Tag Archives: low risk premia

Befuddled Lloyd’s

Lloyd’s of London always provides a fascinating insight into the London insurance market and beyond into the global specialty insurance market, as this previous post shows. It’s Chairman, Bruce Carnegie-Brown, commented in their 2017 annual report that he expects “2018 to be another challenging year for Lloyd’s and the Corporation continues to refine its strategy to address evolving market conditions”. Given the bulking up of many of its competitors through M&A, Willis recently called it a reinvigoration of the “big balance sheet” reinsurance model, Lloyd’s needs to get busy sharpening its competitive edge. In a blunter message Brown stressed that “the market’s 2017 results are proof, if any were needed, that business as usual is not sustainable”.

A looked at the past 15 years of underwriting results gives an indicator of current market trends since the underwriting quality control unit, called the Franchise Board, was introduced at the end of 2002 after the disastrous 1990’s for the 330-year-old institution.

click to enlarge

The trend of increasing non-CAT loss ratios after years of soft pricing coupled with declining prior year reserve releases is clear to see. That increases the pressure on the insurance sector to control expenses. To that end, Inga Beale, Lloyd’s CEO, is pushing modernisation via the London Market Target Operating Model programme hard, stating that electronic placement will be mandated, on a phased basis, “to speed up the adoption of the market’s modernisation programme, which will digitise processes, reduce unsustainable expense ratios, and make Lloyd’s more attractive to do business with”.

The need to reduce expenses in Lloyd’s is acute given its expense ratio is around 40% compared to around 30% for most of its competitors. Management at Lloyd’s promised to “make it cheaper and easier to write business at Lloyd’s, enabling profitable growth”. Although Lloyd’s has doubled its gross premium volumes over the past 15 years, the results over varying timeframes below, particularly the reducing underwriting margins, show the importance of stressing profitable growth and expense efficiencies for the future.

click to enlarge

A peer comparison of Lloyd’s results over the past 15 years illustrates further the need for the market to modernise, as below. Although the 2017 combined ratio for some of the peer groupings have yet to finalised and published (I will update the graph when they do so), the comparison indicates that Lloyd’s has been doing worse than its reinsurance and Bermudian peers in recent years. It is suspicious to see, along with the big reinsurers and Bermudians, Lloyd’s included Allianz, CNA, and Zurich (and excluded Mapfe) in their competitor group from 2017. If you can’t meet your target, just change the metric behind the target!

click to enlarge

A recent report from Aon Benfield shows the breakdown of the combined ratio for their peer portfolio of specialist insurers and reinsurers from 2006 to 2017, as below.

click to enlargeAon Benfield Aggregate Combined Ratio 2006 to 2017

So, besides strong competitors, increasing loss ratios and heavy expense loads, what does Lloyd’s have to worry about? Well, in common with many, Lloyd’s must contend with structural changes across the industry as a result of, in what Willis calls in their latest report, “the oversupply of capital” from investors in insurance linked securities (ILS) with a lower cost of capital, whereby the 2017 insured losses appears to have had “no impact upon appetite”, according to Willis.

I have posted many times, most recently here, on the impact ILS has had on property catastrophe pricing. The graph of the average multiple of coupon to expected loss on deals monitored by sector expert Artemis again illustrates the pricing trend. I have come up with another angle to tell the story, as per the graph below. I compared the Guy Carpenter rate on line (ROL) index for each year against an index of the annual change in the rolling 10-year average global catastrophe insured loss (which now stands at $66 billion for 2008-2017). Although it is somewhat unfair to compare a relative measure (the GC ROL index) against an absolute measure (change in average insured loss), it makes a point about the downward trend in property catastrophe reinsurance pricing in recent years, particularly when compared to the trend in catastrophic losses. To add potentially to the unfairness, I also included the rising volumes in the ILS sector, in an unsubtle finger point.

click to enlarge

Hilary Weaver, Lloyd’s CRO, recognises the danger and recently commented that “the new UK ILS regulation will, if anything, increase the already abundant supply of insurance capital” and “this is likely to mean that prices remain low for many risks, so we need to remain vigilant to ensure that the prices charged for them are proportionate to the risk”.

