Water Footprint and Virtual Water Trade in Spain doc

This analysis can provide a transparent and multidisciplinary framework for informing and optimising water policy decisions, contributing at the same time to the implementation of the EU

Water Footprint and Virtual Water Trade in Spain

David Zilberman

Dept of Agricultural and Resource Economics

University of California, Berkeley

Department of Agricultural Economics and Social Sciences

Technical University of Madrid, Spain

EDITORIAL STATEMENT

There is a growing awareness of the role that natural resources such as water, land, forests and environmental amenities play in our lives There are many competing uses for natural resources, and society is challenged to manage them to improve social well being Furthermore, there may be dire consequences to natural resources mismanagement Renewable resources such as water, land and the environment are linked, and decisions made with regard to one may affect the others Policy and management of natural resources now require an interdisciplinary approach including natural and social sciences to correctly address our societal preferences.

This series provides a collection of works containing the most recent findings on economics, management and policy of renewable biological resources such as water, land, crop protection, sustainable agriculture, technology, and environmental health It incorporates modern thinking and techniques of economics and management Books in this series will combine knowledge and models of natural phenomena with economics and managerial decision frameworks to assess alternative options for managing natural resources and the environment.

The Series Editors

For other titles published in this series, go to

www.springer.com/series/6360

Water Footprint and Virtual Water Trade in Spain

Policy Implications

Department of Agricultural Economic

and Social Sciences

Technical University of Madrid (UPM)

28040 Madrid

Spain

alberto.garrido@upm.es

Consuelo Varela-Ortega

Department of Agricultural Economic

and Social Sciences

Technical University of Madrid (UPM)

28040 Madrid

Spain

consuelo.varela@upm.es

Roberto Rodríguez-Casado

Department of Agricultural Economic

and Social Sciences

Technical University of Madrid (UPM)

Department of Agricultural Economic and Social Sciences

Technical University of Madrid (UPM)

28040 Madrid Spain

paula.novo@upm.es Maite M Aldaya University of Twente

7500 AE Enschede Netherlands m.m.aldaya@ctw.utwente.nl

ISBN 978-1-4419-5740-5 e-ISBN 978-1-4419-5741-2

DOI 10.1007/978-1-4419-5741-2

Springer New York Dordrecht Heidelberg London

Library of Congress Control Number: 2010923233

© La Fundación Marcelino Botín-Sanz de Sautuola y López 2010

All rights reserved This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY

10013, USA), except for brief excerpts in connection with reviews or scholarly analysis Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar

or dissimilar methodology now known or hereafter developed is forbidden.

The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject

to proprietary rights.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

To the co-authors, and in particular

to my mentor, Ramón Llamas

M.M.A.

To Josemaria Escrivá, his example and writings have been a beacon in my work

M.R.L.

Acknowledgements

The authors would like to acknowledge the continuous support of Dr Rafael Benjumea, Director of the Fundación Marcelino Botón (FMB) up until the very last minute the book was sent to press He has been an inspirational source for our work and a tough challenger for some of the ideas developed in this book After numer-ous lengthy discussions, he eventually began to understand why our contributions may offer a fresh look into one of the most thoroughly analysed water economies

in the world Mr Federico Ysart, Director of the Trends Observatory of the FMB, was also an active discussant in the process that led to the FMB’s decision to publish this book We received constant encouragement and valuable service from

Ms Esperanza Botella, FMB’s Deputy Director

The authors would also like to acknowledge Dr Elena López-Gunn’s generosity for reading thoroughly Chap 8 and making valuable suggestions

Finally, we must also acknowledge Profs Anthony Allan (Kings College, England), Mordechai Shechter (Haifa University, Israel) and Arjen Hoekstra (Twente University, The Netherlands) for reading the first version of the manuscript and participating in a seminar held in Madrid in October 2008 Part of the value of this book is due to their critical reading and their valuable comments Some of its weaknesses were not over-looked by them, as in fact they were clearly indicated, but we could not sort them out without delaying more than desired the submission of the volume

Lastly, we also wish to acknowledge Jon Gurstelle, our Springer editor, for believing in our project and helping us with the technical (and legal) details in the publishing stage

