Thursday, 10 September 2015

Sympathy for the Mechanical

Although I’m not always sure of the motivations, there is a politics to the reporting of music industry sectors. In particular, there has been a desire to emphasise the health of the live music industry at the expense of recorded music.
The statistics that have been most widely used in support of this scenario come from Adding Up the Music Industry, a series of reports issued by PRS for Music. Sadly these reports are no longer compiled: the last of them concerns revenues for 2011. It calculates the total money generated by the UK music industry in this year at £3,793m. Out of this figure £1,057 comes from business-to-business income and £2,736m from business-to-consumer. The latter is divided into £1,112m for recorded music and £1,624m for live music.
And so live music triumphs over all other comers. It is this outcome that many analysts have taken up and run with. The reports have their problems, however. Live income is compiled in a different manner to the other streams. As well as documenting the primary market – the money spent on purchasing tickets from ticket agents and venues – the figure includes secondary ticketing and ancillary spend. Secondary ticketing is money derived from the resale of tickets. Although the report states that this business model is ‘legitimate under UK law and is an established practice’ it is somewhat contentious. Moreover, none of the money from these sales goes to writers, musicians, publishers or record companies. At the very least this income is comparable to the second hand sale of records, an income stream that is omitted from the figure for recorded music. Under a different legal system it could be considered more akin to bootlegging. The report calculates its worth at £208m.
Ancillary spend is the extra money that is spent at gigs and festivals: the purchase of ‘merchandise, food, beverages, parking and public transport’. It is questionable whether the money spent on food, drink and transport should be included as part of music industry income. Some money from these sales might trickle through to a few performing artists, but it will be tangential and minimal, and in these instances should be included in the business-to-business figures, rather than business-to-consumer. Moreover, although these sales are included in the live income stream, money spent on food, drink and parking when going record shopping is not included in recorded music. This is the case even when the record shops are serving food and drink themselves.
Artists can make money from merchandise. There are famous cases where bands have made more money from t-shirt sales than they have from ticket sales. It is unfair, however, to include the sales of merchandise in the live tally while omitting it from recorded music: these reports do not include the sale of t-shirts and other merchandise in record shops.
It is difficult to calculate how much ancillary spend is worth as the 2011 report fails to provide precise figures. The best it offers is that festivals and arenas each account for ‘around 25 percent of the market’ and that ancillary spend at festivals is equal to ’95 percent of the average face value ticket per person’ while the ‘level for most other venue sizes was between 35 and 50 percent’. At the very least, then, ancillary spend is equivalent to a third of the money spent on primary tickets. The live income stream can therefore be broken down into £208m for secondary ticket sales, £472m for ancillary spend and £944m for primary ticket sales. It is only the latter figure that can effectively be compared with the statistics for recorded music. Looked at this way, live income is the lower of the two.
Recorded music is unfairly represented in other ways. In 2011 PRS for Music reported £65m income for ‘mechanical revenues’, i.e. the money that songwriters and publishers earn from record sales. Adding up the Music Industry contrasts the ‘continued decline of the recorded music market’ with the ‘phenomenal revenues’ earned in the live sector. However, live income for PRS for Music was £23m, which is lower than the mechanical income derived from its MCPS alliance.
In addition, the summarising table in the report equates ‘recorded music’ solely with business-to-consumer income. The total of £1,112m is made up of ‘payments for physical music products, downloads-to-own and subscriptions’. This obfuscates the fact that there is plenty of recorded music income that is derived business-to-business. A substantial amount of the £448m that is attributed here to PRS for Music comes from the use of recordings. Away from the main breakdown of figures, the report itself allocates £101.6m of PRS income to ‘recorded music’. Sound recordings also contribute to its other income streams: broadcast & online, public performance, and international. It should be conceded that live music also contributes to each of these remaining areas of business-to-business income, albeit that it is recorded music that dominates when it comes to the income generated by radio, television, the internet and the use of music in public premises.
Moreover, while PRS for Music owns the performance right in its members’ works, MCPS does not have complete jurisdiction over the mechanical right. Its members can opt to self-licence the use of recorded music for films, adverts and some television programmes. As such, recorded music makes up a significant proportion of the music publishers’ direct income. Out of the £210m that is allocated to them in this report, £48m comes from these sync rights. Ultimately, the mechanical income for songwriters and publishers is holding up better than much industry reporting would lead us to believe.
There is, in addition, plenty of business-to-business income that makes its way to record companies and recording artists. Recorded music is responsible for all of the £80m income that is attributed to PPL and the £220m accorded to ‘record label direct revenues’. The latter figure, in fact, includes some PPL revenue, as it is made up of ‘music synchronisation, “360 degree” artist deals, concerts, music-related TV production, broadcasting and public performance’.
Because of the way the figures are reported, it is impossible to make an accurate tally for the income derived from sound recordings. Recorded music is, however, worth considerably more than the £1,112m highlighted in Adding up the Music Industry. For the vast majority of professional songwriters and musicians it provides more money than live music does. While the same is obviously true for record companies, it applies to music publishers as well.
There are good reasons why some academics have seized upon PRS’s analysis: they like to emphasise the freedoms of the live music scene in comparison with the tyranny of being signed to a record company. PRS also has a performance emphasis. It is, after all, the ‘performing right’ that is enshrined in the society’s initials. Unlike the Musicians’ Union, however, it does not campaign to keep music live: it collects money from the performance of records as much as it does from performance in person. The reasons why PRS have chosen to under-represent recorded music in their reports are therefore obscure. Finally, I should concede that I have my own biases. Although I do think I’ve provided a fairer way of reading the statistics, I also know that I’m a record man. My main pleasures in popular music have always come from recordings. 

No comments:

Post a comment