Research and analysis

The potential economic impact of ER on performers and the music market in the UK

Published 19 February 2024

ISBN

978-1-915090-60-7

Executive Summary

Scope

The scope of this project was agreed with the Intellectual Property Office’s (IPO) Economics of Streaming Contact Group. The objective was to deliver a flexible interactive economic model to provide a foundational understanding of what Equitable Remuneration (ER) might look like if implemented in the UK.

The purpose of the work was to consider the impact of introducing ER to the investment environment for artists and labels investing in the creation of new recording copyrights. The purpose was explicitly not to consider the fairness of the current situation, nor how the introduction of the proposed changes might lead to an more or less fair outcome.

This work was restricted to modelling:

• the Full Broadcast Model
• the Partial Broadcast Model
• and the Spanish Model

The purpose of this work was not to conclude or provide specific recommendations, but rather to frame a range of disparate discussions on a consistent basis; to consider the impact of each model on different deal models in a repeatable way; and to lay the foundation for future research.

The focus was to consider the impact that potential changes might have, but excluded calculating the cost of implementing any such changes. As such, the model does not calculate the cost of the set up required to enable distribution of streaming royalties by a collective licensing body – and takes 10% as an indicative administrative rate as an accepted midpoint between different rates presented by stakeholders. The set up costs, running costs, and administration rates should be considered for further research.

Equitable Remuneration Calculator

An interactive economic model Equitable Remuneration Calculator is available for users to investigate the outcomes of different variables and different scenarios in relation to the potential implementation of any system of Equitable Remuneration in the UK.

Accessibility statement for Equitable Remuneration Calculator

We know some parts of the Equitable Remuneration Calculator website are not fully accessible:

• you cannot view content on a mobile device
• the website is not compatible with current and future user agents, including assistive technologies
• you cannot modify the line height or spacing of text
• you cannot zoom in beyond 100% without the text spilling off the screen
• there is no logical focus order

If you find any problems not listed on this page or if you need the information on this website in a different format:

• research@ipo.gov.uk
• 0300 300 2000

We’ll consider your request and get back to you in 5 days.

Context

In UK law, when a sound recording is played to the public over the radio, or performed in public, the performers in that recording have a right to “equitable remuneration”. This means the owner of the sound recording must give performers on that recording an equitable share of royalties, which usually means sharing them on a 50-50 basis.

In copyright law, a separate right – the “making available right” – applies to on-demand services. When the UK introduced the “making available” right into its law it excluded it from the scope of the right to equitable remuneration. In doing so, performers were given an exclusive right for making available and not a remuneration right.

In practice, this means that non-featured performers (often referred to as non-featured artists, and hereafter NFA) do not receive additional income from listening on streaming services, and the amount that featured artists (FA) receive will be governed by contract. With the increased attention on the allegedly inadequate amount of income for both FA and NFA from streaming, catalysed by the COVID-19 pandemic, applying ER has been suggested as a way to improve the financial conditions for performers. This has included proposals in a Private Member’s Bill called the Copyright (Rights and Remuneration of Musicians, etc.) Bill, also known as the ‘Brennan Bill’, after its sponsor, Kevin Brennan MP. 

This research sets out to fill some of the gaps in evidence by providing a look at the potential impacts ER could have on the market, and how those affect each of the various stakeholders.

To that end, two economic models are offered for ER: the “Traditional Broadcasting” model and the “Spanish” model. The Broadcast Model comes in two parts. The first considers a ‘Partial’ application, where it is applied only to the portion of streams identified as ‘lean back’ or ‘push’, which are considered less interactive and more akin to radio consumption, and a ‘Full’ scenario where ER is applied to all streams.

The models were built with the intention to capture all important variables. However, many variables are subject to change depending on the negotiations amongst relevant stakeholders, hence only the first order consequences are detailed within this paper. This paper does not look to offer a full resolution, nor to advocate for any outcome, but rather to enable stakeholders to see the impact of different variables and to consider the possible models that would work best in the market.

Deal assumptions and inputs

Initially, 6 deal structures were mapped out, and have been discussed at length with a range of industry stakeholders from the IPO Contact Group. These demonstrated how the impact of ER varies depending on the royalty share of the artist/label. These range from a legacy (10%/90%) and modern (26%/74%) advance deal, to a joint venture (JV) (50%/50%), label services (75%/25%), a distribution deal (85%/15%) and a DIY arrangement (90%/10%).

Each deal model aims to consider the full cost of the deal (including marketing spend and overheads) rather than only recoupable costs. The ratios between recoupable, non-recoupable and ongoing costs are flexible within the model and the published model is based on the assumptions from the best placed stakeholders from within the IPO Contact Group. 

The impact on artist and label income varies significantly depending on whether an artist has recouped their initial advance. As such, each record deal has 4 separate revenue scenarios modelled, giving a stylised loss-making deal, a breakeven deal, a profitable deal and a highly profitable deal. In the cases where an advance is not issued, comparable revenue numbers are input, in the same proportions.

Having created a wide range of models, a range of revenue scenarios, and a range of conditions under which ER might be applied, this paper then focuses on the impact that changing toward ER would have on the two key situations: a modern advance based record deal, and a DIY release. These exemplify the impacts of ER in the clearest way.

Its important when looking at the range of model outputs to consider that not all deal successes are equally likely and that the impact on a label should be viewed in the context of a portfolio of artists.

Key variables

ER introduces many ways in which shares of revenue could change. While there are a range of ways in which ER could be applied, the model reflects the current broadcasting rate, and therefore the rate between the label and artist(s) changes to 50%/50%. This could reasonably not be 50%/50%, but rather another rate giving more favourable terms to the artist(s), or to the label. This can be varied within the model to explore other outcomes and should be a consideration for future work.

The share of streaming income against which labels cannot recoup could also be varied, rather than being all streaming income. This is variable within the model, with the output from 35% of streaming (Partial Broadcast Model) and 100% of streaming (Full Broadcast Model) shared within the paper. 

The split of payments between the featured and non-featured artist could also be varied. This is currently 65%/35% in line with the broadcast splits, but there is flexibility to change the ratios. This can be varied within the model. 

The anticipated third party administration rate could change as the scope of the task changes. The range is likely to be between 15% (broadly in line with PPL’s current UK licensing) and 5% (broadly in line with Sound Exchange’s US online licensing). An indicative, mid-range rate of 10% is applied, and can be varied within the Broadcast Model. Future work should consider the full cost of setting up and running such a system.

Would ER impact the total value of streaming rights?

There is an important debate around whether a change in the law would lead to a change in the total value of the streaming rights in the UK. This paper does not take a view on the outcome of this debate.

The published scenarios do not place a value on this change. That is not to say there would be no change in the value of the rights, but rather to acknowledge that there could be an impact, and that impact is not known and not agreed between the stakeholders. This is included as a variable within the model.

Caveats

The model looks at UK income only, and over a 5 year time period. A further iteration of this discussion would be to look at the impact on the proportion of income coming from international sources (which will vary by deal type, and within deal types), as well potential changes to the share of UK repertoire globally, the growth of streaming and changes to remuneration models around the world.  

The model only captures recorded music income from sales and streams. It does not capture other revenue streams such as merchandise, live, synchronisation and other revenue streams that labels and artists can generate either together or independently. It also does not consider the role of streaming platforms in driving these other revenue streams.

The model looks at the first 5 years of a deal. It is important to consider that a move to ER would mean that investments by record labels take longer to recoup, while FA and NFA would see money sooner than the status quo. 

The assumed third party administration cost of 10% does not capture the setup cost, neither does it consider that rates could fall over time as systems become more efficient. Importantly, applying ER to a smaller portion of streaming would most likely reduce the total cost, but the rate would not necessarily fall by the exact proportion as there is an element of fixed cost. Broadly speaking, the more limited the scope of ER, the higher the administrative rate (for example, the administrative costs as a proportion of ER revenues) is likely to be.