The impact extends beyond soft pricing and could impact Lloyd’s risk profile. The loss of high margin (albeit not as high as it once was) and low frequency/high severity business means that Lloyd’s will have to fish in an already crowded pond for less profitable and less volatile business. The combined ratios of Lloyd’s main business lines are shown below illustrating that all, except casualty, have had a rough 2017 amid competitive pressures and large losses.

As reinsurance business is commoditised further by ILS, in a prelude to an increase in machine/algorithm underwriting, Lloyd’s business will become less volatile and as a result less profitable. To illustrate, the lower graph below shows Lloyd’s historical weighted average combined ratio, using the 2017 business mix, versus the weighted average combined ratio excluding the reinsurance line. For 2003 to 2017, the result would be an increase in average combined ratio, from 95.8% to 96.5%, and a reduction in volatility, the standard deviation from 9.7% to 7%.

click to enlarge

To write off Lloyd’s however would be a big mistake. In my view, there remains an important role for a specialist marketplace for heterogeneous risks, where diverse underwriting expertise cannot be easily replicated by machines. Lloyd’s has shown its ability in the past to evolve and adapt, unfortunately however usually when it doesn’t have any choice. Hopefully, this legendary 330-year-old institution will get ahead of the game and dictate its own future. It will be interesting to watch.


Epilogue – Although this analogy has limitations, it occurs to me that the insurance sector is at a stage of evolution that the betting sector was at about a decade ago (my latest post on the sector is here). Traditional insurers, with over-sized expenses, operate like old traditional betting shops with paper slips and manual operations. The onset of online betting fundamentally changed the way business is transacted and, as a result, the structure of the industry. The upcoming digitalisation of the traditional insurance business will radically change the cost structure of the industry. Lloyd’s should look to the example of Betfair (see an old post on Betfair for more) as a means of digitalising the market platform and radically reducing costs.

London Isn’t Calling

In a previous post, I reproduced an exhibit from a report from Aon Benfield on the potential areas of disruption to extract expenses across the value chain in the non-life insurance sector, specifically the US P&C sector. The exhibit is again reproduced below.

 click to enlargeexpenses-across-the-value-chain

The diminishing returns in the reinsurance and specialty insurance sector are well known due to too much capital chasing low risk premia. Another recent report from Aon Benfield shows the sector trend in net income ROE from their market representative portfolio of reinsurance and specialty insurers, as below.

click to enlargenet-income-roe

It’s odd then in this competitive environment that the expense ratios in the sector are actually increasing. Expense ratios (weighted average) from the Willis Re sector representative portfolio, as below and in this report, illustrate the point.

click to enlargewillis-re-expense-ratios

The 2016 edition of the every interesting S&P Reinsurance Highlights, as per this link, also shows a similar trend in expense ratios as well as showing the variance in ratios across different firms, as below.

click to enlargesp-expense-ratios

Care does need to be taken in comparing expense ratios as different expense items can be included in the ratios, some limit overhead expenses to underwriting whilst others include a variety of corporate expense items. One thing is clear however and that’s that firms based in the London market, particularly Lloyds’, are amongst the most top heavy in the industry. Albeit a limited sample, the graph below shows the extent of the difference of Lloyds’ and some of its peers in Bermuda and Europe.

click to enlargeselect-expense-ratios

Digging further into expense ratios leads naturally to acquisitions costs such as commission and brokerage. Acquisition costs vary across business lines and between reinsurance and insurance so business mix is important. The graph below on acquisition costs again shows Lloyds’ higher than some of its peers.

click to enlargeselect-acquisition-cost-ratios

Although Brexit may only result in the loss of fewer than 10% of London’s business, any loss of diversification in this competitive market can impact the relevance of London as an important marketplace. Taken together with the gratuitous expense of doing business in London, its relevance may come under real pressure in the years to come. London is, most definitely, not calling.

Lessons from Lloyds

There is little doubt that the financial services industry is currently facing many challenges and undergoing a generational change. The US economist Thomas Philippon opined that the finance industry over-expansion in the US means that it’s share of GDP is about 2 percentage points higher than it needs to be although he has also estimated that the unit cost of intermediation hasn’t changed significantly in recent years, despite advances in technology and the regulatory assaults upon the industry following the financial crisis.

The insurance sector has its own share of issues. Ongoing low interest rates and inflation, broader low risk premia across the capital markets, rapid technology changes such as big data and the onset of real time underwriting are just the obvious items. The Economist had an article in March that highlighted the prospective impact of data monitoring and technology on the underwriting of motor and health risks. This is another interesting post on a number of the new peer to peer business models such as Friendsurance, Bought by Many, and Guevara who are trying to disrupt the insurance sector. There can be little doubt that the insurance industry, just like other financial sectors, will be impacted by such secular trends.

However, this post is primarily focused on the short to medium term outlook for the specialty insurance and reinsurance sector. I have been asked a few of times by readers to outline what I think the next few years may look like for this sector. My views of the current market were nicely articulated by Alex Maloney, the Group CEO of Lancashire, who commented in their recent quarterly results statement as follows:

“The year to date has seen a flurry of activity on the M&A front within the industry, much of this, in my view, is driven by the need to rationalise and refocus oversized and over stretched businesses. We also continue to see a bout of initiatives and innovations in the market, the sustainability and longer term viability of which are questionable. These are symptoms of where we are in the cycle. We have seen these types of trends before and in all likelihood, will see them again.”

Lloyds of London has had a colourful past and many of its historical issues are specific to it and reflective of its own eccentric ways. However, as a proxy for the global specialty sector, particularly over the past 20 years, it provides some interesting context on the trends we find ourselves in today. Using data from Lloyds with some added flavour from my experiences, the graphic below shows the dramatic history of the market since 1950.

click to enlargeLloyds Historical Results 1950 to 2015

The impact of Hurricane Betsy in 1965 upon Lloyds illustrated a number of the fault-lines in the structure of Lloyds with the subsequent Cromer report warning on the future danger of unequal treatment between insiders (aka working Names) and “dumb” capital providers (aka all other Names). The rapid influx of such ill informed capital in the late 1970s and the 1980s laid the seeds of the market’s near destruction largely due to the tsunami of US liability claims resulting from asbestos and pollution exposures in the 1980s. These losses were exacerbated by the way Lloyds closed underwriting years to future capital providers through vastly underpriced reinsurance to close transactions and the practice of the incestuous placement of excess of loss retrocession for catastrophe losses within the market, otherwise known as the London Market Excess of Loss (LMX) spiral. There is a clever article by Joy Schwartzman from 2008 on the similarity between the LMX spiral and the financial risk transformational illusions that featured heavily in the financial crisis. Indeed, the losses from the sloppy “occurrence” liability insurance policy wordings and the tragedy of unheeded asbestos risks continued to escalate well into the 1990s, as the exhibit below from a 2013 Towers Watson update illustrates.

click to enlargeTowers Watson Asbestos Claims US P&C Insurers

What happened in Lloyds after the market settlement with Names and the creation of the “bad bank” Equitas for the 1992 and prior losses is where the lessons of Lloyds are most applicable to the market today. The graphic below shows the geographical and business split of Lloyds over the past 20 years, showing that although the underlying risk and geographical mix has changed it remains a diversified global business.

click to enlargeLloyds of London Historical Geographical & Sector Split

Released from the burden of the past after the creation of Equitas, the market quickly went on what can only be described as an orgy of indiscipline. The pricing competition was brutal in the last half of the 1990s with terms and conditions dramatically widened. Rating indices published by the market, as below, at the time show the extent of the rate decreases although the now abandoned underwriting indices published at the same time spectacularly failed to show the impact of the loosening of T&Cs.

click to enlargeLloyds of London Rating Indices 1992 to 1999

As Lloyds moved from their historical three year accounting basis in the 2000s it’s difficult to compare historical ratios from the 1990s. Notwithstanding this, I did made an attempt to reconcile combined ratios from the 1990s in the exhibit below which clearly illustrates the impact market conditions had on underwriting results.

click to enlargeLloyds of London historical combined ratio breakdown

The Franchise Board established in 2003, under the leadership of the forthright and highly effective Rolf Tolle, was created to enforce market discipline in Lloyds after the disastrous 1990s. The combined ratios from recent years illustrate the impact it has had on results although the hard market after 9/11 provided much of the impetus. The real test of the Franchise Board will be outcome of the current soft market. The rating indices published by Amlin, as below, show where rates are currently compared to the rates in 2002 (which were pushed up to a level following 2001 to recover most of the 1990s fall-off). Rating indices published by Lancashire also confirm rate decreases of 20%+ since 2012 in lines like US property catastrophe, energy and aviation.

