As mentioned in the introduction, politics is a key factor in both globalization and de-globalization. The field of International Relations in political science has developed two major theoretical approaches3 that speak to the question of (de-)globalization: liberalism and realism. The former is consistent with most of the discourse on (de-)globalization in the press and among IB scholars, while the latter represents the dominant school of thought in International Relations.
The term “liberalism” has come to mean different things to different people and thus represents a source of potential confusion. In this paper, “liberalism” and any of its derivatives will exclusively denote the International Relations theory by this name as introduced below. It will never imply outcomes such as an open economy, nor will it refer to a political philosophy of individual freedom and rights or an ideology concerning social welfare policies.
Both liberalism and realism have evolved multiple variants. For the purpose of this discussion, I focus on the theoretical formulations growing out of the two foundational pieces of modern International Relations theory, Moravcsik’s (1997) conceptualization of liberalism and Waltz’s (1979) structural realism. For an overview of other variants, I refer the reader to International Relations primers (e.g., Matthews & Callaway, 2017) or any major International Relations textbook (e.g., Pevehouse & Goldstein, 2016).
Table 1 presents an overview of the main characteristics of both approaches, their conceptualizations of (de-)globalization, and their predictions about the future landscape of international economic activity. In the following paragraphs, I first lay out the main concepts and mechanisms envisioned by these theories before turning to the questions of how they relate to (de-)globalization and the kind of future each approach predicts.
Liberalism permits a wide range of actors, interests, and forms of power to determine political outcomes. While countries4 represent the major actors in international politics, possible actors codetermining outcomes further range from the individual to the international, including individuals, firms, non-government organizations (NGOs), and international organizations such as the European Union (EU), the World Trade Organization (WTO) or the United Nations. Liberalism assumes these actors to be rational on average, that is, deviations from rationality are possible.
By engaging in politics, these actors pursue their self-interests, a term that is very broadly conceived. It includes questions of economic welfare, but also permits other preferences including those based on personal and cultural values. Depending on the interests at hand, actors may seek to attain their objectives through positive-sum cooperation (e.g., economic wealth through trade) or zero-sum competition (e.g., national security through military strength).
In pursuing their respective interests, actors have recourse to a wide range of different forms of power. These include hard power, which usually rests on coercive capabilities such as military strength or the ability to inflict economic damage (Nye, 1990); sharp power (Walker & Ludwig, 2017), which draws on distraction and manipulation to weaken opponents and shore up one’s own position; and soft power (Nye, 1990), which relies on co-optation of others by presenting an attractive model or outcome.
In the resulting dynamics, what happens domestically – i.e., inside countries – is at least as important as what happens between countries. Countries may be the major actors in international politics, but, in liberalism, their interests are a reflection of domestic interests, with more powerful actors having greater influence on foreign policy. Depending on the issue area and the structure of the respective country, foreign policy may reflect the interests of different constituents: for instance, those of a few powerful actors in authoritarian dictatorships, those of special interests in countries with strong business lobbies, those of a plurality of the people in countries with first-past-the-poll elections, or of a societal consensus in Northern European corporatist societies. As actors and their interests shift, so does the focus and thrust of foreign policy.
Liberalism thus presents a fairly accurate depiction of reality, with all its complexity. This is also its major weakness as a theory. It is so permissive in terms of the possible actors, their respective interests and the various forms of power that may matter that it can be very difficult to arrive at a conclusive analysis and prediction for specific issues.
The realist school narrows down this wide range of possible actors, interests, and forms of power considerably. Actors are sovereign countries, especially great powers. Countries are unitary, rational actors, with domestic politics being irrelevant for their behavior in the international system. This is because under the additional assumption of anarchy in the international system – the notion that there are ultimately no rules constraining state behavior as there is no overarching authority that could enforce compliance – securing survival becomes the primary concern of states. The main objective of foreign policy is thus forced on countries by the structure of the international system, and countries permitting themselves to get distracted from the objective of survival by domestic-level political concerns risk their survival.
In the anarchic world of realism, securing survival is ultimately a function of hard power. In particular, countries need to acquire sufficient military power to defend themselves against other countries. Since building and maintaining armies requires resources, this implies an attendant need for economic strength. With power representing primarily a tool for securing survival against other countries, the emphasis is on relative rather than absolute power. Economic growth of 10% per year may enable a country to increase its power considerably in absolute terms, but from a realist perspective it is still falling behind if at the same time another state grows by 20%. Relations between countries thus become a zero-sum game.
