Yesterday, Michele Boldrin presented a seminar at the University of Melbourne on his on-going work with David Levine on the need for intellectual property protection. (You can access that work here). Their work is single-themed: economists have typically presumed that under perfect competition, if innovators are not protected with some degree of market power, then they will have no incentive to innovate.

The argument is based on a Bertrand view of competition between imitators and innovators that results in no profits being earned by either and hence, nothing left to pay for sunk production costs. However, if competition does not lead to zero profits as it may do under Bertrand competition with diminishing returns, competition with capacity constraints, or other forms of competition with product differentiation, then this extreme scenario doesn’t hold and protection of the innovator from competition is not necessary for the innovator to have an incentive to innovate. This is a point that Telstra do not seem to understand and it is welcome to see economists bringing it to wider attention. (For non-economists, David Levine and Eric von Hippel will present these arguments in an IPRIA forum here at Melbourne Business School on Friday, 11th August; click here for details). 

From this perspective, there isn’t alot that is truely new in the Boldrine-Levine paper but they are perhaps the first to set the strict goal of showing how you can earn innovative rents under the economist’s extreme definition of perfect competition. What is a little neater is that they do not rely on Bertrand competition with capacity constraints (or even the core with capacity constraints as emphasised by my colleague Michael Ryall in a 2004 Management Science paper with Glenn MacDonald that makes this point). Their motivation is that it is costly to distribute information and ideas and that these costs mean that copying by imitators cannot be instantaneous and has frictions. (Think of someone buying a CD and then making pirated copies to sell; that can take some time). This friction protects the innovator but also protects the imitators too. As a result, early imitators are a source of revenue for the innovator because they will pay a premium to enter into the competitive fray. To be sure, innovators would prefer IP protection but even without it can make some dough.

This has two implications: one which Boldrin and Levine point and and one which they do not. First, they show that as copying technology improves, then it is possible that innovative rents could rise rather than fall. Why? Because early imitators are paying for their ability to compete by making copies. If you improve their ability to make those copies they actually may pay more for the initial master. This is the sort of counter-intuitive result that economists love.

Second, (and this they do not note) as copying technology improves, this will change the structure of pricing. Basically, in their world, prices decline over time with competition. However, when copying technology improves, the innovator can earn more by pricing higher up-front even though prices fall rapidly thereafter because of the improvement in copying technology. This provides an opportunity to test whether frictions in idea diffusion are a constraint that assists earning innovative rents. Just look at the proliferation of copy technology in recent times and see if the impact on prices has occurred.

When pushed Michele Boldrin seem to be in favour of regulation of IP monopolies much the same as we would regulate other monopolies — say through compulsory licensing. Again this is a view that would not be uncommon amongst those of us who spend our time on competition policy issues but it is nice to see it being pushed enthusiastically.