16.1.1 Locally-Tailored Policymaking

A virtue of federalism is often seen as being the way it invites policymaking that is customised to local circumstances, needs and preferences – the ‘decentralization theorem’, as Oates (Reference Oates1972, 35; italics in original) called it. Federal systems provide scope for such customisation to a degree unlikely in a unitary state, where ‘one-size-fits-all’ approaches may prevail.

This requires, of course, that the constituent units enjoy sufficient jurisdictional authority – and the studies in this book suggest that in climate change policy they generally do. Such a degree of autonomy unsurprisingly exists in the European Union’s (EU) supra-national proto-federal system, as Alberton emphasises. Even in Germany, though, with its administrative division of powers where the constituent units have more of an implementational role and less policy autonomy than in dual federations, Eckersley and colleagues make it clear that the Länder have had considerable scope to choose their own climate policy direction.

However, this is not the case everywhere. Pillai and Dubash emphasise the degree to which the highly centralised nature of India’s federal system reduces the scope for independent action by the constituent units. Although India’s states control several important areas of climate policy, financial and capacity constraints limit prospects for bottom-up action. Meanwhile, de Visser and du Plessis show that what they characterise as South Africa’s ‘quasi-federal’ system leaves the provinces with very little policy role at all, and the cities without the resources to pursue the climate action they contemplate. As Di Gregorio and Moeliono explain, the Indonesian case is similar: lacking the constitutional guarantees provided by federalism, the local role in forestry management there has been assumed by central government authorities, with adverse implications for outcomes.

The cases covered in this book provide numerous instances of regionally tailored climate change policy. Constituent units in Australia, Brazil, Canada, Germany, Mexico, Spain, Switzerland, the United States and other countries, have been able to adopt approaches consistent with their specific circumstances and preferences. Most important is how dependent the different constituent units are upon emissions-intensive economic activity – and therein lies the rub. As discussed below, while such flexibility has long been seen as a virtue of federalism, it has more contentious consequences for climate change governance.

16.1.2 Compensatory Federalism

The redundancy of having more than one order of government also creates potential for the constituent units to step up when the central government fails to act; they can play a ‘compensatory’ role (Derthick Reference Derthick and Rabe2010). This is something that sets federal states clearly apart from unitary ones. Given, as noted above, that constituent units typically hold a number of mitigation cards, the potential for compensatory federalism is undoubtedly there. We must be mindful, though, that the question is not whether compensatory action by constituent units is as effective or efficient as an optimal set of policies implemented across the country by the central government. It is simply whether constituent unit initiatives go some way to filling a void created by central government inaction.

The chapters in this book provide evidence of such compensating action occurring across a number of federations. Partisanship has been at the heart of the matter – fiercely so in the United States, but quite evidently so in other federations as well. All policy has its ideological inflection, and climate change is no exception. When central governments have been in the hands of those opposed to climate change action, constituent units have often taken the initiative – typically, those of a different ideological complexion. This has been the case in the United States, as Rabe shows; in Australia; in Canada as Harrison shows; in Switzerland as Kammerer and colleagues show; and Brazil as Barbi and Rei show.

For good reason, California is the best-known case of constituent unit environmental activism, but it is anomalous. There are certainly ‘leader’ jurisdictions in other federations – such as the Mexican state of Yucatan, as Corral and Heredia show – but these do not exercise the kind of systemic leverage that California has enjoyed. Prominent among the measures adopted at constituent unit level have been policies to encourage investment in renewable energy and to penalise emissions through cap-and-trade systems. Occasionally constituent units can go further and endeavour to force the central government to act, as a group of US States did via a legal challenge in 2007. However, there is little evidence of this in our other cases.

Does this compensatory action have a perverse effect by letting central governments off the hook? There is some suggestion this might be the case in Switzerland, but in general it would seem not. In the case of the EU, Alberton describes the leadership role played by those more ambitious member states, setting the agenda for the Union as a whole. To what extent has compensatory federalism filled the void created by inaction at the centre? That is a question addressed below.

