After the Flames: How Exposure Management Responds to Major Events
Key results
Reduced exposure analysis time, enabling swift responses during peak periods.
Delivered real-time analytics to validate assumptions and align exposures with risk appetite.
Improved ability to analyze year-on-year changes and assess profitability with actionable insights.
In the aftermath of a major catastrophe, exposure management (EM) professionals face a dual challenge: deliver a rapid, credible view of potential losses, and remain grounded despite the uncertainties that swirl around incomplete data and a shifting event footprint.
Nowhere was this more apparent than in the response to the recent US wildfires: a complex, high-value loss that stretched across multiple classes and tested EM processes.
At the Allphins April 2025 roundtable, Laurent De La Porte, Allphins CEO, was joined by senior reinsurance professionals to reflect on how EM performed under pressure - and what it revealed about the industry's evolving approach to post-event analytics.
Known Unknowns: Data Complexity and Loss Uncertainty
One of the clearest takeaways from the wildfire response was just how much the usual models still applied perfectly to the losses, but still somehow failed to adequately predict or quantify them. "The property damage losses are well quantified by models," said Mathias Borjesson, SVP Underwriting at RenaissanceRe. "As an industry, we understand buildings and standard replacement costs. But the costs associated with putting wealthy people up in alternative accommodation for several years is both unpredictable and potentially significant. Then, there is potential for high value contents that are unknown or undeclared. I think our traditional methods will hold up well for the well-known portion of the loss. But there is an amplifier specific to this event, which is difficult to quantify from data alone."
James Simpson of Blenheim Syndicate agreed: "Anecdotally, a house 50 metres from the fires cost $5 million to rebuild. But one painting inside was reportedly worth $200 million. That's the kind of asset value that can completely change the profile of a loss. Plus, should an artwork of that calibre be covered by property insurance rather than a specialist art policy?"
These high-value, non-traditional exposures often reside in portfolios where data is thin, contractual clauses are varied, and risk transfer pathways are complex. Vanessa Jones, Head of Exposure Management at Dale Underwriting Partners, noted the limitations of modelling for such risks. "We don't know what's been done in terms of risk management. Was the painting even in that house? Was it moved to another mansion? We have no idea! And even the value of a painting - the value is set by whoever is ready to buy it, so it's really set arbitrarily by its last auction. These are questions that models can't answer."
And then, of course, there's the environment in which a court may decide the outcome. Borjesson cautioned: "California is a difficult place for this because of the very favourable legal environment for the claimants." Cross-class events are particularly challenging because they demand deeper analysis and don't necessarily play out well in a court minded to have sympathy for the claimant.
The Tension between Speed and Accuracy in a media-led world
Reinsurers are under pressure to quantify their losses quickly. De La Porte noted that Allphins' data insights are often used as much for operational reporting as for portfolio analysis. After an event, investors, boards, and capital providers want numbers—often before cedants themselves have completed their assessments. But this need for speed creates tension with the realities of data availability and accuracy.
"There's a lot of value in being able to come up with a range quickly," said Jones. "Not an exact number—but a range with confidence bounds. That allows you to respond to internal stakeholders, capital providers, and of course the bigger picture of reputation management."
Simpson caveats, "We go slow. We do the analysis quickly, but we take our time validating it. We don't have many external pressures forcing us to publish numbers, and frankly, that's a good thing. It's better to be right than fast." This divergence in pace often reflects ownership structure. "If you're reporting loss estimates externally," Borjesson noted, "you have to have capabilities for a one-to-three-day turnaround. We've had events that hit on the last day of the quarter," Borjesson continues. "That's very challenging. Suddenly everyone's looking at you for a number that might take weeks, and sometimes months or even years, to be fully confident in."
The Role of EM in Portfolio adjustment
One of other critical functions of exposure management post-event is in steering portfolio adjustments. This alone is a reason EM analysts are under pressure to report rapidly. Jones said, "You've got to make decisions about your portfolio going forward. So you need to assess your strategy for the remainder of the year given an event has occurred. You may also have your retro coming up in the case of the property treaty book. With most events, we will immediately have the market asking questions of us around our loss estimates. The sooner you can answer that, the better. And the more confidently you can answer that, the better."
If a loss reveals a concentration or unexpected exposure, reinsurers must decide whether to reduce, rebalance, or retreat. Jones again: "Events like the wildfires directly impact portfolio strategy. We run multiple pro forma portfolios throughout the year. When something like this happens, we adjust them. Where are we still competitive? Where are others retreating? Where are the new risks showing up?"
This is where exposure management must shift from passive monitoring to active portfolio management—if the systems and relationships are in place to support it.
Our guest commentators concluded with the fact that the California wildfires endorsed their fundamental positions: firstly that EM has particular value as a rapid-response source of insight in complex events like this; and secondly that EM is, whilst certainly a key tool, one of many in the reinsurer's armoury. Blenheim's Simpson: "It's what I've always said: don't rely on one thing. We use aggregates; we use probabilistic - a range of techniques to get a good rough picture. And we try not to focus on a 'magic number' from exposure management, instead it's analytical. Our job is to give underwriters the tools to help them work towards the number with confidence."
