Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS ( (c) Oxford University Press USA, 2019. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy and Legal Notice).

date: 16 October 2019

The Politics of Trade and Climate Change

Summary and Keywords

Both trade and climate change policies affect the international competitiveness of carbon-intensive industries. This suggests that policy changes in one area may affect politics in the other. Does openness to international trade affect climate change politics? Do climate change policies affect the politics of trade? Does formally linking trade and climate policies via trade sanctions affect the prospects for cooperation in each domain? There are good theoretical reasons to believe that the answer to these questions is yes. Theoretically, each set of policies should affect the other, but these interactions could either encourage or discourage trade and climate cooperation. How trade and climate politics interact is thus an empirical question. Empirically, the overall picture is of a nascent but promising field of research. Extant studies provide indirect tests and suggestive evidence, but little in the way of firm conclusions. Only one point emerges clearly: progress in this area will require more and better data on national climate policies.

Keywords: trade, climate change, environment, issue linkage, globalization

In 2009, the U.S. House of Representatives took up the American Clean Energy and Security Act—which, in the early 21st century, remains the United States’ sole effort to pass comprehensive climate change legislation. During the debate over this bill, concerns about international competitiveness loomed large. For example, Representative Sander Levin (D-Michigan) argued: “As we act, we can and must ensure that the U.S. energy-intensive industries are not placed at a competitive disadvantage by nations that have not made a similar commitment to reduce greenhouse gases” (Broder, 2009). Although the Senate ultimately did not consider the bill, Senators Lindsey Graham (R-South Carolina) and John Kerry (D-Massachusetts) also highlighted trade concerns in a jointly written op-ed in the New York Times:

We agree that climate change is real and threatens our economy and national security . . . [but] we cannot sacrifice another job to competitors overseas . . . There is no reason we should surrender our marketplace to countries that do not accept environmental standards. For this reason, we should consider a border tax on items produced in countries that avoid these standards. This is consistent with our obligations under the World Trade Organization and creates strong incentives for other countries to adopt tough environmental protections.

(Kerry & Graham, 2009)

These statements illustrate three common views about international trade and climate change. First, domestic carbon restrictions—if adopted unilaterally—place domestic industries at an international competitive disadvantage. Second, trade barriers on carbon-intensive imports can level the playing field. Third, the strategic use of such barriers can incentivize other countries to tighten their environmental standards. Whether or not these beliefs are correct, they are prevalent. And their prevalence implies that the politics of trade and of climate change are strongly linked.

Does international trade affect the politics of climate change? Do climate change measures affect the politics of trade? These questions are important, given the salience of climate change and ongoing efforts to liberalize global trade. Yet neither the vast literature on trade politics nor the mounting research on climate change answers them conclusively. This does not reflect a lack of interest: a 2014 poll of international relations scholars found that global climate change was overwhelmingly considered the most important policy issue facing the world in the next ten years (Maliniak, Peterson, Powers, & Tierney, 2014). However, data constraints have limited research on trade-climate policy linkages. As a consequence, we simply don’t know much about causal connections between these policy domains.

This article proposes some hypotheses relating trade and climate policy and discusses the empirical evidence for each. It does not attempt to comprehensively review the literature on climate politics, as this has been done (Bernauer, 2013; Harrison & Sundstrom, 2007; Purdon, 2015). Nor does it devote much space to purely economic links between trade and climate—for example, Grossman and Krueger’s (1993) well-known “scale,” “composition,” and “technique” effects—as these have been studied extensively by economists (Copeland & Taylor, 2004; Tamiotti et al., 2009). Instead, this article addresses three inherently political questions. First, does trade liberalization affect the political feasibility of greenhouse-gas regulations? Second, do greenhouse-gas regulations affect the political feasibility of trade liberalization? Third, would explicitly linking the two policy domains—for example, by using trade sanctions to enforce climate agreements—affect progress in either area? Theory suggests that the answer to all three questions could be yes. However, empirical research on these questions is nascent and suggestive at best. There is therefore tremendous scope for further empirical research.

This article begins by discussing domestic links between trade and climate policies, then turns to international links. Because trade is inherently international, this distinction may seem artificial, and to some extent it is. Nonetheless, it is useful to distinguish two different ways in which trade and climate policies might interact. In the domestic case, the home country’s trade policies and climate policies affect each other even if neither is tied to policies in other countries. For example, if the home country’s tariffs are independent of other countries’ climate policies but nonetheless affect its adoption of carbon restrictions, this is considered a domestic link. In contrast, international links involve some connection between home and foreign policies. For example, if the home country makes its tariffs contingent on foreign climate policies, and if this home–foreign policy linkage affects domestic climate politics, this is considered an international link. After discussing theoretical reasons why each type of link might matter, the article discusses the empirical evidence for each and offers some concluding remarks.

Domestic Links Between Trade and Climate Politics

Why might trade policy affect climate politics, and vice versa? The short answer is that both sets of policies affect the relative competitiveness of domestic and foreign producers. Domestic carbon restrictions—carbon taxes, cap-and-trade systems, petrol taxes, and so on—raise production costs for domestic carbon-intensive industries. This puts them at a competitive disadvantage relative to unregulated foreign producers, which may affect their incentives to seek protectionist trade policies. Conversely, trade liberalization may increase competition from lower cost imports, which could affect the incentives of carbon-intensive industries to oppose domestic carbon restrictions.

Before exploring these connections in greater depth, several scope limitations should be noted. First, the analysis here focuses on trade policies that protect the domestic market from imports (e.g., tariffs and quotas) and does not consider export subsidies, import subsidies, or export taxes. This is not because the latter are unimportant—although import subsidies and export taxes are rare—but because import protection is the focus of most trade-and-climate discussions. Second, the analysis focuses on trade in carbon-intensive goods—that is, goods whose production generates heavy carbon emissions—and does not consider the politics of trade in fossil fuels themselves.1 Although the latter is important to climate change and may affect climate politics in counterintuitive ways (Harrison, 2015), the issues raised by fossil-fuel trade are distinct from those raised by carbon-intensive production. This article thus focuses on the latter for analytical clarity. Finally, the analysis adopts a sector-specific, or Ricardo-Viner, approach to analyzing trade and climate policies, focusing on the interests of particular industries rather than on those of broad factor classes. This is largely because empirical work shows that trade has had sector-specific effects in recent years (Autor, Dorn, & Hanson, 2016; Hiscox, 2002). Claims about the interests of carbon-intensive industries could therefore refer to their workers, their capital owners, or both.

The links between trade policy and climate policy could be either static or dynamic. In this article, static means that a change in one set of policies immediately affects the incentives to lobby for or against the other, taking the economy’s current industrial structure as given. Dynamic means that each set of policies eventually changes an economy’s industrial structure in ways that may alter the political support for the other. This analysis begins with static linkages, then turns to dynamic ones.

To help motivate the discussion of static linkages, consider the economic effects depicted in Figure 1. The downward-sloping solid line D represents domestic demand for a carbon-intensive good; the upward-sloping solid line S represents the domestic industry’s supply curve, and the solid horizontal line PW depicts the world price of this good. The world price is below the domestic autarky price, so this country has a comparative disadvantage in the carbon-intensive good. This reflects the focus on import protection.

