What is the royalty rate in a license agreement – a search for clues

Whenever appraisers chose to apply a?market approach, he or she has to establish?comparability?between observable market transactions and the valuation object. First level of comparability is close proximity of the business model, customer needs and customer benefits.

Like with all market-based valuations, the dodge to find comparability is a ratio or multiple. For real estate it is the price per sqm or the multiple of the annual net rent. For enterprises it is the earnings multiple or the sales multiple, among other. For intangible assets, the preferred ratio is the royalty rate which is a percentage of (net) revenues generated from using the intangible asset. Accordingly, the?relief from royalty method?(RFRM) is the prevalent market approach in the valuation of?intangible assets. It requires the selection of comparable transactions involving the licensing of a comparable intangible asset, and their royalty rates.

Since 1997, data vendors have been providing access to?license agreements?which are filed as exhibits in the?SEC database?as ?material contracts“ according to Item 601(b)(10) of the Securities Act and Exchange Act. Various databases collect, populate and categorize such license agreements from the SEC database. It is the oldest and most popular source for market transactions and royalty rates related to intangible assets.

Broadly speaking, there are approximately 22,000 license agreements accessible in the public domain. Now, as mentioned above, the denominator to make royalty fees comparable is percentage on revenues. In real life, there are many other denominators, i.e. a royalty fee based per units, per weight, on profit, or simply lump sums. They all are not suitable for comparison; leaving these license agreements apart, approximately 11,000 accessible license agreements remain with a royalty rate based on revenues.

But, things are not as simple as many believe. In an ideal world, a license agreement stipulates one clear royalty rate on all revenues for all periods of time for all different territories. Such clear royalty rate is easy and explicit to consider as comparable in a valuation. Sadly, the royalty rate reality in license agreements is different and follows a complex differentiation into different percentage rates for different revenues, or different periods, garnered by minimum or maximum payments. The result is a continuum of royalty rates that fluctuates depending on many factors, within a wide bandwidth. The appraiser must determine which of the different royalty rates to choose. Often, the mean or median value between the highest and lowest observed royalty rate in the agreement is determined to be the one relevant royalty rate. An oversimplification with potentially serious consequences. The determination of the “true” effective royalty rate in a license agreement remains an often-overlooked issue.

Below, we present five different trademark license agreements, their royalty rate clauses, their effective royalty rates, and the likely will of the contract parties. The five sample cases demonstrate the complexity of deriving a royalty rate from a license agreement. We did not have to go far to find these five cases. They are all taken from one landmark litigation case between Amazon and IRS on the taxable value of the Amazon trademark transfered to Europe. The experts of the parties introduced eight different trademark license agreements to the court, as comparable transactions, and we took five out of the eight.

?

Case 1:

License Agreement between The Sports Authority, Inc. and Mega Sports Co. Ltd. (2004)

?This license agreement covers the use of The Sports Authority trademark on sports stores in Japan between licensor The Sports Authority, Inc. (TSA) and licensee Mega Sports Co. Ltd. (MSC). Licensor is the largest full line retailer of sporting goods and equipment, apparel and footwear in the United States under The Sports Authority trademark. The license agreement was concluded in April 2004 for the period until 2014, and a predecessor version was concluded in 1999. The full versions of the two agreements can be accessed in the SEC database here (2004) and here (1999).

The royalty rates stipulated in the license agreement followed a schedule by business segment and contract year as follows:

The 1999 agreement stipulates increasing royalty rates from 1.0% to 1.2%. The 2004 agreement stipulates decreasing rates from 1.2% to 0.8% for free standing stores, and a fix rate of 0.5% for stores within department stores. For the renewal term beginning 2015, the agreement stipulates a maximum rate of 0.5% for everything, subject to negotiation.

Now, what is the ?true“ royalty rate in this license agreement? License agreement data vendors would state both a high end (1.2%) and a low end royalty rate (0.5%), and leave you alone with statistics. In the Amazon litigation case, expert for IRS selfishly proposed to select the highest rate 1.2%. Expert for Amazon proposed the mean rate between 1.2% and 0.5% and arrived at 0.85%. There is also an argument to weight the later years in the agreement with increasing revenues as the business was growing.

But after all, the will of the parties is key – most importantly the will of the licensee. The genesis of the agreements tells a clear story of what the licensee accepted to be a sustainable royalty rate. The first agreement was concluded until 2005. Before the end of its term, licensee requested changes to be made in his favor. Apparently, something turned out to be less positive than initially expected. Could be market success, could be profitability, could be strength of the licensed trademark, or other. The initial royalty rate of 1.2% was too high, and licensee had arguments to convince licensor to accept a substantial decrease. He negotiated 0.5% for stores within stores right away; for free-standing stores, there was a freeze period between 2007 and 2014 before licensor accepted the same 0.5%. Overall, the licensed business – in a developed and sustainable state – supported a royalty rate of 0.5%. Any other rate was an intermediate rate on the path to find out the appropriate rate. The 0.5% rate is 41% lower than the 0.85% median rate, and 58% lower than the 1.2% rate.

?

Case 2:

License Agreement between FAO Schwarz and The Right Start (2002)

This license agreement involves a trademark license agreement between licensor F.A.O. Schwarz Family Foundation (FAOS) and licensee The Right Start, Inc. (TRSI), governing the use of the F.A.O. Schwarz trademark on retail stores. The agreement was signed in 2002.

FAO Schwarz was a well-known chain of traditional toy stores which emerged out of the famous flagship store in New York (originally opened on Broadway in 1870 , later moved to Fifth Avenue). In 1963, the Schwarz family sold their ownership in the retail business, under a license agreement to use the FAO Schwarz trade name. In the years thereafter, ownership in the toy retail business changed hands several times, and the trademark license persisted.

By 2001, the FAO Schwarz business was owned by Dutch retail group Royal Vendex. It operated 41 toy stores and had revenues of $200 million. For various reasons, the business was in distress and making heavy losses. In late 2001, Royal Vendex sold the assets of 23 of ist 41 stores to The Right Start, Inc.; all other stores closed. The Right Start and FAO Schwarz entered into a new trademark license agreement in January 2022. The full version of the license agreement can be accessed in the SEC database here.

The license agreement stipulated a royalty rate of 0.25% on the first $50 million of revenues, and 0.375% on revenues in excess of $50 million.

By the end of 2001, the 23 remaining FAO Schwarz stores acquired by Right Start had revenues of $150 million. During the first year under the license, an additional 8 stores were closed, and revenues were $118 million. Worth to note that the business continued to decline, and that licensee Right Start filed for Chapter 11 protection in 2003. All but two FAO Schwarz stores were closed, and the license with Right Start was terminated.

The contractual royalty rates vary from 0.25% to 0.375%. What is the ?true“ royalty rate in this license agreement? License agreement data vendors would state a low end rate of 0.25%, and a high end rate of 0.375%. Effective royalty rates were 0.33% at the end of 2001, and 0.32% during 2002. The circumstances suggest that negotiations on royalty rate were tough. It looks like licensor came from a (historically) higher rate, but now the business was in trouble. Licensee tried to get a 0.25% rate, which licensor did not accept. The scaled rates were a compromise. Licensee certainly hoped to sustain the acquired business, and eventually to grow it again later. Therefore, the 2001 royalty rate of 0.33% was the most ?true“ rate withing the bandwidth

??

Case 3:

License Agreement between KMart Corporation and KMart Australia Limited (1994)

This license agreement governs the use of the KMart and Super K trade names in Australia.

Licensor Kmart Corp. was an important operator of general merchandise stores (supermarkets, hypermarkets) in the US. In the 60s, Kmart established a JV in Australia with Australian retail firm Coles Myer, to operate Kmart Stores in Australia. Kmart held 51% of the shares of Kmart Australia Ltd. (KMA), Coles Myer 49%. In 1978, Kmart’s 51% shareholding in KMA was switched into a 21.5% direct shareholding in Coles Myer, whereby Coles Myer gained full control over KMA.

In 1994, in an attempt to streamline its portfolio and to divest some non-core businesses, Kmart Corp. sold all of their shares in Coles Myer to Coles Myer. This buyout required new, extended trademark license agreements between the two parties to cater for the full independence of Coles Myer and KMA from licensor KMC. The full version of the license agreement can be accessed in the SEC database here.

The license agreement fixed a royalty rate of 0.188% on KMA’s revenues up to a maximum annual amount (cap) of AU$ 5 million. This is an interesting and rare solution; much more often, license agreements have a minimum annual royalty amount, plus a running percentage royalty on revenues.

Under a 0.188% royalty rate, the maximum royalty amount of AU$ 5 million is achieved at annual revenues of AU$ 2660 million. For revenues higher than this amount, licensee would pay no additional royalties, and the average royalty rate paid on revenues (the ?effective“ rate) would be lower than the 0.188%.

