New champions of preferential trade? Two-level games in China’s and India’s shifting commercial strategies

Abstract Following decades of relative isolation, China and India have become the world’s largest new traders. In this paper, we focus on their Preferential Trade Agreements (PTAs). While the two economies initially followed similar paths, with a growing number of PTAs signed in the first decade of the 21st Century, since 2011 India has taken a U-turn and stopped completing them. China, on the other hand, has widened and deepened its trade agreements. We present a novel theoretical framework to analyze international economic negotiations by emerging economies and use it to study the puzzling divergence of the trade policies of China and India. By adapting the two-level game framework to emerging economies, we argue that there are key differences in the political economies of countries like China and India (compared to Western industrialized ones), which requires a more specific focus on the domestic side of the two-level game. We show that accounting for non-legislative domestic ratification processes and for iterative games and experiential learning by domestic actors are crucial in understanding the trade strategies of emerging economies. While much of the literature explains large emerging economies by looking at external systemic factors, we instead suggest that their domestic politics trumps international politics.


Introduction
As a result of the Sino-US trade war and concerns about global value chains' dependence on China, much has been said about the potential of India as an alternative trade and production partner for the West. In India, Narendra Modi has launched an ambitious campaign (Make in India) aiming to attract foreign direct investment and manufacturing into the country. One would therefore expect to see an increasing share of manufacturing going to India, alongside a widening and deepening of Preferential Trade Agreements (PTAs) to facilitate this. What we observe is the opposite. While at the start of the millennium India followed China in pursuing an aggressive PTA strategy, the country has since reversed course and frozen its PTA ambitions. No new or upgraded agreements have been signed since 2011. What is more, India bargained intensely and ultimately withdrew from the region's most ambitious trade initiative, the Regional Comprehensive Trade Agreement (RCEP) in 2019. All of the above is puzzling, since this would seem a propitious time for India to pursue PTAs.
Drawing on Putnam's (1988) two-level game framework, we seek to answer the question why India has reversed course and stopped pursuing PTAs, while China continues to sign agreements of increasing depth and with a rapidly growing number of partners? To provide an answer, we carry out a comparative analysis of China's and India's PTA strategies, focusing on the interrelation between external and domestic drivers and obstacles. We do so by applying an adapted two-level game framework and, as such, aim to make both a theoretical as well as an empirical contribution to the IPE literature.
Theoretically, we offer a novel framework that adapts the domestic component of the two-level game to emerging economies in both democratic and non-democratic settings. We argue that there are important structural differences in the political economies of emerging economies (compared to Western industrialized ones), which require a more specific focus on the domestic side (inside-out) of the two-level game. The international level (outside in) has already been well covered in the literature (e.g., Hopewell, 2015, Narlikar, 2010. One adaptation is particularly relevant, namely considering non-institutionalized ratification mechanisms (i.e., ratification that does not take place by a parliamentary or congressional vote, as is the case in industrialized economies). We believe these informal ratification mechanisms to be different in authoritarian and democratic emerging economies. In authoritarian settings, the bureaucracy plays a central role, whereas in democratic emerging economies a larger number of social actors are involved in the ratification process. The second major modification we propose is to treat the game as an iterative one, in which negotiators base their negotiation position on information they receive from domestic actors regarding their experience with existing agreements. We claim that such 'experiential learning' plays an important role in PTA negotiations by emerging economies, since they are often part of wider liberalization strategies.
Taking the insights of our modified two-level game framework for emerging economies, we argue that the puzzling divergence in the trade policies of India and China can be explained by a consolidation of the pro-trade coalition of reformist policymakers and trade-dependent firms in China which led to a wider push for liberalization. The Indian case is very different, as the domestic pro trade coalition was less powerful to begin with, relying mainly on large service-goods exporters in its IT sector, and was further weakened due to slow progress in services trade liberalization. This trade-sceptic stance was reinforced by a growing and longstanding trade deficit, which cemented the belief in academic and policy-making circles that India was losing from signing-up to PTAs.

State of the art
Compared to the literature on US and EU trade policy (see e.g., Irwin, 2017;Poletti & De Bievre, 2014), the literature on trade dynamics in Asia is relatively new. Most attention has been paid to the region as a whole rather than individual countries, and has sought to answer why, after being 'committed trade multilateralists' for decades, governments in Asia turned to signing PTAs from the beginning of the 21 st Century (Sol ıs & Wilson, 2017). The shift has been attributed to external economic and political factors such as the inability of existing regional (APEC, AFTA) and global (GATT/WTO) regimes to liberalize trade multilaterally (Feridhanusetyawan 2005) or the fall-out from the 1997 Asian financial crisis (Desker, 2004). Others have argued that regional economic integration in other parts of the world motivated Asian countries to sign PTAs themselves to gain bargaining power (Dent, 2010). Ravenhill (2010) claimed that a 'political domino effect' was set in motion by China's rise and decision to sign PTAs, spurring Japan and other Asian economies to start their own PTA negotiations.
We do not question the 'external shocks' or 'political domino effect' logic behind PTA formation in China and India. Indeed, we think that this existing literature offers a plausible explanation as to why both countries decided to pursue PTAs in tandem in the first place. What is more, as the recently defined 'Indo-Pacific' is the epicenter of an ongoing power transition (Organski, 1958;Hopewell, 2020;Lavenex et al. 2021), systemic, security and geopolitical factors continue to be important to understand PTA dynamics in the region. Yet, we do see some shortcomings in this account. While providing a broad picture of the trade dynamics in the region, by 'black-boxing' the domestic political economy of individual Asian countries, differences in national responses to common external shocks are not accounted for (Aggarwal & Lee, 2011). There is little comparative work; rarely have the trade policies of China and India been evaluated in tandem (for an exception see Hopewell, 2015) and most of the existing research on these two countries have focused on their role at the WTO (Hopewell, 2016;Narlikar, 2010;Sinha, 2019). As such, much of the existing literature does not consider variation between China and India in terms of their PTA strategies. Particularly, little attention has been paid to previous PTA dynamics and/or little consideration has been given to the relationship between multilateral (WTO) and bilateral trade strategies, which could provide useful insights into the link between past trade strategies and more recent initiatives. Some existing works show that path-dependency and 'experiential learning' processes, resulting from iterated interaction, shape future trade agreements (Casas i Klett & Serrano Oswald, 2018;Elsig & Eckhardt, 2015).
