Prof. Stan Liebowitz, University of Texas at Dallas
Sept. 1, 2009

Two pieces of copyright legislation have been floating around Congress during the last two years that are diametrically opposed to each other.  One is the “Performance Rights Act” and the other is the obscurely named “Local Radio Freedom Act.”

The purpose of the first proposed act (supported by a coalition of artists and record companies, known as MusicFirst) is to require radio stations in the United States to pay royalty fees to the artists and record companies when their sound recordings are broadcast over radio.  The purpose of the second proposed act, supported by the National Association of Broadcasters, is to keep radio stations from having to make such payments.

The artists and record companies argue that the United States is the only major country that doesn’t require radio stations to pay artists/record companies for the rights to play sound recordings (radio stations pay composers and publishers when musical compositions are broadcast, with the payments going to performing rights societies such as ASCAP and BMI, but nothing for the musical performance).  Further, American artists and record companies would normally get a share of foreign radio station play of American records, but since the United States has no such domestic right, foreign performing rights societies are not required to pay American performers and record companies.

Radio stations, on the other hand, argue that they provide free publicity to the sound recording companies and therefore they should not need to pay for the right to broadcast sound recordings.  They point to the fact that record companies want to have their records played on radio, sometimes taking actions that are illegal to get their records played on radio, such as engaging in payola (which is defined as record companies surreptitiously paying radio stations to play certain records).

It is very odd to have a dispute where one party is using the product of another as its main ingredient without having to negotiate for the rights to do so.  In most cases, companies could not use copyrighted works belonging to others without the permission of the copyright owner.  For example, book publishers cannot just publish any book they wish without the permission of the copyright owner; neither can television stations broadcast movies or television programs without the permission of the copyright owner.

What is unique in the radio case is that owners of sound recordings have no property right over the (terrestrial) broadcast of their works.  [Note: Internet sites that stream sound recordings are required to pay for their use of the sound recording.]  This appears to be a glaring oversight in the law that might be easily remedied by just creating such a right (which is the suggestion I have made in an academic paper), although there is a potential complication that I will come back to shortly.

If assigning a property right was all that was needed, then the contradictory claims made by the two sides could easily be put to the test.  Radio stations, needing legal access to music before they can broadcast it, might pay hefty sums for the rights to do so.  Alternatively, record companies, if radio broadcasts increase sales, might be willing to pay radio stations to play their recordings (for the market to function properly we would need to make payola legal).

We might well see payments going in both directions as is not uncommon in other markets.  For example, some would-be authors pay to have their books published by vanity publishers.  Yet most authors require being paid to write, and if a publisher expects to make a profit selling a book, it will pay the author.

Radio stations generate revenues selling advertising, and music is the main component of broadcast time, so it is natural to think that radio stations would be willing to pay for the right to broadcast music, just as television broadcasters are willing to pay for the right to broadcast programs and movies.  This favors the position of the MusicFirst coalition.  Nevertheless, radio argues that it increases the sales of sound recordings.  If this were true, then it is possible that record companies might pay radio stations for playing their records, even though radio stations were also generating revenues from advertisers.

This is where a potential fallacy of composition comes into play.  To the best of my knowledge, until recently, there were no studies examining whether the existence of radio increased overall record sales.  Instead, radio’s ability to increase the sale of individual records that had received heavy exposure was noted.  However, radio is perfectly capable of increasing the sales of individual records without at all increasing, or even possibly decreasing, the overall sales of records.  In other words, radio might be capable of influencing the relative popularity of individual songs, increasing sales for songs that are heavily played, but total sales of songs might not be increased by radio.

Take a simple analogy.  Magazines that review products influence sales.  If, for example, Consumer Reports reviewed five small SUVs and concluded that four of them were unacceptable death traps, the sales of the single acceptable SUV would increase relative to other SUVs and most likely in absolute terms as well (although the taint of so many dangerous SUVs might hurt sales even for the SUV with a passing grade).  A positive review in Consumer Reports, then, could be said to increase the market share of SUVs.  Auto companies would want to influence the magazine to write positive reviews and if companies were allowed to influence the ratings, they would be willing to pay the magazine in order to get a better review.  Nevertheless, that does not mean that the existence of reviews in Consumer Reports increases overall SUV sales; to the contrary, in this example, the existence of Consumer Reports reviews would almost certainly reduce overall SUV sales.

