Permanence and Leakage

Permanence and leakage have not been as problematic as additionality,but they have been quite challenging in their own right. Particularly for categories of offset projects like those involving forestry and land-use change.

Offset Permanence

Tackling permanence involves both policy and technical questions that include:

  • How long is permanent enough?
  • Is there monetizable value to temporary carbon sequestration?
  • How does permanence rigor influence offset economics?
  • Should potentially reversible (non-permanent) offsets be treated as a separate commodity from the rest of the offset market?

There are no “right answers” or even “obvious answers to these questions, but an enormous amount of work has been done over the last 20 years to grapple with them.

One thing that is clear is that how offset permanence is handled can dramatically change the economics of carbon offsets from some sectors, and even effectively eliminate those sectors them as a source of offsets.

NOTE: We’ve added a more technical discussion of Permanence to the Climate Web, accessible here:

Offset Leakage

Leakage too has proven to be a challenging criterion as market participants have proposed receiving offset credits for projects like shutting in oil and gas fields, prematurely closing a coal mines, or shutting down a factory. All of these project types would result in close to 100% leakage.

Leakage is a potential problem whenever a proposed offset project is also involved in standard markets, e.g. a reforestation project and its potential interaction with global fiber markets. Closing a coal mine does not result in a reduction of CO2 emissions if that mine’s production is made up for at other coal mines. Closing a factory does not result in a reduction of CO2 emissions if it is simply being moved from the U.S. to China. All of this is leakage.

Let’s take factory example. Assuming for a moment that it might make economic sense to shut down your factory in return for the ability to sell carbon offsets (additionality), what would the net effect be for the sentient atmosphere? Depending on what one produces, pehraps not much. Presumably there is a demand out there for your product; if you stop producing it, someone else will step in and build a new factory , to supply the commodity. The atmosphere may see no difference at all in terms of greenhouse gas emissions; the emissions simply “leaked” from one factory to another. With leakage, a step you put into motion in one place triggers feedbacks to occur somewhere else that counteract in whole or in part the benefit that you seek to achieve.

Not all cases are as simple as shutting down a factory. Global markets and global commodities can be complicated.It can be almost impossible to really know what the leakage potential is for a given kind of carbon offset project. Let’s take a look at forestry projects intended to sequester carbon by growing trees that will eventually go into the market as pulp or as board feet of lumber. (We are distinguishing this from trees planted in a natural area where it’s understood that the trees will never be harvested.) If you plant these trees, and you know that in 45 years they will come to maturity, what’s happening with respect to that sentient atmosphere? First, someone seeing you plant these trees may decide that they won’t plant the trees they themselves were going to plant; perhaps they think prices will be too low in 45 years if everyone plants more trees today. That would be a clear case of leakage. But how do you ever know how the market is responding to a particular tree planting, especially when it is inconsequential to total forest volumes? Many efforts have been made to model timber markets to calculate leakage estimates, but it is largely a frustrating errand. Any estimate of leakage calculated in this way is almost certainly going to be within the noise of any model’s ability to estimate outcomes. At some point, someone has to decide somewhat arbitrarily what leakage to assume for purpose of calculating a carbon offset.

It’s not just forestry. People have tried to design carbon offset projects around buying land that has coal under it, or land that has oil under it. The idea is that placing that coal or oil off limits can then be counted as a carbon offset because the fossil fuel will never be mined or pumped. However, even if you accept that the coal will never be mined, or that the il will never be pumped, what does the sentient atmosphere see? In the short term, other coal will be mined or other oil will be pumped to supply the existing market. The sentient atmosphere would unlikely see any impact.

Like additionality, leakage can usually not be empirically measured when evaluating carbon offsets, and often requires similar policy determinations.