Arrow and Demsetz

This paper is full of interesting insights and ideas, and offers a wonderfully rich set of issues for discussion. Here are two questions that occurred to me.

First: The paper begins with the question, "Why do we have creative fashion industry, if the industry’s main creative output is not covered by an exclusive rights regime?"  But before we know that this is a puzzle, we need to know two things: (1) how much innovation (and copying) in fashion costs, and (2) whether those costs can be recovered by existing appropriations strategies, such as lead-time or status-conferring (see Wendy Gordon’s post) trademark.  Recall the classic work done by Levin and Klevorick et al. about appropriations strategies across industrial fields.  Their surveys show that pharmaceutical firms rely much more heavily on patents as an appropriation strategy than other industries surveyed.  The lesson usually drawn from this is, “That makes sense, because pharma is a field where R&D is very expensive and copying very cheap.”

Returning to fashion, if R&D is cheap (or copying relatively expensive), or if firms can recoup their investment in Karl Lagerfeld’s salary using something like trademark, then there is no economic dilemma, and its not surprising that we have creativity without IP.   

My intuition would be that the costs of innovation in this area are relatively low - but this is nothing more than an intuition.  I suspect that Kal and Chris will have something more informed to say here, so I’ll move on to trademark.  I read the paper to suggest that trademark is not a particularly powerful appropriations scheme in fashion because its not often legible from the outside, and fashion is a positional good.  I very much like the “positional good” argument, but I’m not sure that I’m persuaded on the trademark point for two reasons.  One, there are lots of ways to show trademark to people who matter without it being visible on the surface (“Don’t you love them?  Prada….   I know, I shouldn’t! But I couldn’t resist!”).  Two, fashion is probably also a “self-positional good”  -- i.e. the target of at least some of the positioning is the consumer him or herself.  People buy Prada to be part of the Prada universe, to give themselves a treat, or for a million other reasons that would not be satisfied by cheap knock-off versions.  The medium is the message, so to speak. 

The bottom line is, I’m not sure there’s much reason to think there would be market failure in creative fashion, from an information economics perspective.  Thoughts?

Second point:

Another way to understand the core question of the paper is through Demsetz.  A Demsetzian might say we should see property rights in fashion designs if the value that could be internalized through such protection is greater than the cost of the legal regime.  Many of the paper’s most interesting claims suggest that exclusive rights in fashion designs are likely to be a lot less valuable to innovators than many think (e.g. the “induced obsolescence” point, or the “everyone copies someone” point).  Many of these points are really insightful, but if the argument is about Demsetz (or even if we’re back the land of information economics), I also want to know more about the costs of the imagined legal regime.  My sense is that these costs might be relatively high compared to the benefit, even without factoring in theories of induced obsolescence etc.  While the authors show that design protection for fashion is doctrinally plausible, it also seems plausible that (1) the doctrine might be vague, making litigation costly, and (2) courts might be much slower than fashion cycles.  So why isn’t the cost and torpor of the possible regime the best explanation for the lack of IPRs here, and also the lack of uptake of the EU system?