Contents

1 Introduction 1

1.1 General Framework 1

1.2 Objective 4

2 Literature Review 7

2.1 The Concept of Virtual Water 7

2.2 The Colours of Water 8

2.3 International Virtual-Water “Trade” 9

2.4 Water Footprint Concept and Its Relation to Economic Growth 11

2.4.1 Scale Effects 13

2.4.2 Sectoral Composition 14

2.4.3 Technological Change 15

3 Methodological Approaches 17

3.1 Introduction 17

3.2 Water-Footprint Calculation 18

3.3 Internal Water Footprint 18

3.3.1 Crop Water Use 19

3.3.2 Livestock Water Use 21

3.3.3 Industrial and Urban Water Use 22

3.3.4 Virtual-Water “Exports” 22

3.4 External Water Footprint 24

3.4.1 Virtual-Water “Imports” 24

3.5 Virtual-Water “Flow” 24

3.6 Apparent Productivity of Water and Land 25

3.7 Economic Value of Water 25

3.8 An Econometric Approach 29

3.8.1 Explaining Water Productivity by Water Scarcity and Water Quality 29

3.8.2 Explaining Blue Virtual-Water “Exports” by Water Scarcity and Water Quality 31

3.8.3 Explaining Water Scarcity by Water Quality 32

3.8.4 Exchange Terms of Virtual-Water “Trade” 33

3.8.5 Water Quality Valuation 33

4 Data and Limitations 37

4.1 Data sources 37

4.2 Limitations 39

5 Spain’s Water Footprint 41

5.1 Agricultural Water Footprint 43

5.1.1 Water Footprint: Irrigation and Water Demand 45

5.1.2 Economic Aspects of the Water Footprint 48

5.2 Livestock Water Footprint 53

5.2.1 Livestock Sector’s Water Footprint 55

5.2.2 Water Footprint of Animal Feed Production 57

5.2.3 Economic Aspects of the Water Footprint 63

5.3 Industrial Water Footprint 64

5.3.1 Water Footprint 65

5.3.2 Economic Water Footprint 67

5.4 Urban Water Footprint 67

5.4.1 Water Footprint 67

5.5 The River Basin Scale: The Water Footprint of the Guadiana 68

5.5.1 Guadiana Water Footprint 68

5.5.2 Agricultural Water Footprint in the Guadiana Basin 70

5.5.3 Water Footprint of Irrigated Crops (m3/ton) 70

5.5.4 Economic Aspects of the Water Footprint 74

6 Net Virtual-Water “Flows” 77

6.1 Virtual-Water “Imports” 79

6.1.1 Major Crop-Related Virtual-Water “Imports” 80

6.1.2 Economic Valuation of Virtual-Water “Imports” 81

6.2 Virtual-Water “Exports” 83

6.2.1 Virtual-Water “Trade” 87

6.2.2 Economic Virtual-Water “Trade” 88

6.3 Virtual-Water “Trade” Within the Guadiana Basin: The Regional Scale 90

7 Bringing the Analysis to the Policy Context 95

7.1 Changes in Land Productivity 96

7.1.1 The Ebro Basin 97

7.1.2 The Duero Basin 97

7.1.3 The Guadalquivir Basin 100

7.1.4 The Júcar Basin 100

7.1.5 The Tagus Basin 102

7.1.6 The Guadiana Basin 102

7.1.7 The Sur and Segura Basins 102

7.2 Water Scarcity, Allocation and Economic Efficiency 104

7.3 Water Productivity in Light of Water Scarcity 112

7.4 Explaining Water Scarcity by Water Quality 118

7.5 Virtual-Water “Trade” as an Adaptation to Climate Change 119

7.6 Economic Growth, Water Footprint and Water Exchanges: Can Growth Be Decoupled from Water Use? 121

8 Summary and Conclusions 125

8.1 Virtual Water and Water Footprint of Spain 126

8.2 Water Allocation in Light of Virtual Water 128

8.3 Re-Thinking Water Scarcity Paradigms 130

8.4 Lessons Learned at the River Basin Scale: Guadiana Case Study 132

8.5 Lessons Learned and Avenues for Further Research 134

Glossary 137

References 143

Index 151

A Garrido et al., Water Footprint and Virtual Water Trade in Spain: Policy Implications,

Natural Resource Management and Policy 35, DOI 10.1007/978-1-4419-5741-2_1,

© La Fundación Marcelino Botín-Sanz de Sautuola y López 2010

1.1 General Framework

In most arid and semi-arid countries, water resource management is an issue that is both important and controversial Most water resources experts now acknowledge that water conflicts are not caused by physical scarcity but are mainly due to poor

among others) The scientific and technological advances of the past 50 years have led to new ways to solve many water-related conflicts, often with tools that seemed

This study deals with the estimation and analysis of Spain’s water footprint, both from a hydrological and economic perspective Its ultimate objective is to report on the allocative efficiency of water and economic resources This analysis can provide

a transparent and multidisciplinary framework for informing and optimising water policy decisions, contributing at the same time to the implementation of the EU

mandate of the Spanish Ministry of Environment and Rural and Marine Affairs, which recently issued instructions for drafting river basin management plans in compliance with the EU Water Framework Directive, with a deadline of end of

The water footprint (WF) is a consumption-based indicator of water use

as the total volume of freshwater that is used to produce the goods and services

Closely linked to the concept of water footprint is the virtual-water concept (VW) The virtual-water content of a product (a commodity, good or service) refers to the

on this concept, virtual-water “trade” represents the amount of water embedded in

under-standing of globalisation is whether international trade can save water globally In principle, it does if a water-intensive commodity is traded from an area where it is produced with high water productivity (resulting in products with low virtual-water

Introduction

content) to an area with lower water productivity (Hoekstra and Chapagain 2008)

saved through virtual-water “trade” in agricultural commodities alone Nevertheless, the relevance of global water savings needs a more detailed study, because savings represent only about 5% of the global water footprint and the uncertainties and limitations of the estimations may be greater than this 5% Although virtual-water

“trade” evaluations have taken countries or even bigger regions as the trading partners, the concept can also be applied within countries and even river basins In fact, this

is the dual perspective of this study

At the national or regional level, a nation can preserve its domestic water resources by importing products instead of producing them domestically This is particularly relevant to arid or semi-arid countries with scarce water resources such

as Spain As this study explains, Spain imports water-intensive low-economic value crops (mainly wheat, maize and soybeans and soy products), while it exports water-extensive high-value commodities adapted to the Mediterranean climate, essen-tially olive oil, fruits and vegetables However, most countries, including Spain, import and export the same or very similar commodities, with trade flows that vary

by season, specific varieties and market trends of supply and demand Because water is not the main input in virtually all traded goods, water scarcity and supply costs are poor explanatory factors of virtual-water “trade”, except in very special contexts As basic resources such as water and energy become increasingly scarce, the potential for international trade as a way to promote efficient use of these resources becomes more policy relevant While virtual-water “trade” cannot

be considered as the primary motivation for commodity trade, one can always test whether virtual-water “trade” can enable or facilitate more efficient water allocation among competing ends

In addition to its potential contribution to water savings, it is also important to establish whether the water used originates from rainwater evaporated during the production process (green water) or surface water and/or groundwater evaporated

Traditionally, emphasis has been paid to the concept of blue water through the

“miracle” of irrigation systems However, an increasing number of authors

virtual water (both green and blue) used in the different economic sectors could facilitate more efficient allocation and use of water resources, globally, nationally

or locally, while providing a transparent interdisciplinary framework for policy formulation Furthermore, the Achilles’ heel of the current emphasis of rainfed agriculture (green water) is climate variability, which will increase, as most studies

In order to mitigate drought episodes, water works such as dams and canals have been built, and wells have been drilled to complement surface water supplies In the last half century, however, there has been a silent revolution in groundwater-irrigated

Llamas and Martínez-Santos 2005; Shah et al 2007; Villholth and Giordano 2007)

As a matter of fact in some countries, mainly in India, groundwater development is

While rainfed crops depend only on meteorological conditions, irrigated crops depend both on rain regimes and water supply The combination of these regimes and the interdependencies between international commodity markets and domestic production create opportunities to ensure that water is allocated to the most valuable ends

