How Music Became an Industry

On March 24, 1721, Johann Sebastian Bach presented a bound manuscript of six concertos to Margrave Christian Ludwig of Brandenburg-Schwedt, a minor Prussian nobleman with musical ambitions and a court orchestra too small to perform most of them. Bach had composed the works hoping for a commission — a salary, a position, a patron. He addressed the dedication with the expected groveling: “I humbly beg Your Royal Highness not to judge their imperfection with the rigor of the fine and delicate taste which everyone knows You to have for musical works, but rather to take into benign Consideration the profound respect and the most humble obedience which I thus attempt to show You.”

The Margrave probably never had the concertos performed. Bach received nothing. He had been working at the time as Kapellmeister in Köthen, at a salary of 400 thalers per year, writing music on demand for his employer’s entertainment. The Brandenburg Concertos were a side project — a speculative investment in patronage that didn’t pay off.

This is the ur-form of the music economy: a transaction between a composer-servant and a wealthy patron, with music as a private luxury good consumed exclusively by the people who could afford to maintain musicians. The entire subsequent history of the music industry is the story of how this model broke down and what replaced it — always, in each iteration, through the construction of some new form of artificial scarcity around an art form that is, by its nature, a public good.

The Patronage Model and Its Logic

Understanding the patronage model requires understanding what music actually is as an economic object. Music is non-rivalrous — one person listening to it does not prevent another from listening to it. And in the pre-recording era, music performed in a room was essentially non-excludable at the point of consumption — you could hear it whether you paid or not, if you were close enough to the source.

This is the definition of a public good: non-rivalrous and non-excludable. Economic theory predicts that public goods will be underprovided by markets, because producers cannot capture the full value their goods create — free riders consume the good without paying, and the producer cannot stop them.

The patronage system solved this problem through a mechanism that has nothing to do with markets. The patron purchased music as a private luxury and kept it private — the court orchestra played for the court, the church organist played for the congregation, the chamber musicians played for the noble household. The scarcity was enforced by physical exclusion from the performance space. If you couldn’t afford admission to the Duke’s salon or the cathedral choir, you simply didn’t hear the music. Non-excludability was solved by walls.

But the patronage system created its own political economy: composers were servants. Bach’s employment contracts specified not just the music he was required to compose but his duty to train choir boys, maintain instruments, and not leave town without permission. Handel’s early career at the Hanover court included a clause permitting him to visit England — a clause he exploited liberally, eventually absconding to London permanently and abandoning the Elector of Hanover, who then became King George I of England and had to be placated with the Water Music. Even at the top of the field, composers were dependents, not independent creators.

The patronage model began breaking down in the eighteenth century not because of any philosophical change in attitudes toward artistic freedom but because of a social and economic change: the emergence of a prosperous urban middle class with money to spend on cultural consumption and no access to aristocratic patronage networks.

Public Concert Halls as the First Disruption

The public concert hall was a commercial innovation, not a cultural one. Impresarios in London, Paris, Vienna, and elsewhere discovered that they could charge admission for musical performances — that aggregating an audience of paying customers could substitute for a single wealthy patron. The concert hall created a new revenue model: multiple small payments instead of one large one, with the music as a semi-public good within the space of the hall.

This changed everything about the music economy. Composers who could attract audiences could earn independently of patrons. Haydn’s late London symphonies — composed for Salomon’s subscription concerts in the 1790s — were the product of a new commercial arrangement: Haydn received a fee and a share of ticket revenue in exchange for composing and conducting. The music was written for a paying public, not a noble household. The result, arguably, is more ambitious and innovative than anything he wrote for the Esterházy court, because the incentive was now to impress and retain an audience rather than satisfy an employer.

But the concert hall model had limits. It required physical presence — audience and performer in the same place at the same time. Performance remained the irreducible unit of the music economy. This meant that music’s economic scale was bounded by how many people could fit in a room and how much they would pay to be there. Beethoven could command large audiences at the Theater an der Wien, but his reach was still fundamentally local. The music that circulated beyond the performance hall circulated primarily in the form of sheet music.

Sheet music publishing was the first technology that made musical reproduction possible at scale — the first instance of music as a mass medium. Publishers like Breitkopf & Härtel in Leipzig employed composers to produce editions of their works for sale to amateur musicians who would perform them at home. This was a genuinely new model: the music itself became a commodity, reproducible and tradeable independently of performance. The composer received a flat fee or royalty. The publisher manufactured and distributed the copies.

Sheet music democratized musical access in one direction (more people could buy music and perform it) while concentrating revenue in another (publishers captured most of the economic value). Composers did better than under patronage — they had more commercial options — but they were still dependent on intermediaries who controlled reproduction and distribution. The technology of reproduction had changed; the basic power relationship had not.

Sound Recording and the Great Disruption

The phonograph changed the music economy more fundamentally than any previous development, because it did something that no prior technology had done: it separated the performance from the consumption of music in time and space. For the first time, a listener could consume music without being present at a performance, without reading sheet music, without any musical skill whatsoever. Music became a consumer product in the modern sense — something you purchased and consumed at your convenience.

