Broadcasters and other digital media companies have recently been focused on the royalties that are to be charged by the record labels for public performance of a sound recording in a digital transmission (under the Section 114 compulsory license administered by SoundExchange).  In a Notice of Proposed Rulemaking issued this week, the Copyright Office tentatively concludes that there could be yet another royalty due for streaming – a royalty to be paid to music publishers for the reproductions of the musical compositions being made in the streaming process under Section 115 of the Copyright Act.  This notice was released just as the Copyright Royalty Board is concluding its proceeding to determine the rates that are to be paid for the Section 115 royalty.  While there have been reports of a settlement of some portions of that proceeding, the details of any settlement is not public, so whether it even contemplated noninteractive streaming as part of the agreement is unknown.

How did the Copyright Office reach its tentative conclusion?  First, some background.  The Office for years has been struggling with the question of just what the section 115 royalty covered.  Traditionally, the royalty was paid by record companies to the music publishers for rights to use the compositions in the pressing of records.  This was referred to as the "mechanical royalty" paid for the rights to reproduce and distribute the composition used in a making copies of a sound recording (a record, tape or CD).  These copies were referred to as "phonorecords."  However, in the digital world, things get more complicated, as there is not necessarily a tangible copy being made when there is a reproduction of a sound recording.  Thus, Congress came up with the concept of a Digital Phonorecord Delivery (a "DPD") as essentially the equivalent of the tangible phonorecord.  But just what is a DPD?

Section 115 contains the following definition of a DPD (with my highlights of specific terms or phrases that the Copyright Office addresses in it NPRM in some detail):

A “digital phonorecord delivery” is each individual delivery of a phonorecord by digital transmission of a sound recording which results in a specifically identifiable reproduction by or for any transmission recipient of a phonorecord of that sound recording, regardless of whether the digital transmission is also a public performance of the sound recording or any nondramatic musical work embodied therein. A digital phonorecord delivery does not result from a real-time, non-interactive subscription transmission of a sound recording where no reproduction of the sound recording or the musical work embodied therein is made from the inception of the transmission through to its receipt by the transmission recipient in order to make the sound recording audible. 

From that definition, the Copyright Office in its NPRM goes through a detailed analysis of the meanings of various words and phrases in coming to the conclusion that DPDs encompass not just digital downloads but also on-demand and noninteractive streams (noninteractive streams meaning Internet radio).  The Office first concludes that there is a "delivery" when a person receives a digital transition – seemingly pretty straightforward.

Then it looks at the language that seems to exclude noninteractive streams from the definition of a DPD.  However, the Copyright Office states in its NPRM that it believes that the exclusion only applies where there is no "reproduction" of the sound recording.  The Copyright Office looked at the buffer and RAM copies necessary to make the stream audible, and determined that these were in fact reproductions.  Thus, as the exclusion covered only services that did not make reproductions, Internet radio-type services that stream their music programming to their listeners were not within that exclusion.  .

The Office also looked to determine if these buffer copies were "specifically identifiable."  Essentially, the Office concludes tentatively that the computer being used by the ultimate recipient of the transmission (the listener) can decipher the buffer copies, so those copies must be specifically identifiable.  Essentially, the Office decided that, as the transmission can be heard, the copies must be specifically identifiable.

While the Office claims to be reading the plain language of the statute, one has to wonder if this is what was meant by Congress when the statute was adopted.  Seemingly, the statutory structure of the Copyright Act as amended to cover digital services was trying to equate real time, noninteractive streaming with radio type services, and creating a public performance royalty under Section 114 for the sound recordings and allowing the Performing Rights Organizations (ASCAP, BMI and SESAC) to take care of the composition rights for this real time streaming.  It would seem as if the concept of the DPD was meant to cover the digital equivalent of the pressing of a record or a CD – i.e. a download or similar one-to-one transmission that resulted in a fixed copy that could be used and re-used by the recipient.  To me, the concept of "specifically identifiable" would mean one that can be pointed to and identified, so that a listener can go find the song that they want when they want it, just as I can go to my CD collection, and find a Rolling Stones CD when I want it by looking at the label.  When I want to hear a particular song that is being streamed, I certainly can’t go to my buffer copies to find that song (as, even if it played on an Internet radio station an hour before, it is not there now if I want to hear it again).  But certainly these are issues that will be debated. 

We have written about the recent Court decision that determined that ASCAP had no public performance right in connection with a digital download, as it was not a "public" performance.  Only the music publishers would collect for the composition used in the song being downloaded to avoid double-dipping.  Here, applying Section 115 to noninteractive streaming would seem to be the mirror image of the ASCAP case, yet the Office is reaching a decision that is exactly the opposite of the one reached by the Court.  Of course, that ASCAP case may be appealed, so these issues, too, are unresolved. 

The Copyright Office seemed to recognize that this could be very controversial, and states that it takes no position as to the value of the Section 115 right that it finds to exist.  That would be left to the Copyright Royalty Board to determine royalty rates under Section 115  in its current proceeding – and that the Board could well determine that the copies made in the case of noninteractive streaming have no value at all.  However, one wonders if the issue was fully argued in the case as most parties seemed to acknowledge that the Section 115 royalty would not apply to such activities (the Board even citing in a footnote that the publishers supported such an exemption in a Section 115 reform bill that was introduced and passed by the Judiciary Committee but not adopted by the full Congress in the last Congressional session).

Comments in the Copyright Office proceeding are due on August 15, and Replies are due on September 2. This is a very important proceeding in which parties should make their views known.