18 May 2008

Taxes Are Easier...

In response to my support for carbon taxes, Sheryl over at EDF writes
A carbon tax provides price certainty, but not greenhouse gas emissions certainty. What we need is the latter, and that is what a cap-and-trade system provides - emissions certainty. A well-designed cap-and-trade system can provide adequate price flexibility as well.

Nat Keohane, a Ph.D. economist here at EDF, wrote a great post explaining why a tax is the wrong tool for a job and cap-and-trade is the right tool. He wrote this in response to a phone call from the head of the Congressional Budget Office objecting to his review of a study they’d published. This is a good read - take a look
Oh neat, PhD economist debates! I replied with this:
I disagree about the “certainty” aspect of cap and trade. While true in theory, real emissions will be uncertain to the extent that emissions are unmeasured or mismeasured. Take the measurement of population. In the US, the census is considered to be “fairly” accurate, but naysayers complain about various important groups (homeless, migrants, et al.) being unmeasured.

Now consider how most countries do not even know their populations — and multiply that mismeasurement by — say — 120 (census once per decade versus monthly emissions measurements). Carbon taxes are much simpler to enact, measure, and enforce (except for gathering and burning wood — but they are likely to be missed in cap and trade too).

I took a look at Nat’s post. While I agree with him (and my PhD is younger than his), I think he is leaving out another “transactions cost”-related aspect of cap and trade that does not exist with carbon taxes. Say that permits are issued on an annual basis (by auction), that a bank exists, etc. but it’s “suddenly” discovered that too many permits are out there. The next auction must reduce the number of permits AND policing “black market” emissions must increase.

With taxes, the response is only to increase the tax on carbon inputs. That increase can occur on or before the date for the next auction, and there is no need to police emissions, since taxes are part of the prices at the start — not end — of the carbon-generating cycle. It’s much harder to have a black market input (coal from a different mine) than black market output (coal burned in a burner without a permit).

Taxes are better because they are simpler AND easier to adjust. Forget the theory — these systems have to work, worldwide, in reality.
Tell me what you think -- even if you do not have a PhD :)


  1. Part of the problem with their argument is that it postulates a "well-designed" cap and trade system, as opposed to the one we would likely get in the real world. Given that politicians are going to be the ones ultimately doing the designing, I think a tax makes more sense because it offers less opportunity for them to screw it up.

  2. They're also arguing from a false premise. They've confused terminal value with instrumental value.

    The terminal goal is presumably maximizing social welfare. Minimizing environmentally related costs is the penultimate goal. One instrument for achieving this is reducing carbon emissions. Reduced emissions are not a goal in and of themselves. So we can only demand certainty in this instrument to the extent we are certain about its relationship to the terminal goal.

    Unfortunately, it seems we are very uncertain about the relationship between reducing carbon emissions and increasing welfare. First, we don't seem to have a very precise characterization of the dose-response relationship between emissions and external environmental costs. Second, we have an even worse handle on the tradeoffs caused my emissions controls that might blunt some of their positive benefits.

    Given all this uncertainty about how reducing emissions will help us, it seems a little silly to argue that emissions certainty is a reasonable metric for comparing policies.

    In some sense, certainty may be bad because it's less adaptive to emerging information. A better policy might be Ross McKitricks T3 tax, which is even more uncertaint than a straight up carbon tax, but more adaptive to the actual link between emissions and costs:


Read this first!

Make sure you copy your comment before submitting because sometimes the system will malfunction and you will lose your comment.

Spam will be deleted.

Comments on older posts must be approved (do not submit twice).

If you're having problems posting, email your comment to me