Institutional realities are also a factor in some of our cases. Rabe reminds us how the presidential separation-of-powers system in the United States has helped ensure that ‘no major federal environmental legislation has been adopted in the United States since the 1990 Clean Air Act Amendments’. Kammerer and colleagues point to Switzerland’s unique system of direct democracy as playing an inhibiting role as far as the federal government is concerned. However, as the Australia case shows, it is quite possible to have inaction prevail at the federal level in a straight parliamentary federation if conservative parties hold office there. In India, the central government has taken the position that the country should not have to compromise on economic development for the sake of climate change mitigation, and has refused to implement such policies unless there are clear co-benefits. Given that India is the third largest emitter, this is clearly an issue. Pillai and Dubash show that while the states have compensated somewhat, their ability to do so is limited by the highly centralised nature of India’s federal system.

16.1.3 Laboratory Federalism

Federalism’s greatest promise for policymaking lies in the potential provided by the existence of multiple governments for policy innovation and accompanying interjurisdictional learning. Learning can take place either horizontally among the constituent units, or vertically from one or more constituent units up to the central government. In climate change governance, though, there is little reason to expect a great deal of genuine policy innovation: most of the mitigation measures have been well known for some time and the issue is not coming up with new techniques, but simply making the decision to adopt and implement them (Engel Reference Engel and Robbins2015). In that regard, federalism still provides conditions for second-order policy learning – or ‘political learning’ (May Reference May1992) – in that pioneering jurisdictions can demonstrate the political and economic feasibility of what might be otherwise eschewed as excessively adventurous initiatives.

The accounts provided here are largely consistent with this more modest interpretation of the laboratory federalism idea. There is little mention of genuine policy innovation, but numerous examples of pioneering adoptions that provide a model for other jurisdictions. Even in as centralised a federation as India, there has been some vertical diffusion of state innovations, note Pillai and Dubash. However, there are certainly examples of innovations that failed to inspire emulation. As Harrison notes, British Columbia’s message that a provincial carbon tax can work fell on deaf ears across the rest of Canada. Québec joined with California in an emissions-trading scheme, but other provinces were not queuing up. In Switzerland, the great variation in circumstances between the cantons discourages interjurisdictional learning as far as climate change policy is concerned. In China, Yi and Cao conclude that there have been many instances of innovation and leadership but not necessarily an accompanying diffusion. Leaders without followers does not make for laboratory federalism. In general, we can say that the experience here reflects the chronic challenges that generally constrain interjurisdictional policy learning: the length of time before the ‘success’ of a policy becomes evident; the absence of an objective measure of success; the role of local conditions, circumstances and interests; the impact of ideology and partisanship.

16.2.1 Veto Points

Federalism has been seen as obstructing policymaking because either the relevant government has inadequate jurisdictional authority or institutions of constituent-unit representation in the central government have a veto power. These cases have highlighted some instances of such obstruction. In Spain, Rodrigo and colleagues note the high degree of jurisdictional conflict that has characterised environmental issues. In the EU, Alberton notes the retarding effect of reluctant member states. In Canada, Harrison points to the ability of status quo provinces to block pan-Canadian action because of a convention requiring intergovernmental unanimity. And when the federal government did eventually take unilateral action, passing the Greenhouse Gas Pollution Pricing Act in 2018, that might have succumbed to constitutional challenge (though it was upheld by the Supreme Court of Canada). Underpinning these dynamics is the degree to which Canada stands out as an unusually decentralised federation – a consequence in the largest part of the degree to which it is bicommunal, with Québec having long played a powerfully centrifugal role (Gagnon and Simeon Reference Gagnon, Simeon, Moreno and Colino2010; Lecours Reference Lecours2019). Other than this and some jurisdictional clashes between the states and the federal government in the USA, though, the federations surveyed in this book reveal little by way of such obstacles to effective climate governance.

Indonesia, though, is a different story. There, Di Gregorio and Moeliono describe a situation where two decades of decentralisation have resulted in a misalignment of roles, responsibilities and resources that hamstrings effective action. As we emphasised in Chapter 1, Indonesia is not a federation and indeed has long evinced an ‘aversion’ to federalism. Its experimentation with decentralised governance is recent and the resulting system must be seen as a work in progress.

