Sunday, June 12, 2011

Patent Licensing Fees Modest in Total Cost of Ownership for Cellular

In this, the third in a series of features written for Keith Mallinson (WiseHarbor) for IP Finance, Keith addresses the claim that the aggregate of patent licence fees paid by anyone buying into patented mobile handset technology is prohibitive and stifles competition (Nb if you can't read Exhibits 1 and 3 clearly on-screen, try clicking them).
"Patent Licensing Fees Modest in Total Cost of Ownership for Cellular 
Patented technology is the lifeblood of today’s advanced mobile handsets, network equipment and operator services. As mobile services become increasingly sophisticated, manufacturing of handsets and network equipment represents a declining share of value compared to investments in innovative mobile technologies and software. There is no inherent maximum value share for the IP created with such investments. Aggregate IP fees are a small proportion of handset costs and are very modest compared to operator service charges. Handset costs as a percentage of total ownership expenditures including operator services are 17% in the US and Canada and 13% in Western Europe. 
My previous IP Finance posting showed markets for mobile phones and operator services have flourished with outstanding growth, technological innovation, significant competition and tumbling prices on the basis of (Fair) Reasonable and Non-discriminatory licensing for technologies required to implement mobile communications standards. Despite all these positives, some still complain IP fees are excessive in comparison to other costs. In this article, I evaluate fees paid upstream in technology licensing in comparison to downstream expenditures in supply of handsets and provision of operator services. 
Caps to fix IP charges 
There are concerted attempts to limit licensing fees in standards-essential IP.  For example, downstream equipment manufacturers seek to minimize out-payments for licensing standards-essential IP by promoting aggregate royalty caps.  In 2008, Alcatel-Lucent, Ericsson, NEC, NextWave Wireless, Nokia, Nokia Siemens Networks and Sony Ericsson announced their agreement that aggregate royalties for handsets implementing the 3G/4G LTE standard should be capped below 10% of handset prices. Similarly, mobile operators, who in many cases subsidize handset prices to consumers, also seek to limit these licensing fees.  A common proposal from several mobile operators is to limit aggregate essential-IP charges by establishing an LTE patent pool. Patent pooling will be the topic of my next IP Finance posting. However, one immediate and obvious observation is that if a patent pool is designed to limit aggregate license fees for the benefit of downstream licensees, then it will be unattractive to upstream licensors that depend on licensing revenue to fund continued investments in R&D and earn a return on prior investments.  Also, the major vertically-integrated companies have mostly preferred to enter into bilateral agreements with other vertically-integrated companies in order to be able to negotiate cross-licenses with trade-offs between their business interests and patent portfolios. 
Unproven suppositions of licensing excesses by some technology licensors and resulting harm abound by predominant voices downstream and their cheerleaders. For example, an August 2009 contribution to the European Competition Journal by Philippe Chappatte of Slaughter and May argues that: 
·         There is likely to be an upward spiral of royalty claims for many standards including telecoms standards resulting in higher costs for handsets and other standardised products; and

