Involvement in International Organizations

With regard to international political integration we assume that countries that are already ‘entangled’ in a larger network of international organizations – for example, the Bretton Woods institutions and specialized agencies of the United Nations (UN) – are more likely to adopt this co-operative behaviour also in other issue areas and with regard to other forms of international co-operation. Our argument postulates, therefore, that more extensive membership in international organizations (IOs) motivates states to behave more co-operatively also when it comes to forms of international co-operation that lie outside the scope of specific international organizations they have joined at some prior time.

It is important to recognize the differences between international organizations and global environmental treaties, the specific form of international co-operation we are interested in. To be recognized as an international organization, an organization has to be highly institutionalized, which is reflected, for example, in the fact that IOs possess a permanent secretariat and hold regular meetings.Footnote 25 This implies that IOs constitute the most formal kind of international co-operation. In addition, IOs are set up to deal with a variety of issues that are usually not connected to environmental topics. In contrast, to be counted as an international treaty, the only requirement that needs to be fulfilled according to the 1969 Vienna Convention on the Law of Treaties is that the agreement is ‘an international agreement concluded between States in written form and governed by international law’ in which states express their ‘consent to be bound’.Footnote 26 Hence, multilateral environmental agreements are usually less institutionalized than international organizations and they are, by definition, limited in their scope to environmental topics.

However, both international organizations and multilateral treaties allow states to co-ordinate their behaviour and thereby achieve benefits from mutual co-operation.Footnote 27 Consequently, we argue that membership in international organizations signals a general willingness of states to behave co-operatively in international matters, which states may also carry over to other very particular issue areas such as environmental policy.

Liberal institutionalism posits that, under conditions of interdependence, uncertainty and high transaction costs, states establish international organizations to facilitate co-operation. That is, IOs, by increasing information and decreasing transaction costs and uncertainty, facilitate international negotiations on new agreements or revision of existing ones, and they reduce the risk of opportunism in implementing international commitments.Footnote 28 More generally, IOs are assumed to move states away from pursuing relative gains and towards positive-sum outcomes, help them overcome collective action problems, and promote shared interests.Footnote 29 In addition, they specify legitimate ways for states to handle domestic and international issues,Footnote 30 encourage effective interstate bargaining,Footnote 31 and assist states in solving complex technical problems in more efficient ways.Footnote 32 Membership in IOs thus signals a government’s general willingness to co-operate internationally and adopt rules and regulations that benefit other countries as well.

Although rational states choose to participate in international environmental agreements only when they estimate that the benefits accruing to them by implementing a particular treaty will be larger than the benefits of unilateral efforts, failure to ratify a treaty could lead to reciprocal actions by other states that would undermine the collective effort.Footnote 33 Reciprocity becomes even more important if we take into account that countries interact in many issue areas at the same time and that these issues are linked. For example, when a country fails to ratify an environmental treaty, it may have to worry about other countries’ refusal to ratify, say, a trade agreement. Moreover, being a member of several IOs but refusing to co-operate in the realm of international environmental co-operation may entail ‘audience costs’ in terms of credibility and reputation losses at home and abroad.Footnote 34 As noted by Simmons, credibility has become essential to successful resolution of some of the most important issues states face today, such as environmental degradation.Footnote 35 We assume that such effects tend to grow in importance as a country joins more IOs, and also because many IOs deal with a wide variety of issues and can thus establish more linkages across issues.

The arguments outlined in the previous paragraph are likely to be of special importance in the field of environmental co-operation. Environmental protection and thus also ratification of multilateral environmental treaties means that countries have to forego certain benefits in other areas, such as economic growth. Consequently, low-income countries in particular may often not be willing to join such treaties. However, if these countries are already entangled in a larger network of international organizations, it may be possible to ‘get them on board’ through issue linkages, assistance, reciprocal action or reputational mechanisms. Recent research in fact shows that membership in international organizations and environmental international non-governmental organizations (EINGOs) plays an important role in environmental protection policies of less developed countries.Footnote 36

In summary, the ‘entanglement’ argument relies both on liberal institutionalist and on what one might call ‘sociological’ assumptions. Greater involvement in international organizations fosters co-operative behaviour by reducing transactions costs and creating opportunities for diffuse reciprocity. It also has a ‘socialization’ effect in terms of creating norms of appropriateness. Hence, being a member of international organizations in general might lead governments to value the more generic benefits of international co-operation and thus to adopt this co-operative behaviour to very different issue areas and with regard to different forms of international co-operation.

Integration in the World Economy

The impact of trade on the environment has been the subject of an extensive and controversial debate in both political and academic circles. Moreover, it is amenable to relatively coherent theoretical arguments and empirical testing. This is why, in examining the effects of countries’ integration into the world economy, we focus on trade effects.Footnote 37

Neumayer, relying mainly on arguments commonly associated with the ‘liberal peace’ and arguments pertaining to reputation, coercion and signalling, posits that trade openness promotes participation in multilateral environmental agreements (MEAs).Footnote 38 However, his empirical findings provide only weak statistical support for this hypothesis and he partly retracts by stating ‘countries’ willingness to cooperate in MEAs depends on whether the MEA under consideration is likely to threaten or accommodate the interests of exporters’.Footnote 39 Prakash and Potoski examine participation rates in ISO 14001, a widely used environmental certification system under which firms can voluntarily subscribe to certain standards of green behaviour.Footnote 40 They find ‘that trade linkages encourage ISO 14001 adoption if countries’ major export markets have adopted this voluntary regulation’.Footnote 41 In contrast, Beron et al. and Wagner, in studies on the Montreal Protocol, find that trade interdependence had no significant effect either on ratification or its timing.Footnote 42

Studies such as these offer important insights into whether and how economic integration affects environmental policy in general and environmental treaty ratification in particular. But these insights remain vulnerable to three types of criticism. First, virtually all studies focus on one or very few environmental policy areas; thus, it remains open whether their findings can be generalized across a wider range of environmental policy issues.

Secondly, to explain co-operation, almost all existing studies rely on the ‘liberal peace’ argument, which postulates that trade between two states increases the economic costs of war for both participants and consequently reduces the probability of conflict.Footnote 43 However, we think that the ‘liberal argument’ is not appropriate for the examination of ratification behaviour because decisions on whether to ratify an environmental treaty depend heavily on domestic considerations concerning the effects on one’s own competitiveness. Our argument on the effect of trade on multilateral environmental treaty ratification thus follows standard trade theory.

Thirdly, the positive effect of trade observed for ISO 14001 is closely connected to the trading-up argument.Footnote 44 This argument holds that greener jurisdictions can ‘export’ their preferences and standards to other countries via trade relationships. However, the trading-up effect is likely to materialize only under quite narrowly defined conditions. In a country-to-country context, the principal trading-up mechanism – market access restrictions for polluting goods imposed by the importing country – operates primarily with respect to the environmental properties of products (for example, cars with or without catalytic converters; that is, product regulation), but to a much lesser extent with respect to production processes.Footnote 45 Many international environmental issues, such as climate change mitigation or marine oil pollution, concern production processes rather than the properties of internationally traded products. We posit that a closer look at standard trade theory is necessary, and that this should make us rather sceptical about optimistic (trading-up) views on the effect of trade on international environmental co-operation.

According to the Hecksher–Ohlin model, trade leads to more production of goods that are intensive in the factor that is abundant in the country concerned. Consequently, comparative advantage derives from the distribution of world endowments of the factors of production (the factor endowment theory). If this assumption is correct, developed countries, which are more capital abundant, may become ‘dirtier’ with free trade because capital-intensive production tends to cause more pollution. Therefore, controlling for other influences (such as income or democracy), they will be more reluctant to participate in international environmental agreements that hamper their comparative advantage in the production of polluting goods (regulatory chill effect). If, however, the comparative advantage derives from policy related differences across countries in tolerance of pollution (the pollution haven or risk-shifting hypothesis), then the less developed countries, which tend to be more labour than capital abundant, are likely to be more reluctant to engage in international environmental co-operation. That is, we should then expect poorer countries to engage in more polluting production as a function of growing international trade due to the pollution haven effect; hence, they should also be more reluctant to ratify environmental agreements that hurt their comparative advantage which derives from laxer environmental regulation.

Depending on whether the factor endowment or pollution haven effect dominates, richer or poorer countries will, for reasons of competitiveness, be more reluctant to join international environmental agreements. Consequently, we expect negative trade effects on average because both the factor endowments and the pollution haven effect push in this direction. In other words, the more intensively a country trades, the greater the loss from a reduction in trade. Environmental regulation increases the costs of producing exportables and thus reduces exports (that is, it acts like a tax on exports). Consequently, the trade-off between gains from a cleaner environment and losses from lower exports is more adverse for more open economies.

Contingent Behaviour

Decisions by countries on whether to ratify international agreements are most probably influenced by what other countries do in the respective policy area. In other words, nations pay attention to or even mimic their peers.Footnote 46 In view of the large game theoretic and institutionalist literature on international co-operation, this claim may sound almost trivial. Surprisingly, however, large-N empirical research on international co-operation has – perhaps because it is so obvious – not paid much attention to contingent behaviour.

An exception is the recent research, primarily under the label of ‘policy diffusion’, that has started to explore the role of contingent behaviour more systematically.Footnote 47 Simmons et al. define diffusion as follows: ‘International policy diffusion occurs when government policy decisions in a given country are systematically conditioned by prior policy choices made in other countries (sometimes mediated by the behaviour of international organizations or even private actors or organizations)’.Footnote 48 Most studies take several diffusion mechanisms into account. For example, Elkins et al. find that coercion and competition play a role in the spread of bilateral trade agreements.Footnote 49 Simmons and Elkins report that both competition and learning matter for economic liberalization.Footnote 50 Henisz et al. observe that coercion, common norms and competition contribute to the spread of market-oriented reforms.Footnote 51 Simmons finds evidence that countries are more likely to make and honour a legal agreement such as the IMF’s Article VIIIFootnote 52 if their neighbour countries are doing so.Footnote 53

Even though all of the above-mentioned mechanisms may play a role in motivating states to ratify international environmental treaties, we suspect that most of them are quite highly correlated and also shaped by other factors that we regard as determinants of ratification behaviour (such as trade openness, democracy, income). We believe that it is reasonable to assume that all of the aforementioned mechanisms are at work more prominently in cases where countries share some common characteristics, such as the same level of economic development and location in the same geographic region. In other words, we argue that any given country’s ratification behaviour is influenced by ratification behaviour in its ‘peer group’.

Our study thus contributes to the increasing literature that addresses what is known as Galton’s problem. As outlined above, Galton’s problem points to the difficulty of assessing whether a government makes a specific policy choice because a specific cause or event impacts on all countries or on a given pair or group of countries; for example, pressures arising from globalization may force it to do so, because of internal characteristics, or because this government follows the specific policy choice of another government with whom it shares some common characteristics. By differentiating between contingent behaviour, the effects of international political and economic integration, and the effects of domestic forces driving treaty ratification, the approach taken in this article takes into account Galton’s problem.

Our hypotheses identify peer group effects in three forms:

While the last three sections have discussed the effects of unit-external or international driving forces, we now turn to internal or domestic determinants of countries’ ratification behaviour.

Income

The existing environmental politics and economics literature concentrates on the effect of income on environmental quality (pollution) rather than treaty ratification behaviour. Translation of its arguments on the income-pollution relationship to ratification behaviour is straightforward, however, because we can assume that countries that are more willing to improve their environmental quality are also more willing to join international treaties in this realm.

