How the record
business pays.

In the twelve months after that television interview, the British record industry did something that looks, from a distance, like a collective loss of nerve. It signed nearly everybody.

Bands who had existed for four months were offered contracts. Bands who could not play were offered contracts. A&R departments who had spent 1975 looking for the next singer-songwriter spent 1977 in the Roxy and the Vortex with a chequebook, and a good many of the deals signed in that period were plainly not commercial decisions in any ordinary sense.

The usual explanation is that the industry panicked. That is true and it is not the whole story, and the whole story requires understanding what an advance is, which is the single most useful thing in this book for anyone who has ever wondered how a band with a hit record ends up with nothing.

An advance is not a payment. It is a loan against money you have not yet earned, and you repay it out of your own royalties before you see a penny.

From PUNK, Movement Four. Read the volume →

Time to trace the coin. A teenager puts eighty-nine cents on a counter in 1957 for a hit single on an independent label. The figures below are period-typical rather than audited from any one company’s books (the real deals varied, and most of the books are gone) but the proportions are right, and the proportions are the argument.

The dealer kept roughly a third; call it thirty cents. The distributor took his margin (say a dime) and remitted, eventually, on his own schedule, as Movement Two described. The label’s gross was somewhere around fifty cents. Out of that came pressing at ten or twelve cents, the sleeve, freight, and the two-cent mechanical to the publisher, which, if the label owned the publishing, was the label paying itself. Out of it also came promotion, a word covering the mailed records, the trade advertisements, and the folded handshakes the next chapter is about. What remained, perhaps fifteen or twenty cents on a clean sale, before returns, was the label’s, and the artist’s royalty came out of nobody’s share so much as off the top of the artist’s own future: a contracted two cents, say, credited against an account already carrying the session costs, the free goods and the breakage deduction from four chapters ago.

From ROCK 'N' ROLL, Movement Four. Read the volume →

Every book about the music business has to explain this at some point and most of them do it badly. Here it is as plainly as possible, because once you understand it, a great deal of what happened to the people in this book stops being mysterious.

A song and a recording of a song are two different pieces of property, owned separately, earning separately.

The recording is the specific performance captured on tape. It is normally owned by the record company, which paid for it. Income from selling copies of it flows to the company, which pays the artist a royalty out of it, after the deductions in the last chapter.

The song (the melody and the words, as an abstract thing, independent of any particular recording) is owned by whoever wrote it. That ownership is called the publishing, and it earns in two entirely separate ways.

Mechanical royalties are paid every time a copy of a recording of that song is manufactured and sold. They go to the songwriter, and they are calculated from the retail price, and (this is the crucial part) they are generally not subject to the same deductions and are not recouped against the recording advance. They are a separate stream, from a separate contract, to a different person.

From PUNK, Movement Four. Read the volume →

A song and a recording of a song are two different pieces of property, owned separately, earning separately. The recording (the performance on the tape) belonged, almost always, to the label that paid for the session. The song (melody and words as an abstract thing) belonged to whoever wrote it, or more precisely to whoever ended up holding the copyright, which this movement will show was not reliably the same person. Ownership of the song is called the publishing, and in the 1950s it earned through two channels.

The first channel was the mechanical royalty, and its terms had been set by Congress in 1909, when the recorded-music industry’s dominant product was the player-piano roll. The Copyright Act of that year did two things that governed everything in this book. It fixed the mechanical rate by statute at two cents per copy manufactured, a rate that then sat unchanged for sixty-nine years, through two world wars and the entire history this series covers, quietly transferring value from songwriters to record companies with every year of inflation. And it created the compulsory licence: once a song had been recorded and released with the owner’s consent, anyone could record it, without permission, by paying the statutory two cents a copy. Congress wrote that rule to stop one piano-roll company monopolising songs. Its unintended consequence sits at the centre of the next chapter: no artist and no label could prevent a rival from copying a hit, note for note, the week it charted. The law priced the song and unpriced the performance, an arrangement, it turned out, with a racial geometry nobody in 1909 had imagined.

From ROCK 'N' ROLL, Movement Four. Read the volume →

A race-record artist of the twenties and thirties was, in the standard case, paid a flat fee per usable side, in cash or by cheque, at or shortly after the session, and that was the end of the transaction. Not an advance against royalties, which is the machinery readers of this series know from later volumes, with all its deductions and its long accounting shadow. No royalty existed to advance against. The fee bought the performance outright: the company owned the master, the sales, the reissues, the whole commercial life of the recording, for ever, and the performer’s financial interest in the record ended as the studio door closed. If it sold three hundred copies, the fee was the payment. If it sold three hundred thousand, the fee was the payment.

The numbers, where they survive, are these. Field-trip discoveries and minor artists got amounts in the region of five to twenty dollars a side. Established names commanded more: the figures preserved in Columbia’s files for Bessie Smith (preserved because her biographer went and read them, which is the only reason this book can be precise about anybody) run from one hundred and twenty-five dollars a side in her first contracts to two hundred at her peak, flat, with a written guarantee of so many sides a year. Work the arithmetic that follows from it. Her first release sold, by the figure taken from those same files, some three-quarters of a million copies. Her share of that sale was the session fee she had already been paid (five hundred dollars or thereabouts for the day’s two sides) and her label’s share was everything else. Across her whole decade at Columbia, some hundred and sixty sides, the total the files support is under thirty thousand dollars: the recorded career of the highest-paid Black entertainer in America, bought outright for roughly the price of a large suburban house. She earned her actual living, as the tent-show chapter said, on the road, where the box office could not be filed away from her.

From BLUES, Movement Four. Read the volume →

There is a third party in every deal in this movement and the histories consistently underweight him, partly because he is rarely photogenic and mostly because he does not want to be discussed.

A manager in 1977 typically took twenty per cent. Sometimes fifteen, sometimes twenty-five. The number is not the interesting part. The interesting part is the word that comes after it, and it is one of two words.

Twenty per cent of net means the manager takes a fifth of what is left after the band’s costs, the van, the equipment, the hotel, the crew, the recording. He is in the same boat as his artists and does badly when they do badly.

Twenty per cent of gross means he takes a fifth of every pound that arrives, before any of those costs are paid.

That distinction, one word in a contract most of these musicians did not read and could not have afforded a solicitor to read for them, produces outcomes so different they are barely comparable.

From PUNK, Movement Four. Read the volume →

Take the most famous contract in British independent music (profits split equally after costs, the rights staying with the artist, no long contract, nothing signed) and ask the question almost nobody asks about it: was it generous?

The answer is not obvious, it is arithmetic, and the arithmetic rearranges most people’s intuitions. Fifty per cent of a profit is worth less than twelve per cent of a gross if the gross is fifty times larger, and it is worth a great deal more if the alternative is an advance you never recoup and a catalogue you never see again. Which of those two situations a musician was actually in depended on numbers this movement can now supply.

Recall the major-label machine from the previous volume. A royalty of around ten per cent of dealer price, shrunk by packaging deductions, breakage clauses and free goods to an effective five or six per cent of retail (perhaps twenty-odd pence on a four-pound album) paid only after a large advance and all recording costs had been recouped from the artist’s share alone. The company banked the wholesale margin from the first copy; the artist commonly saw nothing, ever, beyond the advance.

From POST-PUNK & NEW WAVE, Movement Four. Read the volume →

The costs. A first-division progressive album of 1974–76: five to nine months across first-rank London rooms and a residential studio, at rates in the tens of pounds per hour, with orchestral sessions, guest choirs and the mixing marathons of Movement Two. Recording costs in the tens of thousands of pounds, the era’s flagship projects were reported in the trade press at figures that crossed six figures by the decade’s end, a several-fold multiplication inside ten years. Add the deluxe sleeve, gatefold, inner bags, booklet, and tour support for the arms race, both recoupable in the standard agreements of the day. Call the all-in recoupable burden on a flagship album one hundred thousand pounds and upward, and know that the biggest projects exceeded it.