Works looking at domestic factors affecting PTA strategies in China (e.g., Zeng, 2016;Eckhardt and Wang 2021) and India (Chaisse et al., 2011;Sinha, 2019) do exist but have mainly focused on the role of political elites in the policy-making process, and very little comparative work exists. Our own view echoes that of Hopewell (2016: 311), who argues that there is 'considerable variation in [China's and India's] sources of power and behavior in global economic governance'. Scholars typically argue that in the context of PTA strategies pursued by China and India, pressure from societal interests do not play a prominent role. This is because: (a) evidence points to a generally limited economic impact of PTAs in Asia and, as a result, firms have no incentive to put pressure on their government to pursue or not to pursue PTAs (Capling & Ravenhill, 2011); and (b) those domestic economic interest groups winning or losing from PTAs in China and India are usually badly organized and/or not well-represented politically (Wan, 2010). However, with PTAs growing in depth and scope in the Asia-Pacific, the argument that these have little economic or trade effects has been refuted in the literature (see e.g., Manger, 2014). What is more, research has also shown that firms and other domestic interests are in fact actively engaged in lobbying in both China (Kennedy, 2005) and India (Saha, 2019).
We build on this emerging scholarship to gain a better understanding of the different ways in which China and India decide trade policy in general and the role played by domestic actors other than state elites, such as bureaucracies and domestic interest groups. What is more, we also take into consideration policy shifts such as retrenchment and policy feed-backs over time. Such dynamics of experiential learning have not been taken into consideration by most existing accounts of trade policy dynamics in both countries (but see Sinha, 2019) and, as such, the existing literature frequently fails to capture the more nuanced inner workings of both states. We believe that experiential learning and domestic coalitions of pro-and anti-trade forces are central in explaining China's and India's PTA policies in general, and the recent divergence in those strategies in particular.
Adapting the two-level game framework for emerging economies Putnam's well-known two-level game metaphor provides a useful framework to evaluate the role of domestic politics in PTA policies while avoiding the pitfalls of producing ad hoc lists of countless domestic influences or of generic observations attempting to link national and international affairs (Putnam, 1988;Moravcsik, 1993). The framework takes into consideration both the international (outside-in) and domestic (inside-out) levels in negotiations between countries. The two levels exist, since bargaining between negotiators at the international level will result in a tentative agreement which then needs to be ratified domestically. In domestic ratification, particular attention is paid to the distribution of power, preferences and possible coalitions in determining the kinds of agreements that would be able to gain approval domestically. Ratification occurs either by bureaucratic agencies, interest groups, social classes ('public opinion'), or domestic legislatures (Putnam, 1988). Given that ratification in emerging economies does not take place in domestic legislatures, we lay our attention towards bureaucratic agencies, interest groups and public opinion. Our focus of analysis is on the domestic part of the game, and not the international one.
Although initially attention was paid to non-Western countries in the two-level game literature (see Evans et al., 1993), most existing work using the framework in the last three decades has looked at Western cases (L€ utz & Hilgers, 2019;Wolf & Zangl, 1996). As a result, the theory has developed in a Western-centricsome even claim US centric (Morin, 2010) direction. This is not to say that the theory has not been applied to the developing world at all, yet most scholars that have used the two-level game framework in this context have so far mainly used it either implicitly (e.g., Gallagher et al., 2005;Hopewell, 2015;Narlikar, 2010) or follow Putnam's approach without problematizing and/or theorizing the differences between developed and emerging economies, as well as between emerging economies themselves (see e.g., Chung, 2007;Odell, 2006). We therefore agree with Da Conceição and Mello (2017) that the analysis of two-level games can and should be applied to rising powers in a much more explicit and systematic manner.
We adapt the domestic component of the two-level game for emerging economies in both democratic and non-democratic settings by focusing on non-institutionalized ratification mechanisms and by considering negotiations as sets of iterative games instead of static ones. Our adapted framework can be applied to domestic negotiations in a wide variety of political economies of the Global South but, as a first test of our framework, we zoom into recent PTA negotiations of the two largest emerging economies and leading new traders: China and India.
In developed Western economies, legislative action is key for ratification. For example, the US Congress plays a central role through its granting of fast-track authority to the President and through the ratification of agreements. We first need to translate the two-level game framework into contexts where legislative ratification is not central for trade negotiations, as in China and India. Putnam never considered the ratification of agreements reached at the international level to exclusively take place through legislative votes. In fact, he mentioned other possibilities such as public opinion and/or key veto players (Putnam, 1988). We propose that the focus needs to be laid on domestic actors that have the capacity to impact trade policies, together with an understanding of what their interests are, where they come from, and whether they have veto power or not (Da Conceição & Mello, 2017). On this basis, we suggest that non-institutional ratification in authoritarian contexts takes place mainly through the bureaucratic apparatus, since these are the actors wielding veto power. In democratic ones it includes the bureaucracy, but goes beyond it to include public opinion and other societal structures such as business groups or labor unions, all of which may restrict policies.
There is a second relevant aspect to be considered when adapting the two-level game for emerging economies. In these markets, trade negotiations need to be considered as a set of iterative games. The reason for this is that PTAs in emerging economies are often used to lock-in domestic economic reforms, as part of wider liberalization efforts aimed at transforming their developmental states (Eckhardt & Wang, 2021). As a result, each individual negotiation offers lessons to both actors supporting and opposing liberalization strategies, which translate in the adjustment of strategies for the next iteration (i.e., PTA negotiation).
Below we propose two conjectures which focus on our two proposed adaptations to the two-level game framework: a) domestic actors' preferences and their influence; and, b) experiential learning.
With regards to preferences, just like any other trade policy decision, the signing of a PTA has distributional consequences and hence diverging views among domestic actors are to be expected. Economic actors to whom the net result of international trade is beneficial prefer liberal trade policies, while actors who lose overall, favor protection (Baldwin, 2006;Milner, 1997). Forces against PTA membership are typically associated with import-competing firms, which have a structural advantage over other societal interests as they usually have little difficulty in overcoming collective action problems (Hiscox, 2002). There are also domestic economic actors likely to lobby in favor of joining a PTA: export-oriented firms, as they reap the lion's share of the benefits from reciprocal (but not unilateral and multilateral) trade liberalization (Chase, 2003;D€ ur, 2007), as well as certain domestic import-dependent firms (Eckhardt & Poletti, 2016). The latter is related to the growing importance of intra-industry trade and the integration of firms in regional and global value chains (GVCs). Manger (2014) has shown that intra-industry and GVC trade is a particularly important driver of PTA formation in Asia in his detailed analysis of the Japan-Malaysia PTA, and the PTAs of both Japan and South Korea with ASEAN. We propose that the same logic would hold for China's and India's PTA formation.