Lastly, the model only considers the financial interactions between parties. Time spent by the FA and NFA investing in their own development is not captured. 

Importantly, this paper does not consider whether the status quo is fair. It looks at the impact that two key approaches would have on the current situation.

Broadcast Model - Summary of findings 

When looking at applying the Full Broadcast Model to all streaming income, the characteristics of ER are seen at their most stark. It is broadly agreed among stakeholders that Full Broadcast model at current ratios between FA and NFA is not desirable.

There are a number of flexibilities when looking at the scope of ER, including what proportion of streaming income it should be applied to under a Partial Broadcast ER Model, the ratio between featured and non-featured artists, and the administration fee.

Looking only at the Broadcast Model as it stands, and assuming no drop in the total value of rights, the introduction of Partial ER would increase the risks for record labels when investing in new talent development, since there is a value transfer away from labels towards the FA and NFA.

Whether such an intervention is fair – indeed whether the starting point is fair – is outside the scope of this research, but if the intention is to better support the careers of current and future artists then there is a significant risk that introducing the Broadcast Model would make it more difficult for the current label investment model to continue.

While this is most stark when looking at a Full Broadcast Model (applied to 100% of streaming income) it is also holds true when looking at the Partial Broadcast Model (applied to 35% of streams). Whether there are ratios that can better balance the needs of all stakeholders remains in scope for future research.

The real challenge for the Broadcast Model is how it impacts upon DIY artists, and those already on 50/50 deals. They do not benefit from an improved rate of payment from a label and the change would mean a decrease in the FA share of total income, giving a share to NFA and paying administration fees. This takes money away from FA in the early stages of their career while also increasing the administrative burden they experience.

The likely conclusion from this is that the introduction of Full Broadcast Model ER to all of streaming is unlikely to yield a net positive income for the industry at large. However, many of the mechanics from within ER could well have merit in moving the debate forward and applying ER more selectively might drive a more targeted benefit. Further research can examine the optimal levels of payment between labels, FA and NFA in order to better balance the benefits and obligations.

Spanish model - Summary of findings 

The Artist Remuneration Right (ARR) model (often known as The Spanish Model) introduces an additional right to the featured and non-featured artist.

This right can be against the digital service provider (DSP), as is the case in Spain, or applied against the label rights holder as a minima payment outside of the advance. While the DCMS Committee recommendation had initially considered applying this against the record label, the prevailing preference is to apply it to the DSP as in Spain.

The right remunerates featured artists and non-featured artists directly, after administration costs have been deducted by the third party Collective Management Organisation (CMO) (such as PPL).

While the presence of the right in Spain did not lead to a demonstrable change in the licensing position between the labels and the DSP there are concerns that introducing ARR into a larger market, such as the UK, might result in tougher negotiation terms from DSPs towards rights holders as they look to recover the additional value of rights in order to return to the current revenue ratio.

Ultimately, it is possible that future negotiations lead to a situation where the DSP looks to pass the financial obligation back to the rights holders through changes to their label deals. This doesn’t necessarily undermine the potential upside for artists and non-featured artists, but it might.

There is flexibility within the Spanish Model to only remunerate FA or NFA, or to change the ratio between them to better serve an optimal outcome. The administrative costs risk being disproportionately high, since the right is applied to a relatively low proportion income (5%) compared to either Broadcast Model (35% or 100%).

Further points of contention, for additional research:

The impact on advances:

There is a concern that introducing ER would lead to a reduction in the value or volume of advances offered by labels, which would impact upon artists and managers. There is debate between the artist and label communities about the impact on advances.

The impact on label investment:

It is possible that the labels would experience a slower return on investment under ER conditions, since they would be recouping from a smaller proportion of income, often at a lower rate. There is scope for the introduction of ER to change the investment profile, such as encouraging higher risk for a faster return, or investing more safely (in proven artists) to mitigate the increased risk. Similarly, if there was to be a slower rate of return this might lead to a reduced number of investments. Whether such situations would occur, and the likely reaction of the labels to such situations, could not be agreed among the stakeholders and is a consideration for further research.

Additional artist investment not captured:

Further research could also explore the value of investment from featured artists and non-featured artists investing in their own skills and development, as well as the value of their output. The opportunity for them to grow their teams, creating new roles, warrants further investigation in future research.

Implementation risks; costs, speed and accuracy:

While it is beyond scope for this paper, it is important to model what the cost of enabling a Collective Management Organisation (CMO) to distribute funds would be. It is also important to look at the administrative fees passed on to rights holders by the operation of such a body, which may be able to reduce over time. The efficiency (expressed as a lower per cent rate) will be subject to the size of the task and the complexity of implementation.

Speed of payment could be a concern, with many collecting societies distributing revenue every 3 months, which might be considerably slower than labels are currently able to offer. It is also possible that a body could report on the same frequency as the industry is accustomed to.

There are also questions of data accuracy and completeness, especially when it comes to recording metadata for the non-featured performers at the less established end of the market. Currently the Broadcast Model applies to those who have reached a level of success (resulting in their music being used on television or radio) but streaming has a much wider reach, and a much larger number of songs and artists to remunerate, and at a lower price per performance than traditional broadcast mediums.

These three considerations, cost, speed and accuracy, can trade off with each other. This is arguably the crucial question for future research, as the ability to cost-effectively collect and accurately distribute revenue will significantly impact who benefits from any change, and the size of that benefit.

Additional administration burdens:

Under all models there is a new administrative burden created. Where this burden sits and how large or small it is depends on the specific implementation. There is a concern that it may introduce new legal costs for artists who are not signed to a record label, as they structurally operate as their own record label. They might well need to account to their non-featured artists, which could potentially involve legal costs, as well as time costs.

There is also a risk of administration costs imposed on the label, who are now accounting to the artist wholly from their own systems. Similarly, DSPs might also see an increase in the administration burden they experience if they are asked to report to labels in the current format, but to a central body on a different format. Such things might sound trivial, but they could have a considerable negative impact and warrant further research.

The significance of catalogue revenue

Catalogue music, traditionally defined as music older than two years, accounted for 72% of streams in 2021 (up from 53% in 2016). This is a combination of older repertoire that has found a new lease of life on streaming, evergreen classic hits and more recent releases, which are staying relevant for longer.

The dominance of catalogue complicates the streaming picture. On the one hand it means that there is a lot of money to be made from repertoire that is no longer being actively invested in, bringing in income with minimal incremental costs. On the other hand, the risk of investing in new music is increasingly high - since it is fighting for a relatively low share of total consumer income compared to the share they commanded before the advent of streaming. As such the large catalogue owners (the major record labels) are potentially best placed to take the risk on new talent, funded at least in part by their large catalogue revenues.

This complex relationship between old and new means that finding a market level solution that treats both catalogue and new releases equally risks being good for one and bad for the other.

An extremely tight hit-heavy distribution

One other key market condition that complicates intervention is the disparate revenue profiles of various artists partaking in the music industry. The Creators Earnings report found that the top 0.1% most popular tracks achieved more than 40% of all streams. The top 1% of tracks consistently account for between 75% and 80% of total streams from 2016 to 2020.^{[footnote 1]}

The nature of this unequal distribution means that there are a small number of big winners, and that 99% of tracks are fighting for around 20% of revenue, and that there is therefore a broad church of artists who will be differently affected by changes in regulation based on their level of success.

Conclusion

While this paper did not set out to conclude in favour or against any model, the one key conclusion that can be drawn is that ER does not offer a simple solution to the streaming conundrum. There are many unknowns and complex interdependencies within the modelling, and a range of further questions that merit deeper consideration. Importantly, this paper did not set out to determine what is fair and that important debate is outside the scope of this research.

