"The Long Tail" on DRM, downloading and digital media

(Updated Aug 25 pm)

The Long Tail (Chris Anderson) has great insight on the digital media, and in fact, the insight applies to many internet businesses, and business in general as well.

His has a new post about DRM, basically explaining why Movielink and CinemaNow have so far failed. While I do not fully agree with his argument, he made a few interesting points worth discussing

1. "Any protection technology that is really difficult to crack is probably too cumbersome to be accepted by consumers
2. Instead, efficient software and entertainment markets should exhibit just enough piracy to suggest that the industry has got the balance of control about right: not too loose and not too tight. That number is not zero percent (which requires protection methods so invasive they kill demand), and it's not 100% (which kills the business). It's somewhere in-between
3. piracy can actually let you raise your prices: The usual price-setting method is to look at the entire potential market, from the many at the economic lower end to the few at the top, and set a price somewhere in between the top and bottom that will maximize total revenues. But if you cede the bottom to piracy, you can set a price between the top and the middle. The result: higher revenues per copy, and potentially higher revenues overall."

His point #1 is intuitive and is supported by his other points.

I have another good example for his point #2, it proves the "could" but not his "should" assertion. See for example the encryption of DVD, CSS. Free DeCSS softwares such as DVD Decrypter and DVD Shrink are available across the internet (e.g., here). Does this damaged the $15bn DVD market? No. Does the ridiculous region code prevented decrypters at work? No. Those who went to length at DeCSS are not going to buy the legitimate products if pirated copies are not available. The only difference is that the region code has greatly reduced efficiency and increased the cost of manufacturing and distribution. In this case CSS is appealling to consumers, because it is not as cumblesone as DRM. The fact that CSS is cracked does not damage the key areas of its target market.

His point #3 is not really right, and it requires some elaboration and discussion.

• In making this point Anderson has implicitly assumed price uniformity across the world (for software, or media content products), which is fine. Unlike that of pharmaceuticals, it is easy for software and media content to surf cross national coundaries, thanks to the internet (and the small size/weight of the DVD which facilitates shipping). Therefore, we can assume that parallel import has will effectively equalize price, roughly, as we have observed in reality.
• The second assumption he has made is that the long tail (the yellow portion in the chart above) is not very sensitive to price. (i.e stiff price elasticity). This is where I do not agree. The x-axis of the graph is price, the y-axis the volume. Price is the line that separates the two areas. Total revenue is the yellow area times price (the cut-off). Therefore, if the revenue loss of the portion above the cut-off is not compensated by the increase in volume, Total revenue will decrease. (In pharmaceutical, it was possible to apply differential pricing to different markets (Canada and France, vs US) up to this moment, so the argument here does not really work. However, more and more people are importing drugs from Canada.) This is where his logic falls apart. As Brad at Panopticologist correctly pointed out, "When a firm with market power sets their price they are always making a trade-off between receiving a higher price from a few and a lower price from many. The rational firm looks at the entire market and chooses a price which maximises profit. If a portion of that market leaves, revenue will either decrease (if the price was sufficiently low to sell to that portion) or stay the same (if the price was too high for them anyway)." Brad's argument asserts that for a monotonic function volume(price), the revenue function (P * V) has only one maximum. The simple form of his proof assumes uniform pricing, but one can construct a more elaborated proof for tiered pricing as well.

However, what we have observed is that some degreee of piracy is associated with the successful marketed product (e.g., DVD), while none of the fully protected products make any success in the market (e.g., movielink), especially for software and media content products.

Now let me argue in defense of Mr Anderson, that there is an optimum point for tolerating piracy, and that optimum point is not zero piracy.

• There are two choices, 1) abandon the red portion (low end segment) and forbid them to use without purchase (they can't afford to), or 2) cede the low end to piracy. Microsoft's example has proven that ceded it to piracy (hence the total revenue decreased a tiny amount in the short term) will trap the market and train the potential customers, hence helping them the grow of the yellow part much faster when income level increases
• To prove Mr Anderson's hypothesis one needs to assume that the volume curve (the chart above) would expand, which is not difficult to show. Such volume increase due are found in the examples of LD vs VCD vs DVD, in Asia and US. The VCD market in Asia flourished largely as a result of the wide-spread of piracy (also see below). One could even argue that the Hollywood studios made more money over VCD than LD, for which piracy technology was too expensive. In short, the media industry has mistaken that it is a zero sum game, but it is not.
• Or alternatively, to maximize sum of revenue over time, instead of the static sum at current time. This is analogous to the free trial promotion for shampoo from P&G, except that the cost is zero for such promo. Bill Gates cleverly used piracy to accomplish his promotion campaign. The DVD Forum got their free promotion passively (see below).

In addition to Anderson's illustration with Windows OS, we could
also look at the DVD machine market, from DVD Consortium's perspective. It was the Chinese manufacturers
(assemblers) who have commoditized DVD machine, and in turn drived the DVD
market upward. Without the Chinese players, we will still have to pay for
$350-500 per DVD machine. It would not be surprising that DVD suffer the same
fate as LD in North America. And we may still be watching VHS today.

But how did the Chinese managed to cut the cost so much? A few
reasons

1. During the mid-1980s and mid-1990, Chinese companies have
   competed fiercely with each other on the home appliances, from refrigerator to
   microwave machine, to TV set and VCD players. The huge domestic market of 10-40M
   unit is great playground to move on the experience curve when they entered the DVD player
   market around 1999. As we can all see in Wal-marts and Best-buys
   today
2. The learning curve on DVD players is boosted by the learning
   from VCD player assembling (which is entirely absent in the US market, like the
   LDs)
3. Most importantly, before these Chinese players export en mass
   to the US, they perfected their skills within China, where they did not have to
   pay loyalty to the DVD consortium

The results are that the DVD consortium made a lot more money
from licensing to these Chinese factories than they could have otherwise.
Additional sales were also made in the DVD drivers, the chip and other stuffs.
Without some "piracy", Toshiba and Philip would not have profited so much in the
DVD market. This is in support of the Microsoft experience.

The problem of DRM, and that of Movielink and all the pre-iPod online music stores, in my view, lies in their business models. More precisely, the product itself. The restriction DRM imposed on the products just make them repulsive to consumers. There is no "ownership" under DRM, and the pricing for DRM content is totally wrong. If they call it "rental" instead of "sales" it might fare better. But then DRM cannot compete with streaming media. It is caught in the middle of no-man's land.

We learn from the above discussions

1. Highly successful products are often correlated with certain degree of piracy
2. Ceding the lowest segment to piracy might help nurture for future growth
3. Cumblesome anti-piracy measure such as DRM will drive away potential customer, simplifying the technology and making it more user-firendly will help boost volume significantly, even though that might mean losing a portion of potential volume to piracy, as that would be more than compensated by the volume gain in the other segments
4. The failure of DRM related products such as Movielinks are due to their business models, which is in part dictated by the inflexibility of the DRM technology itself

When it comes to new technology, one just has to bear in mind that this is not a zero sum game. Opportunity rewards the risk-takers.