Saturday, February 07, 2009

A Non-Rigorous Calculation: Are E-Books Green?

Just ask Steve Jobs: the best kind of marketing you can get is where your customers do the marketing for you. Buzz is everything.

E-books are one of those markets. Users are absolutely convinced they're right about everything associated with e-books, and are happy to do the marketing on behalf of the industry (how stupid can you be, to act as an unpaid sales rep for anyone?). A quick round-up of e-book arguments runs, roughly:

  • Convenience – Why carry one paper book when you can have an entire library everywhere you go?
  • Cost – E-books are cheaper.
  • Inevitability – This is the way of the future. Anyone who won't get with the program is a luddite.
  • Environment – E-books are greener.

Since the environmental argument is the clincher-of-the-day, I'd like to run it over the back of an envelope and see how it stacks up.

Warning: this is not an academic paper. If you want rigourous analysis, it won't be here, because I don't do rigour for free.

Let's start with the simplest case: all households in Australia have three e-book readers and no longer buy paper books. With roughly 7 million households in Australia, we need 21 million e-books to make paper books redundant.

I don't have a calculation of carbon footprint for e-books, but there's a common enough number for mobile phone manufacture – about 60kg of carbon per phone. Let's give e-books the same number.

That works out to 1.26 million tonnes of carbon to equip Australia for a future free of dead-tree books.

What's the carbon footprint of Australia's book consumption?

That's not too hard to provide, at least as a thumbnail. Australians buy about 130 million books a year, and according to the US Book Industry Study Group, a book's footprint is about 4 kilograms. That comes to 520,000 tonnes of carbon in Australia, if the American figures hold true.

This, however, ignores a couple of things.

The first is the electricity consumption of e-books. This is imponderable, because the one thing I can't tell you is how many people will leave the wall-warts plugged in and warm even when the book is charged.

The biggie, however, is the matter of product turnover. No manufacture can stay solvent by selling a device once, and never selling any more. So let's assume that just to keep the suppliers available to us, we'll need to turn the inventory of e-book readers over every ten years – roughly two million devices sold each year.

That puts a premium on the existence of the market; just for the back-of-the-envelope, we'll put that premium at 120,000 tonnnes of carbon (the footprint of 2 million replacement readers each year).

The other item is the network that delivers the books.

I'm going to use a couple of rough estimates here. The first is from Tom Worthington, who calculates that 20kB crossing the network equates to one gram of carbon (he was talking about e-mail, but let's just borrow the number for PDFs as well).

If each book, as PDF, is 1MB, then its network cost would be 50 grams. If Australia's book consumption remained at 130 million books on e-book readers rather than paper books, the network cost would be 6500 tonnes of carbon per year; nowhere near the millions, but neither is it insignificant.

Let's take the whole thing over five years.

Paper books, at 130 million per year sold in Australia, would generate 2.6 million tonnes of carbon.

For e-books, the assumptions (to repeat myself) are 21 million devices, plus 2 million per year new sales to sustain the market, plus 6500 tonnes of carbon per year for downloading books.

The total, at 1.9 million tonnes, is less than the paper book – but it's not as huge as advocates would have us believe, and it certainly doesn't make the e-book some magical “green” alternative.

And there are a few other cost items that I didn't include.

The first is simple: if the e-book needs more carbon than the mobile phone – just 80 kilos of carbon instead of 60 – then the advantage is nearly zero.

The second is that I've ignored disposal. E-books will either add to the toxic metals stream in landfill, or they'll need energy spent on their disposal.

And finally, there's the assumption that the two markets are exclusive. That is, the e-book isn't just another must-have consumer toy: it really is an exclusive replacement medium.

If this last is not true – if e-books create a distinct market that co-exists with paper books – then e-books are an environmental disaster: they will add to the carbon footprint, not reduce it, while at the same time delivering marginal consumer benefits at the cost of being leg-roped into licensing agreements, dragging publishing under the net of DMCA-type legislation, and reducing our freedom to read, share, resell, and give away our books.