Why ban lightbulbs?

By Eric Crampton 05/09/2012

You could make a case for mandating energy efficiency requirements for lightbulbs and banning less-efficient incandescent bulbs in a country without an effective carbon tax or emissions trading regime.
But can you make the same case where we already have emission trading? It’s a bit more difficult.

It’s hard to make any sense of the reversal of former Government policy on incandescents other than in the most cynical of political terms. It is in direct contradiction to any concern they express to tackle climate change. Lighting has been estimated to use nearly 20% of the world’s electricity and six years ago the International Energy Agency produced a report which concluded that a global switch to efficient lighting systems would trim the world’s electricity bill by nearly one-tenth. It is a low-hanging fruit in the reduction of carbon emissions. Even the US is to phase out incandescents.
Many people are making the switch to efficient bulbs without Government direction. It makes economic sense to do so after all, in addition to the clear environmental benefits involved. But Government also has a responsibility to advance energy efficiency by appropriate regulation, as other free market economies have recognised.

Let’s go back to first principles.
A lightbulb has three sets of associated cost. The first is a fixed cost for the bulb; the second is the ongoing cost of electricity to run it. A third and today largely ignored cost is disposal at end-of-life: some of the fancier lightbulbs come with greater risk of leaching nasty stuff into landfills and so either impose that cost or impose the cost of more careful disposal.
Power generation has some associated external cost to the world through carbon emission. While most electricity generated in New Zealand comes from renewable sources, the marginal unit often comes from coal-fired generation. And so emission abatement has some external benefit. Absent the full costs of power generation being internalised into the price of electricity, you can make a second-best case for regulatory interventions to push people to the choices that they would have been making in a full-carbon-costing world. Now, there’s a problem in that electricity is also used in the production of lightbulbs, and if the non-priced carbon embodied in the production and distribution of more efficient bulbs sufficiently outweighs the non-priced carbon embodied in incandescent bulbs, the result could reverse. I have no clue about either, but it would be awfully surprising if fluorescent and LED bulbs did not have more carbon emissions associated with their production than comparable incandescent bulbs – I would expect that differences in embodied non-priced carbon would be proportionate to differences in the cost of the bulbs. But let’s stipulate for now that the ongoing flow of carbon is lower for the more modern bulbs, especially as they have a longer replacement cycle. Always keep in mind that it’s not easy being green: when prices don’t fully incorporate costs, alternative methods of calculation have non-trivial associated problems.
But this doesn’t hold when we already have a reasonably comprehensive emissions trading scheme. Electricity is in the system, even if farming isn’t quite there. If the permit system is working well, there is absolutely no case for banning incandescent bulbs. Even if incandescents are less efficient at producing light, they’re not all that bad at producing heat. And for two thirds of the year, at least here in the South Island, that isn’t all waste. 
If power prices incorporate carbon charges via the ETS, then there’s no real economic case for pushing consumers to choose bulbs they don’t want. If the ETS isn’t working well, then all kinds of consumer and producer decisions will be out of kilter and we do far better by trying to make the ETS as clean across the board as possible rather than mucking about in individual markets. You’re then forced into a political second-best argument that it’s impossible to fix the ETS but perhaps possible to get political support for pushing on a few important markets. But, again, it’s awfully hard to tell in any of those individual cases whether we’re doing net good in doing so. The UK thought it was doing good in adding food miles; they’d missed that our pastoral systems have lower overall greenhouse gas emissions. 
In very important ways, the problem facing somebody wanting to intervene in particular individual markets to try and fix the problems caused by not having a good ETS are similar to the problems facing somebody trying to run the old Soviet economy. I’m not trying to make a dumb ideological point about Greenies here: rather, it’s about information and its dissemination through a system. The Soviet planners had to figure out how rationally to allocate scarce investment resources in a world where they couldn’t really tell how much consumers valued anything; that’s a hard-to-impossible problem to solve. An environmental planner working in a world without either a comprehensive carbon tax or an equivalent ETS has a parallel problem in trying to figure out all of the environmental upstream and downstream costs of any product, its substitutes and its complements, and of all the processes used to produce it and its substitutes and its complements. Art Carden explains it in more depth. Product-by-product intervention is awfully likely to produce environmental absurdities. You don’t have to be a climate change denier to oppose piecemeal interventions of this sort.