Bhavna Verma, the Appointed Actuary at IndiaFirst Life, shares her expertise on the evolving landscape of actuarial work and the integration of technology in the insurance industry. From discussing notable changes in actuarial tables to highlighting the role of AI and ML, Bhavna provides valuable insights into the key trends shaping the field of actuarial science:
Ravi Lalwani: What are the most notable changes in actuarial tables you have seen in your career?
Bhavna Verma: Actuarial mortality tables are updated periodically in every jurisdiction. In India, studies on mortality experience have been undertaken at the industry level separately for assured lives (the IALM tables) and annuitant lives by the Institute of Actuaries of India (IAI).
During my career, I have seen a shift from IALM (2006-08) to the IALM (2012-14) tables wherein the newer tables were adopted by all companies as prescribed by the Insurance Regulatory and Development Authority of India (IRDAI). Recently, the IAI has also released the IALM (2015-17) table which indicates an overall improvement in mortality rates and highlights certain differences in experience of various segments. The first report on Individual Annuitants Mortality 2012-15 was published by the IAI in 2021.
Such tables form a useful starting point for product pricing and actuarial reporting.
What are the newer sources of data you are looking at for updating actuarial tables?
Actuarial pricing involves a combination of the use of standard tables, layered with the emerging experience of own portfolio and reinsurer expertise, if available.
The size and credibility of internal databases are expanding; this can enable more focused pricing for product categories and target customer segments.
Also, in the post-pandemic world and constantly evolving external environment, it is very important to stay on top of current events and external studies. As an example, the long-term impacts of Covid are a subject that is being studied and will continue to evolve over many years into the future.
Do you find the need to convert subjective information into objective information? Please give examples.
Objective information is generally better appreciated. The peculiarity of actuarial work lies in the fact that it is mostly about estimating the future, therefore it is about layering actual objective information available to date with subjective information about potential future events and making estimates accordingly. This applies to both operating parameters such as persistency, mortality, and operating expenses and as much to economic parameters such as interest rate outlook.
A very relevant example of actuarial work is converting 360-degree information internally and externally into conclusive actuarial assumptions to price a long-term contract, value the liabilities of the company, and even value the company.
How are you shaping the usage of IT in your organization in your domain and related domains?
As an organization, we believe in leveraging technology as much as possible, including the actuarial domain. Specialized actuarial software is used for the production of monthly results. Wherever possible, we utilize technology platforms to carry out tasks that facilitate the actuarial control cycle, such as experience monitoring. Technology is a key component of our IFRS implementation program. There are predictive models in place for persistency management, and we are attempting to use predictive models in other areas as well.
What are the top 3 areas where you see AI & ML making a difference in actuarial work?
Technology is a big enabler for all actuarial work, and its criticality in timely and accurate financial reporting cannot be undermined. With the advent of new complex reporting frameworks such as IFRS, a very high degree of integration and automation through additional tools is necessary to produce comprehensive and well-understood results in a timely and accurate manner.
Globally in insurance, discussions are gravitating towards the hyper-personalization of insurance offerings and embedded insurance in digital ecosystems. AI and ML will play a critical role in the success of such new-age solutions which are dynamic.
AI and ML models are used for critical activities in insurance companies such as fraud prediction which has a direct impact on pricing, and even to improve sales effectiveness by predicting the propensity to buy.
Owing to expanding databases and the need to use these effectively, data science and machine learning are emerging as necessary skills for actuaries of the future.