This book mainly deals with Spain’s water footprint and offers a virtual-water analysis that differentiates green and blue (surface and groundwater) components, both from a hydrological and economic perspective It looks at the potential of these concepts in helping achieve an efficient allocation of water resources First of all, it defines the concepts of virtual water, the colours of water, virtual-water

“trade” and the water footprint and analyses the impact of economic growth on the latter A glossary with key terms is included at the end of the document The study then explores the different economic sectors in detail at the national, provincial and river basin levels Special attention is given to crop production that accounts for about 80% of the total consumptive use (or water footprint) of use of green and blue water resources This is followed by assessments of the footprints of livestock, industry, energy and urban water use Virtual-water “trade” is evaluated both within the EU and with third countries Finally, the policy implications of this analysis are assessed A better knowledge of the water footprint and virtual-water “trade” in Spain and in other arid and semi-arid countries can be very useful for developing a comprehensive instrumental framework across time and space to support water management decisions Ultimately, this knowledge-based tool can be used by the water authorities to achieve a more efficient allocation of water resources Spain has already largely adopted the “more crops and jobs per drop” paradigm, but it struggles to achieve the new goal of “more cash and nature per drop”, because water productivity in many areas of the economy is already high Furthermore, the literature has rarely considered the actual opportunity cost of the water that is used and exported in virtual form For countries suffering continuous water shortages, this poses a serious limitation to drawing policy-relevant conclusions from the concepts of water footprint and virtual-water “trade” In this respect, the generally higher economic efficiency of groundwater irrigation deserves a more thorough analysis, expanding on the earlier assessment of Andalusian irrigation (Hernandez-

For the time being and in almost the entire world, water footprint analyses have focused on hydrological aspects, based on volumetric evaluations A significant innovation of this work is to emphasise the imperative challenge of considering economic and ecological factors, with the aim of moving towards a policy that will enable to balance the trade-off between water for nature and water for rural liveli-hoods, that is to seek for “more cash and nature per drop” Water footprint analyses provide new data and perspectives for a more optimistic outlook on the frequently cited looming “water scarcity crisis” This new knowledge is changing traditional water and food security concepts that most policy makers have held until now

1.2 Objective

The objective of this study is to assess and analyse Spain’s virtual-water “trade” (VW) and water footprint (WF), differentiating the green and blue (surface and groundwater) components, both from a hydrological and economic perspective The research program that provided the results reported in the following chapters was envisioned and designed with the following criteria:

1 A multi-layered perspective – international, national and regional (basin level) is needed to understand and analyse a country’s water policy The geographical analysis casts light on regional controversies lived in Spain since 2000

2 As water use and productivity change over time and vary geographically, a wealth of interpretative data can be gathered, analysed and placed in a global context (both as a cause and an effect of the observed changes at the national level)

3 Agriculture being the largest water consumer, it is of utmost importance to understand how green and blue water components vary with time and from place

to place This variation has implications for water productivity, water allocation and drought management, which in turn are linked to international trade

4 Water is an economic good and provides market and non-market services

“motion pictures” featuring the water footprint and virtual-water “trade” that we are aiming to produce in this study This criterion is entirely consistent with the approach of the WFD and the most recent trends in Spanish water policy.With these points in mind, this study aims to contribute to the WF and VW literature

By evaluating both WF and VW over time and at the provincial scale, the analysis

•

allows for policy-relevant conclusions at the river basin level

By separating green and blue water components and evaluating all crops at the

•

provincial level, the study enables a finer analysis of how WF and VW vary during droughts and water shortages as well as during wet periods The linkage between commodity trade and water scarcity will be explored to determine the extent to which virtual-water “trade” has the potential to deal with water-stressed periods This is a crucial factor for water management in arid and semi-arid countries

By also evaluating WF in terms of m

better integrated in the analysis This provides a distinctive view of WF and allows for a closer linkage between water productivity and water scarcity, in physical and economic terms

Water scarcity is evaluated in terms of opportunity cost, both for virtual-water

•

“trade” and WF, which in this study is corrected with the water quality status of the rivers in each province This analysis, therefore, includes both market and non-market dimensions

This research study mainly builds on an earlier study by Chapagain and

2001 period This study, however, covers 1997–2006 and analyses the Spanish water footprint variations from year to year, not only at a national but also at provincial and river basin levels In both studies, water footprints are assessed following the top-down approach A significant innovation of this work is to emphasise the challenge of considering economic aspects Concerning the spatial dimension, this study explores the different sectors at the national, river basin and provincial levels Furthermore, it refines the methodology of earlier

approach to the Spanish context Results obtained by Rodríguez Casado et al

Finally, an open debate is necessary both on the concept of VW and WF and on the available data This report hopes to make a down-to-earth contribution to this debate through up-to-date, detailed evaluations that enable a closer evaluation of the water footprint and virtual-water “trade” This study will also help explain the roots of regional water conflicts and the role of water markets, through a detailed geographical analysis of water productivity changes across provinces and throughout the study period

-0 - 5-0-0 Million m 3

???

????? 2,500

-4,000 1,500 - 2,500 500 - 1,500

0 - 500 Million m 3

???

????? 2,500

-4,000 1,500 - 2,500 500 - 1,500

0 - 500 Million m 3

Inference potential:

Global Sustainability External

Policy implications:

Water footprint in Spain

Blue water

Green water

Trade Water Policy Sustainability

Time variation (policy, technology change, trade)

Exports Imports Spatial variation (climatic, water conditions)

Year t+2 Year t+1 Year t

Fig 1.1 Schematic description of the project

2.1 The Concept of Virtual Water

volume of water used to produce a commodity, good or service This term can be defined from two distinct perspectives From the production-site standpoint, the virtual-water content of a product is the volume of freshwater used to produce a

From the consumption-site standpoint, it refers to the volume of water that would have been required to produce a product where it is consumed (Hoekstra and

The present study uses the first definition The adjective “virtual” refers to the fact that most of the water used to produce a product is not contained in it; the real-water content of products being generally negligible compared with the

to produce a crop mainly depends on climatic conditions, water management options and agricultural practices While significant research is being done to find

significant changes in water demand can occur, changing the land that receives irrigation water and the crops that are irrigated In the EU, cropping patterns have

now, due to more decoupled modes of farm income support, EU farmers are responding more to market signals And most of these originate from global markets, offering broad opportunities to exploit the connections between food markets and farm trade and water policies

As this book will explain, water shortages and scarcity result from endogenous processes linked to policies and consumption that promote water demand which in turn results in bigger water footprints One of the contributions of this study is to think of virtual water not only as the physical amount of resource embedded in the consumed and traded goods, but also as an economic good with opportunity cost that varies over time and according to quality and location Not all virtual water that is traded – for example, in wheat, oil, meat or automobiles – is equally valuable

Literature Review

A Garrido et al., Water Footprint and Virtual Water Trade in Spain: Policy Implications,

Natural Resource Management and Policy 35, DOI 10.1007/978-1-4419-5741-2_2,

© La Fundación Marcelino Botín-Sanz de Sautuola y López 2010

2.2 The Colours of Water

The virtual-water content of a product consists of three components: green, blue and grey water For the purpose of policy formulation, it is essential to distinguish the

First, the green virtual-water content of a product is the volume of rainwater that

particularly relevant for agricultural products, where it refers to the total rainwater stored in the soil as soil moisture and evaporated from the field during the growing period of the crop (including both evapotranspiration by the plants and evaporation from the soil)

Second, the blue virtual-water content of a product is the volume of surface or ground water that evaporated as a result of its production (Hoekstra and Chapagain

the sum of the evaporation of irrigation water from the soil and the evaporation of water supplied from irrigation canals and artificial storage reservoirs In industrial production and domestic water supply, the blue water content of the product or service is equal to the part of the water withdrawn from ground or surface water that evaporates and thus does not return to the system it came from Evaporated water is considered unavailable for other uses, even though it may come back as rainfall (usually hundreds of kilometres away) Many irrigated crops are also receiving some rainfall, so total water demand is often satisfied by a mix of natural and artificial sources Furthermore, the amount of blue water demanded for irrigation varies because weather conditions vary significantly A technical evaluation in Andalusia (with almost 900,000 ha of irrigated land) found that crop blue water

The distinction between green and blue water originates from Falkenmark

automatically re-allocated to uses other than natural vegetation or alternative rainfed crops, whereas blue water can be used for irrigating crops and also for other urban,

Furthermore, the use of green water in crop production is considered more sustainable

water sources are exploited below their sustainable yield In the semi-arid and sub-humid regions of the world, water is a key challenge in food production, due

to the extreme variability of rainfall, long dry seasons and recurrent droughts, floods and dry spells The key challenge is to reduce the water-related risks posed

by high-rainfall variability rather than coping with an absolute lack of water

generally enough rainfall to double and often even quadruple crop yields in rainfed farming systems, even in water-constrained regions (Comprehensive Assessment of

causing dry spells and much of this improvement is lost The focus of the past 50 years on managing rainfall in farmers’ fields through soil and water conservation cannot by itself reduce the risks posed by frequent dry spells Investments are needed in water resources management in smallhold rainfed farming systems that use supplementary irrigation in combination with rainfall (Comprehensive

Within the blue water component, it is also very important to distinguish between surface and groundwater systems Groundwater plays a significantly different role than surface water In line with existing data, groundwater-irrigated agriculture shows higher productivity when compared to irrigation with surface

part by the greater ability of farmers to control water use and the supply guarantee

or security that groundwater provides against dry spells These two facts in turn allow farmers to invest, without fear of dry periods, in more efficient irrigation techniques and more expensive equipment for cash crops Generally, farmers, who use groundwater, bear all financial, operating and maintenance costs Groundwater users usually pay a higher price per volume of water than irrigators who use surface water, because the latter is usually heavily subsidised

Third, grey water is the volume of water required to dilute the amount of pollutants emitted to the natural water system to such an extent that the quality of the ambient

This component, however, is difficult to evaluate and beyond the scope of this study

In this study, we evaluate the green and blue water components of irrigated crops A detailed modelling approach was developed to evaluate the monthly evapotranspirative demand for each crop, province and year (1997–2006) and the corresponding percentage of blue and green water supplies

2.3 International Virtual-Water “Trade”

Akin to trade theory, international virtual-water “trade” can be evaluated in terms

of comparative advantage (first explicitly formulated by the British economist

unevenly distributed over space and time It is claimed that nations can profit from trade if they concentrate on, or specialise in, the production of goods and services for which they have a comparative advantage, while importing goods and services for

refers to the ability of a country to produce a particular good more efficiently and

at a lower opportunity cost than another country Many water scarce nations save domestic water resources by importing water-intensive products and exporting

the pressure on their domestic water resources and avoids the economic costs and

water savings through product imports can translate into global water savings if

imports originate in countries with higher green and blue water productivity (Allan

within the same region For instance, green water productivity may be very high in

a severe drought because a small amount of soil moisture may be used very ciently by crops In severe agronomic drought water reservoirs and aquifers may have ample reserves, providing blue water to entirely meet the water needs of irrigated crops In this case, blue water productivity in a particular field may

effi-be lower in relative terms than the green water productivity of a rainfed crop just across the road

Whether international trade actually helps alleviate global water stress is still an

The pros and cons of the virtual-water “trade” should be weighed, including the

flows may be more beneficial than others simply because of the higher opportunity cost of the water being saved Consideration of the green/blue water components

of the traded virtual-water volumes is essential to establish how much farm trade should be credited with reducing overall water use For instance, if Australia

Egypt However, since wheat is grown on dry land in Australia, but on irrigated land in Egypt, this is a water-saving exchange in terms of the use of Nile River water and from the economic and global standpoint (Hoekstra and Chapagain 2008)

The virtual-water metaphor addresses resource endowments but not production technologies Hence, the metaphor does not include the concept of comparative

offi-cials to consider policies that will encourage improvements in the use of scarce resources, but comparative advantages must be evaluated to determine optimal

economic considerations often outweigh water scarcity concerns, limiting the

associ-ated with crop trade is about 15% of the total water used in crop production

