How Streaming Will Impact Music Sales

With 2013 now behind us we are beginning to see the first full year sales numbers come if for 2013 and the long anticipated ability to assess the impact of streaming on the market.  Until the IFPI annual revenue numbers come out we are mainly constrained to volume data which only paints half of the picture.  This is especially true for streaming given the massive difference in revenue per stream for free versus paid, YouTube versus Spotify etc.  But even within these constraints we have enough to start establishing a view, one that indicates the headline story may be more about transition than it is growth.

Nielsen’s numbers for the US show that digital track sales were down 5.7% and that digital albums were down 0.1% while albums as a whole were down 8.4%. In the UK the BPI reported that digital track sales were down 4.2% though digital albums were up 6.8%.  Nielsen also reported a 103% rise in audio streams.  Let’s assume that a significant portion of those increased streams will be coming from free users and that the impact on streaming revenue growth will therefore be around the 65% mark. That would translate into total US music market revenue growth of just under 1%, though if free usage is a bigger part of the picture then growth could be negative.

It is important to understand the appropriate context for the shift to streaming: it is fundamentally a transition of spending.  Just as the download was a transition from the CD so streaming subscriptions are a transition from the download.  This is because the majority of subscribers were already digital music buyers before becoming subscribers and the majority of those were iTunes customers.  50% of subscribers buy album downloads every month and 26% buy CDs every month (see figure).  On the one hand this can be interpreted as the fantastic capacity of streaming to drive discovery and music purchasing.  There is some truth in this, but it is an inherently temporary state of affairs.  If streaming services do their job well enough there should be little or no reason for a subscriber to additionally buy music.  They do so because consumers transition behaviour gradually not suddenly.  The fact that a third of download buyers still buy CDs illustrates the point.

In this respect streaming services are strongly competitive with music sales in a way that streaming radio services are not. However what is crucially different from the CD transition is that while downloads drove a decrease in ARPU with consumers cherry picking single tracks from albums, subscriptions drive ARPU upwards. So there is more of an opportunity for subscriptions to drive longer term revenue growth than downloads.  The two key questions that arise are:

1. What download market will be left once/if subscriptions have reached scale?
2. What will the net impact on digital music spending be?

1 – Impact on downloads: The answer to the first question is probably the most straight forward.  Looking at markets like Sweden and Denmark we have strong evidence that streaming subscriptions grew at the direct expense of downloads, but in doing so they transformed the total music markets.  In the US, where the download sector is much more entrenched, streaming has resulted in a worst of both worlds, with streaming eating into downloads but not having enough headway to transform the market Sweden style.  The outlook for downloads in big markets such as the US, UK, France and Germany will be one of subscriptions absorbing the spending of the most valuable download customers.  Downloads as a global sector though will remain strong because they are the natural transition technology from download and will thus have strong long term opportunity in emerging digital markets of scale such as Turkey, Brazil and Mexico.  Downloads will also remain the best tool for monetizing mid tier digital music consumers who like to buy a few singles and the occasional album but do not spend 9.99 a month on music.

2 – Net impact on music spending: This one is a tougher call to make.  If subscriptions only reach scale by converting the most engaged music consumers then there is a risk of reducing ARPU among some of them, changing their spending patterns from buying a few albums a month to spending the equivalent of just one.  This effect will be felt more strongly as the dual-consumption behavior of subscribing and buying naturally fades.  The net positive opportunity lies in converting large swathes of the ‘upper middle’ tier of music buyers with more competitive pricing and also with bundles. Though this will likely come at the expense of further erosion of downloads.

As the RIAA rightly highlighted, even in the US streaming is becoming a really important part of the music market, and there is no doubt that access based models of shapes and sizes are the future.  The next few years though will see some growing pains as we transition away from the old guard in some of the world’s biggest music markets.