Realism presents a much tighter theoretical approach than liberalism. Ironically given its name, realism offers a very reductionist rather than realistic picture of international politics. As a consequence, a salient point of criticism is that it may miss important parts of the picture. For instance, in a strict application of realism, the EU is not a relevant actor, as it does not represent a sovereign nation state. Not everyone finds the fallback position of realists – that the EU is ultimately the expression of the power of one member state (or possibly several of them) – convincing, considering the extent of horse-trading in which even the most powerful EU members must engage to attain their objectives.
Relationship with (De-)Globalization
For liberalists, globalization involves two major ingredients: the agreement of states to cooperate in building interdependence, and a supporting infrastructure that enables such cooperation to occur.
The first component, agreement, requires that there be sufficient political support for building economic interdependence within each country. In other words, proponents of openness need to have more political power than opponents.
While such agreement is an expression of the will to cooperate, putting cooperation into practice often faces challenges because of a risk that other countries will renege on their agreements in order to gain even greater benefits for themselves. Liberalists usually conceptualize this challenge in game-theoretic terms (e.g., Aggarwal & Dupont, 1999; Maggi, 1999), especially with reference to the prisoners’ dilemma: while overall benefits of cooperation may be maximized if everyone cooperates, individual players may maximize their own benefits by defecting while the others continue to cooperate. Since this incentive structure is known, the risk is that no actor agrees to cooperate.
Liberalists claim that countries can overcome the constraints of anarchy and build cooperation in at least three ways. First, because actors, especially states, interact with one another repeatedly, the gains from a one-time defection pale in comparison with the possible gains from long-term cooperation. Since countries realize who can be trusted and who tends to cheat, defecting does not pay. Rational actors will thus learn to cooperate (Axelrod, 1984). Second, cooperation can be helped along through the use of strategies that reward cooperation and punish defection. For instance, adopting a simple “tit-for-tat” approach – doing whatever the partner did last – tends to lead to stable cooperation over time (Axelrod, 1984).
Third, and most important for the understanding of de-globalization, is the argument that just as institutions facilitate cooperation within countries (North, 1990; Williamson, 1985), international institutions (also called “international regimes”) may enable international cooperation (Keohane, 1984, 1989). Institutions in this context are defined as “persistent and connected sets of rules (formal and informal) that prescribe behavioral roles, constrain activity, and share expectations” (Keohane, 1989: 3). This conceptualization is broadly consistent with that by North (1990), commonly used in the international business literature. (Somewhat confusingly, the bodies entrusted with administering international institutions, such as the WTO, are also referred to as “institutions,” while the business literature would see them as “organizations” (defined as groups of people bound together for a common purpose; see North, 1990)).
Liberalists argue that institutions aid cooperation in a number of ways (Davis, 2012; Maggi, 1999; Matthews & Callaway, 2017). They provide a forum for discussion, for identifying mutual interests, and for finding joint solutions. Once institutions are agreed on, the organizations attached to them can help monitor compliance. Some of these organizations also provide for adjudication of conflicts around international institutions and set penalties for violations. Liberalists would not deny that compliance with international institutions is imperfect but would argue that imperfect institutions are better than none, with successful cooperation having the potential to open up further areas of cooperation in the future.
In a liberalist world, there are accordingly two pathways to de-globalization. The first is for the institutional infrastructure supporting globalization to lose its ability to support openness. For instance, diverging interests between countries may prevent the creation or maintenance of the institutions necessary for openness to sustain or advance. We have seen such a divergence of interests in the context of the Doha Round of WTO negotiations, in which advanced industrialized countries essentially pushed for more free trade (except in agriculture), while the emerging markets pushed for what they saw as fair trade. Unlike in previous rounds, in which the advanced industrialized countries were powerful enough to push through their interests, increased power of emerging markets resulted in a stalemate in the Doha Round.
Such stalemate and attendant lack of institutional changes does not only work against the opening up of new areas of economic interdependence. It also contributes to institutional drift (Mahoney & Thelen, 2009), which implies that institutions move out of synch with the issues they should address. For instance, countries have shown great ingenuity in devising ways to renege on their promise of openness while officially remaining in compliance. An example is non-tariff barriers to trade, which many states have thrown up and which the WTO remains ill-equipped to address. Such circumvention of international institutions both reduces interdependence directly – by making trade with and investment in other countries harder – and indirectly – by inducing other countries to reciprocate by limiting openness.