20% of virtual-water “trade” seems to be due to water scarcity (Chapagain et al

2006a; Hoekstra and Chapagain 2008) Therefore, less than 3% of the water “trade” is due to water scarcity This is a fact that needs to be assessed in more detail, because it might mean that the pervasive concept of water scarcity is overstated or perhaps that the scarcity of land and physical or human capital may

virtual-be more important than water scarcity

In addition to the criticisms levelled against the concept of virtual-water “trade”

in terms of international trade theory, there are a number of limitations that must be

nitrogen, phosphorous and potassium, in addition to water scarcity The virtual flow

of nutrients should enter the picture, along with land and water, as another limiting factor in production However, adding in more factors makes this equivalent to an international trade in goods, with considerations of prices and values; mainstream economics would prescribe focusing on competitive advantage instead of just one factor’s productivity The difference between water and other variable inputs such

as fertilisers is that, in the short term, water supply is very inelastic and not tutable in crop production And, more fundamentally, it is not a marketable good, which means that societies must either “produce” it internally with capital goods (infrastructure) or import it embedded in other goods This, in essence, is the underlying principle initially posed by Prof Allan when he developed the idea of virtual-water

substi-“trade”

for a country or region resulting from the choice to rely on food imports instead of investing in infrastructure or subsidising domestic production Basically, their point

is that focusing on virtual-water “imports” is not a neutral policy for a water-scarce country, since this affects, among other things, urbanisation, rural–urban migration

generates overall welfare gains, but also winners and losers among trading partners Through the use of Computable General Equilibrium (CGE) models, the global effects of water supply constraints in major trading partners can be identified, as well as how these constraints affect food prices at a global scale To what extent water resources are mobile across water-scarce sectors has an impact on the size of welfare losses and gains Domestic flexibility, meaning water re-allocation driven

by market signals, is required to create larger welfare gains Again, this idea that a factor’s mobility through trade can generate welfare gains is at the core of interna-tional trade theory This report also shows the importance of agricultural trade distortions in the global welfare effects of virtual-water “trade”

scarcity in water-stressed regions in India In explaining virtual-water flows, these authors identify key explanatory factors other than water scarcity, including per capita gross cropped area (an indicator of land concentration and population density) and access to secure markets (an indicator of institutional performance)

2.4 Water Footprint Concept and Its Relation

to Economic Growth

The water footprint (WF) is a consumption-based indicator of water use (Hoekstra

volume of freshwater that is used to produce the goods and services consumed by

the individual or community (Hoekstra and Chapagain 2008) The total water footprint

in a country includes two components First, there is the internal water print, which is the volume of water taken from domestic water resources to produce the goods and services consumed by the inhabitants of the country (Hoekstra and

water used in other countries to produce goods and services imported and

and external water footprints both in absolute and relative terms (based on Hoekstra

Japan, the UK, Spain and France, which are large importers (and exporters, in the case of Spain and France) of farm products The USA, Canada and Spain stand among the countries with the largest internal per capita footprints In the present study, we particularly highlight the relevance of virtual-water “exports” and “imports” for the economic life of many countries, including Spain Our work also aims to frame water footprint evaluations in a policy-relevant context

growth on water footprints Water use depends on a multiplicity of factors such as global change, including climate change, population growth and social changes, and water makes demands on other sectors such as energy production and ecosystem services A few studies about the influence of economic growth on water use have

withdrawals have remained quite stable, even where population and per capita

0 0.5 1 1.5 2 2.5

UK Russian Fed

Fig 2.1 Internal and external water footprints of several countries Based on Hoekstra and Chapagain ( 2008 )

Gross National Income (GNI) have increased significantly As will be shown in Chap 7, Spain has followed a similar path since the early 1990s.

There is no simple pattern for the impact of economic growth on water prints Depending on the country, markets and prevailing policies, the outcome will

foot-be different and WFs will either increase or decrease as the production of goods and services expands In order to analyse the extent to which economic growth has an influence on WFs, the variables through which growth can affect the footprints outcome must be examined These are scale effects, sectoral composition and technological change

2.4.1 Scale Effects

Economic growth is measured as the increase in the value of goods and services

increased international trade, it may mean that a given country will deplete natural

This is a result of market failures, such as ill-defined property rights, inadequate resource pricing and a failure to incorporate environmental externalities (Brack and

Economic growth is accompanied by changes in consumption patterns In line

GNI growth up to a certain level of income (about US$ 5,000/year) and then becomes less and less sensitive to change in GNI per capita This is the case of emerging countries such as BRIC (Brazil, Russia, India and China), where diets are changing significantly towards water-intensive meat and dairy consumption

On the other hand, the increase in overall financial capacity may both supply more resources for environmental protection and support greater demand for

and raise resources for different objectives, including pollution control and general

p 20) Though still contested, empirical evidence shows that environmental awareness

is often conditioned by the so-called Environmental Kuznets Curve (EKC), which links environmental quality (e.g some specific pollutants) with per capita income Ecological or environmental awareness develops when the country reaches a certain

reductions in developed countries are a result of increased consumption of intensive products imported from developing countries Countries may also reduce their internal water or ecological footprint by increasing the external water footprint

pollution-in exportpollution-ing countries For pollution-instance, the estimated water footprpollution-int of an average

Briton shows that two-thirds of this footprint originates outside Britain (Chapagain

foot-prints and their pattern and evolution from 1996 to 2006

shown how the curve applies to many regions in the world in relation to water resources

shows that economic growth, measured as an increase in both regional GDP and per capita GDP, is apparently not related to variations in water use Nevertheless, this result could lead to misleading interpretations in the absence of footprint evaluations, since the results might be different if other spatial and temporal dimensions were considered

2.4.2 Sectoral Composition

Shifts in economic structures are accelerated by economic growth (Brack and

through processing, to manufacturing and then to services Each step tends to lead

to a reduction in pollution output and resource depletion, though the correct pricing

of environmental externalities is a key factor (ibid.)

