Threat Assessment

by aussi disponible en français

All unite ! Early June, the 36 publishers of the Japan’s Digital Comic Association[1] and a few US publishers[2] announced a coalition to fight the “rampant and growing problem of scanlations”. And it was about time : as Brigid Alverson concluded briefly, “The rise of scan sites like Onemanga and Mangafox has coincided with a dip in manga sales, and no one thinks that’s a coincidence.”. No coincidence either in the timing of the publishers’ reaction : a few days prior, Google had unveiled the list of the top 1000 most-visited websites for the month of April 2010, a list where Onemanga appeared at the 935th place, with dizzying statistics : 4.2 million unique visitors for 1.1 billion pageviews per month. Obviously, the situation is unfolding with irrefutable logic : the market declines, scan sites rise, publishers take notice, publishers react. One can easily imagine what follows : scan sites decline, the market rises back again, and the whole thing reaches a happy ending. Mission accomplished.

Simple, don’t you think ? One even wonders why it has taken them so much time to react. Because, frankly, the US manga market has seen better days. The brains at ICv2 have been very clear : after as strong decline in 2008 (-17 %) 2009 marks another year of draught (-20 %) and sees the market stumble down in two years, going from $210 million to barely $140 million. A third less, and all this because of the scan sites. And indeed, ICv2 is very clear about it, as is written further in the– oh wait. Not exactly. In fact, it’s closer to the opposite :
While comic retailers tell ICv2 that they believe scanlations (translations of scanned manga, which appear on the Web within days of their publication in Japan) are hurting their sales, the evidence is not conclusive. Scanlations were around through the growth of the manga market as well as its decline, and some feel that they actually increase the market for manga collections by creating greater exposure for new properties. While it may be true that more manga buyers are telling retailers that they’re reading online rather than buying, that may be due to economic conditions (they’re buying fewer titles over-all), or to the lack of a major hit that stimulates buying.
Shoot. All of a sudden, things are not so simple. The scanlations are not to blame, but they still play a role, even if there are (maybe) other factors. But still : “the scan sites rise, the market declines”, right ?

Oh, by the way — how much did they rise, the scan sites ? The Google list is great, but there’s only one, for April 2010. Which means it’s a little difficult to draw a trend from it. After some unsuccessful search attempts with comScore and QuantCast, we are left with Alexa, with the obligatory precautions.[3] Alexa, which has Onemanga (proudly) sitting at the 332th rank. Bonus : one can observe the evolution of the traffic rank of the site, as well as other additional indicators (visitors/reach, pageviews, etc.). And indeed, it rises. Erm — wait a second. For sure, the number of visitors sees a solid growth from the launch of the site in April 2007 until early 2009, but after that, things significantly settle down. As for the number of pageviews, it is divided by two between the beginning and the end of the year 2009.
So let sum things up : in 2008, the scan sites rise, the market declines, everything is normal. In 2009, the scan sites decline, and… the market declines again. Oh dear, that wasn’t what was expected. Shoot.

Okay, in order to understand a little what’s really happening, let’s have a look at the sales of Japanese series in the States. As for anywhere else, it’s always difficult to get access to some figures, but Brian Hibbs dives every year in the Bookscan data (covering about 65 % of the market), which provides an interesting starting point. One ponders, compares, makes some calculations. And the conclusion ? Decline. Irrefutable decline. Without a doubt. But careful — decline, before, during, and after 2007.
Decline, because the sales of any series have a tendency to erode over time, losing readers with each new volume, tired with the narrative or distracted by other preoccupations. It is striking to observe that all series show a very similar erosion — be it a mega-hit like Naruto (and its 4.6 million copies sold) or a more modest series like Rosario+Vampire (with only 200,000 copies sold).
And what about the rise of the scan sites ? Frankly, even looking closely, there is no obvious impact on sales. There is no evidence of a “before Onemanga” and a “after Onemanga” that would translate into a significant decline of sales. But Death Note, Bleach, or Full Metal Alchemist remain on the same trend, even though they are regular fixtures of the top popular series on scan sites. Only Naruto seems to indicate some sort of fatigue.

