The Reality-Based Community

 Rupert Murdoch, tree-hugger

A convert speaks:

Climate change poses clear, catastrophic threats. We may not agree on the extent, but we certainly can't afford the risk of inaction...
Today, I am announcing our intention to be carbon neutral, across all our businesses, by 2010...
Our strategy everywhere is the same: first, reduce our use of energy as much as possible. Then, switch to renewable sources of power where it makes economic sense. And, over time, as a last resort, offset the emissions we can't avoid...
We can do something that's unique, different from just any other company. We can set an example, and we can reach our audiences. Our audience's carbon footprint is 10,000 times bigger than ours. That's the carbon footprint we want to conquer.

Al Gore? Ben and Jerry? An Aussie guest character on Sesame Street? No, Rupert Murdoch, declaring a policy for his media empire in May.

It will be fun when the denialist journalists he employs at Fox and the WSJ finally get the memo. Go out and preach climate redemption to the gentiles!

The obvious question is: why?

I can't see this as the old boy losing his grip, and reading out green propaganda that his solicitous children press into his palsied hands. Rupert Murdoch is still the Godfather, a classic right-wing media tyrant. But like Alan Greenspan, he's always been reality-based, and hates being on the losing side. Famously, his British tabloid The Sun endorsed Blair against Major in 1997 - capturing enduring influence with media-obsessed New Labour. He observed in the speech that public opinion on climate change has shifted, even in the USA. If you are forced to move, best make a virtue of it. If he decides to back Hillary or Barack in 2008, I suggested a year ago that some high-profile employees will be out of a job.

It's similarly pure speculation, but I suspect that personal reputation has played a part in the conversion. Self-made tycoons, like other powerful men, would much rather be hated than ridiculed. And Murdoch is a tycoon from Australia, famous for its ancient colonial inferiority complex, the cringe. There comes a point in the public reception of any controversial scientific discovery where denial just makes you a figure of fun in élite society - and Murdoch can't afford to be laughed at by too many of the investment bankers, major advertisers, power brokers, and celebrities his empire feeds off.

The more interesting question is why Murdoch has taken such a strong green position. Kyoto calls for 5% cuts in Co2 emissions below 1990 levels, about 15% from today's. The EU has gone one better, and has agreed on 20% cuts below 1990 by 2020 (para 31), 30% if others agree (para 30), and ultimately "60% to 80%" (idem). Stern (executive summary, p. xi) has called for cuts of 25% below current levels by 2050, and ultimately of 80%. The IPCC has said (Third Assessment Report, executive summary, p. 15) that

Stabilization of CO2 concentrations at any level requires eventual reduction of global CO² net emissions to a small fraction of the current emission level.

But few policymakers are yet calling for such low targets concretely. So in adopting carbon neutrality - zero net emissions - as a short-term corporate goal, Murdoch is taking a splendidly avant-garde and maximalist green position.

Is Murdoch cheating on the targets? Far from it. He claimed the carbon footprint of News Corp. in 2006 was an independently verified 641,150 tons, including electricity and fuels. Does this include the carbon cost of the newsprint used? A back-of-the envelope calculation suggests that it might:

7.5m daily newspapers x 250g per paper x 300kg carbon-eq per tonne paper = 200,000 tons carbon-eq per year.

The methods were not released, but claimed to be in line with good practice. He referred to a green business organisation, the World Business Council for Sustainable Development. Unfortunately this club doesn't actually have an agreed methodology - it just describes an ongoing debate between rival consultants.

The conceptual problem is the boundary. If you just look at the value added, a service business like News Corp. basically consists of a lot of warm bodies in rented offices, with negligible direct carbon footprint. Practically all the carbon is embodied in bought-in goods (paper) and services (electricity). The consultants are worried by this and advocate a more or less myopic accounting for the supply chain. But even if they wanted to be comprehensive, they can't. It's simply impracticable by studying a single firm in some extended paradigm of accountancy.