click to enlargeLloyds of London Rating Indices 2002 to 2015

The macro-economic environment and benign claims inflation over the past several years has clearly helped loss ratios. A breakdown of the recent reserve releases, as below, show that reinsurance and property remain important sources of releases (the reinsurance releases are also heavily dependent on property lines).

click to enlargeLloyds of London Reserve Release Breakdown 2004 to 2014

Better discipline and risk management have clearly played their part in the 10 year average ROE of 15% (covering 2005 to 2014 with the 2005 and 2011 catastrophe years included). The increasing overhead expenses are an issue for Lloyds, recently causing Ed Noonan of Validus to comment:

“We think that Lloyd’s remains an outstanding market for specialty business and their thrust towards international diversification is spot on from a strategic perspective. However, the costs associated with Lloyd’s and the excessive regulation in the UK are becoming significant issues, as is the amount of management and Board time spent on compliance well beyond what’s necessary to ensure a solvent and properly functioning market. Ultimately, this smothering regulatory blanket will drive business out of Lloyd’s and further the trend of placement in local markets.”

So what does all of this tell us about the next few years? Pricing and relaxed terms and conditions will inevitably have an impact, reserve releases will dry up particularly from reinsurance and property, investment returns may improve and claim inflation may increase but neither materially so, firms will focus on expense reduction whilst dealing with more intrusive regulation, and the recent run of low catastrophic losses will not last. ROEs of low double digits or high single digits does not, in my view, compensate for these risks. Longer term the market faces structural changes, in the interim it faces a struggle to deliver a sensible risk adjusted return.

Converts on a comeback

My initial reaction, from a shareholder view-point, when a firm issues a convertible bond is negative and I suspect that many other investors feel the same. My experience as a shareholder of firms that relied on such hybrid instruments has been varied in the past. Whether it’s a sign that a growing firm has limited options and may have put the shareholder at the mercy of some manipulative financier, or the prospect that arbitrage quants will randomly buy or sell the stock at the whim of some dynamic hedging model chasing the “greeks”, my initial reaction is one of discomfort at the uncertainty of how, by whom, and when my shareholding may be diluted.

In today’s low risk premia environment, it’s interesting to see a pick-up in convertible issuances and, in the on-going search for yield environment, investors are again keen on foregoing some coupon for the upside which the embedded call option that convertibles may offer. Names like Tesla, AOL, RedHat, Priceline and Twitter have all been active in recent times with conversion premiums averaging over 30%. The following graph shows the pick-up in issuances according to UBS.

click to enlargeConvertible Bond Market Issuances 2004 to 2014

Convertible bonds have been around since the days of the railroad boom in the US and, in theory, combining the certainty of a regular corporate bond with an equity call option which offers the issuer a source of low debt cost at a acceptable dilution rate to shareholders whilst offering an investor the relative safety of a bond with a potential for equity upside. The following graphic illustrates the return characteristics.

click to enlargeConvertible Bond Illustration

The problem for the asset class in the recent past came when the masters of the universe embraced convertible arbitrage strategies of long/short the debt/equity combined with heavy doses of leverage and no risk capital. The holy grail of an asymmetric trade without any risk was assumed to be at hand [and why not, given their preordained godness…or whatever…]! Despite the warning shot to the strategy that debt and equity pricing can diverge when Kirk Kerborian’s increased his stake in General Motors in 2005 just after the debt was downgraded, many convertible arb hedge funds continued to operate at leverage multiples of well in excess of 4.

The 2008 financial crisis and the unwinding of dubious lending practises to facilitate hedge fund leverage, such as the beautifully named rehypothecation lending by banks and brokers (unfortunately the actual explanation sounds more like a ponzi scheme), caused the arbitrage crash not only across convertibles but across many other asset classes mixed up in so called relative value strategies. This 2010 paper, entitled “Arbitrage Crashes and the Speed of Capital”, by Mark Mitchell and Todd Pulvino is widely cited and goes into the gory detail. There were other factors that exacerbated the impact of the 2008 financial crisis on the convertible debt market such as market segmentation whereby investors in other asset classes were slow to move into the convertible debt market to correct mis-pricing following the forced withdrawal of the hedge funds (more detail on this impact in this paper from 2013).