Now, what is the ?true“ royalty rate in this license agreement? License agreement data vendors would state 0.188%. In the Amazon litigation case, expert for Amazon proposed 0.188%, too. The key information to answer this question is – what are the revenues of licensee? A look into financial statements helps to get a little more background.

In 1993, Kmart Australia operated 147 Kmart stores and paid US$ 3 million of license fees (AU$ 4.5 million, on licensed revenues of AU$ 2400 million), thus little less than the maximum amount. The effective royalty rate at the date when the agreement was signed was 0.188%, but total royalty fees were already close to the maximum cap. By 2004, Kmart Australia had 168 stores and revenues of AU$ 3,500 million, thus more than the revenue threshold. In that year, the effective royalty rate was approximately 0.14%. After 2004, there was a period of distress and stagnation for both licensor and licensee. Licensor went bankrupt and was later merged into Sears. Licensee Coles Myer divested Coles, and the remaining business was acquired by Wesfarmers. Since 2010, Kmart Australia returned to growth, with revenues of AU$ 3,600 in 2010, and AU$ 3,800 in 2014. In 2017, Kmart Australia had grown to little over AUS$ 5,000 million, resulting in an effective royalty rate of 0.10%. In that year, Wesfarmers acquired the trademark rights in the Kmart name and terminated the license agreement.

Apparently, when licensor sold all of its shares in licensee and the license agreement was signed in 1994, the deal was that licensee would pay royalties as before for the then existing revenues, but just that and not more. In other words, licensee expected to get any additional future revenues royalty-free, at 0%. More than 20 years later, revenue increases brought the effective royalty rate down to 0.1%, a decrease of 47% from the rate fixed in the agreement.

It is pure speculation which business expectations (revenues) licensee had in mind when he signed the 1994 license agreement. In any case, the 0.188% rate is the maximum ?true“ rate to be taken from this agreement.

??

Case 4:

License Agreement between Equilink Licensing and Sport Supply Group (2000)

This agreement involves a trademark license agreement between licensor Equilink Licensing (ELC) and licensee Sport Supply Group Inc. (SSG), governing the use of the MacGregor trademark. The agreement was signed in 2000. The agreement involves a series of preceding transactions and a settlement agreement between the parties. The full version of the license agreement can be accessed in the SEC database here.

MacGregor was one of the first US manufacturers of baseballs and baseball gloves, and later added team uniforms, caps and other equipment to its range. By 1989, MacGregor filed for Chapter 11. The remains of MacGregor were sold in parts to different investors in 1991. MacMark acquired major parts of the MacGregor trademark rights. MacMark Corp. is 50% owned by Riddell Sports Inc. (a supplier of football gear), and 50% by Hutch Sports Inc. (a supplier of licensed teamsport apparel). Equilink (ELC) is a subsidiary of Riddell Sports Inc. in charge of sub-licensing the MacGregor trademarks. SSG (previously named Blumenfeld Sports Net Co. or BSN in short) was a direct mail marketer of sports related equipment and leisure products to the institutional market in the US (schools, colleges, universities, government agencies, military facilities, athletic clubs, athletic teams).

In 1992, MacMark had granted a perpetual, fully paid up (!) license to SSG/BSN to use the MacGregor trademark on various ball sport equipment for distribution channels other than store-based retail (i.e. team sport supply), subject to annual minimum turnover. Later, disputes arose between MacMark/ELC and SSG concerning breach of the 1992 TLA and termination by MMC. In 2000, the parties resolved this dispute in a settlement agreement, which involved the conclusion of a new trademark license agreement with a limited term and royalty payments based on annual revenues. Royalties stipulated in the 2000 license agreement were a flat annual minimum of $100.000 for revenues of up to $17 million, 2% on all revenues above $17 million, and 1% on a certain closeout sales above $22 million (limited to 10% of total sales).

What is the ?true“ royalty rate in this license agreement? License agreement data vendors would state a low end rate of 1.0%, and a high end rate of 2.0%. In the Amazon litigation case, expert for IRS selfishly proposed the high-end of 2.0%. The key information to answer this question is – how to consider the flat annual minimum amount, and what are the revenues of licensee?

Financial statements of the licensee suggest that the royalties paid in the first year were $100.000. However, nothing is stated about the revenues achieved under the MacGregor brand – except that SSG ?had converted a substantial portion of its products to the MacGregor brand”. We speculate that the annual minimum revenues of $17 million correspond to the annual minimum revenues stipulated in the 1992 license agreement. We also speculate that SSG missed the annual minimum revenues, which caused licensor to terminate the otherwise perpetual 1992 agreement.