To see why, look at a sample industrial process like papermaking, one of the industries that sells to News Corp. This industry burns fuel, and buys in wood, chemicals, machinery and transport. In turn the chemical firms etc. burn fuel and buy in other inputs, including paper, transport and so on. Take enough steps, and the entire economy is involved iteratively in the production of anything at all. An economy is an ecology.

There is a sound economic approach to the problem, though they aren't using it: the venerable input-output model developed by the Russian exile Wassily Leontief from the 1920s. Skip the next three paragraphs if you hate maths.

To make the exercise feasible, Leontief assumed constant returns and fixed coefficients of inputs (say steel and rubber) for any output (say cars). Notate these like this: the coefficient a_ij is the quantity of the product of industry j required for a unit gross output of industry i. The units can be physical or monetary or mixed, as long as they are used consistently. These coefficients are found empirically from statistics of inter-industry transactions. You assemble these Leontief coefficients into an input-output matrix A, where the columns and rows are the same list of industries. We specify a vector of final output c, and we want to find x, the vector of gross outputs, including those used up by intermediate processes. I is the standard diagonal identity matrix. The matrix algebra of the model, which I copy from Wikipedia, is quite simple:

x = Ax + c
c = x - Ax = Ix - Ax = ( I - A ) x
x = ( I - A)^-1 c

The operation referred to by the superscript "-1" is taking the multiplicative inverse, as 1/5 is to 5. For a large matrix, this is a tedious job that requires a computer. You can play with the function MINVERSE in Excel.

We need a global matrix of about a thousand industries, so as to be able to distinguish between newspapers and television. In addition to the usual row for the ultimate input of labour, we need one for carbon emissions, with physical units. (I don't know how Leontief handled capital inputs.) The total output of "carbon" is an entry in the vector x: x_carbon.

Say we are interested in one industry, newspapers, with final output c _newspapers.
The carbon footprint of this output can be found by setting all rows of c other than newspapers to zero and performing the multiplication. We can then find the unit footprint or that of a typical firm in the industry by simple arithmetic. But we must have the entire inverse matrix ( I - A )^-1 (update correction). You can't just do it with the carbon row of the initial matrix A. This technical requirement reflects the fact of the massive interdependence of the economy, and would reappear in any other realistic model.

Leontief's is probably the simplest imaginable model which can do the job of calculating the true carbon footprint of a single firm. Trying to do it from inside such a firm is a waste of time. Is there a global input-output model augmented for carbon? Not to my knowledge. Seeing the amount of misdirected effort that's going into economically naïve environmental audits, it would be a good investment.

Murdoch's green commitment has an odd ethical implication. News Corp. is adopting a carbon-neutral objective that includes some of the value-added of its papermaker suppliers like Stora Enso. Suppose Stora adopts a similar objective. If all companies do this, the result is a bidding war that leads to a massive carbon reduction, only desirable in a back-to-the-Paleolithic ultra-Green manifesto. We may need this eventually, but for the time being most of us would settle for stabilisation of CO² and temperature.

The Golden Rule requires a fair allocation of burdens within a sensible overall target, which carbon neutrality is. It's reasonable to add in a margin to handle a percentage of backsliders - for example, you should pick up a bit more than your own litter - but altruistic heroism is not usually required, and can never be expected. Experience since Hammurabi indicates that even golden-rule conduct cannot be relied on in business without legal coercion or other strong social controls. Heroic voluntarism on carbon emissions amounts in practice to affordable PR by reputation-sensitive companies like News Corp. and KPMG that don't really produce much of them. You won't see voluntary carbon neutral commitments from Alcoa, Arcelor-Mittal Steel, Lafarge Cement, Delta Airlines, or ConEd - where the real action has to be.

Update correction
I put this badly. You only need one row - indeed only one entry - in the inverted matrix for the multiplication with c_newspapers. But that entry includes information from all over the initial matrix A. It remains true that to calculate any of the footprints of a single industry in this way you have to know and use the whole of A. A different method would require an equivalent model of the supply web of the whole economy.