The income, artist’s side. An established band renegotiating mid-decade commanded royalties well above the punk volume’s debutant ten per cent, twelve to sixteen per cent of retail was the reported range for acts of this rank, sometimes better. Run the standard deductions of the period, packaging allowances, free goods, the archaic breakage provisions that survived in many agreements, and the effective take lands somewhere around forty to sixty pence per full-price British album, split across four or five members and their manager, against every recoupable pound spent first. On that arithmetic a hundred-thousand-pound album must sell in the vicinity of two hundred thousand full-price copies before the band’s account crosses zero, and, unlike the punk case, the bands in this book routinely did: the flagship albums of Movement Three sold in the hundreds of thousands to millions, which is why this genre’s first division actually got rich where punk’s mostly did not. The machine worked as designed, for the few dozen musicians at the top of it. The tier below (the second-division bands signed in the gold rush of 1970–73, spending first-division studio money on second-division sales) lived the punk arithmetic at ten times the stakes: famous, reviewed, touring, and unrecouped for years. Their names fill the genre’s reissue catalogues now; their period bank statements, by every memoir that mentions them, were a monthly humiliation.

From PROGRESSIVE ROCK, Movement Four. Read the volume →

No chapter in this book explains more about where its people physically were, in its final years, than the one thing rock histories least like to discuss: British income tax.

The rates first, plainly, because they sound invented. From 1974, under the incoming Labour government’s first budget, the top rate of United Kingdom income tax on earned income was eighty-three per cent, reached at twenty thousand pounds a year. Investment income above a threshold bore a further fifteen per cent surcharge, taking the top marginal rate on unearned income to ninety-eight per cent. These were marginal rates, not averages, and ordinary earners never met them; but a musician whose album had just done what Movement Three’s records did met them immediately, in the year the money arrived, with no averaging to soften a career that might pay a lifetime’s income in thirty months. At the margin, a pound earned kept seventeen pence, or, on royalty streams structured as investment income, tuppence.

The escape was residence. Tax fell on those resident in the United Kingdom; a full tax year of verifiable non-residence, within rules about counted days that a new profession of advisers grew rich interpreting, took a year’s worldwide income out of reach. So began the era of the tax exile, and the reader has already met its artefacts without the label. ELP decamping to the Bahamas, where their contractual-obligation album was made under the beach-shirt sleeve of Movement Three. The Rolling Stones had shown the way to the south of France at the decade’s start; by its middle years the traffic was general (Switzerland, Monaco, Los Angeles, the Caribbean) and the trade press tracked residencies the way it had once tracked signings. Pink Floyd spent 1979 recording The Wall in the south of France substantially because, as Nick Mason’s memoir sets out with an accountant’s calm, the band’s circumstances that year made non-residence a necessity rather than a choice. Why that year (why the richest band in this book suddenly needed a tax year abroad) is this chapter’s centrepiece, because it is the best-documented case of what the era’s money actually did when it landed.

From PROGRESSIVE ROCK, Movement Four. Read the volume →

It is called a split sheet, and in a writing camp it is the most consequential object in the building.

To see why, you need the two-property distinction this series has now made three times, because it governs everything in this chapter.

A song and a recording of that song are separate pieces of property, owned separately, earning separately.

The recording is the specific performance. The label normally owns it, because the label paid for it. The performer receives a royalty out of it, after the deductions described at length in the punk volume, the dealer-price basis, the packaging allowance, free goods, producer points, and recoupment of the advance and the recording costs.

The song (the melody and words as an abstract thing) belongs to whoever wrote it. It earns in two independent ways. Mechanical royalties are paid on every copy manufactured and sold, at a rate set by statute in the United States, which through the period of this book ran between roughly seven and eight cents per song per copy. Performance royalties are paid whenever the song is broadcast or played in public, collected by the performing-rights societies and paid to writers and publishers.

From TEEN POP, Movement Four. Read the volume →

One narrower legal question decided which parts of this book’s catalogue exist commercially and which do not.

The mixtapes were built on other people’s records. The chapter on the record you did not pay for describes the convention: an artist raps over the year’s biggest instrumentals, the tape is given away, nobody licenses anything, and the industry treats it as promotion.

