A bridge over sustainable water: Politicians’ perceptions about the preconditions for collective action

Sustainable water resource management is a core interest for all societies. As water systems are often common resources, the management of water systems requires coordinated action among actors along the water. For flowing water, a complication for coordination is upstream–downstream relations where what happens upstream affects downstream, but not the other way around. In this study we present results from a survey experiment with politicians in Sweden, focusing on whether and to what extent their willingness to cooperate is affected by their placement upstream or downstream along a fictive water system. Our findings indicate that politicians from upstream and downstream municipalities share the view that upstream politicians bear greater responsibility for undertaking preventive actions and are willing to assume remedial responsibility for problems caused by them. These results challenge the notion that self-interest is the primary obstacle to resolving environmental collective action problems. Supplementary Information The online version contains supplementary material available at 10.1007/s13280-023-01975-5.


Figures and tables
Figure S1 Politicans and Self-interest, likert-scale plot (percent).
Note: The question asked was: How do you think that the costs for decreasing the risks that threatens the drinking water should be distributed between citizens in municipalities upstream (where water runs from) and citizens in municipalities downstream (where water runs to) the river?Seven-point scale was used to measure their attitudes, ranging from 'citizens upstream should pay the full cost' (1) to 'citizens downstream should pay the full cost' (7) with 'equal share' as a middle alternative (4).

Figure S2 Politicans and Equal contribution, likert-scale plot (percent)
Note: The statement was "Every single municipality is responsible for making investments in order to prevent events that pollutes the water, irrespective of the consequences do not affect the municipality's citizens".The respondents were asked to take a stand to by choosing from a scale with the following alternatives: totally agree, partly agree, hardly agree, do not agree at all.

Figure S3
Politicans and Causal responsibility, likert-scale plot (percent) Note: The statement was "Municipalities upstream are responsible to make investments that reduces the risk for pollution of the water that can affect downstream municipalities".The respondents were asked to take a stand to by choosing from a scale with the following alternatives: totally agree, partly agree, hardly agree, do not agree at all.

Figure S4 Politicans and Conditional altrusim, likert-scale plot (percent)
Note: The statement for the figure to the left was "Municipalities downstream should contribute with money to municipalities upstream in order to help out with costs for measures that prevent pollution of the water".The statement for the figure to the right was "If the water is polluted because of something occurring upstream, municipalities upstream should compensate downstream municipalities for the costs".The respondents were asked to take a stand to by choosing from a scale with the following alternatives: totally agree, partly agree, hardly agree, do not agree at all.
(1 = under 30 years old, 2 = 30-39, 3 = 40-49, 4 = 50-59,  5 = 60-69, 6  = 70 or older.Education at a university is a dummy variable (0 = not studies at a university, 1 = studies at a university).Ideology contains of five categories (1 = clear to the left, 2 = slightly to the left, 3 = neither left nor right, 4 = slightly to the right, 5 = clear to the right).Representativeness of the Panel of PoliticiansTo explore the representativeness of the Panel of Politicians, a study was conducted in out to all 12,829 local and regional politicians in late 2016 and included a question about whether the respondent was a member of or would like to join the Panel of Politicians.The respondents were given three alternatives: (1) Yes; (2) I am already a part of the Panel of Politicians; or (3) No thanks. 1 In the analysis, we merged alternative 1 and 2, thus creating a measurement for panel members which provided us an opportunity to examine whether panel members differed in a systematic way compared to the total sample of Swedish politicians at the sub-national level.Figures 1-3 below display the results from different sets of t-tests to detect skewness in the panel in terms of social characteristics and party membership.Significant differences (p<0.001) were found only in terms of gender and being a member of the Centre Party.Men and Centre Party members were found to be overrepresented in the panel, but all other characteristics were not.

Figure
Figure S5: T-tests, gender and place of origin

Table S1 .
Checks for randomization between experimental groups Mean scores (number of participants) Note: The age variable is separated into six categories