"Offsets are an imaginary commodity created by deducting what you hope happens from what you guess would have happened." (Dan Welch quoted in The Guardian, June 16, 2007)
The topic of ‘additionality’ is the most fundamental − and thus contentious − issue in the carbon offset market. In theory, additionality answers a very simple question: Would the emissions reductions have occurred, holding all else constant, if the activity were not implemented as an offset project? Or more simply: Would the project have happened anyway? If the answer to that is yes, the project is not additional.
Additionality makes intuitive sense: If I buy carbon offsets, I make the implicit claim that I forgo reducing my own emissions (i.e. I still drive my car) but in exchange I pay someone to reduce their emission in my stead. If I “neutralize” the emissions I caused while driving my car by buying offsets from someone who would have reduced their emissions anyway, regardless of my payment, I have not eliminated any emissions, but rather have subsidized an activity that would have happened anyway. The following example illustrates this point. The example uses offsets used under a cap-and-trade regime but the same principles apply for the voluntary market.
Company A, a power producer in the UK, operates under a cap-and-trade system, such as the EU-ETS. Company A currently produces more emissions than it holds allowances for. Because of the high price of allowances Company A decides that it is more cost-effective to implement energy-efficiency upgrades in its facilities than to buy additional allowances. The company replaces its gas turbines with new high-efficiency turbines. With this upgrade, Company A reduces its emissions enough that it does not need to buy allowances to meet its quota. The cap-and-trade system has achieved its goal of inducing emissions reductions through a binding cap.
Company A decides that instead of replacing its turbines, it would like to explore buying CER credits at lower cost. Company B in China, also a power producer, is not under a cap-and-trade system, but would like to replace its old turbines, provided the company can obtain financing and access to high efficiency turbine technology. Company A approaches Company B, offering to purchase CDM credits and to transfer technology and expertise. With the additional revenue from the sale of CDM credits to Company A and access to advanced technology, Company B can now undertake the turbine upgrade. This can be considered an additional project.
Company B in China has already determined that it will upgrade its turbines, and has sufficient financing and access to suitable technology. Company A offers to partner with Company B and present this project as a CDM project, creating CDM credits corresponding to the activity that was planned to be implemented anyway. In this case, the CDM project cannot be considered to lead to any additional reductions. If it were to be registered as a CDM project regardless, it would result in the creation of credits that would allow Company A to emit more than it would have without the CDM project without having created any compensating reductions.
The calculation of the number of offsets generated by a project is inherently problematic. The key difficulty lies in the need to compare the projects’ actual emissions to a counterfactual scenario reflecting another reality, one in which the activity is not implemented as an offset project. This scenario is referred to as the “baseline” scenario, and the number of generated credits is equal to the difference between emissions in the baseline scenario and emissions resulting from the project. There is no fail-safe way to divine what the baseline scenario would be. Various methodologies, protocols, and rules-of-thumb can be devised, but ultimately the scenario cannot be known with certainty.
Many different tools have been developed attempting to improve the accuracy of additionality testing and to reduce administrative burden on the project developer and offset program administrator. There are two distinct approaches to additionality testing: Project based additionality and Performance Standards (such as Benchmarks).
Project-based Additionality Testing
Project-based additionality testing looks at the circumstances of each individual project and evaluates them on a case-by-case basis. It is the most accurate, yet also the most labor- and cost-intensive method. The following is a short selection of additionality tests that are commonly used:
Legal and Regulatory Additionality Test
A project can only be considered additional if it is not required to fulfil official policies, regulations, or industry standards. For example, an energy efficiency project might be implemented simply to meet building codes, in which case it would not be considered additional. If the project goes beyond compliance, it might be additional but more tests are required to determine that.
A project can only be considered additional if it is not profitable without revenue from carbon offsets. In other words, the revenue from the carbon offsets is a decisive reason for implementing a project. The financial test is consistent with a microeconomic view of behavior, and in theory would be a perfect additionality test. But in reality there may be projects whose finances make them look non-additional, yet they may still be "additional" because of non-monetary barriers.
A project can only be considered additional if there are barriers, such as local resistance, lack of know-how, institutional barriers, etc, that prevent its being implemented regardless of its profitability. If the project succeeds in overcoming significant non-financial barriers that the business-as-usual alternative would not have to face the project is considered additional.
Common Practice Test
A project can only be considered additional if it employs technologies or practices that are not already in common use. If the technology or practice is already in common use, then implementation as an offset project it presumed not to be necessary to carry out the activity.
Which test or combination of tests is best suited to validate additionality depends on the type of project. An additionality test for one type of project (e.g., a simple regulatory test for methane flaring, where there is no reason to do the project if not required by law) might not be sufficient for other kinds of projects (e.g., energy efficiency, where there could be plenty of reasons for doing a project besides complying with regulations).
Project based additionality tests often rely on information that is inherently difficult or impossible to confirm.
A Performance Standard is a “shortcut” approach to additionality and baselines. It does not attempt to undertake a project-specific inquiry into a project’s additionality, or to determine the specific baseline scenario for each project. Rather, it takes an approximate, aggregate approach. It establishes a generic baseline scenario against which all projects (of a given type) are assessed. This baseline takes the form of a quantitative performance standard − or “benchmark” carbon intensity per unit of output − specific to a given sector, e.g. a electricity carbon intensity in kgCO2/kWh defined for the power sector in China. Any project with emissions below this pre-defined benchmark is automatically deemed additional, and offsets are awarded based on the difference between the project emission rate and the benchmark emission rate.
The advantage of benchmark approaches is that they are simpler and more transparent to apply. They shift the workload from individual project hosts to a centralized entity that collects the necessary sector-specific data and makes a decision about the level at which to set the benchmark. Establishing a benchmark requires comprehensive data collection and verification, as well as regular updates. The political process to approve a benchmark may take a long time and it may only be feasible for certain industries, e.g. small renewable heat and power or small energy efficiency. Benchmark approaches are attractive because they are simple to apply. They may thus reduce cost and administrative burden for the project developer. Yet the devil lies in the details.
The main problem with benchmarks is that they may be too simple and broad. They are crude tools for determining additionality, for example, if the benchmark uses an emission rate as a proxy for determining additionality, all activities whose emissions fall below the benchmark emissions are awarded credits, regardless of whether they would have taken place anyway. If a benchmark is set at, say, the twentieth percentile relative to the range of emissions performance in a given sector, then one can expect that projects amounting to twenty percent of new activity in that sector may be eligible to generate offsets, all of which would be non-additional.
Observations On Baselines and Additionality
No matter how quantitative and objective it appears, any test will create some number of false positives (i.e. projects that appear additional despite the fact that they are not) and some number of false negatives (i.e. projects that appear non-additional despite the fact that they are). The design of the test determines if it will err on the side of false positives or false negative. Deciding which is more acceptable has to be determined through a political process. It is important to understand that while false positives and false negatives both impair economic efficiency only false positives undermine the environmental integrity of offsets. In other words, it is the false positives – offsets from non-additional projects – that lead to increases in emissions and therefore hamper climate protection goals.
Additionality tests can be cumbersome, time-consuming, and expensive. They are, however, necessary, because carbon offsets from non-additional projects sold into the market will actually lead to an increase in the buyer’s emissions, with no corresponding decrease in emissions from the seller, and hence a net increase in GHG emissions. If these projects are fully additional, then there will be a shift in emissions from the seller to the buyer, and zero net change in global emissions. The costs associated with rigorous offset programs are not merely “administrative burden” or “transaction costs” but rather production costs. They are legitimate costs associated with assuring the product has real value.