The challenge of the Paris Agreement is "to achieve a balance between anthropogenic emissions by sources and removals by sinks of greenhouse gases in the second half of this century". It's a 50-year marathon, made up of successive five-year sprints, and we need to approach it as such.

In the rush to create the Paris Agreement and submit their Intended Nationally Determined Contributions (INDCs), Parties have understandably focused on the five-year event and overlooked the fact that the next five-year event is to be harder, and the one after that harder still. It is not sensible to embark upon this marathon without a long-term plan, to ensure that efforts taken in the early stages do not compromise performance in the later stages. And as discussed below, this is a particular risk to developing countries.

To effectively fulfill their commitments under the Paris Agreement, countries around the world know that they must not only submit an initial Nationally Determined Contribution (NDC) but continue to improve on that commitment through periodic submission of increasingly ambitious NDCs. But to avoid locking in the wrong technologies, Parties need to base each successive NDC on a long-term plan as proposed under Article 4 paragraph 19: "All Parties should strive to formulate and communicate long-term low greenhouse gas emission development strategies... ."

These long-term strategies are critical to the process of planning a low-emission economy, but so far countries are at risk of focusing on creating one-off NDCs separate from a well-articulated long-term plan. The failure to elaborate a good long-term plan may result in uncoordinated successive NDCs where an investment in an asset today may seem like a good way of achieving the initial NDC commitments but may in fact start to become a liability in later NDCs.

The chances of this happening are increased when we focus our NDC ambitions on the calculation of greenhouse gas (GHG) emission reductions, particularly in developing countries. These calculations are based on accounting experience gained under the Kyoto Protocol, where estimates of a Business As Usual (BAU) baseline and actual project emissions were used to calculate emission reductions. The concept of setting a baseline and calculating reductions compared to that baseline has been carried into the first round of NDCs whereby, in many developing countries, economy level projections of BAU emissions are made to 2030 or another future date, and targets are set as a reduction against that BAU projection.

For example, at the project level, a developer can argue that conventional coal is a realistic BAU scenario because there are no laws prohibiting it; and, therefore, a supercritical coal fired plant which reduces emissions relative to a conventional coal BAU seems like a good option because it reduces GHG emissions. While we can debate the reliability of the baseline, the bigger questions are what kind of infrastructure will be needed to generate electricity for the economy in 20 or 30 years' time, and is this supercritical plant part of that infrastructure?

At the national level, setting a target based on emission reductions against a projected BAU scenario runs the risk of encouraging Parties to plan to meet their emission reduction targets by summing up the individual project or sectoral contribution to emission reductions. This carries the same risks as the project based approach. First, if the baseline is not correct then the target may be missed and second, it ignores the long-term goal of net zero emissions.

Furthermore, this is a particular risk for developing countries, where NDC targets are generally expressed against projections of future emissions. In most OECD countries the problem does not arise because their targets are expressed against historic emission levels (e.g., 1990 or 2005).

So, to help countries avoid building assets that meet the current NDC but become a liability in 15 or 20 years' time, we need to model the development of economies and prepare projections about the amount of resources a given economy will need in the future. Only by estimating the amount of electricity the economy will require in 2050, and allocating a proportion of national 2050 emissions to the energy sector, can we then start to describe the generating fleet, and in particular the average emission factor per MWh of energy that the fleet must meet.

For example, let us assume that macro-economic models estimate that a given economy will need 100 Terawatt hours of electricity to operate in 2050, and assume that under the country's plan to achieve net zero emissions by 2080, the energy sector will have an allocation of 10 million tonnes CO2e per year in 2050. That tells us that the average emission factor for power generation will need to be 0.1 tonnes per MWh or less. Under this scenario, an investment in a supercritical coal fired power plant (emitting 0.7 - 0.8 tonnes per MWh) which helped achieve the NDC in 2025 will become a major handicap for the 2050-55 NDC. Putting it bluntly, the plant may become a stranded asset which the economy cannot afford to operate.

The same analysis needs to be done for each economic sector, based on estimates of quantities of goods and services which the economy will require, indicative allocations calculated and plans made to develop appropriate infrastructure. It is emissions (or carbon footprints) that matter, not emission reductions. NDC targets need to be set in absolute emissions and, while percentage reduction against BAU can be used as an indicator of ambition, we should understand that this is highly context-specific and not a very reliable indicator of effort.

In conclusion, Parties need to think about the long-term goal and work back from there. A quick start is important because there is low-hanging fruit which need to be cleaned up as soon as possible. A strong case can be made, for example, for bolstering baseload electricity in order to support more renewable energy. The rules allow for some peaking before the sustained decrease kicks in, but thereafter, each five-year period will get tougher. Those who over-reach and commit themselves in the early stages of the race may find their knees weakening as the gradient increases!