After the Flames: How Exposure Management Responds to Major Events
In the aftermath of a major catastrophe, exposure management (EM) professionals face a dual challenge: deliver a rapid, credible view of potential losses, and remain grounded despite the uncertainties that swirl around incomplete data and a shifting event footprint.
Nowhere was this more apparent than in the response to the recent US wildfires: a complex, high-value loss that stretched across multiple classes and tested EM processes.
At the Allphins April 2025 roundtable, Laurent De La Porte, Allphins CEO, was joined by senior reinsurance professionals to reflect on how EM performed under pressure - and what it revealed about the industry's evolving approach to post-event analytics.
Known Unknowns: Data Complexity and Loss Uncertainty
One of the clearest takeaways from the wildfire response was just how much the usual models still applied perfectly to the losses, but still somehow failed to adequately predict or quantify them. "The property damage losses are well quantified by models," said Mathias Borjesson, SVP Underwriting at RenaissanceRe. "As an industry, we understand buildings and standard replacement costs. But the costs associated with putting wealthy people up in alternative accommodation for several years is both unpredictable and potentially significant. Then, there is potential for high value contents that are unknown or undeclared. I think our traditional methods will hold up well for the well-known portion of the loss. But there is an amplifier specific to this event, which is difficult to quantify from data alone."
James Simpson of Blenheim Syndicate agreed: "Anecdotally, a house 50 metres from the fires cost $5 million to rebuild. But one painting inside was reportedly worth $200 million. That's the kind of asset value that can completely change the profile of a loss. Plus, should an artwork of that calibre be covered by property insurance rather than a specialist art policy?"
These high-value, non-traditional exposures often reside in portfolios where data is thin, contractual clauses are varied, and risk transfer pathways are complex. Vanessa Jones, Head of Exposure Management at Dale Underwriting Partners, noted the limitations of modelling for such risks. "We don't know what's been done in terms of risk management. Was the painting even in that house? Was it moved to another mansion? We have no idea! And even the value of a painting - the value is set by whoever is ready to buy it, so it's really set arbitrarily by its last auction. These are questions that models can't answer."
And then, of course, there's the environment in which a court may decide the outcome. Borjesson cautioned: "California is a difficult place for this because of the very favourable legal environment for the claimants." Cross-class events are particularly challenging because they demand deeper analysis and don't necessarily play out well in a court minded to have sympathy for the claimant.
The Tension between Speed and Accuracy in a media-led world
Reinsurers are under pressure to quantify their losses quickly. De La Porte noted that Allphins' data insights are often used as much for operational reporting as for portfolio analysis. After an event, investors, boards, and capital providers want numbers—often before cedants themselves have completed their assessments. But this need for speed creates tension with the realities of data availability and accuracy.
"There's a lot of value in being able to come up with a range quickly," said Jones. "Not an exact number—but a range with confidence bounds. That allows you to respond to internal stakeholders, capital providers, and of course the bigger picture of reputation management."
Simpson caveats, "We go slow. We do the analysis quickly, but we take our time validating it. We don't have many external pressures forcing us to publish numbers, and frankly, that's a good thing. It's better to be right than fast." This divergence in pace often reflects ownership structure. "If you're reporting loss estimates externally," Borjesson noted, "you have to have capabilities for a one-to-three-day turnaround. We've had events that hit on the last day of the quarter," Borjesson continues. "That's very challenging. Suddenly everyone's looking at you for a number that might take weeks, and sometimes months or even years, to be fully confident in."
The Role of EM in Portfolio adjustment
One of other critical functions of exposure management post-event is in steering portfolio adjustments. This alone is a reason EM analysts are under pressure to report rapidly. Jones said, "You've got to make decisions about your portfolio going forward. So you need to assess your strategy for the remainder of the year given an event has occurred. You may also have your retro coming up in the case of the property treaty book. With most events, we will immediately have the market asking questions of us around our loss estimates. The sooner you can answer that, the better. And the more confidently you can answer that, the better."
If a loss reveals a concentration or unexpected exposure, reinsurers must decide whether to reduce, rebalance, or retreat. Jones again: "Events like the wildfires directly impact portfolio strategy. We run multiple pro forma portfolios throughout the year. When something like this happens, we adjust them. Where are we still competitive? Where are others retreating? Where are the new risks showing up?"
This is where exposure management must shift from passive monitoring to active portfolio management—if the systems and relationships are in place to support it.
Our guest commentators concluded with the fact that the California wildfires endorsed their fundamental positions: firstly that EM has particular value as a rapid-response source of insight in complex events like this; and secondly that EM is, whilst certainly a key tool, one of many in the reinsurer's armoury. Blenheim's Simpson: "It's what I've always said: don't rely on one thing. We use aggregates; we use probabilistic - a range of techniques to get a good rough picture. And we try not to focus on a 'magic number' from exposure management, instead it's analytical. Our job is to give underwriters the tools to help them work towards the number with confidence."