One of the most important and most contentious future considerations is to clarify whether a change in the nature of the rights would significantly weaken the rights holders’ ability to negotiate license fees, and whether such a change in total value of rights is worthwhile in order to redistribute funds across the value chain. The hope is that a change could be managed in such a way that would not necessitate a drop in value, but this needs to be considered until it is confirmed that there is no risk of this, or that the risk is mitigated by value added elsewhere in the value chain.

The impact of ER on a label’s ability to meaningfully invest in developing new talent is another key consideration. If ER were to undermine the return on investment (ROI) from new talent development, there is a risk that those labels which had previously played significant roles in artist development might reduce this activity. It might be that the market environment allows for success without label investment, but this is not known and not proven.

The cost of administration is a deadweight loss and is likely to be a significant sum of money. This money is certainly lost from the rights holders’ side of the market, and the upsides of redistribution need to be considered against this backdrop.

The other key consideration is how such an intervention changes incentives between rights holders on the creator side. An exploration of whether there are combinations of ratios that would optimise the relationship between labels, featured artists and non-featured artists is recommended, but was beyond scope for this paper.

While not a satisfying conclusion, it is clear that more research is required into the nuances of how best to apply ER in order to balance the incentives to create with the need to monetise creation.

1. Background

Concerns have been raised around the issues of music creators’ earnings. The issue was brought to the forefront when the COVID-19 pandemic and subsequent lockdowns hit the global economy. This meant that income from live performance stopped, leaving streaming royalties the key source of income for the majority of musicians.

The perceived inequity in the revenue distribution led to a number of musician-led campaigns and initiatives that called for greater income from streaming, claiming that the way streaming revenues are split is not fair, and that the system favours the big players at the expense of the wider creator economy. They requested a review of the way recorded music revenue is shared on the streaming platforms. Later in October 2020, the Digital Culture Media & Sport (DCMS) Select Committee launched an inquiry into the Economics of Music Streaming and published its report in July 2021. In parallel, an independent team of academics, commissioned by the UK Intellectual Property Office (IPO) also conducted research into Music Creators’ Earnings in the Digital Era. The government responded to the Select Committee (published 15 September 2021) with three areas for further investigation: equitable remuneration, contract adjustment and rights reversion.

This report focuses on Equitable Remuneration (ER). ER has been proposed as a simple way to resolve the issues for performers by a number of reports, including the WIPO’s 2021 “Study on the Artists in The Digital Music Marketplace: Economic And Legal Considerations (Castle and Feijoo, 2021)^{[footnote 2]},” as well as the DCMS Committee’s report on the Economics of Music Streaming (Digital, Culture, Media and Sport Committee, 2021). The DCMS Committee’s report concluded;

The right to equitable remuneration is a simple yet effective solution to the problems caused by poor remuneration from music streaming. It is a right that is already established within UK law and has been applied to streaming elsewhere in the world. A clear solution would therefore be to apply the right to equitable remuneration to the making available right in a similar way to the rental right. As such, an additive ‘digital music remuneration’ payment would be made to performers through their collecting societies when their music is streamed or downloaded. This digital music remuneration would address the issues of long-term sustainability for professional performers and the cannibalisation of other forms of music consumption where equitable remuneration applies, whilst also retaining the benefit of direct licensing.

However, there has been a dearth of research into the actual economic modelling of ER and its implications in the market. This model fills in some of the gaps by exploring two existing models of ER currently in place, and considers the likely implications of implementing either within the UK market. The two models for consideration are known broadly as the Broadcast Model and the Spanish Model. The Broadcast Model can be split into the Full Broadcast Model (where ER is applied to all streaming activity and distribution takes place through a collective agency) and Partial Broadcast Model where only ‘lean back’ or ‘push’ consumption is licensed this way and distribution takes place through a collective agency.

The primary purpose of this research is to provide economic modelling of these models to help facilitate an informed debate on ER, highlighting some of the complexities and potential unintended consequences of their potential implementation. The paper will set out the potential changes in money flows, and assess whether these changes would be net positive for the incentive to create, and who the winners and losers might be. This paper will lay the foundation for further analysis of these and other models in the future.

2. Performers’ rights and income

2.1 Evolution of performers’ rights

A performer refers to those who perform for a recording, including a singer or a musician. There are two types of performers: featured and non-featured. Featured performers (or “featured artists”, “FA”) are those whose names are credited in the record release. Non-featured performers (or “non-featured artists”, also known as session musicians, “NFA”) are those whose names are not credited on a recording. They include backing vocalists and orchestral players, who are hired to perform on a recording on a “work-for-hire” basis. They transfer their performers’ rights to the producer (often a record label) in exchange for a one-off payment.

Performers rights fall into two categories: exclusive rights and remuneration rights. Exclusive rights grant the performer the power to authorise or prohibit a certain act. Such acts originally included making a fixation of an unfixed performance (essentially “making a recording”) and making a reproduction of a fixed performance (for example, making a copy of that recording)^{[footnote 3]}. Over time, the range of exclusive rights grew and in 1996 the exclusive right of making available was introduced for performers at the international level by virtue of the WIPO Performances and Phonograms Treaty of 1996 (WPPT).^{[footnote 4]} This is the right most relevant to streaming and is analysed at section 3.2. Exclusive rights are originally owned by the performers, but often are transferred to record labels when signing a record deal.

Remuneration rights are rights that ensure performers receive remuneration when a certain exploitation of their recorded performance occurs. The most common example is the right to receive equitable remuneration when a song is played on the radio or played in public, originally introduced for performers at the international level in the “Rome Convention for the Protection of Performers, Producers of Phonograms and Broadcasting Organizations” of 1961 (“Rome Convention” hereafter).^{[footnote 5]} An important characteristic of the equitable remuneration right in UK law is that it is un-waivable and non-transferrable (other than to a collecting society, such as PPL).

It was only in 1961 that performers’ rights were first recognised at the international level. This occurred with the adoption of the Rome Convention, which laid down minimum levels of protection for performers.

At the EU level, the Rental and Lending Rights Directive (Directive 92/100/EEC) was introduced in 1992. It contained extensive protection for performers including a number of exclusive rights, as well as a right to equitable remuneration for the broadcasting and communication to the public of phonograms.

At this point, the UK provided only limited protection to performers, under the Copyright, Designs and Patents Act 1988 (“CDPA”). However, in order to implement this EU Directive, the CDPA was amended in 1996. Consequently, performers were granted the right to equitable remuneration for the broadcasting and communication to the public of their sound recordings.

By virtue of the CDPA, performers now enjoy non-property rights, property rights, equitable remuneration rights and moral rights. Referring to the CDPA, Arnold (1997) commented that “This statute finally places it beyond doubt that performers have civil remedies by granting them ‘performers’ rights’ which are in effect the long-awaited performers’ copyright.”

2.2 Changes in performers’ income

The way FA are remunerated depends on a number of factors, including their level of contribution to the recording, popularity and career stage, all of which go to determine their contractual terms with intermediaries. It is also important to understand that their level of income is heavily reliant on their commercial success, in which the labels have long played a crucial role.

In a marketplace where labels had a significant role in producing, distributing and promoting album releases, the power dynamics between the performers and labels had long been characterised as skewed toward labels. Digitisation and the subsequent opening of diverse pathways to pursue music careers brought changes to this dynamic, although to what degree is still a matter of debate.

As discussed above, there are two types of artists: FA and NFA. The share of revenue they each receive varies depending on the source of revenue. When entering a traditional deal with a record label, FA typically receive an advance payment, in addition to a royalty share on income from the exploitation of their recording. NFA receive a one-off fee for each session of the recording, without a share in future royalties.