With regards to the influence of domestic interests, as Sol ıs (2013) has shown in the Asian context, the willingness and capacity of domestic actors to influence trade policy depends on their cohesion and access to policy-making processes. That is, only agreements that threaten corporate profitability (or are likely to eliminate economic losses) mobilize businesses (Sol ıs, 2013). Postigo (2016) shows that as PTAs become increasingly complex, governments often lack information on PTAimpacts and, in turn, engage in intense consultations with societal actors. In other words, lobby success depends on the ability of firms to coordinate on joint positions and supply technical expertise to government officials, as well as the fit between business strategies and institutions (access points). This is also the main difference between democracies and autocracies. Lobbying in an autocracy like China mainly takes place through business links to the bureaucracy (Kennedy, 2005), whereas in democracies like India it involves wider coalitions and multiple access points (Kochanek, 1996). On this basis, our first conjecture suggests that: C1: The more concentrated domestic actors affected by PTAs are, the more information they possess, and the easier their access to decision-makers, the more likely they will be able to influence the pursuit and/or depth of PTA policies.
With regards to experiential learning, trade negotiations are complex affairs leading, the actors involved in them to address uncertainty by looking for information (Rosendorff & Milner, 2001). Given that negotiators from developing/emerging economies often lack sufficient knowledge about the future distribution of gains from agreements, these economies often face a problem of 'bounded rationality' (Poulsen, 2015): the complexity of agreements being negotiated, commonly on the basis of templates from Western trading powers, and a lack of experience, makes it difficult for these negotiators to fully grasp the effects until they enter into force. The literature suggests that in general learning happens when (new) information changes the beliefs of actors (whether individuals or firms) about cause and effect (Levy, 1994). If information originates from actors' own past experience the term 'experiential learning' is used (Elsig & Eckhardt, 2015). We suggest that PTA negotiators will draw on information received from domestic actors based on their experience with existing and past agreements and that this information in turn conditions their positions in subsequent negotiations. Negotiators focus both on the overall experience of domestic actors with a country's PTA strategy (whether domestic actors win or lose in general), and lessons learned from specific PTAs. Based on this, negotiators periodically update the information received and, based on their assessment of the distributional consequences for domestic actors, may decide to change PTA strategy or even shift their preference. Our second conjecture is therefore as follows: C2: Domestic actors draw lessons from previous PTA policies (in terms of their perceived gains or losses) and adapt their preferences towards pursuing and/or deepening PTAs accordingly.

Methodology
To explore our research puzzle and test our conjectures, we carry out an in-depth structured comparative analysis of China's and India's PTA negotiations in the 2000-2021 period. Some have questioned the validity of theory building or refinement through case study analysis, however as Odell (2001) has put it, 'case studies have illuminated virtually every subject studied by political economists' and have proven particularly useful for refining 'received theory'. This is certainly true for the work on trade negotiations in general and the two-level game literature in particular, on which we draw in this paper. Case studies of trade negotiations have provided rich raw material for the building and development of the two-level game framework in the past (e.g., Evans et al., 1993;Gallagher et al., 2005;Hopewell, 2015;Narlikar, 2010;Wolf & Zangl, 1996), so our paper falls within a long tradition of case study-led research to further our understanding of trade negotiations. The reason why case studies are a particularly useful method for the refinement of existing theories is inter alia because they are generally better than alternative methods for documenting and understanding processes of continuity and change in IPE. Furthermore, comparative case studies supply more thorough support for testing theoretical conjectures than most single case studies (Odell, 2001), since a structured focused comparison allows to control for similar factors (e.g., in the case of China and India, both countries being large emerging economies with high trade to GDP ratios), while allowing to test conjectures such as those we have proposed regarding domestic ratification and experiential learning.
Our analysis relies on a detailed study of primary and secondary sources as well as 33 semi-structured and in-depth interviews (see Annex 1). 1 It is well established in the literature that expert interviews enable researchers to gain access to difficult (less transparent) fields such as trade negotiations and that they offer opportunities to expand their access to the field (in this case governmental bureaucracies), as well as to cross-validate information (see e.g., Bogner et. al. 2009). We have selected our experts on the basis of their in-depth knowledge about and/or firsthand experience with the trade policies and negotiations of their respective countries (China and India). That is, all interviewees were involved either in domestic discussions determining trade policies or in trade negotiations themselves. This also applies to the various academics we have interviewed.
Emerging economies, as a result of their rapid economic growth, have had to swiftly build up capacity in multiple and complex trade issue-areas (see Lavenex et al., 2021). They have often addressed this gap by drawing on existing expertise in their research institutions. In China, this often happens through (non-public) workshops and study sessions where academics interact with policymakers and party members. In India, academics are often directly involved in policy-shaping processes. Given the difficulty of obtaining access to actors directly involved in sensitive policies such as trade negotiations, we spent several months through numerous stays in Beijing and Delhi, in order to expand our access to further experts through a snowball sampling technique, in which information provided by experts helped us identify further potential interviewees. In both cases, we were able to carry out follow-up interviews with actors involved in trade policy making at various points in time, thus giving us a better sense of the evolution of trade policies in China and India respectively. The interviews were conducted in person by the authors between 2010 and 2021 at various locations in China, India and Switzerland. One of the interviews took place by videoconference, due to the Covid-19 pandemic. In the case of India, most of our interviews were carried out in 2014, however we were able to carry-out additional interviews in 2016 and in 2021 with a longstanding contributor to India's trade strategy. Our final data source is the Design of Trade Agreements (DESTA) Database, 2 which we use as our main source to evaluate the partners and depth of PTAs signed by China and India. Although the paper introducing the database was published some time ago (D€ ur et al., 2014), the data has been updated until January 2022.

China's and India's PTA strategies
China and India began engaging in deep economic reforms in the early 1980s and 1990s. Part of this reform process was a gradual opening-up of their domestic markets, turning both countries into the worlds' largest and most consequential new traders. According to World Bank data, 3 China's trade to GDP ratio grew from 11 percent in 1978, when reforms started, to 64 percent at its highest point, in 2006. In 2019, the year before the Covid-19 pandemic began, trade still made for well over a third (35.8 percent) of China's GDP. Less studied, but equally important is that India followed a similar path, albeit a decade later: its trade to GDP ratio rose from 15 percent in 1990 to nearly 56 percent in 2011; by 2019 it was 39.3 percent. Based on the data from the World Bank, India's trade to GDP ratio has been higher than China's for every single year since 2011.
