First is the Pandora study that found a small "Pandora Effect" among its listeners. Using a sort of A/B analysis that prevented listeners in some areas from hearing certain songs, Pandora showed that listening to music on Pandora has a positive effect on music sales. For both new releases and catalog, listening to a song on Pandora has a slight but positive effect on music purchases. What's more, the "Pandora Effect" grows as a song is streamed more.
Not all listening is free at Pandora. But the service's roughly 3.3 million subscribers are a small part of its roughly 76.5 million monthly listeners. And while Pandora carried out this study to influence the Copyright Royalty Board's decision of royalties for 2016 to 2020, the results suggest the value of free music is greater than the royalty received.
Second is a study by the CMA that found that streaming services such as YouTube or Spotify are three times better at driving music sales than AM/FM radio. About 25 percent of survey respondents purchased new music after streaming it compared to 8 percent of radio listeners.
Of course, not all of this streaming is free. Spotify and Pandora both have paid tiers, and services like Rhapsody and Beats Music are available only to paying customers. But the two streaming giants are free or mostly free. YouTube is completely free (for the time being) and Spotify has more free than paying users. In the context of free music, this survey's takeaway is free music has a positive promotional benefit.
Third is a variation on the notion of a promotion effect. A new user study from BitTorrent says BitTorrent users are more likely than average Internet users to pay for a music subscription and buy digital music. This theme -- peer-to-peer downloads spend more than average people -- isn't a new one. Great attention was given to a 2006 survey on the impacts of peer-to-peer downloads on the Canadian music industry found that music purchases increased 0.44 CDs for every 12 songs downloaded via peer-to-peer services. More surveys and studies have since touted the buying habits of peer-to-peer downloads.
BitTorrent is different from streaming services that license the catalogs of record labels and publishers. It is a popular tool for peer-to-peer file sharing, and lately it has made the jump into paid downloads. And its study shows correlation, not causation. In other words, BitTorrent users' purchases may or may not occur because of the free content they download. But the company's survey drives home the point that some active music consumers engaged in both paid and free options.
The larger question is if free music is a net positive or a net negative to the music business. It's a complicated issue. The benefits of free music must be weighed against its possible detriments. For example, consumers could consider a free but personalized music service -- one in which users create channels based on favorite songs or artists, for example -- to be an adequate substitute for a paid option. Indeed, some music industry trade groups want to tighten the definition of "non-interactive" that currently applies to webcasters like Pandora. A change in this legal definition could require personalized, non-interactive webcasters to pay a higher royalty than the less-personalized webcasters.
Even if streaming confers a promotional effect, many questions still arise. Do streaming services induce consumption beyond purchases of downloads, CDs or vinyl LPs? What happens when music purchases are a small fraction of their current numbers? Will the services that claim promotional effects continue to boost revenues by, say, encouraging paid subscriptions? That question could be easier to answer when YouTube launches its paid service, Music Key. And answers will be easier in the coming years as services and label analyze data and market researchers have more time to track consumer behavior.
Until then, the debate will rage on.