News flash: we all make errors when analyzing forestry investments.  Which are the most common and how can we avoid them?

At the end of the day, most analytic errors in forestry relate to the data used – the inputs – or the math in a spreadsheet.  Errors associated with inputs affect the cash flows in the analysis and how it impacts value when discounted back to the present.  When reviewing models for clients, we commonly find errors associated with the assumed costs, the estimated revenues, the timing of cash flows and the expectations associated with inflation and asset appreciation.  Specifically, in our work and the work of others, we focus on and observe three major categories of errors in Excel models and spreadsheets.

First, we find that analytic errors in Excel are common.  Of all of the spreadsheet models we review, approximately one-third have an error in a formula (that we find). Most often, these are associated with “relative” and “absolute” referencing, where a formula is pulling in data from the wrong cell or worksheet.  Professor Ray Panko at the University of Hawaii, who actually conducts spreadsheet research (http://panko.shidler.hawaii.edu/ssr/), noted in a 2006 interview with the Wall Street Journal that “you’re going to have undetected errors in about 1% of all spreadsheet formulas.”

Second, we observe application errors, which basically reference issues with the thinking behind the spreadsheet, the formula chosen, the use of a formula, or a comparison made.  For example, analysts commonly make mistakes with inflation by mixing real and nominal discount rates and cash flows.  Also, spreadsheet models of forestry investments may inappropriately compare before and after-tax results and investments of differing duration (time periods). Finally, using the incorrect metric can be problematic, such as misapplying cash-on-cash return, pay back analysis and internal rate of return (IRR).  We typically find errors of this type when analysts are comparing timberland and forestry investments with alternative asset classes, such as stocks, bonds, agricultural commodities and commercial real estate.  Historically, the day-to-day metrics for forestry differed in application, so a bit of time spent checking assumptions and thinking through the communication of results can be helpful.

Third, errors of omission are especially problematic (and embarrassing).  We work hard to avoid the situation of being asked “did you check _____?” and, if relevant, having to answer “No, we didn’t think of that.”  Ugh.  Unfortunately, we find that third-party spreadsheets often omit key facts or considerations, including relevant costs and potential revenues.  Just as important can be confirming that the assumed costs and revenues are current and reflect the best available, accessible information.  In short, know what’s knowable.

We observe a few key practices in our work with clients to minimize the chance and occurrence of errors.  First, we label tabs and worksheets, and date all files. That way, if we revisit a model several weeks or months later, we can retrace our steps and know what we’re working with.  Also, if we correct an error or make another improvement, we know which version is the most current.  Second, we try to set aside time to check each other’s work before sending results “out the door.”  Admittedly, this is also the most challenging.  We find that imposing and embracing milestones – midpoints for deliverables or reviewing work – throughout a project institutionalize a level of quality control.  Finally, at the end of the day, we ask ourselves the question “where could we have blown this?” Some level of paranoia and self-awareness is required for quality analysis of forestry, or any other, investments.

Click here to register for “Applied Forest Finance” on February 7^th in Atlanta.  The course details skills and common errors associated with the financial and risk analysis of timberland and other forestry-related investments.