Friday, October 24, 2014

Empirical IP Research Conference: TM counterfeiting

Trademark breakout session #2: Counterfeiting and Its Effect on the Market for the Genuine Article

Facilitator: Scott Hemphill (Columbia; Visiting Professor, NYU)

Why we care: innovation policy; enforcement policy.  Debate over magnitude: 10s of billions, 100s of billions in the US?  Is that all damage?  Lost profits: what monetary effects does it actually have on TM owners?

Counterfeiting scenarios: (1) seller fools the buyer: drugs, aircraft parts.  Mostly we aren’t talking about that, but rather (2) buyer fooling everyone else: handbags, sunglasses, footwear.  Or (3) nobody is fooled. Is that even counterfeiting? Dilution?

Standard story: substitution. Though substitution is highly contested.  Overuse—becomes uncool. Positive stories: Sampling: I try it, I like it, and ultimately I buy the original (as with sampling in music downloading).  Advertising: Might drive word of mouth.  Still might be a case for enforcement—the legitimate seller still needs some way of recovering; illegality can be a form of price discrimination by segmenting the market.

Sprigman: law on books v. law on streets. In criminal counterfeiting, DOJ maintains and most courts accept that once they prove the D’s mark is identical to/substantially indistinguishable from the registered mark, confusion is conclusively presumed.  Civil counterfeiting plaintiffs are now taking the same position.  Real example: if I take a wallet and fake up an expensive LV emblem, and stitch “fake” all over it, so no consumer would be confused, the gov’t says that’s criminal counterfeiting.  The Q of harm from counterfeiting has to be assessed w/in that framework.

McKenna: Other ways in which context matters: many kinds of products can be counterfeited, and some are much more harmful than others. Yi Qian has empirical evidence: similarity of price matters a lot. Where price is quite different, advertising effect tends to swamp substitution effect, but reverses as prices converge.

Joel Waldfogel: Deadweight loss can turn into consumer surplus if I can get a cheap copy of a handbag; no loss unless other people react badly to the fact that I own a copy.  Can look across products: but are there systematic data for fake Gucci purses?  Can look across individuals to see what their stock of fake/real stuff is. Do people who buy fake stuff start buying less real stuff over time?  (Drawing on music downloading studies.)

Stefan Bechtold: can present consumers with well-known brands and knockoffs whose similarity is exogenously varied. Ask about perception of knockoff and brand—determine whether there is some advertising/spillover effects.

Mike Meurer: Child labor/slave labor producing counterfeit goods? 

Sprigman: terrorism, narcotics, organized crime: it’s a problem w/manufactured goods generally!

Rahul Telang: comparative pressure on original seller to innovate to avoid copying? Pressure to lower prices that could be good?

Hemphill: Qian gets into that in her work on footwear in China. Firms most affected by counterfeits respond by going up vertically in quality—fancier, invest in vertical integration.  Natural experiment in that 1995 marked an exogenous fall in TM enforcement as inspectors shifted to tainted food. Also some companies were more in with the Chinese gov’t than others, and could get more help—measured distance from gov’t. At the high end advertising effect dominated over substitution, opposite at the low end. One of the Megaupload papers finds an effect like this: substitution dominates for blockbusters but advertising dominates for smaller movies.

Sprigman: The potential is for lower cost advertising when other people sell copies than when you advertise for yourself. Other people give you credibility by choosing to consume the counterfeit. But the proper function of law here is a relevant consideration. 

Telang: Bill Gates says, if you copy any product, copy mine.  I’ll eventually get you to shift over/buy complement.

Hemphill: generating some illegality may enable your price discrimination.

Orly Lobel: Different things you can do in this realm: Dishonesty of honest people.  Who is willing to do what?  Also can measure willingness to pay: and can measure deliberate ignorance/degree of certainty about whether something is a counterfeit.  Also: secondary markets. If there’s less demand/supply for counterfeits that can affect the size of the secondary market.

Hemphill: WIPO 2010 has a literature review on empirical studies—a few in the area of why people buy.

Glynn Lunney: categories of counterfeits. If the law allowed counterfeiting when no direct purchaser is confused the line might become more blurred.  Could become difficult for buyers to tell the difference.  If substitution and overuse are balanced by sampling/advertising, we should be able to allow the TM owner to enforce and TM owner would enforce efficiently; the question is consumer welfare—deadweight loss, desire for prestige brand.

RT: Dan Ariely on honesty—if you feel like a fake, you behave more dishonestly: another separate effect and it’s not clear how much law affects this but to the extent law affects magnitude it may matter.  Also, Lunney’s theory is unlikely to materialize because TM owners will overenforce: they will often see things in a property frame whether or not allowing some counterfeiting would be beneficial, if they have the right to enforce and are not forced to give up on the idea of control by practical realities.  (See Jessica Silbey’s work.)

Strandburg: might be that counterfeiting doesn’t matter as long as it’s illegal. Companies might want it to be illegal while being satisfied with a certain level.

Hemphill: much variation in reactions to buying counterfeits.  122 attendees at counterfeit purse parties, with no previous exposure to brand. 46% subsequently bought an original from the same line. Some felt very bad about having bought/being complimented.

McKenna: note how much this conversation doesn’t have to do w/TM law.  The trouble is situations we all know there’s no confusion.  It either is or isn’t another problem. Ought to be a question of whether we want copyright or design protection. If the mechanism of harm isn’t confusion, it doesn’t belong in TM.  Sui generis design choices are very different.  If we all recognize that there are problematic forms of counterfeiting, that’s where it’s most likely the case there’s genuine confusion.  So long as TM does its traditional role, no problem. Next we ask whether TM should address another type of product.

Sprigman: maybe we know it’s not confusion, but INTA doesn’t know.  There aren’t studies that fully contextualize shopping experiences. Need to know more about shopping intention. Lack empirical infrastructure to say where confusion really is.  His intuition agrees that in large swathes people know what they’re getting and it’s just reducing deadweight loss, but there’s work to be done.

Strandburg: relates to panel session: confusion is all related to context, which includes price, but no TM study will ever ask whether people think the $20 thing comes from Louis Vuitton: decontextualization of confusion.

No comments: