Understanding urban residential water use in Beijing and Tianjin, ChinaIntroductionMethodologyOverview of study areasFindings and discussionUnderstanding the differences in residential water useHousehold socioeconomic characteristicsHousehold housing environmentHousehold water amenity and facilityHousehold water uses and behaviorHousehold water perceptionHousehold environmental attitudesExploring the patternsModelsAnalysisSummary of major findingsConclusionsAcknowledgementsReferencesHabitat International 29 (2005) 469–491Understanding urban residential water use inBeijing and Tianjin, ChinaHenry H. Zhanga,*, David F. BrownbaThe Maryland-National Capital Park and Planning Commission, 14741 Governor Oden Bowie Drive, Upper Marlboro,MD 20772, USAbSchool of Urban Planning, McGill University, Macdonald Harrington Building, 815 Sherbrooke Street West,Montreal, Quebec, H3A 2K6 CanadaReceived 20 October 2003; received in revised form 11 February 2004; accepted 11 April 2004AbstractResidential water use (RWU) is directly linked with the health of the general public and constitutes wellover half of the total municipal water use in many countries. Because RWU is the most importantcomponent of municipal water use, it is always being given the top priority in municipal water supply andplanning. But little is known about RWU in developing economies, especially in the Chinese context. Inorder to obtain the determinants of water use and consumption, empirical study is widely pursued. In atransition economy, rapidly changing socioeconomic conditions profoundly alter household water use inBeijing and Tianjin. A social survey on urban household water use and consumption is judged as anecessary starting point to an efficient municipal water planning and management in the two cities. A totalof 806 in-house interviews were completed in Beijing and Tianjin by using stratified sampling method basedon the predominant housing typologies. A multivariate analysis has also been performed in order to revealthe structured relationship between the water use and consumption and postulated explanatory variables invarious urban households. For the first time, this empirical study will bridge the information gaps regardingurban household water use patterns and compositions, household water amenities and facilities, householdwater using habits and behaviors, household water perceptions and environmental attitudes as well as ahousehold’s capacity and willingness to respond to water conservation policy in different housingtypologies in the Chinese context.r 2004 Elsevier Ltd. All rights reserved.Keywords: Beijing; Tianjin; Household; Residential water use (RWU); Housing typology; SurveyARTICLE IN PRESS*Corresponding author. Tel.: +1-301-952-4151; fax: +1-301-952-3749.E-mail address: [email protected] (H.H. Zhang).0197-3975/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.doi:10.1016/j.habitatint.2004.04.0021. IntroductionResidential water use (RWU) can constitute over half of total municipal water use andconsumption in many developed countries (Howe & Linaweaver, 1967; Grima, 1972). RWUrequires higher quality and reliability, which in turn leads to higher costs. Historically, RWU isdirectly linked with the health of the general public (Lee, 1969; Grima, 1972) and is crucial to thewell being of a society.RWU shows a strong positive correlation with living standards and is one of the importantproxies for economic development. Many factors affect residential water demand. Both empiricalinvestigations and analytical studies of industrialized countries suggest several determinants ofRWU: policy variables, such as water price or water rate; household economic variables, such asincome; physical features and technological variables, such as water amenities and metering, orwater-saving plumbing fixtures; environmental variables, such as temperature and precipitation;and demographic variables, such as household size and attitudinal variables (Howe &Linaweaver, 1967; Lee, 1969; Grima, 1972; Katzman, 1977; Holtz & Sebastin, 1978; Danielson,1979; Clouster & Miller, 1980; Hanke & de Mar!e, 1982; Jones & Morris, 1984; Jones, Boland,Crews, DeKay, & Morris, 1984; Vickers, 1991; Baumann, Boland, & Hanemann, 1998).Specifically, water demand has an inverse relationship with the price charged per unit of water.Income, which is usually surrogated by market value/assessed value of residence (Howe &Linaweaver, 1967; Grima, 1972; Camp, 1978) or purchasing power (Lee, 1969), has a positivecorrelation with household water use.1Indoor use is less price-responsive than outdoor use.Annual and domestic water demand is inelastic, but shows higher elasticity for peak summerdemand (Jones et al., 1984). Comparisons between metered and unmetered water use suggest thatconsumers tend to alter their water use behavior to avoid extra charges and choose water-savingalternatives when they make water use decisions. This is especially apparent when consumers arefacing conservation-oriented water rates (for example in Cuthbert, 1989). In addition, per capitaluse is inversely correlated with household size (Danielson, 1979). Water use per residence isinversely correlated with rainfall per unit of time and is positively correlated with averagetemperature (see, for example, Morgan & Smolen, 1976). Water use per residence is positivelycorrelated with irrigable area, i.e. lot size (Linaweaver, Geyer, & Wolfe, 1967).However, little is known about RWU in developing economies, and especially in the Chinesecontext. One reason may be that most of the developing economies are in transition. Swiftlychanging demand patterns usually make the study very difficult. Many former studies suggest thatprojections of future residential water demand should not be based solely upon historical percapita use, which is the widely accepted method, but should incorporate a range of variables thatinfluence water demand. In order to obtain these determinants of water consumption, empiricalstudy is widely pursued. The basic hypothesis is that a good understanding of residential waterrequires the inventory and analysis of household water use and consumption patterns. Thispremise is particularly relevant to Beijing and Tianjin, where rapidly changing socioeconomicconditions have profoundly altered household water use. A social survey on urban householdwater use is judged as a necessary starting point in order to understand RWU. Empirical study isARTICLE IN