The Politics of Trade and Climate ChangeClick to view larger

Figure 1. Import tariffs, carbon taxes, and producer surplus.

Consider first the impact of imposing a carbon tax t in an open economy.2 The carbon tax increases the domestic industry’s production costs and shifts its supply curve inward, from S to S + t. This reduces the industry’s producer surplus from area a + b to area a, as consumers shift to lower cost imports. The industry’s loss from the carbon tax is thus area b. It could recoup some of this loss by lobbying for an import tariff, τ‎, that raises the domestic price of imports to PW + τ‎. This increases the industry’s producer surplus by area c. This implies that import tariffs could compensate domestic carbon-intensive industries for losses imposed by carbon restrictions. If this occurs in practice, domestic carbon restrictions will lead to higher import tariffs. Let us call this the tariff-compensation hypothesis, since tariffs compensate domestic industries for costly environmental protection.

Now consider the effect of tariff liberalization in a regulated economy that maintains a carbon tax t. Removing the tariff τ‎ reduces the price of imports from PW + τ‎ to PW. This reduces domestic producer surplus from area a + c to area a, as consumers again shift to lower cost imports. The industry can recoup some of this loss if it successfully lobbies against the carbon tax, which would shift the supply curve outward to S and increase producer surplus by area b. In other words, domestic industries could seek carbon deregulation as compensation for losses imposed by trade liberalization. If this occurs in practice, then trade liberalization will lead to weaker carbon restrictions. Let us call this the carbon-compensation hypothesis, since governments allow domestic industries to emit more carbon to compensate them for costly trade liberalization.

Note that the carbon-compensation hypothesis is similar but not identical to the familiar race-to-the-bottom argument. Both assert that international competitive pressures lead governments to relax domestic environmental regulations. However, the carbon-compensation hypothesis is agnostic about why those pressures exist: foreign producers could have cost advantages due to factor or resource endowments, domestic economies of scale, or other reasons. In other words, this hypothesis asserts only that (a) governments relax domestic regulations to help their industries compete against foreign producers, and (b) that a reduction in trade barriers increases the incentives to do so. Trade liberalization, not foreign environmental policy, is the “moving part” in this argument. In contrast, the race-to-the-bottom argument states that the home government weakens domestic environmental regulations in response to weak regulations abroad. International trade plays a role, but it is a mediating variable between home and foreign environmental policies, and one that scholars often implicitly treat as fixed (i.e., trade is assumed to be open). In this argument, foreign environmental regulations are the moving part. Because the race-to-the-bottom argument posits a link between home and foreign environmental policies, it is discussed in the section on international linkages, “International Links Between Trade and Climate Politics.”

In the foregoing arguments, tariffs can be raised and lowered, and carbon restrictions can be tightened or weakened. However, in the real world, tariffs are bound by the World Trade Organization (WTO), and most countries have not adopted carbon taxes or other broad-based greenhouse-gas regulations. This means that certain outcomes—carbon regulations leading to large tariff hikes or trade liberalization causing carbon deregulation—are unlikely to occur. Nonetheless, this caveat does not alter the intuition behind these compensation effects. Although new greenhouse-gas regulations might not cause tariffs to rise, they could impede further trade liberalization—a comparable outcome in a world of downward-trending tariffs. And while trade liberalization might not cause governments to abandon (nonexistent) carbon restrictions, it could make adopting the latter politically more difficult—also a comparable outcome when a growing number of governments seek to mitigate climate change. In these plausible scenarios, both compensation hypotheses still imply a positive relationship between tariffs and greenhouse-gas restrictions.

Although these hypotheses seem reasonable, they do not emerge unambiguously from Figure 1. In fact, the figure could motivate contradictory predictions. Note that if a carbon tax is in place, imposing a tariff increases producer surplus by area c. However, in the absence of a carbon tax, the tariff increases producer surplus by area c + d. In other words, the marginal gain from the tariff—and presumably the return to protectionist lobbying—is higher when no carbon tax exists. Intuitively, the domestic market is more valuable without the carbon tax, so producers have stronger incentives to seek tariff protection. This implies, counterintuitively, that stricter carbon restrictions could lead to lower tariffs. According to this tariff-reduction hypothesis, tariffs and carbon restrictions should be negatively related.

Similarly, if no tariff exists, removing the carbon tax increases producer surplus by area b. However, if a tariff is in place, removing the carbon tax increases producer surplus by b + d. The marginal gain from carbon deregulation is higher when trade barriers exist. Intuitively, higher tariffs make the domestic market more valuable, increasing the incentives to lobby against a carbon tax. In this carbon-regulation hypothesis, trade liberalization leads to stronger carbon restrictions, producing a negative relationship between tariffs and carbon restrictions.

As this discussion indicates, tariffs and carbon restrictions could in theory be either positively or negatively related. Which relationship prevails depends on whether carbon-intensive industries and their political representatives treat the two policies as complements or substitutes. The word “treat” is critical because, in a purely economic sense, tariff protection and carbon deregulation are complements: the former raises the returns to the latter, and vice versa. Hence, if industries simply maximize income in an ahistorical, context-independent way, they should lobby for policies that produce a negative relationship between tariffs and carbon restrictions. However, it is easy to imagine how history, context, or both could produce the opposite result.

History might matter because, as behavioral economists note, individuals do not always exhibit the well-behaved utility functions of standard economic models. For example, Kahneman and Tversky (1984) argue that people are often loss averse, seeking above all to minimize losses relative to the status quo. This means that history matters: instead of seeking the best of all possible outcomes at all times, individuals may “settle” for the historically derived status quo and only mobilize politically when their current incomes are threatened. If industries behave this way, then losses in one policy domain—due to trade liberalization or environmental protection—would not lead them to scale back their efforts to deregulate or protect the now-less-valuable domestic market. Rather, loss-averse industries would attempt to offset losses in one domain with gains in the other. This would produce compensation effects and a positive relationship between tariffs and carbon restrictions.

The political context can also produce such a relationship. For example, legislators in “green” districts might face strong incentives to support climate legislation. However, legislators from “dirty” districts might oppose such legislation as detrimental to their local industries. Green legislators might support tariffs for carbon-intensive industries—tariffs they might otherwise oppose—to obtain dirty legislators’ support for carbon restrictions. Conversely, dirty legislators might acquiesce to carbon restrictions as a quid pro quo for higher tariffs. In this case, green legislators’ need to build a legislative coalition would produce a tariff-compensation effect.

The same logic extends to societal coalitions. Although the discussion has focused on industry pressures, environmental groups provide the impetus for climate legislation. Moreover, these groups often voice concerns about carbon leakage and hence international trade. For example, the Citizens’ Climate Lobby, a grass-roots environmental movement, explicitly advocates border taxes to ensure that carbon restrictions do not produce carbon leakage.3 This suggests that a coalition of industry and environmentalists might support protectionist policies that neither group could obtain alone. Such a “bootlegger and Baptist” coalition would also lead to tariff compensation.

