Modernising underwriting: from segregation to integration


Yannis (John) Spanoudakis

Modernising underwriting: from segregation to integration

bearsky23 /

Separation between lines of business might be rational and convenient but it is inevitable that the differentiation will gradually disappear and integration will take its place, writes Yannis Spanoudakis from Matrix Brokers.

The insurance market is admittedly a conservative environment so it comes as no surprise that historically it has been accused as being bureaucratic, unimaginative, slow-moving, etc. Despite the fact that a market offering long-term commitments should have stability and not flexibility as its primary priority, many of those accusations do hold water.

This atmosphere has triggered multiple efforts in the last two decades to change the traditional ways towards a more aggressive approach for ‘modernisation’ or ‘evolution’. Such efforts essentially focused not so much on the ways of doing the business, but merely on the ways of running the business (two different things which are often confused).

“Different legislation and exposures might justify such segregation technically but the clients do not care how we name the cover, they just want/need to be covered against any relevant mishap.”

In the context of evolving the industry, examples emanating from other more (allegedly) innovative business areas have been considered, the banking sector being the usual suspect. Understandably the weight was shifted towards most visible and easily influenced areas. One of them was administration: processes have been streamlined, average costs have been reduced, speed has been increased. Probably it has happened with some shortcuts here and there, but there has still been some success.

One other target was sales & marketing. We all are, to some extent at least, wholesalers now and mass-marketing and sales methods are applied. Catchy ‘company statements’ and well-engineered advertisement campaigns are now part of everyday life.

But what about the core of insurance: underwriting? Times have changed but underwriting—this unique insurance function—has basically remained unchanged and heavily dependant on the human factor, the almighty ‘underwriter’. When everything else is advancing, staying the same (or almost the same) actually means that you are losing ground. Not everyone would admit this, but in most lines of business underwriting is still performed in the same old ways—and at this point and for clarity I stress that simplification of underwriting is not to be considered ‘evolution’. Despite all the modern tools, data analysis, etc, underwriting seems to remain basically unchanged, although the ‘risk’ is becoming more complex every day.  


The way forward

In the aftermath of 9/11, everybody was crying out that basic safety nets such as accumulation control and risk allocation had failed. Since then hundreds of millions of dollars have been spent on IT systems, customer-managed relationships (CMRs), etc but still ‘real-time accumulation control’ remains a dream. With GPS now easily available on mobile devices, fast and easy accessibility to all types of networks, high computing speeds at low cost, cheap huge data banks (only two decades ago ‘available memory’ was our daily nightmare, remember?) and, last but not least, more efficient and highly educated personnel, this should (and could) have been the first problem to be resolved.

There is nowadays no real technical (or financial) obstacle to successful real-time risk accumulation control, so the problem remains not because of lack of technology or resources but because of lack of priority. We refer only to the absolute basics—risk assessment and quantification and pricing should follow and these are by far more difficult and diversified across the lines.

The truth is that many leading companies are now focusing on improving/modernising (read ‘automating’) underwriting. There are many reasons for such sudden anxiety: necessity for quicker response, error minimisation, regulatory requirements on the one hand, highly demanding risks and (let’s be honest) the lack of experienced underwriters on the other.


The best laid plans

Even with some delay, are we on the right path? Well, in reality it does not matter what you do but how you do it. The worst plan might be brilliantly implemented and work whereas the best ideas may be destroyed by poor execution. Challenges—risks—are increasingly changing and it is not easy to keep up. Even in commodity lines such as motor, we are facing a real revolution as autonomous driving is just a few years away and the insurance market is still discussing how to deal with it.

Fortunately, reality is simple but unfortunately, simple does not mean easy. Irrespective of the multiple and complex technical issues, the essence is that the owner/driver of the vehicle will no longer be solely responsible for an accident—the car manufacturer (and indirectly all kinds of sub suppliers) would also become involved on a daily basis. In other words, motor liability is to become heavily influenced by professional and products liability.

How do we deal with it as direct insurers or reinsurers? The good old-fashioned way, step one is proof of fault, isn’t it? Well, that’s good money for surveyors and lawyers. Step two is cross-line allocation of loss—again good money for lawyers and consultants. Is all this really meaningful or is a dead end?

In essence, what we are really facing is a mix of various liabilities. Until now we approached different risks with different products, different pricing, different methodology (ultimately through different underwriters/underwriting). But such a separated approach is no longer adequate, not only because risks are evolving but also because clients (and regulators) are becoming more demanding.

Grey areas in insurance contracts are nowadays rarely accepted even if totally explainable from our (the insurer’s) point of view. Clients (all kinds, in all lines of business) do not really care to hear about how insurers separate and draw barriers between different lines of business, different terms and exclusions, various rates and multiple deductibles, etc.

We should prepare ourselves that in the not-too-distant future we will have to provide clients with more ‘spherical’ insurance products—not just packaged policies where different products are bound together but single common uniform wordings with seamless multiline coverage underwritten through a single common process. Someone may note that this is already happening in certain cases with customised policies, but it is quite selective and rare and actually still segregated underneath.

Segregation of insurance lines of business is artificial in some cases—take casualty for example: public, motor, aviation, marine, professional, etc. Different legislation and exposures might justify such segregation technically but the clients do not care how we name the cover, they just want/need to be covered against any relevant mishap. And crossing between lines is sometimes blurred even to us—I myself have tried many times to get a simple but precise description about the actual boundaries of ‘aviation liability’, with no satisfactory result.


Joining forces

One contemporary example is cyber. There always has been concern about the threat from hackers and their ilk but it was the Millennium virus that terrorised the whole industry in the late 1990s—and this in practice non-real risk cost billions (literally) down the drain to all kinds of industries.

The insurance market has evolved from the Millennium threat—classic but now obsolete cyber exclusion clauses can nevertheless still be found—and is trying to develop specialised cyber products for mass sell. There is demand, yes, and some decent products are already available but penetration remains low. This effort is mainly focused on the ‘pure’ cyber risk but perhaps while trying to develop a specialised product we should also be trying to incorporate cyber (or cyber-like) exposure in common policies (hopefully at a price).

If we do not do so, it is inevitable that we shall end paying for what we consider to be cyber losses under the normal (eg, property or casualty) policy. Production systems are now ‘open’ and hackers are (or soon will be) in a position to cause actual material damage—not only data loss. Will the usual property policies—and their pricing—cater for such situations or in-between scenarios? What if a production line just ‘freezes’ after a cyber attack and as such there is no direct material damage—how would the business interruption section react in a ‘standard’ BI wording? Or how would we deal with actual material damage caused through malicious programming?

Before going there, have we asked ourselves as insurers or reinsurers whether we intend, or really want, to cover such cases? The truth is that the ‘normal’ property underwriter cannot fully assess such risks and the cyber underwriter probably cannot fully quantify them. The only way to deal with them is by combining experiences, bonding know-how, brainstorming around the same table, and developing common tools. In other words, ‘integrating’ the lines is the only way out of it.

Separation between lines of business might be rational and convenient but it is inevitable that (maybe with very few exceptions) the differentiation will gradually disappear and integration will take its place. Are we prepared for it, both psychologically and technically? What will it take to move from segregated lines to integrated areas?

To do it effectively IT tools will certainly be required, provided they are designed on that principle. That would involve not only taking advantage of the multiple available resources (Google Earth, social media, technical databases, news media, etc) but also incorporating a totally new approach to the way all such data are processed. Nevertheless, what the transition from segregation to integration requires first is more open ears and minds. It is primarily a matter of culture, of the way of thinking, of taking the big step of crossing the frontiers to ‘boldly go where no-one has gone before’. Evolving on a human business level is necessary in order to evolve technically and that is the most difficult and time-consuming part.

It is what ‘innovation’ is really all about, and it is how tomorrow’s winners will emerge.

Yannis (John) Spanoudakis can be contacted under

Underwriting, Insurance, Technology, Yannis Spanoudakis, Matrix Brokers

Intelligent Insurer