As part of an extensive trade liberalization drive which began when both countries joined the World Trade Organization (WTO), albeit with the significant difference that India had been a GATT member, the two countries have become involved in the worldwide push to sign PTAs. Both countries have been actively pursuing such agreements during the past two decades, placing them amongst the countries which have signed the most PTAs per country in the Asian region. 4 Tables 1 and 2 give an overview of China's and India's PTA activity over time, including their depth (defined here as the level of obligation and the stringency of enforcement mechanisms), which tells us whether or not agreements only cover trade in (industrial) goods or also include sensitive products like agriculture, as well as how much room for manoeuvre PTA partners have and how formalized their interactions are (Hicks & Kim, 2012;Hofmann et al., 2017). The PTA strategies of both countries show similarities: such as a deepening over time since the mid-2000s, and their widening to partners beyond the Asian region, particularly to Latin America. However, after 2011 India stopped its drive to pursue PTAs, while China intensified its PTA policy.
India's de facto moratorium on PTAs and the resulting divergence between the two countries' PTA strategies is surprising, especially since India has become a major trader. Most of the older Indian PTAs were in fact shallow with a narrow scope and with relatively small trading partners, and had limited trade-liberalizing effects. However, when measured by their depth, several more recent agreements show a similar path of liberalization to that of China in the first decade or so of the 21 st Century: namely, those with Singapore (2005), South Korea (2009), Malaysia (2011) and Japan (2011). Yet, India has not initiated any trade negotiations since 2011 and gridlock is observable in all major trade initiatives where India takes part, such as the RCEP (from which India recently withdrew), the EFTA-India PTA, or the EU-India PTA.
China, on the other hand, continued to be actively engaged in expanding an ever-growing network of PTAs after 2011. While being a relatively new player to the PTA game, as Table 1 shows, China's agreements have become increasingly comprehensive and deep over time. There has also been an interesting evolution in terms of China's PTA partner choice. China at first solely signed agreements with  small trading partners, yet in recent years it has concluded agreements with developed economies in Asia Pacific (Australia, New Zealand and south Korea) and Europe (Iceland, Switzerland). It has also launched negotiations with Japan; was the driving force behind the RCEP, which includes Japan, New Zealand and Australia (Smith, 2021); signed an investment agreement with the EU in 2020 as a first step towards establishing PTA; and has sought a Bilateral Investment Treaty (BIT) (and possibly a PTA) with the US since 2013.
A two-level game analysis of China's PTA policies In China the main actors involved in discussing and ultimately approving PTAs are bureaucracies, in particular the Ministry of Commerce (MOFCOM), but also the National Development and Reform Commission (NDRC) and other affected units in the State Council such as the Ministry of Agriculture (MOA), or the Ministry of Information Industry (MII). The NDRC in particular retains strong planning attributes (it succeeded the former State Planning Commission) and is often referred to as the 'small' State Council (Cabinet) since it has 26 departments and bureaus and its jurisdiction spans over a large number of public policies including monetary, infrastructure, industrial, agriculture and foreign economic relations (Yeo, 2009). Other important veto players are bodies in the Chinese Communist Party (CCP) related to these bureaucracies. One interviewee explained the different trade positions in the bureaucracy as follows: "Chinese interests are mainly domestic; this also has to do with the bureaucratic set-up. So, for example, the Ministry of Finance, the Central Bank, MOFCOM and the Foreign Affairs Ministry are all pretty much in favour of reform and trade openness and this may have to do that they commonly interact with foreign actors, while the NDRC, MOA and MII, tend to be more protectionist" 5 Think tanks, which tend to be supportive of economic reform and liberal trade policies, also play an important role in advising these actors on their economic policies. 6 Differences also exist at the local (provincial) level, with coastal areas generally more open towards economic reform and trade (e.g., Guangdong, Shanghai, Beijing, Tianjin) and the rustbelt provinces in the Northeast (e.g., Liaoning, Jilin, and Heilongjiang) more closed. 7 Within the CCP, factions historically hold different positions on trade policy, with some strongly resisting changes to the domestic economy (Breslin 2003). However, under Xi Jinping, and due to concerns of China falling into a 'middleincome trap', a centralization has taken place ('top-level design') which seeks cautious but far-reaching economic reforms. 8 This centralization is important since, as Young (2011) argues, it matters whether there are multiple agenda setters or not at the executive level; and this also determines whether executives are able to go beyond the limits set by domestic actors or not (L€ utz, 2011). Under the administration of Hu Jintao, a more collegial set-up at the Standing Committee of the Politburo meant that a wider number of agenda setters existed, something which was at times interpreted as unnecessarily slowing-down the policy-making process. This is an important distinction between China and India, where despite centralization attempts under Narendra Modi, multiple agenda setters persist (see next section). 9 Other actors influence China's economic policies beyond party and government elites. Kennedy (2005) critiques the often, and wrongly, shared assumption that vested interests do not influence public policy since they are allegedly 'subject to a state-controlled association system that effectively blocks their ability to defend their interests at the national level or because they establish patron-client ties with bureaucrats to obtain privileged access [and] avoid onerous regulations' (Kennedy, 2005: 2). China's political economy has in fact changed substantially throughout its reform and opening period, impacting relations between state-and societal actors. China's economy gained in complexity, with a variety of public and private actors at different levels winning or losing from the signing of trade agreements, and willing to protect their interests (Wan, 2010). There is a growing body of work, which shows that that firms in China not only mobilize but also have the ability to successfully influence policy-making via trade associations, various intermediaries, and even through the direct lobbying of regulators and bureaucratic agents (Kennedy, 2005;Jiang, 2010).