Finally, positive and negative relationships could potentially coexist. For example, consider two economies: a clean one with strict carbon regulations and a dirty one with weak regulations. Carbon-intensive industries in the dirty economy might lobby harder for tariffs than those in the clean one because the former have more to gain from trade protection, as in the tariff-reduction hypothesis. However, a tightening of carbon regulations in both countries could lead loss-averse industries in both to seek higher tariffs, as in the tariff-compensation hypothesis. In this example, a cross-national comparison would support the tariff-reduction hypothesis, while within-country over-time analyses would support the tariff-compensation hypothesis. Although the asymmetries in this example—industries responding one way to regulatory levels and another way to changes—are not based on a “clean” economic model, they seem empirically plausible and have some basis in behavioral economic research. To the extent this is true, empirical tests of these hypotheses might need to examine both cross-sectional and temporal variations.

In sum, the relationship between tariffs and carbon restrictions is theoretically ambiguous. Even if we restrict our focus to static effects, theory alone cannot tell us whether these policies should be positively or negatively linked. This theoretical complexity grows when we consider dynamic effects.

In the long run, trade and climate policies affect not only industry incentives to lobby but also the size of the lobbies themselves. By altering the incentives to invest in clean (non-carbon-intensive) and dirty (carbon-intensive) sectors, such policies cause each sector to grow or decline over time. These long-term structural changes could eventually alter the balance of political forces, with consequences for policies in both domains.

Suppose a country has a comparative advantage in clean goods. In the long run, trade liberalization should lead to growth in this country’s clean industries and to decline in its dirty ones. If an industry’s political clout reflects its economic size—either because of employment or financial resources—then this long-term process should reduce opposition to climate legislation. In this case, trade liberalization would eventually promote carbon restrictions. Let us call this the dynamic green-trade hypothesis, since trade liberalization promotes environmental protection over time. It implies a negative long-run relationship between tariffs and carbon restrictions.

The opposite would occur if a country had a comparative advantage in dirty goods. In this case, trade liberalization would promote growth in dirty industries, which would eventually increase resistance to carbon restrictions. In this dynamic dirty-trade hypothesis, liberalization leads over time to weaker environmental protection. This implies a positive long-run relationship between tariffs and carbon restrictions.

Conversely, carbon restrictions should lead to the decline of dirty industries and the growth of clean ones: this is, after all, their main rationale. How this affects trade policy depends again on a country’s comparative advantage. If a country has a comparative advantage in clean goods, the process should strengthen the pro-trade lobby, promote liberalization, and produce a negative long-run relationship between carbon restrictions and tariffs. Call this the dynamic-liberalization hypothesis. Conversely, if a country has a comparative advantage in dirty goods, this process should weaken the pro-trade lobby, discourage liberalization, and produce a positive long-run relationship between carbon restrictions and tariffs. Call this the dynamic-protectionism hypothesis.

These hypotheses are summarized in Table 1. The top half of the table (A) summarizes possible effects of tariffs on carbon restrictions. The bottom half (B) summarizes possible effects of carbon restrictions on tariffs. The first column indicates the hypothesis; the second shows the predicted direction of each variable’s effect (positive or negative); the third indicates the time frame (short run or long run) in which these effects should occur; the last summarizes the conditions under which we would expect each type of effect.

Table 1. Domestic Politics of Trade and Climate Change

A. Effects of Tariffs on Carbon Restrictions (CRs)


Direction of Effect

Time Frame




Short run

Tariffs and CRs treated as substitutes


Short run

Tariffs and CRs treated as complements

Dynamic Green-Trade

Long run

Comparative advantage in clean goods

Dynamic Dirty-Trade


Long run

Comparative advantage in dirty goods

B. Effects of Carbon Restrictions on Tariffs


Direction of Effect

Time Frame




Short run

Tariffs and CRs treated as substitutes


Short run

Tariffs and CRs treated as complements


Long run

Comparative advantage in clean goods



Long run

Comparative advantage in dirty goods

These hypotheses are not meant to be exhaustive. This simple framework could be extended in a number of ways, for example, by asking how these relationships might be influenced by domestic political institutions.4 Nonetheless, the discussion illustrates two points. First, there are multiple ways in which trade and climate politics could interact. Second, these interactions could produce either a positive or a negative relationship between trade barriers and carbon restrictions. Both points underscore the need to study these connections empirically.

International Links Between Trade and Climate Politics

The discussion has so far focused on how changes in domestic trade policies could affect domestic climate policies, and vice versa. However, neither set of policies is formed in an international vacuum. Governments adopting climate change measures are keenly aware of how other countries’ environmental regulations—or lack thereof—may affect their home industries’ international competitiveness. They might therefore condition their domestic trade policies on other countries’ climate policies—for example, imposing higher tariffs against countries with lax carbon regulations. This would amount to a trade sanction against environmental “laggards,” which could incentivize the latter to adopt stricter carbon restrictions. Foreign climate policies might thus affect domestic trade policies, and vice versa.

In fact, proponents of carbon tariffs—border taxes on carbon-intensive imports from laggard countries—typically cite this incentive effect as one rationale. As noted earlier, Senators Graham and Kerry advocated a border tax in part because it “creates strong incentives for other countries to adopt tough environmental protections” (Kerry & Graham, 2009). Similarly, a recent proposal by the Climate Leadership Council—a conservative group that included former secretaries of state James Baker and George Shultz, former treasury secretary Henry Paulson, and former chairmen of the President’s Council of Economic Advisers Martin Feldstein and Gregory Mankiw—advocated a tax on carbon-intensive imports to “protect American competitiveness and punish free-riding by other nations, encouraging them to adopt carbon pricing of their own” (Baker et al., 2017, p. 1). Notably, these dignitaries are all pro-trade. That they nonetheless support targeted border taxes underscores the broad appeal of tying domestic trade policies to foreign climate change measures.

The argument for explicitly linking trade and climate policies is straightforward. International trade agreements can be effectively enforced through reciprocal trade policies: members can reward foreign liberalization with domestic liberalization and punish foreign protection with domestic protection. Such tit-for-tat strategies are possible because trade policies are targetable: a country can open its market to a single trading partner or levy tariffs on only one partner’s goods. Most scholars believe that the use or promise of such targeted rewards and punishments has helped sustain a high level of trade cooperation (Keohane, 1986; Maggi, 1999).

International climate agreements cannot be enforced in this way. On the one hand, rewards—in the form of carbon restrictions and climate change mitigation—cannot be targeted at cooperative countries. If a country cuts its carbon emissions, it rewards all other countries—including polluters—with a more stable climate. This encourages free riding and undermines cooperation. On the other hand, punishments—increased carbon emissions and worsening climate change—also cannot be targeted at particular offenders. A country that spews carbon punishes all countries, including itself. This inability to target rewards and punishments almost certainly impedes climate cooperation.

Linking trade and climate policies offers a possible solution. Climate “leaders” could employ trade policy as both carrot and stick, rewarding other leaders with market access and punishing laggards with high tariffs. This solution has several strengths. First, the threat to impose trade sanctions against environmental laggards is credible. Within leader states, such sanctions would be supported by both domestic industries seeking import protection and environmental groups seeking to reduce carbon leakage. For these reasons, “most studies of environmental treaty formation now assume that trade sanctions are self-enforcing” (Cirone & Urpelainen, 2013, p. 315). Second, because trade sanctions are both targetable and credible, they should create incentives to comply with international climate agreements. As Barrett’s (1997, p. 347) game-theoretic analysis indicates, “The credible threat of imposing trade sanctions may be sufficient to deter [environmental] free-riding completely.” Third, trade sanctions may increase political support for climate agreements. Bechtel and Scheve’s (2013, p. 3) large-scale experimental survey shows that “the public prefers a small sanction over an agreement that involves no sanction for failing to meet emission reduction targets.” Similarly, Tingley and Tomz (2013) find broad public support for punishing climate laggards with economic sanctions. These results suggest that incorporating trade sanctions into climate agreements may make the latter more politically feasible in leader states.

