The Relief from Royalty Enigma - Newsletter May 2007

They say that all good things come in threes. Over the past few months I have been employed as an expert witness on three separate occasions. Strange that that should happen in the space of just a few months. This kind of work is fascinating because it takes you into another world. Not just the intriguing ability of court lawyers to shift their attention from one industry and set of circumstances to another, but also the nature of what people sue others for. Naturally the ones I was asked to advise on were to do with brands and brand equity.

In one case, a firm was suing another because they had jointly embarked on a merger which did not proceed. The plaintiff (see, I know the terms now) was aggrieved because they felt the other firm had broken the agreement and walked away from the deal. Because the price was related to the brands that would be part of the transaction, the value of the portfolio was a major component in the estimation of damages. An expert was appointed to work out what this was. Of course the “other side” wanted to shoot down this calculation; hence my appointment. That’s the nice part: you are asked to be a legal assassin.

I don’t know how this will be argued and what a judge might decide, but I found the situation quite conducive to my well honed sharpshooter skills. Quite frankly the work was sloppy and full of holes. I won’t go into detail for fear of identifying the perp. But if you know anything about discount rates you will be amused to hear that rather than use a risk free rate, such as the R153 government bond, this person stated in all seriousness that the rate would be based on the inflation rate at the time which was “about 9%”. Actually it wasn’t. I looked it up and according to the Reserve Bank Governor and FNB Economic Research, the average for that year was 5,7%. Two early big hits.

It was not, as must be clear, too hard to find fault with this valuation. But the effort did produce some interesting spin offs because I made three discoveries that I will mention later.

The valuation expert used a method to conduct the calculation that I abhor. Relief from Royalty has been around for many yeas and is used by professional firms to value brands. I believe it to be fatally flawed, but somehow it continues to be used. In preparation for my attack, I made the first of my discoveries. According to the papers I was given, written by my opposing expert, Relief from Royalty was developed by Price Waterhouse in New York under the title of Royalty Savings. But, some pages I downloaded from the net seem to imply that it was in fact PW’s rival, Deloitte, that used it first. The fact that the approach was developed by one or the other adds nothing to its appeal. It simply fails too many tests.

For example:

• The approach applies a royalty rate to projected brand turnover. That disregards the bottom line and if the brand actually makes any money.

• A royalty rate is chosen and applied; the theory being that the royalty would have to be paid to a third party for use of the trademark if the brand owner didn’t own the brand. Since he does, the royalty does not have to be paid and therefore the capitalised present value of the royalty stream that would have been paid is saved and this is the value of the brand. Even if you accept that rather convoluted explanation, there is another problem. The whole thing about brands is that they differentiate one product from the other. Royalty Relief requires the valuer to find what they call a “comparable” rate. They choose and use a rate typical of the category. One consistent rate allows no scope for differentiation. All brands become equal.

• This leads to another problem. What royalty rate is appropriate? This is the source of my other two discoveries. In the case in point the valuator (as he describes him or herself), kind of made it up. After a long explanation of how they are derived, the rate suddenly appears, unexplained, in the text.

My second discovery was a firm in New Jersey in the USA called RoyaltySource. For a modest payment I was able to acquire a set of comparable rates paid by firms in a broadly similar category. Thus had I been conducting the valuation, I could at least have provided some support for my chosen rate. Although, I have to add, that I would not like to justify such an important input to my model with such a wide spectrum of numbers. I would prefer to use the 25% rule. My third and final discovery.

Over the years I have heard of this approach, but not with any great consistency. It is always a ratio between net sales and profit, but whether this is gross or net profit seemed, until recently, to be debateable. The principle is that you calculate the ratio and the royalty rate is 25% of that. Not that it mattered to me, because it is not a device I am likely to use. Chatting to the very helpful people at Royalty Source I discovered they had contributed to a recent article about this strange sounding tool. The paper appeared in an esoteric journal, which you will not easily find. I do have a copy.

Now I learn that a Mr Robert Goldscheider first developed the approach in the 1950s. He was a co-author of the paper which is dated 2002. The concept is based on an empirical study he conducted where he found that this seemed to be the rule: licence fees tended to equate to one quarter of the profits that the licensee earned from the trademark or patent. The calculation is based on an estimate of future net profits before tax. This is divided by net sales to produce a percentage, one quarter of which is the royalty rate. The amount is sufficient to make it worth while for the licence owner to make it available to the user, who in turn is still able to generate a positive return.

It may appear a little crude, but it is related to the profits that the user will earn, and it does sound very reasonable.

• The royalty is applied to projected sales, discounted to present value and capitalised. But for how many years? Relief from Royalty users are confronted with a problem of whether the brand has a finite or infinite life. If it is finite, perhaps limited by the licence agreement, or expiry of a patent, the number of years in the projection is clear. But if the brand is for all time, which is what most brand owners hope for, the longevity must be taken into account. But how?

In the case in point the valuer chose ten years and chose is the correct term. There was really no justification for it other than most licensors would expect a brand to live for at least ten years. I wonder which brand owners he spoke to.

The way most knowledgeable users deal with this is to project the sales for five years, using management budgets and country growth rates. They then divide a sixth year by the discount rate to produce what they call an annuity, but what is formally a growing perpetuity. The former represents the early years and the latter is a proxy for the brand’s long life: discounted to infinity.

What strikes me as problematic in this approach is that the annuity invariably accounts for a larger portion of the value than the early years; sometimes as mush as two to three times as much. Further, many of the valuations I have seen using this approach, including the one I was dealing with, apply a single growth rate for all five or ten years. As a finance colleague at university pointed out, exponential growth like this will cause the brand eventually to be larger than the economy of the country. Any business can have a few exceptional years of growth but it is unlikely that they will sustain that for more then three or four years. The prudent way to deal with growth is to use reasonable management forecasts for two or three years and then step the growth down to the projected growth of the country itself.

This is the approach used for many very serious valuations and unless I am missing something or am unnaturally biased against it, I cannot see why it survives.

Especially when there are such wonderful alternatives.