How Music Works

by David Byrne

  • $50 to $60 million: Advances for up to three new albums. As with a regular record deal, this money only gets handed over when Madonna delivers the music for each record. So it’s possible that the company won’t have to pay this out entirely. The Material Girl may not be moved to record another thirty-six to forty-five more songs, which is what it would take to fulfill that contract and get the entire advance.

  • $50 million in cash and stock for the right to promote Madonna’s concerts and license her name. Note that Live Nation still has to share concert and licensing revenue, which will give her 90 percent of concert sales (probably net) and 50 percent of the licensing money it collects.

  I, for one, would not want to be beholden to Live Nation. They’re a spinoff of Clear Channel, the radio conglomerate that turned much of the US radiospectrum into pabulum. But Madge is a smart cookie; she’s always been adept at controlling her own stuff, so we’ll see.

  According to my manager David Whitehead, “A new band, now at EMI, signed a deal like that. They didn’t have a track record—no one had heard of them, so they had no negotiating leverage. EMI presumably negotiated a nice portion of album sales, merch, and a cut of touring.”

  One can see the logic in this model as record sales decline and the profits from downloads don’t offset the loss. Record companies (and even concert promoters) feel that since they are the ones who helped create a popular artist/brand, they should naturally see a percentage of the profits from all possible revenue streams. And that seems fair, especially when the upfront investment is so high. If I’d spent millions bankrolling Lady Gaga’s records, producing those elaborate videos and marketing plans (and I know nothing of her finances—she may have bankrolled it all herself), I’d sure want a piece of her live shows and any other lucrative sources of revenue that might come down the road.

  All the major labels these days tend to want to sign artists to 360° deals. The question is whether the deal is “passive” or “active.” In a passive deal, the label skims off some percentage of sales from the licensing income but isn’t involved in an artist’s business in other ways. As long as the label gets their money when they come around asking, they won’t be telling the artist how to run her career.

  Labels, however, tend to prefer an “active” deal. For example, since all the majors have affiliated publishing divisions, they solicit interest in the artist from them and the publishers, then make a separate offer to the songwriters alongside the record deal. If the artist resists and wants to retain their publishing rights, the label will accept a passive participation in the publishing income and counter this less lucrative (for them) deal by offering lower rates on “mechanical” income as a tool to drive the artist to agree to the deal with their affiliated publisher.

  Mechanical licenses, a mandatory requirement for any record deal, grant the right to mechanically reproduce a recording. This generally runs as high as 9.1 cents per song. The writer of a song, who is not necessarily the performer, receives that 9.1 cents per track for songs under five minutes (in addition to the royalties, if they are also the recording artist). Typically these songwriting cuts are negotiated—in favor of the record company—to be limited to ten tracks. Even if there are twelve or more songs on a CD, they agree to pay only on ten of them. Skits—like the short dramatic or comedic interludes on hip-hop records—don’t count. If the artist writes his own songs, he has the option to negotiate this mechanical percentage, and often as a result it gets lowered to 7.1 cents. Mechanicals are an important source of income, as I’ll discuss later.

  Back to the active deal. If the company can manage to own a portion of the publishing of a song, then the label stands to reap additional income if that song is licensed by an ad or covered by another act. Typically the label will try to get a 10 percent passive participation in the publishing income, if they’re not able to get an affiliated (active) publishing deal, which will therefore mean that they also get a piece of the mechanicals they are paying to the songwriter.

  Touring is of course a big topic in these all-inclusive deals. Normally, touring falls under “passive” participation. The label doesn’t actively promote the tours or help organize them; it’s too much work. They just take a piece of the profit, although some labels try to exert more control and actually make deals directly with concert promoters. The tour participation in 360° deals are all over the place, ranging from those that take 5 to 15 percent of gross tour income to even higher amounts of net income. “Shelters” are often built into the deals so that the label only starts to participate above a certain net-income threshold. If, for example, you aren’t netting more than a certain amount on your tour—if you’re only playing clubs, for example—the label won’t be terribly interested and won’t commission your income. This benign neglect can be contractually formalized, and the artist is therefore “sheltered” somewhat.