This reneging is linked to the second pathway: a change in national political interests leads countries to opt out of economic interdependence. Consistent with the notion that shifting interests may be connected with the current de-globalization, public support for globalization has dropped in many economies since the early 2000s (OECD, 2017). In a November 2016 survey across 23 advanced and emerging markets, an average of 53% of respondents had little or no confidence in international institutions, 61% had little or no confidence in large firms, and a large minority was concerned about immigration (with no majority in favor) (IPSOS, 2017).
As to the causes of this shift, given the flexibility of liberalism with respect to interests, a wide range of causes is conceivable. For example, values underlying policy preferences may undergo ideological shifts away from open to protected markets. Since ideologies legitimate positions of power (Mannheim, 1936), such shifts usually require a failure, perceived or real, of the prevailing ideology. Arguably, the financial crisis of 2008 and the European refugee crisis of 2015 represented such watersheds against pro-openness ideology in the Western world. The former empowered left-wing critics of globalization, who object to openness in its present shape, which they see as unjust (Fisher & Ponniah, 2003; Santos, 2013; Verbeke et al., 2018). In particular, leftist activists object to an investment- and corporation-led form of globalization in which investors and firms are seen as benefiting at the expense of common people, some countries, and the environment. The latter reinforced support of right-wing nationalist groups linking globalization to unwelcome immigration (Rodrik, 2018), which seems to have played a role in the Brexit vote of 2016. Jointly, the rise of such opposition and attendant challenges to pro-market ideology are likely to weaken support of openness at the domestic level, with concomitant implications for a country’s ability to pursue international economic interdependence.
Second, opening up trade or finance internationally has distributional consequences within countries, with some sectors gaining and others losing. For instance, international trade or financial liberalization will generally hurt previously protected sectors and their workers (Buckley & Ghauri, 2004; Frieden, 1991; Stolper & Samuelson, 1941). Low-skilled workers suffer in terms of wage depression and job losses – and resultant increases in inequality – when exposed to competition from emerging markets, as has been shown for US workers facing competition from China (e.g., Acemoglu, Autor, Dorn, Hanson, & Price, 2016; Autor, Dorn, & Hanson, 2013). Unless compensated, the losers from globalization may mobilize and seek to reverse economic openness (Autor, Dorn, Hanson, & Majlesi, 2016; Frieden, 1991). If these actors then have sufficient power – for instance, by being numerous enough to elect a leader aligned with their interests – a country may shift its foreign economic policy in favor of de-globalization. Arguably, this accounts at least partially for the election of Donald Trump as U.S. President and the global rise of populism more generally (Rodrik, 2018).
A third example is that the payoffs from cooperation may change over time, and interests supporting globalization may weaken as a consequence. For instance, while importing goods or offshoring production may be economically efficient in the short term, the attendant shrinkage of production in the home market may be undesirable in the longer term for reasons such as a reduction in expertise and production capacity in industries important for national defense (Berger, 2013). We have also observed this mechanism in recent years, for example, in the context of growing numbers of Chinese foreign direct investment projects blocked in Western nations on grounds of national security.
Overall, the picture drawn by liberalism is consistent with de-globalization. International institutions appear to be weakening, and domestic political interests seem to have shifted to favor reduced interdependence.
For realists, the main pathway to globalization involves coercion. This is the thrust of a sub-theory of realism called “hegemonic stability theory” (Keohane, 1980; Kindleberger, 1973; Krasner, 1976). Hegemonic stability theory argues that periods of globalization occur when an overwhelmingly powerful country, a “hegemon,” creates and maintains, for its own benefit, sets of international institutions (“regimes”) that govern aspects such as trade and investments. The hegemon will keep this system in place as long as it remains strong enough to do so and the benefits from keeping the system exceed the costs. Other states may or may not benefit from the system.
Once the hegemon declines – i.e., it loses power relative to other countries to the point that it is no longer overwhelmingly powerful – the system becomes unstable. While hegemonic decline has been a persistent pattern in history (Gilpin, 1983; Organski, 1958), it is not necessarily clear precisely when a state ceases to be a hegemon. As a result, an economic order may outlive the hegemony of the country that created it – it is “sticky” (Krasner, 1976). It will fail once a shock to the system, such as an economic crisis, reveals that the hegemon has lost its power to maintain that system.
