A water stress perspective on the UK’s water footprint

ABSTRACT

The global water requirements of economic activities have been studied through the concept of the water footprint, which can be captured in multinational Input–Output (IO) frameworks. While useful for looking at aggregate requirements, this paper links the UK’s blue and green water footprint with a country-level water stress index. We show that more than half of the UK’s blue water footprint was in areas of high water stress, while its green water footprint is more concentrated in areas of lower water stress. These findings show how water stress can be incorporated alongside measures of water footprint. This can be critical in evaluating the success of policies aimed to improve national and international water security and reduce water stress.

KEYWORDS:

• Water footprint
• water stress
• multi-regional input-output analysis

JEL CLASSIFICATION:

• Q56
• Q25
• C67

Acknowledgments

RL input to this research while an MSc student at the University of Strathclyde. Errors and omissions are the responsibility of the authors.

Disclosure statement

The authors have no interests to declare.

Notes

¹ These are, respectively, where more than 80% of the available water supply is withdrawn by agriculture, industry and municipalities or where over 40% of the available water supply is withdrawn (World Resources Institute 2019).

² For a recent review of virtual water research, see Sun et al. (2021).

³ At a global scale albeit with different regional and sectoral disaggregation, this includes the databases from WIOD, EXIOBASE and EORA, amongst others.

⁴ Blue water refers to water ‘water that has been sourced from surface or groundwater resources and is either evaporated, incorporated into a product or taken from one body of water and returned to another, or returned at a different time’ while green water refers to ‘water from precipitation that is stored in the root zone of the soil and evaporated, transpired or incorporated by plants’ (Water Footprint Network 2011).

⁵ This was presented in basic prices which is recommended for use in environmentally-extended IO analysis. The 26 sectors are given in Appendix 1.

⁶ Specifically, ‘households’, ‘non-profit institutions’, ‘government’, ‘investment’, ‘inventories’ and ‘acquisitions’.

⁷ Blue, green and grey water use is provided by the EORA environmental accounts, however we exclude grey water from the analysis.

⁸ While the huge range of countries in the EORA database exceeded the number of countries for whom Aqueduct had provided water stress index, this was not a major issue for the UK’s blue and green water footprints. With the identification of the UK’s water footprint from the EORA database, and then its allocation to countries, we were able to allocate 99.96% of the UK’s blue water footprint and 99.80% of the UK’s green water footprint in 2015 to a country with a baseline water stress measure. Our measure of water stress compares to those in other studies. For instance, Zhang et al. (2019) adopt a water stress indicator from Zhao et al. (2015) which calculates water stress at the province level in China. Our water stress index provides considerable international coverage which aligns well with the country level coverage of the EORA.

⁹ For instance, the World Bank measure of ‘level of water stress’ relates to 2014 for most countries and 2015 for some (World Bank 2021).

¹⁰ As we do our analysis for blue and green water, we use vectors for blue water use per unit of output and green water use per unit of output respectively. $\hat{\mathrm{\Omega }}$ denotes a diagonalized matrix with sectoral water-output coefficients on the diagonals and zeros elsewhere.

¹¹ Note that this necessarily means that the water embodied in the production of goods and services consumed outside of the UK is attributed to the consuming country and is not included in the UK’s water footprint.

¹² The blue and green water footprints are simply the summation of these 26 separate calculations.

¹³ Note that we are not seeking to explain a countries water stress levels – which will reflect a host of resource endowment, geographical, climate and infrastructural issues – by looking at the external demands for goods and services produced (using water resources) in that country, but to identify the location of water stress and the extent to which water stress areas provide inputs to the consumption of goods and services in different countries through the footprint calculation.

¹⁴ Our argument here is similar to that of Vanham and Mekonnen (2021) against a scarcity-weighted water footprint.