurers.

In terms of the underwriting approach in these emerging markets, do you subscribe to the view that it’s better to have boots on the ground, or do you think it’s better to have centralised underwriting hubs such as Bermuda or Lloyd’s?

Becker: In emerging markets we would prefer to be closer to the ultimate client in order to have greater visibility of what’s happening in that market. You also have to balance that with the desire of emerging markets to require you to be close to the market. In Brazil you’ve got to have a local reinsurer or you’re screened out of 40 percent of the market. Argentina is passing similar legislation. They have recognised that there’s a lot of capital in our industry and it’s nice to have that capital in their local economy as opposed to exporting it all abroad. In the emerging markets, whether it is your intentional strategy or whether it’s a reaction to where the governmental bodies go, you are going to have to be more local than less.

Driscoll: The biggest challenge with any local office is the ability to export underwriting culture from the home office to the global office, and to rationalise the underwriting culture with the current capabilities of the market—whether it be data quality, product construct, overall pricing, or how you position yourself.

"You have to decide where you're going to take the risk: on the asset or the liability side of the business. There are some viable business models now... which are designed to take more asset-side risk, but you have to temper that with a lot less risk on the liability side."

We’re massively intrigued by Asia and Latin America. We’ve got a presence in both areas. We keep a really tight rein on matters because we have questions about the overall data quality, and about how the product is sold. They tend to be heavy proportional markets andwe’re not a huge fan of proportional business. But they will evolve as they get bigger and as the underlying capital bases of the ceding companies strengthen.They’ll evolve into a more excess market and we want to be positioned for that. We spend a lot of time trying to export our underwriting culture, our philosophy, our approach to rating business, and it is hard. It is hard for people in the local markets to say ‘no’ more often than they say ‘yes’. But if you can work your way through that, there’s no reason not to have a local presence, but that’s not a small challenge.

Few: Maintaining underwriting standards is key, but with modern systems reinsurance can be run in a relatively flat manner, so it is achievable. The reason it is better to have the underwriting at the front end is that with better knowledge and service you’re more likely to get the really plum business.

The other challenge is the expense point that Marty brought up. You can’t replicate every centre of excellence globally. You’ve got to use technology and systems to back up local offices. There isn’t always enough attractive business available in emerging markets topay for fully staffed offices but if you’re building for the five or 10-year picture, then perhaps there will be.

Cooper: These remain very small markets. The entire Latin American market will, in three years’ time, be about the size of the German market, and when you have a lot of capital chasing a small amount of business it creates obvious problems with pricing.

Few: My point would be that the people who are going to get the best deals in such a market are going to be the ones who are closest to the original buyer, particularly in strong relationship-driven places like Latin America. But it won’t be easy. The effort expended in a dollar of profit in an emerging market is considerably greater than the effort expended for a dollar of profit in the US.

Do you think that regulators and rating agencies are being overly prescriptive in their rulings as regards those investments the industry can pursue, and do you think reinsurers should be allowed to pursue more aggressive asset strategies?

Becker: I haven’t run into a lot of regulatory restraint on asset mix. You get rating agency constraint, but with the rating agencies it’s not so much the asset category you’re investing in, it’s just how much capital they’re charging you to be in that asset class. It’s a classic cost:benefit analysis. What do you think the incremental return is likely to be versus the amount of extra capital you’re going to have to carry? As regards our regulators, I can’t remember the last time I had a conversation regarding their concern about the mix of the assets on our balance sheet. We remain pretty high quality fixed income for the bulk of our assets.

Rentrup: I completely agree. We at Hannover Re Bermuda have a rather boring investment approach and pursue a purely fixed income strategy. That way we don’t have any discussions with regulators or rating agencies. But can reinsurers go into a more aggressive asset strategy? Yes, of course they can be more risky but it depends on their risk appetite and if they want to allocate their capital more to the investment or reinsurance risk.

Cooper: But you have to be extremely careful. As a business you have to decide where you’re going to take the risk: on the asset or the liability side of the business. There are some viable business models now, like PAC Re, which are designed to take more assetside risk, but you have to temper that with a lot less risk on the liability side.

Few: There is, given a sustained period of low investment returns, the notion that the way that the industry is invested through fixed income is not going to deliver the kind of return on equity that investors are used to. This is largely because underwriting rates haven’t responded sufficiently to make up for the lack of investment returns.

If we’re in an extended period of low interest rates, which most people think we probably are, then management teams are naturally considering whether a conservative asset side of the balance sheet isstill appropriate and whether there should be some kind of movement to taking a little more risk to try and improve yield. What I am aware of however is the industry moving a little from very highly-rated debt to fairly highly-rated debt.

Cooper: To get any kind of real improvement on the asset side you have to invest in a cat bond or similar instrument and that simply doesn’t make any sense.

Driscoll: As an entity PAC Re is pursuing a purely alternative investment strategy with significantly less liability. It’s a strategy that’s very specific to its circumstances. Validus, on the other hand, has an incredibly vanilla approach. Our investment philosophy is to maintain liquidity and conservatism. We never want a situation where the chief financial officer has to tell an underwriter to put down his or her pen because something has occurred in the global financial markets.

How are the non-cat lines faring? People seem to be getting into speciality and casualty lines more and more. Are these lines living up to expectation?

Becker: We probably write as much of that as anyone here. Bermuda’s non-cat business is a fairly narrow segment of the marketplace. It’s typically excess products, largely Fortune 2000 or Fortune 3000 kind of business. Historically for us, and even today, such lines have remained our most profitable business segments. They’ve shrunk in size as we’ve left some business as the marketplace pricing has changed, but the return on equity on that business remains incredibly attractive, and it hasn’t been that volatile.

Few: We’re running businesses that are diversified globally, with Bermuda a specialist market and the world’s leading market for excess of loss products. As Marty says, it is particularly adept at property, but there’s also plenty of capacity here for excess of loss products that don’t necessarily need to be sold close to the customer. This is a real centre of excellence for that kind of business.

Rentrup: A stand-alone company created in Bermuda which might have outside shareholders on the hunt for returns, cannot be successful by writing property catastrophe only. Catastrophe reinsurance requires significant capital to support the business and it would be very challenging to produce the returns investors are expecting. This is also reflected in the market—there is no Bermuda standalone company writing property catastrophe business only. All market players have diversified into other lines of business, merged with other players or left the market.

Cooper: Pure property catastrophe is turning into much more of a fund management structure where it’s not just your own capital that you’re managing, it’s third party capital, and that can help you get the kind of returns that you need. The problem with excess of loss business is the interest rate environment we’re in, and particularly with excess casualty, the liability reserves on that business. If you look at your cost of capital of holding those liabilities down relativeto the interest rates you’re earning—if you’re earning 2 percent and the cost of capital is 12 percent, that difference over a long period is challenging. There’s a lot of room for improvement in that market.

Finally, is it a tough writing specialty business in today’s environment?

Driscoll: There’s significantly more competition in the specialty segment than there was a year ago, certainly five or six years ago. There’s more interest in it, with more people chasing a pie that hasn’t significantly increased. We’re focused on non-cat exposed low frequency, high severity classes such as aerospace. They’re probably our best performing markets at present, although major losses have not occurred in those areas for some time.

Specialty is a concept that’s growing sideways rather than vertically. But we like the class, we’ll continue to be as active as we can in trying to grow our footprint in specialty, but some of these classes are pretty difficult to underwrite, establish a costs of goods sold, and manage your true exposure, so you’ve got to navigate these waters carefully. There are good returns, but these are volatile classes.