Ralf Becker United Nations Statistics Division

1. Adapting statistical classifications to green growth measurement Ralf Becker United Nations Statistics Division

2. Green growth is about fostering economic growth and development while ensuring that the natural assets continue to provide the resources and environmental services on which our well-being relies. To do this it must catalyse investment and innovation which will underpin sustained growth and give rise to new economic opportunities. • (OECD: Towards Green Growth, 2011)

3. Green growth encompasses two paths: "greening growth" and harnessing new growth possibilities from environmental considerations • Policies need to be supported by data • measurability becomes key point • combining data/indicators from different areas • existing indicators and new indicators

4. Economic and environmental data exist, frameworks exist (SNA, FDES) but: • it is often difficult to integrate them due to differences in • Classifications • Terminology (also within classifications) • Timeliness • Units (“natural” units vs. administrative units) • Needed: a consistent environment-economy accounting framework • With the SEAA and FDES, such systems are now being developed / updated • Classifications that link relevant concepts/categories

5. Within these frameworks, classifications of various components are being developed to support a general measurement • These are sometimes new classifications, targeted particularly at measuring environmental components or links to specific economic components • Example: classifications of environmental goods and services • Many of these will be useful for measurement issues related to green growth

6. Classifications used / developed within these systems: • Environment statistics • Classification of Water Use • Classification of Freshwater Quality for the Maintenance of Aquatic Life • Classification of Marine Water Quality • Classification of Land Use • Classification of Wastes • Classification of Ambient Air Quality • Classification of Flora, Fauna and Biotopes • Classification of Environment Protection Activities and Facilities • all are ECE classifications developed ~1990 • not pure classifications in the strict sense - multiple hierarchies, include measurement issues • have been used for international data collection extensively by ECE, OECD/Eurostat and UNSD

7. Classifications used / developed within these systems: • SEEA(currently under revision) • Classification of assets • Classification of physical flows • Classification of Environment Protection Activities and expenditures (CEPA) • Classification of Resource Use and Management Activities and expenditures (CRUMA) • Classification of Environmental Activities (CEA = CEPA + CRUMA) • Linkages between the environmental and the economic classifications facilitate the integration of data

8. Integrating data • Problems arise when data are collected / are available through existing programmes, using existing classifications • In such cases: • a link to green growth issues may not be directly possible • a link to classifications developed for such purposes may be difficult • This applies to most economic data (e.g. production, trade) • Example: Production / trade of goods / prices • uses classifications like CPC, CPA, Prodcom, HS • detailed information collected from producersetc.

9. Adapting classifications • Can existing economic classifications be adapted to suit the needs for data on green growth? • Key question: What information related to green growth do we want to reflect in the classification? • What criteria do we use to identify relevant industries/products/expenditures/programmes etc.? • Are these criteria compatible with the principles of the classification? • Are resulting classifications still applicable for their original purposes? • How are links to other classifications affected?

10. Existing efforts: • Identify “environment-related products” in existing classifications • E.g. done by WTO (using Harmonized System) • Synthesis of submissions on environmental goods (TN/TE/W/63) • Based on certain characteristics of the final product, typically related to use • Does not (yet) affect the classification itself • Reflects an “alternative aggregation” • Preferable method if necessary product detail exists

11. Can products already be classified as "green products" or not? • Depends on definition of "green product“ • if all inputs are to be considered, this will be near impossible • A decision on whether or not a product is “green” will depend not only on its characteristics, but also on the inputs and processes used to produce it • Reflecting the "greening growth“ component • Example: aluminum refining • requires electricity • electricity could come from a solar power plant or coal fired plant • the characteristics of the product itself are not affected • Do we want to make a distinction based on the source of electricity? • Where can this be observed? Where/how can the classification be applied? (e.g. point of production, trade)

12. Example: • Manufacturing of a good resulting in waste as a by-product • The process can reuse the waste, treat the waste or just discard the waste • the characteristics of the produced product are not affected • Do we want to make a distinction based on impact of the product’s production on the environment? • Example: • Carbon footprint • Can be measured for each product (maybe), but can a distinction be embedded in the classification?

13. Reasons for caution: • It may be possible to define detailed categories in a classification that correspond to data needs, but: • Are these categories still observable? • E.g. for activities – does sufficient specialization exist? • Are existing rules for the application of the classification supportive of the desired purpose • E.g. rules for treatment of multiple activities or bundled products or multi-purpose expenditures

14. Classifications (even the multi-purpose type) are usually designed for describing objects in certain settings • E.g. at point of production, point of trade or point of consumption • Applying the classification at a different point may create measurement issues • E.g. carbon footprint issue in trade • “Eco-labeling” may help • Depends on local regulation -> comparability issues • May define a larger scope than appropriate for green growth

15. Classifications affected • Traditional classification families that are potential targets for review under a “green growth” agenda: • Industry/activity • Products • Occupations • Education • R&D programmes • Financial flows

16. Additional considerations: • International Standard Classifications • Require stability over time • Should be applicable to most countries • Smaller countries may not be able to use detailed categories of the classification • Limits are often tested with current classification details • Results: • Use of higher levels as proxies • Use of different aggregations • Loss of comparability of data

17. Conclusions • Information on green growth requires data from different fields, based on solid classifications • Existing classifications can be exploited through use of alternative aggregations, if necessary detail exists, to provide standard definitions/aggregates • Introducing new concepts in existing classifications requires caution • Introducing additional detail in existing classifications requires considerations of conceptual consistency, measurability and of the targeted users of the classification (e.g. data providers) • Not all new data needs require a change in classification