This discussion suggests two testable hypotheses. The first, which builds on the last point, is a linkage-support hypothesis: Climate agreements that incorporate trade sanctions are more likely to be ratified by states with ambitious climate policies than are agreements without sanctions. The second is a linkage-enforcement hypothesis: Climate agreements that incorporate trade sanctions should enjoy higher compliance rates than agreements without sanctions. Both hypotheses imply that formally linking trade and climate policies could facilitate climate cooperation.

Despite the appeal of such arguments, the case for linking trade and climate policies is not ironclad. Opponents of this idea point to potentially negative consequences in both policy domains. First, however justified they may be, trade sanctions constitute protectionism. Climate agreements that permit trade sanctions may thus worry free traders. Moreover, governments sometimes invoke environmental concerns to mask what is really interest-group protectionism (Henson & Loader, 2001; Kono, 2006; Otsuki, Wilson, & Sewadeh, 2001). This raises the concern that governments could exploit trade-climate linkages to evade the rules of the World Trade Organization (WTO). We must therefore consider a linkage-abuse hypothesis: Climate agreements that incorporate trade sanctions will lead to more widespread protectionism than ones without such sanctions or no agreements at all.

A second concern is that more effective enforcement of climate agreements could deter participation at the negotiation stage. As Fearon (1998) observes, a high likelihood of tit-for-tat retaliation may lead to better treaty enforcement. However, for the same reason, it may discourage states from making binding commitments at the negotiation stage. Rosendorff and Milner (2001) similarly argue that inflexible agreements with little “wiggle room” are harder to ratify than more flexible agreements. These arguments suggest a linkage deterrence hypothesis: Climate agreements that incorporate trade sanctions are less likely to be ratified by laggard states than agreements without sanctions.

Finally, Cirone and Urpelainen (2013) argue, counterintuitively, that trade sanctions could reduce the incentives for climate cooperation. The logic is as follows: Trade sanctions increase the efficacy of unilateral climate action by reducing carbon leakage. At the same time, they reduce the cost of unilateral action by shielding domestic industries from international competition. By increasing environmental benefits and reducing political-economic costs, trade sanctions boost the incentives for leader states to take unilateral climate action. This, however, reduces the credibility of leaders’ threats to renege on international climate agreements. Knowing this, laggard states have stronger incentives to free ride on leaders’ efforts. The result could be weaker climate cooperation.

Although Cirone and Urpelainen raise an interesting point, they do not directly challenge the claim that trade sanctions can help enforce climate agreements. This is because, in their model, the presence or absence of trade sanctions does not depend on other countries’ climate policies. Put differently, they are not sanctions at all, since they are not used to punish polluters. Punishment is instead carried out via climate policy: “Players adopt environmental policy until one defects, at which point both players refuse to abate [emphasis added]” (Cirone & Urpelainen, 2013, p. 319). Although this is a reasonable modeling choice, the model consequently does not address the claim that trade sanctions can deter environmental defections. Nonetheless, it generates an important insight concerning the effects of trade–climate linkage: Adopting carbon tariffs may encourage leader states to take climate action but at the same time discourage laggard states from following suit. This prediction is observationally equivalent to the linkage-support and linkage-deterrence hypotheses.

Although these linkage hypotheses seem straightforward, several caveats are in order. First, though they refer to characteristics of climate agreements (the presence or absence of trade sanctions), agreement-level sanctions remain but a theoretical possibility: no climate agreement to date authorizes retaliatory sanctions. Realistically, then, the linkage hypotheses refer to trade sanctions incorporated into national climate legislation. This blurs the line between the domestic and international arenas: for example, both the linkage-support and the carbon-compensation hypotheses predict that domestic trade barriers increase support for domestic carbon restrictions. Nonetheless, these hypotheses are not identical: linkage support requires a formal link between domestic trade and foreign climate policies; carbon compensation requires no such link.5 And other linkage hypotheses are more clearly international: for example, linkage deterrence predicts that trade sanctions in leaders reduce support for climate commitments in laggards.

Second, while some linkage hypotheses refer to the likelihood of making international commitments, they could also refer to commitment depth. For example, linkage deterrence could alternatively predict that trade sanctions lead to weaker commitments by laggard states. This formulation may make more sense when climate agreements allow different obligations for different states.

Finally, this article has assumed that climate-related trade sanctions are possible. However, their legality under the WTO remains unclear (Jaspers & Falkner, 2013; Kaufmann & Weber, 2011). As this question may not be resolved for some time, note simply that the feasibility and empirical relevance of these trade–climate linkages may ultimately depend on future WTO rulings.

No analysis of trade–climate linkages would be complete without a discussion of the race-to-the-bottom argument and its counterpart, Vogel’s (1995) “California effect.” These arguments are being considered last because they are analytically distinct from previous ones. Whereas the previous arguments treat trade policy as either an independent or a dependent variable, the race-to-the-bottom and California-effect theories treat trade as a mediating variable between home and foreign environmental regulations. That is, such regulations affect each other to the extent that home and foreign industries compete via international trade.

The race-to-the-bottom argument states that governments—or rather, the industries they represent—compete with each other by lowering their regulatory standards. Moreover, this regulatory competition arises from and grows with international trade. The race-to-the-bottom hypothesis thus states that weak carbon restrictions abroad lead to weak restrictions at home, and that this tit-for-tat behavior grows stronger with the degree of trade competition. Note that this hypothesis implies more than just a positive relationship between home and foreign carbon restrictions. The positive relationship must also reflect downward competition; that is, environmental laggards must lead the race to low standards.

Arguments for a California effect turn this logic on its head, stating that environmental laggards face pressures to adopt the standards of leaders if they must do so to sell their goods in the leaders’ markets. This is only true, of course, if leaders restrict imports that do not meet their domestic environmental standards. Because WTO law has historically prohibited process standards, scholars often assume that the California effect applies only to product standards—and indeed, this is where evidence for California effects tends to be found (Prakash & Potoski, 2006). This is important to trade–climate debates, since most emissions are generated by the production rather than the consumption of carbon-intensive goods.

Nonetheless, Urpelainen (2011) argues that California effects can occur for process standards if economies are deeply linked via trade. This is perhaps not surprising, as his model assumes that governments can apply both process and product standards to imports, with equivalent effects. While this does not resolve the legal questions surrounding process standards, the abovementioned linkage arguments similarly assumed that governments can use process standards like carbon tariffs. It therefore makes sense to add the California-effect hypothesis to the list: stricter carbon restrictions in leaders promote stricter regulations in laggards if (a) leaders employ trade sanctions against laggards, and (b) laggards depend on leaders’ export markets.

These hypotheses are summarized in Table 2. The first column lists the hypotheses; the second describes the independent variable; the third describes the mediating variable (if any); the fourth describes the dependent variable, and the last indicates the hypothesized relationship between the independent and the dependent variables.