Furthermore, allocative efficiency gains from specialisation in the production of goods or services where a country has a comparative advantage can lead to a reduction

in global WF accounts, if correct national and international incentives and/or

literature does not offer any example that shows water allocation efficiency gains resulting from changes in water use in a given sector In fact, the causality is probably reversed, so that as the tertiary sector economy grows in relative terms at the expense

of the agricultural and industrial sectors, water is generally re-allocated, either through water markets or by government agencies This can even occur within the agricultural sector, as has already been seen in Spain and in Australia

At the global level, neo-Malthusian predictions have recently gained prevalence, partly as a result of sharp food price increases in 2008 (Formas 2008) Doubts exist

as to whether the world will be able to provide enough food for all its people on the horizon of 2030–2050, but few analysts today have concluded that there will be technical advances preventing this Kuylenstierna et al (2008) point that total water

a diet of 3,000 kcal per person with 70% plant and 30% animal components UNESCO (2008) has estimated that irrigated land should increase 30% by 2030 for

a similar diet If, to this increased water demand, we add the projected impacts of

reasons for concern The increase in irrigated acreage does not seem to be limited

infrastructure and the lack of human capital required to move from subsistence to commercial agriculture

2.4.3 Technological Change

Exogenous technological change is generally due to new production methods Endogenous technological change is determined by trends in output and input prices, market structure, economic incentives and improvements in physical and human capital The potential of economic growth to gain access to modern tech-nologies in the international markets and employ less resource and pollution-intensive technologies may help reduce WFs Growth may be beneficial for the environment due to its potential effects on the kind of technology used by domestic

advanced rainwater harvesting and supplementary irrigation techniques Furthermore, growth-induced trading regimes that are open to foreign competition and the constant need for technological progress force a country’s producers to stay

technology generally consumes fewer water resources and generates less pollution

The increased dissemination of more efficient and less polluting technologies can lighten water footprints

Finally, the hypothesis that economic growth might benefit the environment is

load on the environment, but it does not guarantee smaller WFs The net effect on the water use, as well as on the environment, will depend on the kind of economic

that for almost all environmental indicators, economic growth must be based on

environmental indicator divided by € of GDP exhibits a downward trend, but in absolute terms many of the indicators are still growing Since, in mature water

Latin America and Caribe

East and Southeast Asia

km 3

Fig 2.2 Water uses and renewable resources in selected world regions (Based on Comprehensive Assessment of Water Management in Agriculture 2007 )

economies, domestic water resources are generally limited, it is instructive to see whether a country’s external water footprint grows along with its economy If this

is the case, then its economic progress could still be coupled to water resources, though abstracted and integrated in the exporter’s production processes This study asks whether and to what extent Spain’s economy is still dependent on its internal and external water footprints

3.1 Introduction

The purpose of this chapter is to describe the methodologies applied in this study and to discuss other methods that are also applied in virtual-water studies Within the virtual-water literature, the general approach is based on calculations of the virtual-water content of products in order to estimate the water footprint (WF) of a sector or economy However, the examples found in the literature show certain

essential to clarify the place and period under consideration, as well as the point of measurement and the method for attributing water inputs at different production

of developing an input–output model to analyse the interrelations between the

in the Spanish economy, as well as their impact on the availability of the resource The authors focus on the role of each sector as a forward or backward linkage and

on the importance of the internal use of water, studying water intensity/productivity

by sector and the impact on income generation

south of Spain The model allows direct and indirect water consumption by sector

to be determined, as well as to what extent water limits the production potential of each sector of the economy Once the input–output model of water consumption is

Although this matrix accounts for the direct relationships among sectors, it does not include the indirect ones

models for water is expressed only in cubic metres, leaving aside its economic value Nevertheless, this approach is complementary to ours, since it provides valuable information both on the relationships among sectors and the technical coefficients to estimate these relationships A new application of the virtual-water and water-footprint

Methodological Approaches

A Garrido et al., Water Footprint and Virtual Water Trade in Spain: Policy Implications,

Natural Resource Management and Policy 35, DOI 10.1007/978-1-4419-5741-2_3,

© La Fundación Marcelino Botín-Sanz de Sautuola y López 2010

concepts and methods is in the Life Cycle Analysis (LCA) (Chapagain and Orr

through-out the whole economy, taking into account not only direct and indirect tion but also waste and re-utilisation

consump-3.2 Water-Footprint Calculation

in Chap 1, the main contributions of this study to previous WF analyses are (1) the differentiation and separate estimation of green water and blue water in rainfed and irrigated agricultural production; (2) the economic valuation of water in terms of apparent water productivity and economic scarcity and (3) the time and spatial dimension, as the study covers the period 1997–2006 and all Spanish provinces and river basins These improvements provide a more dynamic view of the water foot-print, connecting it to drought cycles and the flows of virtual-water “trade”

As described in the literature, a country’s water footprint can be assessed from the bottom up or from the top down In the bottom-up approach, all goods and services consumed by a country’s inhabitants are multiplied by their water needs at the site of production The top-down approach estimates total water use in a country and then subtracts the water used for producing export commodities and adds the water used to produce imports at the production site We have applied the top-down approach because it fits better with our data sources and provides more direct connections with water, agriculture and trade policies

volume taken from internal water resources to produce commodities consumed by

volume of water resources used in other countries and consumed by the country’s

3.3 Internal Water Footprint

3.3.1 Crop Water Use

Crop water use refers to the volume of water used for crop production One of the

Green water is the infiltrated rainwater stored in unsaturated soils (Falkenmark

and surface water systems (ibid.) These two categories are kept separate because their relative proportion is extremely sensitive to climate variability Furthermore, within the blue water component, the distinction between groundwater and surface water is essential to address the stabilisation role of groundwater under stochastic variations

in surface water levels

a 1

The green and blue components of crop water use are summed over the number n

irriga-tion, including both open air and covered systems

Both green and blue crop water use are calculated in a number of steps First,

mm/month) under standard conditions, which means that water does not limit plant

Green water use (m 3 /ha)

Internal agricultural water footprint (m 3 )

Blue virtual water content (m 3 /ton) Blue water use (m 3 /ha)

External agricultural water footprint

Virtual water ‘imports’ (m 3 ) Green virtual water ‘exports’ (m 3 )

Crop water requirements (mm/month)

Green virtual water content (m 3 /ton) Blue WF of production (m 3 )

Total area (ha) Irrigated

area (ha) crop yield (ton/ha)Irrigated/rainfed

Reference evapotranspiration Crop parameters

Effective rainfall (mm/month)

From Chapagain and Hoekstra (2004)

