Performance Perspectives Blog

The ever befuddling world of materiality

by | Dec 8, 2012

Today’s post was inspired by an op-ed piece in this weekend’s WSJ: “Madoff Got Away, But Netflix Won’t,” by Holman W. Jenkins, Jr.

Jenkins discusses how Netflix CEO Reed Hastings’ Facebook post that the company’s “customers had streamed a billion hours of video in the month of June” was determined to be worthy of scrutiny by the U.S. Securities & Exchange Commission. He suggests that one “ask just how crazy is the SEC in trying to extend its concept of ‘materiality’ to the news that Netflix actually disclosed.”

In our industry, the Global Investment Performance Standards (GIPS(R)) calls for the disclosure of material items multiple times, including the requirement that firms define a level of materiality for a certain threshold of errors (which happen to be classified as “material”).

Of all the things we wrestle with, defining materiality is probably one of the most challenging, because it is entirely subjective, and there is no real guidance or “best practices” to help.

My favorite word source,, offers definitions for “materiality” that refer to “material,” which is of no real value, and so we’re forced to check out what this latter term means. And while several definitions are offered, the one that applies here is “of substantial import; of much consequence; important.” An example is included: Your support will make a material difference in the success of our program.

And so something is material when it’s important and involves much consequence. Now, let’s translate that into numbers; ah, there’s the rub.

I tend to think that something is material if it causes us to look differently upon something than we would otherwise have.

An extreme example: Wilt Chamberlain, the late Hall of Fame National Basketball player claimed to have slept with 20,000 women; quite a tally. If we were to learn that the actual number was 19,000, or 21,000, would we think any differently about his achievements in this area? I think not. And yet, it would still be a 5% error over what he reported.

Interestingly, his actual height is unknown. He claimed to have been 7-foot tall, as I recall, but was apparently even taller (by a few inches, perhaps). Hardly a 5% difference from the actual height, but he was somewhat intimidated by his height and didn’t want to share the actual figure, fearing, perhaps, that it might cause some embarrassment.

When trying to define rules for materiality, try to consider what others would consider a noticeable difference, which could cause them to rethink their prior conclusions based on the earlier result: would it cause someone to say something like “well, upon further reflection I must reconsider my earlier response”? If yes, then it’s material.

As always, we’re open to your thoughts and ideas on this subject.

Free Subscription!

The Journal of Performance Measurement

The Performance Measurement Resource.

Click to Subscribe