Importantly, while the international system under hegemonic stability may look much like an institutional structure along the lines of that envisioned by liberalism, these institutions are in reality epiphenomenal: they do not exist in their own right but reflect the interests and power of the hegemon. In essence, they are a matter of convenience for the hegemon: it is easier to hand out a rule book than to tell each country on a case-by-case basis what to do.
The theoretical parsimony of realism makes it much more straightforward to illustrate how the current period of de-globalization coincides with hegemonic decline. Hegemonic stability theory links the openness of the international economic system to the preponderance of the most powerful country, which in recent history has been the United States of America. Conversely, de-globalization accompanies a decline in power of the strongest state, not necessarily in absolute terms, but relative to other states. If such decline is present, one would expect de-globalization.
To assess this possibility, I evaluate the relative power of the United States as compared with the rest of the world and its closest rivals. I focus on two main dimensions that are central to the realist concept of power: military strength and economic power. The latter is easy to operationalize as GDP. Military power, on the other hand, is difficult to measure without a detailed, and probably classified, understanding of the capabilities of the world’s militaries, including the number and capabilities of soldiers and equipment. As a proxy, I rely instead on the extent of military spending. This measure is imperfect in that it does not account for classified budgets; does not differentiate between spending for new equipment as opposed to ongoing operations; and does not account for purchasing power differences. At the same time, the resultant biases all work against finding a decline in power in the current context: budget transparency is greater in the US than in its next closest military rivals, China and Russia; the US has been spending much more on ongoing wars in the past 15 years than any other nation, with concomitant expenses for veterans; and much of the world, but certainly China and Russia, should see greater purchasing power per dollar spent on the military than the United States. Any decline in the US budget relative to other states is thus likely to be a conservative depiction of actual dynamics.
Both metrics paint a clear picture of relative decline of US power. Figure 8 shows US GDP at PPP and constant 2011 international dollars relative to all other countries in the world combined, as well as relative to the respective next largest economy in the world (Japan until 1998, China from 1999 onward). The time series begins in 1990 because the World Development Indicators database used does not report GDP at PPP for earlier years. Relative to the rest of the world, US economic strength in this period peaked in 1999 at 20.6%. By 2017, that figure had declined to 15.2% of world GDP, a drop of 26.2%. The picture is even clearer for US GDP relative to the next largest. Again, US strength peaked in 1999 with its GDP at 289.4% that of China’s. By 2013, China had reached parity, and in 2017, US GDP was down to 83.2% of China’s, a decline by 71.3%. While one can argue whether the United States was an economic hegemon in 1999, it is clear that, by 2017, it was far from hegemony. At the same time, it is also clear that China is not (yet?) an economic hegemon, either.
Figure 9 shows US military spending relative to all other countries in the world combined, as well as relative to the respective next largest spender (USSR 1988–1990, France 1991–2000, China 2001 onward). It is measured in constant 2016 international dollars, with PPP adjusted data unavailable. The time series begins in 1988 because this is the earliest year for which the SIPRI database (Stockholm International Peace Research Institute, 2018) contains a budget for the Soviet Union, which during the Cold War was the closest military contender to the United States. Military spending followed a similar pattern to GDP, though peaking earlier. Relative to the rest of the world, US military spending reached its high point in 1992 at 46.4%. By 2017, it had declined to 36.1%, a drop by 22.2%. Again, the picture is even clearer relative to the next closest contender: US spending peaked in 1992 at 921.7% that of the next closest, France. By 2017, it had decreased to 261.7% relative to the Chinese military budget, a drop of 71.6%. As already mentioned, especially the more recent figures are likely to underestimate the attendant shift in military power.
Overall, these data show a picture that is consistent with de-globalization. To the extent that the United States used to be a hegemon that could erect and maintain an open international economic system, it is unlikely that it remains powerful enough today. Its economy is already smaller than China’s, and the Chinese military is rapidly catching up.
Both approaches would thus expect de-globalization under the present conditions. However, their predictions of the future landscape of the international political economy vary considerably.
As laid out above, liberalism suggests that de-globalization results from a combination of failing international institutions and interests shifting away from economic openness. To the extent that this change of interests is not uniform across countries – which is probably a realistic assumption – the most likely outcome is a globalization patchwork in which different pairs or groups of countries provide for varying levels of interdependence between them.