Prior to the crisis, convertible arb hedge funds dominated the convertible bond market responsible for up-to 80% of activity. Today, the market is dominated by long only investors with hedge funds only reported to be responsible for 25% of activity with those hedge funds operating at much lower leverage levels (prime brokers are restricted to leverage of less than 1.5 times these days with recent talk of an outright rehypothecation ban for certain intermediaries on the cards). One of the funds that made it through the crash, Ferox Capital, stated in an article that convertible bonds have “become the play thing of long only investors” and that the “lack of technically-driven capital (hedge funds and proprietary trading desks) should leave plenty of alpha to be collected in a relatively low-risk manner” (well they would say that wouldn’t they!).

The reason for my interest in this topic is that one of the firms I follow just announced a convertible issue and I wanted to find out if my initial negative reaction is still justified. [I will be posting an update on my thoughts concerning the firm in question, Trinity Biotech, after their Q1 results due this week].

Indeed, the potential rehabilitation of convertible bonds to today’s investors is highlighted by the marketing push from people like EY and Credit Suisse on the benefits of convertible bonds as an asset class to insurers (as per their recent reports here and here). EY highlight the benefit of equity participation with downside protection, the ability to de-risk portfolios, and the use of convertible bonds to hedge equity risk. Credit Suisse, bless their little hearts, go into more technical detail about how convertibles can be used to lower the solvency requirement under Solvency II and/or for the Swiss Solvency Test.

With outstanding issuances estimated at $500 billion, the market has survived its turbulent past and it looks like there is life left in the old convertible bond magic dog yet.

Insurance & capital market convergence hype is getting boring

As the horde of middle aged (still mainly male) executives pack up their chinos and casual shirts, the overriding theme coming from this year’s Monte Carlo Renez-Vous seems to be impact of the new ILS capacity or “convergence capital” on the reinsurance and specialty insurance sector. The event, described in a Financial Times article as “the kind of public display of wealth most bankers try to eschew”, is where executives start the January 1 renewal discussions with clients in quick meetings crammed together in the luxury location.

The relentless chatter about the new capital will likely leave many bored senseless of the subject. Many may now hope that, just like previous hot discussion topics that were worn out (Solvency II anybody?), the topic fades into the background as the reality of the office huts them next week.

The more traditional industry hands warned of the perils of the new capacity on underwriting discipline. John Nelson of Lloyds highlighted that “some of the structures being used could undermine some of the qualities of the insurance model”. Tad Montross of GenRe cautioned that “bankers looking to replace lost fee income” are pushing ILS as the latest asset class but that the hype will die down when “the inability to model extreme weather events accurately is better understood”. Amer Ahmed of Allianz Re predicted the influx “bears the danger that certain risks get covered at inadequate rates”. Torsten Jeworrek of Munich Re said that “our research shows that ILS use the cheapest model in the market” (assumingly in a side swipe at AIR).

Other traditional reinsurers with an existing foothold in the ILS camp were more circumspect. Michel Lies of Swiss Re commented that “we take the inflow of alternative capital seriously but we are not alarmed by it”.

Brokers and other interested service providers were the loudest cheerleaders. Increasing the size of the pie for everybody, igniting coverage innovative in the traditional sector, and cheap retrocession capacity were some of the advantages cited. My favourite piece of new risk management speak came from Aon Benfield’s Bryon Ehrhart in the statement “reinsurers will innovate their capital structures to turn headwinds from alternative capital sources into tailwinds”. In other words, as Tokio Millennium Re’s CEO Tatsuhiko Hoshina said, the new capital offers an opportunity to leverage increasingly diverse sources of retrocessional capacity. An arbitrage market (as a previous post concluded)?

All of this talk reminds me of the last time that “convergence” was a buzz word in the sector in the 1990s. For my sins, I was an active participant in the market then. Would the paragraph below from an article on insurance and capital market convergence by Graciela Chichilnisky of Columbia University in June 1996 sound out of place today?

“The future of the industry lies with those firms which implement such innovation. The companies that adapt successfully will be the ones that survive. In 10 years, these organizations will draw the map of a completely restructured reinsurance industry”

The current market dynamics are driven by low risk premia in capital markets bringing investors into competition with the insurance sector through ILS and collaterised structures. In the 1990s, capital inflows after Hurricane Andrew into reinsurers, such as the “class of 1992”, led to overcapacity in the market which resulted in a brutal and undisciplined soft market in the late 1990s.

