Insurers need to overcome trust concerns towards the analytics ‘black box’ to achieve the full benefits of automated loss reserving.
New analytics technologies are permeating the actuarial function and are showing great promise particularly in loss reserve analysis. Many insurers are considering the value of integrating cognitive computing capabilities into their actuarial processes. Yet, to realize the full benefits of automation, Property and Casualty (P&C) insurers will first need to learn to trust increasingly complex systems.
The automation of loss reserving could solve a number of key challenges facing most insurers and potentially reduce costs, improve business flexibility and market share gains.
Indeed, data scientists are using algorithms to produce estimates that out-perform traditional approaches while also standing up to rigorous `back-testing'. Besides, automation assigns equal weight to each piece of data, which can help remove the risk of human bias where prior conclusions are given greater weight than newer information.
Similarly, actuaries can create real value for the insurance organization by providing higher level-analysis of data and applying their deep actuarial judgement to the insights, and decision makers can view current and in-depth analysis to spot trends, manage risks and respond real-time. The pace of business can therefore be dramatically improved as reports can be developed in hours rather than weeks, and business performance and efficiency can be tracked to uncover improvement opportunities.
Automated loss reserving clearly offers P&C insurers efficiency and efficacy gains, but to achieve its full benefits, executives will need to overcome their concerns about the transparency of the so-called `black box' of analytics.
Most management teams and their boards rely heavily on the insights and analysis offered by their actuarial function, and their trust in those insights is largely based on their historical human interaction, rather than the complex machines that they don't understand and can't relate to.
While the actuarial function needs to build trust in the tools and more sophisticated analytics capabilities will certainly be required, the bigger barrier will be culture: actuaries need to understand that automation is here to help them become more productive, not to replace them.
While trust may currently be low, we believe that P&C insurers can help raise the level of trust by applying key concepts that we refer to the Four Pillars of Trusted Analytics:
The path to improved trust in reserve model automation is an ongoing process. It will require continuous end-to-end consideration of the data, assumptions, calculations and resulting value. As value is generated and results are validated, the level of trust will improve.
Over the coming year, we expect the perceived loss of transparency towards the increasingly complex systems to be replaced by the confidence that can be gained through rapid feedback created by automated structures. With continuous feedback of actual results, acceptance and adoption of automation will quickly start to rise.
We firmly believe that the P&C sector is on the cusp of an exciting new wave of actuarial innovation, and those who recognize and respond to the opportunity will likely see significant competitive advantage.
Find out more about the automated actuarial in our publication, and contact a KPMG advisor on how you can leverage trusted analytics to capitalize on new methods and technology for your organization.
© 2017 KPMG International Cooperative (“KPMG International”), a Swiss entity. Member firms of the KPMG network of independent firms are affiliated with KPMG International. KPMG International provides no client services. No member firm has any authority to obligate or bind KPMG International or any other member firm vis-à-vis third parties, nor does KPMG International have any such authority to obligate or bind any member firm. All rights reserved.
KPMG International has created a state of the art digital platform that enhances your experience, optimized to discover new and related content.