  Understandably, labels don’t want to be stuck with the “short” dollars (meaning the debt or losses that the artist might incur while on tour), which will only increase the artist’s tour-support demands. To get their act into the venues that will pull in serious money, and therefore be worth commissioning, the labels that sign artists to 360° deals aim for blockbuster hits—just like movie studios. If a song is a hit, then the performance venues tend to increase in size, and the act will actually make money on the tour—which will then be commissioned by the label.

  Labels offering 360° deals also like to participate in sponsorship and endorsement deals, whether tour-related or not. Sometimes those deals can be limited to ones brought to the table by the label, but often not. Again, the company’s commission on these deals ranges from 15 to 20 percent of net income. Labels are bolstering their staffs in this area, since they believe relationships with advertisers and corporate sponsors will be key to future profitability. Needless to say, this means artists signed to these deals will be pressured into associating themselves with sponsors and the products they’re selling. The line between music as a creative act and music as a means of getting you to buy something will become even fuzzier. As more artists sign these deals, we will have a hard time knowing whether or not we are listening to a song or a commercial—or whether there is any difference between the two.

  2. STANDARD ROYALTY DEAL

  This is more or less what I lived with for many years as a member of Talking Heads, and even as recently as 2004, when I released Grown Backwards with Nonesuch. In this model, the record company bankrolls the recording and handles the manufacturing, distribution, press, and promotion. The artist gets a percentage of record sales. The label does not a have financial stake in live shows, T-shirts, or endorsements.

  In a typical deal of this type, the record label owns the copyright to the recording. Forever. This doesn’t mean they own the “song,” though. This distinction is confusing for many people—we tend to think of a recording as being the same as the song. However, the song itself and the version that the artist recorded are not always the same thing. It could be someone else’s song, for example. In that case, the song copyright is shared by the writer(s) and the publisher(s) of the song. This goes back to the era before recordings, when sheet music was the published version of a “song.” With the advent of the recording industry, sheet music now brings in a minuscule amount of income, but the recording of a song—particularly one that becomes a hit—is a valuable commodity. Since the record company in this model typically finances the recording, they claim 100 percent ownership of it, with an agreed-upon percentage of record sales going to the artist.

  Obviously the cost of all the services a record company provides, along with their overhead, accounts for a big part of the price of a CD. You, the buyer, are paying for all those trucks, all those CD-pressing plants, all those warehouses, and all that plastic. Only a small percent of the retail price is for the music. Theoretically, as digital distribution increases and much of that overhead goes away, those costs should no longer be passed along to the consumer—or to the arti
st. Theoretically!

  Much of the income for songwriters like myself doesn’t come from record sales, though we do get the mechanical and publishing income before expenses like video budgets, recording costs, and tour support are repaid. However, I’m going to focus on sales of recordings for now, because the other sources of income—like touring and licensing songs to films or commercials—are optional. I could make a lot more money if I decided to license songs to commercials. Here is the breakdown of how I did on a record that was made under a more or less standard distribution deal.

  Talking Heads spent many years with Warner Brothers, and in 2004 I released Grown Backwards through their boutique Nonesuch label. Part of the attraction of Nonesuch was that we felt in good company with their eclectic roster of acts, like John Adams, the Black Keys, Laurie Anderson, Caetano Veloso, Wilco (until they left to start their own label, dBpm Records), Buena Vista Social Club, and the Magnetic Fields. Like the Warner Brothers of old and indie labels today like Warp, 4AD, Tomlab, Daptone, and Thrill Jockey, Nonesuch’s taste is reflected in their roster of artists. If you like one record on the label, you just might like another.