Fig 3.1 Diagram to calculate the internal and external water footprint of agriculture Based on Hoekstra and Chapagain ( 2008 ) Source: Own elaboration

multiplying the reference evapotranspiration (ETo, mm/month) by the crop coefficient

methodology developed by the Food and Agriculture Organization (Allen et al

1998)

o

Effective rainfall is defined as the amount of rainfall water actually available to

irrigation requirement is zero if effective rainfall is greater than crop water requirements

value between effective rainfall and crop water requirements Similarly, for irrigated crops, blue water evapotranspiration is equal to the difference between crop water requirements and green water evapotranspiration This calculation is carried out by crop, by Spanish province and by month It is assumed that surplus water in a given month is lost to deep drainage and runoff and cannot be used dur-ing following months Therefore, the water balance is calculated monthly The type

of soil is not considered; therefore, the estimations of green and blue water transpiration are subject to a certain error

to crop evaporative demand, that is it, perfectly matches crop water requirements

3.3.2 Livestock Water Use

Within the livestock sector, water is consumed both directly and indirectly: this includes water consumed directly by farm animals and indirectly in the production

of animal feed In order to calculate the livestock’s sector water use, we multiply the virtual-water content of live animals by the number of animals registered in the animal census (in the case of pork and broiler chicken, we counted the number of

on the virtual-water content of their feed and the volumes of drinking and service water required during their lifetime

The online publication of the “Canada Statistics Division” (Statistics Canada 2003) is used as the main feed data source for livestock In order to adjust these data

to the specific parameters of the Spanish livestock sector, we adjusted animal ages

for each species

virtual-water content, we obtain the theoretical volume of virtual-water used for the feed production These data were compared with those of feed production published in 2008 by INTERAL, the Spanish association of feed producers Based on the tons of feed production and amount used by each animal species, we obtained an approximate

evaluation of the virtual-water content of each species feed In the case of broilers and pigs, we used the weighted average feed as specified by FAO (2006) for France.

This refers to the water needed to produce goods and services within a country

of agricultural production accounts for all of the water used for agricultural purposes in a country regardless of where the products are actually consumed The water footprint of production can be used to examine the stress placed on a country’s

3.3.3 Industrial and Urban Water Use

Industrial and urban water use refers to total withdrawals of blue water Industrial water withdrawal includes treated water required for the entire industrial production process Urban water withdrawal includes household, commercial and municipal water consumption In addition, the industrial and commercial sectors have been divided into the following sub-sectors (1) agro-food, (2) textile, (3) lumber, (4) paper, (5) chemical industry, (6) plastic and rubber, (7) other non-metal products, (8) metallurgy and metal products, (9) machinery and mechanical equipment, (10) electric, electronic and optic equipment, (11) transport and (12) other manufacturing industries

3.3.4 Virtual-Water “Exports”

Virtual-water “export” involves the export of products produced within the country

To determine virtual-water “exports”, commodity export volumes (ton/year) can be

is calculated as the ratio between green crop water use and crop yield (Y, ton/ha)

crop yield Since yield is different for rainfed and irrigated lands, each has been estimated separately: a single green component for rainfed crops and separate green and blue virtual-water components for irrigated primary crops It should be noted that irrigated production includes both open air production and covered systems

,

g g

CWU V

Y

b b

CWU V

Y

The total virtual-water content of a primary crop (V c, m3/ton) is the sum of the green and blue components To determine the virtual-water “export” that corresponds

to a crop, we need to know the overall green and blue virtual-water content, since the available data do not differentiate exports from rainfed and/or irrigated production Overall green and blue virtual-water content is calculated by weighting rainfed and

irrigated virtual-water values in terms of relative production (P, ton/year).

A primary crop might be processed into a number of crop products (e.g wheat into wheat flour) In such cases, we calculate the virtual-water content of the processed product by dividing the virtual-water content of the primary product by the product fraction When two or more products are derived from the same crop root, it is necessary to distribute the virtual-water content of the primary crop among its by-products This is done by including a value fraction which is propor-tional to the value of the processed product Furthermore, process water use (PWU,

Following the methodology described above, we can also calculate the

a a

V V P

volume of blue water withdrawals divided by the production value in monetary terms, except for textile, paper and lumber industries In these cases, the virtual-water content of the industrial product is calculated based on the virtual-water content of the primary commodity Thus, for example, the virtual-water content

of textile products is estimated as the virtual-water content of cotton multiplied

by the ratio between cotton-derivative products (ton) and total textile products (ton) Although it may be a crude estimate, it is done this way because industrial products and processes are highly heterogeneous Furthermore, it is difficult to find reliable statistics related to water use in the industrial or commercial economy

j

where j denotes the product traded (e.g crop, livestock and industrial products), V

3.4 External Water Footprint

The external water footprint is equal to the difference between virtual-water

3.4.1 Virtual-Water “Imports”

outside the country Following the methodology described above, virtual-water

“imports” are estimated by multiplying the quantity of the export by its estimated virtual-water content at the production site

The virtual-water content of imported primary crops and live animals, as well as

of their respective by-products, has been taken from Chapagain and Hoekstra

content of industrial products is assumed to be equal for all countries concerned Therefore, the virtual-water content of an imported product is equal to the virtual-water content of the same exported product

j

where j denotes the product traded (e.g crop, livestock, industrial products), V

importing country

1 Virtual-water content of industrial products is expressed in m 3 /€.

2 Trade in industrial products is expressed in Euros.

Furthermore, net virtual-water imports (NVW I, m3/year) are estimated as the difference between virtual-water imports and virtual-water exports Net virtual-water “import” to a country might be either positive or negative The former would

be a net importer and the latter a net exporter

3.6 Apparent Productivity of Water and Land

The virtual-water concept is linked to water productivity, geographical location and the site-specific socio-economic setting By shifting production to areas with higher water productivity and lower opportunity cost, scarce water resources may be

this in mind, it is worth going one step further in virtual-water studies and adding

a new economic dimension to the previous estimates

In this study, we have included the concept of apparent water productivity (WAP,

metre of water required when producing the commodity This parameter is mated as:

ton In agronomic terms, this is the inverse of water productivity

In a similar way, apparent land productivity represents the economic value of farm output per hectare cultivated (€ /ha)

