Local factors driving the adoption of municipal voluntary environmental programs: the case of Sweden’s eco-municipalities

Abstract

In the past few decades voluntary environmental programs (VEPs) have gained traction as a tool to promote environmental performance beyond regulatory requirements. However, such programs have been largely studied in the context of the private sector with comparatively less consideration given to the possibility that local governments can also join such programs to improve both their reputation and environmental performance. We consider Sweden’s Eco-municipality association as an instance of a public sector VEP and find that, even after accounting for spatial dependence; environmental consciousness, municipality type, the level of education, industry structure and environmental vulnerability (as proxied by proximity to the coast) are significant determinants of the municipality’s decision to participate in the program.

Keywords:

• Voluntary environmental programs
• eco-municipality
• public sector
• local government

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental data

Supplemental data for this article can be accessed here.

Notes

1 Lauf et al. (2020) compare regional variation in municipal land-use policy in Germany and Sweden; they note that compared to Germany as well as other European countries Swedish municipalities have a lot more autonomy.

2 Adopting the definition set out in the UN report Our Common Future of 1987 we take sustainable development to mean, “Development that satisfies today’s needs without jeopardizing the ability of future generations to satisfy their needs.”

3 An example of an industry association sponsored VEP is the Responsible Care program sponsored by the American Chemistry Council (formerly The Chemical Manufacturers Association). The program promotes pollution prevention, waste minimization and energy conservation. It has mandatory transparency reporting for its members (Prakash and Potoski 2006, 25).

4 Interestingly another study of the transnational Covenant of Mayors program found that the number of environmental NGOs in a country increases the proportion of its population living in a city signed up for the program. This provides additional support for the idea that civic capacity may be important for program participation (see, Dolšak and Prakash 2017).

5 One alternative that Allison (2009) suggests is to exclude those observations whose values for the response variable do not change over the sample period. This would allow us to still make use of a municipality-level fixed effects model. It is easy to see how doing this would negate the entire point of our analysis, as all the municipalities that never joined SEKOM have a time invariant response/dependent variable.

6 An alternative is to use a distance spatial weight matrix instead. Two municipalities would be considered neighbors if they fell within a certain distance of each other. This method avoids missing neighbors by picking the smallest distance, such that every municipality has at least one neighbor. In our case, that distance is about 128 miles which results in the median municipality having 38 neighbors, as opposed to the 5 neighbors we obtain using queen contiguity-based weights.

7 In addition to the previously mentioned argument by Allison (2009), another reason for opting for county-year rather than municipality-year is that the Markov Chain Monte Carlo (MCMC) algorithm employed by the spatial probit model did not converge when municipality-year fixed effects were used.