In our interviews with policymakers, academics and trade negotiators, we find general support for the idea that firms are indeed also politically active and influential actors in China's PTA decision-making. More specifically, objections to PTAs usually come from import-competing agricultural firms and service providers, while firms in most other sectors actively lobby in favor of PTAs to further their interests. Of the pro-trade interests, export-oriented firms are typically most actively engaged politically, according to our interviewees. 10 It does matter whether a PTA partner is an important export destination for the sector/firms in question in this regard and, as hypothesized above, access to local or national bureaucrats is indeed crucial for wielding influence on decision making. 11 Our interviews also revealed the relevance of size when it comes to political strategy and firm influence. One interviewee explained: "Global firms know much better how to push for their interests. However, they focus on small issues that are crucial for them … On the other hand, one also sees a pull model (government driven) on some issues. Usually the pull-side covers 'big-issues' or 'big talk' in part reflecting that sometimes the government does not know much about particular industries." 12 In relation to this, another interviewee mentioned the rise of big Chinese multinationals as a key driver behind China's PTA policy: "An important change is the rise of big private firms such as Huawei, or Lenovo. These companies are generally pro-trade and well connected and hence able to influence trade policy." 13 Another of our interviewees added to this: "There is plenty of lobbying, however this takes place in different forms. Small firms mainly lobby at the local level. For example, Huada Jierui (华达杰瑞) is an important biotechnology firm based in Shenzhen, but it has no strong networks at the national level. Hence, the Mayor of Shenzhen will meet once a month with representatives from the local firms and hear their concerns (in Beijing local government offices act as important actors lobbying at the national level). Big firms such as Huawei or Haier have very good networks both locally and nationally. Personal networks with the different ministries are particularly important. In this sense smaller firms are mainly reactive and do not shape FTA negotiations. But if they are hurt afterwards, they will try to get something." 14 Yet another interviewee pointed at the fact that foreign firms often play a much more significant role in the decision-making process than Chinese firms: "A substantial part of private firms in China do not really care about trade policy, as they mainly produce for foreign companies and hence are not independent operating entities. In other words, their trade interests are basically in line with what is most beneficial for the companies they are producing for and they expect those companies to raise any concerns regarding trade policy.
[These] foreign (owned) companies are economically very important and they can have a big influence on China's trade policy. They may put pressure on the Chinese government through direct lobbying or they ask their home state to raise a certain issue with the Chinese … or they could (threaten to) leave the country." 15 The factors outlined above support our first conjecture, that the degree of concentration, information, and access to bureaucrats are relevant factors in explaining Chinese PTA strategy. We also find some evidence for our second conjecture regarding policy feed-backs from experiential learning by domestic actors. In the literature (e.g., Yang, 2009) there is evidence suggesting that as a newcomer, Beijing chose to sign PTAs with relatively small players to learn how to negotiate trade agreements. This point about diplomatic learning also came up in several of our interviews. 16 One concrete example mentioned was that the PTAs with EFTA countries would provide valuable negotiating and diplomatic lessons for a possible Sino-EU PTA in the future. 17 However, what we are interested here is the learning process of Chinese domestic actors based on their experience with previous PTAs and how this in turn has informed China's PTA strategy.
Various surveys conducted in the 2010-13 period show that in the first 10 years after China started to sign PTAs in earnest, such experiential learning took place. Based on one survey of more than 1000 exporting firms in China, Japan, Malaysia, the Philippines, Singapore, South Korea and Thailand, Kawai and Wignaraja (2013) show that, in this period, there was a sharp increase in Chinese exporting firms making use of PTA preferences and that Chinese firms were the greatest users of PTA preferences in Asia. Zhang (2010) surveyed 232 Chinese export-dependent firms (including foreign owned firms) and showed that 45% indicated they use PTA preferences and, when plans for using PTAs were also taken into account, almost 80% of Chinese firms said they either used or planned to use PTA preferences. What is more, the majority of firms indicated that their perception of PTAs was positive, as 'they benefited from an increase in export sales due to widening market access followed by greater convenience due to easier facilitation' (Zhang, 2010: 26). These figures on usage of PTA preferences and perceptions of PTAs among Chinese firms are substantially higher than those shown by studies carried out a just a couple of years prior and show that, in relatively short period of time, exporters in China became much more positive about the economic potential of PTAs (Kawai & Wignaraja, 2013). These findings are also in line with data gathered around the same time on the economic effects of China's PTAs, which shows that their net benefits are much more significant than previously thought. Urata and Okabe (2010) find clear evidence for trade creation effects in the context of China's PTA network, while others have shown that, in 2013, 27% of total Chinese trade took place within the setting of a PTA, which meant that China's PTA coverage at that point was already higher than that of the US (26%) and only slightly lower than that of the EU (34%) (Kawai & Wignaraja, 2013).
Our interviewees confirmed that experiential learning by domestic actors, including foreign owned enterprises, has taken place and has played a role in China's PTA strategy since then. 18 The recently signed agreements with developed economies like Australia, New Zealand and South Korea can be seen in light of the Chinese leadership taking into consideration the (positive) experiences of domestic actors with previous PTAs. Experiential learning based on these recent PTAs with relatively small, highly industrialized countries is in turn feeding into current negotiations with more advanced economic partners like Japan, the EU and the US. 19 Experiential learning does not only seem to play a role in PTA-partner choice but also in the substance and design of China's PTAs. One concrete example has to do with China's shifting position on liberalization of agriculture. China originally refused to discuss agricultural products during PTA talks because of strong resistance from the agriculture sector, which led for instance to a long stalemate in PTA negotiations with Australia (Ravenhill & Jiang, 2009). However, China has become more willing to discuss agricultural products in trade negotiations. The first-time China accepted the elimination of tariffs on agricultural products was in its PTA with New Zealand, accepting to phase-out by 2019 tariffs on imports on almost all main agricultural products (Kawai & Wignaraja, 2013). Since then, China has accepted a comprehensive coverage of agricultural products in almost all its PTAs, including the agreement with Australia, which was eventually signed in 2015 (Salidjanova, 2015). As in the Japan-Mexico agreement (Sol ıs & Katada, 2007), in China, trade liberalization through PTAs has also over time tipped the domestic balance in favor of pro-liberalization forces; in particular, by showcasing PTAs' advantages and limiting the blocking potential of the agricultural lobby. Trade liberalization (and liberal reforms in China more generally) has strengthened (often private) export oriented firms vis-a-vis more protected State Owned Enterprises (SOEs) and this has helped to consolidate domestic pro-trade coalitions, thus creating a virtuous circle of trade liberalization. 20 It must be noted that at the same time China's agricultural subsidies have risen dramatically in the last two decades (Hopewell, 2019) and this could be seen as a way to better protect domestic sectors affected by liberal reforms and trade liberalization, leading to reduced opposition to PTAs.
