Institutional Pressures and Voluntary Environmental Behavior in Developing Countries: Evidence From the Costa Rican Hotel Industry

Abstract

Abstract This study aims to identify how institutional forces, such as regulatory and stakeholder pressures, are related to proactive environmental behavior by hotel facilities participating in Certification for Sustainable Tourism, a voluntary environmental program established by the Costa Rican government. This program is among the first third-party performance-based environmental certification initiatives implemented in the developing world. Findings suggest that voluntary environmental programs that include performance-based standards and third-party monitoring may be effective in promoting beyond-compliance environmental behavior when they are complemented by isomorphic institutional pressures exerted by government environmental monitoring and trade association membership. These results are consistent with neo-institutional theory from the organizational sociology literature. Surprisingly, findings also indicate that compared to locally owned hotels, foreign-owned and multinational subsidiary facilities do not seem to be significantly correlated with higher participation and superior environmental performance in Certification for Sustainable Tourism. Keywords: Costa Ricavoluntary environmental programstourismhotel industryinstitutional theoryLatin Americamultinational corporationsnational parks I received invaluable comments from four anonymous reviewers, Mark Starik, Andrew Hutson, Christine Martel, and Linda Kalof. I am also very grateful to Jennifer Oetzel, Rodolfo Lizano, Robert Healy, William Ascher, and Stuart Hart for their support and ideas. Notes 1. Using power analysis and assuming “small” effect size for the independent variables, it was determined that a minimum sample of 138 observations was necessary to have an 80% chance of rejecting a false null hypothesis at 95% confidence (Cohen and Cohen 1983 Cohen, J. and Cohen, P. 1983. Applied multiple regression: Correlation analysis for the behavioral sciences, Hillsdale, NJ: Lawrence Erlbaum Associates. [Google Scholar], 59). 2. No comprehensive list of all the hotels operating in Costa Rica was found. Thus, a sample frame list including 649 hotels was prepared. Sources of information consulted for building the sample frame included archival data available at the Ministry of Tourism, the Costa Rican Chamber of Tourism, and the Association of Small Hotels, the 2000 Costa Rican Phone Directory, and the most popular travel book guides to Costa Rica. The travel guides consulted were A New Key to Costa Rica (Blake and Becker 1998 Blake, B. and Becker, A. 1998. The new key to Costa Rica, Berkeley, CA: Ulysses Press. [Google Scholar]), The Berkeley guide to Central America (Nystrom and Smith, 1996 Nystrom, A. D. and Smith, W. 1996. The Berkeley guide for Central America , 2nd ed., New York: Fodors Travel. [Google Scholar]), Lonely Planet (Rachowiecki, 1997 Rachowiecki, R. 1997. The Lonely Planet guide: Costa Rica , 3rd ed., Hong Kong, , China: Lonely Planet. [Google Scholar]); and Fodor's Costa Rica 99 (Rockwood, 1998 Rockwood, C. 1998. Fodor's 99: Costa Rica travel guide, New York: Fodor's Travel. [Google Scholar]). Subsequently, we categorized the sample frame list into 6 geographic groups and randomly drew observations from each group to build a survey sample of 250 hotels. 3. The author thanks one of the anonymous reviewers for highlighting the limitations of using hotel location as a measure for government monitoring. 4. I am grateful to an anonymous reviewer for pointing out this alternative explanation. 5. An anonymous reviewer suggested this alternative rationale. 6. OLS regression model. Condition index and variance inflation measures for the independent variables revealed weak to moderate dependencies among the independent variables. Hence, it was concluded that harmful multicollinearity did not affect the regression (Belsley et al. 1980 Belsley, D. A., Kuh, E. and Welsch, R. E. 1980. Regression diagnostics: Identifying influential data and sources of collinearity, New York: John Wiley and Sons. [Crossref] , [Google Scholar], 105). Lack of heteroscedasticity was also determined by White's chi-square test (White 1980 White, H. A. 1980. Heteroskedasticity-consistent covariance matrix estimator and direct test for heteroskedasticity. Econometrica, 48: 817–838. [Crossref], [Web of Science ®] , [Google Scholar]). 7. Two studies assessing participation in ISO-14001 and the Chemical Industry's Responsible Care Program have found that multinational subsidiaries operating in China, Mexico, and Brazil are more likely to participate in voluntary environmental programs (Christmann and Taylor 2001 Christmann, P. and Taylor, G. 2001. Globalization and the environment: Determinants of firm self-regulation in China. J. Int. Business Stud., 32: 439–458. [Crossref], [Web of Science ®] , [Google Scholar]; Garcia-Johnson 2000 Garcia-Johnson, R. 2000. Exporting environmentalism: U.S. multinational chemical corporations in Brazil and Mexico, Cambridge, MA: MIT Press. [Crossref] , [Google Scholar]). It is important to note, however, that ISO-14001 and Responsible Care lack performance-based standards and do not require third-party overseeing of the environmental behavior of participants. 8. These exceptional cases include five foreign-owned or multinational subsidiary hotels that have received some of the highest CST environmental scores. 9. “An increase in sample size (assuming no change in population standard deviation) results in a decrease of the standard error, thereby affording more precise estimates” (Pedhazur and Schmelkin 1991 Pedhazur, E. J. and Schmelkin, L. 1991. Measurement, design, and analysis: An integral approach, Hillsdale, NJ: Lawrence Erlbaum Associates. [Google Scholar]).