https://orcid.org/0000-0001-6195-7664
![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
ABSTRACT
With the federalism reform in 2006, the German Länder gained the competence to deviate from the nationwide standard legislation in various areas. In this study, we focus on nature conservation policies and see that some Länder are raising their regulative standards while others are lowering them. Based on an original dataset that measures this direction of policy change, our empirical analysis shows that the party-political composition of the state governments, and particularly the presence/absence of Green parties, plays a crucial role in explaining the variation. Moreover, we find that the strength of nature conservation associations and geographical conditions are relevant, too. In sum, our findings contribute to the growing literature on both the policy effects of the German federalism reform and the partisan effects on public policies, as well as the question of responsive decision-making in multilevel settings.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Theoretically, federalism and de-centralization are two different concepts and the efficient provision of public goods at the local level may also be possible in unitary states (Biela, Hennl, and Kaiser Citation2013). However, the politico-institutional framework of federalism is often linked to decentralized structures (Keman Citation2000) and federalism itself has also been ascribed positive economic effects (Weingast Citation1995).
2 However, this deviation can be overruled by new federal law and regional laws – while the most recent law adopted prevails (Scharpf Citation2009, 99–101).
3 It is true that this simple idea gets more complicated when we introduce the differentiation between the de-centralization of resource allocation and taxes on the one and federalist decision-making on the other hand (Biela, Hennl, and Kaiser Citation2013).
4 In practice, however, this ‘ban’ did not survive for a long time given that the Länder needed financial support from the federal government (Kropp and Behnke Citation2016, 675).
5 Art. 72.3 GG further states that ‘federal laws in these areas come into force at the earliest six months after their promulgation, unless otherwise determined with the consent of the Bundesrat.’ This leaves the Länder with a certain time horizon to adapt their regulation and prohibits quickly changing regulations. A complete list of the policy areas and the respective changes can be found in Scharpf’s excellent review of the reform process and outcomes (Scharpf Citation2009, 106–107). We thank the anonymous reviewer his/her clarifying comments.
6 While it is true that the budgetary situation differs between the richer (e.g. Bavaria) and poorer states (e.g. Saarland), these differences are moderated by the equalization system and specific subsidies, which is why economic differences are minor as compared to cross-national comparative studies (e.g. Renzsch Citation2010). Moreover, environmental policies, and nature conservation policies in particular, are regulative in nature and should therefore be less affected by smaller differences in the economic situation.
7 While we do not test whether voters of Green parties for indeed prefer protective nature conservation policies, survey data indicate that this expectation is not far-fetched: In the Politbarometer surveys in 2019, around 49.6 percent of the respondents who said that nature conservation is an important issue also intended to vote for the Greens at a federal election – followed by 21.6 percent for the CDU/CSU and 11.6 percent for the SPD (Forschungsgruppe Wahlen Citation2020).
8 The data comes from the website ‘buzer’ which provides a searchable juridical database which documents all national laws and links the deviating regional laws for the respective paragraph. This database is well-established in juridical-scientific circles within Germany and is considered reliable because it is regularly cross-checked with the Federal Ministry of Justice’s texts for quality assurance purposes. Previous studies in the context of deviating nature conservation law also used this platform (e.g. Böcher and Töller Citation2016).
9 We only consider Länder which made use of their right to deviate from the federal standard, i.e. Brandenburg, Bremen, Saarland and Thuringia, are excluded. Moreover, we did not opt for any qualitative weighting that may account for the content of the deviation, as this would involve making (even) more qualitative choices along different analytic dimensions. Unweighted indices are common in the absence of strong a-priori-reasons for weighting (e.g. Hooghe, Marks, and Schakel Citation2010). As shown by the robustness tests of our outcome calibration (see Appendix D), adjusting the calibration does not lead to major changes of the results.
10 Similar to correlations, set relations per se are not causal which is why the plausibility of any assumed causal relationship should be corroborated through additional evidence from the underlying cases (Schneider Citation2018, 253).
11 The configuration representing Hamburg 11/13 exhibits similar levels of consistency for protective and permissive policy change. Including it into both analyses would pose a contradictory statement since the same combination would be used to explain protective and permissive policies. We only include it into the former based on three reasons: i) the consistency value for protective policy change is higher than for permissive policy change, with the latter being below 0.8 and at the lowest end of what is usually deemed sufficient; ii) the PRI parameter is higher for protective than permissive policy change; and iii) Hamburg 11/13 itself shows protective policy change. We test for the effects of alternate decisions in our robustness tests (Appendix D).
12 We use the parsimonious strategy which is driven by the objective to produce minimally sufficient conditions and therefore includes all logical remainders that lead to more parsimonious results (Baumgartner Citation2015; Schneider Citation2018). In the Appendix we also present the results of the complex and intermediate strategy. Since our data contains only one logical remainder, the differences between the chosen strategies are at the margins (see Appendix C).
13 In addition, having either small areas of nature reserve or weak environmental associations also turns out as a consistent superset. However, we again do not interpret this substantively because i) the relevance parameter points towards empirical trivialness with a score close to 0.5, and ii) the conditions cannot be integrated into an overarching concept.
14 Expert interview, 3.9.2021.
15 Expert interview, 7.9.2021.
16 We are aware that these are political and not factual statements.