2.3 Changes in the market

In a predominantly physical world, record labels funded the majority of artist development, including touring, in order to sell high priced albums. At that time, the cost of creation was expensive, and mainstream marketing channels were finite and few. In return for their investment, record labels took a large share of the royalty income. Generally, the majority of income would be realised in the first 6 weeks of an album being released, with listening being paid for up front in the form of a purchase. Put another way, regardless of how much a CD or vinyl was listened to, the revenue was the same and the money was received in advance of listening.

This model began to be challenged in the digital era, as technology enhancements offered a boom in Do-It-Yourself (DIY) tools which enabled music creators to produce, market and distribute their music on their own, bypassing the structural help of the traditional intermediaries.

As well as innovation at the grassroots, there were some stand-out successes for well-established acts, most famously Radiohead’s Pay What You Want innovation for their seventh album In Rainbows. This is often exemplified as a new way of achieving commercial success in a digitally networked environment, but it is also considered that its success owed much to Radiohead’s existing fandom from their career to that point, rather than solely to the power of digital innovation. 

iTunes opened the gate of consuming digitised music legally, making many changes to the value of the digital music. For example, à-la-carte music choice enabled track-based music consumption where every song could be unbundled from album-based music consumption. Consumers still paid an up-front payment for the product, regardless of the number of listens, but for a smaller unit at a lower unit cost.

Spotify’s “freemium” subscription-based music streaming business model (both “free” and “premium”) changed the way music is consumed and monetised. Its monthly subscription for almost all the music in the world changed unit-based music sales to a monthly unlimited amount of consumption model.

In the case of “paid-for” streaming, this changed the unit of payment from a fixed fee for a finite number of songs (regardless of listens per song) to a new model where a flat fee was divided across total listens within a geographic area. While the pay-per-unit model increased market size in proportion to the growth of record sales, the pay-per-listen subscription model meant that more music being sought after did not necessarily grow the revenue coming into the industry. Instead, increased listening (with no increase in the fixed fee) actually reduces the financial value of each listen, as the pool of money being generated in the market stays the same.^{[footnote 6]}

As shown in Figure 1, from 2012, the revenue from streaming has gradually increased, becoming the dominant consumption platform and revenue source, and accounting for 78% of consumer spend on recorded music in 2022.

Figure 1. Recorded Music Income by Format. Unit: GBP millions. Source: All about the Music (BPI, 2023)

In parallel, the massive growth of DIY music gave rise to the explosion of repertoire on streaming platforms, with digital streaming services now adding 100,000 tracks per day (Peoples, 2023^{[footnote 7]}). This meant that there is a staggeringly large number of creators fighting over a fixed monthly pie, leading to diluted income for the majority of music creators. In addition, the listening-based remuneration also made for a slower payment schedule over a long period of time, in stark contrast to the previous model of upfront payment.

Another important aspect to consider is that the abundance of choice and the fixed total income further affirms, if not intensifies, the heavily concentrated nature of music industry income. The Music Creators Earnings Report (Hesmondhalgh et al., 2021)^{[footnote 8]} found that “the top 0.1% most popular tracks achieved more than 40% of all streams in all years (based on October) and the top 0.4% of tracks accounted for more than 65% of all streams from 2016 onwards. The top 1% of tracks account for between 75% and 80% and the top 10% for between 95% and 97%, in all years from 2016 to 2020.”

When considering the number of unique tracks generating at least one stream in the UK in October 2020, they found “[o]f the 4.28 million tracks achieving at least one stream on UK streaming services in October 2020, 1% (42,835 tracks) accounted for around 78% of streams.” Similarly, the BPI have reported that catalogue music (defined as tracks released two or more years ago) accounted for “a record 72.0% of [UK] audio streams in 2021, up from 53.3% in 2016 (BPI, 2022).” This grew to 75% in 2022^{[footnote 9]} and creates a more challenging environment for new music to make money, increasing the risks for artists looking to break through, and for those investing in their success.

3. Economic modelling

Despite the heightened interest in this field, the debates around ER have predominantly been discussed in legal perspectives, and there has been little research on the potential impacts ER might have on the market economics. This report fills part of that gap by transparently demonstrating different scenarios of ER and their impacts on various parties in the music marketplace.

In this section, there are the three most debated models, Partial Broadcast Model, Full Broadcast Model and the Spanish Model. Although the model aims to have included as many variables as possible, the actual music market is much more complex and it is highly difficult, if not impossible, to capture all the nuances and complexities in economic models.

The contentious nature of the debate added another layer of complication, as the variables contain contingencies. Henceforth, these models can best be understood as simulations that may be subject to market conditions and negotiations amongst diverse stakeholders.

3.1 Model descriptor

Before we delve into the modelling, we first describe how these models have been made. We explain the types of deals we took into consideration in this report, followed by breakdowns.

3.1.1 Types of deals

This economic model is focussed on two key deal types. They are the modern advance deals and the DIY deal.

Initially, 6 deal structures were mapped out, and have been discussed at length with a range of industry stakeholders from the IPO Contact Group. These demonstrated how the impact of ER varies depending on the royalty share of the artist/label. These range from a legacy (10%/90%) and modern (26%/74%) advance deal, to a JV (50%/50%), label services (75%/25%), a distribution deal (85%/15%) and a DIY arrangement (90%/10%).

Each deal model aims to consider the full cost of the deal (including marketing spend and overheads) rather than only recoupable costs. The ratios between recoupable, non-recoupable and ongoing costs are flexible within the model and the published model is based on the assumptions from the best placed stakeholders from within the IPO Contact Group. 

The impact on artist and label income varies significantly depending on whether an artist has recouped their initial advance. As such, each deal has four revenue scenarios modelled. In the cases where an advance is issued, revenues have been modelled at 0.5x, 1x, 3x and 7x the value of the advance, giving a stylised loss-making deal, a breakeven deal, and profitable deals. In the cases where an advance is not issued four revenue numbers are input, in the same proportions.

Since there are a range of models and a range of revenue scenarios, your author focussed on the impact that changing toward ER would have on a modern advance based record deal and a DIY release. While not covering every possible deal situation, these exemplify the most significant impacts of ER in the clearest way.

These are broadly indicative of the key deals available within the industry at large and are based on detailed discussions with the stakeholders in the IPO Contact Group.

Advance deals

Advance deals refer to deals where artists receive a payment in advance of expected royalties from a record label.​​ Under a traditional “advance and royalties” deal, record labels collect revenue from DSPs and once the advance has been repaid, royalties are paid to their signed artists at a contractually agreed royalty rate, in return for a transfer of rights. The royalty rate and the advance amount are negotiated between the label and the artist, alongside many other terms. In this model, we offer two different advance deals: legacy deals, and modern deals. The legacy deals offer a stylised view of some of the older contracts which might pre-date the digital era and reflect deals of that time. The modern deals fall in line with the CMA Music and Streaming report, which denotes that artists signing contracts today receive a 26% royalty share on average (CMA, 2022)^{[footnote 10]}.

Do-It-Yourself (DIY) set up (90:10)

Under a DIY set up, the artist takes on the role traditionally borne by the record label. They cover the cost of recording and marketing. There is no advance and they pay a third party to distribute their music to global music services. This may be in return for a flat fee per track or album, or as a share of royalties collected. In DIY deals, an intermediary takes between 10% and 15% for the distribution of the music, and the artist takes the remainder. Other services take an annual subscription and no share of revenue. There is no advance paid, and usually no transfer of rights.

Other deals that are not included in the final paper but have been considered as part of this analysis are:

Joint Venture deals (50:50)

Under a joint venture or profit share arrangement, the record label and the artist put the revenues and agreed costs into a joint entity or accounting, and then share the resulting profits. In this deal, the risk is shared by both the label and the artist. The label typically pays a lower advance and splits profits 50/50 with the artist. The artist receives a lower up-front payment than in the advance deals, sharing the risk with the label in return for a higher royalty rate.