Some (re)insurers sought to diversify their business base by embracing innovation in transaction structures and/or by looking at expanding the risks they covered beyond traditional P&C exposures. Some entered head first into “finite” type multi-line multi-year programmes that assumed structuring could protect against poor underwriting. An over-reliance on the developing insurance models used to price such transactions, particularly in relation to assumed correlations between exposures, left some blind to basic underwriting disciplines (Sound familiar, CDOs?). Others tested (unsuccessfully) the limits of risk transfer and legality by providing limited or no risk coverage to distressed insurers (e.g. FAI & HIH in Australia) or by providing reserve protection that distorted regulatory requirements (e.g. AIG & Cologne Re) by way of back to back contracts and murky disclosures.

Others, such as the company I worked for, looked to cover financial risks on the basis that mixing insurance and financial risks would allow regulatory capital arbitrage benefits through increased diversification (and may even offer an inflation & asset price hedge). Some well known examples* of the financial risks assumed by different (re)insurers at that time include the Hollywood Funding pool guarantee, the BAe aircraft leasing income coverage, Rolls Royce residual asset guarantees, dual trigger contingent equity puts, Toyota motor residual value protection, and mezzanine corporate debt credit enhancement  coverage.

Many of these “innovations” ended badly for the industry. Innovation in itself should never be dismissed as it is a feature of the world we live in. In this sector however, innovation at the expense of good underwriting is a nasty combination that the experience in the 1990s must surely teach us.

Bringing this back to today, I recently discussed the ILS market with a well informed and active market participant. He confirmed that some of the ILS funds have experienced reinsurance professionals with the skills to question the information in the broker pack and who do their own modelling and underwriting of the underlying risks. He also confirmed however that there is many funds (some with well known sponsors and hungry mandates) that, in the words of Kevin O’Donnell of RenRe, rely “on a single point” from a single model provided by to them by an “expert” 3rd party.

This conversation got me to thinking again about the comment from Edward Noonan of Validus that “the ILS guys aren’t undisciplined; it’s just that they’ve got a lower cost of capital.” Why should an ILS fund have a lower cost of capital to a pure property catastrophe reinsurer? There is the operational risk of a reinsurer to consider. However there is also operational risk involved with an ILS fund given items such as multiple collateral arrangements and other contracted 3rd party service provided functions to consider. Expenses shouldn’t be a major differing factor between the two models. The only item that may justify a difference is liquidity, particularly as capital market investors are so focussed on a fast exit. However, should this be material given the exit option of simply selling the equity in many of the quoted property catastrophe reinsurers?

I am not convinced that the ILS funds should have a material cost of capital advantage. Maybe the quoted reinsurers should simply revise their shareholder return strategies to be more competitive with the yields offered by the ILS funds. Indeed, traditional reinsurers in this space may argue that they are able to offer more attractive yields to a fully collaterised provider, all other things being equal, given their more leveraged business model.

*As a complete aside, an article this week in the Financial Times on the anniversary of the Lehman Brothers collapse and the financial crisis highlighted the role of poor lending practices as a primary cause of significant number of the bank failures. This article reminded me of a “convergence” product I helped design back in the late 1990s. Following changes in accounting rules, many banks were not allowed to continue to hold general loan loss provisions against their portfolio. These provisions (akin to an IBNR type bulk reserve) had been held in addition to specific loan provision (akin to case reserves). I designed an insurance structure for banks to pay premiums previously set aside as general provisions for coverage on massive deterioration in their loan provisions. After an initial risk period in which the insurer could lose money (which was required to demonstrate an effective risk transfer), the policy would act as a fully funded coverage similar to a collaterised reinsurance. In effect the banks could pay some of the profits in good years (assuming the initial risk period was set over the good years!) for protection in the bad years. The attachment of the coverage was designed in a way similar to the old continuous ratcheting retention reinsurance aggregate coverage popular at the time amongst some German reinsurers. After numerous discussions, no banks were interested in a cover that offered them an opportunity to use profits in the good times to buy protection for a rainy day. They didn’t think they needed it. Funny that.