A “Muddy Confluence” of Appointed Actuaries

A small actuarial contingent facing growing regulatory requirements provides an opportunity for CAS collaboration.

Recently, I had the opportunity to attend an excellent two-day workshop targeted to property-casualty appointed actuaries and their staff members. The CAS co-organized the workshop, and all three of the workshop presenters proudly carry CAS credentials. I had met only one of the presenters prior to connecting with them via the workshop. I gained some really useful knowledge, interesting perspectives and fresh insights from attending. The workshop attendees were quite engaged; I enjoyed hearing their questions and comments, and working with a dozen colleagues during the breakout session, as well as chatting with many other participants more informally during the breaks.

I know. Those elements don’t sound like the ingredients of headline news; many other CAS seminars could be described similarly. But consider this: The workshop was conducted in Kuala Lumpur, capital city of Malaysia, half a world away from the vast majority of CAS members.

“Kuala Lumpur” literally means “muddy confluence” — thus the title of this article. The confluence refers to the rivers that flow down from the inland areas. In the 1850s, rich sources of tin were discovered in the area. That tin, and the pewter of which the tin is a key ingredient, have been significant to the ascendency of the city as a center of commerce.

So how and why did the CAS come to be involved in organizing and presenting a workshop in Malaysia in November 2014?

The roots of this workshop go back a year earlier, to the 2013 East Asia Actuarial Conference (EAAC) in neighboring Singapore. During that 2013 conference, senior CAS representatives leveraged the presence of more than 500 actuaries from all over Asia to schedule a series of bilateral conversations with leaders from various actuarial organizations in the region, including the Actuarial Society of Malaysia (ASM). The ASM’s leaders briefed us on the general insurance¹ actuarial marketplace in Malaysia (30 general insurance companies, including seven takaful operators, serving a population of 30 million people, but fewer than 20 credentialed actuaries focusing on general insurance). ASM’s leaders also outlined some of the challenges of meeting the rapidly evolving actuarial educational needs of a small and young profession in a country where the insurance industry is in a state of significant development and evolution. We, in turn, described the CAS’s appetite to work collaboratively with local organizations to deliver excellent education for general insurance actuaries around the globe. A seed of an idea was planted.

That seed took root and sprouted in the early months of 2014 under the leadership of Kelvin Hii, ASM vice president and leader of ASM’s General Insurance Committee, who invited me to continue the EAAC conversations. A shared but not well-defined interest in educational collaboration came into clear focus when Bank Negara, the Malaysian regulatory authority, announced a new requirement for appointed actuaries: to issue an annual financial condition report on each general insurance company. The ASM and the CAS decided to join forces to produce a two-day workshop for Malaysia’s general insurance actuaries, with a bit more than half of the workshop to be focused on financial condition reports and the remainder on loss reserving methods and considerations. The CAS agreed to identify, recruit, provide and support the speakers; ASM very skillfully managed all of the local marketing, budget and finance, site selection and management, logistics and on-site support. We also were pleased to welcome the Institute of Actuaries of Australia (IAAus) as a co-sponsor and to have the support of IAAus, particularly with respect to endorsing and marketing the event to its members through its significant communication channels.

Our solicitation for potential CAS speakers produced numerous enthusiastic volunteers with a healthy diversity of backgrounds, experiences and job responsibilities. Ultimately, we selected a team of three:

• Judy Ng, a 2013 FCAS working in Singapore, was, with her active consulting work for a variety of insurers, an excellent choice to lead the basic loss reserving sessions.
• Herb Desson, a 1992 ACAS who has worked for several insurance companies in Thailand since relocating there in 2006, brought a depth of real-world experience to his sessions on the more subtle considerations that enter into the selection of loss reserving methods and the calibration of parameters.
• Daniel Tess, a 1997 FCAS and also a Fellow of IAAus, has lived and worked in Australia since 1998 and has extensive familiarity with implementing and delivering financial condition reports as defined in Australian regulation. He was very effective throughout the seminar in using that platform to illuminate key issues and to suggest key areas of focus for appointed actuaries in the Malaysia market.

The loss reserving presentations in Malaysia also benefited significantly from tapping the substantial CAS library of presentation materials used at past Casualty Loss Reserve Seminars. (Our thanks to CAS Director of Professional Education and Research Dave Core for great help in finding the materials we needed, and to the past CAS authors and presenters who created those materials.)

The 75 participants themselves contributed tremendously to the effectiveness and the success of the workshop. One of the highlights of the first day was a breakout session at the end of the day. The attendees were divided into groups and each group was assigned a question or issue to discuss. There was not a right or wrong answer, but rather a complex situation requiring thoughtful weighing of different ideas.

The team to which I was assigned took on the following issue: “You are the loss reserving actuary for Malaysia Airline. How do you analyze the airline’s loss reserve requirements at the end of 2014?” This group decided to segregate the reserving analysis into two major elements. The first would deal with routine daily claims (minor injuries, damaged property and so forth), and would likely use typical forms of aggregated data and conventional actuarial methods to examine and extend patterns in that aggregated data. The second element would focus on catastrophic claims and likely would be customized to deal with the particulars of each catastrophe, the process and status of identifying and resolving all claims and obligations, and the knowledge of experts. This element would also contain detailed information about legal responsibilities and liability limits, insurance and reinsurance protection, and the involvement of other insurers in the various layers of exposure. Another theme of this discussion centered on communicating with experts in the business being analyzed. This communication is vital because it gives the actuarial analysis some context and shows that the analysis is not simply a mathematical process applied to abstract numbers.

One of the teams considered how an actuary might use generalized linear models (GLMs) for loss reserving. This group saw the potential of GLMs as predictive models to estimate ultimate losses from partially developed claims. For example, you may apply three dimensions in traditional triangulation data: accident years, development years and financial year factors (e.g., inflation). However, you need to be careful of overparameterization. Because external factors may result in past trends being not relevant to future expected developments, caution is needed when using GLMs. With any projection model, it is important to understand the intricacies of the model and its limitation, as opposed to using the model as a “black-box.” Data quality remains an important consideration, as bad data may produce results that are not credible.

I was very impressed with the grasp of key loss reserving concepts, as well as the subtle issues, that came out in our group’s discussion. And I was equally impressed with the energy of the participants! During the second day, some of the liveliest discussions during the general sessions related to how the appointed actuary can clearly convey the key elements of the technical analysis, the nature and magnitude of uncertainty associated with future projections, and the most important business implications of the actuarial work and conclusions. These are communication challenges that I still face on almost every project! Equally energetic were the informal exchanges and networking among participants and instructors during the various breaks.

On all fronts, the workshop was an extraordinarily successful first collaboration of the CAS and the Actuarial Society of Malaysia. And already, the CAS and the ASM are following through on an intention to repeat our joint experiment with another workshop in Kuala Lumpur in the fourth quarter of 2015; this one will be focused on ratemaking and predictive modeling. With the Malaysian regulators currently laying the groundwork for relaxing the centralized determination of insurance prices—and allowing more competitive pricing processes—what could be more topical and more essential for Malaysia’s general insurance actuaries?

Bob Conger is a CAS past president and consultant with Towers Watson. He served as CAS International Ambassador from 2013 to 2014 and is a member of several CAS international committees.

¹ While “property-casualty” is the North American collective label for the insurance products that comprise the work of many CAS members, “general insurance” is the corresponding term in Malaysia and many other countries. The remainder of this article will refer to general insurance.