Eventually, effective revenues were somewhere between $10 and $15 million, causing an effective royalty rate of between 1.0% and 0.67%. This has however little substance and is speculative. In any case, the 2% royalty for revenues above $17 million is very clearly not the true royalty rate in this agreement, mathematically it is not even achievable.

?

Case 5:

License Agreement between Tandy Corp. and InterTan Australia (1999)

This agreement involves a trademark license and sourcing commission agreement between licensor Tandy Corporation (TC) and licensee Intertan Australia Limited (ITAL), governing the use of the Tandy Electronics trademark on electronic retail stores in Australia. The agreement was signed in 1999. The agreement involves a complex relation and a series of previous transactions between the parties. The full version of the license agreement can be accessed in the SEC database here.

Licensee was a former international distribution division of licensor. Licensor (TC) was a manufacturer of personal computers until the early 90s under the Tandy brand. Moreover, TC operated the RadioShack retail stores for consumer electronics in the US and Canada, and Tandy stores outside the US, i.e. in UK and Australia. InterTan Inc. is the former retail business of Tandy Corp. outside the US which was spun-off to the Tandy shareholders in 1986. Through the spin-off, InterTan became a fully independent company. However, InterTan remained under strong influence of Tandy through various collaboration agreements. After the spinoff, Tandy remained InterTan’s principal supplier, the licensor of the Company's principal trade names, and a secured creditor.

In 1986, TC and InterTan entered into a merchandising agreement whereby a Tandy subsidiary was the exclusive exporter of products to InterTan. Given the different local market requirements, InterTan sourced an increasing share of its inventory from other suppliers. By 1995, InterTan sourced only 30% of its inventory through TC; former synergies and economies of scale between the two firms went lost. To cater for the new situation, Tandy and InterTan entered into a new merchandising agreement and into some license agreements to use various trade names owned by Tandy in 1999. Among other, these license agreements permitted InterTAN Australia Ltd. (ITAL) to use the "Tandy Electronics" trade name in Australia and New Zealand on electronics retail stores. In order to encourage common sourcing, the royalty fee for the trade name was a function of the share of ITAL’s inventory sourced through TC, where the trademark royalty decreases with an increasing share of inventory sourced from TC.

In addition to the trade name royalty, there was a sourcing commission of 1.0% on revenues from all inventory sourced through TC, so licensee had to pay the sum of trade name royalty rate plus sourcing commission.

What is the ?true“ royalty rate in this license agreement? License agreement data vendors would state a low end rate of 0.350%, and a high end rate of 1.0%. In the Amazon litigation case, expert for IRS selfishly proposed the high-end rate of 1.0%.

There was a maximum percentage of 1.47% TC could earn from both trade name license and sourcing (at a royalty rate of 1.0%), and a minimum of 1.22% ITAL had to pay, at a royalty rate of 0.60% and a sourcing share of 62%.

By 1999, ITAL generated revenues of US$ 105 million under the Tandy Electronics name. ?The royalty rate ITAL paid in 1999 was 1%, meaning that the share of products sourced from TC was 46.9% or less. Eventually, the parties had envisaged to increase the share of sourced products to ~60%, resulting in a royalty rate of 0.6%. However, the license agreement was terminated in 2001, and InterTAN sold its Australian business to Dick Smith/Woolworths. Therefore, the ?true“ royalty rate from this agreement is likely 1.0%.

?

Summary

What do these five case studies tell us?

1.???License agreements are rarely about one single and clear royalty rate. Often, they involve complex mechanism based on different variables.

2.???When there is no single royalty rate in an agreement, data vendors typically provide the lowest and highest rate stipulated in the agreement. The valuation analysis would then conclude on the mean rate between the highest and lowest rate. Using the mean is far too simple to describe agreement reality.

3.???Understanding the full royalty rate mechanism requires full-text reading of the agreement, and calculations. The bandwidths can be substantial.

4.???Concluding on the effective or ?true“ royalty rate requires a lot of background information, including effective licensed revenues, negotiation details, and other. Often, determining the effective or true royalty rate is simply impossible.

5.???Using royalty rate data from license agreements in royalty relief valuation requires much more background search and analysis than what appraisers usually do.

6.???Some license agreements should be rejected as it is impossible to draw reliable conclusions on royalty rates.