·         Operators will be reluctant to invest in new technologies or upgrade their networks to endorse faster and higher quality networks and the quality and range of services that will be available to consumers may be prejudiced. 
Contrary evidence is that handset prices and royalty costs have actually fallenwith handset prices, upon which royalty fees are based, declining 77% on average since 1993despite the addition of many new technologies and increasing demand for advanced features and functionality. 
Estimates for “cumulative royalties” vary widely. In 1998, International Telecommunications Standards User Group (representing some operators and manufacturers) complained to the European Commission that “when GSM handsets first appeared on the marketplace cumulative royalties amounted to as much as 35 percent to 40 percent of the ex-works selling price”. Much lower estimates for the cumulative GSM royalty rate paid, by companies that do not have any patents to trade, include 10-13 percent (IP Law and Business reporting PA Consulting Group estimate, July, 2005). In September 2005, CSFB’s “3G Economics” report estimated cumulative royalties had fallen to single digits and predicted 17.3% cumulative royalties in WCDMA “for those vendors without an IPR position to trade off”. Whereas ABI Research described average WCDMA cumulative royalties of 9.4% in 2007 “a most challenging barrier... ...to the development of more affordable devices”, the market-leading handset manufacturer with 37% share was paying much less: Nokia stated that “until 2007 it has paid less than 3 percent aggregate license fees on WCDMA handset sales under all its patent license agreements”. 
In addition, there have been various attempts to determine aggregate fees sought by licensors for new technologies. In 2007, the Next Generation Mobile Network (NGMN) Alliance, an industry group led by mobile operators and including major 4G equipment vendors, established a confidential process for the ex ante disclosure and aggregation of expected licensing fees for a number of upcoming 4G standards including LTE. The process concluded in 2009 and the results are confidential. However, commentators have suggested the individual disclosures of expected licensing feeswhich were in several cases accompanied by public disclosures on company websitesproduced misleading and unrealistic figures.
Aggregate figures derived are not actual prices paid including cross-licensing and do not reflect other realities in negotiations such as identification of patents that are weak or inapplicable. Patent strengths and “essentiality” were not validated. In 2003, the 3G Patent Platform Partnership (including 19 telecommunications operators and equipment makers) estimated “that several hundred different patents, among several thousand publicly claimed as essential, will actually be determined to be ‘essential patents’ in implementing 3G standards”. Some candidate licensees would rather risk being sued than pay “rack rates” in these circumstances. Licensors prefer to negotiate settlements than litigate and subject their patents to invalidity and non-infringement claims. Vertically-integrated licensors are particularly concerned about their product revenues with the risk of being counter-sued for infringement. 
Mobile operators are as eager as ever to invest in new technologies to improve performance and lower total costs. New technology cost savings outweigh licensing fees. For example, while mobile operators spend billions of dollars on spectrum, technological advancements have mitigated this cost with 20-fold spectral efficiency increases and much improved voice encoding since 1G analogue cellular. Operators worldwide are investing extensively in advanced technologies HSPA+ and LTE that have increased network capacity and maximum end-user data speeds 1,000-fold since the introduction of 2G technologies around 1993. In the US, for example, all the major operators (and smaller ones too) claim to have introduced “4G services” over the last couple of years. Operators are also making major investments in associated devices by significantly subsidising end-user prices. With demand for HSPA+ and LTE so strong, IP cost issues can be no more significant than they were with previously and currently successful 2G and 3G technologies. 
Increasing value share in software and patents 
There is no reason why any arbitrary percentage limit should be imposed on IP costs. It is widely accepted that when one pays, for example, $25 for a hardback or $10 for a paperback book, production costs in printing account for but a small proportion of these figures. Royalties to authors, illustrators and agents as well as costs in distribution, marketing and the publisher’s profit margin account for the vast majority of these prices. Similarly, other IP-intensive products, as illustrated in Exhibit 1, have a significant proportion of costs in the intangibles. 
Exhibit 1: Manufactured content value varies substantially by product category
                    Source: WiseHarbor
I have predicted a marked trend of increasing value with the intangibles in mobile devicesincluding embedded and aftermarket software predominating over hardwaresince Apple’s 2008 3G iPhone launch. The success of the iPhone including its Apps store proves my point. The iPhone leads the smartphone market and has a manufacturing cost around just one third of its $600 average wholesale pricing (before operator subsidies to consumers). Gross profit margins approaching 60% provide a significant return on investments in software, brand and distribution, while Apple largely relies on the essential IP developed and contributed to mobile standards by others. 
Handset, network and services-essential IP 
Mobile phones are inextricable from the networks and operator services with which they are used: licensing fees should be considered in this broader context. In contrast to technologies that can be used offline, such as in audio and video players, standards-essential IP is implemented end-to-end in handsets and network equipment with the provision of cellular voice and data services. In addition to increased speeds and network capacity, end-to-end innovations include voice encoding, encryption, automatic roaming and location tracking. A handset in isolation from a network cannot make calls or receive data, let alone exploit any of these capabilities. By convention, licensing fees are charged on wholesale mobile phone prices. Whereas this royalty base is simple and convenient to administer in licensing, it overlooks where most ecosystem value is generated—in operator service revenues. In fact, phone prices are commonly subsidised—to substantial extent in many cases—by operators in anticipation of these revenues. 
The average service life of a phone from purchase until retirement is around 20 months in the US where postpaid contracts predominate and 34 months in Western Europe where most users have prepaid or SIM-only service with unsubsidised phones. Exhibit 2 shows that during a handset’s service life, consumers spend on average around five or six times more on service fees than they or their operators spend on the handset. Handset costs in the US/Canada and Western Europe represent 17% and 13% respectively of total ownership expenditures including handset costs and operator service charges. 
Exhibit 2: Handsets, a small proportion of total ownership expenses

US and Canada
Western Europe
Average service revenue per user (per month)
$50
$32
Service life (in months)
20
34
Total operator services expenditures
$1,001
$1,087
Average unsubsidised wholesale phone price
$207
$167
Total lifecycle expenditures
$1,208
$1,254
Handset cost/total expenditures
17%
13%
       Source: WiseHarbor, based on 2009 and 2010 market figures
Royalty rates expressed as a percentage of total ownership lifecycle expenses are therefore much lower than rates based on handset prices. Exhibit 3 shows that converting aggregate handset cost-based royalty rates to rates based on total ownership expenditures reduces the rate to 13% and 17% of the rate based on handset costs for Western Europe and US/Canada respectively.  More frequent handset upgrades in the US account for most of the differences between the two regions. 
 
Source: WiseHarbor Research   * For companies with no IP to trade

Competitive advantage with IP 
It is not the average level of IP charges that affects competition; it is the different rates paid among competitors. Aggregate royalty rates are significantly less than European Union VAT rates that have mostly ranged from 15% to 25% in recent years.  Applied uniformly among competitors, taxing phones and services at these VAT rates has not significantly impeded their sales versus nations where consumption taxes on phone sales are much lower. 
The asymmetry in licensing costs between manufacturers with IP who can cross-license to minimise their licensing expenditures and manufacturers without essential-IP patents who must pay more is a significant competitive factor. Manufacturers are faced with a business choice: bear the up-front costs and risks of investing in technologies with the aim to cross-license for much of the essential IP required, or pay to license others’ IP. Investing up to several billions of dollars per year in R&D in the hope that some of it will prove effective enough to be accepted in leading mobile standards merits competitive benefits and commercial returns. Nevertheless, latter-day cellular market entrants including Research in Motion, HTC, Apple and others succeeded with little or nothing in the way of essential IP at the outset".
Keith Mallinson‘s recent clients include several mobile phone technology IP owners. His work includes various other commercial issues as well as IP. He provides advisory services including market analysis and forecasts for operator services, network equipment and handset. He also has significant testifying expert witness experience in the cellular sector, but has not yet testified on matters relating to standards-essential IP.