While Canadian provinces were able to block central government action because of their power in the federation, they have no formal veto authority through representation in the Canadian parliament. The situation is very different in Germany, where the second chamber of the federal parliament, the Bundesrat (Federal Council) is made up of delegated representatives of the Länder governments. There, the Länder do enjoy a formal veto power and a coalition of coal states could stand in the way of national mitigation strategies. Eckersley and colleagues find little evidence to this point, though, of the much-discussed politikverflechtungsfalle or ‘joint-decision trap’ being an obstacle to federal climate change policymaking in Germany – though they suggest it could become so. The Bundesrat is, however, a very unusually ‘federal’ second chamber, and thus we would not expect this kind of veto point to be a factor in other federal systems. The only other case here with German-style Council governance is the EU, where similar dynamics can be found.

16.2.2 Policy Patchwork

While constituent units may well step up and compensate for central government inaction, quite possibly only some of them will – and to varying degrees. This pattern is evident from the case studies, particularly the USA and Canada, where several jurisdictions have taken notable climate change initiatives, but, equally, a number remained steadfastly attached to the status quo. Eckersley and colleagues describe a similar situation in Germany, Alberton does likewise for the EU, Rodrigo and colleagues do for Spain, and Yi and Cao do for China – reflecting what the latter describe as ‘the conflict between local economic interests and national goals for climate governance’.

In many cases the problem goes beyond the mere existence of diverse responses. Adela and colleagues describe a situation in Ethiopia where spillover problems between ethnically defined constituent units jealous of their powers and interests undermine mitigation and adaptation efforts. Oates (Reference Oates1972) noted that the decentralisation theorem is predicated on costs and benefits of a given policy being internalised, and there being no significant external effects, or spillovers. Decentralisation becomes more problematic the more serious those externalities.

In Canada, the issue is the degree to which those jurisdictions that have not taken action are the ones where action is most needed. As Harrison emphasises, no matter how concertedly Québec or British Columbia might work to reduce their emissions, those efforts cannot compensate for the large and increasing contribution from Alberta. ‘With only 12 percent of the population, Alberta contributes over 40 percent of Canada’s emissions’ (Harrison, Chapter 4 this volume), and its economic welfare is tied to its high-emission industries. As a consequence, Canada’s emissions have continued to rise. Similarly, as Barbi and Rei show, the efforts of the southern states in Brazil cannot compensate for the large and rising emissions of their Amazonian counterparts.

In Chapter 1, we noted the view that such policy discrepancies will be fatal. ‘Attempts to reduce greenhouse gas … emissions by one jurisdiction are meaningless if others allow emissions to increase by an equal (or greater) amount. An effective response … requires vertical coordination’ (Gordon Reference Gordon2015, 122–23). This would, however, seem to exaggerate. If emissions increase in some jurisdictions on a business-as-usual basis while other members of the federation are implementing policies to reduce theirs, those reductions are nonetheless reductions in the net federation-wide output, just as they would be if there were no shirkers. While, as Harrison emphasises, mitigation efforts by BC, Québec and Ontario are overshadowed by Alberta’s large and growing emissions and thus the leaders cannot compensate for the laggards, Canada’s net emissions are still lower than they would be if those three provinces had also continued on a business-as-usual path. None of the achieved reductions is ‘meaningless’. But that’s cold comfort, Harrison argues; in a federation such as Canada’s, the central government’s heavy hand is required if emissions are to be reduced. This would seem to be the case for any federation where there are jurisdictions wedded to emissions-intensive industries of such scale.

A very different scenario is represented by the Australian case, where states have varied in their commitment but not such that some jurisdictions effectively undo the efforts of the others. Because of their coal and LNG exports, Western Australia and Queensland play a role analogous to Alberta’s, but not to the same degree. The difference lies not in the respective federal systems, but in the type and distribution of resources within those systems. Both the fossil fuel resources and the renewable alternatives are more evenly distributed in Australia and none are as dirty to produce as the oil from Alberta’s tar sands.

16.2.3 The Collective Action Problem

Given that no constituent unit ‘acting alone, is even capable of adopting emission controls that would make a dent in global emissions’, and assuming that mitigation efforts are costly, there is good reason to think that all jurisdictions would be tempted to ‘free ride’ on the efforts of others (Adler Reference Adler2008, 448). The studies in this book provide little of evidence, though, of such an effect prevailing. That seems consistent with findings that jurisdictional action on climate change is determined more by internal dynamics than by the external collective action dilemma (e.g., Aklin and Mildenberger Reference Aklin and Mildenberger2020; Matisoff Reference Matisoff2008).