The large body of theoretical and empirical literature that focuses on economic determinants of environmental quality has led to the identification of an important empirical pattern, the so-called environmental Kuznets curve.Footnote 54 Many (but not all) forms of environmental degradation first become worse and then improve as income per capita increases. The turning points of the curve vary considerably across pollutants and countries. The standard interpretation of this pattern is that environmental quality is a luxury good in the initial stages of socio-economic development. Poor countries facing a trade-off between protecting the environment and improving material living standards opt for the latter. Once significant gains have been made in living standards, the opportunity cost of stricter environmental policies becomes (relatively) smaller and constituencies are prepared to accept lower economic or personal income growth (the two may not be identical) to obtain less pollution. That is, environmental quality becomes a ‘normal’ good. Assuming that this pattern applies not only to local but also to transboundary environmental goods, and that international treaties are a necessary (though not sufficient) condition for achieving improvements in environmental quality, we expect that a country’s willingness to ratify an international environmental treaty is positively correlated with income. The empirical analysis will examine both linear and non-linear income effects.

Democracy

Many authors have argued that democratic countries are more likely to make credible international policy commitments than their non-democratic counterparts.Footnote 55 The reasons are that democratic institutions are stronger compared to non-democracies, democratic decision-makers are more accountable to their electorates, audience costs in democracies are higher, and so is the transparency. This implies that democratic dyads are more likely to be able to solve transboundary problems through mutual international commitments.Footnote 56 We cannot directly deduce from this argument, however, that (in a monadic sense) democratic countries are more likely to join international agreements. We submit, nonetheless, that democracy is likely to have a positive effect on participation in international environmental agreements, but the reasons are somewhat different from the aforementioned ones. Arguments relating democracy and international environmental commitment can be grouped into demand and supply side arguments.Footnote 57

As to the demand side, democratic political systems offer a much higher degree of civil liberties, such as freedom of speech, freedom of the press and freedom of association.Footnote 58 Such liberties imply that citizens are better informed by independent mass media and other sources (such as NGOs) about environmental problems and government policies. They also have more opportunities to express their opinions freely and organize around alternative political views, and thus they can impose higher audience costs on policy-makers who renege on promises.Footnote 59 Consequently, at any given level of (objective) environmental risk exposure and/or income, public demand by the median voter and/or politically influential interest groups for risk mitigation is likely to be stronger in democracies than in non-democracies. That is, our expectation is that the higher the level of civil liberties, the higher the probability that a country ratifies an environmental treaty.

As to the supply side, many authors have argued that non-democratic political systems are likely to under-provide public goods, including environmental quality.Footnote 60 They are typically governed by small elites that use the resources of their country to generate personal wealth and funnel income from the population into their own pockets. If the costs of stricter environmental policies mandated by an international environmental treaty fall disproportionately on the governing elites in the sense of opportunity costs from spending tax revenue on environmental protection instead of accumulating rents while the benefits are uniformly dispersed throughout the population (for example, cleaner air), then these elites would have little incentive to ratify this treaty. Conversely, the median voter in a democracy incurs lower marginal cost from environmental policies relative to the economic and political elite in non-democracies.Footnote 61

Congleton argues, however, that policy makers may have a short time horizon, leading to less stringent environmental regulation.Footnote 62 Many forms of environmental degradation in fact develop slowly and over long periods of time (for example, climate change, loss of biodiversity, air and water pollution). Hence, their mitigation requires a long time horizon. Assuming that authoritarian rulers tend to have a shorter time horizon, we can conclude that democracies enact stricter environmental regulation than non-democracies. But quite the reverse, one might also argue that elected governments have shorter planning horizons than non-elected governments because of political myopia (maximizing votes at the next election).Footnote 63 Since the social costs of current economic behaviour and political choices often materialize only over the long term and burden future generations and future politicians, democratic leaders may refrain from ratifying international environmental treaties that impose high short-term costs. Their autocratic counterparts, in contrast, do not face democratic elections and can take more costly decisions (stricter environmental policies) with longer term benefits without fear of been punished by myopic voters. Consequently, democracies might be less willing to ratify international environmental treaties.

Overall, public demand for environmental risk mitigation is likely to be stronger in democracies than in non-democracies (demand side). Whether democratic political elites are more inclined to satisfy such demand than non-democratic elites remains an empirical question. The existing literature has, thus far, found mainly positive effects.Footnote 64 Even though these studies concentrate on only one or very few international environmental agreements, we follow those findings and expect (albeit with caution) that more democratic countries are more likely to ratify international environmental agreements.

