Risk-Based Capital Line of Business Diversification: Current RBC Approach vs. Correlation Matrix Approach

Abstract
The NAIC RBC Formula treatment of line of business (LOB) diversification (referred to in this paper as the CoMaxLine% Approach) is very different from the Solvency II Standard Formula treatment. In this paper we show that, notwithstanding the differences, the NAIC RBC Formula, the correlation matrix approach used in Solvency II1 and the Herfindahl-Hirschman Index (HHI), widely used in economics, all produce similar risk-based capital underwriting risk values, for most companies. To the extent that there are differences between the CoMaxLine% and correlation matrix approaches, the differences are due, in part, to the fact that CoMaxLine% calculates diversification based on premium or reserve volume while the correlation matrix approach calculates diversification based on premium risk or reserve risk. To examine this feature of the RBC Formula, we also apply the CoMaxLine% idea to risk by LOB rather than volume by LOB. We refer to that as CoMaxLine%-Risk. The differences between CoMaxLine%-Risk and the correlation matrix approach are smaller than the differences to the RBC CoMaxLine% Approach. This is one of several papers being issued by the Risk-Based Capital (RBC) Dependencies and Calibration Working Party.

Keywords: Risk-Based Capital, Capital Requirements, Analyzing/Quantifying Risks, Assessing/Prioritizing Risks, Integrating Risks, dependency, correlation.

Volume
Winter, 2019
Page
1-32
Year
2019
Categories
Actuarial Applications and Methodologies
Enterprise Risk Management
Processes
Analyzing/Quantifying Risks
Actuarial Applications and Methodologies
Enterprise Risk Management
Processes
Assessing/Prioritizing Risks
Actuarial Applications and Methodologies
Enterprise Risk Management
Processes
Integrating Risks
Actuarial Applications and Methodologies
Capital Management
Capital Requirements
Actuarial Applications and Methodologies
Regulation and Law
Risk-Based Capital
Publications
Casualty Actuarial Society E-Forum
Authors
Allan M Kaufman
Sholom Feldblum
Jennifer Wu