Actuarial training

Your actuarial training is not enough

By Insight.

How IT education can help actuaries stay relevant and support digital transformation


“If you want your actuarial digital transformation projects to succeed,” says Faheem Suban, CTO of Insight Actuaries and Consultants, “your actuaries need to become more comfortable with technology.”

And Suban should know.  As an actuary turned digital solutions architect, he has witnessed first-hand that the road to digital transformation failure is paved with good intentions and littered with buzz words. “Future-proof, synergies, digitisation, automation…these are all great concepts,” he says, “but why are most insurers still so far away from that holy grail that is the ‘one-click valuation’?”


Darwinism and digital disruption

The conversation around transforming systems and processes to free up actuaries’ time and enable them to focus on value-adding analysis has been circulating – often actually going in circles – for years.  A thousand transformation projects have been launched based on recognition of the need to streamline clunky models and modernise legacy systems, spurred on by waves of regulatory and accounting reform.  But transformation projects – actuarial and other – notoriously fail more often than they succeed[1].  Examining the reasons for this is a research field on its own, but Suban is convinced that the secret to transforming actuarial teams in a sustainable way is for actuaries to better understand technology.

Improved technology skills will allow actuaries and hence insurance companies to compete in an industry that is ripe for disruption.  This view is bolstered by the findings of McKinsey’s Digital disruption in insurance: Cutting through the noise report: “With competitive landscapes changing fast, it can be hard to know just how digital technology will play out, and hence where to place big bets. Yet hesitation is not an option. In insurance, as in other industries that have felt the force of digital disruption, those that move fastest to adapt are likely to take a disproportionate share of the profits.”

Embracing technology and digitisation is not just good for profits and projects, it may even be the key to ensuring that actuaries remain relevant in a digital world.  Quoting Charles Darwin, Suban believes that “It is not the strongest of the species that survives, nor the most intelligent that survives. It is the one that is the most adaptable to change.”


The faster horse fallacy

There is a famous quote (somewhat dubiously) attributed to Henry Ford: “If I had asked people what they wanted, they would have said faster horses.” Customers can easily describe a challenge they’re facing – in the case of the horse, wanting to get somewhere faster – but not the best solution: a car.  We don’t know how to ask for solutions that are beyond the realms of our current knowledge and understanding.  Suban applies this scenario to actuaries seeking transformation in a world where keeping up with technological progress is a full-time job.  What distinguishes this case from the horse analogy is that the car – fit-for-purpose technology – already exists.  Actuaries just need to be made aware of that and know which kind of car to ask for. “Actuaries want better systems, models and processes without knowing the full extent of what is possible.  They may, for example, ask for more processing power in the hope of improving model runtime.  What they may not realise is that those models might be improved to function much more efficiently, either on the same or a more suitable platform.”  He believes that many actuaries do not know what is available and therefore are unable to articulate what changes they want to see in their teams.  “We’re not looking to reinvent the wheel,” he continues.  “There are many examples of mature open-source data projects that can be looked to for guidance.  Know when to stand on the shoulders of giants.”



The art of the possible

“Start small,” Suban advises.  “Send your actuaries on a coding course and work through case studies of how new thinking and technology have been used to improve performance.  There are countless affordable resources online: open-source projects, free courses, YouTube videos…without too much effort you can enable yourself to use the technologies that do everything you need.”

He also advocates teaching actuaries about data analytics and business intelligence.  “Once they know how to code, empower them to build their own dashboards and destroy the black box of financial reporting.  The only way to extinguish the mystery around where results come from is to build the reports yourself, from the ground up.”

In this way, actuaries will start to understand what is possible and how to fix processes themselves instead of having to rely on other people to advise and implement.  “And choose a general purpose language that can be applied to other products, practice areas, even industries,” he continues.  “Only having expertise in a system that exclusively models life insurance cash flows, for example, really restricts one’s way of thinking and career options.”  The benefit of learning a high-level or general purpose language, such as Python, Java or C++, is that, architecturally, they are all very similar.  Once the general logic and programming paradigm is understood, it can, relatively easily, be transferred to other languages.

In addition, actuaries will no longer need to push Excel to solve problems it was not designed to handle.  “When all you have is a hammer,” Suban says, “every problem looks like a nail.  Excel is a revolutionary tool with countless applications.  Processing and analysing large volumes of data is not one of them.”


