A METHOD TO CREATE CARBON FOOTPRINT ESTIMATES CONSISTENT WITH NATIONAL ACCOUNTS

^†The views expressed are those of the authors and do not necessarily reflect the policies of Statistics Netherlands or PBL Netherlands Environmental Assessment Agency.

Abstract

Although multiregional input–output (MRIO) databases use data from national statistical offices, the reconciliation of various data sources results in significantly altered country data. This makes it problematic to use MRIO-based footprints for national policy-making. This paper develops a potential solution using the Netherlands as case study. The method ensures that the footprint is derived from an MRIO dataset (in our case the World Input–Output Database (WIOD)) that is made consistent with Dutch National accounts data. Furthermore, usage of microdata allows us to separate re-exports at the company level. The adjustment results in a foreign footprint in 2009 that is 22% lower than the original WIOD estimates and a significantly altered country allocation. We demonstrate that already in the data preparation phase due to the treatment of re-exports and margins, large differences arise with Dutch national statistics, which may help explain the variation in footprint estimates across MRIO databases.

Keywords:

• Multiregional input–output analysis
• Carbon footprint
• Official statistics
• Uncertainty analysis

FUNDING

Part of this work was carried out in the e-Frame (European Framework for Measuring Progress) project funded by the European Commission's FP7 program [grant number 290520]. The authors gratefully acknowledge this funding.

SUPPLEMENTARY DATA

Supplemental material for this article is available via the supplemental tab on the article's online page at http://dx.doi.org/10.1080/09535314.2015.1048428

Notes

^† The views expressed are those of the authors and do not necessarily reflect the policies of Statistics Netherlands or PBL Netherlands Environmental Assessment Agency.

¹This article builds upon Hoekstra et al. (2012; 2013a; 2013b) and Edens and Hoekstra (2013).

²A number of statistical institutes have produced footprints, but these are generally less sophisticated than the MRIO-based calculations (see Hoekstra et al., 2014 for an overview).

³The export totals in the WIOD and Statistics Netherlands are related as follows: starting with the WIOD export data of 310 (panel 2(a)) billion euro we deduct the difference in re-exports (54) and difference in treatment of margins (28) and add the margins on re-exports (14) and obtain (after rounding) the Statistics Netherlands export value of 244 billion euro (panel 1(b)).

⁴As explained in Timmer et al. (2012) there are two types of IntSUT tables, we use the so-called analytical set.

⁵“Re-exports are goods that are transported via the Netherlands during which they are (temporarily) owned by a resident and that do not undergo significant industrial processing. When no change of ownership occurs we are dealing with transit trade” (Edens et al., 2011). When there is change of ownership, but no transportation via the Netherlands occurs we are dealing with merchanting (see 2008 SNA para 26.50-54 and A3.158; EC et al., 2009). According to the 2008 SNA, in case of goods sent for processing (without a change of economic ownership) only the service fees for processing are recorded as imports/exports, and not the value corresponding with the physical flows. This is different from the 1993 SNA which allowed for imputed changes of ownership in the case of substantial changes in goods.

⁶In fact, we have used the IO database in the ISIC rev. 4 breakdown, as for 2009 this is the only format available. At the 35 industries that WIOD discusses this causes not too many problems, although we did split up for instance the leather and clothing industries. As a result, the allocation to industries is less certain and not discussed in this article.

⁷This relatively simple reconciliation procedure could in the future be replaced by more advanced methods that take into account the differences in reliability of figures or the relation of figures over time (in case of the availability of multiple periods), such as the multivariate Denton method (see, e.g. Bikker et al., 2010), but this is beyond the scope of this research.

⁸The differences in country total output are all zero; the macro country totals for domestic produced intermediate consumption and domestic produced final demand differ less than 1% for all countries (on average 0.1% and 0.2%, respectively). The differences on macro totals for exports and imports are slightly larger averaging in both cases 0.5%.

⁹Although the carbon footprint was estimated at the time at a mere 187 Mton CO₂.

¹⁰Thanks are to Glen Peters for providing the data set with country footprints (Personal communication). The underlying data sources are: Peters et al. (2011), Peters et al. (2012), ESSD (Earth System Science Data) 2012 and 2013, Le Quéré et al. (2013), Eora, Lenzen et al. (2012), GRAM, Wiebe et al. (2012), WIOD, Timmer et al. (2012).