Risk and insurance managers need to assess a diverse range of risk. Fortunately, there are plenty of tools available to analyze and predict the risk, one of which is loss triangles. Loss triangles are used to analyze historic risk data and make predictions regarding the future costs of claims. Here’s a breakdown of what loss triangles are, why they’re important, and how to use them.

What Is a Loss Triangle?

A loss triangle is a triangle-shaped chart that shows loss data spread out over different periods. With every loss triangle, there are rows and columns. The rows represent the accident year, which is the year in which an incident occurred.

The columns refer to the maturity, or development, period. The maturity may be represented in months, such as 12, 24, 36, etc. This number tells you how many months have passed since the start of the accident year.

The cells in the loss triangle contain the loss data you want to analyze. For example, suppose you want to see how the total amount you’ve paid out in claims changes as time progresses. Each cell may contain a number representing the total paid in claims a certain number of months (the maturity period) after the start of each accident year.

For instance, suppose you want to compare the amount of loss paid out after two years for 2019 and 2020. You would look at the 24-month maturity column and see what it is for the 2019 row and the 2020 row.

Now, let’s say that the number is $900,000 for 2019 and $950,000 for 2020. That means that the amount you pay out in losses two years after the accident year is $50,000 higher for incidents that occurred in 2020 than those that happened in 2019.

Why Loss Triangles Are Important

Loss triangles are important because they give you a quick view of how losses mature over time—and for specific years. They make it very easy to spot trends, particularly because they unify data across several years and present it in a single chart.

For insurers and companies with risk management information systems, loss triangles can make the difference between allocating just enough money to cover losses and not enough. They can also identify situations where you’re charging too much for certain types of coverage.

Who Uses Loss Triangles?

Insurance companies, risk management, and mitigation teams, and actuaries use loss triangles to spot trends and predict future expenses. They help decision-makers decide how to arrange resources and charge for services. Loss triangles also come in handy for statisticians looking for trends that could indicate room for improvement in a claim management system.

Challenges with Loss Triangles

Loss triangles aren’t without their challenges, particularly when it comes to the quality of data and DATA SILOES.

Poor Data Quality

If data is accurate, in any way, it can render a loss triangle useless and/or misleading. For example, if some departments don’t report all of their claims data, the figures for a specific year and maturity period won’t reflect what’s actually going on.

Siloed data.

Data siloes can also put a wrench in an otherwise strong loss triangle calculation. For instance, suppose a decision-maker is trying to figure out how much to charge for cybersecurity insurance. They have cyber attack data from manufacturers they insure, but the department that handles cyber claims for financial companies hasn’t provided their information. There’s simply no way to set up a comprehensive set of loss triangles because they need all cyber loss data, regardless of which business sector it applies to.

Different Variables to Use for Loss Triangles

The variables you can use in your loss triangles are nearly unlimited, but here are some that many companies find useful:

  • Geographic data. You can track loss data and payment history according to different locations, such as towns on the coast and those at least 20 miles inland.
  • The kind of loss incurred. Loss triangles can be helpful when figuring out the loss history of claims related to fire as opposed to flooding, or data center failures as opposed to power outages.
  • The year in which policies originated. A company looking to improve its claims management system may use a loss triangle to see if newer ones do a better job of meeting loss goals than older ones.
  • The type of business being insured. You could use loss triangles to see how losses are paid out over time for automotive companies versus professional service providers. This could surface data that may justify increasing marketing efforts for certain kinds of businesses.

Benefits of Loss Triangles

Loss triangles benefit any person or team that needs to see how loss data evolves. They’re especially helpful for actuaries, who use them to estimate the kinds of risks a business is most likely to encounter and how to allocate resources accordingly.

For instance, by using loss triangles, an actuary can help their company decide which lines of insurance have the highest loss overhead, which business sectors present the most risk, or how the speed at which claims get settled varies over time.

Loss triangles also streamline the process of figuring out the ultimate cost of losses incurred during different years. This makes them good tools for analyzing the costs you expected to incur versus what you ended up having to pay.

Armed with a loss triangle, you can also decide how to set your reserve amounts. For example, suppose personal injury losses tend to get settled at 85% by the end of 60 months. You may want to apportion more reserves for these kinds of incidents than you would for natural disasters, which may get settled within a year or less.

Put Loss Data to Work with Riskonnect’s Integrated Risk Management Information System

Riskonnect’s Integrated Risk Management (IRM) solution provides you with a full transactional history that you can use to develop accurate loss triangles. Because you can centralize all of your data under Riskonnect’s single umbrella, you can say goodbye to siloes that make actionable info hard to come by. To see how Riskonnect can position you to make more effective decisions, SET UP A DEMO TODAY.