Making Risk Flow | The Future of Insurance
00:07:06 · Juan de Castro
[00:07:06] S0: And that is one of the reasons why I wanted to bring both of you into this episode. I think listening to your introductions, you talk a lot about insights, about predictive analytics, right? Which is, are things that probably often are not top of mind when people think about an actuary. Well, probably wrongly so, but reality. So again, we'll deep dive into that into a second. But perhaps following up with you, Jamie, let's start with what is an actuary? Like what is the role of an actuary in an insurance company? And I'm sure many people are familiar with it, but other listeners might not be that familiar.
00:07:44 · Jamie Wilson
[00:07:44] S2: Yeah. I'm never too offended when someone doesn't know what an actuary is. I feel like it's quite a niche job. But yeah, I guess one of the ways I'd describe it, so kind of maybe a little bit of history of actuaries really. So kind of like the actuarial profession really started in kind of life insurance and pensions. So kind of it was effectively this statistical study of when are you most likely to die? And like, how can I then like use that analysis to figure out what money do I have to hold in order to pay out for life insurance and things like that? So it kind of very much started in that space. And it was very much embedded on this idea of statistical analysis, but kind of starting in mortality. And then I think kind of over time, there was this realisation that actually, well, within general insurance, there's a similar need for kind of like statistical analysis to try and figure out, okay, I'm gonna have to pay out claims over a number of years. So therefore I need to know how much money I need to hold back for that. And so kind of you started to get the movement of actuaries into kind of the insurance world, but very much in that kind of reserving space. So very much in like a regulatory, how much money do I need to hold in order to pay out claims? And so kind of much more like that third line of second, third line of defense of, okay, well, once I've written the business, how much money do I need to hold? And then I think kind of what's happened over the last twenty years is that kind of the actuarial profession has kind of become more and more embedded in the kind of the insurance industry and moved away from just reserving. And now you kind of hear about capital modelling actuaries, reserving actuaries, then and crucially kind of the latest breed, let's call it, of pricing actuaries who are actually there, who are meant to be kind of like helping underwriters to figure out how they should be pricing business. And that obviously looks very different. A pricing actuary looks very different in retail insurance to kind of SME insurance to kind of large commercial specialty insurance. But kind of always, I think kind of coming from that background of like trying what you would hope that they're kind trying to apply statistical analysis to figure out some kind of outcome for the business.
00:09:47 · Juan de Castro
[00:09:47] S0: Okay, so it sounds like even within the actuarial teams, there's been quite a lot of evolution in the last few years. And again, we don't often talk about transformation of the kind of the actuarial teams. So Simon, like when you think about this evolution that Jamie has referred to, like how do you see the evolution of the actuaries in insurance companies?
00:10:09 · Simon Cooper-Williams
[00:10:09] S1: Yeah, I mean, think looking at what's happening in Markel, the demand from the business for data and analytics has just increased so much. So we talk about actuaries evolving in the insurance industry, but I see the underwriting part of the insurance industry has evolved for them to be a lot more demanding for that analytics. So, I mean, we have a program in Markel called Underwriter of the Future. That wasn't something there when we started. And a lot of that is around sort of technical capability of underwriting. So it outlines segmentation, it outlines expectations of portfolio management capability that various levels of underwriting divisions should be doing. So we've now got underwriters who are looking at these frameworks and saying, okay, well, I should be able to segment my book. I should be able to use pricing data to stitch on top of it. The people who are really well placed to do that are the actuaries who've been sort of been trying to do that for years, maybe on a different floor historically or somewhere else not sitting with them, but I think they've evolved with the underwriting business at the same time. So you've now got that demand and actuarial skill sets are very well placed to help with that. Like the bread and butter is slicing and dicing data, stitching things together, dealing with really messy sort of sets of data where they're not easy to analyze. It's not standardized. It's picking it apart. It's making sense of it. It's trying to put judgment on top of it as well. And I think that's really brought them to evolve with the underwriting function at the same time. And that's where I think where traditionally, actually, it may have been in more the sort of second line functions. You can see them in that first line now because they're there to help make the decisions on front line. I think that's where they can add a lot more value compared to historically.
00:11:54 · Juan de Castro
[00:11:54] S0: Yeah, and that's probably, I think, of the evolution also Jamie was referring to, right? So from being a reserving actuary where you probably need to work less day to day with underwriters to being more pricing actuaries, where, as you said, at the end it's about enabling underwriters to make the best decisions about the risks. You are mentioning there the kind of the evolution from what actuaries used to be more in the past, which was reserving actuaries, and you mentioned them being in a different floor, etcetera, to working closer together with underwriters. And I guess that also has meant that the actuaries themselves have have had to evolve and also like probably become more cross functional in the way they those pricing actuaries work. So I'm sure based on your kind of experience in the market, what type of actuarial teams do you see are most effective? And perhaps also on the other side, some examples of when they are not that effective.
00:12:52 · Simon Cooper-Williams
[00:12:52] S1: Yeah, really what I've seen after moving out of a traditional team is working with people from other professions across insurance. Back when I was in my earlier days in actuarial, a lot of what we were trying to do, we try and do a lot of data ourself. So we'd be using spreadsheets, we'd be using access databases. We were trying to force solutions through. A lot of it was wrangling data, but we didn't have a traditional training in that. What I've really seen is doing what I do now. In our team, we have data architects, data engineers. We have computer scientists on the HX side, actually. We've got software engineers as well. So when we've merged those together and we've done a small sort of cross functional team, they actually are really good at using the analytics part, but that data prep, data pipelining, stitching it all together, sourcing it from multiple sources, I think having worked in now a team with people who are from a different background and are able to do that, it's been sun lining. We've been able to run compared to walking before. To come to a couple of examples that we've done, just something as simple as our dashboarding solutions. Business intelligence has moved on so much in the last sort of five years compared to what we saw before. There's no more spreadsheet tools. We're using Power BI. That's completely transformed wholesale and international where it's on all the screens. You walk around the floor, everyone is using it. And we had actuarial people design the dashboards. They sort of knew what underwriters would like to see as an insight, or as a portfolio management tool if they're trying to segment their book or look for trends that aren't coming out from other avenues. And then working with the people who are able to source data and are able to structure it sensibly, we've gone from seeing projects where a dashboard may take a year to launching it in a month. And we've really just seen those skill sets come together and work together instead of passing work on and putting it to another team. And that's been absolutely transformational. I think my second thing is just around our portfolio managers that we've got in the underwriting division. I think a common frustration for me was, I would like to do that, but I don't have the data.