Notes from the underground: what the ailing record industry can learn from a successful subway musician.

Every two weeks or so, I pack up my Taylor acoustic guitar, fill my backpack with CDs of my music, and head down into the New York City subways to busk away. I make good money, and I get to watch and study people, too. For example, I can now tell from about 50 feet away whether a woman is likely to give me money.

If she's walking fast, wearing headphones, angrily porting a briefcase, or chasing down one of her children, that's an easy no. She wouldn't throw a dime into Jimi Hendrix's case. Other women, who are more aware of their surroundings, have greater possibility. Usually it boils down to makeup and midriffs. If the woman is decked out, she may look at me, but only to see if I'm looking at her. But if a woman is dressed casually, walking slowly, and thinking about something beside herself, she's likely to listen for at least a few moments, and then I have a decent chance she'll enjoy the music, stop, and maybe buy an album.

This is but one of the lessons I've learned from performing in train stations that I think could be helpful to the floundering music industry, or at least to the many talented musicians stifled by it. These lessons haven't gotten me rich, but I've sold about 500 records in the subways playing sporadically since releasing my new album in January. I make more money down there per hour than I do as a journalist. And while my sales and profits have gone up, the music industry's have gone down. Sales of recorded music in the United States have dropped by more than a 100 million units in the past two years, and, after decades of steady gains, industry revenues have dropped 15 percent over the last three years.

Different experts will give different reasons for the decline. The music industry itself blames its customers, or more specifically young people who download music for free from increasingly popular file-sharing networks. Others, such as Josh Bernoff of Forrester Research, blame competition from video games and other entertainment. Whatever the reason, it's clear that the music industry's old model of doing business isn't working so well in today's market. That old model relied on labels plucking out a handful of bands they believed would sell big, and investing millions of dollars in production, promotion, and marketing to get them the time on the radio dial or the space in the record stores they required to catch fire. The industry defended itself against complaints by saying they were simply meeting the demands of popular taste.

In truth, there was always a tautological element to this argument. The music industry functions like a cartel, and the public's preferences have always been limited by the choices they were given. Now that the market for music has changed, and CD sales are declining, the record industry is hiring lawyers and lobbyists to squelch the new technologies that are changing the music business. Over the last few months, the Recording Industry Association of America has issued hundreds of subpoenas to college kids who swap music over the Internet. Meanwhile, the industry's lobbyists have convinced Sen. Orrin Hatch (R-Utah), himself a songwriter, to float the idea of allowing companies to remotely slag computers whose owners used them to download movies mad songs.

But the industry's efforts are counterproductive. About 60 million people in the United States have already swapped copyrighted material

over the Internet, and that number isn't likely to shrink. The times are a changin', and record companies should learn to how to profit in this new environment. With all the modesty required of a guy who doesn't make enough money on a given night to buy front-row seats at a Mariah Carey concert, let me offer a few pointers.

Number One: Drop the price of CDs

When I first started playing in the subways, I experimented with different prices for my albums: $2, $5, $8, $10. I sold slightly more CDs at $2, but far fewer at $8 or $10. The sweet spot seemed to be a price of $5. Later, I gave up pricing my CDs altogether when I got a ticket for selling CDs in the NYC subways. Now I post a sign saying that my CDs are technically free, but they cost something to make, and people should pay what they want. Occasionally someone will take one for free. Sometimes a passerby will drop $20 into my guitar case. But the mean, median, and mode are all, again, $5. This could merely be the measure of what my music is worth. But my strong sense is that five dollars is what people will pay for a CD they like by a musician that they've never heard of.

So why does the average CD sell for more than $17? It's not the manufacturing cost. My last album cost me about $1.10 for each CD manufactured. I had high quality work done at every step, from the mastering to the packaging. People who print larger runs have vastly lower costs.

True, recording solo acoustic guitar is vastly easier than a full band, but even industry-produced acoustic guitar albums sell for that price. That means the companies are either reaping cartel prices or most of the money is just going to the layers of middlemen: the redundant producers paid by the hour to fine-tune albums, the marketers, and the promotional photo shoots. If you're U2 or the Rolling Stones and sell millions of albums, such expenses may be perfectly worthwhile. But drooping industry sales figures and my own experience suggest that for the vast majority of artists, the industry is going to have to drop the price of its CDs--maybe not to $5, but certainly closer to that than to $17.