Actuaries: the original data scientists

Suban feels that actuaries should not have to rely on data scientists to produce reports because actuaries are the original data scientists.  “Data has been the bread and butter of actuarial work since Edmond Halley compiled the first life table – actuaries have always been data scientists and good technical actuaries thoroughly know their data.” 

What has changed though, is the volume, variety and velocity of data and the availability of technology platforms to handle it.  “This recognition of the value of data has given rise to the label ‘data scientist’: someone who harnesses technology to derive insights from unprecedented volumes of data.  Actuaries can do that work – no-one has better knowledge of the life, health and other insurance domains than they do.  They just need to keep up to date with technological developments and become familiar with data tools and platforms so that they no longer need to rely on first principles and Excel alone.”  Not doing this, Suban cautions, may result in actuaries losing their status as experts in data analysis.  “On a personal level, I have had to hire applied mathematicians and engineers because I struggle to find actuaries with the skills I need.”

When asked why actuaries are reluctant to lean into the data science domain even though they have the skills required to do so, Suban is stumped.  “Perhaps it’s a comfort zone thing, a fear of the (perceived) unknown…Or maybe they just don’t have the time.  Ironically the lack of time is often due to the use of inefficient technologies.”  A lack of time caused by unsuitable tech, but no capacity to improve the tech because there’s not time.  And so the cycle continues.


How to make the time

Actuarial managers may lament that the reason we need transformation in the first place is that actuaries do not have time to focus on the things that matter.  How, then, will they find the time to learn and implement these new skills? 

Suban urges teams to see the cost as an investment. “Whichever way you look at it,” he says, “transformation programmes are expensive.  Why not invest in your team and its future?  Actuaries have always been self-taught coders to some extent: the knowledge required to build complex cash flow models in purpose-built modelling software or even Excel is often learnt on the job.  Buy time for your actuaries to learn about the latest technology and technological thinking, instil some IT rigour and good habits and then have them go through the exercise of applying their learnings to their own products.” 

Suban quotes Eskom CEO, Andre de Ruyter’s bicycle analogy: “we’re a bit like a cyclist running next to the bicycle with the chain off, and when asked why we say, well we’re too busy to put the chain back on.  So we do need to slow down, put the chain back on, so that we can get back into the saddle and start pedalling again.”


The role of the IT team

It is important to make the distinction between the IT department and actuarial teams; the former still have a crucial role to play.  Actuaries should be solving business problems; they do not need to be network administrators, infrastructure experts or cyber security gurus.  But they need to become solutions architects to solve the business problems they face.  To get there, all they need, as a start, is a solid understanding of high-level coding languages and data management systems.  The approximate manifestation of Moore’s Law (that the number of transistors on a microchip doubles about every two years, though the cost of computers is halved) has meant that the technology now allows anyone with a solid understanding of high-level languages to work with big data.  This ability will allow actuaries, themselves, to help their companies meet their strategic objectives, without having to rely on the IT department to work with the data.

So, what is the role of an insurer’s IT department?  “Actuaries will not need to be involved in the administration of an IT environment,” Suban believes.  “They will have a better understanding of the elements thereof, but the IT department will still manage the functional aspects of policyholder administration systems, cloud services and other non-end-user applications.”  In fact, empowering actuaries to understand and own their models and systems may assist in improving the relationship between the two departments.  “Actuaries will understand the IT world and be able to better communicate with the IT team.  The latter will enjoy the release of capacity that will result from fewer requests from the business.  And we’ll be able to eliminate that layer of cost that exists due to the need for interpreters between the two.  The business analysts can go.”


Back to Redington

The idea of upskilling actuaries to maintain relevance is nothing new: Frank Redington himself famously said that “An actuary who is only an actuary is not an actuary.”  In an environment that increasingly values data expertise and IT skills, it is in actuaries’ best interests to proactively combine the solid base of their traditional actuarial training with the power of technology. |

[1]  John P Kotter’s paper, Leading Change: Why Transformation Efforts Fail, in the 1995 Harvard Business Review is still a respected exploration of this topic.  There are numerous more recent papers looking at transformation failure in the contexts of digital transformation, lean transformation and other angles.




Share this post