  Trusting that I’d sell some records, Nonesuch offered me an advance of $225,000. If I had recorded with just myself and a few other musicians, my expenses would have been lower and I could have pocketed the leftover money. Instead, after my recording studio and musician costs, all that was leftover was about $7,000. Was I crazy? That’s not much to live on for all the time that writing and recording takes, which in this case amounted to almost a year, though not of continuous work. That record did cost a significant amount to make, because I mixed a rhythm section with strings, winds, and horns on a lot of the tracks. There were lots of arrangements, players, and big studios for recording them. When you do a record like that with a corporate-owned label, you have to pay at least union rates for musicians. Though high, those rates are generally fair. In this case, recording costs (all the musicians, studio time, technicians, and arrangements) were $218,000, which seems like a sizable sum. I was glad Nonesuch was covering those expenses with their advance, but still, what was I to live on? Was I being foolish and naïve?

  Presumably they didn’t give me that advance knowing or even caring what the record would cost. Rather, for them the amount was based on projected sales. Needless to say, this “loan” would need to be paid off—it wasn’t a gift for signing with them! I would begin to see income from my record only after that sizable sum was repaid.

  There are two ways of handling a royalty deal, but they both come out more or less the same for the artist. In one form of accounting, the artist gets their percentage only after a lot of others get to the feed trough first. The other standard model involves the retailer and record company taking a lump sum off the top, with the artist receiving a fixed royalty on what’s left. I’m going to focus on the first form of accounting, since it’s more transparent.

  A big chunk of the price a consumer pays goes to the retailer—either the physical store (those that are left) or to iTunes or Amazon. Then the record’s producer gets some percentage (3 percent is common). Any tour support the record company advanced to the artist gets paid back, as do video costs, which can be as high as the cost of making a record and are often higher, like a million dollars for a really big-budget video. Then promotional costs are shared, including payola—which is essentially bribes of one form or another to radio stations. So half of what the record company pays to get your record played and marketed is your own money—you just don’t have to front it. (To be fair, they usually ask at each step along the way.) It goes on and on. Returns (meaning the records that are pressed and shipped out but remain unsold and need to be returned), limos, those dinners they bought you that you thought were such a nice gesture—this all gets deducted before the artist’s percentage kicks in. A lot of accounting work is required if an artist decides he or she wants to actually investigate where all that money went.

  PHYSICAL ROYALTY BREAKDOWN

  Wired Magazine

  Here are the figures that show the cost of making my Grown Backwards record versus the advance I got from Nonesuch:

  EXPENSE BREAKDOWN FOR GROWN BACK WARDS ($225,000 ADVANCE)

  RZO Music Ltd.

  This was an expensive, and therefore risky record to make in the current economic climate. The record business was heading down a slippery slope, so betting that I’d make that hefty advance back was in no way a sure thing. I felt lucky just to be able to make a record with strings, winds, and the host of great musicians I loved, and I was prepared not to reap much in the way of profits as a result. We all knew that records were costing more, but relative to what my record sales were at that time, this was really pushing it a bit. I’ve talked recently to some emerging musicians who are still watching the industry tank, and when I asked them why they even wanted to make a record, their feeling was, “I want to do it while they still exist.” I may have been operating under a similar impulse: Let me sneak under the wire before the whole game is over.

  How many copies of that record would I have to sell to actually make money? The record retailed for $18 (before inevitable discounts, but for the sake of this exercise let’s use the full price). Eight dollars of that goes straight to the retail store selling physical CDs, which leaves $10. If my royalty percentage were a not-uncommon 14 percent, I’d get $1.40 from that wholesale price. If my royalty rate were a fairly high 19 percent I’d get $1.90 per CD sold. If I had a bigname producer on the record then I’d typically be obliged to give 3 percent to him or her, because producers get paid off the top—not after I go into profit. Really big producers get advances, too. T Bone Burnett often gets an advance in the six figures, which comes off the artist’s royalty share, though many artists would say that he is worth the price.