Another area where a clear shift has occurred in China's position is the inclusion of trade in services PTAs. In its first-generation PTAs, China only agreed to include the General Agreement on Trade in Services (GATS) principles. The country was forced to include these because of its WTO membership, but refused to go further due to pressure from domestic service providers (Ravenhill & Jiang, 2009). More recently, China has become one of the few Asian countries willing to move beyond its WTO commitments, by including GATS-plus commitments in PTAs. 21 The most comprehensive Chinese PTA in this regard is the agreement with Singapore where coverage of services trade includes a chapter on the movement of natural persons (Kawai & Wignaraja, 2013). In its agreement with Australia, Switzerland and Korea, China has also accepted the inclusion of additional sectors and improvements in areas such as environmental, financial, air transport, tourism, health care and logistics services, as well as for providers of short-term contractual services (The Economist, 2014).
We deem it plausible that this inclusion of service provisions in PTAs is related to experiential learning by domestic interests. Apart from the more general logic described above about the strengthening of domestic pro-trade forces, liberalization in services is particularly important to facilitate GVC trade and investments. There has been a drastic increase in the integration of Chinese firms in GVCs, which requires the elimination of cross-border impediments and facilitation of trade and FDI in the region and beyond. As Zeng et al. (2021) have shown, recent Chinese PTAs with strong services provisions have fostered both backward and forward GVC linkages for Chinese firms. We therefore believe that Beijing has come to view its PTA network as part of a supporting policy framework for deepening production networks and supply chains formed by foreign MNCs operating in China and Chinese owned exporters. We suggest that this is one of the main reasons why China's PTAs have become much deeper over time. There may be alternative or complementary explanations for the latter. As recently suggested by Hopewell (2020), China has been able to curtail US institutional or rule-making power in the global trading system as part of the ongoing hegemonic struggle between these two powers. While Hopewell's analysis is based on the WTO and the OECD, one could envisage that PTAs are another avenue through which China engages in rule-making, and thus an important motivation for the Chinese leadership to initiate, champion, and conclude PTAs.
A two-level game analysis of India's PTA policies As in China, legislative ratification is not necessary for signing PTAs in India. International treaties (including PTAs) are executive acts that can be concluded trough the written approval of the Cabinet and which accordingly give central roles to bureaucracies, especially the Ministry of External Affairs and the Ministry of Trade and Industry. However, if PTA obligations entail changes to domestic law, legislative action is necessary (Ministry of External Affairs, 2018). Another difference between the two countries is that, unlike China where the number of vetoplayers in the executive have been reduced under Xi Jinping, in India, despite Narendra Modi's centralizing attempts, cabinet members continue to have veto roles. Even if the Ministry of Commerce (MoC) leads trade negotiations, five different ministries (with different constituencies) are usually involved in them. The process was explained by one of our interviewees: "The Ministry of Commerce will circulate a white paper among the other ministries. Ministries have in a way veto power in their areas, when they are concerned, their views prevail over those of the Ministry of Commerce (e.g., Ministry of Agriculture). Once this first round is completed, the Ministry of Commerce consults with the major stakeholders (affected business associations e.g., car, textiles etc.). Once these views are taken into consideration a proposal is drafted (usually at the sub-secretary level). This policy proposal is then sent again to the other ministries for a second round of consultation (they might have been approached by stakeholders or raise particular issues). Once agreement is achieved, this goes to the cabinet where the issue is discussed (all ministers plus the Prime Minister). Usually, ministers will respect the position brought by the other ministers if it is in their field (unwritten rule/gentleman's agreement). So, all in all domestic negotiations tend to be more cumbersome and complicated than the international ones." 22 Even within the MoC divisions exist, as it includes both an industrial policy and a trade branch. Foreign direct investment, regulated by the Finance Ministry, spurs turf-wars. These blockages are in line with the expectations of the literature examining institutional constraints on policy-making and the role of veto players (see e.g., Tsebelis, 2002). As in China, where provincial representative offices in Beijing play important lobby functions, representative offices of the Indian States in New Delhi lobby for regional interests. However, states in India are much more powerful than Chinese provinces, an interviewee described it as follows: "State governments are extremely important. The 28 states have their police forces, are in charge of education (primary) and jointly with the Federal government for higher education. Perhaps most importantly, they are also in charge of subsidies. So, if the Federal government enacts a particular policy state governments might follow it if it serves them, or they might ignore it if it does not serve them. Another important issue has to do with party allegiances, so if the party in power at the central level is similarly in power in a given state, they are more likely to follow the policies of the government. If, however it is not, they will do whatever they want and ignore the federal government." 23 States can be very important for specific issues included in PTAs, for example on government procurement, as another interviewee noted: "This is a big problem why India is not able to negotiate PTA's. At the federal level this is not an issue, and India has already opened government procurement at the federal level through the WTO. However, this is a huge issue at the state level. Some states are very unlikely to agree to open government procurement at the local level. And here, as with many other issues, states matter, given our federal decentralized system. For other parties who would be willing to sign an FTA with India, given that significant investment is made at the state level, this is a very important issue. One has to be reminded that some of the states are gigantic, with populations of hundreds of millions of people." 24 Business associations and informal groups are relevant domestic actors too. Amongst the latter, the Bombay Group has played a central role in reversing the government's early trade liberalization efforts in the mid to late 1990s (Sinha, 2019). Protectionist interests are relevant. The influence of import competing visa-vis export sectors looms large in India given its long history of closed markets and import substitution industrialization (ISI). For the first thirty years after independence, India followed a restrictive licensing regime whose intent was virtually to eliminate imports of consumer goods (Krueger, 2010). India's tariff levels remain among the highest for a large economy. That being said, pro-trade interests exist as well. Some industries have become strongly competitive since the reforms of the 1990s, particularly in the services sector. One would expect these firms to be the most active in applying pressure towards trade liberalization, given that in this sector India has rapidly increased its export share. For example, exports of services to the United States grew from 1.9 billion USD in 2000 to 16.9 billion in 2011more than eight-fold; and from 12.7 billion USD in 2009 to 29.7 in 2019nearly 15 times the 2000 figure. In contrast, exports of goods to the United States amounted to 10.7 billion USD in 2000 and 'only' 36.2 billion in 2011; by 2019 the figure was 57.7 billion USD, displaying a more modest growth pattern -nearly six-fold from the 2000 figure. 25 Compared with China, a different picture emerges from the manufacturing sector: trade with all main partners, except for the United States and China, is on primary products. Little incentives to further PTAs exist, since primary products are easily traded and do not face major tariffs. While the economic reforms of the 1990s onwards increased manufacturing exports, no category or categories of manufactures has shown exceptional growth; and hence no growth engine may be found among them. Roy (2020) shows that the liberalization of the 1990s instead of leading to a regain in manufacturing, due to the high-costs of capital in India (the banking sector is mainly stated-owned making access to credit often a matter of good connections) led to a large wave of bankruptcies. Bardhan (2020) similarly sees rent and cronyism in public banks as a main reason for India's modest industrial development.
