THE INFLUENCE OF SOCIAL CAPITAL ON WILLINGNESS TO PAY FOR THE ENVIRONMENT AMONG EUROPEAN CITIZENS

ABSTRACT

Social capital has been recognized as one of the most influential concepts in social and political sciences. It is assumed that it has significant influence on several issues such as health status, educational achievement and environmental attitudes. In the present article we attempt to investigate the influence of social capital on the tendency of European citizens to contribute money for the environment using data from EVS 1999/2000. Firstly, an estimation of individual social capital is calculated for European citizens with the use of Confirmatory Factor Analysis. Secondly, the relation between individual social capital and willingness to pay (WTP) is examined. Thirdly, further evidence is provided by introducing in the analysis the influence of other factors such as income, age, gender and educational level. In addition, the influence of contextual factors on WTP is explored. Through the results of the study it is observed that most European citizens are willing to pay for the environment and their decision is influenced significantly from elements of social capital. Furthermore, on a contextual level only some of the factors investigated explain the tendency to contribute money for the environment.

Keywords:

• social capital
• willingness to pay (WTP)
• environmental attitudes
• national differences

Acknowledgements

An earlier draft of the present paper was presented at the 8th Conference of the European Sociological Association, 3–5 September 2007, Glasgow. The authors gratefully acknowledge the financial support of the MSc program ‘Environmental Policy and Management’ (Department of Environment, University of the Aegean) in allowing them to attend the conference.

Notes

1. Free-riding behaviour refers to the situation where citizens accept the benefits from the use of an environmental good but are unwilling to share the cost that it implies (Olson 1965).

2. The selection of the 24 countries was based on the availability of variables included in the model of the study.

3. Specifically, the latter methodology is a three-step approach that in the first step estimates thresholds from the categorical observed variables, in the second step the polychoric correlations for the given thresholds of the first step are estimated and in the final step the CFA model is estimated from the polychoric correlations using the method of weighted least squares.

4. In addition, indices values of goodness-of-fit of the model obtained from LISREL 8.8 also indicate that the second-order factor model tested provided a good fit to the 13 observed variables: RMSEA: 0.043 (accepted boundaries 0.00–0.06), GFI = 0.99, AGFI = 0.99, NNFI = 0.95. NFI = 0.96, CFI = 0.96 (accepted boundaries >0.90).

5. Several categorizations of European countries have been proposed in the literature (Esping-Andersen 1990; Ferrera 1996). In the present study countries were divided according to Ferrera (1996) categorization with the addition of the category of post-socialist countries (see also Kaariainen and Lehtonen 2006). The Netherlands, which has been analyzed both as a conservative and a socio-democratic country, is treated as a ‘conservative’ country in our study. Finally, Iceland, which has been included both in the Scandinavian and Liberal countries (see Arnason 2004; Olofsson et al. 2006), it is treated here as a Scandinavian country.