Useless metric of the decade

Does the idea of "embodied energy" convey any useful information whatsoever in the vast majority of its applications?

I ask because I recently read about the difficulties faced in calculating this metric. For the uninitiated, it means the energy consumed in creating one unit of product X. People concerned about climate change and the supply of energy, of which I am one, sometimes suggest that it should be used as a reference to encourage the conservation of energy. For example, some say it should be printed on the labels of consumer goods, or even used as a basis for pricing.

But nobody can agree how to define it. Do you include the entire energy usage of the factory? Heating? Lighting? Street lighting? Do you include the energy usage of the workers on their way to work? If yes, why not include the energy they use at home - after all, they are spending their wages from one lot of energy use to buy electricity, gas, and vehicle fuel? Quite quickly you get into a theological level of debate. Should BT take responsibility for the power used to route every IP packet that crosses its network? After all, it's buying the power to drive the routers and switches. But a lot of those packets, even a majority, are transiting from - say - AT&T to DE-CIX, and BT didn't explicitly choose to carry them. The contractual relationships don't include them, and they didn't start it. Shouldn't either DE-CIX or AT&T pay?

Worse, when you try to implement an energy tax on this basis, you hit some horrible policy landmines. Why should I pay for the embodied energy in X? Why should the manufacturer get away with it? Or their workers - if you include their journey to work, why should they be able to get away with their energy use? Imagine if every issue of Mobile Communications International's price included a tax to represent the electricity South West Trains uses on my behalf to get me to work. Why should the reader pay so I can go to work? Why can't I use the same principle and charge my season ticket to expenses? This is serious. I don't control the use of the energy, so a tax on me can't change the behaviour of the people who do. There's no incentive for the manufacturer (or their suppliers, or the distributor..) to save energy, unless you double (triple or worse) count. But there is worse.

The whole thing is a lot like VAT carousel fraud. There, the fact that the tax is collected or refunded at various points between the manufacturer and the end-user means that it's possible to make a good living shunting the liability from level to level. The best solution I've seen is just to charge it at the final sale to the end user.

Now there's an idea.

Rather than the intellectual struggle to define embodied energy, the even tougher one to avoid creating stupid distortions, opportunities for fraud, and perverse incentives, and the cost of the bureaucracy needed to police it, why not just impose a tax on energy at the point of use? Or rather, on nonrenewable and CO2-generating energy? We've already got a highly efficient system for the collection of indirect taxation, run by the least corrupt civil service in the world. It's called HM Revenue and Customs, and the taxation is VAT, as mentioned above.

There is one kind of product where "embodied energy" is useful, of course. That is anything intended to convert energy available in nature into a useful form, like a solar panel or a wind turbine. If they produce less energy in their design life than they use (nonrenewable) energy in production, they are useless. Seeing as there are costs of production and operation beyond energy, though, they are very unlikely to be worth producing.