Label services deals (75:25)

A label services deal typically involves a range of services (such as marketing and promotion) and associated fees which typically range from 20% to 40% of revenues. This is lower than an advance deal, but higher than a distribution deal. If an advance is paid, this is typically lower than that which is paid in an advance deal. The services offered vary significantly between deals, but would usually include marketing spend and some artist development expertise, alongside distribution infrastructure.

Distribution deals (85:15)

Under a distribution deal, the artist forms their own label company and recruits a distributor to get their music on to digital platforms. The distribution company takes a share of the royalties in exchange for their services. They may also provide a recoupable advance or a commitment to marketing spend and other costs.

3.1.2 Investment breakdown

There are costs that artists cover (recoupable costs) and other costs that the label bears (including marketing, but also office costs). Recoupable costs are agreed as part of the advance recovered from the artist’s share of royalties under the advance model. This is a relatively simple view of recoupable costs, which can vary by deal and vary depending on the which marketing levers are activated.^{[footnote 11]}

Within the model, recoupable costs are set to c52% and non-recoupable costs to c48% based on input from record label stakeholders in line with input from stakeholders from the IPO Contact Group.

3.1.3 Revenue breakdown

Streaming revenue is separated from non-streaming revenue, as the two behave differently under new ER conditions. It is also broken down pre/post recoupment. This is important as there are no royalties paid to an artist before the artist recoups 100% of their advance payment.

The full model considers four stylised revenue situations. In the case of Advance deals, the first assumes that only half the advance is recovered, exemplifying a loss-making situation for the project. The mid-income scenario assumes that the advance is recovered but that there is no additional income, and two larger income scenarios where the project income is broadly three times and seven times the advance respectively. For the deals where there is not an advance then broad assumptions are made to offer a loss making, break even and profit-making scenario. These are examples to demonstrate the impact of ER rather than models of specific deals.

3.1.4 Comparable metrics

For each deal structure, under each revenue assumption, the model presents 4 key metrics:

• label net position: Label revenue minus total label costs (both recoupable and non-recoupable).

• Featured Artist (FA) income: Sum of advance paid and any royalty income paid to the FA. The value of recoupable spend from the label is also plotted.

• Non-Featured Artist (NFA) income: This is the sum of royalties paid to musicians who are not named as part of the act but play on the recordings. This figure does not capture any session fees which would have been paid for the recording process, as described earlier in this paper.

• admin third party cost: This captures any additional costs arising from a new or existing body distributing funds to labels, FA and NFA.

3.2 Impact of Broadcast Model ER

In this section, we give a summary of the 2 different types of deals when ER is applied. We explain the impact on labels, featured artists, non-featured artists and third party admin expense by comparing the status quo (in blue) after ER is introduced.

3.2.1 How Broadcast ER changes revenue flows:

Before diving into the model outputs, it’s important to consider the various impacts of ER, and how they affect different parties – and how they affect those different parties differently depending on the deal structures they are operating under.

Payment rate

The first impact of ER is to change the rate of payment between labels and artists. Where an artist has signed a deal for less than a 50% share of streaming income, the rate would be adjusted to equal 50%, reducing the label share accordingly. This is in line with the broadcast rules PPL apply to radio royalties.

Where an artist has signed a deal in which they retain more than 50% (such as a label services deal, or a distribution deal) their rate will be unchanged. Similarly, if an artist is taking a DIY approach their share of royalties will remain unchanged.

The rate does not necessarily need to change to a 50/50 split to be considered equitable. This is the application of the current Broadcast Model.

Non-recoupable income

Importantly, the introduction of ER would mean that the artist share of streaming revenue would not be used to repay their advance, because ER payment is paid directly to the artists. This would mean royalty payments would be received by artists as consumption takes place, rather than receiving royalties only after they have recouped their advance. It would also mean that record labels are receiving their return on investment from a smaller share of the total revenue.

Non-Featured Artist payment

A further impact of ER is to introduce a payment for non-featured artists. Currently, non-featured artists receive no payment for the use of their music on streaming services. They are likely to have received a session fee, as described above, but no incremental royalties. The introduction of ER would mean a payment for all musicians listed as playing on a recording, not only the featured artist. Under the current Broadcast Model, record labels receive 50% of royalty income, with the other 50% split, 65% to the featured artist and 35% to the non-featured artist (resulting in 32.5% and 17.5% of the total respectively).

Administrative cost

There would be a cost to administering these rights. The anticipated third party administration rate could change as the scope of the task changes. Should ER only be applied to 50% of streaming income, then the lower absolute income might lead to a higher comparative rate. There are also trade-offs between accuracy, speed of distribution and cost effectiveness of the third party administration, which are highlighted in section 4.

The range is likely to be between 15% (broadly in line with PPL’s current UK licensing) and 5% (broadly in line with Sound Exchange’s US online licensing). There is scope for this rate to change over time, potentially starting higher and reducing as the costs of setting up the third party diminish. An indicative, mid-range rate of 10% is applied, and can be varied within the model.

The scale of the task is a significant jump from broadcast royalties, with streaming services serving between 40 million and 50 million unique tracks per month, compared to tens of thousands through broadcasting. This may require a significant amount of financial investment especially for the initial set-up. As discussed in section 3.2, in implementing ER for traditional broadcasting, similar issues have arisen which led to setting up new CMOs, which later were merged with PPL. There might also be a cost to labels and DSPs for administering rights in a new way, but these are not captured within the model. Only the new cost of paying a third party an admin fee of 10% is captured. There are other costs which warrant further research, as discussed in section 4.

3.2.2 Key variables and assumptions:

In devising the economic modelling for ER, we had placed a few variables that might arise in the market when ER is applied. Although informed by the interviews conducted for this research, these variables are not definite but only indicative and therefore may be subject to change. Here we provide the background of these variables.

Market value of rights under ER

During the interviews, stakeholders from the labels have argued that the introduction of ER would impact on the rights holders’ ability to negotiate the same total value of streaming. This concern is held by label representatives, and holds that it might not be possible to maintain the current value of rights should ER be applied to the making available right. This is contested by the Council of Music Makers (CMM), where the view is that the legal change is ultimately flexible and could potentially be structured in a way which does not compromise the labels’ ability to negotiate for the value of the master right.  

While this is a crucial consideration, there was no agreement across the stakeholders and as such the impact of a potential change in the value of streaming rights is not published here. Unlike the administration rates, where it was agreed there would be a rate but debate about what that rate would be, there is no consensus on whether change would take place and how impactful that change would be. The accompanying model offers flexibility to consider the impact that a change in the value of rights would have, but it is not published here.

What proportion of streaming income should ER apply to?

Stakeholders from the labels have also argued about the share of total streaming that should be subject to ER. For the example modelled below we have applied ER to all of streaming. This is chosen in order to show the maximum impact of ER, and also to avoid inadvertently siding with one of many other models which are in discussion within the industry currently.

As mentioned above, the models below consider partial broadcast (an indicative 35% of streams) and full broadcast (100% of streams). There is debate around the share of streams that are to be considered ‘lean back’, and scope for the share of such consumption to change as listening habits on streaming platforms mature, and as new users join the platforms. Broadly speaking, lean back consumption includes consumption from playlists that are not your own, artist radio functions, and music that plays once your own selections have been played.

What should the split be between labels and artists?

Under the broadcast ER model, a 50%/50% split between the artist and the label is assumed. This is reflective of the current shares under broadcast licensing. However, it is important to consider that there is scope for flexibility in how a new law could be applied and other ratios could be applied.

This would have two key effects. Firstly, it would reduce the impact of the change in the rate of payment (see above) and potentially dampen the downside impact on the record labels, while decreasing the upside for the artists. However, it keeps the benefit of direct payment even against a lower proportion of income.