The service sector on the other hand has been active in shaping trade policy, particularly the National Association of Software and Services Companies (NASSCOM) and the Software Technology Park in India (STPI). These sectors sought the inclusion of service-export issues such as labor mobility for service suppliers under GATS mode 4 in PTAs. 26 However, trade partners, due to concerns about immigration, have been reluctant to include this aspect in PTAs (with a few exceptions such as Japan). 27 In the cases where services have been included, such as the agreement with Singapore, its implementation has disappointed (supposed to take full-effect after 90 days, and still unfulfilled). 28 The difficulty of including services in PTAs appears to be one of the main reasons behind the lack of interest in signing new PTAs and the stalled negotiations on, for example, the India-EU agreement and the India-Australia agreement. Furthermore, the benefits of existing PTAs have been limited for most Indian exporters. 29 In terms of relative gains the fact that most countries already have low tariffs, whereas those of India remain comparatively high, means that the costs of signing new PTAs for India are substantial. The more surprising then, that India's deeper agreements are like those of other large traders in their rules of origin and dispute settlement. As explained by an interviewee: "Most RTA's focus on manufactured goods but India's main interest is services. Everyone thinks of IT; however, this is only a small part, and is very strong in construction, tourism, and education. For example, on the RTA with South Korea they didn't want to sign an FTA on services because we are very competitive on education. For this reason, an agreement with the EU will probably not happen. They are asking for things we will never agree to, and we are asking for labour mobility (mode 4), which I don't think they will agree to." 30 Another interviewee argued along similar lines: "Clearly services. Maybe in a decade or so, if what people are saying that some manufacturing will move from China to India, maybe then this could change. But for this to happen there are many obstacles, such as building the necessary infrastructure. At the moment, it is clearly services. When I say services, one needs to think beyond IT, which is what everyone thinks about, or even highly qualified labour. A lot of services have been moving inland so they are not even concentrated in the big cities like Bangalore. A lot of this is low skill, inputting things in computers and the like, but also other services like construction. Here one big advantage has been language capabilities (English)." 31 The agricultural sector has not had an interest in pushing for trade liberalization either. In part this is because until the mid-1990s agricultural exports were not allowed. This changed with the Uruguay Round, after which India began exporting high-value added agricultural products such as fruits and flowers. However, these exports never constituted more than 18% of agricultural production (Roy, 2020). Another important reason why agricultural liberalization has been limited is the long history of famines in India. This was a main reason why India traditionally limited exports of agricultural produce. A vast majority of the population still lives in rural areas and Indian politics are still strongly local affairs, where most important policies are implemented (Bardhan 2020: 33).
The expectations of our first conjecture are thus partially supported as the wellorganized service industry has managed to have access and influence policymakers. However, given that trade negotiations did not lead to meaningful concessions, this coalition weakened. Another reason for the weakening of the pro-trade coalition is that it did not include other important sectors such as industry and agriculture, which either suffered from openness (manufacturing) or failed to gain substantially from it (agriculture). A longstanding adviser to the government on trade matters (both in the Singh and Modi administrations) explained that a lack of consultation with broader economic sectors was to blame for a lack of a clear PTA strategy, but that this has changed with the RCEP negotiations: "A key problem for India's policymakers has been figuring out what domestic preferences are … The only consistent red line has been agriculture; but other than that, a general problem of India's PTA strategy has been coming up with a clear negotiating strategy, objectives and non-negotiable issues. India's negotiating partners clearly do so (e.g., the EU with labor rights and the environment), but this has been more difficult for India. RCEP was in this sense very important because it led to the kind of domestic debate, which made it much clearer what our red lines were. So, this was an important learning process to uncover domestic preferences, to know what domestic interests are important. This sounds strange, but in reality, for a long time the government was able to sign PTAs without much opposition because they did not really matter, no-one was too much affected. The first time there was some opposition was with the South Asian Free Trade Area (SAFTA) in 2004. Then, the state of Kerala was worried about competition from Sri-Lanka (e.g., on textiles and agricultural products). The central government was able to compensate Kerala and together with a strong push for liberalization by PM Singh, who had put his career on the line for this objective, the agreement went through." 32 As to why RCEP failed, the interviewee added: "RCEP from an Indian perspective was in essence an India-China PTA, since our trade with ASEAN is limited and we already have an agreement with Japan and South Korea. This was also the problem, since even before the border issues and other political tensions India had concerns about China's "cheating." For example … when China enacted a policy of 0% port-fees for ships carrying export goods from China to India. This was thus a central concern in the RCEP negotiations, and more generally, the fact that India has had a trade-deficit with RCEP partners for the past 20 years. So even liberals in the government supporting the negotiations had to admit, this is a very long time to have a widening deficit (you can't keep this going on forever). There were significant discussions inside the government, with both voices in favor and against the deal, but in the end, we realized we could not have a deal with China." 33 The RCEP case shows that learning from iterative negotiations plays a role in India's trade strategy. However, the case also shows that our second conjecture only applies for a limited number of domestic actors (especially policy-makers and academics, including the NITI Aayog, of which more below) and not, as we had expected, for businesses, public opinion, NGOs, or labor unions due to a lack of consultations with these actors. This finding is surprising inasmuch as we expected such results in autocratic settings like China, but not in democratic ones like India. In fact, as explained above, the RCEP negotiations were the first ones in which a broader number of domestic actors were able to shape trade policy.