Table 2. International Politics of Trade and Climate Change


Independent Variable

Mediating Variable

Dependent Variable

Direction of Effect


Trade sanctions in leader climate legislation (all linkage hypotheses)

Likelihood/depth of leader climate commitments



Laggard compliance with climate commitments



Leader trade barriers



Likelihood/depth of laggard climate commitments


Laggard carbon restrictions (CRs)

Leader-laggard trade competition

Leader CRs



Leader CRs with trade sanctions

Laggard dependence on leader market

Laggard CRs


As before, these hypotheses are not exhaustive. For example, it is possible that linking trade and climate negotiations could lead to deeper cooperation in both areas by permitting governments to exchange concessions across issue areas. This type of issue linkage has promoted cooperation in other policy domains (Davis, 2004; Keohane, 1984; McKibben, 2015). However, it would require climate talks to be linked with broad-based trade negotiations, which seem unlikely anytime soon. Instead of considering all possible connections between trade and climate policies, this article focuses on the most obvious and currently plausible ones.

Empirical Evidence

Efforts to test the hypotheses presented here face two empirical challenges. First, we still lack comprehensive cross-national time-series datasets on climate policies. As Bernauer (2013, p. 435) has noted, systematic research on national climate policies “requires, first and foremost, better data on climate policies, particularly panel data . . . [Existing datasets] offer aggregate measures of ambition levels of climate policies but no information on the adoption of specific climate policy instruments.” Scholars have begun to respond to this need: for example, Steves and Teytelboym (2013) developed the Climate Laws, Institutions and Measures Index (CLIMI) as a comprehensive climate policy measure. However, it provides only a cross-section for 95 countries and does not permit panel analysis. Moreover, as the authors acknowledge, “CLIMI measures the policies that countries have adopted to mitigate climate change, but does not provide an assessment of the quality of implementation of those policies” (p. 8, emphasis in original).

This data problem partly reflects the multidimensional nature of climate policies. A wide range of policies influence carbon emissions: broad-based carbon taxes, cap-and-trade schemes, fuel-efficiency standards, clean energy (or fossil fuel) subsidies, and so on. An ideal measure of carbon regulations would aggregate these policies into a single comparable metric, such as a national carbon tax equivalent, but this is challenging and has not been done. A second problem is that explicit climate-mitigation policies are a recent development: broad-based carbon taxes and cap-and-trade programs did not appear until the 1990s, and in most cases later. This further impedes efforts to conduct longitudinal analyses. Finally, these data constraints become still more severe if we are interested in the relationship between trade and climate policies. The European Union (EU) countries have adopted the lion’s share of ambitious climate policies, but these countries must typically be omitted from analyses of trade policy because they share a common external tariff and do not set their trade policies independently (Weinberg, 2016).

For this reason, studies of the trade–climate relationship tend to focus on links between trade flows and carbon emissions rather than on trade and climate policies (Aklin, 2016; Cole, 2004; Cole & Elliott, 2003; Frankel & Rose, 2005; Kleemann & Abdulai, 2013; Managi, Hibiki, & Tsurumi, 2009). While these studies have told us much about the trade–emissions relationship, they tell us little about that between trade and climate policies. Take, for example, the finding that trade openness has reduced CO2 emissions in rich countries but increased them in poor ones (Aklin, 2016; Cole, 2004; Kleemann & Abdulai, 2013; Managi et al., 2009). If we assume that rich countries have a comparative advantage in clean goods, whereas poor countries have one in dirty goods—which seems likely, given trends in the emissions embodied in international trade (Davis & Caldeira, 2010; Peters, Minx, Weber, & Edenhofer, 2011)—this result is consistent with the dynamic environmental effect. That is, perhaps trade liberalization has led to the growth of clean industries in rich countries and of dirty industries in poor countries, which have pushed for stronger and weaker carbon restrictions, respectively. However, while this interpretation is plausible, there is a simpler and more plausible one: namely, this result simply reveals a composition effect (Grossman & Krueger, 1993). If trade liberalization had such effects on clean industry and dirty industry output, it would produce the observed changes in emissions even in the absence of an environmental policy response. As this example shows, we cannot test the political hypotheses presented here without data on actual climate policies.

The second challenge is endogeneity. Endogeneity concerns are endemic to social science, but they are particularly acute in this area because we have strong theoretical reasons to expect bidirectional causation. For example, both the carbon-compensation and the tariff-compensation hypotheses predict a positive relationship between tariffs and carbon regulations: in the former, trade liberalization weakens carbon regulations; in the latter, carbon regulations impede trade liberalization. Finding a positive relationship would therefore reveal little about the direction of causation. Endogeneity problems could also produce spurious nonresults. For example, the linkage-support hypothesis states that climate-related trade sanctions permit deeper climate commitments. However, we may not observe this relationship if both sanctions and commitments are influenced by domestic carbon-intensive industries. These industries might lobby for trade sanctions and against climate commitments, weakening or even reversing the relationship between the two. This result would be misleading if it obscured a positive effect of sanctions on commitments.

Previous work on environmental policy shows that such endogeneity can strongly influence our results. For example, both Ederington and Minier (2003) and Levinson and Taylor (2008) find that when industry-level environmental regulations are treated as exogenous, they have no effect on net imports. However, when these regulations are treated as endogenous, they lead to significantly higher net imports. This difference arises because internationally uncompetitive industries lobby harder against environmental regulations, producing a negative association between imports and environmental regulations that offsets the positive effect of the latter on the former. This example illustrates the importance of addressing endogeneity problems before reaching any conclusions about the causal links between trade and climate politics.

Given these challenges, what do we know about these political links? As before, this review begins with domestic linkages before turning to international ones, classifying studies as “domestic” if they examine the relationship between domestic trade and environmental policies, and as “international” if they involve a link between policies in one country and those in another.

Despite the aforementioned caveats, it is not entirely true that the economic research on trade and the environment tells us nothing about politics. For example, though the finding that trade openness has reduced CO2 emissions in rich countries but increased them in poor ones (Aklin, 2016; Cole, 2004; Kleemann & Abdulai, 2013; Managi et al., 2009) does not directly support any hypothesis discussed here, it does imply that trade has promoted clean industries in rich countries and dirty industries in poor ones. This point is reinforced by research on emissions embodied in trade (Davis & Caldeira, 2010; Peters et al., 2011)—which indicates that rich and poor countries are net importers and exporters of carbon emissions, respectively—and by “pollution haven” research showing that environmental regulations cause dirty-industry output to migrate from high-regulation to low-regulation countries (Brunnermeier & Levinson, 2004; Levinson & Taylor, 2008; Millimet & Roy, 2011). Together, this research indicates that (a) environmental regulations place domestic polluting industries at a competitive disadvantage, and (b) that international trade leads to predictable shifts in economic structure. It therefore supports the foundational assumptions of the political hypotheses discussed here.