It would also mean that the administrative burden falls disproportionately on the label, with them paying the majority of the administrative costs.

What proportion of the ‘artist pool’ should be paid to non-featured artists?

Under traditional broadcasting, 65% of income is paid to the FA, with 35% of income paid to NFA. We applied the same rate to the ER in streaming in this model. This is not to say that this is fixed, and it is reasonable that another rate could be applied in order to ensure a different outcome between FA and NFA.

What fee would a third party charge?

In this model, we applied a 10% administrative charge from the third party distribution body, in line with feedback from the IPO Contact Group. We also drew upon the current market rates as a benchmark, with PPL (c14%) and SoundExchange (c5%). However, there is the potential for a range of administration rates.

3.3 Model outputs: Partial Broadcast ER

Since ER changes how pre-recoupment income is distributed, its impact varies significantly depending on how much revenue is generated, in proportion to the advance. To demonstrate this, we explore four revenue scenarios, depicting a loss making deal, another where the advance is returned in full, but not other costs, a third where 3x the advance is generated in income and a fourth where 7x the advance is generated in income.

For ease of comparison, we have laid out four revenue scenarios for two situations: the modern advance deal, and DIY. This demonstrates the impact of ER for those artists recently signed and for those forging their own path in the music industry.

The focus of the most recent discussions centre around the Partial Broadcast Model and the Spanish Model. As such we first explore the impact of the Partial Broadcast Model, and for completeness your author has included the analysis with the Full Broadcast model under two mid-revenue assumptions for comparison.

3.3.1 Partial Broadcast Model - Modern Advance Deal

Modern Advance deal – Project income = 0.5x Advance

Model parameters:

• ER applied to 35% of streaming, with no reduction in total streaming value
• streaming accounts for 78% of artist income, with 22% from physical
• label invests £150,000: £80,000 recoupable costs (split £60,000 cash advance, and £20,000 in spend) and £70,000 non-recoupable costs (including marketing)
• label’s share of income post recoupment is 74% (100% before recoupment)
• artist receives £60,000 cash advance, and £20,000 of in-kind spend
• project income is £40,000 (half of the recoupable advance)

Figure 2. Modern Advance - Half Advance

In the status quo scenario, the record label makes a loss of £110,000. The Featured Artist receives no additional income beyond the cash advance paid up front, since the advance has not been recouped. There is no payment to the non-featured artist, nor to a third party. There is no change to the value of rights.

Under ER, applied to 35% of streaming income, the record label loss increases by £6,006 (5%) to £116,006. The featured artist benefits from additional income, with their total value received (including recoupable costs spent) growing to a total of £83,194, of which £63,194 is cash. Non-featured artists who played on the record in question would share £1,720. The cost of distributing the money to the relevant stakeholders through a third party is £1,092.

Modern Advance deal – 1x Advance

Model parameters:

• ER applied to 35% of streaming, with no reduction in total streaming value
• Streaming accounts for 78% of artist income, with 22% from physical
• label invests £150,000: £80,000 recoupable costs (split £60,000 cash advance, and £20,000 in spend) and £70,000 non-recoupable costs (including marketing)
• label’s share of income post recoupment is 74% (100% before recoupment)
• artist receives £60,000 cash advance, and £20,000 of in-kind spend
• project income is £80,000 (equal to the recoupable advance)

Figure 3. Modern Advance - 1x Advance

Under status quo scenario – where project revenue is equal to the advance paid – the record label makes a loss of £70,000. The featured artist receives no additional income beyond the cash advance paid up front, since the advance has not been recouped. There is no payment to the non-featured artist, nor to a third party. There is no change to the value of rights.

Under ER, applied to 35% of streaming income, the record label loss increases by £12,012 to £82,012. The Featured Artist benefits from additional income, with their total income received growing to £66,388. Non-featured artists who played on the record in question would share £3,440. The cost of distributing the money to the relevant stakeholders through a third party is £2,184.

Modern Advance deal – 3x Advance

Model parameters:

• ER applied to 35% of streaming, with no reduction in total streaming value
• streaming accounts for 78% of artist income, with 22% from physical
• label invests £150,000: £80,000 recoupable costs (split £60,000 cash advance, and £20,000 in spend) and £70,000 non-recoupable costs (including marketing)
• label’s share of income post recoupment is 74% (100% before recoupment)
• artist receives £60,000 cash advance, and £20,000 of in-kind spend
• project income is £240,000 (3 times the recoupable advance)

Figure 4. Modern Advance - 3x Advance

Under the status quo scenario the record label makes a profit of £90,000. The Featured Artist receives no additional income (beyond the cash advance paid up front), since the advance recoups only from the artist share of income and has not been fully recouped. There is no payment to the Non-Featured Artist, nor to a third party.

Under ER, applied to 35% of streaming income, the record label moves from profit of £90,000 to a reduced profit of £53,964. The FA benefits from additional income, with their total value received (including recoupable costs spent) growing to £99,165,  with income of £79,165. NFA who played on the record in question would share £10,319. The cost of distributing the money to the relevant stakeholders through a third party is £6,552.

Modern Advance deal – 7x Recoupable Advance

Model parameters:

• ER applied to 35% of streaming, with no reduction in total streaming value
• streaming accounts for 78% of artist income, with 22% from physical
• label invests £150,000: £80,000 recoupable costs (split £60,000 cash advance, and £20,000 in spend) and £70,000 non-recoupable costs (including marketing)
• label’s share of income post recoupment is 74% (100% before recoupment)
• artist receives £60,000 cash advance, and £20,000 of in-kind spend
• project income is £560,000 (7 times the recoupable advance)

Figure 5. Modern Advance - 7x Advance

In the status quo scenario, the record label makes a profit of £344,400. The Featured Artist receives £65,600 additional income beyond the cash advance paid up front. There is no payment to the non-featured artist, nor to a third party. There is no change to the value of rights.

Under ER, applied to 35% of streaming income, the record label profit decreases by £44,335 (-13%) to £300,065. The Featured Artist benefits from additional income, as their total income received grows to a total of £130,569. Non-Featured Artists who played on the record in question would share £24,079. The cost of distributing the money to the relevant stakeholders through a third party is £15,288.

3.3.2 Partial Broadcast Model – DIY artists

DIY – Low income

Model parameters:

• ER applied to 35% of streaming
• streaming accounts for 78% of artist income, with 22% from non-streaming
• there is no label, so no investment and no recoupable costs
• distributors share of income is 10%
• artist receives £0 cash advance, and no in-kind spend
• project income is £5,000

Figure 6. DIY - Low

In a low revenue scenario (£5,000 total income), there is very little change for the label/distributor. Their revenue falls by £13, corresponding to the administrative deductions from the CMO collecting the ER share of revenue on their behalf. FA would see a more significant reduction in their revenue from £4,500 to £3,990, a fall of £510 (an 11% drop). NFA would see their revenue increase from zero to £387 and the third party administration cost would be £137.

This redistribution away from the FA in favour of NFA is a very important consideration for whether ER is actually a simple method for redistributing income towards artists. It might well not be the case for DIY artists and those at the earliest stages of their careers.

DIY – Middle income

Model parameters:

• ER applied to 35% of streaming
• streaming accounts for 78% of artist income, with 22% from non-streaming
• there is no label, so no investment and no recoupable costs
• distributors share of income is 10%
• artist receives £0 cash advance, and no in-kind spend
• project income is £10,000

Figure 7. DIY - Middle

Given the structure of the DIY model, the trend is consistent as revenue grows. In a mid-revenue scenario (£10,000 total income), there is very little change for the label/distributor. Their revenue falls by a relatively small £27. FA would see a reduction in their revenue from £9,000 to £7,980, a fall of £2,120, an 11% drop.

NFA would see their revenue increase from zero to £774 and the third party administration cost would be £273.

DIY – High income

Model parameters:

• ER applied to 35% of streaming
• streaming accounts for 78% of artist income, with 22% from non-streaming
• there is no label, so no investment and no recoupable costs
• distributors share of income is 10%
• artist receives £0 cash advance, and no in-kind spend
• project income is £30,000

Figure 8. DIY - High

In a high revenue scenario (£30,000 total income), there is a small change for the label/distributor. Their revenue falls by £82. The Featured Artist would see a reduction in their revenue from £27,000 to £23,941, a fall of £3,059, an 11% drop.

Non-featured artists would see their revenue increase from zero to £2,322 and the third party administration cost would be £819.

DIY – Very high income

Model parameters:

• ER applied to 35% of streaming
• streaming accounts for 78% of artist income, with 22% from non-streaming
• there is no label, so no investment and no recoupable costs
• distributors share of income is 10%
• artist receives £0 cash advance, and no in-kind spend
• project income is £70,000

Figure 9. DIY - Very High

In a very high revenue scenario (£70,000 total income), there is a small change for the label/distributor. Their revenue falls by £191. The Featured Artist would see a reduction in their revenue from £63,000 to £55,862, a fall of £7,138, an 11% drop.

Non-featured artists would see their revenue increase from zero to £5,418 and the third party administration cost would be £1,911.

Summary of Partial Broadcast ER

A change to Partial ER would make record label investment more risky and more difficult to justify. That is not to say that other opportunities and models for funding creativity could not be found, but rather that it would be very disruptive to the current investment paradigm.  

For an artist taking a DIY approach, the introduction of ER increases their obligations to pay others without improving the total income for FA.

This redistribution away from the FA in favour of NFA is a very important consideration for whether ER is actually a simple method for redistributing income towards artists. It might well not be the case for DIY artists and those at the earliest stages of their careers. It might also discourage the use of NFA in the early stages of artists careers in order to keep a larger share of their income.

3.4 Model outputs: Full Broadcast ER

3.4.1 Full Broadcast ER – Modern Advance Deal – 3x Advance

Having demonstrated the impact of applying ER to a portion of streaming, the following scenarios consider the impact of applying ER to 100% of streams. This approach has more recently taken a back seat to the Partial Broadcast ER Model and Spanish Model and as such only 2 like for like scenarios are offered here for comparison with the Partial Broadcast Model above.

Model parameters:

• ER applied to 100% of streaming, with no reduction in total streaming value
• streaming accounts for 78% of artist income, with 22% from non-streaming
• label invests £150,000: £80,000 recoupable costs (split £60,000 cash advance, and £20,000 in spend) and £70,000 non-recoupable costs (including marketing)
• label’s share of income post recoupment is 74% (100% before recoupment)
• artist receives £60,000 cash advance, and £20,000 of in-kind spend
• project income is £240,000 (3 times the recoupable advance)

Figure 10. Modern Advance - 3x Advance

Under Full Broadcast ER, applied to 100% of streaming income, the record label position moves from £90,000 profit to a loss of £12,960. The featured artist benefits from additional income, with their total income growing to £114,756. Non-featured artists who played on the recording would share £29,848. The cost of distributing the money to the relevant stakeholders through a third party is held constant at 10%, costing £18,720.

3.4.2 Full Broadcast ER – DIY – High Income

DIY – High income

Model parameters:

• ER applied to 100% of streaming
• streaming accounts for 78% of artist income, with 22% from non-streaming.
• there is no label, so no investment and no recoupable costs
• distributors share of income is 10%
• artist receives £0 cash advance, and no in-kind spend
• project income is £30,000

Figure 11. DIY - High

The impact of ER on DIY artists is in contrast to those signing advance deals. Since the starting rates of a distribution deal are better than 50%/50% there are limited upsides (if any) for the DIY artist. The obligation to pay NFA out of initial income has the potential to be problematic for DIY artists, as well as increasing their administrative responsibilities.

That is not to say that ER is not an option for the market at large, but it does bring into sharp focus the scope for it to make life more challenging at the grassroots level. Such a change needs to balance the requirements at the grassroots with the more established part of the market.

3.5 Model outputs: The Spanish Model (ARR)

Currently in Spain, 5.6% of streaming income is shared out between featured artists and non-featured artists (NFA). Internal rules at local CMO AIE^{[footnote 12]} determine the split where 2/3 of distributions go to FA, and1/3 to NFA, which is in line with the broadcast split between FA and NFA in the UK.

In Spain, the right is held against the DSP. When the DCMS committee first considered the Spanish model there was a consideration that it could be held against the label. However, this model reflects the latest preference that this right should be held against the DSP in the UK and assumes a 5% rate.

Contrasted with the full broadcast ER model, the Artist Remuneration Right (ARR) model offers a much less significant shift in revenue. This has the benefit of reducing disruption to the current investment model but raises a question as to whether it does enough to make a material difference to featured and non-featured artists.

In line with AIE’s (Artistas Intérpretes o Ejecutantes) submission to the streaming enquiry a 10% administration rate is assumed. However further research is required to confirm whether such an assumption would be achievable given the comparatively low base to which the rate is applied and whether such a rate is achievable in the UK.

Impact of ARR – Against the DSP

The impact of ARR has been modelled at a market level. For simplicity, we have modelled the Modern advance deal (26%/74%) and the DIY situation (10%/90%) only, in favour of the label/artist respectively.

Figure 12. Modern Advance: Spanish Model

Under an advance deal, there is a 5% change for the DSP (1/6 of their income), and no change for songwriters and publishers (collectively authors). There is no change in the record label share of total revenue.

The 5% from the DSP would be distributed to the featured artist who increases their share to 17% from 14% and the non-featured artist growing to 1.6% share.

The third party administration cost is 0.5% of total income, reflecting the relatively low proportion of income it is applied to. Given the scale of setting up an organisation to operate in this way it is unlikely that the UK would see such a low rate from the outset, since there are set up costs and fixed operational costs to consider.

Figure 13. DIY - Spanish Model

Under a DIY deal, there is no change for the record label, and no change for songwriters and publishers (collectively authors).

There is a significant reduction in DSP share (dropping from 30% to 25%), which is distributed to the featured artist who increases their share to 51.5% from 48.6%, with the non-featured artist growing to 1.6% share.

The third party administration cost is 0.5% of total income.

In order to scale back up to a market level, the broad assumption that 80% of revenue comes from advance style deals, while 20% comes from DIY deals was applied. The reflection at a market level helps convey the absolute value of the marginal percentage gains, alongside the relative percentage changes.

Figure 14. Market: Impact of ARR

At a market level, there is no change for the record label, and no change for songwriters and publishers in the short run. This is not to say that streaming services would not seek to renegotiate existing deals to recover the additional expense they are now experiencing.

There is a reduction in DSP share (dropping from 30% to 25%), which is distributed to the featured artist who increases their share from 21% to 23.9%, with the non-featured artist growing to 1.6% share.

Figure 15. Total Market: Spanish Model

Looking at this in revenue terms, there is a shortfall of around £42 million for the DSP, a gain of around £25 million for featured artists, and a gain of around £13 million for non-featured artists.

The administration cost – held at 10% for consistency with other models - is very low at £4.19 million per year. This emphasises the need for further scoping of the annual cost of implementation. It might well be difficult to reach the 10% assumed administration with this relatively low level of revenue being distributed by the third party.

Additional considerations for the Spanish model

The current model shares revenue between featured artists (65%) and non-featured artists (35%) but there is flexibility for a UK implementation to apply another ratio. It is also possible for a UK implementation to exclusively remunerate featured artists, or exclusively remunerate non-featured artists.

The second order impacts of market intervention also warrant serious consideration. While there are debates as to whether the intervention in Spain has led to significant changes in the deals done, there is no basis upon which to say what the implications might be for the UK.