That being said, experiential learning did play a role in India's trade policy reversal, albeit in a different manner to what we had expected in our second conjecture. Policymaking circles, informed by academic studies demonstrating little trade effects of PTAs for India, shifted trade strategies accordingly. This resonates with our expectations about experiential learning. Shortly after Narendra Modi came to power, he replaced the existing planning commission (established after India became independent and in charge of its five-year plans) with a policy commission (Aayog in Hindi). The latter is known as the National Institution for Transforming India (or NITI Aayog). NITI Aayog shares a similar origin with China's NDRC (both succeeded former state planning agencies), but the two institutions are not comparable. 34 Even if they both play important roles in PTA strategies, the NDRC (created in 2003) is a major regulator with a large bureaucracy in both central and provincial levels. NITI Aayog was only created in 2015, it does not have a large bureaucracy, is only present at the national level, and has a small budget of ca. USD 50 million. 35 NITI Aayog is essentially the government's thinktank, although it plays an important role in policy-decisions. The NDRC competes with other ministries (e.g., MOFCOM) to set up policy, whereas NITI Aayog has no similar peer-competitor. The key role of the institution is gathering views from different stakeholders to guide policy-decisions. As such, rather than permanent staff, NITI Aayog outsources studies to third parties (albeit to well-known academics and policymakers).
By the mid-2000s, academic studies had shown that India was not benefiting from its PTAs (e.g., Akhter & Ghani, 2010, Joshi, 2012, Srinivasan & Archana, 2009, revealing that even deeper agreements which were expected to have significant effects did not lead to trade increases, as with India-South Korea, or had only marginal effects, as in the case of India-Japan (Pant & Deepika, 2015: 36). NITI Aayog was given the task of evaluating India's trade policies. In 2018 it made public a critical and influential report 'A Note On Free Trade Agreements and their Costs' (Saraswat et al., 2018). The report concluded that 'India should carefully review existing Free Trade Agreements (FTAs) before negotiating new ones'. Drawing on an evaluation of India's recent PTAs it suggests 'unfavorable gains to our trade partners'. The main critical points of the report include: a) that India's exports to PTA partners have not outperformed overall export growth or exports to rest of the world; b) that PTAs have led to both increased imports and exports, but that the increase has been larger in imports; c) that India's trade deficit with ASEAN, Korea and Japan has widened post-PTAs; d) that its main beneficiaries have been metal importers and textile exporters; e) that tariff reductions do not boost exports significantly (since Indian exports are highly responsive to income changes and not to price changes); and f) that the PTA utilization rate is very low (between 5% and 25%). This last point is notable when compared to China where a growing number of exporters has made use of PTAs.

Conclusion
In this article, we provided a novel theoretical framework to analyze and understand the trade strategies of emerging economies such as China and India. By building on the two-level game literature, we argued that we need to take into account non-legislative ratification processes and view trade negotiations as iterative games to explain the preferential trade strategies of emerging economies. In other words, to meaningfully apply the two-level game framework to countries like China and India, we need to adapt the domestic side (inside-out) of the equation. A deepening and widening of PTAs should reflect, first of all, lobbying from domestic actors benefiting from them (if they were able to form coalitions and gain access to policymakers) but we propose that there is a difference in this regard between authoritarian (China) and democratic (India) emerging economies. In the former, business ties to the bureaucracy plays a pivotal role, whereas in the latter a larger number of social actors (including the bureaucracy) are involved in ratifying trade agreements. We also proposed that domestic actors would learn from the effects of previous PTAs and adjust their preferences accordingly. Our findings partly support these theoretical expectations.
In the case of China, we find that economic actors with strong links to local or national bureaucrats that benefited from China's PTAs have indeed pushed for a continuation of PTA policies and deeper agreements. In part, these actors have been successful since the executive (CCP leadership and bureaucracy) have less divergent preferences than what was the case when China began signing PTAs. Policymakers in MOFCOM, but also within the CCP, appear to have drawn lessons about the positive effects of trade liberalization for domestic economic reform, and have sought to continue widening and deepening them, albeit carefully. The rising number of firms taking advantage of PTAs has translated into a strong lobby for further liberalization being able to take on protectionist interests from some government ministries, SOEs and import-competing sectors. Widespread subsidies have allayed the concerns of sectors and bureaucracies opposed to liberalization, even if the latter often has harmful effects on trade partners (see e.g., Hopewell, 2019).
In the case of India, the evidence is more mixed. Although initially we did find evidence of a strong pro-trade coalition as a result of India's services industry in areas such as IT (but also beyond that in sectors such as education and construction), this coalition has weakened. Despite concentrated interests in business associations such as NASSCOM and STPI, a lack of results in services-PTA negotiations has dampened its interest and influence. More generally, until the RCEP negotiations, the number of domestic firms and other relevant actors involved in trade policy was very limited. This reflects a lack of consultation from government ministries and the limited effects of PTAs, as seen by very low utilization rates. We find partial support to our second conjecture in that (central) government and (a limited number) of domestic actors have drawn lessons from previous PTAs and adjusted trade strategies accordingly. However, rather than PTAs tipping the domestic balance against liberalization as the perceived costs of these agreements rose over time (as we expected), in fact we find little involvement of most bureaucratic agencies, interest groups and public opinion. Instead, the widening trade-deficit and the lack of success in services negotiations led to a re-think in about the benefits of PTAs, mainly in the cabinet and the planning agency NITI Aayog. That being said, there is one partner which has an interest in negotiating intra-services agreements with India: the United Kingdom. Services are very relevant in the UK economy and Brexit shifted the balance of economic power in favor of India and made the UK more willing to sign trade agreements with large emerging economies. This makes an India-UK PTA in the near future likely.
Frequently, economic relations of large emerging economies such as China and India are examined from the perspective of international politics and systemic factors. Our study suggests that their trade policies cannot be explained without understanding their domestic political economies. This aspect is even more relevant when, as a result of the Sino-American trade war, attention is overwhelmingly focused on US-China dynamics. Other large countries have agency too, as we have seen in the case of India with regards to RCEP negotiations. This is not to say that the international dimension is irrelevant, such as China's hegemonic struggle visa-vis the US (Hopewell, 2020); in fact, in our view, these hegemonic ambitions reinforce existing political economy dynamics in China, translating into a deepening and widening of PTAs. Less understood is that India also has (regional) hegemonic ambitions, as can be seen by its rapprochement with the US and its participation in the Indo-Pacific Strategy, both of which go against India's long held policy doctrine of strategic autonomy. What is crucial is that India's hegemonic struggles have not led to a similar outcome as in China because of domestic politics. Thus, even if international politics play an important role, China's and India's trade policies should not be primarily evaluated under this perspective. In this study, we have shown that the size and complexity of these countries means that their domestic politics will continue to trump international politics; this should not be overlooked by scholars seeking to understand the shifting commercial strategies of these large new traders.