Kono (2017) tested the tariff-compensation and carbon-compensation hypotheses by examining the relationship between tariffs and carbon emissions in a large sample of countries from 1988 to 2013. This study gets closer to the outcomes of interest here in that it employs a measure of trade policy rather than trade flows. Using instrumental variables to address the endogeneity of trade and CO2 emissions, Kono finds that higher CO2 emissions led to lower tariffs on carbon-intensive manufactured goods but had no effect on tariffs in the less carbon-intensive primary sector. These results are consistent with the tariff-compensation hypothesis: countries that increased their carbon emissions reduced tariffs on carbon-intensive goods more quickly than ones that restrained emissions. In contrast, he finds no evidence that tariffs of any kind affect CO2 emissions and hence no evidence of a carbon-compensation effect. He explains these asymmetric results—emissions affect tariffs, but not vice versa—by invoking the different collective-action problems inherent in each domain.

Although the Kono (2017) study’s focus on tariffs is useful, its reliance on CO2 emissions rather than greenhouse-gas policies still limits the inferences that can be drawn from it. It seems likely that carbon restrictions indeed impede trade liberalization: if carbon restrictions affect emissions (as they should), and if the latter affect tariffs (as they seem to), then carbon restrictions must affect tariffs. Given this, these results provide reasonable if indirect evidence of a tariff-compensation effect. However, it is less clear that the null results reject the carbon-compensation hypothesis. It is possible that tariffs affect carbon restrictions but that this effect does not show up in emissions because it is outweighed by other nonpolicy factors. Moreover, as noted earlier, trade policies could affect emissions directly in ways that offset their effects on carbon restrictions. Conclusively rejecting the carbon-compensation hypothesis will therefore require tests based on actual climate policies.6

The broader literature on trade–environment policy linkages also provides suggestive evidence. Ward and Cao (2012) show that integration into the global economy reduces “green taxes”—that is, taxes levied on environmentally harmful activities. As the authors conclude, “countries more tightly integrated into the globalized economy . . . have lower tax rates. Rather than competition with specific other countries . . . it seems likely that the perception of policy makers about a country’s general competitive position matters” (p. 1095). This is consistent with the carbon-compensation hypothesis, in that internationally exposed economies are less likely to regulate pollution via taxes. However, the authors’ measures provide only indirect evidence of a trade–carbon policy linkage. Their measure of economic exposure incorporates not only trade but also portfolio and foreign direct investment, so it is not clear which type of openness matters. And while the green-tax measure captures some climate policies (e.g., carbon taxes, petrol taxes), it also incorporates taxes on pollutants that do not contribute to climate change, and it does not incorporate some climate policies (e.g., regulatory caps on emissions). The results thus suggest the plausibility of the carbon-compensation hypothesis but do not provide a conclusive test.

Research on international climate commitments also speaks indirectly to our questions. Bättig and Bernauer (2009) find that trade openness (imports + exports/GDP) is negatively related to a multidimensional index of climate policy based largely on international commitments. This is consistent with the carbon-compensation hypothesis: more open economies are less likely to regulate carbon. However, because trade openness is weakly correlated with most measures of trade policy—economies can be open for many nonpolicy reasons—this result is at best an indirect test. Moreover, Neumayer (2002) finds little evidence that trade and trade policy predict ratification of the Kyoto Protocol. The evidence based on international commitments is thus inconclusive.

In sum, there is suggestive evidence for at least two of the domestic-level hypotheses. Kono (2017) shows that carbon emissions affect tariffs in ways that are consistent with the tariff-compensation hypothesis. Ward and Cao (2012) and Bättig and Bernauer (2009) show that international economic exposure affects environmental policies in ways that are consistent with the carbon-compensation hypothesis. However, the imperfect fit between these empirical studies and both hypotheses requires us to interpret these results with caution.

To my knowledge, there is no evidence for the tariff-reduction or carbon-regulation hypotheses. We should not assume they are incorrect, however, because the above-mentioned studies arguably do not test them. This is because of not only the measures but also the research designs themselves. As discussed, we might expect to detect compensation effects in longitudinal analyses but tariff-reduction and carbon-regulation effects in cross-sectional analyses. Kono (2017) and Ward and Cao (2012) both employ country fixed effects and thus examine only within-country, over-time relationships. This may bias their results toward compensation effects.

Similarly, the dynamic hypotheses remain untested. Examining the long-term relationship between trade and climate policies is difficult because climate change became a salient political issue only in the late 1990s. The Kyoto Protocol was adopted in 1997 and went into force in 2005, and few countries outside the EU have adopted broad-based climate change measures. This may mean that the long-term impact of climate policies on trade policies—captured in the dynamic-liberalization and dynamic-protectionism hypotheses—cannot be studied until existing climate change measures have had time to shape industrial structures. We might, however, test the dynamic green-trade and dynamic dirty-trade hypotheses by first measuring the historical impact of trade on industrial structures, and then examining the impact of the latter on climate change policies.

Turning to international linkages, there is abundant anecdotal evidence for the linkage-support hypothesis. As noted earlier, U.S. politicians from both parties have publicly argued that U.S. greenhouse-gas regulations should be accompanied by trade sanctions against laggard countries. In fact, such sanctions appear to be a political prerequisite for climate legislation: thus far, “all legislative proposals for a U.S. greenhouse gas (‘GHG’) emissions cap-and-trade system . . . have recognized the need to safeguard the [international] competitiveness of U.S. firms” (Kyo, Janzen, & Smith, 2010, p. 12). Moreover, calls for such linkage are not limited to the United States: for example, during the Copenhagen climate conference in 2009, French president Nicolas Sarkozy called for carbon-based border taxes to protect EU industries and jobs (Kanter, 2009), and in 2017 the EU implemented its first-ever border tax on carbon (The Economist, 2017). In all these cases, efforts to tie domestic trade and foreign climate policies seem designed in part to build domestic support for climate legislation.

The linkage-support hypothesis also enjoys some micro-level support. In a large-scale experimental survey in four developed countries, Bechtel and Scheve (2013) find that citizens prefer climate agreements that include small economic sanctions for noncompliance over agreements with no sanctions at all. Similarly, in a sample of 26 countries, Tingley and Tomz (2013) find broad public support for “extrinsic reciprocity”—that is, punishing climate laggards with economic sanctions. Interestingly, they find little support for “intrinsic reciprocity” in which home climate policies are conditional on foreign climate policies. This suggests that linking home trade and foreign climate policies may be a more effective way of building domestic political support than trying to get other countries to play along in the climate realm. These micro-level results are important, because they show that the linkage-support hypothesis rests on solid microfoundations. However, because we do not know if these preferences translate into policy outcomes, we need to test this hypothesis at the macro level as well.

Direct tests of the linkage-enforcement hypothesis are difficult, because climate legislation containing formal trade sanctions is such a recent phenomenon. However, research on nonclimate outcomes provides some suggestive evidence. For example, the Montreal Protocol’s sanctions against trade in ozone-depleting substances seem to have helped that agreement succeed (Barrett, 1997; Cirone & Urpelainen, 2013). Moreover, evidence on trade sanctions in other policy areas shows that such sanctions can work. For example, Hafner-Burton (2005) found that preferential trade agreements with “hard” human rights standards—that is, standards linked to market access—improve human rights performance in member states. This is noteworthy because human rights practices—like climate change policies—cannot be effectively enforced via reciprocity in the same policy domain. Hafner-Burton’s results support the linkage enforcement hypothesis for human rights. Whether it holds up for climate change, however, remains to be seen.