The UK is a much larger market and as such if ARR was levied against the DSP then they might look to recover some or all the ARR revenue from either their label deals, or from their deals with the publishing and songwriter community. This is difficult to fully anticipate and would vary with the value of the levy and who was remunerated. It is an important topic for further research when considering the Spanish model.

4. Unintended consequences and further research

This research set out to examine the economic impact the introduction of ER might have. In the complex web of music business dynamics, a change on one side may have a knock-on effect on other relevant stakeholders which then may impact upon the performers. Some of the unintended consequences the introduction of ER might have are listed here.

4.1 The impact on the value of streaming rights

As discussed within the paper, label stakeholders have raised concerns that implementing ER on streaming might have a negative impact on the total value of the streaming market by reducing the negotiation power of the labels. Contrasting viewpoints were put forward by CMM who argue that there are many ways of implementing ER and it does not necessitate a drop in total value.

Since the exact application of the law is yet to be phrased, it is not possible to take a view on this. Your author has been at pains to emphasise that this paper does not conclude on this matter, and only maps out scenarios where value does and does not fall.

4.2 The impact on record label investment

Labels also suggested that the reduced share for labels caused by ER might lead to a reduced number of advances offered by labels, which then would impact upon artists and managers negatively. They also raised concerns over a slower return on investment under ER conditions, since they would be recouping from a smaller proportion of income, often at a lower rate, which then might lead to a reduced number of investments, or a reduction in the value of the advance, or both.

It is debatable, however, whether this would happen or, if it does happen, whether the impact would be negative. The artist community argues that it is not clear whether the advance level will fall.

Further research could therefore explore the value of investment from featured artists and non-featured artists investing in their own skills and development, as well as the value of their output. The opportunity for them to grow their teams, creating new roles, warrants further investigation in future research.

4.3 Implementation risks; costs, speed and accuracy

Under this economic modelling, an estimated admin cost of a CMO at a rate of 10% was applied. In further research, it is important to consider the trade-offs between cost, speed and accuracy when setting distribution policy.

Cost

It is important to model what the cost of enabling a Collective Management Organisation (CMO) to distribute funds would be, given the continued increase of tracks becoming available. It is also important to look at the administrative fees rights holders will have to bear, although this may reduce over time. The efficiency (expressed as a lower percentage admin rate) will be subject to the size of the task and the complexity of implementation.

Speediness

Speed of payment could be a concern, with many collecting societies distributing revenue every 6 months, which might be considerably slower than labels are currently able to offer. It is also possible that a body could report on the same frequency as the industry is accustomed to.

Data accuracy

There are also questions of data accuracy and completeness, especially when it comes to recording the non-featured performers at the less established end of the market. Currently the broadcast model applies to those who have reached a level of success (resulting in their music being used on television or radio) but streaming has a much wider reach, and a much larger number of songs and artists to remunerate, and at a lower price per performance than traditional broadcast mediums. The volumes of data involved also bring additional cost and complexity.

These three considerations can trade off with each other. For example, an inaccurate distribution can be done quickly and cheaply. The more accurate the distribution is, the more likely it is to take more time and require more money. This is recommended for further research.

4.4 Revenue concentration

Streaming revenues are heavily concentrated around the top 1% of artists and as such the majority of the royalty income would be concentrated around the top few artists. Data quality tends to improve the more popular a piece of repertoire is, and so, when it comes to matching usage to revenue in order to distribute, there is a risk that the most successful artists benefit disproportionately.

The scale of this and any mitigation measures is a consideration for future research when looking at the costs and benefits of third party administration and the impact on all non-featured artists of introducing ER.

4.5 Additional administration burdens

Under both the Broadcast and the Spanish models there is a new administrative burden created. Where this burden sits and how large or small it is depends on the specific implementation. Beyond the administration fees captured within the model there is a concern that it may introduce new legal costs for artists who are not signed to a record label. They might well need to account to their non-featured artists, which could potentially involve legal costs as well as time costs.

There is also a risk of administrative burdens being imposed on the label, who are currently accounting to the artist wholly from their own systems. Similarly, DSPs might also see an increase in the administration burden they experience if they are asked to report to labels in the current format, but to a central body on a different format. This could have a considerable negative impact and warrants further consideration.

Where this burden sits and how large or small it is will depend on the specific implementation, which is open to discussion. One thing that was not considered, but will have to be investigated in future research is how to accommodate artists who are not signed to a record label and how to account to their non-featured artists. The internal administration costs borne by the labels and DSPs also have not been considered in this model. Although it might be only the initial cost, future research might consider this additional cost.

5. Conclusion

While this paper did not set out to conclude in favour of or against any particular model, the one key conclusion that can be drawn is that ER does not offer a simple solution. There are a large number of unknowns and complex interdependencies within the modelling, and a range of further questions that merit deeper consideration. Importantly, this paper was not intended to determine what is fair, and that important debate will continue.

One of the most important and most contentious future considerations is to clarify whether a change in the nature of the rights would significantly weaken the rights holders’ ability to negotiate license fees, and indeed whether such a change in total value of rights is worthwhile in order to redistribute funds across the value chain. The hope is that such a change would not necessitate the drop in value, but this needs to be considered until it is confirmed that there is no risk of this, or that the risk is mitigated by value added elsewhere in the value chain.

The impact of ER on a label’s ability to meaningfully invest in developing new talent is another key consideration. If ER were to undermine the ROI from new talent development, there is a risk that those labels who had previously played significant roles in artist development might reduce this activity. The future market environment  may allow for success without label investment, but this is not known and not proven.

The cost of administration is a deadweight loss and is likely to be a significant sum of money. This money is certainly lost from the rights holders’ side of the market, and the upsides of redistribution need to be considered against this backdrop.

The other key consideration is how such an intervention changes incentives between rights holders on the creator side, and explore whether there are alternative ratios that would help optimise the relationship between labels, FA and NFA.

While not a satisfying conclusion, it is clear that more research is required into the nuances of how best to balance the incentives to create with the need to monetise creation.

1. Music Creators’ Earnings in the Digital Era 2021 (IPO)  ↩

2. The WIPO report stated that, “What remains is that performers transfer value to streaming services beyond that which is compensated by market centric royalty payments. It seems that the policy goals and principles of equitable remuneration are best fulfilled by a streaming remuneration in the nature of a communication to the public royalty that is outside of any recording agreement, is not waivable by the performer and it is collected and distributed by performers’ CMOs (49-50).”  ↩

3. Rome Convention 1961, article 7  ↩

4. WIPO Performances and Phonograms Treaty (WPPT), article 10  ↩

5. Rome Convention 1961, article 12  ↩

6. It is worth noting that some services have recently put their prices up, but the point here is that more total consumption does not necessarily lead to more total income.  ↩

7. How Much Music Is Added to Spotify & Other Streaming Services Daily? – Billboard  ↩

8. Music Creators’ Earnings in the Digital Era 2021 (IPO) - GOV.UK (www.gov.uk)  ↩

9. BPI_All About The Music 2023  ↩

10. Music and streaming market study: final report  ↩

11. Note that in response to the DCMS Economics of Streaming inquiry, Sony provided a sample contract in which overheads of 25%, and marketing costs of (at least) 15% were deducted from the artist’s share throughout the duration of the contract, regardless of recoupment: “Sony’s information suggests that, whereas the costs of advances and recording are fixed, those relating to marketing will increase as a recording achieves popularity. Also, while a label’s overheads are fixed, the contribution that each release makes to those overheads will be based on 25% of that release’s gross revenue and will therefore rise as its revenue rises (Hesmondhalgh et al., 2021: 220)”  ↩

12. committees.parliament.uk/writtenevidence/15300/pdf/ ↩