  Some producers also demand a share of the publishing, essentially claiming they co-wrote the songs. This might in fact be the case at times: the beats and sounds they contribute are often so integral to the success of the song that they could legitimately be classified as composition. There’s no law that says beats are composition, though, so this demand is completely negotiable.2a In a typical situation, I might be paying off my recording debt (and tour support money, if I took any) with that $1.40 per record, minus the 30 cents the producer would be getting from the very first record sold. So I’m left with $1.10 per record sold.

  Foreign sales work a little differently. My royalty in this deal was 75 percent of the U.S. dealer price in Europe and 50 percent in the rest of the world. So even if I were hugely popular in Japan, it would take me twice as long to recoup my recording costs using those record sales as it would in the United States. This seems completely arbitrary and unfair to me, especially when downloads are increasingly the way folks buy music. But this is the standard offer on such deals.

  If I wrote every song on a CD (which I didn’t on Grown Backwards), I would get, from the mechanical income mentioned above, 91 cents (9.1 cents per song multiplied by the ten songs the labels usually limit these mechanicals to), in addition to the sales royalty percentage and my publishing income. If my sales royalty was 14 percent of the $10 dealer price and I wrote all the songs, then I would get $1.40 + 91¢ = $2.31 per record sold. Things are looking a little better.

  I believe it was the Beatles and other singer-songwriters of the sixties who realized that recording your own songs was far more lucrative than doing record after record covering other people’s songs, as had often been the norm in pop music. This incentivized songwriting, and it was partly due to this insight that there was suddenly an explosion of creativity and innovation in pop music in the sixties. But it also made a few too many musicians feel more or less obliged to consider themselves songwriters. I’m as guilty as many others in feeling that I, or my bandmates, “had” to write every last song on a record, even though covering an underappreciated gem might have been a better choice than recording one of our not-so-stellar writing efforts. However, even notso-good songs
generate income from album sales, as long as there are a couple of hits on there that motivate folks to buy the whole album. The “filler” goes along for the ride and still generates money for the artists and publishers.

  In the case of my Nonesuch record, the math is pretty simple: if I had written all the songs on that record, then I would have had to sell almost 100,000 records to break even on that $225,000 advance. (But remember, if I had recorded that album for a fraction of the actual cost, I would essentially have had that advance as income.) That might not have seemed a huge number of records to sell back in the day for a popular act, but it’s far more than most records sell now. Few acts sell millions of copies anymore, and the artists who do tend to have more debts to pay off than just the recording costs—massive promotional budgets, percentages to managers and video producers. Say a top pop-act does a video that costs half a million dollars (which is not unusual). They’ve then got to sell way more than my 100,000 records to break even; they’ve got to sell more like 750,000. Not all do, of course, and their debts begin to mount up quickly.

  Here are some sobering facts from SoundScan via Billboard: only thirtyfive albums released in 2006 sold more than one million copies within the calendar year; twenty-seven in 2007; twenty-two in 2008; twelve in 2009; and ten in 2010. Only 2,050 of the 97,751 albums released in 2009, or 2.1 percent, sold over 5,000 copies. That sort of puts a different spin on the dream of living large off of record sales. If that hypothetical record with its expensive video isn’t successful, the artist is suddenly about a million dollars in the hole. The pressure to have a hit with the next record is then immense.

  What about downloads? Aren’t they picking up some of the slack as CD sales dwindle? Nope. Typically an album downloads for $10, and Apple’s iTunes Store, for example, takes 30 percent of every sale. The record label applies the artist’s royalty percentage to that $10 retail price, so if the artist is getting the traditional 14 percent, he or she is left with $1.40 per album download. So an artist isn’t better off, especially when you think about the way people buy music online—they tend to buy songs, not entire albums. Artists are understandably trying to negotiate better royalty percentages for downloads, arguing that the record companies don’t have the same overhead and expenses and nor do the “stores”—therefore the royalty given to artists should be higher. But there is, of course, a lot of resistance from the record companies.