The Lost Tail

The myth of the Long Tail is possibly fading away as the digital book market grows, and it is operated by few mega e-retailers.


In a limitless world of digital goods, powerful search engines, near-zero marginal cost of digital production, storage and distribution, niche products shall get much more market relevance. ‘Selling less of more’ is part of what the ‘Long Tail’ theory has been preaching.

Does it apply to the creative industries to? And how? Should digital book publishers reduce attention on blockbusters and increase focus on the Long Tail as the source of the most profitable growth? Is there a space for unlimited growth of niche e-books? Who is going to consume a potentially unlimited supply of creative goods?

It is interesting to notice that the Long Tail theory was first published 10 years ago (October 2004), a few years after the dot-com bubble, when Internet was still at its infancy (it was 11 years old then). Amazon had not launched the Kindle yet (it did it only at the end of 2007) and the e-book market was still awaiting the ignition. The digital music had just started, as Apple launched its iTunes Store only in April 2003 and that was the single most important booster to the digital music market in the years thereafter. When the Long Tail theory first appeared, detailed sales data regarding the digital music in US were not available yet. Only in 2005 Nielsen Soundscan made first sale data available and only at end of that year Billboard started to take into account paid downloads in the music charts in US.

Historically, the Long Tail theory was born before the economy of digital creative goods exploded (digital music, digital books, digital movies, and so on). Internet offered at that time the first opportunity in the history to distribute an ‘unlimited’ inventory of physical goods. Unlimited inventory, of course, in term of virtual shelf space from a retailer’s perspective. It clearly was not meant as endless inventory from a production standpoint, as marginal production costs of physical goods obviously are not zero and therefore the supply is limited by financial and economic opportunities. The Long Tail theory deserved its merits for clearly identifying unprecedented opportunities at that time.

Thus, an interesting question arises when the marginal production cost of digital creative goods virtually drops to zero. An ‘unlimited’ supply of digital goods could potentially be unlocked. Is this really happening? And if it is happening right now, will it happen forever in the future? Even if the supply were to exceed demand? That is not possible, of course. Eventually, some economic mechanisms will have to emerge to self-regulate any market of digital creative goods as well.

Actually, by increasing the variety of supply a larger number of niche opportunities can be addressed, nobody contends that. We agree with the Long Tail theory. That is for sure. Nevertheless, these opportunities are not ‘unlimited’ and might at very best exist only if supply does not approach or exceed demand. Excess of supply cannot be consumed. However, there are other economic factors that heavily affect the book consumption. More than plain money, the available reading time is single most relevant scarce resource that affects the book consumption. Even the wealthiest person of the world cannot consume all available books, even though he/she own them all. Simply for a question of time, not money. Book reading is not like music listening. When you read a book, you invest your time and forgo many other opportunities for other activities. On the contrary, listening to music does not strictly require you to stop doing many things (you can drive, do housecleaning, dance, work, even read or write a book). Therefore, most individuals potentially have more time that is available to music consumption than for book consumption. It cannot be overstressed the fact that even if the supply of digital books could potentially be huge or ‘unlimited’’, the potential for the demand is not, because it is constrained by key scarce resources, such as time and people among others. Even giving away all world’s books will not make the demand ‘unlimited’’. From a macroeconomic perspective, price matters less than number of interested readers and their available time for reading. In fact, the price can be adjusted for any given market. The reading time cannot. In average it takes the same time to read a book, no matter what the price is.

It’s also important to clarify that the ‘unlimited’ supply of creative goods, such as digital books, is meaningful only in the strict sense of unlimited copies of the same title, as marginal production, storage and distribution cost is near-zero. In fact, the cost of creation of any new title typically is nowhere near-zero. The variety of digital books (i.e. different titles and not different copies of the same title) come at cost. Such cost depends on many elements (competences, talent, forgone opportunities, expenses, etc.) that very often are fundamentally unrelated to the internet and the digital technologies. Save, of course for the digital books ‘created’ by artificial intelligence programs or mash-ups of previously created, contents. The marginal cost of creating a new title in most cases matters. Therefore, such ‘unlimited’ supply of creative goods does not mean unlimited variety of digital books. For the sake of creating a new content for a book, the author must still find a compelling reason and the proper resources. The Long Tail has never zeroed in the implication of creativity costs. It assumed the goods exists, as a starting point.

Incidentally, because of the near-zero marginal cost of storage and distribution, the variety of digital books available in the market is bound to grow. In fact, once a digital book gets published, it can potentially stay forever in the market, as the marginal costs of staying ‘published’ are virtually zero. Over time, new releases of digital books might encounter more competition from the growing catalogue of previously published books. This is something new in the book publishing industry, as most books used to stay on the market until demand was high enough to make it worthwhile keeping it available.

One decade after the Long Tail theory was originally conceived, the world wide web has left its infancy and has entered its adult stage. It is 21 years old now. It is about time to do a reality check and challenge such sound, fascinating and very popular theory that has not been proved true yet from a scientific standpoint (at least in the digital book business). Even if a theory makes sense and some empirical evidences confirm it, it does not mean it holds true. We have believed for thousands of years, that the Earth was flat or the Sun gravitated around the Earth, the gravity law was ‘universal’’ and many others sound theories that passed many evidence tests over the past centuries until they were proved wrong.

In the white paper, The Lost Tail, we will show how we applied a novel econometric methodology – devised on purpose – to start tackling this challenge. The preliminary results, based on actual sales in Italy of several thousands of titles over the past four years, show that the Tail is getting less relevant (not more) as the digital market grows. The head has been growing quicker than the tail, even though both head and tail have been growing significantly.

As shown in the paper, our findings are very consistent with those of Harvard Professor Anita Elberse, which were published in October 2013 in her book Blockbuster: Hit-making, risk-taking and the big business of entertainment.’ It is worthwhile mentioning that less than 1% of the digital songs in US makes more than 80% of sales (source the above mentioned book by Anita Elberse). This seems to exclude the current observability and relevance of Long Tail theory in the market of the digital songs.

Interestingly our findings also are highly correlated with the growing concentration of the e-retail market. As the market share of the global e-retailers grows, the sales impact of small and independent retailers diminishes dramatically. A highly concentrated e-retail market appears to be less capable to foster and grow a long tail of digital books than it is a not concentrated e-retail market. In fact, the overall e-book market has been growing very significantly, but the bestsellers have been taking a growing lion’s share. Of course, correlation does not mean causation. One cannot easily jump to the conclusion that the e-retail market concentration affects the the Long Tail. This could be, though, an opportunity for future researches.

There are not final answers yet. This paper is more a methodology study to advance our fact-based knowledge. Any publisher or distributor with a large number of titles can easily replicate the proposed methodology with its own data and get some insights. Of course, not all markets nor all publishers are the same. There might very well be different, even opposite, results.

Final remark: a theory that sometimes work and sometimes does not, might need either a narrower scope or to be fixed/updated… You will be able to judge if there are reasons to be skeptical about this long-standing and very popular theory, at least as regards as digital creative goods, such as e-books or digitial songs.

‘Science must begin with myths, and with the criticism of myths.’ –Karl Popper

Here the link to download the white paper on The Lost Tail: http://www.allbrain.org/lost-tail/

Leave a Reply

Your email address will not be published. Required fields are marked *

I have read the Privacy Information Notice and I give my consent (to my data processing)

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>