August 25, 2015
The 2 New Realities Because of Big Data
by Bret Shroyer
The bad news because of big data: If you have a profitable niche, everyone else will soon find it. The good news: You can still win.
I have some bad news. There are no longer any easy or obvious niches of sustained, guaranteed profits in insurance. In today’s environment of big data and analytics, all the easy wins are too quickly identified, targeted and brought back to par. If you’ve found a profitable niche, be aware that the rest of the industry is looking and will eventually find it, too.
Why? The industry has simply gotten very good at knowing what it insures and being able to effectively price to risk.
Once upon a time, it was sufficient to rely on basic historical data to identify profitable segments. Loss ratio is lower for small risks in Wisconsin? Let’s target those. Today, however, all of these “obvious” wins stand out like beacons in the darkness.
To win in a game where the players have access to big data and advanced analytics, carriers should consider two new realities:
- You can’t count on finding easy opportunities down intuitive paths. If it’s easy and intuitive, you can bet that everyone else will eventually find it, too.
- Sustainable opportunities lie in embracing the non-obvious and the counter-intuitive: finding multivariate relationships between variables, using data from novel sources and incorporating information from other coverages.
Just knowing what you insure is only the start. The big trick is putting new information to good use. How can carriers translate information on these new opportunities into action? In particular, how can carriers better price to risk?
We see two general strategies that carriers are using in pricing to risk:
- Put risks into categories based on predicted profitability level
- Put risks into categories based on predicted loss
The difference appears subtle at first glance. Which approach a given carrier will take is driven by its ability to employ flexible pricing. As we will now explore, it’s possible for carriers to implement risk-based pricing in both price-constrained and flexible-rate environments.
Predicting Profitability: Triage Model
In the first strategy, carriers evaluate their ability to profitably write a risk using their current pricing structure. This strategy often prevails where there are constraints on pricing flexibility, such as regulatory constraints, and it allows a carrier to price to risk, even when the market-facing price on any given risk is fixed.
The most common application here is a true triage model: Use the predicted profitability on a single risk to determine appetite. Often, the carrier will translate a model score to a “red/yellow/green” score that the underwriter (or automated system) uses to guide her evaluation of whether the risk fits the appetite. The triage model is used to shut off the flow of unprofitable business by simply refusing to offer coverage at prices below the level of profitability.
A triage model can also be implemented as an agency-facing tool. When agents get an indication (red/yellow/green again), they start to learn what the carrier’s appetite will be and are more likely to send only business that fits the appetite. This approach has the added benefit of reducing underwriting time and expense for the carrier; the decline rate drops, and the bind/quote rate rises when the agents have more visibility into carrier appetite.
A final application carriers are using is in overall account evaluation. It may be that a carrier has little or no flexibility on workers’ compensation prices, but significant pricing flexibility on pricing for the business owners policy (BOP) cover. By knowing exactly how profitable (or unprofitable) the WC policy will be at current rates, the carrier can adjust price on the BOP side to bring the entire account to target profitability.
Predicting Loss: Pricing Model
If a carrier has pricing flexibility, pricing to risk is more straightforward: Simply adjust price on a per-risk basis. That said, there are still several viable approaches to individual risk pricing. Regardless of approach, one of the key problems these carriers must address is the disruption that inevitably follows any new approach to pricing, particularly on renewal business.
The first, and least disruptive, approach is to use a pricing model exclusively on new business opportunities. This allows the carrier to effectively act as a sniper and take over-priced business from competitors. This is the strategy employed by several of the big personal auto carriers in their “switch to us and save 12%” campaigns. Here we see “know what you insure” being played out in living color; carriers are betting that their models are better able to identify good risks, and offer better prices, than the pricing models employed by the rest of the market.
Second, carriers can price to risk by employing a more granular rate structure. This is sometimes referred to as “tiering” – the model helps define different levels of loss potential, and those varying levels are reflected in a multi-tiered rate plan. One key advantage here is that this might open some new markets and opportunities not in better risks, but in higher-risk categories. By offering coverage for these higher-cost risks, at higher rates, the carrier can still maintain profitability.
Finally, there is the most dramatic and potentially most disruptive strategy: pricing every piece of new and renewal business to risk. This is sometimes called re-underwriting the book. Here, the carrier is putting a lot of faith in the new model to correctly identify risk and identify the correct price for all risks. It’s very common in this scenario for the carrier to place caps on a single-year price change. For example, there may be renewals that are indicated at +35% rate, but annual change will be limited to +10%. Alternatively, carriers may not take price at all on renewal accounts, unless there are exposure changes or losses on the expiring policy.
Know What You Insure
Ultimately, the winners in the insurance space are the carriers that best know what they insure. Fortunately, in an environment where big data is becoming more available, and more advanced analytics are being employed, it’s now possible for most carriers to acquire this knowledge. Whether they’re using this knowledge in building strategy, smarter underwriting or pricing to risk, the results are the same: consistent profitability.
Sometimes there are pricing constraints that would, at first glance, make effectively pricing to risk challenging. As we have discussed, there are still some viable approaches for carriers facing price inflexibility. Even for carriers with unlimited price flexibility, pricing to risk isn’t as easy as simply applying a model rate to each account; insurers must take care to avoid unnecessary price disruption. We’ve discussed several approaches here, as well.
Effectively pricing to risk gives carriers the opportunity to win without relying on protecting a secret, profitable niche. In the end, this will give them the ability to profit in multiple markets and multiple niches across the entire spectrum of risk quality.