Publishing as an industry is a bit of an odd beast. Your usual model for an industry is to take some raw materials, combine them in some way with labour to add value, this gives your product which you sell.
Of course there are a lot of industries that are exceptions, and where exceptions arise there is usually some scope for markets to work very poorly, or at least inefficiently. Take the music industry for example. Artists produce music but recorded music is not selling CDs primarily, it is selling the information on them. The producer cannot limit supply and so legislation needs to be passed to protect the industry.
Publishing has a different problem. For example take a look at the process of publishing a novel, admittedly in a slightly out of date way.
Firstly a prospective author will write a book, or close to a complete book. They need to secure a deal with a publisher and most will go through a literary agent. The agent performs two key functions. Firstly they represent the prospective author to publishers and negotiate to get the author the best deal that they can, a very traditional agent function.
The second function is to screen prospective books. Manuscripts submitted to agents will vary enormously from the unreadable to those ready to publish and be commercially viable; the vast majority of submissions are rejected (reported figures are that about 0.2% of subissions go further). There is huge value to the reader in knowing that all books that reach the shelves have passed this screening process. Of course not all will be classics nor every book be to the taste of every reader but this screening process adds immense value to the output of the industry as a whole.
The interesting point here is to consider where the added value of labour resides. Eliminating all the books that are not ready to be published is a lengthy and difficult process, it is also not unskilled labour. In a “normal” industry the investment of labour is into the end product; in this case the investment is as much into what is discarded as what is passed through. The end result of this stage of publishing is that a book recommended by an agent is more valuable than it was beforehand even though it has not undergone any change. This endorsement changes the nature of the book.
This raises the question of who gains the benefit? Both the publisher and the author gain; the author because their manuscript is seen as higher quality, the publisher because they are passed a higher quality script.
Given the benefits to the author of having their book represented it is natural to ask why does the author simply not pay the agent to endorse the book? This type of bribery looks like it should be common. An agent, even for a small company is exposed to a tragedy of the commons type dilemma when faced with an offer of a payment to recommend a book (I assume here that this is a modest payment to pass a book from not quite good enough to good enough so there is still a good chance of the publisher accepting the manuscript). The agent can accept the payment: they benefit directly from this; the company for which they work may take a marginal hit to its reputation but the cost of the lower reputation is borne by all staff. Given this structure we would expect payments to become the norm to grease the wheels and that agents reputations would fall so low as to stop any but the smallest companies from working in the marketplace. It may not be entirely coincidence that most literary agents tend to work for very small companies of under twenty people (although factors like constant returns to scale, low capital requirements and the ability to provide a differentiated product may be more significant here). Why have agents not just become shills?
This is a fairly similar situation where corruption has essentially occurred. The value of approval from an agent or DJ is greater than the cost of the payment needed to secure their approval. The DJ by accepting and acting upon a payment lowers the quality of their output to the detriment of their employer and consumers. I find it remarkable that one industry has a reputation for corruption and one has not.
Of course there are differences; in music the value of bumping a record up the playlists may be much more valuable and warrant sufficiently tempting payments. Maybe in publishing the prospective bribes are just too small. Maybe also in publishing an agent’s individual reputation is also very important and protecting this by only recommending the highest quality work is paramount.