WSJ Law Blog and the WSJ report that Universal Music has now settled with the NY AG’s office for $12 million as a result of Spitzer’s continued attack on what he describes as “corrupt practices” in the music industry. (HT: Bill) The settlement also requires Universal, like Sony BMG and EMG before it, to cease and desist all payola-related activities. The WSJ story reports that EMG, the final major industry player, should settle within the month, making final Spitzer’s payola ban.
In the “Assurance of Discontinuance,” Spitzer explains the mechanism through which it is alleged that intense competition for radio airplay harms consumers:
“Intense competition among record labels for the relatively small number of valuable play list slots has caused a variety of aggressive pay for play mechanisms to emerge. All labels share the common objective of advancing the circulation of record labels’ products to the listening public, without regard to the artistic value of those products. In each case, music consumers remain unaware of the extent to which radio programming and record popularity statistics are being manipulated and compromised . . .. By engaging in such an elaborate scheme to purchase airplay, increase spins, and manipulate the charts, [Universal Music] and the other record labels present the public with a skewed picture of the country’s “best” and “most popular” recorded music.”
So, let me see if I understand the mechanics driving the consumer harm, here?
(1) Radio airplay is a scarce promotional input;
(2) Record labels pay radio stations for this scarce promotional input with various forms of compensation ranging from cash to vacations;
(3) These practices harm consumers, who otherwise believe that radio stations are playing the country’s “best” and “most popular” recorded music.
I have been fairly critical of attempts to ban payola (see, e.g., here, here and here) on several different grounds. Now seems like an appropriate time to reiterate two of those objections in greater detail below.
1. The investigation demonstrates little to zero interest in the actual underlying economics of payola, and most importantly, its impact on the consumers the settlements purport to protect.
The fact that record labels pay radio stations for a valuable promotional input that increases highly profitable record sales should be about as unobjectionable as any other form of advertising. Spitzer apparently concedes that these payments are part of the competitive process (a point I have made previously). It is difficult to reconcile this concession with the absolute ban imposed on these contracts.
Benjamin Klein and I have written extensively about the economics underlying contracts in which the manufacturer purchases promotional inputs in the grocery store context, i.e. slotting contracts in which the manufacturer purchases retailer shelf space. We demonstrate that slotting contracts, which resemble payola contracts in large part, are a consequence of the normal competitive process when the promotional input primarily promotes incremental manufacturer sales rather than shifts sales between retailers. Despite the obvious differences between airplay and shelf space, the economic forces underlying the record label payment for promotional input are analogous and illustrate some important points regarding the importance of competitive process for distribution in many markets.
If Spitzer understands that payola is a part of the competitive process, perhaps he would not be surprised that the history of the music industry is replete with failed attempts to collude by enforcing agreements to cease payola. Luckily, collusive pricing agreements are notoriously difficult to enforce. As I pointed out previously:
One need only read Coase’s seminal account of the many failed (and documented) attempts at collusion by music publishers in 1890, 1916-17, 1933, 1960, and 1986, to garner some appreciation for the difficulty of the task. Importantly, each of these attempts was initiated by the music industry. Many of these collusive endeavors failed because cartel members could not credibly commit to reducing output.
I also noted that collusion is made substantially easier when a third party can enforce the agreement by punishing those who would deviate, and that an ideal third party enforcer candidate would be the AG’s office:
Successful collusion often takes a third party to regulate the agreement and punish defectors. Occasionally, would-be cartel members are able to persuade the government to take the job. It appears that Spitzer may succeed where the recording industry has failed for over a century by stepping up to police the industry restriction on competitive payments for spins.
Consistent with over a century of history suggesting that the music industry would like nothing more than put an end to payola, most reports suggest that that labels have been more than happy to cooperate with the investigation, even agreeing to cease and desist competitive conduct outside the scope of the payola regulations.
In the meantime, there is no evidence that payola contracts have resulted in any tangible harm to consumers in terms of higher prices or lower quality music. It is this latter proposition, that consumers are harmed because payola results in worse music on the air that is at the heart of Spitzer’s attack and many commonly voiced objections to payola. But this comparison necessarily requires knowledge of the counterfactual: what would music playlists look like under a payola ban?
Coase pointed out in his seminal payola analysis that the assumption that a music meritocracy would prevail in the absence of payola was naive. As I wrote previously:
An ancillary point of Coaseâ€™s analysis was that a payola ban â€œmay result in worse record programsâ€? because station song selection will depend solely on maximizing advertising revenues. I am not confident that I know what the mix of songs that attracts the best demographic audience for radio commercials looks like. Nor am I confident that I, or any regulator for that matter, could figure out whether such a mix of songs would be better or worse than what we have now.
Without some analysis of what playlists would look like in this counterfactual world, the crusade against payola is a classic example of the regulators falling victim to the Nirvana Fallacy. I should note here that my empirical analysis of shelf space contracts, Slotting Contracts and Consumer Welfare, attempts exactly this type of analysis (finding that while slotting contracts impact the mix and shelf allocation of products, they are not associated with any tangible consumer harm in terms of prices, quality or variety).
So, what exactly do consumers get for Spitzer’s troubles? It is not clear that playlists will change in a good way (whatever that means). On the other hand, it is likely that the reduction in competition between record labels will harm consumers.
2. On its own terms, the “deception” theory of consumer harm supports stronger enforcement of payola disclosures, not a ban.
One might argue that the true harm to consumers from payola is not a change in prices or playlists, but that consumers are deceived because the payments are not disclosed. In other words, consumers’ expectations regarding their music selection assume some sort of unbiased and unfettered process which is upset when labels make pay for play arrangements without disclosure. I am not convinced that this argument justifies the payola settlements.
I am skeptical that consumers’ expectations are so naive, given the long, colorful, and well documented history of payola arrangements in the United States. But taking the argument on its own terms, that theory of harm supports more vigorous enforcement of the disclosure regulations in the payola statute, NOT a ban on payola contracts.
Perhaps the largest consumer welfare costs associated with these settlements derives from their underlying theory that undisclosed payments are always bad for consumers. The theory has apparently become a popular one in Spitzer’s office (see, e.g., the recent settlement with Liberty Mutual regarding insurance steering fees, payments from insurers to producers to promote products to ultimate consumers).
I should also note, in full disclosure, that I am currently working on a number of reseach projects regarding the economics and empirical nature and consequences of undisclosed payments such as payola, insurance steering, and slotting contracts (including projects testing some of the ideas expressed here). I will look forward to blogging about some of these in the near future.