Ah-ha ! Here is proof ! The scan sites rise, and Naruto declines, right ? Yes, but no. In fact, a closer look reveals that the varying fortunes of Naruto have more to do with a rather “original” release schedule than other external factors. Consider this : 12 volumes published over four months between September and December 2007, as part of the “Naruto Nation” campaign, and again in 2009 with 11 volumes between February and April. The main objective was to catch up with the Japanese publication, to then stabilize on a quarterly schedule. For the fans, this amounts to 29 volumes over just 20 months, for a total purchase of about $230. Without surprise, this is way beyond their usual yearly budget, this causes shelving problems at retail, and in the end, more than a few readers are lost along the way, unable to keep up with this onslaught of new releases. Only remain the true fans.
Indeed, it appears that Naruto now functions on cruise control, as sales of the early volumes decline, indicating a stabilized base of readers around 125,000 readers … since volume 14, published in May 2007. The creation and impressive growth of Onemanga (on which Naruto is inevitably ranked #1) has not changed a thing. Of course, now that we are back to 4 releases a year, sales are bound to somewhat decline, as that was already the case in 2008 which had seen “only” six new releases, with sales down 18 % — and this, despite some catching up from the laggards generated by the “Naruto Nation”.

Wait a minute — do you mean that the number of releases could have an impact on sales ? That’s quite interesting. Because, you know, ICv2 does more than just estimate the size of the global market, they also monitor the number of new releases on the market since 2005. Taking 2005 as the reference year, this gives us the graph on the right. And all of a sudden, everything becomes clear : “the number of releases rises, the market rises ; the number of releases declines, the market declines.” Which seems to me a better alternative to “the scan sites rise but not really, the market declines but in reality series are selling normally according to the expected erosion,” which doesn’t sound as convincing.
Of course, now it’s the question of the chicken and the egg. Obviously, between the scan sites (that rise) and the market (that declines), there is not much of a debate,[4] the culprit is well-known. But regarding the number of releases, things become more complicated. Publishers are quick to blame it on the declining market (because of the rise of the scan sites, remember) and that they are only reacting. But less new releases mean less sales, and therefore a declining market. QED. More or less.

Let’s sum up. Without a doubt, the popularity of scanlation sites is quite impressive, or a reason to fear Apocalypse if you are a publisher. Today, it still seems difficult to prove that this alternative (and often illegal) “offer” actually impact the dynamics of physical sales, which evolutions generally appear to be linked to structural elements (publishing schedule, number of releases, series erosion). But this “offer” that scanlations constitute also corresponds to a demand from the fans. And instead of trying to suppress this “offer” in the (illusory) hope to bring back the stray sheep to a service (the book) that they do not want, publishers should better acknowledge the existence of this demand, and try to provide a corresponding answer.
As the release of Apple’s iPod certainly signals the beginning of a revolution of our reading habits, not unlike Napster’s impact on music in its time, publishers are now facing a delicate period to navigate. It’s up to them to decide whether they’ll do it with the fans, or against them.


  1. Namely : Akane Shinsha, Akita Shoten, ASCII Media Works, East Press, Ichijinsha, Enterbrain, Okura Shuppan, Ohzora Shuppan, Gakken, Kadokawa Shoten, Gentôsha Comics, Kôdansha, Jitsugyô No Nihonsha, Shûeisha, Junet, Shôgakukan, Shôgakukan Shûeisha Production, Shôdensha, Shônen Gahôsha, Shinshôkan, Shinchôsha, Take Shobô, Tatsumi Shuppan, Tokuma Shoten, Nihon Bungeisha, Hakusensha, Fujimi Shobô, Fusôsha, Futabasha, France Shoin, Bunkasha, Hôbunsha, Magazine House, Media Factory, Leed-sha, and Libre Shuppan.
  2. Viz Media, Tokyopop, Vertical, and Yen Press.
  3. Alexa bases its Internet statistics on the tracking of the use of its free toolbar. For reference, the representativity of the corresponding sample (with an over-represented anglo-saxon population) as well as its accuracy for websites with low traffic have often been questioned.
  4. Even if a few scattered voices sometimes attempt to put forward the impact of a good “buzz” for helping putting a book in the spotlight.
Humeur de in June 2010