Do Patent Pools restrict innovation?

This post on TechDirt by Mike Masnic agrees with recent research by Stanford academics Ryan Lampe and Petra Moser “Do Patent Pools Encourage Innovation?
Evidence from the 19th-Century Sewing Machine Industry” that patent pools restrict innovation, essentially by "locking-in" innovation at a level beyond which there is no incentive to improve.

Patent pooling is increasingly common in high-tech where thousands of patents from multiple owners have to be licensed in a device such as a mobile phone. Patent pooling such as in this announcement "Patent Pooling Could Reduce LTE Competition" for LTE next generation mobile technology are meant to ease the traditional bilateral negotiations between patent owners - to agree fees, avoid litigation and reduce product release delays – all potentially expensive outcomes.

But are they good or bad for innovation? As Mike says,

“The core problem, once again, is confusing innovation (finding something that the market actually wants in a way that it wants it) with invention (coming up with something new). When you get a patent, or set up a patent pool, what you're effectively doing is declaring a stop on any incremental improvements above that. It ignores the fact that real innovation is an ongoing and never ending process of improvements and tweaks, often made by others. But, in blocking out the ability of others to make those improvements -- those real innovations -- you slow down, drastically, the pace of innovation. In the end, a patent pool is just like a bigger patent, and has the same negative impact on innovation, just on a larger scale.”

So What?

I don’t disagree with Mike that patent pools may restrict further innovation in their own domain but he seems to forget the fact that in at least some domains, like mobile, it is essential to have a stable and economic platform at a viable cost in the market and patent pooling exists to accelerate this process. Further activities in the mobile value chain depend on it, such as applications and content, and given the complexity and pace of change in products is there a viable altenaive?

For example, without an economically viable standard for LTE (for which the patent pooling announcement above is intended to help deliver) ‘dependent’ innovation in operating systems, applications and content technologies will be delayed, more expensive or less universal as a result and thus delaying uptake, usage and value to the consumer – the ultimate purpose of innovating in the first place.

Maybe patent pools are a constraint on innovation but maybe they are necessary if we are to generate innovation in other domains further up the value chain? Their use may help to release this innovation more quickly, cheaply or widely and as a result an inferior innovation (or standard that results) is a price worth paying?