Much the same can be said of the linkage-abuse hypothesis. Research on technical barriers to trade, such as health, safety, and environmental standards, has shown that such barriers are sometimes abused. For example, Henson and Loader (2001) find that sanitary and phytosanitary standards constitute a growing obstacle to trade; Kono (2006) finds that such standards tend to be driven by protectionist rather than public-welfare concerns; and Otsuki, Wilson, and Sewadeh (2001) find that EU aflatoxin standards have huge effects on African food exports without appreciably improving the health of EU citizens. Although we do not know if carbon tariffs would be similarly abused, there is no obvious reason to think they would not be.

The linkage-deterrence hypothesis is currently impossible to test because no international climate agreement has included trade sanctions. Some climate legislation, such as that of the EU, incorporates trade sanctions. However, it is not clear why this would deter anyone from signing an agreement like Kyoto or Paris, since the sanctions are based on countries’ domestic emissions rather than (non)participation in an international agreement. For the time being, evidence of linkage deterrence must probably come from other policy areas: for example, by showing that hard human rights provisions deter states from signing trade agreements.

Finally, scholars have found that both race-to-the-bottom and California-effect dynamics influence environmental standards, though they have not examined climate policy specifically. Prakash and Potoski (2006) find evidence for a California effect in the adoption of ISO 14001 environmental standards: firms are more likely to adopt this standard when many firms in their major export markets have done so. Though interesting, this result says little about government policies, since ISO 14001 is a voluntary standard adopted by firms rather than national governments. On the other hand, it is worth noting that ISO 14001 is a process standard. This study thus supports Urpelainen’s (2011) point that California effects can occur for process as well as product standards, at least if no legal obstacles stand in the way.

Cao and Prakash (2012) examine the impact of trade competition on de jure environmental policies and de facto environmental outcomes. They find that structurally equivalent countries—that is, “competitor” countries with similar export profiles—influence each other’s de facto outcomes but not their de jure policies. Specifically, when a competitor country increases its air or water pollution, it induces other competitors to follow suit. This seems like a race to the bottom, since the authors’ measure of trade exposure—structural equivalence rather than export dependence—should capture incentives to weaken rather than strengthen environmental standards. The authors also find that races to the bottom are weakened by domestic veto players. However, this only occurs for highly visible pollutants. This is relevant here because greenhouse gases have negligible local effects and are thus essentially invisible. Given this, the authors’ results provide little hope that veto players would mitigate a race to the bottom in climate policy.

Although Cao and Prakash (2012) provide good evidence of races to the bottom, their empirical measures limit what we can infer about climate policy. Their measure of de jure policies incorporates a wide range of international environmental policy commitments, while their de facto outcomes are sulfur-dioxide emissions and water pollution, neither of which is a greenhouse gas. We thus cannot be sure whether these results apply to greenhouse gases.


There are good theoretical reasons to believe that trade and climate policies affect each other. In the short run, policies in each domain affect industry incentives to lobby in the other domain. In the long run, each set of policies affects industrial structure in ways that could alter political support for the other. Politics in each area might also be affected by explicit links between one country’s trade policy and another country’s climate policy. Although theoretically, it is not clear whether trade liberalization and climate change mitigation are complementary or competing goals, it seems clear that the pursuit of one could influence pursuit of the other. We therefore need to study these links empirically.

The empirical literature on trade–climate policy links is easily summarized: there are many indirect tests and no conclusive ones. Most of what we “know” about trade and climate politics is based on proxy measures of trade or climate policy (Bättig & Bernauer, 2009; Kono, 2017) or research on related but nonetheless different policy outcomes (Cao & Prakash, 2012; Hafner-Burton, 2005; Ward & Cao, 2012). Hence, while there is suggestive evidence for many of the hypotheses considered here, none of it is conclusive. There is enormous scope for further research on trade and climate politics, and also a clear agenda: collect more data on national climate policies.


Aklin, M. (2016). Re-exploring the trade and environment nexus through the diffusion of pollution. Environmental and Resource Economics, 64, 663–682.Find this resource:

Autor, D. H., Dorn, D., & Hanson, G. H. (2016). The China shock: Learning from labor-market adjustment to large changes in trade. Annual Review of Economics, 8, 205–240.Find this resource:

Baker, J. A. I., Feldstein, M., Halstead, T., Mankiw, N. G., Paulson, H. M. J., Shultz, G. P., . . . Walton, R. (2017). The Conservative Case for Carbon Dividends.

Barrett, S. (1997). The strategy of trade sanctions in international environmental agreements. Resource and Energy Economics, 19, 345–361.Find this resource:

Bechtel, M. M., & Scheve, K. F. (2013). Mass support for global climate agreements depends on institutional design. Proceedings of the National Academy of Sciences USA, 110, 13763–13768.Find this resource:

Bernauer, T. (2013). Climate change politics. Annual Review of Political Science, 16, 421–448.Find this resource:

Broder, J. M. (2009, June 28). Obama opposes trade sanctions in climate bill. New York Times.Find this resource:

Brunnermeier, S. B., & Levinson, A. (2004). Examining the evidence on environmental regulations and industry location. Journal of Environment and Development, 13, 6–41.Find this resource:

Bättig, M. B., & Bernauer, T. (2009). National institutions and global public goods: Are democracies more cooperative in climate change policy? International Organization, 63.Find this resource:

Cao, X., & Prakash, A. (2012). Trade competition and environmental regulations: Domestic political constraints and issue visibility. Journal of Politics, 74, 66–82.Find this resource:

Cirone, A. E., & Urpelainen, J. (2013). Trade sanctions in international environmental policy: Deterring or encouraging free riding? Conflict Management and Peace Science, 30, 309–334.Find this resource:

Cole, M. A. (2004). Trade, the pollution haven hypothesis and the environmental Kuznets curve: Examining the linkages. Ecological Economics, 48, 71–81.Find this resource:

Cole, M. A., & Elliott, R. J. R. (2003). Determining the trade–environment composition effect: The role of capital, labor and environmental regulations. Journal of Environmental Economics and Management, 46, 363–383.Find this resource:

Copeland, B. R., & Taylor, M. S. (2004). Trade, growth, and the environment. Journal of Economic Literature, 42, 7–71.Find this resource:

Davis, C. L. (2004). International institutions and issue linkage: Building support for agricultural trade liberalization. American Political Science Review, 98, 153–169.Find this resource:

Davis, S. J., & Caldeira, K. (2010). Consumption-based accounting of CO2 emissions. Proceedings of the National Academy of Sciences USA, 107, 5687–5692.Find this resource:

The Economist. (2017, February 17). Are carbon tariffs a good idea? The Economist Explains [blog].Find this resource:

Ederington, J., & Minier, J. (2003). Is environmental policy a secondary trade barrier? An empirical analysis. Canadian Journal of Economics/Revue Canadienne d'Economique, 36(1), 137–154.Find this resource:

Fearon, J. D. (1998). Bargaining, enforcement, and international cooperation. International Organization, 52, 269–305.Find this resource:

Frankel, J. A., & Rose, A. K. (2005). Is trade good or bad for the environment? Sorting out the causality. Review of Economics and Statistics, 87, 85–91.Find this resource:

Grossman, G. M., & Krueger, A. B. (1993). Environmental impacts of a North American free trade agreement. In P. M. Garber (Ed.), The U.S.-Mexico free trade agreement (pp. 13–56). Cambridge, MA: MIT Press.Find this resource:

