In collaboration with our partner company, Typhon at the Lloyd’s market, we’ve pulled together some preliminary thoughts on how simulation can help underwriters gain a better understanding of where there may be significant unknown risks in a property portfolio. Based on our experience, we believe that a combination of systems thinking, simulation and modelling can give underwriters:
greater insights into your risks and
the ability to act to control them.
We are interested in exploring if these technologies can help identify substantial but unknown risks lying hidden in a current property portfolio. We also want to explore the development of a model whereby new and difficult to evaluate risks could be added as time goes by.
Our perspective on property portfolio risk-exposure
While the majority of the risk accepted falls within the category of random instances of individual variation from agreed standards, there have been occasional situations (e.g. subsidence) where a single cause has widespread effects across the portfolio and which could therefore lead to very substantial claims against the syndicate. Illustrative historical examples might include:
The mandated introduction of condensing boilers, which was followed by a spate of failures because plumbers were unused to dealing with the condensate drains. This led to considerable rework and warranty claims across the domestic boiler industry.
The inappropriate use of joist hangers, especially coupled with incorrectly sized joists, which led to some spectacular failures. The issue is sufficiently widespread that the NHBC issued a technical note advising on good practice.
The drive for energy efficiency means that new products are appearing with increasing frequency – such as windows with new coatings, integrated photovoltaic tiles or systems that use heat pumps or solar panels. These devices, being new, are subject to faulty installations that could lead to longer term unforeseen common-mode failures that could be expensive to rectify.
The UK house building industry is not so much an industry as an amalgam of crafts, trades, large scale contractors and professionals such as architects and surveyors. Each of these actors focus on their own concerns and seek to pass risk off to others.
Whilst there is a framework of rules and regulations supervised by local authorities, there is no genuine control over the total building process – quite unlike what one finds in industries such as pharmaceuticals or automotive manufacturing. The UK house building industry is an extraordinarily complex ecosystem.
The contractor selection process and the inspection regime seem focused more on ensuring that the contractors are not comprising on quality to achieve their cost and performance measures. Considerations such as uncovering and managing the systemic risks to which the syndicate may be exposed are not likely to be a high priority for either the contractors or the inspectors.
We therefore believe that if there are major but as yet unaddressed risks for the syndicate, they are most likely to arise, not from isolated elements but from failures in the way the system as a whole operates – in other words, problems in the ways in which the various elements combine. Indeed, both the examples mentioned above with the boilers and the joists are essentially systemic, originating in deficient processes.
We believe that these unanticipated risks are most likely to arise from this fragmented industry structure of actors with separate agendas and the consequent lack of strong and enforceable communications and risk monitoring processes across, as well as within, these domains.
There is, therefore, a justifiable doubt left with those exposed to the final result that there may be substantial but unknown (at least to them) risks lying hidden in the current portfolio as well as the concern that new ones could be added as time goes by.
Our experience suggests that the greatest sources of unknown systemic risk will arise through either:
weak communication between the various entities involved in this fragmented industry, especially the high impact upstream processes of design, material selection, building practices etc., or
within a domain, such as contractors, where increasing cost and time pressures may lead to a consistent and cumulative drift in deviations from agreed standards.
While these drifts might be acceptable by themselves as random deviations from the norm, when taken together, they could significantly change for the worse the overall risk being accepted by the syndicate - the long stop at the end of all these individual steps.
Current changes in the UK environment may further exacerbate these tendencies as builders’ profits plummet. The pressures on them may add yet further risks. Risk is further increased when new approaches to design and delivery are introduced in response to the need to increase productivity or reduce costs without adequate understanding of their full systemic effects.
If this system level analysis seems plausible, we suggest there are real benefits to syndicates to be gained from building a syndicate-focused high-level model of the overall UK system and its attendant risks (as outlined in the attached diagrams).
In the short term, such a model will provide syndicates with valuable new insights into where significant unanticipated risk may have entered the portfolio in the recent past, signalling where the expense of more detailed inspection and data gathering is justified.
In the longer term, using this approach will help syndicates to illustrate to the upstream players where they are creating risks for the syndicate - and enlist their support for any procedural changes deemed necessary.
End-to-end process control using system-aware approaches has been successfully adopted by our clients in a variety of industries where the appropriate insight and executive authority exists to make the changes needed. Opportunities to extend this to the building industry are already being explored by networks of professionals such as the International Group for Lean Construction.
We believe that a focused systemic approach, given that there are a relatively small number of high impact risks, has an excellent chance of succeeding.
To take things forward, we suggest that the three next steps for any syndicate with a property portfolio are:
Explore with Typhon and appropriate colleagues within the syndicate to define risk types, instances and dependencies. This will enable Typhon to build an integrated view of the ecosystem within which house building occurs, the house building system itself and the overarching risk assessment and management system.
Prioritise risk categories and of their possible evolution in different categories of housing (or specific developments).
Model and simulate the possible outcomes of risk types to identify specific sensitivities, to understand the dynamics of long-term embedded risks and to evaluate the potential for better predicting risks and claims.
During this work, we would advise involving experts from each area where risk arises. These experts would need to come with practical “on-the-ground experience”, open minds and plenty of (possibly unattributable) hard facts on what can happen in practice.
There may be, as you fear, a single cause to a welter of damages. There is also the potential that our simulation will find truly unanticipated risks. The simulation (and the preparatory work) will be designed to model the entire system of risks and the elements that compose it.
The simulation program will move the syndicate who invests in its development to a higher level of understanding. It will also make it possible for them to manage, or distribute risks, in better ways, e.g. avoiding putting too many similar eggs in a particular basket (or by re-pricing a particularly frequent risk element). Conversely, the syndicate’s better understanding of your “risk portfolio” may also allow you to adjust premiums thus enlarging your business base.
