Greenhouse Gas Emissions Forecasting: Learning from International Best Practices

Successful climate policies are those that achieve forecasted greenhouse gas (GHG) emissions reductions. But, forecasting expected GHG emissions reductions from specific policies and measures is difficult, as noted by the National Round Table on the Environment and the Economy (NRTEE) in its Response to its Obligations under the Kyoto Protocol Implementation Act (2007 KPIA Response). ^[3] It has not only been a challenge for Canada, but for all countries to generate accurate projections of expected GHG emissions reductions from policies over a given time period. However, Canada has consistently produced inaccurate forecasts over the past 10 years, in each instance underestimating the growth of domestic GHG emissions under business-as-usual (BAU) conditions.^[4]

The importance of accurate forecasting cannot be overstated. It is the forecast upon which climate policies and programs are developed and measured. Policy makers can be driven to differing policy choices depending upon the forecasted emissions reductions expected to be realized. Inaccurate estimates make it difficult for the federal government to design policies that will result in expected forecast emissions reductions. Some countries, particularly those with similar resource-based economies to Canada’s, have had more success in forecasting GHG emissions reductions in some policy areas. While no country produces forecasts that are 100 per cent accurate, some countries utilize approaches that appear to be more reflective of actual emissions events. Therefore, it is well worth examining and understanding how these countries approach their GHG emissions projections, both from methodological and governance perspectives, and then assessing whether and how these might be beneficially applied to Canada.

2.1 Background and Purpose

In its review of the government’s climate change plan under its KPIA obligations in summer 2007, the NRTEE found that differing and inconsistent forecasting methods were used among various federal departments to describe the emissions reductions accruing from a particular initiative, leading to issues of additionality, free ridership, rebound effect, and policy interaction effects.^[5] It is important to note that in its just-released 2008 KPIA Plan, the government has taken significant steps to address these issues.^[6] The areas of concern identified by the NRTEE in its evaluation of the government’s 2007 KPIA climate change plan are primarily those of methodology and governance. In its 2007 KPIA Response, from which this report originated, the NRTEE emphasized the importance of transparency and clarity with respect to key assumptions and methods, and the consideration of important sensitivities and uncertainties. It also emphasized the importance of consistency in approaches across different departments, programs, and measures, and the need to integrate the findings in a holistic framework. In light of these conclusions, the NRTEE felt it could be useful for the federal government if international best practices could be identified and highlighted in the forecasting of emissions reductions resulting from government policies, from both a methodological and a governance perspective.

At its November 2007 plenary meeting, the Round Table accordingly approved a proposal to develop a best practices guide or backgrounder that would be submitted to the government along with the NRTEE’s next response under its KPIA obligations in 2008.

From a methodological perspective, the NRTEE identified certain forecasting methods for estimating GHG emissions reductions that did not result in the expected realized emissions, as the difference between stated reductions and reference case (or BAU) emissions. The sum of stated reductions for a given year should correspond to the expected difference between the reference case and forecasts of realized emissions. As noted above, deviating from the basic computation of measuring reductions against the reference case can lead to issues of additionality, free ridership, rebound effect, and policy interaction effects.

From a governance perspective, the NRTEE identified inconsistent forecasting methods across departments for estimating GHG emissions reductions and gaps in the on-going evaluation of forecasting approaches. The NRTEE also found that a consistent definition for "reductions" for policy impacts was not applied in all instances in the plan--some policy impacts were stated in different terms (e.g., in terms of their cumulative impact). Using a consistent model and definitions for forecasting the emissions reductions resulting from specific programs and initiatives is necessary to address this.

For both the methodological and governance areas of concern listed above, it is important to note that in its 2008 KPIA Plan, the government has taken steps to address these issues. Nevertheless, the NRTEE believes the application of certain best practices, as outlined in this report, will ensure more accurate forecasting moving forward and, by extension, inform effective policy choices that will achieve GHG emissions reductions.

2.2 Analytical Approach

In order to ensure optimal climate policy outcomes, accurate and realistic forecasting is vital. The NRTEE believes this report could be useful in assisting the government to develop effective policy by providing examples of how other governments forecast GHG emissions reductions from policies and measures.

Flowing from the key findings of the Round Table’s 2007 KPIA Response, the approach to this report has been to not only define and provide examples of international best practices from a methodological perspective, but also to highlight best practices from a governance perspective.

Specifically, we sought answers to two key questions:

• Methodology: What is the most effective forecasting methodology that countries should use?
  
• Governance: What are the optimal forms of governance to ensure that best practices in forecasting are followed?

A case study approach was chosen as most appropriate. To the extent possible, countries with economic or national situations most applicable to Canada have been examined. It is important to note that the selection of case studies to provide examples of international best practices in GHG emissions forecasting is not as straightforward as it seems. The original intention was to select countries with similar challenges to those facing Canada in forecasting GHG policies and that followed best practices in methodology and governance as outlined in sections 4.1 and 4.2. What these case studies highlight is that some countries utilize best practices from a methodological perspective and some from a governance perspective, but not all utilize best practices in both areas.^[7] Another initial purpose of this report was to highlight best practices in sub-national jurisdictions, particularly some Canadian provinces and U.S. states. No province, however, has released detailed plans with emission policy forecasts,^[8] raising questions about the methodological basis behind their plans.

In order to ensure rigour in our research, analysis, and findings, this paper was then peer reviewed by several well-respected economic experts.^[9] Beyond the peer review, the final report was reviewed and considered by NRTEE members themselves.

3 For further information, see http://www.nrtee-trnee.ca/eng/publications/c288-response-2007/index-c288-response-2007-eng.htm.

4 Simpson, J., Jaccard, M. and Rivers, N. (2007). Hot Air: Meeting Canada’s Climate Change Challenge. Toronto: McLelland and Stewart, p.165.

5 Please see Appendix B for a description of these effects.

6 Environment Canada (2008), A Climate Change Plan for the Purposes of the Kyoto Protocol Implementation Act, p. 30.

7 This issue is further discussed in section 4.4.1.

8 This issue is elaborated upon in section 4.1.

9 Please see Notice to Reader for details.