Today on Blogcritics
Home » Culture and Society » Science and Technology » A New Way: Business Model for File Sharing

A New Way: Business Model for File Sharing

Please Share...Print this pageTweet about this on Twitter0Share on Facebook0Share on Google+0Pin on Pinterest0Share on TumblrShare on StumbleUpon0Share on Reddit0Email this to someone

We’ve been waiting for a comprehensive plan to replace the current business model with regards to music and other content over the Internet. Check out this one from Harvard University Professor Terry Fisher, as discussed in the Register:

    Fisher actually lays his philosophical armory down for us to inspect at a very early stage, and it’s thus. Economies have had a lot of trouble with public goods that are ‘nonrivalrous’ – if you use it you’re not depriving someone from access – and ‘nonexcludable’ – it’s actually hard to make them exclusive. Examples of the latter include roads, defense, and culture. It’s a real danger that if no one pays, then nothing gets done: the roads crumble, the country becomes vulnerable, and aspiring pop stars give up their dreams of one day snorting cocaine from an expensive prostitutes thighs.

    But our flippant illustration of the final example is not entirely accidental. Many artists forsake fame for fame’s sake – but the beauty of alternative reward models is that there’s no disincentive for them to become popular, either. To cite an example, when we’ve discussed flat fees before, someone usually writes in with some anguish to complain how this would only reward Michael Jackson for being popular. But what’s wrong with that? There was a time when he was very popular, and deservedly so.

    So let’s start at an accountant’s year zero.

    Calculating lost revenue
    In the year 2000, the record labels earned $7 billion on retail sales of $13 billion. For the sake of argument, let’s assume that in the first year 20 per cent of retail sales were lost to unlimited copying. That’s $1.4 billion, although they’d save $210 in manufacturing costs, and approximately $145 million in mechanical royalties. That brings the compensation to $1.045 billion for the recordings royalties and $138 million for songwriters, plus an amount for lost radio-related royalties.

    For the movie industry, calculating the potential loss is extremely difficult. Firstly it’s hard to estimate how much the industry earns now from DVD and VCR sales and rentals, and cable and satellite deals. And it’s even harder to gauge the loss from file swapping. Even with the advent of Bitorrent, downloads are slow, and few have the patience or resources to find value in them compared to the availability on offer at plentiful late night retail outlets. Fisher reckons five per cent, rather than twenty per cent for the music business, of a $10 billion industry, or $479 million.

    So combined, that’s $1.677 billion to keep the RIAA and the MPAA happy.

    But of course that’s not all it would cost: the model requires an organization to calculate and distribute the royalties, performing the duties of ASCAP or BMI today. ASCAP reported that its 1998 administrative overhead was 16 per cent, so Fisher generously estimates 20 per cent. (It’s pretty generous, as we’ll see, because the digital overhea ds may actually be much lower). This takes – and bear with us, because it also generously throws in a 10 per cent charge for inflation between 2000 and 2004 – the net result to $2.306 billion.

    So who pays?

    Raising the money

    If it was implemented as a regressive poll tax, with 87 million household filing IRS returns, each household would pay a mere $27 extra a year: a little over $2 a month, or 51 cents a week. That’s half the price of a single iTunes Music Store song.

    That’s the most efficient way, with the lowest overheads.

    However, any kind of income tax increase is obviously a hard sell, especially in God and Gubbment-fearing America. And there are many sound objections. Why should the poor subsidize the rich? Why, notes Fisher (who clearly must remember the culture wars of the early 1990s), should a proportion of the population which finds the entertainment products blasphemous be asked to subsidize their creators? And why should Net-free households want to subsidize the broadband users who are actually taking advantage of the system?

    Fisher then exhaustively discusses four other options: taxing the playback devices and/or burners, levies on the physical media, levies on the delivery service, such as KaZaa, or on the Internet access point (your ISP). The latter is by far the largest: spending on broadband in the US in 2004 is estimated to be $16.4 billion. By contrast, blank media sales generate $2 billion in revenue. In total, these four categories gross $20.248 billion. And so to get our $2.3 billion to compensate artists, studios and labels would require an 11 per cent hike.

    But what if it fell entirely on broadband users? Some might find the figure surprising: excluding all of the other penny taxes we’ve just mentioned, the cost will be $6 per broadband user per month. Um, is that all? Well, actually, yes it is.

    So what do consumers gain from suddenly being able to exchange music? It’s perhaps the most delicious question that’s ever been asked – and there are so many advantages to the free exchange of culture that we may have forgotten what they are.

    ….The simple idea is very powerful. Fisher identifies four constituencies necessary to accept the model: consumers, artists, device manufacturers and finally the intermediaries: the studios and labels. The model has huge advantages for three of the four. And what incentives, we asked, would the labels and studios have?

    After hearing his presentations, Fisher says industry is intrigued but hardly feels impelled to jump. The biggest ‘carrot’ is that it would see its revenues guaranteed at 2000 levels. If it believes its own rhetoric, that could be a very powerful incentive indeed.

Extremely interesting – read the whole thing.

Powered by

About Eric Olsen

  • http://www.foliage.com/~marks Mark Saleski

    it’s an interesting idea…but so far i don’t hear a lot about this approach coming out of either record companies or the riaa.

    what i do see are more & more items taking the punitive & “gotta stop this” route…

    as much as i’m opposed to paying extra in my broadband bill (because i download nothing), i probably wouldn’t complain if the end result is that the record labels get back to getting music out there (though i’m not convinced that would happen either…at least not with the big labels)

  • Eric Olsen

    The labels will be the last to agree to something like this because they have the least to gain – especially in their minds where they still think they can end file sharing with litigation DRM and threats.