Hafner-Burton, E. M. (2005). Trading human rights: How preferential trade agreements influence government repression. International Organization, 59.Find this resource:

Harrison, K. (2015). International carbon trade and domestic climate politics. Global Environmental Politics, 15(3), 27–48.Find this resource:

Harrison, K., & Sundstrom, L. M. (2007). The comparative politics of climate change. Global Environmental Politics, 7(4), 1–18.Find this resource:

Henson, S., & Loader, R. (2001). Barriers to agricultural exports from developing countries: The role of sanitary and phytosanitary requirements. World Development, 29(1), 85–102.Find this resource:

Hiscox, M. J. (2002). Commerce, coalitions, and factor mobility: Evidence from congressional votes on trade legislation. American Political Science Review, 96, 593–608.Find this resource:

Jaspers, N., & Falkner, R. (2013). International trade, the environment, and climate change. The Handbook of Global Climate and Environment Policy, 412–427.Find this resource:

Kahneman, D., & Tversky, A. (1984). Choices, values, and frames. American Psychologist, 39, 341–350.Find this resource:

Kanter, J. (2009, December 17). At climate talks, trade pressures mount. New York Times.Find this resource:

Kaufmann, C., & Weber, R. H. (2011). Carbon-related border tax adjustment: Mitigating climate change or restricting international trade?. World Trade Review, 10, 497–525.Find this resource:

Keohane, R. O. (1984). After hegemony: Cooperation and discord in the world political economy. Princeton, NJ: Princeton University Press.Find this resource:

Keohane, R. O. (1986). Reciprocity in international relations. International Organization, 40, 1–27.Find this resource:

Kerry, J., & Graham, L. (2009, October 10). Yes, we can (pass climate change legislation). New York Times.Find this resource:

Kleemann, L., & Abdulai, A. (2013). The impact of trade and economic growth on the environment: Revisiting the cross-country evidence. Journal of International Development, 25, 180–205.Find this resource:

Kono, D. Y. (2006). Optimal obfuscation: Democracy and trade policy transparency. American Political Science Review, 100(3), 369–384.Find this resource:

Kono, D. Y. (2017). Tariffs and carbon emissions. International Interactions, 43(6), 895–919.Find this resource:

Kyo, S., Janzen, B. G., & Smith, H. M. (2010). Border adjustment measures in proposed US climate change legislation: “A new chapter in America’s leadership on change?” Sustainable Development Law and Policy, 9(3), Article 8.Find this resource:

Levinson, A., & Taylor, M. S. (2008). Unmasking the pollution haven effect. International Economic Review, 49(1), 223–254.Find this resource:

Maggi, G. (1999). The role of multilateral institutions in international trade cooperation. American Economic Review, 89, 190–214.Find this resource:

Maliniak, D., Peterson, S., Powers, R., & Tierney, M. J. (2014). TRIP 2014 Faculty Survey. Williamsburg, VA: Institute for the Theory and Practice of International Relations.Find this resource:

Managi, S., Hibiki, A., & Tsurumi, T. (2009). Does trade openness improve environmental quality?. Journal of Environmental Economics and Management, 58, 346–363.Find this resource:

McKibben, H. E. (2015). State strategies in international bargaining: Play by the rules or change them? Cambridge, U.K.: Cambridge University Press.Find this resource:

Millimet, D. L., & Roy, J. (2011). Three new empirical tests of the pollution haven hypothesis when environmental regulation is endogenous (IZA Discussion Paper No. 5911). Institute for the Study of Labor, Bonn, Germany.Find this resource:

Neumayer, E. (2002). Does trade openness promote multilateral environmental cooperation? World Economy, 25, 815–832.Find this resource:

Otsuki, T., Wilson, J. S., & Sewadeh, M. (2001). Saving two in a billion: Quantifying the trade effect of European food safety standards on African exports. Food Policy, 26, 495–514.Find this resource:

Peters, G. P., Minx, J. C., Weber, C. L., & Edenhofer, O. (2011). Growth in emission transfers via international trade from 1990 to 2008. Proceedings of the National Academy of Sciences USA, 108, 8903–8908.Find this resource:

Prakash, A., & Potoski, M. (2006). Racing to the bottom? Trade, environmental governance, and ISO 14001. American Journal of Political Science, 50, 350–364.Find this resource:

Purdon, M. (2015). Advancing comparative climate change politics: Theory and method. Global Environmental Politics, 15(3), 1–26.Find this resource:

Rickard, S. (2017). Electoral systems and policy outcomes. Oxford Research Encyclopedia of Politics [Online]. New York, NY: Oxford University Press.Find this resource:

Rosendorff, B. P., & Milner, H. V. (2001). The optimal design of international trade institutions: Uncertainty and escape. International Organization, 55, 829–857.Find this resource:

Scott, R. E. (2009). Climate change policy: Border adjustment key to U.S. trade and manufacturing jobs (EPI Briefing Paper 241). Economic Policy Institute, Washington, DC.Find this resource:

Steves, F., & Teytelboym, A. (2013). Political economy of climate change policy (Working Paper 13-06). Smith School for Enterprise and the Environment, Oxford, U.K.Find this resource:

Tamiotti, L., Teh, R., Kulacoglu, V., Olhoff, A., Simmons, B., & Abaza, H. (2009). Trade and climate change: A report by the United Nations Environment Programme and the World Trade Organization. Geneva, Switzerland: World Trade Organization.Find this resource:

Tingley, D., & Tomz, M. (2013). Conditional cooperation and climate change. Comparative Political Studies, 47, 344–368.Find this resource:

Urpelainen, J. (2011). A California effect for international environmental externalities?. International Interactions, 37(2), 170–189.Find this resource:

Vogel, D. (1995). Trading up: Consumer and environmental regulation in a global economy. Cambridge, MA: Harvard University Press.Find this resource:

Ward, H., & Cao, X. (2012). Domestic and international influences on green taxation. Comparative Political Studies, 45, 1075–1103.Find this resource:

Weinberg, J. (2016). European Union member states in cross-national analyses: The dangers of neglecting supranational policymaking. International Studies Quarterly, 60, 98–106.Find this resource:


(1.) Scott (2009) identifies iron and steel products, nonferrous metal products, textile products, paper products, petroleum and coal products, chemical manufactures, and nonmetallic mineral products as the most carbon-intensive U.S. industries. This list largely dovetails with the dirty industries identified by Copeland and Taylor (2004).

(2.) For convenience, the focus is on a carbon tax, but the logic also applies to other types of carbon restrictions.

(3.) Laser Talk, Border carbon adjustment, updated June 1, 2018, Citizens Climate Lobby.

(4.) See, e.g., Rickard (2017) for a discussion of electoral systems and policy outcomes.

(5.) In this sense, the linkage-support hypothesis is essentially a special case of the carbon-compensation hypothesis in which domestic climate policies are explicitly linked to trade sanctions.

(6.) Kono (2017) performs a robustness check using a measure of carbon restrictions: a dummy for a carbon tax or cap-and-trade program. The results are consistent with the emission-based ones. However, because data limitations precluded the use of instrumental-variable techniques, these results are more suggestive than conclusive.