Data Challenges | Angela Druckman

The delights of data: deficiencies in the quagmire? Angela Druckman and Tim Jackson RESOLVE University of Surrey Carbon Accounting Conference Heriot-Watt University, Edinburgh 11 March 2009 www.carboncounting.org.uk

Plan What are we trying to measure? What methods do we use? What data do we need? What are the problems? Recommendations.

What are we trying to measure? Examples: (a) What are the carbon footprints of different types of households?

(b) What activities do we use our carbon for?

(c) What real progress is being made towards reducing the carbon footprint of UK lifestyles? The view from different perspectives:

Accounting from different perspectives Production Perspective Carbon emissions attributable to UK production; Used for reporting under Kyoto Protocol; Territorial basis; Includes exports, excludes imports; National Emissions Inventory. Consumption Perspective Carbon emissions attributable to UK consumption; Life-cycle approach; Includes imports, excludes exports; Environmental Input-Output Analysis

UK carbon emissions: trends from different accounting perspectives Example of some results obtained from the Surrey Environmental Lifestyle Mapping model (SELMA) Consumption perspective emissions estimated using a quasi-multi-regional input-output (QMRIO) model.

Trends in UK CO 2 emissions from different accounting perspectives Source: Druckman and Jackson 2008.

EIO is an economy-wide approach that is used to map carbon emissions through an economy for the purposes accounting from the consumption perspective. EIO is based on economic Input-Output Tables which show sales and purchases between industry sectors, and final demand consumption . The economic Input-Output Tables are combined with environmental data for each industry sector. EIO can be applied to: resource use such as oil, water, land-use etc. emissions such as CO 2 , greenhouse gases, waste etc. Environmental Input-Output (EIO) Analysis

Production sectors 1 2 3 n 1 2 3 n Final demand I III II Input-Output Framework (a) Adapted from: Moll et al 2004. .

Input-Output Framework (b) The 1st quadrant, called the production or transformation matrix, shows inter-industry flows, ie sales and purchases between industrial sectors. Technical coefficients which show the combination of products from each industrial sector required to produce the output of a sector can be calculated from this. This matrix, in theory, captures all inter-industry transactions to an infinite order. It provides the basis for obtaining information on all upstream resource flows of semi-manufactured and manufactured goods. Adapted from Moll et al 2004. .

Input-Output Framework (c) The 2nd quadrant, called the final demand matrix, shows which final demand category purchased the products and services. Final demand categories: Private households Government Capital investments Exports The 3rd quadrant, called the primary input matrix, shows the primary inputs (e.g. labour wages and profit, depreciation, taxes minus subsidies) required by each industrial sector in order to produce its output. Adapted from Moll et al 2004. .

Data requirements for Environmental Input-Output Analysis (single region) Final demand Technical coefficients: A-Matrix Carbon emissions attributed to final demand Carbon emissions per unit monetary output

Imports and exports Estimating the upstream emissions due to imported goods and services presents a challenge. Traditionally, EIO models assumed that the technology used for producing imports was the same as domestic technology (the “Domestic Technology Assumption”). Much work is currently focusing on developing input-output models in which emissions due to imports are estimated with improved accuracy.

Two-region input-output model with domestic technology assumption Region 1 Y A 1 u 1 Region 2 A 1 u 1

Quasi-multi-regional input-output (QMRIO) model Region 2 A 1 u 2 Region 3 A 1 u 3 Region 4 A 1 u 4 Region 5 A 1 u 5 Region 1 Y A 1 u 1

Multi-regional input-output (MRIO) model with uni-directional trade Region 2 A 2 u 2 Region 3 A 3 u 3 Region 4 A 4 u 4 Region 5 A 5 u 5 Region 1 Y A 1 u 1

Multi-regional input-output model with multi-directional trade Region 2 A 2 u 2 Region 3 A 3 u 3 Region 4 A 4 u 4 Region 5 A 5 u 5 Region 1 Y A 1 u 1

Data requirements For a two region model using Domestic Technology Assumption we need: UK Input-Output Table which includes Inter-industry flow matrix (A 1 -matrix) Final demand in basic prices (Y) Environmental emissions per unit monetary output (u 1 ) Additionally, for a full multi-regional model, data requirements include: Inter-industry flow matrix for each region (A n ) Environmental emissions per unit monetary output for each sector for each region (u n ) Trade flows between all sectors for all regions.

Sector disaggregation Every industrial sector is assumed to be homogenous with regard to its input requirements, the commodity it produces, and the emissions from the firms within the sector. Therefore the fewer the sectors, the more errors occur.

Data classifications EIO is fraught with data classification issues. For example UK economic datasets are provided in different classification from the environmental datasets. Different countries use different classification systems for their data. Cross-mapping between classification systems is therefore required and, as sectors often do not map directly, inaccuracies arise.

Data conversion Economic datasets are provided using different price bases (eg Purchasers’ v Basic Prices). Conversion is required – but often incomplete information is available for this. Modelling imports requires conversion between different currencies. 2 types of exchange rates: Market exchange rates Purchasing power parity (PPP) eliminates the differences in price levels between countries. statistical constructs rather than precise measures. may differ markedly from market exchange rates.

Availability of Input-Output Tables Production of Input-Output Tables is highly time-intensive and hence costly. They are therefore produced relatively infrequently. Technical coefficients are assumed to change slowly. Under EU regulations, member states are required to produce Input-Output Tables at 5 year intervals (European Commission 2007 ). The availability and disaggregation levels of data for different countries varies. UK: 123 industrial sectors. Scotland:127 sectors. Produced annually since 1998, latest table 2004. USA: >400 sectors; latest table 2002. International datasets: Global Trade Analysis Project (GTAP) EXIOPOL project

Recommendation Up-to-date economic and environmental data are required in a consistent , compatible and highly disaggregated format.

Recommendations specific to UK Up- to-date, authorised input-output tables are urgently required. Failing this, full information should be disclosed in order for tables to be produced by other organisations under supervision of the ONS. This information should include : Data on which to calculate final and intermediate demand in basic prices; High sectoral disaggregation; Detailed trade information ie imports and exports of goods and services to/from world regions/countries at sector level; Detailed Supply tables (minimum suppression or privileged access); Information on capital investment (gross capital formation) by sectors. These recommendations are largely in line with those made by Wiedmann et al (2008).

UK as a leader? http://www.ukwatch.net/files/GordonBrown.jpg

Data Challenges | Angela Druckman

More Related Content

What's hot (20)

Viewers also liked (9)

Similar to Data Challenges | Angela Druckman (20)

More from icarb (20)

Data Challenges | Angela Druckman

Editor's Notes