That works perfectly as long as nobody is being paid. The moment a record earns (which is what streaming did, retrospectively, to a decade of free releases) every uncleared use becomes a liability, and the record cannot be put on a service at all.

So the era split into three.

Records that were always original and cleared moved onto the services and have been earning ever since.

Records that were reissued with the problems removed (replaced instrumentals, re-recorded sections, cleared samples where the owners would deal) exist in altered form. Listeners who know the originals frequently regard the reissues as a different and worse record, which is fair, and the alternative was that they existed nowhere.

From TRAP, Movement Four. Read the volume →

The guest verse is this genre’s most distinctive commercial instrument, and it is conducted almost entirely by invoice.

How it works.

An artist wants another artist on their record. A price is agreed (usually through managers, sometimes directly) and paid, frequently in advance and frequently in full before the verse is delivered. The guest records their part, sends it, and is paid.

The range across this book’s period is wide and the figures circulate freely in the scene: a working artist with regional standing might command a few thousand dollars; an artist with a current hit, tens of thousands; a major star, a six-figure sum for sixteen bars.

Three properties make it unlike anything else in this series.

It is paid up front and it is not recoupable. Unlike a royalty, unlike an advance, unlike almost everything in the previous twenty-nine volumes, the feature fee is money that arrives and stays. For an artist unrecouped at a label (which is most of them) it may be the only music income that behaves like a wage.

From TRAP, Movement Four. Read the volume →

Follow the T-shirt.

A band T-shirt in 2004 cost somewhere between four and six dollars to produce in quantity and sold at a show for twenty to twenty-five. That is a gross margin no other product in this book approaches, better than a compact disc, far better than a ticket, and immeasurably better than a royalty against an unrecouped balance.

Now consider who is buying it. The audience described in the chapter on who was actually there is thirteen to nineteen, wears its affiliations visibly as a matter of social necessity, attends shows with friends, and treats the shirt as the point of having gone. Merchandise conversion at shows in these genres was, by every account from people who ran the tents, extraordinarily high compared with other kinds of music.

And the crucial structural fact: this was frequently the band’s own money.

Under the recording contracts of the period, merchandise income mostly belonged to the artist and was not recoupable against the record advance. So a band with a debt of two hundred thousand dollars at the label could nonetheless come off a summer tour with real cash, because the tent’s takings went into a different pocket.

From EMO & NU-METAL, Movement Four. Read the volume →

If the arena was the market, then the only shop window in the genre was the forty minutes before the headliner, and it is worth dwelling on how that forty minutes worked, because it was the apprenticeship system of the entire industry described in this book, and because its terms tell you, more nakedly than any other document, who had leverage and who had none.

A support act on a major American arena tour in the middle seventies was paid a flat fee per night, and the fee was small: the figures that surface in memoirs and trade retrospectives run from a few hundred dollars to perhaps a thousand for a valued opener, and this book passes them on as the shape everyone remembers rather than as audited payroll. Against that fee the opener carried real costs (travel, crew, equipment cartage, hotels) that reliably exceeded it. Supporting a big tour, in other words, lost money by design. The nightly deficit was met, for signed bands, by tour support: an advance from the record label to cover the shortfall, which was (and here is the word that rules this movement) recoupable. The label lent the band the money to lose money in front of the right audience, and would take the loan back later out of the band’s own record royalties. The economics of apprenticeship, formalised: the young band paid for its own exposure, with borrowed money, at interest paid in ownership.

From HEAVY METAL, Movement Four. Read the volume →

If the record no longer pays and the tour and the T-shirt and the advert do, then a record company facing a halved business has one obvious move: stop being a record company.

That is the 360 deal, and it is the most consequential contractual change in this volume.

In a conventional agreement a label participates in record income and nothing else. Touring, merchandise, publishing, sponsorship and licensing belong to the artist. In a 360 (sometimes called a multiple-rights deal) the label takes a percentage of some or all of those other streams as well, in exchange for a larger advance and, in principle, a longer-term commitment.

The reported milestones are public. In 2002 EMI signed Robbie Williams to an arrangement widely reported at the time as being worth around eighty million pounds and as covering revenue well beyond records. In 2005 EMI’s deal with Korn was reported as buying a share of the band’s other income for a substantial guaranteed sum. Then the live-entertainment companies moved in from the other direction: Live Nation’s arrangements with Madonna in 2007 and with Jay-Z in 2008 (reported figures in the hundreds of millions between them) were record deals signed by a concert promoter, which is the whole shift in one sentence.

From INDIE & THE BLOG YEARS, Movement Four. Read the volume →

This is the hardest sum in the series and the answer is not a number. It is a formula, and the formula is the finding.

There is no per-stream rate. That sentence contradicts almost everything written about streaming, including by people who should know better, so here is the actual mechanism used by the major subscription services.

In a given month, in a given country, the service collects its subscription and advertising revenue. It keeps a share (commonly reported in the region of thirty per cent) and places the rest in a pool. It then counts every stream in that country that month, works out what fraction of them belonged to each recording, and distributes the pool in those proportions to the rights holders.

That is called pro-rata, and it has one consequence that almost nobody’s intuition supplies: the value of a stream depends on how much money is in the pool it is drawn from, which depends on what subscribers in that country pay.

So a play is worth different amounts in different countries, and the difference is enormous. A subscriber in Norway or the United States pays a monthly fee several times what a subscriber in Nigeria, India or Brazil pays, because prices are set to local purchasing power, which is a sensible and defensible commercial decision. The pool in a high-price market is therefore much larger per listener, and every stream in it is worth much more.

From GLOBAL POP, Movement Four. Read the volume →

Payola was not invented by rock 'n’ roll and was not, when this book’s span began, a crime. Song pluggers had been paying vaudeville singers to perform their publishers’ tunes since the sheet-music era; the word itself was trade slang decades old. In the radio economy Movement Two described, payment for play was the small independent label’s marketing department: a fifty-dollar handshake, a weekly retainer described as a consultancy, a listening fee, a Christmas envelope. The last three chapters showed its upmarket forms (the cut-in credit, the publishing share) which paid better and photographed worse. Through most of the fifties, none of this violated any federal law. It was, at most, a tax problem, and among the disc jockeys of America it was as close to universal as makes no difference.

What turned standard practice into a national scandal was not an outbreak of ethics. It was a war between two licensing societies, and the war was about exactly the money described three chapters ago.

ASCAP’s old-line writers had watched, across the decade, as the airplay economy they had dominated for forty years re-routed itself through BMI, through, as they saw it, a broadcaster-owned rival that paid hillbillies and rhythm-and-blues men for music no civilised station would play unless someone had rigged the game. Their counter-offensive ran the length of the fifties: a hundred-and-fifty-million-dollar antitrust suit by ASCAP-aligned songwriters in 1953; House hearings in 1956 into the broadcasters’ ownership of BMI; a Senate bill in 1958 to force the divorce, at whose hearings the era’s most quotable witness statement was entered. Frank Sinatra, in a written broadside first published in a magazine the year before, dismissed rock 'n’ roll as music sung, played and written for the most part by cretinous goons, and the sentiment, if not the phrasing, was the campaign’s whole theory: the music was trash, therefore its popularity must be purchased, therefore the purchasers should be found. Each effort failed on the same evidence (the sales data kept showing that the records the children bought were the records the children wanted) but the theory was now on file in Washington, waiting for a vehicle.

From ROCK 'N' ROLL, Movement Four. Read the volume →

Every music economy in this series has one asset class at its centre. Punk’s was the master tape of a cheap single; disco’s, when this series reaches back to it, the twelve-inch mix; hip-hop’s the sampled break. Jamaica’s is the riddim, and it is the strangest and most instructive of them all, because it is the only one that behaves like land.

Define it with an example, the most-worked plot on the island. In 1967 the Studio One session crew cut an instrumental called Real Rock, a bass figure, drums, organ and guitar, two bars of identity repeated for three minutes. That recording, under and after its original song, became a named thing in itself: the Real Rock riddim. It carried deejay cuts and dub cuts in the seventies; it was recut, re-recorded from scratch in the new era’s drum language, at Channel One and everywhere else; it carried fresh vocals in the dancehall era and the digital era and carries them still. Nobody has audited the full count of records built on it; every attempt runs into the hundreds. The rocksteady and early reggae catalogues of Studio One and Treasure Isle together (a few hundred rhythm tracks cut by session men at day rates across perhaps six years) constitute the ground on which a measurable share of all subsequent Jamaican music has been built.

From REGGAE, Movement Four. Read the volume →

The reggae volume described the dubplate’s invention: the one-off acetate, cut on a lathe, that let a Kingston sound system own music nobody else possessed, exclusivity as a manufacturing standard, in an economy where the dance, not the record, was the product. Forty years later and four thousand miles away, the jungle scene ran the identical instrument, for the identical reasons, at the identical position in the economy, and, in a detail that would strain a novelist’s nerve, frequently for the grandchildren of the same families.

The mechanics, as Movement Two sketched them. A producer cut a finished track to acetate at Music House for a fee in the region of thirty pounds a plate; the plate went to one DJ, or a handful, and the tune then lived (for weeks, months, in famous cases north of a year) as an exclusive, played on pirates and in the rooms of Movement One, unavailable at any price, known to thousands of people who could not buy it. The acetate itself was consumable: soft-surfaced, good for a few dozen plays before the top end audibly died, the same physics as the reggae volume, unimproved because it did not need improving. A working DJ of the first rank carried a box holding hundreds of pounds of cutting fees and, in trade terms, priceless inventory; participant memoirs of the period return constantly to the box, guarded, insured by vigilance, occasionally stolen, the theft of a plate box being the scene’s equivalent of a bank job.

From RAVE & JUNGLE, Movement Four. Read the volume →

On the first of August 1942, the American recording industry stopped. Not slowed, stopped. Every union instrumentalist in the United States, which for practical purposes meant every professional instrumentalist, ceased to make records: no sessions, no sides, no exceptions for stars or friends or unfinished projects. The stoppage held, at the two biggest companies in the business, for two years, three months and ten days. It is the longest and most complete shutdown of an art form’s documentation ever undertaken on purpose, and it was undertaken over the question this movement has been circling since the jukebox chapter: what does the machine owe the musician it replaces?

The union’s case, stated in Petrillo’s blunt public arithmetic, ran as follows. A record was made once, by a handful of musicians, for a flat fee. It was then performed endlessly (on several hundred thousand jukeboxes, on the disc-jockey programmes that were replacing radio’s live orchestras) and every one of those performances was a bandstand nobody hired. The companies profited from each pressing; the composer collected his statutory pennies; the performers whose playing was the entire content of the object received, in perpetuity, nothing. The law offered no performance right, Congress had declined to create one, and individual musicians could not strike against their own session fees. Therefore the federation would do collectively what no member could do alone: withhold the product until the companies paid, not to the session men individually, but into a union fund, financed by a royalty on every disc sold, to employ musicians displaced by the very records they were making.

From JAZZ, Movement Four. Read the volume →

Begin with the emblem. Robert Johnson’s twenty-nine sides, bought flat in two hotel-room blocks for sums nobody recorded, sold modestly and went out of print. In 1961 Columbia (corporate heir, through the ARC acquisitions, to masters it had never much valued) issued an album of them for the folk-revival market; the record became a foundational text for the sixties guitar generation, described in the ending movement. In 1990 the label issued the complete recordings as a boxed set, which went platinum, a certification, for a catalogue of pre-war 78 sides, that no participant in this book’s first half could have parsed as a sentence. The composer royalties that had accrued into escrow, because the writer credit, for once, was uncontested and unowned by any middleman, became the object of a decade of estate litigation, resolved when Mississippi’s courts recognised as heir a retired gravel-truck driver named Claud Johnson, born to a woman the singer had known in 1931, whose paternity evidence included a witness to the conception. He received, by the settled accounts of the case, well over a million dollars. The man himself had been paid, in his lifetime, in cash, per side, amount unrecorded.

From BLUES, Movement Four. Read the volume →