08 November 2012

Fragmented Patent Pools Will Not End Smartphone ‘Wars’

Photograph © Mark Summerfield 2008Depending on where you are in the (developed) world, the next generation of mobile communications networks is either already with you, or on its way.  Colloquially known as ‘4G’, it is more accurately termed ‘LTE-Advanced’, an upgrade from the Long Term Evolution (LTE), or ‘3G’ technology which is now widespread.  LTE-A is an improvement in every respect over 3G – higher peak data rates, greater spectral efficiency, reduced latency and increased throughput.

And, of course, there are more patents than ever before covering networks and devices capable of communicating using the full range of available standards.

Considering that the holders of patents covering the previous generation technologies seem unable, or unwilling, to resolve their existing disputes over access and licensing terms for access to standards-essential patents (SEPs), it may be difficult to imagine how there could ever be peace as the number of such patents – and associated patent holders – only seems to be increasing.

The creation of ‘patent pools’ – collections of SEPs owned by multiple proprietors, and made available as a block via a single licence agreement – may therefore be seen as an encouraging sign.  Two such pools have recently been established covering subsets of the LTE SEPs.  And while we would like to think that this is the beginning of a more peaceful coexistence among competitors in the mobile communications space, a closer look at the LTE patent landscape suggests that there is a long way to go before these pools will represent a realistic option for achieving a truce in the industry.


When we looked at the issue of SEPs, and requirements for licensing on ‘fair, reasonable and non-discriminatory’ (FRAND) terms early this year (see our three-part series commencing with The Fraught Issue of FRAND I: A FRAND Primer), we suggested that patent pools could be a significant part of the solution to the current disputes.  However, we placed an important caveat on this suggestion, noting (towards the end of part III in the series) that:

While there seems little prospect of all of the parties getting together to form a single pool of patents accessible to everyone, the formation of a manageably small number of pools – perhaps three or four – seems realistic.

If the number of pools is not small, or their coverage of the necessary patents not reasonably comprehensive, the improvement over the current situation will be only marginal.  Companies requiring access to all of the relevant SEPs will still need to obtain licenses from multiple sources, and the cumulative royalties may remain prohibitive to those prospective market entrants who bring no valuable patents of their own to the table (a problem commonly termed ‘royalty stacking’).

In this context, recent reports of the formation of two new pools comprising LTE SEPs are of interest.  On 3 October 2012, US-based technology licensing company Via Licensing launched its LTE patent pool, comprising SEPs owned by AT&T, Clearwire Corporation, DTVG Licensing, HP, KDDI Corporation, NTT DOCOMO, SK Telecom, Telecom Italia, Telefónica, and ZTE Corporation.  Then, on 5 October 2012, European patent aggregator and licensing company Sisvel announced its own pool comprising patents owned by Cassidian, an EADS company, the China Academy of Telecommunication Technology, the Electronics and Telecommunications Research Institute, France Telecom, TDF, and KPN, as well as patents originally filed by Nokia that Sisvel acquired in late 2011.


Interestingly, both pools are being open about the licence fees for access to their respective portfolios.

Sisvel’s model is very simple – a flat royalty rate of €0.99 (about US$1.30) per unit.

Via’s fee structure is a little more complex, with different rates for different types of device, and discounts for higher volumes.  A useful point of comparison, however, is the royalty rate of the first half-million ‘terminal devices’ (i.e. smartphones, tablets, or similar) each year, which is US$3.00 per unit.

So any manufacturer taking licences from these two pools initially is already up for US$4.30 per unit.  And, frankly, the coverage provided by these licences falls well short of what is actually required to implement the relevant LTE standards.


Patent analysis and consulting firm iRunway recently published a report of its study of the 4G-LTE technology patent landscape'.  The table below lists the top 14 patent holding companies amongst the major industry participants identified in the report (assuming the Motorola holdings are combined), the total number of 4G-LTE patents held by each, and the percentage share of all 4G-LTE SEPs.

Top LTE Patent Holders

The iRunway report notes that the Nokia patents were acquired by Sisvel in January 2012, while the Nortel patents were acquired by the Rockstar consortium (Apple, Ericsson, RIM, Microsoft and EMC) in 2011.  According to some reports (e.g. here), Apple has acquired control of Nortel’s LTE patents as part of this deal.  Motorola continues to own its LTE patents, however it is, in turn, now wholly owned and controlled by Google.

In addition there is one significant non-practising patent holder which is not listed in the above table.  The non-profit government-funded Korean research organisation ETRI (Electronics and Telecommunications Research Institute) holds 249 LTE patents, which it licences to a range of companies.

The main message here, however, is that aside from the 2.33% of LTE SEPs now held by Sisvel along with the 1.99% owned by ETRI (also in the Sisvel pool), not one other of the patents represented in the above table is covered by either the Via or Sisvel pools.  These comprise just over a third of all LTE patents, with the remaining LTE patents being owned by around 300 minor contributors, of which only 15 are currently participating in the two pools.


To put it bluntly, then, a fee of US$4.30 per unit currently buys rights through the Sisvel and Via pools to use about 5% of all LTE SEPs.  Assuming all patents are of equal value, licences providing full coverage would cost over US$85 per device.  However, this would be an extremely conservative estimate. 

The iRunway analysis takes into account the fact that all patents are not equal.  In fact, iRunway estimates that Samsung and Qualcomm between them own around 25% of the most valuable LTE SEPs (which they call ‘seminal patents’).

Adjusting for relative portfolio value would most likely therefore push the total royalty estimate well above US$100.  This price makes Samsung’s reported offer of a 2.4% royalty rate to Apple look to be within a ball-park established by the Sisvel and Via patent pool rates.  Even if it is a little high, it is not, comparatively speaking, an unreasonable early offer in the course of a good-faith negotiation.

Of course, this assumes that the current pool rates are ‘reasonable’.  Clearly, however, they are not.  The idea that more than half the cost of a low-end smartphone or tablet should be payable in patent royalties is manifestly untenable.  At some point, the industry as a whole is going to need to sit back and take a long, hard, look at the issue of reasonable royalty rates for SEPs in the mobile communications space.  And, at a rough guess, these rates are going to have to come down to no more than about a quarter of the price that major LTE patent holders currently appear to believe they are worth.


Despite some excitement at the formation of the first LTE patent pools, it is fairly clear that, as of right now, these pools are not going to resolve any of the current issues in the industry.

The Sisvel and Via pools between them account for no more than about 5% of the total number of LTE SEPs, and the existing royalty rates are excessive relative to the value of the licensed patents.

Qualcomm alone controls roughly the same proportion of LTE SEPs as are covered by the two pools combined, and Samsung owns nearly twice as many.  There is currently absolutely no incentive for either of these companies – or indeed most of the other top market participants (i.e. practising entities) – to join in with either of the pools.  This would serve only to dilute the value of their portfolios, and reduce their level of control over licence fees, terms and conditions.

Yet, unless the vast majority of SEP holders agree to participate in a very small number of pools (i.e. no more than four or five), the dream of easy, affordable and substantially complete access to patented LTE technologies will remain out of reach.  Every licensing program has associated overheads and inefficiencies, and the total stacked royalties payable on a large number of licences from small pools and individual patent holders will most likely continue to exceed the real market value of the aggregate portfolio, as measured in terms of the price consumers will be prepared to pay for LTE-enabled devices.

Patent pools are likely to be at least part of the resolution to the current disputes, but only if the industry is able to overcome the problem of fragmentation.


Post a Comment

Copyright © 2014
Creative Commons License
The Patentology Blog by Dr Mark A Summerfield is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Australia License.