H11O-04:
Managing the financial risk of low water levels in Great Lakes with index-based contracts

Monday, 15 December 2014: 8:55 AM
Eliot Meyer, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, Gregory W Characklis, University of North Carolina, Chapel Hill, NC, United States, Casey M Brown, University of Massachusetts Amherst, Amherst, MA, United States and Paul Moody, US Military Academy, West Point, NY, United States
Abstract:
Low water levels in the Great Lakes have recently had significant financial impacts on the region’s commercial shipping, responsible for transporting millions of dollars’ worth of bulk goods each year. Low lake levels can significantly affect shipping firms, as cargo capacity is a function of draft, or the distance between water level and the ship’s bottom. Draft increases with weight, and lower lake levels force ships to reduce cargo to prevent running aground in shallow harbors, directly impacting the finances of shipping companies.

Risk transfer instruments may provide adaptable, yet unexplored, alternatives for managing these financial risks, at significantly less expense than more traditional solutions (e.g., dredging). Index-based financial instruments can be particularly attractive as contract payouts are directly linked to well-defined transparent metrics (e.g., lake levels), eliminating the need for subjective adjustors, as well as concerns over moral hazard. In developing such instruments, a major challenge is identifying an index that is well correlated with financial losses, and thus a contract that reliably pays out when losses are experienced (low basis risk).

In this work, a relationship between lake levels and shipping revenues is developed, and actuarial analyses of the frequency and magnitude of revenue losses is completed using this relationship and synthetic water level data. This analysis is used to develop several types of index-based contracts. A standardized suite of binary contracts is developed, with each indexed to lake levels and priced according to predefined thresholds. These are combined to form portfolios with different objectives (e.g. options, collars), with optimal portfolio structure and length of coverage determined by limiting basis risk and contract cost, using simulations over the historic dataset. Results suggest that portfolios of these binary contracts can substantially reduce the risk of financial losses during periods of low lake level at a cost of only 1-3% of total revenues.