The economic implications were enormous and took decades to work out. The initial reaction of the music industry — which at this point meant primarily publishers and performing artists — was panic. If consumers could listen to recordings rather than attending concerts, the concert business would collapse. If people could hear music at home rather than buying sheet music, the publishing business would collapse. Both predictions proved partly wrong and partly right.

What actually happened is that the recorded music industry built a new model on a crucial peculiarity of the medium: physical reproduction costs. A vinyl record cost money to manufacture, distribute, and retail. This physical cost structure created a revenue model: consumers paid for the physical object, with a portion of the payment flowing back to the rights holder. This was artificial scarcity — the music itself had zero marginal reproduction cost after the master recording was made, but the physical format imposed a cost structure that made pricing possible.

The model was extraordinarily profitable for everyone except the artists. Record labels captured most of the value because they controlled the capital-intensive steps of the chain — recording studios, manufacturing plants, distribution networks, and retail relationships. Artists received royalties, but those royalties were structured to minimize payouts: recording costs were charged against royalty accounts, returns were generous to retailers but not to artists, and contract terms were written to extract maximum value from successful acts while the label bore minimum risk. The industry that emerged from the Golden Age of recorded music — roughly 1950 to 2000 — was one of the most profitable in history, and it was built on manufactured scarcity in a medium that had, after the initial capital investment, essentially zero reproduction cost.

Streaming and the End of Manufactured Scarcity

Digital technology eliminated manufactured scarcity completely. The transition was not gradual — it happened in roughly a decade, between the arrival of Napster in 1999 and the commercial establishment of Spotify in 2008. Within that decade, the entire revenue model of the recorded music industry — built on the physical cost structure of vinyl, cassette, and CD — collapsed. The music was free at the point of copying, and no amount of legal action could prevent it from being treated as free.

Streaming was the industry’s attempt to reconstruct a revenue model on the ruins of the old one. The logic was straightforward: if music is freely available as digital files, perhaps consumers could be induced to pay for convenience, curation, and legitimate access rather than for the music itself. This works — streaming services have hundreds of millions of paying subscribers — but the revenue model generates dramatically less per-listen revenue than the album sale model it replaced.

The arithmetic of streaming royalties illustrates why. Spotify pays approximately $0.003 to $0.005 per stream to rights holders — which is then split between the label, the publisher, and the artist according to contract terms that typically leave the artist with fifteen to twenty percent of the total. A million streams — an impressive number suggesting significant popularity — generates perhaps three to five thousand dollars for the rights holder pool, of which the artist might see five hundred to a thousand dollars. A platinum album sale in 2000 would have generated far more. The orders of magnitude difference is not a policy choice that could be reversed — it reflects the underlying economics of a medium with zero marginal reproduction cost and no artificial scarcity.

The Permanent Problem

The music industry has never solved the fundamental economic problem that defines it: music is naturally a public good, and creating sustainable livelihoods for the people who produce it requires constructing some artificial mechanism that limits either access or reproduction.

Patronage did this through social exclusion — music for the wealthy only. Concert halls did it through physical presence requirements — you had to be there. Sheet music did it through the cost of reproduction. Recorded physical media did it through manufacturing costs. Streaming attempts to do it through subscription bundling and the inconvenience of piracy. Each of these is a genuine solution to the economic problem, but each generates a different distribution of value among creators, intermediaries, and consumers.

The pattern across all these models is consistent: intermediaries capture most of the economic value, creators receive a fraction of what their work actually generates, and consumers pay some price that lies between zero and what the marginal consumer would pay. The specific form of the artificial scarcity determines who the intermediary is — patron, concert promoter, publisher, label, streaming platform — but the presence of an intermediary capturing the majority of value is a constant.

Streaming is not an aberration or a failure of music industry policy. It is the logical endpoint of a process that has been underway since the phonograph: the progressive elimination of physical friction in musical reproduction, which steadily erodes whatever artificial scarcity mechanism the industry relies on. The only artificial scarcity remaining is the subscription wall, and the subscription model generates far less revenue per unit of consumption than any prior model.

The musicians who blame Spotify for their low royalties are identifying the wrong culprit. Spotify is a symptom. The disease is the fundamental economics of a non-rivalrous, non-excludable good in a world where reproduction is free. Bach’s situation — talented creator, dependent on institutional intermediaries who captured most of the value — has not fundamentally changed in three hundred years. What has changed is the identity of the intermediary and the fraction of value the creator receives, which has moved in the wrong direction.

The music industry will not find a stable model because there is no stable model to find. Music is a public good. Every revenue model for public goods is inherently unstable, dependent on maintaining some artificial constraint that technology or social change can always erode. The next disruption will not solve this problem. It will merely change which artificial scarcity is being maintained and who gets to maintain it.