In previous posts, we’ve addressed how a modernized actuarial platform can push organizations towards an optimal reserving process. We’ve discussed the benefits of having a reserving database, owned and operated by the actuaries, the benefits of modern reporting tools, and the benefit of automating as much of the reserving process as possible.

These are all key components of the reserving actuary’s job, that each take steps towards addressing the increased regulatory demands and the change in the economic and technological world in which finance professionals operate. However, all the automation capabilities, the fancy graphical reporting technology, and the granular claim level loss database don’t change the fact that at the end of the day actuaries need tools that allow them to work confidently and efficiently.

While many reserving actuaries still do most of their work in spreadsheets, it’s long past time to acknowledge the many risks associated with spreadsheets. Even the most locked-down spreadsheets, expertly built by Excel wizards, still needs to be updated every quarter. The files still need to communicate with one another, and they still need to allow for actuarial judgment and expertise to override predetermined calculations. And often when the critical person who built the spreadsheet or wrote the underlying—and often opaque—VBA code moves on or is unavailable, companies can be exposed to "key-person risk," crippling confidence in the numbers and in the team’s efficiency.

Reserving actuaries deserve better tools

• Tools to help them be more efficient and to get them the information they need faster and more reliably, allowing them to focus more deeply and in the areas that can provide the most value to their organizations.
• Tools that allow actuaries to exercise their expertise and judgment – rich data diagnostics, collections of industry-standard reserving methods to choose from, quick algorithmic techniques such as curve fitting and factor smoothing, flexible data visualizations to depict variability and trends, and the ability to drill into more granular levels of data.
• Tools that are not a chore to maintain, update, and enhance–reserving departments spend way too much time honing their tools, and not enough time using them.

Like spreadsheets, purpose-built reserving solutions easily support the work of the sophisticated reserving team, while also reducing or eliminating the inherent risks that spreadsheets present and adding efficiencies to all phases of the actuarial analysis process. In addition, such tools provide users with the built-in ideas, suggestions, and experience of perhaps hundreds of additional users, benefits that most spreadsheet analyses could never take advantage of.

Modern reserving software provides a multitude of ready-made exhibits, diagnostics, and reports for users to immediately begin taking advantage of. They typically encompass a broad array of analytical tools that most actuaries don’t have the time or in-house resources to build for themselves. New users should find ready access to exhibits and methods that can reveal new insights while requiring little or no additional effort on their part. The reserving team should be able to focus on better and deeper analysis rather than building, editing, and maintaining spreadsheets.

At the same time, modern reserving software cannot and should not be an “actuary-in-a-box.” Exhibits and reports must be easy to create and modify to meet users’ unique analysis needs. They must be easily scalable to all shapes and sizes of reserving projects, and easily sharable with other team members, making everyone more efficient. A modern actuarial platform should track (and if necessary, restrict) who can update the custom calculations, and should keep links and calculations locked down once implemented. This effectively eliminates almost all spreadsheet risk, together with much of the human error that can creep into even the best managed spreadsheet processes.

Strong controls over both standard and customized calculations eliminate much of the technical review time required to assure the thousands of calculations in your old-fashioned spreadsheet aren’t broken by an errant click. Rather than spending the first part of the reserving process updating the templates, analysts and their managers can focus on the analysis itself, with mindful resource allocation enabled by no longer having to untangle the web of spreadsheets, only to re-tangle them again for the next analysis.

But what about the analysis itself? Eliminating inefficiencies in spreadsheet management only works if the actuaries can still work in a tool set that allows them to do their job effectively. They’ll need an environment that easily caters to their needs. The analyst performing initial selections needs different information than the manager reviewing those selections. The personal lines team may need different methodologies than the commercial lines team. The modern reserving tool needs to allow teams, or even individuals, to organize the exhibits in a workflow that’s intuitive to them, while still keeping all of the methods (including the custom-built ones) on the shelf for near instant access.

The modern tool should allow actuaries to unravel the tangled web of calculations, allowing users to quickly trace every calculation back to the original data source or underlying assumption. The tool needs to provide graphical representations of key assumptions and results, which not only help identify trends and trouble spots, but can also help focus users and managers. Reviewing hundreds of loss development factor selections across hundreds of projects is much easier with a series of graphs than a series of triangles. It’s easier to explain and exclude outliers if you can drill into the claim level detail to see what’s really driving the development.

Of course, to satisfy regulatory and audit requirements, decisions and selections need to be well documented and retained within the tool. Retention of this information has distinct benefits beyond the regulatory needs, quickly enabling comparison to previous periods with instant actual-vs-expected analyses, and further assisting managers in actuarial resource allocation.

Not only will a modern tool allow managers to allocate analysis resources more effectively, it enables the actuaries to be actuaries, and not spreadsheet jockeys. Imagine the time saved by not needing to update hundreds of spreadsheets and then having to perform technical reviews of all of them. Imagine the time saved by your actuarial staff having on-demand access to the methods and tools they need. Imagine your time saved by being able to see exactly what you need to see for a given set of projects, without navigating all over your network.

What are you or your team going to do with that extra time? The possibilities are almost endless. That time could be spent further optimizing the process by implementing a “drivers of change” analysis or an early warning system, even further reducing the time spent on unnecessary tasks. You could develop advanced management dashboards that tell the reserving story from start to finish, giving key stakeholders access to important business information early in the reserving process. You could review important information in the underlying claims data using new tools like machine learning to validate your reserve selections, optimally segment your data, and provide early identification of potential large claims for triage.

You can even look beyond the reserving team: expand your existing workflow to provide scenario analyses to assist with capital planning; provide early warning of inflation risk to assist the pricing team; or monitor changes in settlement time to help the claims team.

By replacing risky and time-consuming spreadsheets with modern reserving tools, the reserving team can evolve from its traditional reporting-focused role to one with a deep understanding of the loss function, adding value throughout the enterprise.

This article is part of a longer series on rethinking the loss reserving process. Read the next article here, or view the rest of the series here.

This article was originally published on April 19, 2022.