Friday, July 20, 2012

How to lie with statistics: Netflix percentages not false/misleading, just framed favorably


Cullen v. Netflix, Inc., 2012 WL 2906245 (N.D. Cal.)
(I am ignoring the ADA and related claims.)  Cullen alleged that Netflix engaged in false advertising under California law through omissions and misrepresentations about its streaming video library's closed captioning, which had the “effect of conveying to Netflix's deaf and hard of hearing members that Netflix would meaningfully subtitle its streaming library within a reasonable period of time.”  In fact, Cullen alleged, its streaming video library was effectively useless due to the small amount of captioned content, lack of support tools, and slow captioning rate.
“Meaningful” and “reasonably” were subjective and vague words of puffery.  The court looked at Netflix’s specific statements.  On a Netflix blog, a June 2009 entry said that Netflix would “expect to deliver subtitles or captions to Silverlight clients sometime in 2010, and roll the same technology out to each CE device as [Netflix is] able to migrate the technology.”  Cullen alleged that this falsely attributed failure to caption to technical difficulties, but didn’t sufficiently plead falsity.  The complaint alleged that other providers were captioning with existing tech, which Netflix also could have used, but “the fact that another, better technology existed does not contradict the representation that Netflix's captioning rate was limited by technical difficulties, much less state a plausible claim that Netflix's statement was false or misleading.”
Cullen also identified other statements between October 2009 and November 2010, such as that Netflix said it would “continue to work on closed captioning,” that there was “much more to come,” similar technology would soon be released for Netflix's other platforms, and closed captioning was offered on a growing number of titles. Again, the complaint failed to sufficiently allege falsity, since the complaint alleged that Netflix’s captioning did speed up at least a bit. 
Cullen challenged a specific February 2011 blog statement that there were “more than 3,500 TV episodes and movies” with subtitles in Netflix's streaming library “representing about 30% of viewing.  More subtitles are being added every week and we expect to get to 80% viewing coverage by the end of 2011.”  However, the complaint alleged that only 6% of the streaming video programming was captioned.  The court noted that the captioned percentage of total titles might be smaller, but Netflix gave figures for video viewed, which of course skewed popular.  (The court did reject Netflix’s argument that a “we expect” statement wasn’t actionable, since the caselaw didn’t exclude statements about future facts from the coverage of the California consumer protection statutes.)  In opposition to the motion to dismiss, Cullen argued that a reasonable consumer would assume that the number meant that 30% of titles were captioned and that 80% would be (after all, an individual consumer probably wants to know how much is available to her; even if every Netflix consumer watches Breaking Bad such that the show is a big percentage of Netflix’s streaming time, the consumer is more likely to want to go on to the next thing than to rewatch).  But the complaint didn’t allege that these statements were misleading by being usage-based, though the court did grant leave to amend.
Cullen also argued that failure to caption was an unfair practice under the UCL, creating a “deaf tax” “because the DVD-by-mail plans that provide sufficient access to video programming were sold at a significant premium to Netflix's streaming subscription.”  However, Cullen failed to allege facts about the utility/benefits of charging higher prices for the more accessible DVD-by-mail plans, and thus hadn’t plausibly alleged that the gravity of the harm outweighed the utility of the challenged activity as required to make the price difference immoral or unscrupulous.

No comments: