Comparing apples and pears

With an increasing use of electronic underwriting engines, how do reinsurers compare the quality of different provider’s rulebooks? Andy Smith investigates risk assessment and operational performance

Historically, much credit or otherwise has been given to electronic underwriting offerings on the basis of the software in use. Undoubtedly, there are many praiseworthy underwriting engines in the market, whether self built or off the shelf.

However, the rulebook within varies considerably from implementation to implementation.

Ultimately, it is the rules themselves that form the coal face, or rather interface, with the customer, and it is this that really matters. In that respect, the actual engine used is perhaps not as important as the content, provided the functionality of that engine does not hinder rule design.

So, how do reinsurers measure the performance of differing client rulebooks? Can’t we just run each client’s rules through one ‘master’ reinsurer rulebook to compare? Is it not possible to gather the data outputs and feed these into the reinsurer’s underwriting engine?

Major obstacles

One major obstacle prevents such a solution. Providers do not assess risk in the same way. Different insurers use different criteria and therefore different rule structures in designing their own rulebooks. As a consequence, the data these rules produce will only reflect the rule structures and assessment criteria it has arisen from. This will not then be compatible with an engine containing a different rulebook.

As an example we can use an old favourite, asthma (and why not considering it regularly makes up 10% of all disclosures). One company may use the following criteria to assess the risk and therefore build rules and actions based around these:
Childhood asthma or late onset

• Number of attacks per annum
• Use of oral steroids
• Hospitalisation – ever
• Occupation induced or complicated by
• PFR (Peak Flow Rate)
• Smoker / non smoker

Another decides to assess the same disclosure in a different way:

• Frequency symptoms inhibit daily activities in the last two years
• Frequency of nocturnal symptoms per week
• Intensive care treatment required in the last five years
• Number of cigarettes per day
• Number of treatments required to control symptoms
• Frequency inhalers used per week