For anyone who has taken a business valuation course, or is a student of equity investment, you know that at least theoretically, a company’s stock price should represent the sum of all expected future cash flows, discounted for the time value of money. If we apply this to royalty rates, common sense would dictate that a license agreement with a longer term (let’s say perpetuity, at this end of the spectrum) should have a different royalty rate than a short-term agreement (let’s say three years)—all other factors, including the intangibles being licensed, equal. Should the royalty rate for the longer-term agreement be lower? Higher? Arguably, the longer-term agreement should have a lower royalty rate, given the licensee has a longer timeline to earn the same cash flows—again, we are keeping all factors other than duration equal.
So, does the data prove this out? While I have no easy way of controlling for all other factors than duration, i.e. find licensing agreements for the same exact intangible property licensed by one party to many different parties, all with different durations—and arguably, even if I did, it would not be a statistically significant sample size. What I do have is access to RoyaltyStat’s extremely useful statistics tool. If we take the same search criteria we used in our last experiment around exclusivity and run analytics showing the interquartile ranges of the agreements returned by their duration, we see the following results.
What is most striking about these statistics is that the median of agreements with different durations is fairly stable at around 5 percent. The only dip we see is for agreements with durations between 15-30 years. When we examine the quartiles, however, we do see that there is a trend showing lower rates as the duration increases—at least for this search focusing on apparel, textiles, beauty, and consumer products. Other industries could be different, going against the commonsense theory that rates should decrease with duration. For example, sampling 500 agreements from 523 returned in a search encompassing the Telecom, Computer hardware, software, semiconductor, and electronics industries, shows that the royalty rates actually increase with a longer agreement duration, as shown below.
There could very well be other industries where the search results show that duration has no impact on the range.
The conclusion we should draw from this exercise is that we cannot automatically use duration as a reason to reject potentially comparable agreements from our set. Rather, we must examine the data, using the analytics RoyaltyStat provides and determine whether the conventional wisdom holds true: Does duration play a role as a comparability factor for our search? Only after doing so can we properly use it as a filter in our search.
So now that we’ve examined duration, what about the actual date of inception of the agreement? One of the mantras I’ve always adhered to when leading transfer pricing teams is to reject agreements that are too old. But given our findings so far, perhaps that’s not necessarily the right thing to do. We’ll have a look in my next post.
Myth: Plausible