Radio play is like a positive review — the station, by playing the song and having the DJ say a few nice words about the song, is giving it a seal of approval as well as allowing the listener to become familiar with it.  However, that does not mean that sales of all sound recordings, which include sound recordings not being played on radio, will increase even though radio does increase the market share of songs being heavily played.  Therefore, the fact that radio play can enhance a particular song’s sales is irrelevant to a question about radio’s overall impact.

I have performed two studies attempting to examine whether or not radio increases the overall sale of sound recordings that appear to be the first such studies ever undertaken.1  The results indicated that radio is more likely to decrease record sales than to increase record sales.  [Contrary to claims made on the NAB website, neither of these studies was commissioned by the RIAA.]

The intuition supporting this result is simple.  Consider, for example, that on average, people spend an hour a day listening to radio in their cars.  If radio were not available, most drivers would want to listen to something.  Would not most of this former radio listening turn into the playing of prerecorded music?  Since, on average, people spend less than an hour a day listening to prerecorded music, this hypothetical case of radio disappearing in automobiles would likely lead to a very large increase in prerecorded music listening and that would almost certainly increase sales of prerecorded music.  This is an example of how radio could decrease the overall sales of sound recordings.

The NAB commissioned a study whose data and methodology were similar to the second of my studies.2  The NAB study concluded that radio play increased overall record sales.  The results were so different between the two studies that I assumed that an error must have been made in one of the studies and offered to exchange data (and details of the analysis) with the author of the study to see whether an error had occurred in one of the studies.  The NAB refused to allow the data exchange, saying that they and he no longer had access to their data.

These dueling studies aside, it would still seem reasonable to provide a property right to copyright owners when radio stations broadcasts their sound recordings and then let the chips fall where they may regarding the flow of payments.  The complication I promised to come back to, however, has to do with the feasibility of copyright owners to negotiate, monitor, and enforce their property rights when there are many potential users of the works.  In order for markets to function, not only do the property rights need to be assigned and well defined, but the cost of transacting needs to be lower than the benefits of trade.  The case at hand, of musical performances, might be an instance where the transaction costs are so great that just providing a property right would not lead to efficient solutions.

A close analogy exists for musical compositions.  Although composers have a right to be paid when their songs are played on the radio, it was thought that private contracting would to be too difficult for individual composers having to keep track of the radio play of their songs by thousands of individual radio stations.  Instead of requiring individual composer to negotiate with each station, a performing rights organization was created to keep track of the broadcasts, negotiate with radio stations, and collect the money on behalf of the composers.

Because of concerns that a collection society might have monopoly power in price negotiations, the prices that are paid for broadcasts of musical compositions in most countries, including the United States and Canada, is overseen by some form of government body or court that has the authority to set the prices.  It is sometimes claimed that technological advances now allow individual composers to negotiate with radio stations and monitor radio usage of their songs, but I am not aware of any country where such negotiations take place.

If property rights to recorded musical performances were granted, it is not clear whether copyright owners would be in a position to monitor the uses of their works, any more than was the case for composers.  Thus, just granting a property right might not solve the problem.  If transactions costs were too high, then some system with a performing rights-type collection organization would need to be in place and a determination would need to be made to assess whether and how much radio would pay to use music if the market could function efficiently.

The current fight is not about introducing a property right so much as it is about introducing a collection society.  The current proposed laws assume that transactions costs are too high for the market to function.

That is where the economic studies on the impact of radio play on sound recording sales become important.  If radio play enhances record sales, the price that radio stations would be expected to pay goes down, perhaps becoming negative, meaning that record companies would pay radio stations.  If radio play decreases record sales, this is less likely to happen.

The dueling laws mentioned at the top of this article will be decided in the political arena.  Hopefully, the discussion will be at least a little informed by analysis and not just politics.  Interestingly, many of these issues arise again in the case of the Google book project.  That discussion, however, will have to wait for another time.