A transformative shift in pricing is almost tangible—embracing new technologies and data analytics can help insurers in creating long-term value generation. McKinsey spoke with Elena Pizzocaro, a partner in the Milan office, to understand more about pricing and current challenges and opportunities.
McKinsey: What are the latest pricing imperatives for insurers?
Elena: Pricing can be the source of a sustained competitive advantage and a primary differentiator for long-term value generation in property and casualty insurance (P&C). Profitable P&C insurers emphasize pricing innovation and underwriting excellence—and invest in them.
We have recently investigated multiple pricing transformations at global insurers and have codified different possible archetypes reflecting carriers’ maturity level and the strategic priority attributed to pricing. Most frequent archetypes include:
- Consistent application of generalized linear models (GLMs) with the highest focus on risk modeling
- Use of AI-based and machine-learning pricing tools
- Implementation of robo-pricing for enhancing market understanding and transparency, and adjusting prices automatically and dynamically (subject to regulations)
A full-scale pricing transformation focuses on building the right infrastructure to achieve substantial and sustainable improvements across the entire pricing value chain, including underwriting strategies and risk selection, technical pricing, market-based and behavioral pricing—where it’s allowed by regulations. Pricing execution and governance are equally important. Key enablers here are organization and talent, as well as advanced analytics and digital capabilities.
Embracing artificial intelligence (AI) throughout the pricing value chain means conquering the ability to recalibrate models in any selectable frequency and transforming the pricing team operating model.
McKinsey: Which opportunities does a full-scale pricing transformation offer to insurers?
Elena: The opportunities for more sophisticated pricing and immediate P&L impact have never been better. Pursuing pricing sophistication can enable transformative shifts for insurers towards advanced analytics, automation, new data sources, and the ability to rapidly react to changing market environments.
Pragmatically speaking, embracing artificial intelligence (AI) throughout the pricing value chain means conquering the ability to recalibrate models in any selectable frequency and transforming the pricing team operating model. Actuaries can then spend most of their time improving data quality and identifying new sources of data, resulting in top management easily understanding the pricing challenges due to easy visualization and augmented intelligence.
Insurers should not refrain from experimenting with advanced techniques, and they should push the knowledge frontier further. For example, the application of deep learning—a subset of machine learning techniques—allows the processing of rich and complex data sets, including images and raw machine data.
McKinsey: How can insurers reconcile pricing sophistication with a need for “explainability”?
Elena: There is an increasing demand for the explainability of models and an easier way to access and understand models, even from a regulators’ point of view. Complex models may, at some point, be seen as less transparent. Traditionally, this would imply a trade-off between model performance and explainability, and would result in prioritizing or eliminating certain types of models from the analysis.
There is an increasing demand for the explainability of models and an easier way to access and understand models, even from a regulators’ point of view.
Nowadays, state-of-the-art methods aim at explaining each sample post-hoc, effectively overcoming the trade-off between interpretability of performance. In particular, “Explainable AI” (XAI) focuses on the development of post-hoc explanatory methods that can be applied to black box machine learning models.
This typically involves the creation of additional explanation algorithms which augment the machine learning model, providing insight into how certain predictions or outputs are reached. If high prediction accuracy is paramount, using a black box model and explaining it after training and optimization with the explanation algorithms is the best-performing approach. This way, AI is not a science for data scientists only, but becomes intelligible and challengeable to a larger group, including insurers, internal and external stakeholders, actuaries, product developers, commercial strategists, risk managers, and intermediaries.
Elena Pizzocaro is a partner in the Milan office.
For more on pricing in insurance, see: