How can public settlement information be used to estimate exposure for my client?

Economic experts often employ two approaches that use public settlement information to estimate potential exposure for a client. The first approach is a “top-down” approach, which relies on analyst reports or a global settlement, and apportions a share of the total settlement to a single entity (the client). The second approach is a “bottom-up” approach that estimates statistically potential exposure based on the characteristics of prior settlements, specific data and assumptions about the client’s products and sales.

Top-down Approach

The top-down approach involves starting with public settlement information and/or analyst reports as a starting point to estimate the global liability exposure.  The next step is estimating the client-specific sales share and apportioning a share of the global exposure to the client based on the calculated sales share. Using relevant estimates of client sales and sales shares, the economic expert can develop calculations of exposure under different scenarios.  For example, different scenarios may include different estimates of market shares at either the local or global level, different assumptions about geographic or product markets, and assumptions about pass-through if the litigation involves downstream claimants.

Bottom-up Approach

If the number of available public settlements in similar matters is large enough, a bottom-up approach can be used to estimate exposure. In a bottom-up approach, the expert identifies influential factors in prior settlements and uses statistical models to estimate their relationships between those factors and the settlement amount. These relationships can be used for constructing a predictive regression model.  This process can be carried out in two steps:

1. Build a statistical model: The model takes into consideration prior settlements and judgements to draw correlations between the settlement amounts and factors that could influence the settlement. For example, the statistical model should likely predict that the higher the sales shares, the greater the settlement amount. Regression analysis is often used to model statistical relationships between the settlement amount and various factors such as: sales share, proceeding type, entity size and type, product characteristics, and other factors. Once the statistical model has been built and tested, it can then be used for prediction.
2. Predict client exposure: The estimated influence of each factor derived from the statistical model is then used to forecast the potential settlement amounts for the client.  The specific factors at issue in a particular litigation will be inputs into the statistical model and the result will be an estimate that is based on the experience associated with prior settlements applied to the client’s specific situation.

Both approaches can be useful for providing clients with estimates of potential exposure from a litigation.  Which approach to apply often depends on the facts and circumstances of the issues facing the client and the data and information available.