Improved Analytics Revolutionize Catastrophe Planning

Unlike when Hurricane Katrina struck in 2005, companies today can use Big Data to make more informed risk financing decisions.
Claude YoderJuly 2, 2015
Improved Analytics Revolutionize Catastrophe Planning

This year marks the 10th anniversary of Hurricane Katrina, the costliest hurricane on record in the United States with $41 billion in insured losses and more than $100 billion in economic damage, according to industry and government estimates. For many businesses, Katrina’s destruction exceeded the scope of their business resiliency and disaster recovery plans.

Claude Yoder

Claude Yoder

Major disasters typically spark an evaluation of the way risk is managed. Following the 2005 hurricane season — which also included Hurricanes Rita and Wilma — catastrophe (CAT) models came under intense scrutiny by insurers that relied on them.

Ten years of technology advances and big data and analytics are helping companies today to not only gain a better perspective and understanding of their unique hurricane risks, but also of their risk tolerance levels and the potential impact of a storm on their return on capital. That can help them make more informed risk financing decisions to protect their bottom lines.

Drive Business Strategy and Growth

Drive Business Strategy and Growth

Learn how NetSuite Financial Management allows you to quickly and easily model what-if scenarios and generate reports.

CAT Models

One positive outcome from Hurricane Katrina was a flight to data quality. Initial loss estimates from the storm were inadequate in part because some of the data in the models wasn’t sufficient. Today, insureds are gathering and providing much more detailed information about their properties — from the exact latitude and longitude to the number of stories of each building and type of construction, for example. In addition, unique construction characteristics, like flood defense systems and first-floor building elevations, can be captured to more accurately depict and differentiate risk.

At the same time, the CAT modelers themselves have made many improvements to their loss assumptions, especially as related to business interruption and storm surge.

So while an insured 10 years ago was able to model loss expectancies for a 1-in-10-years hurricane on up to a 1-in-500-years storm, those loss projections are improved today and give risk managers and CFOs better insight into their hurricane exposures and risk protection needs.

Risk Financing Optimization

Just as companies can use CAT modeling results to consider how much property insurance to purchase, so too can they model whether the cost for that coverage provides increased value versus alternative structures or even self-insurance.

As with any use of capital, it’s important to be able to estimate the expected return on it. Today’s analytics models can help companies answer questions concerning expected return. To do this, models look not only at premium, risk retention, and policy limit sufficiency, but also at the cost of volatility of the potential loss scenarios the company may face.

From a capital perspective, when an adverse event occurs that hasn’t been fully insured, there might be an impact on earnings, capital may need to be diverted from ongoing projects, and the company could be forced to raise additional capital. Because no company is fully protected against the unexpected, every organization bears an implied cost for its unexpected risk.

Analytics techniques can be used to calculate a company’s implied cost of risk by examining the potential frequency and severity of unexpected losses as well as understanding its cost of capital. This implied risk charge, in addition to a company’s expected retained losses and insurance premium, reveals a more complete picture of the true value of the insurance program — its Economic Cost of Risk (ECOR) — and whether the program makes sense from a risk tolerance and capital outlay perspective.

Analytics in Use

Consider an example where the CAT model output projects that a 1-in-250-years hurricane event would result in an estimated $200 million loss for Company X. Based on that outcome, Company X spends $5.5 million to purchase $200 million in property coverage above a $5 million risk retention.

Is $200 million in coverage worth the $5.5 million capital expenditure? Or would Company X be better off purchasing less coverage, perhaps choosing a higher risk-retention level?

Consider that Company X has a highly volatile hurricane-exposed risk portfolio — it expects to face one large hurricane loss every 10 years — and as a result its implied risk charge is calculated at $130,000. The chart below compares Company X’s ECOR over that 10-year period scenario for a $200 million windstorm program with two different retention levels.

18456 Chart for 06-2015_v1

In this instance, under a $10 million retention structure, Company X would retain an additional $450,000 in expected loss within its deductible over the 10 years and could expect its premium to decrease commensurately. However, its implied risk charge — the expected costs associated with uninsured losses — increases to $250,000 from $130,000.

This makes the total ECOR for the $10 million retention option higher than the cost of the $5 million retention, suggesting that Company X is economically better off by staying with the $5 million retention.

Although one cannot predict when the next Hurricane Katrina will hit, advances in data, analytics, and technology allow companies to better understand their hurricane exposure and risk potential in addition to the true economic cost of risk underlying their property programs. As a result, companies today are making more informed risk financing decisions and seeking to protect their companies’ bottom lines with more confidence during the hurricane season.

Claude Yoder is a managing director with Marsh, based in New York City, where he leads firm’s Global Analytics practice. In this role, he is responsible for enhancing and evolving Marsh’s analytical offerings on behalf of clients, colleagues, and markets.

This information is not intended to be taken as advice regarding any individual situation or as legal, tax, or accounting advice and should not be relied upon as such. You should contact your legal and other advisors regarding specific risk issues. The information contained in this publication is based on sources we believe reliable but we make no representation or warranty as to its accuracy. All insurance coverage is subject to the terms, conditions, and exclusions of the applicable individual policies. Marsh cannot provide any assurance that insurance can be obtained for any particular client or for any particular risk. Marsh makes no representations or warranties, expressed or implied, concerning the application of policy wordings or of the financial condition or solvency of insurers or reinsurers.