Geographically confined agreements are thus likely to gain in importance. The past decades have already seen the emergence of such a globalization patchwork in the shape of numerous bilateral and regional, rather than global, trade agreements (Aggarwal and Urata, 2006; Shadlen, 2008). While these agreements were effectively attempts at increasing interdependence above the global standard, de-globalization implies an additional layer of complexity in that some countries begin to opt out of the system to reduce their levels of interdependence. To the extent such departures threaten the viability of the remaining global infrastructure, countries willing to maintain high levels of interdependence may find a need to devise suitable bilateral and regional agreements. Similar to past agreements, these alternative arrangements are likely to be tailored to the interests of the countries negotiating them. Given diverse interests, it seems likely that variance in the extent and characteristics of economic openness will increase.
In the extreme, this may result in the reemergence of economic blocs seen in the 1930s, with currency and trade restrictions in place (Jones, 2005, 2014). While these restrictions between blocs curtailed overall FDI, some countries still registered growing FDI from other blocs, as profits from existing operations could not be repatriated and were instead reinvested (Jones, 2005). The attendant rise in local assets and production may have helped MNEs mitigate the impact of trade barriers. Unfortunately, it also implied larger losses during World War II, when the combatants confiscated enemy assets (Jones, 2005; Jones & Lubinski, 2012).
Realism essentially predicts globalization in the presence of a global hegemon. In a multipolar world, in which at least two superpowers represent regional (in the sense of not global) hegemons and neither is strong enough to become a global hegemon, realism would expect the emergence of economic blocs around each regional hegemon, supported by different sets of institutions (regimes).
The most likely outcome under realism for the near to medium future is thus a repeat of the Cold War configuration of the international political economy: a bipolar world in which two superpowers erect two different systems in their respective spheres of interest. In this context, the United States remains less than a global hegemon, but strong enough to deny China the mantle of global hegemony. This may be a transitional phase for the next decades, or it may become a permanent condition if China fails to break through the middle income trap and thus does not grow much stronger (Lewin, Kenney, & Murmann, 2016; Witt, 2016).
In the longer run – presumably some decades hence – the world may see the emergence of China as a new global hegemon. For this case, hegemonic stability theory would suggest the creation of a new global order reflecting Chinese preferences. The smoothest transition would be for the United States to concede international leadership voluntarily to China. This seems unlikely. For a realist, it is more probable that leadership will be wrested from the United States. Possibly, this may occur in the context of a hegemonic war (Allison, 2017; Gilpin, 1983), but, given the threat of nuclear annihilation, conflict may be confined to economic or perhaps digital warfare.
How China would then refashion the international system is difficult to predict beyond such obvious changes as the Chinese yuan replacing the US dollar as reserve currency and new institutions replacing existing ones. Jacques (2012), for instance, extrapolated from history to suggest the emergence of a neo-tributary structure in which Chinese views of different races play an important role. Whether this is a credible prediction and specifically what this would entail remains unclear.
However, the longer run might also see the emergence of other superpowers and thus a multipolar world with multiple economic regimes. Since population size and thus the attainable total size of GDP is a key indicator of power in realism, the most likely candidate to attain superpower status in the future is India. This is contingent on continued economic development of the country, which is not a foregone conclusion given past economic performance and the possibility that especially China may seek to prevent the emergence of India as a competitor.
An unlikely candidate for superpower status is the European Union. The early 2000s saw a series of aspirational works that suggested that the EU might indeed be a coming superpower (Leonard, 2005; McCormick, 2007; Verhofstadt, 2006). The euro crisis put an end to such hopes, at least for now (Webber, 2016). In particular, the past 10 years have revealed that even though the EU would in principle have a sufficiently large economy and population, it lacks the “power conversion capability” requisite for superpower status (Nye, 2015: 25). The problem is that EU norms of near-consensual decision-making paired with the frequently diverging interests of member states drastically reduce the ability of the EU to act (Webber, 2016). Unless this changes as a result of further integration, presumably into a United States of Europe, superpower status is likely to remain beyond reach.
Other countries are generally too small to become poles in a multipolar world. This includes Russia. Given its relatively small population of 144 million, it could not match the power of any of the other superpowers even if it managed to attain per capita GDP levels as in the advanced industrialized world. By the same token, Japan with a shrinking population of currently 127 million will not be able to muster sufficient power.