Banning the bulb – the information critique

By Eric Crampton 17/09/2012

I’d noted a few problems with banning incandescent lightbulbs. First, we can’t be sure that one bulb is really friendlier than another without very comprehensive information on how they’re manufactured; second, where New Zealand’s ETS has things roughly right, at least when it comes to power generation, it’s pretty hard to make a case for banning lightbulbs.

Bryan Walker notes a couple of studies suggesting that the cheaper-to-run bulbs are also friendlier to manufacture:

However I had a look to see what I could find, and came across this assessment of CFLs from a writer initially inclined to be sceptical about them, and this report on LEDs. It doesn’t look to be an issue.

A commenter at Offsetting found this one too.

I’ve no particular dog in this fight; if CFLs and LEDs are friendlier to manufacture, so much the better. But I’m still not sure that it’s actually knowable. For example, Bryan’s first link provides this table:

Here’s a summary of the embodied energy in a light bulb (all numbers represent energy in kWh):

CFL Incandescent 8 Incandescent bulbs
Glass 0.17 0.11 0.88
Plastic 0.68 0 0
Electronics 0.66 0 0
Brass 0.18 0.18 1.44
Operation* 120 60 480
Recycle** 1.69 0 0
Total 123.38 60.29 482.32

* This assumes the CFL bulb operates for 8000hrs and the incandescent bulb operates for 1000hrs
** This assumes that the energy required to recycle a CFL bulb is equal to its production 

Let’s think about the plastics in the CFL and assume that the table has everything right about the direct energy costs of making plastic. But what about the energy costs of the machines that had to be bought to make the plastics? What about the costs of the machines that made those machines? We’d need to know everything about all the materials that go into all of the pieces of equipment that make the machines that make the machines that make plastics, and then everything involved in the materials used in making that prior set of machines, and so on all the way back.

Read Leonard Reed’s I, Pencil. If a pencil’s that hard to figure out, an LED bulb isn’t going to be easier.

That’s just the information problem on the supply side. What about heterogeneous customer demand based on their having very different uses for lightbulbs in different places? In large parts of the country in large parts of the year, waste heat from incandescent bulbs is not waste. It’s just a less efficient way of partially heating your house. When I spend a dollar in power heating my house with my lightbulbs, I’m wasting about fifty cents if my heat pumps are twice as efficient as radiant heat; I’m not wasting the whole dollar. I don’t want CFL bulbs in some outlets because they take just too long to wake up and provide light; in other spots in the house, it doesn’t matter if it takes a couple of minutes to get useful lighting levels. A ban says there is no possible reason for a consumer to prefer an incandescent bulb that can outweigh the difference in power cost, and that just isn’t true.

Walker continues:

Crampton’s second point was that an ETS which is functioning well removes the need for any regulatory interference in the market. “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.” He goes on to say that if the ETS isn’t producing the desired effect the answer is to improve the ETS, not make piecemeal interventions. It crossed my mind when I was writing the post that if the ETS was functioning at a level designed to drastically reduce carbon emissions there mightn’t be a need to bemoan the Government’s action on incandescents. But it is not functioning at that level, and the Government seems determined to ensure that it never will, or will only so far in the future as to be much too late.

If the ETS is broken and unfixable, then you can start making second-best cases for all kinds of stuff. But I’d thought it was least broken when it came to electricity generation.

But should New Zealand’s ETS really go beyond that which everybody else is doing? “Drastically reduce” seems a pretty tough standard. Maybe it’s the right one if everybody does it at the same time and agrees to be bound by it, but surely NZ going it alone in pushing for drastic reductions does a lot more to ruin the NZ economy than to delay global warming; we’d have to expect the rest of the world to be remarkably strongly swayed by New Zealand’s example to expect otherwise. And that’s just not going to happen so long as the mess in Europe and the looming potential economic mess in China are the headlines.

Walker continues:

Reining in carbon emissions has become a matter of high urgency, far outweighing concerns about government intervention in the economy. For that matter the ETS itself is an intervention, designed in its original intention to make markets assume the environmental costs which left to themselves they ignore. I see no reason why it should not be accompanied by other government directives which ensure that markets are not permitted to operate in areas that clearly slow the transition to a decarbonised economy.  We accept government mandates in many parts of the economy such as the compulsory insulation of new buildings and we rue failures in regulation such as allowed the emergence of leaky buildings.
Banning incandescents does not to my mind invoke the spectre of a centrally planned economy. It’s simply part of boundary setting for markets to operate within, a proper function of government and one buttressed by the urgency of the climate crisis.

The terms of the argument I think here have shifted a bit. First, my critique of lightbulb banning wasn’t that it was interventionist; rather, that it was a worse regulation than having a working ETS. If the ETS were working correctly, there would be no efficiency case for banning lightbulbs; I’m not even convinced that there is a case for banning lightbulbs given the problems in our actual ETS.

I was hardly making the case that banning bulbs leads to a centrally planned economy. Rather, the knowledge requirements for assessing whether a ban is desirable and being really sure about it aren’t far from the knowledge required to make central planning feasible.

If we want a shot at drastic reductions, though, we could do well to take another tack entirely. The ETS imposes some costs on the economy. Not huge ones, but they’re real. Ditch the ETS and pour money into ag biotech research into improved pastoral systems for low methane; provide a free licence for anyone to use the resulting research. If it does nothing, then we’ve hastened global warming by maybe a day a century from now relative to NZ’s having kept the ETS.* If it works, we substantially abate global agricultural methane emissions. A small country in the middle of nowhere with little influence might do better with the high variance play

* My baseline assumption here is that if New Zealand as a whole were shot into outer space tomorrow, with no further emissions of any kind, we’d at most delay whatever carbon concentration or temperature milestone we’d have otherwise achieved a century from now by at most two or three days. We’re a pretty small dot.