Number Two: Branch out

The beauty of the subway system is that it's about as free a market as there is. If I play good songs that I'm passionate about and that I have down cold, I make money. If I play junk, I don't make a dime. Naturally, though, I've had to learn a few things about how to place myself. The New York City subway has two good places for musicians to perform: the platforms where the trains stop, and the hallways leading between platforms and up to street level. The advantage of a hallway is that everyone passing by hears you for a few seconds; in a platform, you get far fewer people, but they hear you longer.

Hallways, it turns out, work great for playing music that's instantly familiar. In my favorite station there's frequently a talented hallway musician who plays Beatles songs, the kind of music that in three notes can jar a pleasant memory for a huge number of people. The instrumental guitar music I play, on the other hand, is a little unusual. To people who know the genre, it sounds like a poor man's version of Leo Kottke or Michael Hedges. To people who don't know the genre, it's a poor man's version of Duane Allman's Little Martha or Jimmy Page's Bron-y-aur. In any event, it's fairly complicated, and I literally don't make a penny when I try to play in the spot where the Beatles troubadour sings. But on a platform where I have roughly three minutes between trains coming, I can get folks' attention long enough to make some sales.

One might think that my only audience would be the Birkenstock nature-lovers and 14-year-old kids porting around their first guitars--and I do do well with that crowd. But I also do well with middle-aged black couples, 40-year-old white couples with kids, white blue-collar workers, and the Ecuadorian immigrants who sell jewelry in my favorite station. In fact, I have a much better chance of telling whether someone will like the music based on the way that they walk than based on their age, sex, or apparent income.

The music industry tends to divide both bands and audiences into broad, set formats: alt-music, hip-hop, and modern country. There is an obvious reality to these categories, but in truth, they exist largely for the benefit of record companies, which can then narrow and target their promotion efforts. Unfortunately, most bands and artists can't get to first base unless their music fits one of these formats, and there are many other bands and other types of music-like mine-that don't fit into any set genres. Many people's tastes stretch well beyond formats, and might they want to buy some of this music if they heard it. Indeed, it's almost guaranteed that somewhere between these formats, the next big thing in music is brewing. But figuring out how to profitably micro-market heterogeneous bands to scattered audiences is something the music industry has not yet figured out how to do.

Number Three: Embrace file-sharing

Fortunately, the Internet allows a wide audience to inexpensively sample a huge array of music. File-sharing networks like Kazaa, and artists who allow free downloads off their Web pages, are roughly like playing in the subway. The Net allows artists access to a substantial potential audience at almost no marginal cost, while providing listeners with short samples of a wide variety of artists and musical styles they may not hear on the radio with a low investment of time and almost no investment of money.

As I'm writing this, my computer is downloading songs by Nigerian afro-pop innovator Fela Kuti in anticipation of a trip to a museum show about him. Last night, I tried to download music from several artists currently performing in New York to decide if I wanted to see any of them live. Unfortunately, I couldn't find free downloads of most of their music, so I watched a movie instead. I was happy, though, when I saw that someone had downloaded a song off my album.

That's a risk, but it's one worth taking. I profit tremendously when people download my songs. It makes them more likely to pay to come to my concerts. It helps raise my profile and makes it more likely that people will call radio stations to request my songs--which could one day be a source of album sales and my ultimate transition from a Washington Monthly contributing editor into a major music icon.

Big artists do indeed lose with file sharing, and it's their profits on which the industry depends for survival. That's why they're fighting it so hard. But it's a fight they will eventually lose, and that won't be a bad thing either for bands or fans. Eventually much as the movie industry eventually figured out how to profit from the VCR, record companies should look harder to find ways to make money off of the Internet--probably once they learn to be more nimble and pay more attention to the particular tastes of a diverse audience. Meanwhile, file sharing helps small artists with limited distribution find their audience and make a decent living--a truer expression of the free market in music. If everyone could download tracks from groups playing in their city that weekend, the best band would draw the most fans and a lot of unknown bands would surge. That may scare big groups who have become popular under the current system, but it shouldn't scare fans or scare the industry. Record companies will just have to get better at serving their customers.

Nicholas Thompson is a Washington Monthly contributing editor. His latest CD, Lend Me Your Ears, can be purchased at nickthompson.com.