The economics of superstars: Can Taylor Swift her AI era?
Claims that AI will make it harder for working class talent to succeed in the creative industries miss the point – the arts are inherently unequal, and not because of technology, says Paul Ormerod
The cultural and creative industries have been very much in the news over the past week
A report fronted by the crossbench peer Baroness Kidron warns of the threat to the sector from AI.
It does not so much as warn that it screams from the rooftops that the jobs of 2.4m people in these sectors are already at risk from the spread of AI.
The sector as defined in this report is a broad one, embracing the arts, media, design, publishing and entertainment worlds.
The problem is that large tech companies have the ability to scrape the work of composers, artists and writers and use the results to generate offers which are more competitive. This seems to be done more or less without regard to the laws of copyright.
A second report was led by Nazir Afzal, Chancellor of the University of Manchester.
Class ceiling?
Entitled Class Ceiling, it argued that there are major barriers which prevent working class talent from succeeding in this sector, such as low pay, class-based discrimination and a lack of the social connections needed for entry into the sector.
But this sector of the economy has an inherent tendency to generate a distribution of income across its workers which is massively unequal.
Taylor Swift, for example, has become a billionaire. In contrast, most people involved in the music industry make less than national average earnings.
This huge disparity in earnings is itself a key reason why many young people are excluded from participating in the cultural and creative industries.
Most people in the sector are low paid, and so family support is important for young people trying to make their way, hoping for the big break. Without it, aspirants simply give up through economic necessity.
The driving force behind the inequality isn’t technology. The internet era has intensified the trends, but the trends were already present well before the current hyperconnected web-based world appeared.
The American economist Sherwin Rosen captured some of the key arguments in a brilliant paper called, quite simply, “The Economics of Superstars”. He published it in 1981, years before the internet really existed.
He starts with the very reasonable assumption that people prefer a product which is slightly better quality than its competitors, even if the difference is small.
We would not necessarily expect huge differences in sales just because of this. Some marginally inferior offers might set lower prices, for example, and take lower profits.
But in many cultural and creative industries which supply mass markets, technology enables small differences to be magnified, generating “winner take all” outcomes.
The key is what is known in the jargon of economics as “joint consumption”.
If there is just one bunch of bananas left on a market stall and you get there just before me, you get to eat them and I don’t. If you watch a film on Netflix, this does not prevent me, or anyone else, from watching it as well. And it makes virtually no difference to the cost of streaming the film whether it is watched by just ten people or by 10m.
This effect is strongly reinforced by the fact that in this type of market, people often attach considerable value to joint consumption. Having watched the same film separately gives you something to talk about.
This leads to offers which are already popular becoming even more popular.
Taylor Swift is good, but is she really a thousand times better than a performer who makes a mere million? Obviously not, but the market structure and technology make it so.
The creative industries have already been massively impacted by the internet. AI is just one more challenge they have to cope with.
Paul Ormerod is an Honorary Professor at the Alliance Business School at the University of Manchester. You can follow him on Instagram @profpaulormerod