Environmental Liabilities and Simulation

0
- By:

Courtesy of Oracle Crystal Ball GBU

Environmental costs are incurred by organizations in response to requirements of federal statutes (e.g., Clean Water Act, Clean Air Act, Oil Pollution Act) or state and local laws, and may be incurred either voluntarily or as part of a program to comply with these statutes and laws.  Examples of environmental costs include compliance programs, fines, penalties, legal fees, new pollution control equipment, natural resource damage claims, or remediation costs.

Estimating environmental costs presents a unique challenge because of the range of uncertainties that can affect the actual outcome.  Along with the typical uncertainties associated with cost estimates on construction projects (e.g., timing, production, labor, etc.), uncertainties associated with estimating environmental costs include:

  • Regulatory interpretation and approval,
  • Subsurface quantities,
  • Technology efficacy, and
  • Accuracy of historic unit cost information.

A number of factors may trigger the need to estimate an organization’s environmental liabilities.  Examples include business decision-making, negotiating transactions (e.g., sale or purchase of properties), disclosing environmental liabilities to shareholders, or complying with the organization’s Environmental Management System. Sarbanes-Oxley Act of 2002 requires CEOs and CFOs of publicly traded companies to certify both the financial statements and the procedures used to prepare those statements.

While disclosure requirements are not new, increased public attention on financial disclosures and environmental performance creates a need for standard methods for estimating environmental costs and understanding the uncertainty associated with these costs.

The (American Society for Testing and Materials) ASTM has established standard methods for estimating environmental liabilities. ASTM E2137 describes the following estimating methods in order of decreasing comprehensiveness:

  • Expected Value is the sum of all the probability weighted outcomes, or the mean value of a simulation model (e.g., Monte Carlo simulation methods).
  • Most Likely Value is the cost of the most likely outcome.
  • Range of Values is defined as the low and high cost outcomes, defining a range of values.
  • Known Minimum is the sum of the cost components that can be reasonably expected to occur.

With MS Excel and risk analysis software such as Crystal Ball®, practical tools are available to estimate environmental costs.  Further, these tools allow the user to more effectively include the uncertainty associated with environmental costs and more accurately communicate that uncertainty to the users of the information.

Example of Estimating the Cost of a Portfolio of Environmental Conditions

As an example of the application of simulation to estimating costs for environmental liabilities, an energy company wanted to know the potential environmental cost associated with soil and groundwater remediation at 44 distinct properties.  Table 1 is an excerpt from the full inventory of 44 sites.  The table includes information about the site including the estimated area and volume.

In this modeling example, key uncertainties included the proportion of soil or groundwater to be treated in situ, and the unit costs for these activities.  Other uncertainties, such as quantity, were not simulated as part of this exercise, although the uncertainty associated with quantities could easily have been included in the model.

The proportion of the soil treated in situ rather than removed was modeled assuming the proportion will vary between 0% and 20% with the most likely proportion either 5% or 10% depending on the surface area at the site.  The balance of the soil or ground water was removed.  A triangular probability distribution, defined by the minimum, most likely and maximum values, was used for this assumption.  The setup for this portion of the model is shown in Table 2.

Unit costs for the soil and groundwater treatment were modeled using the lognormal probability distribution.  The lognormal distribution is a good fit for simulating environmental costs because it is skewed right reflecting the residual probability of cost growth, and the result must be greater than zero.

Model Results and Sensitivity

The model was calculated for 3,000 trials to simulate the range of possible outcomes.  A frequency distribution of the total remediation cost for soil and groundwater is shown in Figure 1.

In terms of the ASTM standard, the following values describe the distribution:


Expected Value (Mean):           $  9,812,921
Most Likely Value (50% CL):     $  9,697,637
Range of Values (Minimum):     $  6,412,654
Range of Values (Maximum):    $13,632,211


Based on the simulation, sites were rank ordered to find the largest contributors to the total cost (Figure 2).  The top 5 sites accounted for 65% of the total cost.  These sites were then targeted for more detailed cost estimates to validate and refine the model.

Conclusion

Estimating environmental costs offers unique challenges because of the uncertainties involved in predicting the outcome.  There are a variety of reasons for an organization to develop an estimate of the environmental costs, from negotiating transactions to disclosing environmental liabilities to shareholders.  ASTM has developed standard methods for addressing the unique challenges of estimating environmental costs.  With the power of MS Excel and risk analysis software, these methods can be implemented in a practical and robust manner.

Jon Hoogenboom can be reached at jon@hightree.com

Customer comments

No comments were found for Environmental Liabilities and Simulation. Be the first to comment!