3.7 Economic Value of Water

There is a growing body of literature focusing on virtual water and water footprints However, to the best of our knowledge, none of these studies deal with the economic valuation of virtual water From a water resources perspective, the analysis

of potential gains from international trade must take into consideration economic and policy conditions, in addition to the spatial and temporal variations

socio-of blue and green water

From an economic perspective, only blue water is valued Green water certainly has an economic value both for agricultural production and natural ecosystems

attach an opportunity cost to green water, since it cannot be easily allocated to other uses The environmental benefits of using green water instead of blue water for food production and trade can be evaluated globally

Blue water can be assessed using various methodologies that provide different layers or levels of analysis For instance, a global evaluation of water within a basin could incorporate all services, including biodiversity, landscape and production

value of world natural resources and assets to humankind This framework guishes an ecosystem’s functions from environmental services Ecosystem func-tions refer to system properties and processes Services represent the benefits that society derives, directly or indirectly, from ecosystem functions A summary of these authors’ evaluation of annual flows of water-related ecosystems on a world

non-commercial water services and to delimit the services we will be focusing on here

As the numbers show, humans enjoy many different services from water-related ecosystems in addition to water supply We note, for example, that one hectare of wetlands can generate almost $4,200 per year in waste treatment services While this evaluation was certainly preliminary at the time it was produced, it conveys a clear idea of the costs and damages that water scarcity can provoke The mere recognition of many services identified as being valuable for society has huge implications for drought policy design and implementation Chief among these is the fact that many of these services have public natural features, which means that they are non-rival and non-exclusive goods As scientists have learned to identify and value them, water policy must take them into account and ensure that decisions included compromises among both productive and non-productive services

For the purpose of this study, the economic value of blue water is defined in terms of shadow prices or scarcity values Shadow price refers to a consumer’s willingness to pay for an extra unit of water and is equivalent to the marginal value

of available water endowments, which measures the benefits derived from an increase in water availability In order to estimate the marginal value of water, it is

shadow price is also a useful tool to measure, in economic terms, the effects of resource depletion and degradation In such a way, it could serve as a guide for

measure the economic value of blue water seems consistent with the analysis of virtual-water “trade” in arid countries, where the distinction between green and blue water is essential to relate land and water management to drought and climate variability

been selected based on a comprehensive review of the literature Blue water values are defined for each river basin and scarcity level Each Spanish province

is identified with a specific river basin, although the administrative and basin boundaries do not perfectly overlap Blue water scarcity value varies according

to scarcity level, which in turn depends on the volume of water stored in each river basin Scarcity levels are defined on a scale of 1–4, the latter being the scarc-est level For each river basin, storage thresholds are based on a percentile analy-sis for the period 1997–2006 Thus, when in a certain year the volume stored in May is higher than the 50th percentile, the scarcity level is 1 Scarcity level 2 corresponds to a volume stored between the 25th and 50th percentiles Scarcity level 3 is defined as being between the 10th and 25th percentiles and scarcity level 4 occurs when the stored volume is lower than the 10th percentile Groundwater storage has not been considered in this stage of our study because

of the lack of reliable data from the Ministry for Environment However, probably this deficiency can be solved during the hearings and public participation processes in pursuant to the water plans required by the Water Framework Directive (end of 2009) In any case, previous analysis of the Guadalquivir Basin

the dry spells In summary, the Andalusian study showed that during the dry spells, when the surface water storage collapses, farmers using groundwater make more profit

3.8 An Econometric Approach

3.8.1 Explaining Water Productivity by Water Scarcity and Water

Quality

Our data generation process allows for testing the hypothesis that water productivity

is dependent on water scarcity Basic economic theory would suggest that as water becomes scarcer, users would be more efficient

The simplest test one can think of with our database makes use of spatial and

temporal variations of both water scarcity and blue water productivity With i denoting a province, and t a year, we can pose the following model, only relevant

for irrigated agriculture:

which varies according to year and basin, using the parameterisation shown in

ratio between green crop water use and total crop water use in irrigated production

in province i and year t.

The time series and panel structure of our database can be best estimated using Feasible Generalized Least Squares, assuming heterocedastic but uncorrelated panels.

2

0

I V

estimations as well as regional estimations This strategy will be pursued by estimating the model for all provinces, Mediterranean provinces only and interior provinces only In terms of the model’s variables and crop demand, these major

“regions” differ in two essential aspects (a) the much greater percentage of green water in the evapotranspirative demand of crops in the interior regions than in the Mediterranean regions and (b) the fact that water is scarcer in economic terms in the Mediterranean regions than in the interior regions Clearly, the most direct

in elasticities

2

could be either positive or negative Positive/negative means that worse water quality would reduce/increase the efficiency of water use by increasing/reducing water use

in contexts where the green water component is less important or crucial, such as

more water is needed to meet the total water demand of crops

Since both water scarcity and water quality values differ among river basins, a set of dummy variables is introduced to explain as much as possible about these inter-basin differences:

i

variables in the model Once the geographical differences are controlled by

is correlated with higher blue water productivity

Regarding the valuation of water resources in terms of the opportunity cost of quantity and quality, we can pose the following models in order to explain the

differences between Mediterranean and interior provinces on one hand and to control for latent geographical conditions on the other:

expressed in terms of opportunity cost of quantity and quality, that is € of scarcity

water use and total crop water use in irrigated production in province i and year t

measured in € of crop value per hectare

assuming heterocedastic but uncorrelated panels

negative, which means that a higher proportion of green water would reduce the

Positive/negative means that higher irrigated land productivity would increase/reduce the opportunity cost of water in terms of quantity and/or quality

In order to control for the geographical differences among river basins, we can use the following model:

i

3.8.2 Explaining Blue Virtual-Water “Exports” by Water

Scarcity and Water Quality

Making use of the spatial and temporal variations in both water scarcity and land productivity, we can use the following model to test the hypothesis that blue virtual-water exports are dependent on water scarcity and land productivity:

production in province i and year t, measured in € of crop value per hectare.

As in the previous model, the time series and panel structure of our database can best be estimated using Feasible Generalized Least Squares, assuming heterocedastic, but uncorrelated panels (provinces)