In last year’s Energy Market Review we pulled no punches in outlining why conditions had changed so dramatically in the Downstream market. 12 months further on, we are finding, if anything, an intensifying of the hardening process as insurers, buoyed by their ability to finally instigate a market turnaround in 2019, continue to press for even more draconian rating increases - pretty much across the board. Their determination to do so is fuelled by relentless management pressure to deliver improved underwriting results within the context of poor performances across the Property & Casualty (P&C) spectrum. Buyers are therefore being faced with the unenviable choice of either scaling back on the cover purchased – leaving them with potentially significant new risk exposures – or accepting that current insurance budgets will have to be revised.
The scales tip further… A quick glance at Figure 1 on page one shows why the “scales” have tipped even further in favour of the insurance market as we have moved further into 2020. Where once capacity levels and hunger for premium income were the key drivers in this market, those days are long gone as the key driver is now firmly to underwrite this portfolio profitably at all costs. As a result, we now not only looking at a reduction in available capacity; we are also looking at a myriad of factors, not least of which are the continuation of the recent appalling loss record and increasingly severe management pressures that are turning the screw even further on buyers (as we suggested in our November 2019 Update).
Time to reassess risk transfer strategies? But in this new decade, as the market hardening process shows no signs of abating, are there ways in which this process can be managed and mitigated? Are there ways in which buyers can offset some of the worst effects of the upturn in market rates? Let’s have a look at the overall situation in more detail and then conclude with some suggestions that may form the basis of an enhanced risk transfer strategy.
Capacity reduces for the second year in succession Figure 2 overleaf shows overall capacity levels in the Downstream market during the last 20 years. It shows that the long period of year-on-year capacity increases, which began in 2005 in the aftermath of that year’s appalling hurricane losses, finally came to an end last year with the first decrease in overall capacities in 14 years. Most market observers will not be surprised to see that theoretical capacity levels for 2020 have reduced still further, down to US$5.978 billion from last year’s US$6.428 billion. In practice, much will depend on the geography of the risk in question but in general terms it is becoming increasingly challenging to access capacity in the way in which buyers have become accustomed during the last 14 years. What are the besetting reasons behind this continued reduction in capacity? It’s not really a case of multiple insurer withdrawals. During the last 12 months only one insurer of note (Pioneer) has withdrawn from this class of business and we have only had one significant merger, between Arch and Barbican. It’s much more the case that the artificially high figures offered in theory by major insurers in the past – to attract the interest of buyers and their brokers – have now been withdrawn and replaced by more realistic figures.
Line size restrictions a factor Insurers are also finding that their participations are being limited by two other significant factors:
US$3 billion still achievable - at a price However, based on our own brokers’ experience of trading in this market, we still think it is possible to access US$3 billion for a well-regarded International programme (US$2 billion for US programmes). We identified these realistic maximum figures in our November 2019 Update, and we see no reason to revise these figures downwards at this stage in the market cycle - so long as buyers are prepared to accept the inevitable pricing increases currently being demanded. One of the reasons for this is the impact of the introduction of Downstream specific reinsurance treaties; this development has had the effect of releasing a degree of additional capacity to the direct market.
At the beginning of 2019, some Downstream insurers might have been hoping for a respite following three years of virtually unprecedented losses. However, a glance at the data outlined in Figure 3 overleaf tells a very different story; the 2019 data that we hold on our database shows a further deterioration in the loss record and, while the figures are not quite as bad as they were for 2017, last year’s figures (which may be even worse if revised loss forecasts materialise during the next few months) are already in excess of any other year apart from 2017 since 2008, the year of hurricane Ike.
Existing loss record deterioration No wonder these figures have produced consternation in the market. The difficulty that underwriters are facing is the continuing deterioration in individual loss figures and explaining these to senior management, who generally hate bad surprises; loss adjusters are often reporting initial advices of one figure and then recommending a final settled figure at several multiples of the original figure.
We believe that the discrepancy between the initial and final figures can generally be put down to three factors:
Premium income on the rise In the meantime, 2019 saw an increase in overall premium income as the hardening process instigated at the beginning of last year produced some much-needed increased revenue to the market. However, as Figure 3 illustrates, this looks to have been insufficient to prevent another loss-making year for the Downstream portfolio in overall terms.
Finally, Figure 4 to the right depicts the major losses to have impacted the market since the beginning of 2018. It is interesting to note the number of losses involving a vapour cloud explosion (VCE) at either a refinery or a petrochemical plant, the degree to which the Business Interruption (BI) element often outweighs the Physical Damage (PD) element and the preponderance of losses emanating from North America.
Are buyers absorbing more losses? What our database does not show, however, is the degree to which these figures represent insured rather than uninsured losses. This year we have been astonished to see at least one of our clients suffer a major loss to find that a very significant part of it has been uninsured, due to the buyer electing to purchase only minimal Business Interruption cover. What seemed like a prudent decision before the loss to reduce premium spend has ultimately resulted in a very large uninsured loss something that must make for a very difficult conversation between the Risk Manager and his or her Financial Director.
