In a previous post I looked at the UK government’s suggestion that infrastructure proponents value the resiliency that infrastructure brings and that investors ask for this analysis to ensure that climate adaptation is included in the project design and the risks of climate change are accounted for.

This amounts to building a business case or cost-benefit analysis for the insurance that resiliency brings – identifying and quantifying the benefits of risk mitigation and climate adaptation. In this post I explore whether financial-type stress tests could be developed to model infrastructure’s resiliency and what is wrong with some of our design standards.

Stress Testing

Climate change can manifest itself in many ways: higher average energy prices, more volatility in energy prices, rising sea levels, worse storms, more frequent storms, and different geographical patterns for storms (less predictability). Therefore to nail down which risk factors you are going to be prepared for is tough as you could be prepared for the next hurricane but not an increase in gas prices.

To translate the word resiliency into a numerical or modeling context means that a project (or city) is able to react to large (and perhaps growing) shocks or changes in many risk factors. Resiliency is the ability, flexibility, or capability to withstand large changes in risk factors — these may be physical (1 in 200 year flood) or financial (energy prices greater than three standard deviations from the historical mean). To be able to make comparisons across projects or cities as to their preparedness would require a definition of a set of standard risk factors moving together to some agreed-upon set of extreme values.

This as akin to bank regulators defining standard stress scenarios to determine the resiliency of the banks to financial shocks. Financial stress tests typically allow not only for testing the ability to deal with extremes of individual risk factors, but also combinations of different events. These may be historical worst-case scenarios or worse.

Could a central agency or organization define these “stress scenarios” so that investors could assess resiliency or preparedness? An example stress scenario might be: over the next 20 years, energy price levels and their volatility increase at two times the historical rate of the last 20 years and population growth in cities is twice that of last 20 years. Or as above, ability to deal with 3 standard deviation move in prices of 1 in 500 year flood.

Just like infrastructure is built to withstand 1 in x year floods, there are sufficiency ratings[1] that are used by engineers for bridges. However, as the following describes, getting an engineer to define redundancy is as hard as asking an economist to define resiliency:

“It is not an exact science,” said Charles Roeder, professor of structural engineering and mechanics at the University of Washington, Seattle. “The probability of failure depends upon imponderable factors. I don’t think anyone knew that this truck would hit the Skagit River Bridge. Even if by some miracle someone knew that this specific truck would hit this specific bridge, they would not know that the bridge would collapse. Sufficiency ratings take into account many characteristics of bridges, including ones not directly related to their chance of failure.”[2]

The combination of flooding from extreme weather and the resiliency of bridges was brought into sharp focus when “Six cars were left dangling precariously over the Bow River Thursday when the CPR Bonnybrook rail bridge partially collapsed due to damage from flooding”[3] in Calgary.

Politically Charged Ratings Serving Too Many Masters

Sufficiency ratings and 100-year floodplains have unfortunate regulatory and political uses. These numbers are used to determine what gets funded and what is covered by insurance.

“… a bridge’s rating, 45% of which is derived from factors other than the bridge’s condition, such as whether it was built to current standard and how vital it is to local traffic flow. A drawback of sufficiency ratings is that by compiling information on so many characteristics, major deficiencies can have trouble standing out. The ratings are composites, and problems with a bridge’s condition stand alongside other factors, such as average daily traffic and detour length if the bridge were out of service. So one bridge could be more vulnerable to collapse than another with a lower rating. Two other related classifications, of bridges that are structurally deficient or functionally obsolete, also don’t directly translate into the chance of collapse, say engineers.”[4]

Do One Thing Well

New legislation seems to be heading towards letting engineers do what they do best, predict failure without worrying about he cost of failure:

“In legislation called Moving Ahead for Progress in the 21st Century, or MAP-21, that took effect last October, it ended the use of ratings for funding thresholds…. Engineers expect the new federal law will push an approach called bridge management. That involves using software to predict how bridges will change, and possibly fail, then to apply cost-benefit analysis to optimize spending on maintenance and repair. That could represent an advance from using sufficiency ratings as a crude screen, engineers say.”[5]

This approach seems to make sense as the cost of failure is a local calculation. NY’s financial district is worth more than Toronto’s. It does make sense though, to standardize the cost-benefit methodology and parameters. This can be done in the same way that the US DOT did in evaluating stimulus fund grant applications under its TIGER programs.

A professional engineering association, perhaps with a few economists for good measure, could take the lead and define resiliency standards in terms of a complete set of stress scenarios of extreme risks that infrastructure should be designed to. Then we economists can help legislators set priorities.

Breaking sufficiency of bridge collapse, or definition of 1 in 100 year floodplains, into probability of failure or occurrence (defined by standardized stress tests) and cost of failure (defined by costs of adaptation, benefits of avoided costs) seems to allow for a better understanding of resiliency and an aid to tough decision on climate change preparedness.

[1] “By having a complete and thorough inventory, an accurate report can be made to the Congress on

the number and state of the Nation’s bridges” Recording And Coding Guide For The Structure Inventory And Appraisal Of The Nation’s Bridges” FHWA Report No. FHWA-PD-96-001 December 1995 http://www.fhwa.dot.gov/bridge/mtguide.pdf

[2] “No Shortcuts in Measuring Bridge Safety” The Numbers Guy Wall street Journal Blog ByCarl Bialik May 31, 2013, 8:38 PM ET http://blogs.wsj.com/numbersguy/no-shortcuts-in-measuring-bridge-safety-1244/?mod=WSJBlog

[3] “Politicians point fingers over rail bridge failure” By Sherri Zickefoose And James Wood, Calgary Herald June 29, 2013 6:17 AM http://www.calgaryherald.com/news/alberta/Politicians+point+fingers+over+rail+bridge+failure/8595456/story.html#ixzz2Xu8Y7p7L

[4] “Will This Bridge Fall? It’s Hard to Say” – WSJ.com – Wall Street Journal Online May 31, 2013 http://online.wsj.com/article/SB10001424127887324412604578515980033291880.html

[5] WSJ.com – Wall Street Journal Online May 31, 2013 Op. Cit.