Power

We expect that power, defined in terms of a country’s economic size and/or population, has a relevant although theoretically ambiguous effect on ratification behaviour. Neumayer, for instance, argues that powerful states are more likely to participate in multilateral environmental agreements ‘in order to demonstrate their importance in world politics, of which the environment represents one part. In other words, important countries want to be seen as good citizens and leaders in world environmental affairs’.Footnote 65 This is only one possibility, and even in this case we suspect that participation is likely to be motivated primarily by countries’ concerns for their own safety and well-being, rather than the desire to be seen as good world citizens by someone else. The other, not so glamorous possibility, is that more powerful states may choose not to ratify because they are likely to get away with such behaviour at lower cost.Footnote 66 In other words, the effect of power is theoretically ambiguous, but potentially important. We control for this effect without a prior assumption concerning its direction.

Domestic environmental quality

Countries’ willingness to participate in international environmental agreements may reflect the degree to which environmental degradation impinges upon their national welfare. Sprinz and Vaahtoranta argue that ‘the worse the state of the environment, the greater the incentives to reduce the ecological vulnerability of the state’.Footnote 67 If we subscribe to this argument, we should expect that countries with bigger domestic pollution problems are more likely to join international environmental agreements because there is greater public demand for more stringent environmental policies, and/or because the government seeks to tie its hands through international commitments in order to be able to impose stricter policies on opposing domestic interests. In addition, we may also expect that states whose environmental quality is improving or has improved as the result of strict domestic environmental regulation are likely to face lower domestic political and economic hurdles in ratifying international environmental commitments. However, given that more polluted countries also face higher abatement costs, it is reasonable to assume that these costs might deter governments from ratifying a cost imposing treaty. Consequently, we expect domestic environmental quality to have an ambiguous effect on ratification behaviour.

Geographic Region

While Hypothesis 3b (see above) focuses on the effect of regional ratification behaviour in terms of contingent behaviour, we also include regional dummy variables to capture any effects that the geographical location of a country might have on its propensity to ratify an agreement. For the empirical analysis, we select from the very large number of international environmental agreements only those that are, in principle, open to all countries globally. However, some agreements may, because of the very nature of the issue they deal with, attract more countries from some regions than from others. For example, African countries might be less eager to join agreements to protect arctic seals and polar bears (even though these agreements are open to them), whereas many Western European countries may be less interested in efforts to cope with desertification.

Dependent and Independent Variables

The following paragraphs briefly describe the data used. Tables 1 and 2 provide descriptive statistics for all variables and the web appendix contains more detailed descriptive statistics for the dependent (ratification) variable.

Table 1

Summary Statistics: Continuous Variables

Table 2

Frequencies of Dummy Variables with Percentages in Parenthesis

The dependent variable is defined in terms of the ratification of a multilateral environmental treaty. We coded ratification in binary form. For each year in which a treaty is not ratified by a particular country, the respective treaty–country pair is coded as zero (0). The dependent variable takes the value one (1) in the year the country ratifies the treaty, whereupon this particular treaty–country pair leaves the dataset. The latter approach is necessary because leaving the treaty–country pair with code 1 in the dataset after ratification would indicate that the respective country ratified the treaty again and again in each subsequent year until 2000 (where our dataset ends). The data on treaties and ratifications were retrieved from the datasets by CIESIN and Mitchell.Footnote 69 We then eliminated from this data treaties that, in our assessment, are not environmental treaties or deal with environmental issues only at the margin, and treaties that are not open to all countries globally; and we re-coded the remaining data to fit the particular format used for this analysis. Our dataset includes 255 treaties.Footnote 70 Whether particular treaties are not only in principle but also de facto open to all countries globally is not entirely clear in some cases. We have thus assessed the robustness of our statistical findings with alternative sample compositions (presented in the web appendix).

Control Variables