Bordereaux have a well-deserved reputation for being difficult to work with. Constant formatting changes, inconsistent column labels, different conventions for data layout, and the sheer heterogeneity of casualty risks all make it challenging to convert bordereaux data into a single, consistent format.

Our bordereaux processing pipeline is designed to combine the best aspects of three distinct approaches to provide a fast, robust, and high-quality process:

• AI assistance –can help remove much of the drudgery from semantic labeling of policy and claims data. Intelligent suggestions for the interpretation of common column labels can greatly reduce the amount of time required to set up a new bordereaux format.
• Human expertise – helps to maintain a high level of quality. LLMs are not yet foolproof, and no one in their right mind would let an AI prepare their corporate financials. By keeping humans in the loop, we eliminate the risk of silly mistakes from blindly trusting a useful but imperfect LLM.
• Rule-based semantic validation – helps ensure that an interpretation of bordereaux data is logically consistent. Is the "total incurred" column in a claims bordereaux gross or net of salvage and subrogation? If we think it's net, we can check by summing the paid loss, loss reserve, and recovery columns and ensuring that the total is always the same as "total incurred". Our advanced engine automatically identifies such redundant information in bordereaux files and ensures that data labels are internally consistent.

Collateral Calculation

Compared to traditional reinsurance, collateralized reinsurance has the added complexity of ensuring that sufficient collateral is held at all times per the terms of the reinsurance agreement. We have years of experience in designing, implementing and applying complex collateral structures.

Our financial modeling engine can handle the intricacies of collateral calculations, while producing clear and intelligible outputs that help all parties to a transaction understand the fundamental reasons behind changes in collateral requirements over time.

Financial Statements

Translating premium and losses into a P&L for a collateralized reinsurer or casualty ILS investor is far from trivial. On top of the challenges posed by data aggregation, application of contract terms, and collateral calculations, one must also contend with accounting for investment returns, brokerage fees, and other associated administrative expenses. Our powerful and flexible is able to automate this work and integrate with the rest of your accounting workflow. Whether you prefer to view outputs in Excel or directly import journal entries into your accounting system, we can supply exactly what you need.

Basic dashboards can tell you what has happened with a book of business, but not what is going to happen or why it's happening. Our team of industry veterans, data scientists, and visualization experts has carefully crafted a rich suite of policy and claim data visualizations designed to help you answer these deeper questions faster.

• Premium – How far ahead or behind plan is the underwriting team on writing premium? How do cancellations, endorsements, and audit transactions affect premium after policies are written? What is the net effect of these amendments? What's the risk that an underwriter will exceed their premium cap?
• Claims – What does the distribution of claim severity look like? Are losses on this book of business driven more by frequency or severity compared to its peers?
• Large Losses – What's the typical reporting lag for large losses? How long does it take for case reserves on large losses to be increased beyond the large-loss threshold? Once reserves are raised, do the case reserves tend to be optimistic or pessimistic?
• Loss Development – What do paid and reported loss development patterns look like for this book of business? Do losses tend to develop more or less quickly compared to other programs in the same line of business?