Figure 5 to the left represents the only data available to estimate the overall profitability of the Downstream portfolio. These statistics, issued by Lloyd’s, depict Incurred Ratios (i.e. premium received versus paid and outstanding claims) for the Onshore Energy audit code over the last 27 years. It is important to note that the figures for 2019 are much too immature to be germane; however, the data for the years preceding 2019 is sufficiently robust for us to be able to estimate overall portfolio probability. Obviously, any figures in excess of 100% demonstrate unprofitability; however, as explained in previous editions of this Review, anything in excess of 50% is also likely to result in overall portfolio losses.
2014 the last profitable year? The data shows that the last overall profitable year for Lloyd’s Onshore Energy was back in 2014; since then Lloyd’s has recorded Incurred Ratios of 57%, 88%, 124% and 87.5%. While these figures include some element of onshore Exploration & Production (E&P) losses, by common consent the vast majority of these losses pertain to the Downstream sector. Given these gloomy figures, it is perhaps not so surprising that underwriters remain in a determined mood to bring about an escalation of the hardening process in the market. So what can buyers expect as we move further into 2020?
“What did it pay last year?” At the time of writing (March 2020) there is no question that the momentum established by the market during 2019 has continued to escalate. Underwriters are generally under instructions from their management to insist on rating increases across the board, with the first question on every underwriter’s lips being: “What did it pay last year?” Gone are the days when brokers could introduce a programme to an insurer who did not participate on the previous year as “new business” – even if this is the first time the insurer has been involved, no one can be seen to be offering a rating reduction on last year’s price. Programmes that were paying rises in the region of 30% last year can probably expect a rise of 40% or so this year. Those programmes that are attracting the most attention from the market are refinery and petrochemical programmes, particularly from North America; moreover, these programmes are likely to suffer from less advantageous terms and conditions, particularly with regard to lower sub-limits for such coverages as Contingent Business Interruption.
LNG, Midstream and Russia less affected However, other areas of the portfolio, including LNG plants and other midstream business such as conventional gas plants and onshore pipelines, are likely to be treated more leniently by the market, although some form of rate increase is virtually unavoidable. In particular, LNG programmes have one advantage over the market, in that these programmes often feature significant premium income and captive participation, as well as the possible involvement of the Upstream market as these assets are not excluded from Upstream reinsurance programmes. These factors serve to reduce the apparent leverage of the insurance market and we have seen rating increases for this business average out at a lower percentage figure than for refinery and petrochemical programmes. Another area of the portfolio to be less affected by the hardening market conditions is Russian business; such is the scale of the premium income involved, the long-established relationships with key market leaders and the relatively benign loss record for this portfolio that the hardening process has turned out to be much smoother than for other geographical areas.
It could have been worse… And yet the current situation from a buyer perspective could be so much worse. Certain major European composite insurers that already boast significant capacity have watched rating levels on “quota share” (i.e. unlayered) placements increase substantially over the course of the last 12 months. Previously these (re)insurers have preferred to participate on an excess basis, believing that the lower or quota share layers were under-priced for the risk involved. Now, they seem to be more willing to offer their capacity on a quota share basis, injecting more capacity into this area of the risk spectrum and mitigating the overall percentage rise that would have been applied but for their new participation. Added to this should be the effect of the new Downstream reinsurance treaties mentioned earlier in this chapter. So although prices continue to rise, the extent of the increases is at least being offset by the appetite of some insurers who are beginning to sense an opportunity to capitalise on today’s underwriting climate.
Differentiation dynamic intensifies Apart from this, the market continues to adopt essentially the same stance as during the latter half of 2019. What has not changed has been the way that the market continues to differentiate in favour of those buyers that have maintained their relationship with key carriers. In these challenging market conditions, these buyers are being treated more favourably than those who took advantage of the previous soft market conditions to drive down the overall cost of their programme. In contrast, we are now sensing an increased determination in the market to ensure that those buyers who regularly tendered their programme to drive maximum competition during the soft market should now be put in the same position as the rest of their peers - in other words, to be singled out for more draconian rating increases than the rest of their peer group. There is no doubt that some buyers have benefitted from continuing to tender their programmes on a regular basis; however, it is equally clear that others have not been so fortunate. Indeed, we are aware of several instances recently when a programme has been labelled as “distressed” in the market following a tender process where the price quoted by the broker as been proven to be totally unrealistic. As a result, the market understood that the buyer had no place to run and no place to hide; as a result, the eventual price on which the programme was placed was almost certainly in excess of what the buyer would have paid had they persevered with the existing programme.
Natural catastrophe cover limited and expensive Meanwhile natural catastrophe (nat cat) cover remains limited and relatively expensive to purchase. Indeed, for those buyers that are suddenly forced to solicit nat cat cover mid-term, for example because they have taken over assets in a specific geographical location, the cost can be overwhelming – as much as a 10% rate on line – if insurers’ aggregate exposure limits have already been reached. At a time when climate change is likely to drive ever more frequent windstorms and floods across the globe, there is currently a clear disconnect between the risk transfer offering that is available from the conventional insurance market and the future risk transfer requirements of the energy industry. Bridging that gap will require considerable collaboration between brokers, insurers and buyers in the years ahead.
Business Interruption restrictions We mentioned in last year’s Review that one leading insurer was attempting to introduce an Actual Declared Value (ADV) basis of coverage for Business Interruption, to replace the traditional “Gross Profit” form widely used in the Downstream market during the prolonged soft market of 2005-18. During the last 12 months other insurers have also found that they have been basing their premium calculations on one set of values, declared by the buyer at inception as their estimate for the forecast calendar year, but finding that they are actually paying a loss later in the process on the basis of a completely different set of figures. The reasons for this kind of disparity are varied, but one possible reason is the increase in the use of forensic accounting specialists to quantify the actual loss. Be that as it may, we are now finding that a larger number of insurers are now seeking to impose an annual cap of say approximately 110% of the values declared at inception, with a monthly cap of say 125% (although there is a conversation to be had with insurers to increase this for clients that provide the right information to the market). As we intimated last year, it seems clear that the solution to this issue is to forecast more accurate numbers to insurers at the inception of the programme. However, there is clearly a tension between ensuring an accurate payout in the event of a loss and keeping insurance costs to a minimum in the first instance. Moreover, it seems clearly to be in buyers’ long-term interests to have their insurers accept accurate figures at the inception of their programme rather than to find that in the event of loss the final settlement is disputed - and therefore at the very least delayed. It seems clear to us that a professional deployment of risk engineers, forensic accountants and valuation consultants will eventually lead to greater trust and certainty for both buyer and insurer, speedier settlement times and less of a need to resort to lawyers.
Resultant cyber cover clarified As well as a focus on Business Interruption, there is a new development to report regarding cyber coverage. At the culmination of the January 1 renewal season, Lloyd’s management became concerned at the potential total aggregate cyber risk to which the Corporation might be exposed and as a result there has been a general introduction of the LMA 5400 or LMA 5401 Cyber Exclusion clause. The original Electronic Data Endorsement clause, NMA 2915, provided a fire and explosion buyback but didn’t talk about cyber-attack. Furthermore, in the NMA 2915 clause the only mention of “malicious” is in the definition of Computer Virus, whereas the LMA5400 and 5401 specifically do exclude Cyber Attack, but use the term “malicious” without defining what it means. In practice, it may be difficult to establish and define the word “malicious” - does it, for example, include the actions of a disgruntled employee, thereby putting such actions at the same level as a specifically planned terrorist attack? The LMA 5401 is a total cyber exclusion, but even where the lesser exclusion LMA 5400 is applied, the written back fire and explosion (and sometimes further named perils) are still excluded if they result from Cyber Acts. Furthermore, the Data Processing Media cover in the LMA 5400 is similar to that given in the NMA 2915, although most re/insurers insist on a sub-limit for loss of data.
To give some sort of wider perspective on today’s Downstream market dynamics, it’s worth having a look at our data that we have collated over 27 years and determine exactly where we are in the overall market cycle. Figure 6 to the right shows maximum capacity levels set against estimated average rating levels, using an index of 100 for 1992. The historical data by no means presents a neatly ordered cycle of supply contraction and expansion, but instead suggests a slowing down of the cycle itself. Up until about 2003, market volatility was fairly pronounced, with prices reaching a record low in 1999 and a record high just four years later.
Volatility has flattened out compared to previous eras However, during the last ten years this volatility has flattened out considerably. The capacity increases prevalent in the last soft market (2010–18) have been steadier, while the accompanying rating decreases have also been less dramatic than in previous eras. So market dynamics have become more of an ocean liner – heavier, steadier but more difficult to stop – rather than a speedboat, capable of turning on a sixpence.
Now that the market has turned, buyers can therefore expect the current state of affairs to continue for some time yet and should adjust their expectations accordingly. We see no sign of any end to the hardening process at present, and certainly no indication of any fresh competition that would threaten insurers’ resolve to bring market rating levels back to what they consider to be profitable levels.
It is perhaps worth pointing out that, even with two years of rating increases under their belts, rates are still only back to where they were five years ago - halfway through the last soft market. As we stated in last year’s Energy Market Review, even at these inflated rating levels compared to 2018, the Downstream market continues to offer clients excellent value for money.
Brokers and buyers must work more closely together So in the absence of any additional competition, and in light of the performance of other areas of the Heavy Industry portfolio, buyers and their brokers will have to work together to offset the worst effects of this continued market hardening. This will involve working as a partnership to develop the right strategy to offset this process; it may involve the increased use of risk engineers, forensic accountants, analytics specialists and others.
As we move further into 2020, buyers continue to have the option of engaging with their broker to develop this process to smooth any future volatility or to carry on as before, hoping for optimal cover from an increasingly intransigent market that itself is increasingly apprehensive about the future.
Steve Gillespie is Head of Downstream broking at Willis Towers Watson Natural Resources in London.
